Article
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Multidisciplinary Design Optimization of a vehicle system for safety, NVH (noise, vibration and harshness) and weight, in a scalable HPC environment, is addressed. High performance computing, utilizing several hundred processors in conjunction with approximation methods, formal MDO strategies and engineering judgement are effectively used to obtain superior design solutions with significantly reduced elapsed computing times. The increased computational complexity in this MDO work is due to addressing multiple safety modes including frontal crash, offset crash, side impact and roof crush, in addition to the NVH discipline, all with detailed, high fidelity models and analysis tools. The reduction in large-scale MDO solution times through HPC is significant in that it now makes it possible for such technologies to impact the vehicle design cycle and improve the engineering productivity.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... The structural optimization design of truck cab crash safety is a multi-objective optimization problem, which requires a repeated finite element analysis [36][37][38][39]. In the automotive industry, A-pillar impact is usually used to design and verify the safety of vehicle structures [40,41]. Therefore, as an example, the surrogate model of a truck cab is constructed, and the accuracy of the model is tested in this paper. ...
... According to the ECE R-29 standard, the crash safety performance of the cab is tested by measuring the living space of the fiftieth-percentile male manikin model after the crash test [40]. Specifically, the living space is represented by characteristic dimensions, which means the minimum distances between the manikin and any inelastic part of the cab. ...
Article
Full-text available
In order to overcome the problem of the low fitting accuracy of the expected improvement point infill criteria (EI) and the improved expected improvement point infill criteria (IEI), a high-precision surrogate modeling method based on the parallel multipoint expected improvement point infill criteria (PMEI) is presented in this paper for solving large-scale complex simulation problems. The PMEI criterion takes full advantage of the strong global search ability of the EI criterion and the local search ability of the IEI criterion to improve the overall accuracy of the fitting function. In the paper, the detailed steps of the PMEI method are introduced firstly, which can add multiple sample points in a single iteration. At the same time, in the process of constructing the surrogate model, it is effective to avoid the problem of the low fitting accuracy caused by adding only one new sample point in each iteration of the EI and IEI criteria. The numerical examples of the classical one-dimensional function and two-dimensional function clearly demonstrate the accuracy of the fitting function of the proposed method. Moreover, the accuracy of the multi-objective optimization surrogate model of a truck cab constructed by the PMEI method is tested, which proves the feasibility and effectiveness of the proposed method in solving high-dimensional modeling problems. All these results confirm that the Kriging model developed by the PMEI method has high accuracy for low-dimensional problems or high-dimensional complex problems.
... Given that there are variables in the MDO problem that are not shared between all disciplines, i.e. have their own relevant subsets of design variables, approximations can be built on only the sub-set of the design variables that is significant to each discipline whilst the optimisation problem is solved in the full design variable space. There are several examples of the use of this approach in the automotive industry, e.g. by Sobieszczanski-Sobieski et al. (2001), Kodiyalam et al. (2004), Ollar et al. (2014) and Ollar et al. (2015) and Ryberg et al. (2015). The benefit is that sampling and approximation building can be carried out in a space of reduced dimensionality. ...
... Such judgement may be based on, for instance, engineering experience or design variable ranking studies. In the simplest form of sub-space approximations, used in the work by (Sobieszczanski-Sobieski et al. 2001), Kodiyalam et al. (2004), Ollar et al. (2014) and Ryberg et al. (2015), deficiencies in sub-space partitioning, i.e. by failing to identify significant variables, can result in approximation errors that cannot be resolved by additional sampling. This is due to changes in response values as a consequence of changes in the insignificant variables. ...
Article
Full-text available
An approach to solving multidisciplinary design optimisation problems using approximations built in sub-spaces of the design variable space is proposed. Each approximation is built in the sub-space significant to the corresponding discipline while the optimisation problem is solved in the full design variable space. Since the approximations are built in a space of reduced dimensionality, the computational budget associated with building them can be reduced without compromising their quality. The method requires the designer to make assumptions on which design variables are significant to each discipline. If such assumptions are deficient, the resulting approximations suffer from errors that are not possible to reduce by additional sampling. Therefore a recovery mechanism is proposed that updates the values of the insignificant variables at the end of each iteration to align with the current best point. The method is implemented within a trust region based optimisation framework and demonstrated on a multidisciplinary optimisation of a thin-walled beam section subject to stiffness and impact requirements.
... where s 0 e, i is the objective function, representing the sensitivity value of the i order modal frequency v i subvariable b of the element e. 12 The values of car body mass and stiffness depend on the material properties and its specific structural makeup. Based on the above modal frequency sensitivity theory, the first vertical bending frequency is set as the objective function s 0 e, i . ...
... According to the solutions to equation (12), the natural frequencies can be expressed as follows ...
Article
Full-text available
High structural modal frequencies of car body are beneficial as they ensure better vibration control and enhance ride quality of railway vehicles. Modal sensitivity optimization and elastic suspension parameters used in the design of equipment beneath the chassis of the car body are proposed in order to improve the modal frequencies of car bodies under service conditions. Modal sensitivity optimization is based on sensitivity analysis theory which considers the thickness of the body frame at various positions as variables in order to achieve optimization. Equipment suspension design analyzes the influence of suspension parameters on the modal frequencies of the car body through the use of an equipment-car body coupled model. Results indicate that both methods can effectively improve the modal parameters of the car body. Modal sensitivity optimization increases vertical bending frequency from 9.70 to 10.60 Hz, while optimization of elastic suspension parameters increases the vertical bending frequency to 10.51 Hz. The suspension design can be used without alteration to the structure of the car body while ensuring better ride quality.
... Multidisciplinary design optimization (MDO) has gained popularity in both industry and academia due to its success in using numerical optimization techniques to find improved designs. Notably, MDO has been successful in designing aircraft [1,2], automobiles [3,4], wind turbines [1,5], launch vehicles [6] and satellites [7,8]. ...
Conference Paper
Full-text available
Recently, the graph-accelerated non-intrusive polynomial chaos (NIPC) methods have been introduced to effectively address a variety of uncertainty quantification (UQ) challenges in multidisciplinary and multi-point models. The essence of these graph-accelerated NIPC methods lies in leveraging tensor-grid input points for sampling the random space and using a computational graph transformation method, AMTC, to efficiently perform the tensor-grid evaluations by taking advantage of the inherent sparsity within the computational graph of the computational model. Despite their advancements, the effectiveness of these methods in tackling optimization under uncertainty (OUU) problems remains unexplored. This paper presents a detailed case study for a laser-beam-powered aircraft design problem. The focus is on applying the graph-accelerated NIPC methods to solve a large-scale multidisciplinary design optimization problem under uncertainty. This study not only compares the results of multidisciplinary optimization (MDO) and OUU but also highlights the effectiveness of the graph-accelerated NIPC method in tackling OUU challenges. The numerical results show that the OUU-optimized design is more robust under the variations of the flight conditions. Additionally, the AMTC method accelerates the optimization time by a factor of five, making the computational cost of the OUU problem only twice that of the MDO problem.
... The complexity of engineering systems arises from their involvement of multiple disciplines that interact with one another. This interaction in multidisciplinary systems is common in various fields, including aerospace [1], marine applications [2,3], automobile engineering [4], and renewable energy [5,6]. For instance, a system with coupled fluid and structure disciplines [7] and another with coupled aerodynamic and structure disciplines [8] represent some of the many multidisciplinary systems encountered in engineering. ...
Conference Paper
Many engineering systems involve multiple interacting disciplines or subsystems. For a design or analysis task, unknown linking variables, which are those variables that are outputs of some disciplines and inputs of other disciplines, are obtained by solving the system of implicit interdisciplinary compatibility equations for a given set of system inputs. This study creates surrogate models for linking variables using label-free training with neural networks. The compatibility equations are embedded in the cost function of the model training. They are calculated and are not solved for given input training variables, thereby avoiding label acquisition. To quantify the prediction errors of the surrogate models, we build their error models with Gaussian Process regression, which uses the existing training points and the derivatives of the compatibility equations at the training points. The error models are then used to compensate for the errors of neural network surrogate models of the linking variables, producing more accurate predictions of linking variables with quantified model uncertainty for predicting system responses. The linking variables with quantified model uncertainty are then used to predict the system responses and associated prediction errors. We demonstrate the effectiveness of the proposed method by the application to a propane combustion problem.
... Currently, the natural frequency of a body is transformed through changing wall thickness and cross-sectional shape of the structure [4,5], causing the wall resonance to thereby reduce the noise. Jian [6] has applied the reinforcing rib and changed wall thickness to alter the stiffness of a SUV body panels, declining the interior noise. ...
Article
Full-text available
Interior noises of vehicles would be caused when the vibration of body panels excites the indoor air. In the paper, the vibration load of engine was obtained firstly through experiments. Secondly, the engine load was applied in the finite element model of body in white to compute the vibration velocity and realize virtual reality, indicating that the front support of the body had large vibration velocity when the frequency was lower than 60 Hz. The boundary element was then adopted to compute the interior noise and extract the sound pressure at a point near the driver's head. Two obvious peaks were shown in sound pressure curves, at 270 and 310 Hz, respectively. The body panels that had obvious impact on the interior peak noise were determined through the panel contribution analysis, and the interior peak noise was remarkably reduced after applying sound absorption materials on these panels. Nevertheless, many more additional sound absorption materials were not always better. If a multilayer of sound absorption materials was needed, an optimal value was existed in the thickness of sound absorption material. And a great impact would be played toward the interior noise of the cabin by the reasonable selection of different sound absorption materials and their thicknesses. Finally, the neutral network (NN) was also used to predict interior noises, which was compared with the result of the boundary element. The maximum difference between the prediction values of NN and boundary element was within 5 dB, indicating that the neural network was feasible to predict the interior noise. Subsequently, the neural network method would be applied to conduct the optimization analysis for the interior noise.
... Optimizing the design of an aircraft requires an integrated optimization framework that considers how the different disciplines interact with each other. MDO has also been applied to other complex engineering design problems for automobiles [12][13][14], wind turbines [15][16][17], launch vehicles [18], satellites [19,20], electric aircraft [21], and electric vertical take-off and landing (eVTOL) aircraft [22]. ...
Article
Full-text available
For engineering design optimization, the full-space formulation offers the potential for greater efficiency than the more commonly used reduced-space formulation. This potential is greater when the numerical model involves discretized partial differential equations or coupled disciplines. However, the full-space formulation results in a larger optimization problem with at least a factor of two increase in the number of optimization variables and equality constraints. Using Newton-type methods to solve such problems involves solving a large-scale and, often, ill-conditioned Karush–Kuhn–Tucker linear system at each optimization iteration. This can be time-consuming to solve even with a Krylov solver. If the number of iterations is reduced, the full-space formulation could be applied to a broader class of problems. This paper presents an inexact quasi-Newton algorithm with an adaptive extension for solving large-scale equality-constrained optimization problems. The new algorithm inexactly solves the Karush–Kuhn–Tucker system using new inexactness criteria that are derived to ensure a descent direction. The adaptive extension chooses the stopping condition of the Krylov solver by also taking its convergence rate into account. The paper presents results of numerical experiments applying this algorithm to three types of problems: six constrained optimization problems from the widely used CUTEst test suite, a bar thickness optimization problem, and a two-dimensional topology optimization problem. For all problems, the new algorithm consistently shows a roughly 50% reduction in the total number of Krylov solver iterations and a minimum of roughly 15% reduction in the optimization time. Moreover, the proposed approach for selecting the Krylov solver tolerance shows an improvement in all cases, whereas the existing forcing-parameter approach shows an increase in the number of Krylov iterations in some cases. These results indicate that this new method for selecting solver tolerances is effective and robust, and a good choice in algorithms that use a Krylov solver for solving the Karush–Kuhn–Tucker linear system.
... Furthermore, it is highly unlikely to find a global optimum because of the noisy nature of crash events. Using metamodels trained from a design of experiments has proven to be a good option for the solution of crash problems as shown by Redhe et al. (2002); Kodiyalam et al. (2004); Forsberg and Nilsson (2005). The application of metamodel-based optimization to the design of TRPs is presented by Chuang et al. (2008); Duan et al. (2015). ...
Article
Full-text available
The majority of parts in modern car bodies is manufactured from sheet metal. Rarely these parts are fully stressed due to design space restrictions and complex requirements. The usage of tailor rolled blanks (TRB) enables the reduction of sheet thickness in areas less loaded and thus reduces part weight. Technically most sheet metal parts are potentially suited for the application of TRB. Economic circumstances like the additional flexible rolling process and technology-specific nesting constraints limit the application to a subset of parts. The search for the best candidate parts taking mass and cost into account is currently challenging. This article presents an optimization strategy for the selection of the parts in a vehicle structure that are best suited for the application of TRB. As a first step, a priori preferencing is performed to select parts based on engineering rules. Using a reduced number of candidate parts, high quality metamodels are trained to perform multiobjective optimizations of all possible combinations of remaining parts, revealing the most efficient part selection under consideration of mass and cost.
... These disciplines are often strongly coupled and cannot be solved independently. The MDO research was mainly driven by aircraft design (Giunta et al., 1997;Benaouali & Kachel, 2019), and its applications have been expanded to other engineering systems, such as racing cars (Kodiyalam et al., 2004) and wind turbines (Grujicic et al., 2010;Ashuri et al., 2014). To solve the MDO problem, the MDO architecture, including problem formulations and organization strategies, plays a critical role. ...
Article
Full-text available
Design for additive manufacturing (AM) involves decision making in various design domains, including product design, material selection, and process planning. In practice, engineers typically adopt a sequential design process to optimize these design domains in consecutive order. However, coupling factors, e.g. shared variables, related constraints, and conflicting objectives, are not sufficiently considered within the sequential design process, resulting in an inefficient workflow and suboptimal design solutions. To address the above issues, this paper proposes a multidisciplinary design optimization framework to simultaneously optimize different domains, which enables rapid exploration and complete exploitation of the AM design space under complex constraints. More specifically, the proposed framework is based on the concurrent optimization method, which coordinates the optimization of different design domains by allowing an automated exchange of design information. Also, the framework utilizes the surrogate modeling approach to approximate high-fidelity simulations for facilitating the iterative process. The effectiveness of the proposed framework is validated with two examples, a plate with a hole design and a hook design, which involve multiple design objectives from both process and structure domains, i.e. the print time, print area, strain energy, and maximum von Mises stress.
... Systems of computer models constitute the new frontier of many scientific and engineering simulations. These can be multiphysics systems of computer simulators such as coupled tsunami simulators with earthquake and landslide sources [23,32], coupled multiphysics model of the human heart [26], and multidisciplinary systems such as automotive and aerospace systems [7,16,34]. Other examples include climate models where climate variability arises from atmospheric, oceanic, land, and cryospheric processes and their coupled interactions [12,15], or highly multidisciplinary future biodiversity models [31] using combinations of species distribution models, dispersal strategies, climate models, and representative concentration pathways. ...
Article
The state-of-the-art linked Gaussian process offers a way to build analytical emulators for systems of computer models. We generalize the closed form expressions for the linked Gaussian process under the squared exponential kernel to a class of Mat\'ern kernels, that are essential in advanced applications. An iterative procedure to construct linked Gaussian processes as surrogate models for any feed-forward systems of computer models is presented and illustrated on a feed-back coupled satellite system. We also introduce an adaptive design algorithm that could increase the approximation accuracy of linked Gaussian process surrogates with reduced computational costs on running expensive computer systems, by allocating runs and refining emulators of individual sub-models based on their heterogeneous functional complexity.
... The results showed that the crashworthiness and the lightweight effect were improved by multidisciplinary optimization design. Kodiyalam et al. (2004) carried out multidisciplinary optimization design for the whole vehicle under the constraints of crashworthiness, noise, and durability, and the better overall performance was achieved successfully. Furthermore, compared with other optimization methods, the hierarchical optimization method has several advantages: parallel optimization calculation, unlimited series, and strict convergence proof. ...
Article
Full-text available
Door system plays a very important role in the domain of automobile passive safety. In order to improve its side crashworthiness, some affiliated impact components are assembled. Conventional impact parts can make the door system possess enough stiffness and strength so as to ensure the integrity of car body when side collision occurs. But for occupant protection, the excessive rigidity may increase the risk of occupant injury. To address this problem, this work first introduces a kind of negative Poisson’s ratio (NPR) structure, and proposes a novel door system which is composed of NPR energy-absorbing block, NPR impact beam, inner panel, outer panel, and reinforcing plate. Next, parameter sensitivity analysis for each performance index is conducted to determine the corresponding variables when constructing the approximate model. Then, aiming at the disadvantages of the Technology for Order Preference by Similarity to Ideal Solution (TOPSIS) and the Mean Square Deviation Method (MSDM) in processing performance index data, a novel integrated weighting method is used to determine the weighting coefficient of each performance index. Finally, considering side structural crashworthiness, occupant protection, and lightweight, the hierarchical optimization for the novel door system is conducted to further enhance its overall performance. The result demonstrates that compared with the conventional door, the optimal door can improve the performance of occupant protection and ensure the side crashworthiness more effectively.
... The performance of an automotive can be improved in a cost-effective manner by integrating optimization techniques in the design process. First, studies were realized by simple Monte Carlo search approaches, e.g., Kodiyalam [13] Kodiyalam and Sobieski [14]; however, it became soon evident that it could be extremely time-consuming, especially when the number of design variables, load cases, and domain nonlinearity increases in the design consideration. A part of this difficulty might be removed by improving computing facilities. ...
Article
A novel optimization technique was implemented to investigate the effects of vibrations on comfort of occupants to maintain oscillations in an acceptable zone in accordance with the International Organization for Standardization 2631 standard. In this regard, a newly introduced comfort indicator was defined as discomfort criterion (DiC). The effectiveness of the proposed measure was investigated throughout the suspension optimization of an in-wheel motor electric vehicle which almost doubled the unsprung mass by adding an electric motor to the wheel assembly. First, a spatial oscillatory model of the electric vehicle with eight degrees of freedom was developed, and the linear quadratic regulator control scheme is selected to control an actuator in an active suspension. Road excitations were then generated by applying the power spectral density of road class B–C provided by the International Organization for Standardization 8608 standard. The exceedance from the reduced comfort limit (in accordance with the International Organization for Standardization 2631 standard) and wheel travel (WT) of the vehicle were considered as design objectives. Finally, using a novel optimization procedure, the optimum condition and impact factor of the design variables, as well as counterplots of the design objectives with respect to the effective design parameters, were extracted and analyzed. Results proved the proposed indicator, that is, discomfort criterion (DiC) as a reliable measure to assess suspension systems’ performance effectively.
... 21 Other scholars have also contributed the field of body structure weight loss. [22][23][24][25][26][27] However, since the body structure of a vehicle is a complex engineering system, the statics and dynamics, crash safety, and the influence of fatigue strength on the multidimensional performance of the system are typically considered independently. A comprehensive multi-objective design method to minimize weight is lacking and presents an urgent problem that needs to be solved. ...
Article
Full-text available
In this study, the integrated MSOT (M-Multi-dimensional factor autobody model, S-Screening autobody component, O-Optimization of plate thickness, T-Testing, and validation) integration method is adopted to optimize the automobile body structure design for weight reduction. First, a multi-dimensional factor body model is established, then components of the vehicle are screened for the most important targets related to weight reduction and performance, and a multi-objective optimization is performed. Virtual experiments were carried out to validate the analysis and the MSOT method were proposed for lightweight design of the automobile body structure. A multi-dimensional performance model that considers stiffness, modality, strength, frontal offset collision, and side collision of a domestic passenger car body structure. Components affecting the weight of the vehicle were identified. Sheet metal thickness was selected as the main optimization target and a multi-objective optimization was carried out. Finally, simulations were performed on the body structure. The comprehensive performance, in terms of fatigue strength, frontal offset collision safety, and side collision safety, was verified using the optimized Pareto solution set. The results show that the established MSOT method can be used to comprehensively explore the weight reduction of the body structure, shorten the development process, and reduce development costs.
... Nevertheless, structural optimization usually engages costly function evaluations. For example a simulation of a traveller automobile for crash acquires around 27 hours with an probable computational cost of about US Dollars 5,200/- [15]. As a result, alternate technique of function evaluations; for instance, DOE and RSM are generally engaged in design engineering to minimize the computational costs. ...
Article
Full-text available
Wing is the most critical part of an aircraft whose structural weight holds prime importance. It is always desirable to reduce the weight of an aircraft. In this work, a detailed analysis of wing is carried out and results pertaining to stress distribution are recorded. Commercial software ANSYS is used to perform simulations related to stress, strain and displacement distributions. Moreover, the optimization design of an aircraft wing was conducted using design of experiments (DOE) and response surface method (RSM) technique for the reduction of computational cost. The aircraft wing is efficiently optimized by changing thickness under the same loading and flight conditions. Results show that this optimization technique is useful to reduce stresses as well as overall weight of the structure with minor variation in thickness of components.
... Systems of computer models constitute the new frontier of many scientific and engineering simulations. These can be multi-physics systems of computer simulators such as coupled tsunami simulators with earthquake and landslide sources ( Ulrich et al. 2019, Salmanidou et al. 2017), coupled multi-physics model of the human heart ( Santiago et al. 2018), and multi-disciplinary systems such as automotive and aerospace systems (Fazeley et al. 2016, Kodiyalam et al. 2004, Zhao et al. 2018). Other examples include climate models where climate variability arises from atmospheric, oceanic, land, and cryospheric processes and their coupled interactions ( Hawkins et al. 2016, Kay et al. 2015), or highly multi-disciplinary future biodiversity models ( Thuiller et al. 2019) using combinations of species distribution models, dispersal strategies, climate models, and representative concentration pathways. ...
Preprint
We generalize the state-of-the-art linked emulator for a system of two computer models under the squared exponential kernel to an integrated emulator for any feed-forward system of multiple computer models, under a variety of kernels (exponential, squared exponential, and two key Matérn kernels) that are essential in advanced applications. The integrated emulator combines Gaussian process emulators of individual computer models, and predicts the global output of the system using a Gaussian distribution with explicit mean and variance. By learning the system structure, our integrated emulator outperforms the composite emulator, which emulates the entire system using only global inputs and outputs. Orders of magnitude prediction improvement can be achieved for moderate-size designs. Furthermore, our analytic expressions allow a fast and efficient design algorithm that allocates different runs to individual computer models based on their heterogeneous functional complexity. This design yields either significant computational gains or orders of magnitude reductions in prediction errors for moderate training sizes. We demonstrate the skills and benefits of the integrated emulator in a series of synthetic experiments and a feedback coupled fire-detection satellite model.
... -Automobile and Aeronautics: This field has a lot of simulation and modeling, model prediction and verification including probabilistic modeling, computer aided drawing, graphic designing, design automation, the design of structures, automated plan building, analysis of design, and concrete modeling [66,73,98,130]. -Astrophysics and Quantum Physics: A lot of applications based on physics, especially on quantum physics and astrophysics, has very large computations as they receive a large input data. Load-balancing of Spin-image Algorithm on CPU and MIC have been studied in [22,34]. ...
Chapter
High-performance computing (HPC) plays a key role in driving innovations in health, economics, energy, transport, networks, and other smart-society infrastructures. HPC enables large-scale simulations and processing of big data related to smart societies to optimize their services. Driving high efficiency from shared-memory and distributed HPC systems have always been challenging; it has become essential as we move towards the exascale computing era. Therefore, the evaluation, analysis, and optimization of HPC applications and systems to improve HPC performance on various platforms are of paramount importance. This paper reviews the performance analysis tools and techniques for HPC applications and systems. Common HPC applications used by the researchers and HPC benchmarking suites are discussed. A qualitative comparison of various tools used for the performance analysis of HPC applications is provided. Conclusions are drawn with future research directions.
... High Performance Computing (HPC) has long been used by the manufacturing industry to support product design, optimization or testing. The automotive industry has perhaps a dominant position with a number of applications but other industries have seen some HPC use in their production processes [8,11]. In the USA, the current 'High Performance Computing for Manufacturing' (HPC4Mfg) programme aims to grant access to HPC facilities for selected industry partners to address key challenges in manufacturing [5]. ...
Conference Paper
Significant effort is currently being invested on enhancing digitization in all aspects of industrial processes and manufacturing on the quest for the next transformational change towards sustainable manufacturing and the factories of the future. This vision, commonly termed as Industry 4.0, is at the core of a number of funded projects in Europe. This short paper briefly reports experiences from the EU-funded project DISRUPT and focuses on the identification of some key challenges in the interplay between Industry 4.0 and High Performance Computing.
... The use of decision-support tools (DSTs) that optimize process, quality, and costs in silico has been remarkably powerful at mitigating scale-up challenges in a wide range of industries (Kodiyalam, Yang, Gu, & Tho, 2004;Schmidt, 2005), recently including the production of cell-derived therapeutics (Rekhi et al., 2015;Tan et al., 2014), but not EVs. Systematic and modular modeling of manufacturing processes offers great insight into cost structure and allows consideration of case-specific needs and constraints (Hassan et al., 2015;Simaria et al., 2014). ...
Article
Full-text available
Newly recognized as natural nanocarriers that deliver biological information between cells, extracellular vesicles (EVs), including exosomes and microvesicles, provide unprecedented therapeutic opportunities. Large‐scale and cost‐effective manufacturing is imperative for EV products to meet commercial and clinical demands; successful translation requires careful decisions that minimize financial and technological risks. Here, we develop a decision support tool (DST) that computes the most cost‐effective technologies for manufacturing EVs at different scales, by examining the costs of goods associated with using published protocols. The DST identifies costs of labor and consumables during EV harvest as key cost drivers, substantiating a need for larger‐scale, higher‐throughput, and automated technologies for harvesting EVs. Importantly, we highlight a lack of appropriate technologies for meeting clinical demands, and propose a potentially cost‐effective solution. This DST can facilitate decision‐making very early on in development and be used to predict, and better manage, the risk of process changes when commercializing EV products.
... Especially since various crash load cases have to be considered for the development of a body structure. The influence of the crash load cases on the computational costs are described in Kodiyalam et al. (2004). In the case of a full vehicle optimization, the time required to obtain the optimization results can quickly increase to several weeks (Duddeck 2008). ...
Article
Full-text available
A challenge in the design and optimization of vehicle structures is the high computational costs required for crash analysis. In this paper an automated model generation for simplified vehicle crash models is presented. The considered crash load cases are the US NCAP (100%, 56 km/h), the Euro NCAP (40%, 64 km/h) and the IIHS Small Overlap (25%, 64 km/h). The generation of the physical surrogate vehicle models is based on different sub-steps which were automated using a process chain. With this process chain it is possible to evaluate very efficiently the influence of structural modifications on the global crash behavior. During the model generation the crash behavior of the surrogate model is directly compared with the full vehicle model to enable a direct assessment of the model quality. Since the interface, where the model is cut, is an important factor for the obtained correlation, different interface positions were analysed. With obtained solutions it is possible to identify the interface position, which fulfils the required correlation by a given computational time. Additionally, the interface discretisation is analyzed to identify the model configuration with the highest correlation. This investigation was performed for three different vehicle models.
... The performance of an automotive can be improved in a cost-effective manner by integrating optimization techniques in the design process. First, studies were realized by simple Monte Carlo search approaches, e.g., Kodiyalam [13] Kodiyalam and Sobieski [14]; however, it became soon evident that it could be extremely time-consuming, especially when the number of design variables, load cases, and domain nonlinearity increases in the design consideration. A part of this difficulty might be removed by improving computing facilities. ...
Article
Full-text available
N oise, vibration, and harshness (NVH) attribute is needed to be included in the vehicle structure design since improving the NVH characteristics enhances the ride quality experienced by the occupants. In this regard, an efficient method was proposed to investigate the structural dynamic response of an automotive body considering low-frequency NVH performances. Moreover, the improvement of an automotive structure under the constraint of NVH behavior was investigated by using the design of experiments (DOEs) method. The DOEs methodology was for screening of the design space and generating approximation models. Here, the thicknesses of panels consisting of a body-in-white (BIW) of an automotive were employed as design variables for optimization, whose objective was to increase the first torsional and bending natural frequencies. Central composite design (CCD) for DOEs sampling and response surface methodology (RSM) were employed to optimize the dynamic stiffness. Moreover, the effects of the selected variables as well as their binary interactions were modeled and the optimum conditions for rigidity improvement were obtained via the RSM. Furthermore, the validity of the proposed optimization scheme was verified through CAE analysis. The results indicated that the first torsion and bending natural frequency were improved compared to the baseline design. Additionally, precise surrogate models in polynomial terms for the first bending and torsion natural frequencies were obtained.
... The multidisciplinary design optimization (MDO) is widely used in designing complicated coupling system. [1][2][3][4][5] The traditional MDO is performed based on deterministic conditions and ignores the uncertainties in real life. Uncertainties such as manufacturing tolerances, material properties, and boundary conditions can reduce the system reliability. ...
Article
This article presents a procedure for reliability-based multidisciplinary design optimization with both random and interval variables. The sign of performance functions is predicted by the Kriging model which is constructed by the so-called learning function in the region of interest. The Monte Carlo simulation with the Kriging model is performed to evaluate the failure probability. The sample methods for the random variables, interval variables, and design variables are discussed in detail. The multidisciplinary feasible and collaborative optimization architectures are provided with the proposed method. The method is demonstrated with three examples.
... However, one type of such a study found in the literature involves minimising the mass of the vehicle body considering noise, vibration, and harshness (NVH) and crashworthiness. Examples are described by Craig et al. [6], Sobieszczanski-Sobieski et al. [7], and Kodiyalam et al. [8], which are all executed using Optimiser Analyser … Analyser System Optimiser Subspace Optimiser + Analyser … Subspace Optimiser + Analyser metamodels and single-level optimisation methods. There are automotive MDO studies that use direct optimisation [9] but metamodel-based design optimisation is the most common approach when computationally expensive models are involved, as described in the recent article by Rakowska et al. [10] since it is more computationally efficient [11]. ...
Article
Multidisciplinary design optimisation (MDO) can be used as an effective tool to improve the design of automotive structures. Large-scale MDO problems typically involve several groups who must work concurrently and autonomously in order to make the solution process efficient. In this article, the formulations of existing MDO methods are compared and their suitability is assessed in relation to the characteristics of automotive structural applications. Both multi-level and single-level optimisation methods are considered. Multi-level optimisation methods distribute the design process but are complex. When optimising automotive structures, metamodels are often required to relieve the computational burden of detailed simulation models. The metamodels can be created by individual groups prior to the optimisation process, and thus offer a way of distributing work. Therefore, it is concluded that a single-level method in combination with metamodels is the most straightforward way of implementing MDO into the development of automotive structures. If the benefits of multi-level optimisation methods, in a special case, are considered to compensate for their drawbacks, analytical target cascading has a number of advantages over collaborative optimisation, but both methods are possible choices.
... High performance computing (HPC) systems are critical for complex large scale design studies (Kodiyalam et al. 2004). While HPC systems deliver high computational power (capability computing) or high throughput (capacity computing), they are static resources with little scalability or flexibility. ...
Conference Paper
Service ORiented Computing EnviRonment (SORCER) is a Java-based network-centric computing platform. SORCER provides a service oriented architecture, which enables the implementation of parallel algorithms in a dynamic distributed computing environment. SORCER is often used for multidisciplinary aircraft design analysis and optimization. However, the current approach often assigns intense optimization algorithms to run entirely on single overloaded nodes, rather than evenly distributing the workload. The goal of this work is to provide lower-level optimization algorithms as integrated SORCER services and study the overhead of doing so. VTDIRECT95, a Fortran 95 implementation of D. R. Jones' algorithm DIRECT, is a highly parallelizable derivative-free deterministic global optimization algorithm. QNSTOP is a parallel quasi-Newton algorithm for stochastic optimization problems. The potential benefit of integrating VTDIRECT95 and QNSTOP into the SORCER framework is to provide dynamic load balancing among computational resources at the optimization level, resulting in a dynamically scalable process.
... Aircraft analysis and design entail complex simulations, some I/O intensive, others requiring high floating point performance and high memory bandwidth. High performance computing (HPC) systems are critical for large scale design studies (Kodiyalam et al. 2004). While HPC systems deliver high computational power (capability computing) or high throughput (capacity computing), they are static resources with little scalability or flexibility. ...
Article
With rapid growth in the complexity of large scale engineering systems, the application of multidisciplinary analysis and design optimization (MDO) in the engineering design process has garnered much attention. MDO addresses the challenge of integrating several different disciplines into the design process. Primary challenges of MDO include computational expense and poor scalability. The introduction of a distributed, collaborative computational environment results in better utilization of available computational resources, reducing the time to solution, and enhancing scalability. SORCER, a Java-based network-centric computing platform, enables analyses and design studies in a distributed collaborative computing environment. Two different optimization algorithms widely used in multidisciplinary engineering design-VTDIRECT95 and QNSTOP-are implemented on a SORCER grid. VTDIRECT95, a Fortran 95 implementation of D. R. Jones` algorithm DIRECT, is a highly parallelizable derivative-free deterministic global optimization algorithm. QNSTOP is a parallel quasi-Newton algorithm for stochastic optimization problems. The purpose of integrating VTDIRECT95 and QNSTOP into the SORCER framework is to provide load balancing among computational resources, resulting in a dynamically scalable process. Further, the federated computing paradigm implemented by SORCER manages distributed services in real time, thereby significantly speeding up the design process. Part 1 covers SORCER and the algorithms, Part 2 presents results for aircraft panel design with curvilinear stiffeners.
... Hence, weight reduction appears inevitable and urgent in order to improve the endurance and performance capability of new energy vehicles, with the urgent need of environment protection and its growth momentum. Many studies have focused on wei ght reducti on of the vehicl e body (Sobieszczanski-Sobieski et al. 2001;Kodiyalam et al. 2004) but, as well as reducing weight, improving crashworthiness properties is a pivotal issue. To date, many studies have focused on structure optimization or material matching. ...
Article
Full-text available
With the rapid development of the vehicle industry, crashworthiness has become a crucial aspect in vehicle body design. In fact, crashworthiness is a multivariable optimization design problem for a vehicle body, regardless of structure or material. However, when crashworthiness involves a large number of design variables, including both material and structure variables, it is more difficult to deal with. In this paper, an integrated design technique for materials and structures of vehicle body under crash safety consideration is suggested. First, a finite element model of the vehicle body is established according to relevant vehicle safety standards. Then, the material parameters of the vehicle body are set as analytical factors for factor screening. Next, significant factors are obtained using a three-level saturated design integrated with multi-index comprehensive balance analysis and the MaxUr⁽³⁾ method, with an improved evaluation method. These screened material parameters along with the corresponding continuous variables of the structure, are considered as the design variables of the integrated design of the vehicle body. Both the weight and the crashworthiness properties are set as the design objectives. Optimal Latin hypercube sampling and radius basis functions are utilized to construct highly accurate surrogate models. Furthermore, the non-dominated sorting genetic algorithm II is implemented to seek the optimal solutions. Finally, two cases considering the roof module and the frontal module of a vehicle body are analyzed to verify the proposed method.
... Although still in its infancy, mathematical optimization techniques are increasingly being applied to the crashworthiness design of vehicles. Early crashworthiness studies were followed by response surface-based design optimization (Etman et al. 1996;Avalle et al. 2002;Kurtaran et al. 2002;Fang et al. 2005;Yang et al. 2005;Sinha 2007), component-level optimization (Marklund and Nilsson 2001;Deb et al. 2004;Zou and Mahadevan 2006) and full vehicle crash simulation and optimization (Eskandarian et al. 1997;Sobieszczanski-Sobieski et al. 2001;Kodiyalam et al. 2004;Youn et al. 2004;Duddeck 2008). ...
Article
Design optimization is presented for the crashworthiness improvement of an automotive body structure. The optimization objective was to improve automotive crashworthiness conditions according to the defined criterion (occupant chest deceleration) during a full frontal impact. The controllable factors used in this study consisted of six internal parts of the vehicle’s frontal structure in a condition that their thickness was the “design parameter”. First using the Taguchi method, this study analyzed the optimum conditions in discontinuous design area and impact factors and their optimal levels of design objectives were obtained by analyzing the experimental results. Next to model a precise understanding of the explicit mathematical input–output relationship, fuzzy logic is utilized which make use of full factorial design set of experimental test cases resulted from Taguchi predicting formulations. Interestingly, the optimum conditions for automotive crashworthiness occurred with 2.72% improvement in the defined crashworthiness criterion in comparison with the baseline design while selected structural parts experienced mass reduction by 8.23%.
... Currently, the natural frequency of a body is transformed through changing wall thickness and cross-sectional shape of the structure [4,5], causing the wall resonance to thereby reduce the noise. Jian [6] has applied the reinforcing rib and changed wall thickness to alter the stiffness of a SUV body panels, declining the interior noise. ...
Article
Full-text available
Interior noises of vehicles would be caused when the vibration of body panels excites the indoor air. In the paper, the vibration load of engine was obtained firstly through experiments. Secondly, the engine load was applied in the finite element model of body in white to compute the vibration velocity and realize virtual reality, indicating that the front support of the body had large vibration velocity when the frequency was lower than 60 Hz. The boundary element was then adopted to compute the interior noise and extract the sound pressure at a point near the driver’s head. Two obvious peaks were shown in sound pressure curves, at 270 and 310 Hz, respectively. The body panels that had obvious impact on the interior peak noise were determined through the panel contribution analysis, and the interior peak noise was remarkably reduced after applying sound absorption materials on these panels. Nevertheless, many more additional sound absorption materials were not always better. If a multilayer of sound absorption materials was needed, an optimal value was existed in the thickness of sound absorption material. And a great impact would be played toward the interior noise of the cabin by the reasonable selection of different sound absorption materials and their thicknesses. Finally, the neutral network (NN) was also used to predict interior noises, which was compared with the result of the boundary element. The maximum difference between the prediction values of NN and boundary element was within 5 dB, indicating that the neural network was feasible to predict the interior noise. Subsequently, the neural network method would be applied to conduct the optimization analysis for the interior noise.
... Martins and Lambe [36] surveyed various methods encountered within the field of multidisciplinary design optimization, a field of research that studies the application of numerical optimization techniques to the design of engineering systems. Multidisciplinary design optimization is commonly used to address engineering design problems (see, e.g., [37][38][39][40][41]); some such problems use traditional FEA and are hence forced to script mesh generation procedures and to manage separate geometry descriptions. ...
Article
Isogeometric analysis (IGA) fundamentally seeks to bridge the gap between engineering design and high-fidelity computational analysis by using spline functions as finite element bases. However, additional computational design paradigms must be taken into consideration to ensure that designers can take full advantage of IGA, especially within the context of design optimization. In this work, we propose a novel approach that employs IGA methodologies while still rigorously abiding by the paradigms of advanced design parameterization, analysis model validity, and interactivity. The entire design lifecycle utilizes a consistent geometry description and is contained within a single platform. Because of this unified workflow, iterative design optimization can be naturally integrated. The proposed methodology is demonstrated through an IGA-based parametric design optimization framework implemented using the Grasshopper algorithmic modeling interface for Rhinoceros 3D. The framework is capable of performing IGA-based design optimization of realistic engineering structures that are practically constructed through the use of complex geometric operations. We demonstrate the framework’s effectiveness on both an internally pressurized tube and a wind turbine blade, highlighting its applicability across a spectrum of design complexity. In addition to inherently featuring the advantageous characteristics of IGA, the seamless nature of the workflow instantiated in this framework diminishes the obstacles traditionally encountered when performing finite-element-analysis-based design optimization.
... These problems could become very complicated given their multi-disciplinary nature. MDO has been used in different engineering systems, for instance, in underwater autonomous vehicles [20], spacecraft launch vehicles [21] and different aerospace design optimization [22], vehicle suspension and dynamics [23]- [25], automotive design [26], [27], linkage mechanism design [28], and rocket design [29]. ...
Thesis
Full-text available
Vehicles are major sources of air pollution and greenhouse gas emissions, so any improvement in their fuel efficiency can have a significant impact on the environment and economy. In this study, a regenerative auxiliary power system (RAPS) is proposed to prevent engine idling in service vehicles. Besides preventing idling, the proposed RAPS reduces vehicle total fuel consumption by utilizing regenerative braking and an optimal charging strategy. This system employs waste energy during braking and provides the demanded auxiliary power for a service vehicle to prevent idling. In addition, the system can be retrofit onto existing vehicles. In addition, necessary tools, algorithms and methods to arrive at an optimum RAPS for anti-idling of service vehicles are designed, developed, and implemented. A generic, modular, and flexible vehicle model is created using scalable powertrain components. This model is used by the optimizer to simulate the energy efficiency of the vehicle system in order to minimize the total cost of the system during its expected life cycle. Multi-disciplinary design optimization is applied to optimize the system’s component sizes and power management control logic with respect to a cost-based objective function. Two different optimization methods, Genetic Algorithms (GA) and Simulating Annealing (SA), are utilized to find the optimal solution. Different limitations and constraints of utilizing the Electrical Storage Systems (ESS) are considered in the optimization for more accurate results. Expected changes in power consumption and fuel efficiency in the service vehicle equipped with RAPS is presented as a case study. It is shown that utilizing the RAPS has a significant impact on the total fuel consumption of the vehicle. Based on the results from case study, a prototype model of RAPS (containing generator, battery, auxiliary load, and control system) is developed for laboratory evaluation. Using the RAPS prototype as a hardware-in-the-loop (HIL) of a service vehicle, the proposed system performance is evaluated and the model is validated. It is shown that there is a close match between the experimental and simulation results. The results show that the RAPS is capable to eliminate the idling in service vehicles with considerable fuel saving.
... The main challenge in solving this optimization problem is related to the very large computational costs of performing FE simulations, and the use of surrogate models is a common and practical approach. Several researchers performed structural optimization of the vehicles or their components by using surrogate models (e.g., [24][25][26][27][28][29][30]). In particular, some researchers focused on energy absorption capabilities of the thinwalled tubes, and used surrogate models to perform crashworthiness optimization of these tubes [10,[31][32][33][34][35][36]. ...
Article
In this paper, the effects of introducing lateral circular cutouts on crash performances of tapered thin-walled tubes are explored within a simulation-driven surrogate-based multi-objective optimization framework. The crash performances of the tubes are measured using the crush force efficiency (CFE) and the specific energy absorption (SEA) criteria, which are computed using the finite element analysis code LS-DYNA. Surrogate-based optimization approach is followed to find out that optimum values of the wall thickness, the taper angle, the cutout diameter and the numbers of cutouts in horizontal and vertical directions to maximize CFE and SEA. Four different surrogate models are used: polynomial response surfaces, radial basis functions, and Kriging models with zeroth- and first-order trend models. It is found that the optimum CFE of the tubes with lateral circular cutouts is 27.4% larger than the optimum CFE of the tubes without cutouts. It is also found that the optimum SEA of the tubes with lateral circular cutouts is 26.4% larger than the optimum SEA of the tubes without cutouts. It is observed that the optimum SEA design has slightly reduced wall thickness, significantly reduced taper angle, significantly increased cutout diameter, increased number of cutouts in horizontal direction and slightly reduced number of cutouts in vertical direction compared to the optimum CFE design. In addition, multi-objective optimization of the tubes is performed by maximizing a composite objective function that provides a compromise between CFE and SEA. It is found that the CFE dominates the behavior of composite objective function.
... Fang et al. (2005) also used RBF to achieve crashworthiness optimization using a vehicle model. Kodiyalam et al. (2004) studied multidisciplinary design of vehicles based on approximation models by the Kriging method. ...
Article
The energy absorption characteristics of diamond core sandwich cylindrical columns under axial crushing process depend greatly on the amount of material which participates in the plastic deformation. Both the single-objective and multi-objective optimizations are performed for columns under axial crushing load with core thickness and helix pitch of the honeycomb core as design variables. Models are optimized by multi-objective particle swarm optimization (MOPSO) algorithm to achieve maximum specific energy absorption (SEA) capacity and minimum peak crushing force (PCF). Results show that optimization improves the energy absorption characteristics with constrained and unconstrained peak crashing load. Also, it is concluded that the aluminum tube has a better energy absorption capability rather than steel tube at a certain peak crushing force. The results justify that the interaction effects between the honeycomb and column walls greatly improve the energy absorption efficiency. A ranking technique for order preference (TOPSIS) is then used to sort the non-dominated solutions by the preference of decision makers. That is, a multi-criteria decision which consists of MOPSO and TOPSIS is presented to find out a compromise solution for decision makers. Furthermore, local and global sensitivity analyses are performed to assess the effect of design variable values on the SEA and PCF functions in design domain. Based on the sensitivity analysis results, it is concluded that for both models, the helix pitch of the honeycomb core has greater effect on the sensitivity of SEA, while, the core thickness has greater effect on the sensitivity of PCF.
... Although the algorithms mentioned earlier are effective in obtaining the optimal solution, comprehensive design is difficult to achieve by single discipline. Therefore, multidisciplinary design optimization (MDO) theory is necessary to achieve the comprehensive design of crankshaft [14,15]. ...
Article
Full-text available
The feasibility design method with multidisciplinary and multiobjective optimization is applied in the research of lightweight design and NVH performances of crankshaft in high-power marine reciprocating compressor. Opt-LHD is explored to obtain the experimental scheme and perform data sampling. The elliptical basis function neural network (EBFNN) model considering modal frequency, static strength, torsional vibration angular displacement, and lightweight design of crankshaft is built. Deterministic optimization and reliability optimization for lightweight design of crankshaft are operated separately. Multi-island genetic algorithm (MIGA) combined with multidisciplinary cooptimization method is used to carry out the multiobjective optimization of crankshaft structure. Pareto optimal set is obtained. Optimization results demonstrate that the reliability optimization which considers the uncertainties of production process can ensure product stability compared with deterministic optimization. The coupling and decoupling of structure mechanical properties, NVH, and lightweight design are considered during the multiobjective optimization of crankshaft structure. Designers can choose the optimization results according to their demands, which means the production development cycle and the costs can be significantly reduced.
... The previous shape optimizations for TWD were conducted without considering the influence of flow and heat transfer of the cooling air, i.e., the optimization process is not based the concept of the multi-disciplinary design of optimization (MDO). As a typical multidisciplinary problem, the design process of the HPT disk is attracted by the Multidisciplinary Design and Optimization (MDO) methods (Sobieszczanski-Sobieski and Haftka 1997), which has obtained success in many industrial product designs (Barsi et al. 2015;Berci et al. 2014;Chuang et al. 2008;Coelho et al. 2008;Huang et al. 2014;Kodiyalam et al. 2004;Liem et al. 2014;Zhu et al. 2014). ...
Article
Full-text available
With higher operation temperature required by the advanced aero-turbine, the conventional single web turbine disk (SWD) has reached its limits. At this circumstance, a twin-web disk (TWD) has been proposed as a breakthrough by J Eng Gas Turbines Power Trans ASME 124:298–302, (2002) for its improvements in heat transfer, structural strength and weight loss. However, this novel structure needs new cooling process which brings problems with pressure loss. Fins are designed in this paper in order to increase the outlet pressure and enhance the heat transfer, at the same time demonstrated by the computational fluid dynamics (CFD) analysis. Then, the multidisciplinary design of optimization (MDO) has been performed to find a proper shape and layout of the fins with the minimum stress and maximum outlet pressure. A Kriging surrogate model is also used to accelerate the optimization pace. Because it is a typical multi-objective optimization problem (MOP), the Pareto Front set is obtained in this paper. The results show that the TWD with fins exhibits a better performance in heat transfer and outlet pressure than the one without fins. This structure would be a future trend in TWD design.
... In order to design a vehicle body with high strength and high stiffness, multi-disciplinary contraints should be considered; for instance, static stiffness (bending / torsional), durability performance, noise and vibration, crashworthiness and weight reduction, as seen in Figure 1. However, since the problem that analysis has to be done repetitively until the differences of the conflicting design schemes among each component are minimized, it is hard to deduct optimal solution which satisfies requirements by using existing optimal design technology [4,5]. Therefore, in the current research, the development of the BIW and its components is greatly facilitated by the use of computational engineering analyses and simulations. ...
Chapter
In order to design a vehicle body with high strength and high stiffness, a multi-disciplinary design process should include careful consideration of multi-disciplinary design constraints to properly account for vehicle static stiffness (bending/torsional), durability, Noise/Vibration/Harshness (NVH), crash worthiness, light weight vehicle structure during the early stage of vehicle design process. With this approach, fast development of new vehicle body structures can be achieved with minimal number of iterations to match the conflicting design goals from each discipline. In the current research, a multi-disciplinary design optimization (MDO) based on a meta model is developed and refined to apply for the design of body structure. In an effort to apply the MDO for vehicle body structure, 4 phase procedures were established in the current research. In Phase I, a base model is created. In Phase II, an effect analysis is carried out. In Phase III, a meta model is created. Finally in Phase IV, using the optimization algorithm, the meta model created in Phase III is eventually refined through the process of optimization. In this research, static stiffness (bending / torsional), dynamic stiffness (1st torsion mode) were used for constrained conditions and the mass minimization was the object function for optimization.
... other aspects of aerospace design [33,53,79,30], automotive design [34,39,38,26,57], and civil engineering [3,18,9,17]. ...
Article
During early stages of design of large scale engineering systems, such as ships, competing objectives of multiple discipline analyzers must be taken into account. Currently, tools such as Multidisciplinary Design Optimization (MDO) provide an automated method for analyzing trade-offs between competing discipline design objectives. While automated design methods are useful in the design world, automation removes the human ability to interpret results of analysis. Current design tools lack the ability to incorporate the intent of experienced designers in the communication between disciplines, so MDO results are dependent on the fidelity of the analysis models being used rather than human input. When evaluating designs, a team of experienced engineers communicates information about the design space to each other with respect to their own disciple's objectives and design constraints. This thesis presents a new method of MDO which provides a framework to emulate the intent of an human designer in a MDO optimizer. The method uses the ideas of managing trade-offs and modeling discipline interactions from MDO. Typical MDO algorithms utilize a system level optimizer, which receives information from discipline optimizers and moves toward a globally optimal solution. Generally, the information communicated to the system level optimizer is point data about the individual discipline's optimal point, which is then interpreted as the potential improvement from the current system design state. The proposed algorithm would communicate the discipline's preferred areas of the design space to be combined with the other discipline's preferences in an attempt to find a globally optimal solution. This design intent modeling is introduced through fuzzy logic systems and fuzzy logic controllers. Fuzzy logic systems provide a method to interpret design analysis tools and glean more information to use in the decision making process. Fuzzy logic controllers are often used to emulate human decision models in the control of physical systems. These methods are extended to evaluation of designs to combine data from multiple disciplines to find the optimal system design while considering trade-offs between multiple disciplines. This dissertation will present the development of a novel method of MDO which incorporates fuzzy logic systems and controllers in the MDO optimizer.
... The design of the HPT disk is a typical multidisciplinary problem due to its thermo-fluid, corrosion and high speed rotation working environment. The designers have been attracted by the Multidisciplinary Design and Optimization (MDO) methods (Sobieszczanski-Sobieski and Haftka 1997) and obtained success in designing some industrial products (Huang et al. 2014;Berci et al. 2014;Liem et al. 2014;Chuang et al. 2008;Kodiyalam et al. 2004;Coelho et al. 2008). Nagendra et al. (2005) proposed a technique developed by RAPIDDISK which provides an optimal preliminary shape and design using MDO method. ...
Article
Full-text available
The newly designed structures are employed to improve the high pressure turbine (HPT) disk to the expected performance. The twin-web disk (TWD) has been proven to be the future trend of the HPT disk due to its breakthrough in weight loss, strength and heat transfer efficiency compared to the conventional single web disk (SWD). Because of the multi-physics working conditions and intense coupling of multiple disciplines, the conventional design of the HPT disk is a labor intensive work. A series of design procedures, including asymmetrical computational fluid dynamics (CFD) analysis, inverse distance weighted (IDW) interpolation method, multidisciplinary feasible method (MDF), and design of experiments (DOE), are proposed to obtain the proper design for both TWD and SWD in an efficient way. In the present work, the multidisciplinary design of optimization (MDO) has been performed to find the proper shape of the TWD disk with the minimum mass. The results showed that the TWD exhibits a better performance in heat transfer and weight loss than SWD. The modeling and optimization procedure of this work can be referred for engineering design.
Article
Full-text available
Supercomputing power is one of the fundamental pillars of the digital society, which depends on the accurate scheduling of parallel applications in High-Performance Computing (HPC) centers to minimize computing times. However, precedence-constraint task scheduling is a well-known NP-Hard optimization problem, and no optimal polynomial-time algorithm exists to solve it. Therefore, as accurate as possible to the optimal values, heuristic algorithms are of relevant interest. Our new scheduling proposal name is EFT-GVNS, which stands for Earliest Finish Time - General Variable Neighborhood Search. EFT-GVNS uses a Composite Local Search (CLS), making our proposal more efficient than traditional GVNS. EFT-GVNS accuracy against four high-performance algorithms from the state-of-the-art (EDA, EFT-ILS, GRASP-CPA, MPQGA) and one reference algorithm in the literature (HEFT) is studied. Experimental results over four real-world applications (Fpppp, LIGO, Robot, Sparse) and 14 synthetic instances from the literature} show that EFT-GVNS outperforms {in terms of the median achieved results, with a global improvement of 37.6%, 27.4%, 17.8%, 6.1%, 2.2% to HEFT, EDA, EFT-ILS, GRASP-CPA, and MPQGA, respectively. EFT-GVNS achieves all the 14 optimal values of the synthetic benchmark.
Article
In autonomous excavation, design and optimization of the unmanned cable shovels (UCS) are important issues in the full life cycle of the equipments. However, the design of physical structure and control system of the UCS are preformed at different stage, which makes it difficult for traditional sequential optimization strategy to generate global optimal solution. To enhance the working performance of the UCS, in this paper, a multi-stage multi-objective (MSMO) Co-design optimization strategy is proposed to perform global optimization considering excavation and loading processes by simultaneous optimization for the structure and control parameters of UCS. Under this framework, first, a point-to-point motion model based on 4-5-4 piecewise polynomial is proposed to describe the motion trajectory and the dynamical model of the working device is established to predict the energy consumption in working process. Then, the physical and geometric constraints in actual working are analyzed and a multi-objective optimization model considering excavation and loading processes is established to improve mining efficiency and reduce energy consumption in unmanned excavation scenarios. Finally, the structural and control parameters are optimized synchronously to generate optimal physical structure, excavation and loading trajectory. Numerical results show that the proposed MSMO Co-design method can further improve the operational performance of UCS compared with the traditional optimization strategy.
Conference Paper
View Video Presentation: https://doi.org/10.2514/6.2021-3041.vid Large-scale multidisciplinary design optimization problems often involve thousands of design variables and tens of thousands of state variables. If formulated using a simultaneous analysis and design architecture, these optimization problems would have tens of thousands of optimization variables and tens of thousands of equality constraints. Such problems can be particularly difficult to solve even with a gradient-based approach. This paper presents an adaptive, inexact quasi-Newton algorithm for solving large-scale multidisciplinary design optimization problems using a simultaneous analysis and design architecture. This algorithm is novel in two ways. First, this algorithm uses new inexactness criteria that are easy to compute and can be used with any Krylov solver with or without a preconditioner. Second, this algorithm adaptively chooses the stopping criteria for the Krylov solver that makes the best use of its performance. This algorithm is applied to an equality-constrained cantilever bar problem with up to 3,000 design variables. We observe that this algorithm is robust and that it yields a reduction greater than 50\% in terms of the total Krylov iterations, compared with the exact quasi-Newton method.
Article
A challenge in the design and optimisation of the vehicle front structures is the high computational costs required for crash analysis. A methodological approach to simplifying finite element vehicle models and crash barriers is presented in this paper. The methodology uses global deformation characteristics of structures which are obtained from the global crash model. For the simplification of the vehicle crash model, structural regions which sustain only elastic deformations during the frontal crash are replaced by kinematic numerical representations which describe both stiffness and load paths at the interface of the substituted structures. Verification studies of the simplified vehicle model show a very good agreement of the global and local structural response during the frontal crash. Further simplifications were applied to the offset deformable barrier by replacing its detailed crushing behaviour by kinematic descriptions. Through the combined use of both simplified numerical representations, the computational cost of a Euro new car assessment program (NCAP) offset crash analysis can be reduced by around 92%. With the obtained time reduction, structural optimisation studies of the remaining structure can be conducted efficiently for the identification of weight reduction potentials.
Article
For the coach industry, rapid modeling and efficient optimization methods are desirable for structure modeling and optimization based on simplified structures, especially for use early in the concept phase and with capabilities of accurately expressing the mechanical properties of structure and with flexible section forms. However, the present dimension-based methods cannot easily meet these requirements. To achieve these goals, the property-based modeling (PBM) beam modeling method is studied based on the PBM theory and in conjunction with the characteristics of coach structure of taking beam as the main component. For a beam component of concrete length, its mechanical characteristics are primarily affected by the section properties. Four section parameters are adopted to describe the mechanical properties of a beam, including the section area, the principal moments of inertia about the two principal axles, and the torsion constant of the section. Based on the equivalent stiffness strategy, expressions for the above section parameters are derived, and the PBM beam element is implemented in HyperMesh software. A case is realized using this method, in which the structure of a passenger coach is simplified. The model precision is validated by comparing the basic performance of the total structure with that of the original structure, including the bending and torsion stiffness and the first-order bending and torsional modal frequencies. Sensitivity analysis is conducted to choose design variables. The optimal Latin hypercube experiment design is adopted to sample the test points, and polynomial response surfaces are used to fit these points. To improve the bending and torsion stiffness and the first-order torsional frequency and taking the allowable maximum stresses of the braking and left turning conditions as constraints, the multi-objective optimization of the structure is conducted using the NSGA-II genetic algorithm on the ISIGHT platform. The result of the Pareto solution set is acquired, and the selection strategy of the final solution is discussed. The case study demonstrates that the mechanical performances of the structure can be well-modeled and simulated by PBM beam. Because of the merits of fewer parameters and convenience of use, this method is suitable to be applied in the concept stage. Another merit is that the optimization results are the requirements for the mechanical performance of the beam section instead of those of the shape and dimensions, bringing flexibility to the succeeding design. © Chinese Mechanical Engineering Society and Springer-Verlag Berlin Heidelberg 2016.
Chapter
The evolution of aerospace systems has progressed rapidly and the type of problems that engineers are faced with is becoming more and more complex each year. The traditional phases of conceptual preliminary and detailed design have now been complemented with the need to integrate more than one discipline optimisation method as well as uncertainties in all phases of the design process. This chapter provides an overview of multi-disciplinary and robust design optimization tools.KeywordsRobust DesignLift CoefficientMultidisciplinary Design OptimisationComputational ExpenseAircraft DesignThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Article
A finite element model for car door is created with Hypermesh software and a sensitivity analysis is conducted by using Optistruct solver. Taking thicknesses of key panels as design variables, 100 sample points are generated and calculated by using Latin hypercube design of experiment, and Kriging modeling technique is applied to the construction of the surrogate models for mass, torsional stiffness and first-order natural frequency based on calculated results. An optimization is then performed by using sequential quadratic programming with minimizing mass as objective, torsional stiffness and the first-order natural frequency as constraints, and a set of optimum combination of panel thicknesses is finally obtained, making the mass of car door reduces by 2.38% while meeting the requirements on torsional stiffness and modal performance. The combination of sensitivity analysis and Kriging model ensures the efficiency and accuracy of optimization.
Article
In order to reduce abnormal vibration and noise of a domestic light bus, its finite element simulation and order tracking analysis of road test were performed to identify vibration sources, it was found that the rotation angular frequencies of the wheels and the first two natural frequencies of the body structure are overlapped, resonances occur, they lead to increased vibrations. To stagger the first two natural frequencies of the body and excitation frequencies, the thicknesses of sheet metal and skeleton of the body-in-white were chosen as the design variables, to raise the first two natural frequencies of the body-in-white was taken as the optimization objective, the optimization design and sensitivity analysis of the body-in-white was conducted. According to the modal sensitivity and mass sensitivity of sheet metal and skeleton, the optimal scheme was designed and test analysis was performed. Comparing the test results before and after optimization, the effectiveness and rationality of the optimization were verified.
Article
A finite element model for the cab of a heavy-duty commercial vehicle is established, its computational modal analysis is conducted, and a topology optimization is performed on cab structure. Three improving schemes are put forward on the basis of optimization results, from which the improving scheme of adding reinforced plates is chosen through comparison. Then a size optimization is carried out on the thickness of new reinforced plates. The final results show that after structural optimization, the first modal frequency of the cab is raised from 17.952 Hz to 22.200 Hz with its bending stiffness and torsional stiffness also improved.
Article
This research investigated the safety performance of a passenger car in side collisions, using a side air-curtain protective device to minimize the risk of head-brain injuries. Traumatic brain injury in passenger vehicle side collisions is one of the common injury patterns with fatal consequences in traffic accidents. A dummy head has limited capability to evaluate vehicle safety performance with detailed injury related physical parameters. Continuous development of mathematical models has provided efficient tools for assessing the safety performance of a protective device. A validated head-brain FE model and air-curtain model was employed to simulate the impact responses of head-brain to B-pillar and further quantify the effect of the air-curtain on the protection of the human head in side collisions. A parametric study was conducted using the design of experiment approach. The kinematics of a head impacting the B-pillar and the air-curtain are presented. The calculated physical parameters from the head-brain FE model include the distribution of von Mises stress, shear stress, coup pressure, contrecoup pressure, and accelerations during vehicle side collision. Study outcomes suggested that using a side air-curtain protective device can reduce von Mises stress, shear stress, coup pressure, contrecoup pressure, and deceleration rate during vehicle side impact. The efficiency of the protective device for head-brain injuries is identified by reducing the HIC (Head Injury Criterion) to 80% of original values.
Article
Automotive companies continuously strive to design better products faster and more cheaply using simulation models to evaluate every possible aspect of the product. Multidisciplinary design optimization (MDO) can be used to find the best possible design taking into account several disciplines simultaneously, but it is not yet fully integrated within automotive product development. The challenge is to find methods that fit company organizations and that can be effectively integrated into the product development process. Based on the characteristics of typical automotive structural MDO problems, a metamodel-based MDO process intended for large-scale applications with computationally expensive simulation models is presented and demonstrated in an example. The process is flexible and can easily fit into existing organizations and product development processes where different groups work in parallel. The method is proven to be efficient for the discussed example and improved designs can also be obtained for more complex industrial cases with comparable characteristics.
Article
The vehicle interior noise reduction was focused. The acoustic-structure property was analyzed based on white body, acoustic and acoustic-structure FEM models. Those body panels contributing most to interior noise were determined according to acoustic contribution analysis. To reduce the vibration and noise radiation, an optimization topology model was developed and Evolutionary Structural Optimization (ESO) method was introduced to obtain the optimal topology configuration of damping material. The results show that the optimal topology configuration can highly improve the efficiency of damping material. The noise reduction measure which requires 100% damping material coverage in the original design can be achieved by the use of 50% damping material coverage. The optimization design for damping structure supplies theoretical support to the vehicle interior noise reduction. ©, 2015, Chinese Vibration Engineering Society. All right reserved.
Article
Based the quadratic subdivision mathematical model of the braking severity, a design method of the predictive model for parallel cooperative braking system is investigated systematically. An off-line optimization stream is designed. First, the continuous design space, which is constituted of the vehicle speed v, battery State-of-Charge (SoC) and braking severity z, is sampled by Design of Experiments(DOE); two subsystems of regenerative braking energy and braking stability are defined and optimized by collaborative optimization method(CO) based on the sampling points; finally, the predictive model is established for the cooperative braking system. Simulation results show that the proposed model can realize real-time optimum control and maximize the regenerative braking energy recovery efficiency under the condition of ensuring the braking stability. Accordingly, it has high engineering application value. ©, 2015, Editorial Board of Jilin University. All right reserved.
Conference Paper
Full-text available
Despite the steady and continuing growth of computing power and speed, the complexity and computational expense of engineering analysis codes maintains pace. Statistical techniques are becoming widely used in engineering design to construct approximations or metamodels of these analysis codes which are then used in lieu of the actual codes, facilitating optimization and concept exploration. Our purpose in this paper is to report results of ongoing research aimed at increasing the efficiency of computer-based engineering design through the use of spatial correlation metamodels to build global approximations of computationally expensive computer analyses. Three structural design examples are presented to test the predictive capability of these metamodels for use in design optimization. The reported results confirm that these spatial correlation metamodels can produce sufficient accuracy for optimization when used as global approximations.
Conference Paper
A feasibility study of using numerical optimization methods for crash analysis is presented. Software required for numerical optimization includes parametric modeling (Pro/ENGINEER), automatic mesh generation (PDA/PATRAN3), nonlinear finite element analysis (RADIOSS), and optimization methods (CONMIN & DOP). Both single- and multiple-objective formulations are used for numerical optimization and result in better designs. It was found that crash optimization is feasible but costly and that finite element mesh quality is essential for successful crash analysis and optimization. A simplified front horn was used to demonstrate this approach.
Conference Paper
The present paper extends an efficient mathematical programming optimization procedure, and includes wing twist and panel buckling constraints with earlier developed material strength, displacement, and minimum-gage constraints. In this procedure, the panel buckling constraint for laminated composite panels is included on the global design level for the first time. The design procedure is used to provide several examples of aluminum and composite multispar high aspect ratio wing designs subject to a number of diverse design constraints. The effects of various constraints on the minimum-mass designs of these wings are evaluated and the advantages of the composite designs over the aluminum designs are assessed.
Article
1. Abstract: A vehicle is an engineering system whose successful design requires harmonization of a number of objectives and constraints that, in principle, can be modeled as a constrained optimization in the space of design variables. However, dimensionality of such optimization and the complexity and expense of the underlying analysis suggest a decomposition approach to enable concurrent execution of smaller and more manageable tasks. In order to preserve the couplings that naturally occur among the elements of the whole problem, such optimization by various types of decomposition must include a degree of coordination at the system level. Multidisciplinary Design Optimization (MDO) is a body of methods and techniques for performing the above optimization so as to balance the design considerations at the system and detail levels. The paper is an overview of a few MDO methods selected for their applicability to vehicle systems.
Article
A car body structure was optimized for minimum weight under the constraints of noise, vibration, and harshness (NVH), and a crash event, using up to 254 concurrently operating processors. The crash analysis alone, if executed on a single processor and repeated the number of times this optimization required, would have taken 257 days of elapsed computing time. Parallel processing has compressed the elapsed time to one day demonstrating how a multiprocessor machine may be useful in solving engineering tasks that heretofore were regarded as intractable. The optimization procedure transformed the structure initially infeasible to one having its weight reduced and all the constraints satisfied. The experience gained in the reported application indicated it is important to tailor the solution method to the characteristics of the multiprocessor computer architecture and to understand the data handling options offered by that architecture. Another conclusion drawn from this case is that the coarse-grained parallelism whereby an existing code is being replicated over an array of processors should be regarded as an effective way of utilization of multiprocessor machines, immediately available in the interim before solutions are redeveloped from ground up specifically for that class of machines.
Article
An unstiffened panel buckling constraint for balanced, symmetric laminated composites is included on the global design level in a mathematical programming structural optimization procedure for designing wing structures. Constraints are introduced by penalty functions, and Newton's method based on approximate second derivatives of the penalty terms is used as the search algorithm to obtain minimum-mass designs. Constraint approximations used during the optimization process contribute to the computational efficiency of the procedure. A criterion is developed that identifies the appropriate conservative form of the constraint approximations that are used with the optimization procedure. Minimum-mass design results are obtained for a multispar high-aspect-ratio wing subjected to material strength, minimum-gage, displacement, panel buckling and twist constraints. The material systems considered for the examples are all graphite-epoxy, graphite-epoxy with boron-epoxy spar caps, and all aluminum. The composite material designs are shown to have an advantage over the aluminum designs since they can often satisfy additional constraints with only small mass increases.
Article
A definition of the multidisciplinary design optimization (MDO) is introduced, and functionality and relationship of the MDO conceptual components are examined. The latter include design-oriented analysis, approximation concepts, mathematical system modeling, design space search, an optimization procedure, and a humane interface.
Article
Two methods of creating approximation models are compared through the calculation of the modeling accuracy on test problems involving one, five, and ten independent variables. Here, the test problems are representative of the modeling challenges typically encountered in realistic engineering optimization problems. The first approximation model is a quadratic polynomial created using the method of least squares. This type of polynomial model has seen considerable use in recent engineering optimization studies due to its computational simplicity and ease of use. However, quadratic polynomial models may be of limited accuracy when the response data to be modeled have multiple local extrema. The second approximation model employs an interpolation scheme known as kriging developed in the fields of spatial statistics and geostatistics. This class of interpolating model has the flexibility to model response data with multiple local extrema. However, this flexibility is obtained at an increase...
Structural optimization in occupant safety and crash analysis Proceedings of Conference Opticon’99: Optimization Software
  • U Schramm
  • D Schneider
  • H Thomas
Crashworthiness technology using response surface methodology and massively parallel programming, Poster paper at Optimization in Industry II
  • N Stander
Stander, N. 1979: Crashworthiness technology using response surface methodology and massively parallel programming, Poster paper at Optimization in Industry II, Banff, Canada Starnes, J.H.;
Preliminary design of com-posite wings for buckling, strength and displacement con-straints Feasibility study of crash optimization
  • R T L Haftka
  • J Cheng
Haftka, R.T. 1979: Preliminary design of com-posite wings for buckling, strength and displacement con-straints. J. Aircraft 16, 564–570 TNO Automotive 2001: MADYMO Users Manual Yang, R.J.; Tseng, L.; Nagy, L.; Cheng, J. 1994: Feasibility study of crash optimization (ASME) 69(2), 549–556
Computer aided engineering for improved vehicle crashworthiness, Poster paper at Optimization in Industry II
  • M Chargin
  • H Miura
Chargin, M.; Miura, H. 1999: Computer aided engineering for improved vehicle crashworthiness, Poster paper at Optimiza-tion in Industry II, Banff, Canada, 169–196
Optimization of car body for noise, vibration and harshness (NVH) and crash
  • R J Yang
Yang, R.J. 2000: Optimization of car body for noise, vibration and harshness (NVH) and crash. Proceedings of the AIAA/ASME/ASCE/ AHS/ASC 41st Structures, Structural Dynamics and Materi-als Conference, AIAA, Atlanta, AIAA Paper Number: AIAA-2000-1521
Optimization of car body for noise, vibration and harshness (NVH) and crash
  • J Sobieszczanski-Sobieski
  • S Kodiyalam
  • R J Yang