No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Overview of computational intelligence for building energy system design
... Published review papers specifically focus on evaluating an individual area within the wide range of built environment problems solved using AI approaches. This includes building energy consumption forecasting [21][22][23], [30][31][32][33][34], integration with building energy management systems (BEMS) [23][24][25], building design optimisation [26] and occupancy detection [26][27][28][29]. Bordeau et al. [21] reviewed data-driven and machine learning techniques for modelling and forecasting buildings' energy consumption. ...
... The studies [23][24][25] reviewed AI applications based on their design and integration with BEMS. The review covered different AI frameworks and workflows used for HVAC design and optimisation process and control. ...
The built environment sector is responsible for almost one-third of the world's final energy consumption. Hence, seeking plausible solutions to minimise building energy demands and mitigate adverse environmental impacts is necessary. Artificial intelligence (AI) techniques such as machine and deep learning have been increasingly and successfully applied to develop solutions for the built environment. This review provided a critical summary of the existing literature on the machine and deep learning methods for the built environment over the past decade, with special reference to holistic approaches. Different AI-based techniques employed to resolve interconnected problems related to HVAC systems and enhance building performances were reviewed, including energy forecasting and management, indoor air quality and occupancy comfort/satisfaction prediction, occupancy detection and recognition, and fault detection and diagnosis. The present study explored existing AI-based techniques focusing on the framework, methodology, and performance. The literature highlighted that selecting the most suitable machine learning and deep learning model for solving a problem could be challenging. The recent explosive growth experienced by the research area has led to hundreds of machine learning algorithms being applied to building performance-related studies. The literature showed that existing research studies considered a wide range of scope/scales (from an HVAC component to urban areas) and time scales (minute to year). This makes it difficult to find an optimal algorithm for a specific task or case. The studies also employed a wide range of evaluation metrics, adding to the challenge. Further developments and more specific guidelines are required for the built environment field to encourage best practices in evaluating and selecting models. The literature also showed that while machine and deep learning had been successfully applied in building energy efficiency research, most of the studies are still at the experimental or testing stage, and there are limited studies which implemented machine learning strategies in actual buildings and conducted the post-occupancy evaluation.
... The white-box modeling methods (known as the physical-based or engineering methods) use physical principles to solve the calculation of thermal and energy behaviors on the whole-building level or for sublevel systems in buildings [24]. A series of mathematical models are built up step-by-step based on elaborate physical functions or thermodynamics of the mass and energy balances, momentum, and flow balance [25]. The common way of making a white-box model for building energy modeling is as follows: building geometry/envelope, internal heat gains (e.g., lights and occupants), sublevel systems (e.g., HVAC and renewable systems), and control and management parameters. ...
Buildings use up to 40% of the global primary energy and 30% of global greenhouse gas emissions, which may significantly impact climate change. Heating, ventilation, and air-conditioning (HVAC) systems are among the most significant contributors to global primary energy consumption and carbon gas emissions. Furthermore, HVAC energy demand is expected to rise in the future. Therefore, advancements in HVAC systems’ performance and design would be critical for mitigating worldwide energy and environmental concerns. To make such advancements, energy modeling and model predictive control (MPC) play an imperative role in designing and operating HVAC systems effectively. Building energy simulations and analysis techniques effectively implement HVAC control schemes in the building system design and operation phases, and thus provide quantitative insights into the behaviors of the HVAC energy flow for architects and engineers. Extensive research and advanced HVAC modeling/control techniques have emerged to provide better solutions in response to the issues. This study reviews building energy modeling techniques and state-of-the-art updates of MPC in HVAC applications based on the most recent research articles (e.g., from MDPI’s and Elsevier’s databases). For the review process, the investigation of relevant keywords and context-based collected data is first carried out to overview their frequency and distribution comprehensively. Then, this review study narrows the topic selection and search scopes to focus on relevant research papers and extract relevant information and outcomes. Finally, a systematic review approach is adopted based on the collected review and research papers to overview the advancements in building system modeling and MPC technologies. This study reveals that advanced building energy modeling is crucial in implementing the MPC-based control and operation design to reduce building energy consumption and cost. This paper presents the details of major modeling techniques, including white-box, grey-box, and black-box modeling approaches. This paper also provides future insights into the advanced HVAC control and operation design for researchers in relevant research and practical fields.
... The authors employed swarm particle algorithm in a decentralized RHVAC system to eradicate existing limitations and obtain equalimproved performance to conventional centralized sensor fault detection and diagnosis systems. Huajing et al. (2019) reviewed applications of RHVAC system optimization designs. Three approaches, namely white box, black box and grey box approaches, were identified. ...
Artificial intelligence (AI) models for refrigeration, heat pumps, and air conditioners have emerged in recent decades. The universal approximation accuracy and prediction performances of various AI structures like feedforward neural networks, radial basis function neural networks, adaptive neuro-fuzzy inference and recurrent neural networks are encouraging interest. This review discusses existing topographies of neural network models for RHVAC system modelling, energy prediction and fault(s), and detection and diagnosis. Studies show that AI structures require standardization and improvement for tuning hyperparameters (like weight, bias, activation functions, number of hidden layers and neurons). The selection of activation functions, validation, and learning algorithms depends on author’s suitability for a particular application. Backpropagation, error trial selection of the number of hidden layer, and hidden layers’ neurons, and Levenberg–Marquardt learning algorithms, remain prevalent methodologies for developing AI structures. The major limitations to the application of AI models in RHVAC systems include exploding or/and vanishing gradients, interpretability, and accuracy trade off, and training saturation and limited sensitivity. This review aims to give up-to-date applications of different AI architectures in RHVAC systems and to identify the associated limitations and prospects.
... Building energy system designs usually need to confront complexity issues. Resolving the complicated details may be expedited by the utilization of advanced computational aids [92]. While examining a residential architecture, an energy efficiency analysis should list and quantify as many implicated effects as possible; it reduces the energy consumption uncertainty and determines its optimal resource usage [93]. ...
Buildings consume a large portion of the global primary energy. They are also key contributors to CO2 emissions. Greener residential buildings are part of the ‘Renovation Wave’ in the European Green Deal. The purpose of this study was to explore the usefulness of energy consumption screening as a part of seeking retrofitting opportunities in the older residential building stock. The objective was to manage the screening of the electromechanical energy systems for an existing apartment unit. The parametrization was drawn upon inspection items in a comprehensive electronic checklist—part of an official software—in order to incur the energy certification status of a residential building. The extensive empirical parametrization intends to discover retrofitting options while offering a glimpse of the influence of the intervention costs on the final screening outcome. A supersaturated trial planner was implemented to drastically reduce the time and volume of the experiments. Matrix data analysis chart-based sectioning and general linear model regression seamlessly integrate into a simple lean-and-agile solver engine that coordinates the polyfactorial profiling of the joint multiple characteristics. The showcased study employed a 14-run 24-factor supersaturated scheme to organize the data collection of the performance of the energy consumption along with the intervention costs. It was found that the effects that influence the energy consumption may be slightly differentiated if intervention costs are also simultaneously considered. The four strong factors that influenced the energy consumption were the automation type for hot water, the types of heating and cooling systems, and the power of the cooling systems. An energy certification category rating of ‘B’ was achieved; thus, the original status (‘C’) was upgraded. The renovation profiling practically reduced the energy consumption by 47%. The concurrent screening of energy consumption and intervention costs detected five influential effects—the automation type for water heating, the automation control category, the heating systems type, the location of the heating system distribution network, and the efficiency of the water heating distribution network. The overall approach was shown to be simpler and even more accurate than other potentially competitive methods. The originality of this work lies in its rareness, worldwide criticality, and impact since it directly deals with the energy modernization of older residential units while promoting greener energy performance.
... In addition, as the purpose of this study focuses on the optimization of passive design solutions in the early design stage of buildings, the different HVAC system types and energy use modes are not considered. However, the performance of HVAC systems and occupants' behavior significantly impact the building's operational building energy consumption [56,57]. Fig. 19 illustrates the interactions between EUI H&C , building energy efficiency improvement and occupant comfort demand. ...
China, with the largest energy consumption system in the world, faces numerous challenges in achieving the government’s commitment to reach a carbon-peak and carbon-neutral target. As the most common public building type in terms of floor area, office buildings have great potential for energy saving and emissions reduction. To meet this target, building designers target passive solutions that can meet the thermal comfort needs of occupants and also reduce energy consumption. This study aims to develop a decision-making method to select optimal solutions from among tens of thousands of design options considering the factors of energy consumption, comfort, and cost. We developed a novel optimization decision approach with the above-mentioned three objectives. The model consists of three stages: 1) the establishment of the reference building model, 2) sensitivity analysis to identify the main influencing variables and 3) the establishment of the optimization and decision-making model by applying NSGA-II and TOPSIS methods. By applying this three-stage decision-making model, this paper first proposes cost-effective passive design solutions for office buildings throughout the Hot Summer and Cold Winter climate zone. Finally, an office building in Shanghai was chosen as a case study to demonstrate the practical implementation of the proposed solutions through a post-occupancy evaluation with a two-year energy auditing and thermal comfort survey. It is evident that the proposed solutions provide support for the new low energy building design guide for office buildings along with necessary revisions to the existing standards for the hot summer and cold winter climate zone in China.
... The ANN can learn and generalize by using the data samples [80]. In [81], the primary ANN method's description and various types such as 'feed-forward and feedback networks' are discussed. In [82], the authors noted that the NN structure, which has two hidden layers, can generate optimal peak, daily, and monthly LF solutions. ...
Load forecasting (LF), particularly short-term load forecasting (STLF), plays a vital role throughout the operation of the conventional power system. The precise modeling and complex analyses of STLF have become more significant in advanced microgrid (MG) applications. Several models are proposed for STLF and tested successfully in the literature. The selection of a forecasting method is mostly based on data availability and its objectives. This article presents a survey of the latest analytical and approximation techniques reported in the literature to model STLF in an MG environment. This article mainly focuses on the review on important methods applied to forecast renewable energy availability, energy demand, price and load demand. Different models, their main objectives, methodology, error percentage, etc., are critically reviewed and analyzed. For quick reference, we have highlighted the important points in the form tables. The researchers can quickly identify and framed their research problem related to the LF area by reading this review paper.
... In particular, the high-level performance of the HVAC system is critical as it serves as the respiratory system of the building. Sha et al. [81] reviewed the computational intelligence (CI) technology application in solving HVAC design optimization problems and proposed an integrated HVAC automation and design optimization framework to realize the information transmission between various design stages, and ultimately improved the efficiency of the HVAC system. Xiao et al. [82] reviewed the automatic commissioning of HVAC systems and considered it as an important part to achieve sustainable buildings. ...
Healthy buildings are a deep-level development of green buildings, which can effectively
help relieve stress and improve occupants’ physical and mental health. In addition, they are is likely to play an important role in preventing the spread of respiratory infectious diseases. Therefore, healthy buildings have attracted worldwide attention. This article reviews the research and development of healthy buildings in China. First, it briefly introduces the definition of healthy buildings, the key elements of evaluation standards, energy conservation measures and new technology applications for healthy buildings, and lessons learned from the global outbreak of SARS-CoV-2. Secondly, it presents the milestones of healthy building development and healthy building projects in China,
and the benefits of healthy buildings were also discussed. Finally, the differences in the evaluation systems of healthy buildings between China and other countries were analyzed, the problems of the current policy system of healthy buildings in China were identified, and suggestions for future development were provided.
... The work flow of this new concept can be divided into four tasks: pretreatment of BIM (Building Information Model), automatic zoning, system selection, and detailed configuration. Some researchers in recent years have emphasized the importance of the task of system selection, which means they want to generate the appropriate design scheme of an HVAC system automatically (as highlighted by the yellow circle in Figure 1) [24]. The current project will support this idea by programming the method for a BIM software plug-in such as Revit. ...
During the design stage of an HVAC (heating, ventilation, and air conditioning) system in a construction project, designers must decide on the most workable design scheme for the plant room in the building based on the evaluation of multiple aspects related to system performance that need to be considered, such as energy efficiency, economic effectiveness, etc. To solve this problem, this paper proposes a comprehensive evaluation method for the plant rooms of centralized air-conditioning systems in commercial buildings. This new method consists of two analyses used in tandem: Building Performance Simulation (BPS) models and a collection of real HVAC design cases (the carried-out design solutions). The BPS models and a knowledge of the reduction approach based on Rough Set (RS) theory are used to generate data and weight factors for the indices of energy efficiency; and the real design cases are employed with a heuristic algorithm to extract the compiled empirical information for other evaluation items of the centralized HVAC system. In addition, this paper also demonstrates an application in an actual case of a building construction project. By comparing the expert decision-making process and the evaluation results, it is found that they are basically consistent, which verifies the reasonability of the comprehensive evaluation method.
... Applications of machine learning algorithms to energyrelated problems show an exponential increase, due to the rapid developments in the field of machine learning and the growing availability of big datasets [22]. Machine learning applications become more and more common in wind energy [31,41,61,91], solar energy [88,89,101], water resources [98], energy in buildings [80,86,112], renewable energy in general [82,102], fossil fuels [70], energy demand [100] and development of materials [24] to name a few disciplines. ...
Machine learning algorithms have been extensively exploited in energy research, due to their flexibility, automation and ability to handle big data. Among the most prominent machine learning algorithms are the boosting ones, which are known to be “garnering wisdom from a council of fools”, thereby transforming weak learners to strong learners. Boosting algorithms are characterized by both high flexibility and high interpretability. The latter property is the result of recent developments by the statistical community. In this work, we provide understanding on the properties of boosting algorithms to facilitate a better exploitation of their strengths in energy research. In this respect, (a) we summarize recent advances on boosting algorithms, (b) we review relevant applications in energy research with those focusing on renewable energy (in particular those focusing on wind energy and solar energy) consisting a significant portion of the total ones, and (c) we describe how boosting algorithms are implemented and how their use is related to their properties. We show that boosting has been underexploited so far, while great advances in the energy field are possible both in terms of explanation and interpretation, and in terms of predictive performance.
... The building energy system of modern buildings provides a healthy and comfortable indoor environment [1]. However, in cities, buildings consume 70% of the primary energy [2]. ...
Accurate prediction of the building load is crucial to ensure the energy saving and improve the operational efficiency of the heating, ventilation, and air conditioning (HVAC) system. In this study, the heating load (HL) and cooling load (CL) of buildings are analyzed using the Spearman method considering eight influencing factors: relative compactness, surface area, wall area, roof area, overall height, orientation, glazing area, and glazing area distribution. The ant colony optimization (ACO) method is used to optimize the ability of a wavelet neural network (WNN) to predict the HL and CL values of residential buildings. The linearly decreasing inertia weight and self-adaptive mutation operator are introduced to improve the optimizing capability of the ACO. An improved ACO-WNN (I-ACO-WNN) model is proposed to achieve a high-precision building load forecasting, and the formulas including the influence factors of the building load, are proposed. The regression coefficient values of the proposed forecasting model of HL and CL are 0.9714 and 0.9783, respectively. Compared to the traditional WNN model, the root mean square error values of HL and CL predictions by the I-ACO-WNN model are decreased by 66.01% and 73.28%, respectively; while the mean absolute error values are decreased by 82.44% and 84.82%, respectively; also, the mean absolute percentage error values are reduced by 81.21% and 85.31%, respectively; lastly, the mean square error values are reduced by 88.44% and 92.86%, respectively. The proposed prediction model can be used as a reliable tool for HL and CL estimation in future intelligent urban planning.
... The importance of creating one-stop-shop BIM to manage and control life cycle sustainability issues of construction projects from the beginning to the end was mentioned in all these publications [5]. Although a framework of an integrated Heating, Ventilation, and Air-Conditioning (HVAC) automation and optimization tool was developed based on BIM, it is not a comprehensive framework covering all aspects of energy simulation [6]. Building Energy Modeling (BEM) needs to be done based on BIM models. ...
Development of Building Information Modelling (BIM) is assisting engineers with automating design/construction processes in the Architectural, Engineering and Construction (AEC) industry. Lack of such a comprehensive decision-making framework which utilizes BIM, Management Information Systems (MIS), simulation, and automation tools to choose between different construction alternatives have been addressed in the research background. As an example of decision-making objective, optimized smart building's equipment combination for a certain project should be selected in the feasibility study phase. In this research, a comprehensive decision-making framework was developed to choose smart building's equipment based on energy consumption and cost trade-off. Subsequently, smart building alternatives were considered as a decision-making example to choose the best alternative using BIM, MIS and simulation tools. The research contributes to the automating of some parts of the decision-making framework by developing an Application Programming Interface (API). This API helps with making the appropriate automating permutation out of possible options, totalizing the cost of each combination and sorting data in the research database developed for the recommended framework. This framework, database and API can be used similarly for any other decision making objective. Finally, a model representing the average conditions of residential buildings in Tehran was developed to choose the optimized smart building's equipment combination of 31 considered options. This optimized combination which included all possible smart building options except for smart lighting has an investment return of about 7.5 years which is more than that of similar projects in other countries due to Iran's low energy carrier tariff.
... Saha et al. [8] reviewed how data analytics could be used for occupancy sensing in buildings. Sha et al. [9] summarized how computational intelligence could be used to improve building energy system design. Guyot et al. [10] reviewed how artificial neural networks (ANNs) could be used for energy-related applications in the building sector. ...
Fueled by big data, powerful and affordable computing resources, and advanced algorithms, machine learning has been explored and applied to buildings research for the past decades and has demonstrated its potential to enhance building performance. This study systematically surveyed how machine learning has been applied at different stages of building life cycle. By conducting a literature search on the Web of Knowledge platform, we found 9579 papers in this field and selected 153 papers for an in-depth review. The number of published papers is increasing year by year, with a focus on building design, operation, and control. However, no study was found using machine learning in building commissioning. There are successful pilot studies on fault detection and diagnosis of HVAC equipment and systems, load prediction, energy baseline estimate, load shape clustering, occupancy prediction, and learning occupant behaviors and energy use patterns. None of the existing studies were adopted broadly by the building industry, due to common challenges including (1) lack of large scale labeled data to train and validate the model, (2) lack of model transferability, which limits a model trained with one data-rich building to be used in another building with limited data, (3) lack of strong justification of costs and benefits of deploying machine learning, and (4) the performance might not be reliable and robust for the stated goals, as the method might work for some buildings but could not be generalized to others. Findings from the study can inform future machine learning research to improve occupant comfort, energy efficiency, demand flexibility, and resilience of buildings, as well as to inspire young researchers in the field to explore multidisciplinary approaches that integrate building science, computing science, data science, and social science.
Introduction. The paper indicates that the further development of methods for calculating the thermal regime of disturbances in the automation of microclimate systems is still relevant. The aim of the work is to find an approximate analytical dependence of the air temperature in the conditions of a jump in the heat flow in air-conditioned rooms on time for the integral law of regulation of compensatory heat exposure from the climatic equipment. As a scientific hypothesis, the position is put forward about the possibility of expressing this dependence through trigonometric functions with sliding parameters. Materials and methods. The work involves the use of the basic differential equation of the thermal balance of a room serviced by climate systems equipped with continuous integral regulators, with a sudden change in the thermal disturbance. The linearization of the equation due to the freezing of variable coefficients and the Heaviside operational method are used, as well as the normalization method to eliminate the influence of the solution feature on the possibility of taking into account the initial conditions. Results. Analytical expressions are found for the coefficients and parameters of the approximate dependence of the temperature change in the room during the integral regulation of air conditioning systems under conditions of a jump in heat supply. Simplified formulas for the maximum deviation of the air temperature and the sliding frequency for the trigonometric functions involved in the solution are obtained, and their error is estimated in comparison with the exact solution on the example of one of the rooms in a residential building for the climatic conditions of Moscow. Conclusions. It is shown that the dependence of the air temperature in a room serviced by climate systems regulated by an integral law on time can be represented in a simplified analytical form on the basis of trigonometric functions with variable frequency. Additionally, the validity of the previously obtained exact solution for temperature changes in the form of an infinite series in degrees of time since the beginning of the thermal disturbance is confirmed.
Accurate building energy consumption prediction is essential for achieving energy savings and boosting the HVAC system's efficiency of operations. Therefore, in this work, a novel ensemble predictive model, which combines the weighted linear aggregation of Gaussian process regression (GPR) and least squared boosted regression trees (LSB), leading to WGPRLSB, is proposed for the accurate estimation of energy usage in the cases of Heating Load (HL) and Cooling Load (CL). Marine predator optimization (MPO) is used to evaluate the optimal values of the design parameters of the proposed methodology. Further, predictive models based on linear regression (LR), support vector regression (SVR), multilayer perceptron neural network (MLPNN), decision tree (DT), and generalized additive model (GAM) are also designed for comparison purposes. The results reveal that the value of RMSE is reduced by 12.4%–70.7% (HL) and 39.7%–64.9% (CL) for WGPRLSB in comparison to the other predictive models. The results of the performance index (PI) also confirm the effectiveness of the proposed model energy consumption prediction for HL and CL. Furthermore, the performance investigation on the second dataset reveals that WGPRLSB achieves the highest value of VAF (97.20%) compared to other designed models. It may be concluded that the proposed WGPRLSB accurately forecasts building energy demands.
The effective use of building information modeling (BIM) in the design phase generates countless benefits that contribute to risk management (RM). However, a better understanding of the relationship between the critical-success factors (CSFs) in the design phase has not yet been addressed. This study aims to investigate the influence of BIM CSFs in the design phase in the RM process, exploring the mediating effect played by BIM knowledge, RM knowledge, and BIM maturity. The research design applies the partial least-squares structural equation modeling technique, and the variables were collected by a survey with a sample of 195 respondents from different countries.
The results pointed out that earlier and accurate three-dimensional (3D) visualization of the design was the top-ranked recognized design factor in the use of BIM. The findings also indicated that BIM design CSFs have a positive impact on the RM process. Furthermore, there is a positive and significant indirect effect of BIM knowledge, RM knowledge, and BIM maturity through the path of BIM Design CSF on RM.
Building heating, ventilation, and air conditioning (HVAC) systems consume large amounts of energy, and precise energy prediction is necessary for developing various energy-efficiency strategies. Energy prediction using data-driven models has received increasing attention in recent years. Typically, two types of driven models are used for building energy prediction: sequential and parallel predictive models. The latter uses the historical energy of the target building as training data to predict future energy consumption. However, for newly built buildings or buildings without historical data records, the energy can be estimated using the parallel model, which employs the energy data of similar buildings as training data. The second predictive model is seldom studied because the model input feature is difficult to identify and collect. Herein, we propose a novel key-variable-based parallel HVAC energy predictive model. This model has informative input features (including meteorological data, occupancy activity, and key variables representing building and system characteristics) and a simple architecture. A general key-variable screening toolkit which was more versatile and flexible than present parametric analysis tools was developed to facilitate the selection of key variables for the parallel HVAC energy predictive model. A case study is conducted to screen the key variables of hotel buildings in eastern China, based on which a parallel chiller energy predictive model is trained and tested. The average cross-test error measured in terms of the coefficient of variation of the root mean square error (CV-RMSE) and normalized mean bias error (NMBE) of the parallel chiller energy predictive model is approximately 16% and 8.3%, which is acceptable for energy prediction without using historical energy data of the target building.
Although computer technologies have greatly advanced in recent years and help engineers improve work efficiency, the heating, ventilation, and air conditioning (HVAC) design process is still very time-consuming. In this paper, we propose a conceptual framework for automating the entire design process to replace current human-based HVAC design procedures. This framework includes the following automated processes: building information modeling (BIM) simplification, building energy modeling (BEM) generation & load calculation, HVAC system topology generation & equipment sizing, and system diagram generation. In this study, we analyze the importance of each process and possible ways to implement them using software. Then, we use a case study to test the automated design procedure and illustrate the feasibility of the new automated design approach. The purpose of this study is to simplify the steps in the traditional rule-based HVAC system design process by introducing artificial intelligence (AI) technology based on the traditional computer-aided design (CAD) process. Experimental results show that the automatic processes are feasible, compared with the traditional design process can effectively shorten the design time from 23.37 working hours to nearly 1 hour, and improve the efficiency.
Artificial intelligence (AI) controls are commonly used to save energy. However, excessive diversity in technological development has resulted in the inability to provide consistent energy-saving effects. Therefore, this study analysed 164 academic papers, with a total of 113 AI methods applied in six different fields of application. A universal workflow was developed to identify workable AI technologies for energy saving.
The concept of a universal workflow originates from machine learning (ML). To approach an ML problem, a workflow is constructed to assist researchers in defining their problem and assembling a dataset. In this study, the developed universal workflow adopted a hierarchical structure to guide users to choose learning, optimisation, and control tools to achieve energy saving. Based on the data from various studies, the developed workflow provides qualitative and quantitative energy-saving effects for application in diverse fields.
Universal workflow has contributed to the development of ML for commercial applications, and this research is also targeted to facilitate the commercial application of AI in the field of energy saving. Through a comprehensive analysis of experimental data, the universal workflow can confirm 35% energy cost saving in the building; 25% energy saving of the heating, ventilation and air conditioning equipment; 50% artificial lighting system energy saving; up to 70% reduction of information transfer and communication power; a continuous output of 30% peak power from the renewable energy device to the microgrid; and 20% power demand reduction in the factory. Corresponding to the choice of application technology, to the qualitative and quantitative benefits, and to the difference in control response, the universal workflow developed in this study provides a workable method to assist the use of AI in various applications. With workable design guidelines, the acceleration of commercial applications of AI for energy saving can be expected.
Distributed energy station receives multiple energy inputs to satisfy multiple energy demands of customers in district energy system. Their planning and design faces with challenges of high dimension, multivariate, and nonlinearity in comparing and selecting multiple components, connecting and coupling multiple energy flows, and evaluating whole life performance. This work analysis the characteristics of whole life operation of district energy system, and reveals the space-time helix and component turntable from its multivariate, high dimension, and nonlinearity. Deep integrated method is proposed to optimize all variables in interactive models to searching the optimum from all potential possibilities on system scale, component type, and operation mode. Free connection model, serial structure model, and modular operation model interact each other. The selfish optimum of gradual methods is eliminated and the nested algorithms of iterative methods is avoided. Through case study on a district energy system with three commercial buildings, compared with other integrated method, this method improves about 8.2%, 7.9%, and 9.5% benefits on economic, energy, and environmental sectors. The structuration and operation of distributed energy station designed by this method are more flexible and reliable. Compatibility and expansibility of models are presented. Also, the computation time of this deep integrated method is acceptable for engineering applications.
Introduction. It is noted that the improvement of the technology for calculating the thermal regime of premises in the conditions of automation of climate systems is still relevant. The aim of the work is to find a universal dependence of the air temperature in the building premises on time in conditions of a jump in heat access or heat loss for the integral law of regulating the compensatory heat flow from microclimate systems. In the form of a scientific hypothesis, we consider the statement about the power nature of such a dependence with the presence of a maximum and an asymptotic tendency to zero.
Materials and methods. The study involves the use of basic equations that relate the most significant components of heat flows in rooms serviced by microclimate systems equipped with integrated controllers under conditions of abrupt changes in thermal disturbances. The method of dimension analysis is used to identify dimensionless complexes that are essential for obtaining engineering dependencies, as well as software calculation of sums of infinite series with a given accuracy, numerical solution of nonlinear equations, and the method of power series economization.
Results. An analytical expression is found for changes in room temperature under integrated control of climate equipment under conditions of a jump in heat availability, which has the form of an infinite series in degrees of a dimensionless parameter that characterizes the properties of the room and the automation system. A simplified expression for the deviation of air temperature is obtained and a formula for the required control time is derived, as well as its estimation is given on the example of one residential building in the climatic conditions of Moscow.
Conclusions. It is shown that the dependence of the air temperature in a room serviced by microclimate systems with integrated controllers on time is represented in a universal dimensionless form, suitable for any objects regardless of their specific characteristics. The previously discovered relationships for the moment of maximum deviation and the value of the dynamic control error depending on the air exchange of the room, the transmission coefficient of the controller and the room’s own heat stability are confirmed and refined.
Artificial Intelligence (AI) and Machine Learning (ML) are currently hot topics in industry and business practice, while management-oriented research disciplines seem reluctant to adopt these sophisticated data analytics methods as research instruments. Even the Information Systems (IS) discipline with its close connections to Computer Science seems to be conservative when conducting empirical research endeavors. To assess the magnitude of the problem and to understand its causes, we conducted a bibliographic review on publications in high-level IS journals. We reviewed 1,838 articles that matched corresponding keyword-queries in journals from the AIS senior scholar basket, Electronic Markets and Decision Support Systems (Ranked B). In addition, we conducted a survey among IS researchers (N = 110). Based on the findings from our sample we evaluate different potential causes that could explain why ML methods are rather underrepresented in top-tier journals and discuss how the IS discipline could successfully incorporate ML methods in research undertakings.
Multi-energy flexibility measures comprising energy substitution and demand-side management (DSM) can enhance the control of buildings and help them participate in the energy market, obtaining greater profit margins. However, the application of these flexibility measures is also subject to many limitations. For example, DSM is a kind of load redistribution process in which the energy payback constraints associated with comfort or usage demand should be considered. Therefore, this work studies the optimal energy management of a building energy system (BES) considering multi-energy flexibility measures, specifically under the energy payback effect, to better guide the peak shaving strategies. First, energy substitution measures are proposed involving energy conversion and storage modeling. Second, a novel dynamic two-step DSM measure is modeled for the reduction and recovery process. Then, a mixed-integer and linear programming (MILP)-based energy management model is developed to optimize the operation of smart buildings for peak shaving. The case studies demonstrate that 1) a BES can obtain better profit by utilizing multi-energy flexibility measures; 2) optimized multi-load recovery strategies can enhance the flexibility potential of a BES; and 3) a reasonable multi-load recovery mechanism should be established to offset the energy payback effect.
Nowadays, most areas of human activity should be reviewed with the aim of reducing CO2 emissions, since these activities are producing the majority of these emissions. Specifically, the building sector is one of the main responsible activities. In order to minimize the ecological footprint and ensure energy sufficiency, European Union created the nearly-Zero Energy Building (nZEB) concept. More than ten years have elapsed and it worth to review the current state around the concept, considering the new advances in computer development that are already applicable to this field. Accordingly, recent researches published in reputed indexed journals and international conferences have been reviewed. This paper explains the nZEB concept and reviews research articles focused on achieving it. A research gap is detected, so enabling concepts and technologies as Building Energy Performance Simulation (BEPS) tools and Model Predictive Control (MPC) are recalled, and relevant researches where used are included in a specific state-of-the-art for each concept, since the academia considers that these tools should be applied in building air conditioning to achieve nZEB. After this deep analysis, we conclude that the possibilities to optimize the energy consumption are huge combining properly in a holistic way BEPS tools for modeling and simulation and MPC for control strategies. It is possible to manage a Heating, Ventilation and Air Conditioning (HVAC) system using Renewable Energy Sources (RES) in an effective means, reducing CO2 emissions problems worldwide and reaching considerable energy savings.
In this paper, the free vibration of axially-loaded, multi-cracked Timoshenko beams with differing
boundary conditions, namely, hinged-hinged, fixed-fixed, fixed-hinged, and fixed-free is studied. The cracked
beam system is represented as several beam segments connected by massless rotational springs with
sectional flexibility. Each segment is assumed to obey the Timoshenko beam theory. A simple transfer matrix
method is used to derive the characteristic equation of the axially-loaded, multi-cracked beam with differing
boundary conditions. The characteristic equation and corresponding mode shapes are a function of natural
frequency, crack size and location, and physical parameters of the beam. In this paper, the effects of crack
depth, number of cracks, position of cracks, axial load, shear deformation and rotary inertia on the dynamic
behavior of multi-cracked beams are studied in detail. It is found that there is good agreement between the
results obtained in this study and results available in the literature. Additionally, interesting observations
overlooked by other researchers are obtained.
Building performance simulation (BPS) may provide valuable support to the planning of more energy efficient buildings, but the effort needed to create a complete model can hinder this potential. In particular, an idealized representation of HVAC systems is often used instead of a more explicit and insightful model, because of the difficulty of creating the latter. While automated methods are available for the translation of building geometry for BPS, or for the sizing of delivery components, there is no generally accepted way of determining the HVAC distribution subsystems. This paper presents a method allowing models of these subsystems to be automatically created for the purpose of energy simulation. Based on an input consisting of a zoned building model and sized delivery components, networks of potential distribution components are determined and components corresponding to subsets of these networks are created and sized. The output is a component-based model of delivery and distribution HVAC systems, to be completed with generation components and translated into an input for a building performance simulation engine. The method is tested in a case study where its results are compared to an existing heating system for an office building.
Modern houses are more and more frequently characterized by the presence of “smart” metering devices, capable of measuring air temperature, relative humidity, air quality, and in the more sophisticated cases, even electric equipment consumptions. In addition, other relevant parameters such as illuminance may often be determined and they can be used as proxy variables to account for other important aspects (such as solar irradiance) influencing the energy balance of a building. Such information, in combination with weather data which can be retrieved by other sources (or by additional sensors), may conveniently contribute to the creation of a “black box” model in which, given a few input variables it is possible to output a variable which would result from otherwise complex calculations (e.g. an energy balance) requiring many data. The availability of such a “black box” could be helpful under many points of view, such as benchmarking energy consumptions and stimulating virtuous behavior from the occupants. To test whether such approach can be feasible, an EnergyPlus model of a real house was made, trying to accurately reproduce building features, systems set-points, and occupant behaviors. The overall simulated energy consumptions were compared with the real ones resulting from energy bills, thus ensuring a good agreement with reality. The dataset resulting from EnergyPlus was then used to train an artificial neural network (ANN) capable of yielding hourly energy consumptions based on limited input data. Finally, the relative importance of the different input variables was analyzed to understand which might influence prediction accuracy most.
Energy consumption in buildings contributes to 41% of global carbon dioxide emissions through electricity and heat production, making the design of mechanical systems in buildings of paramount importance. Industry practice for design of mechanical systems is currently limited in the conceptual design phase, often leading to sub-optimal designs. By using Generative Design (GD), many design options can be created, optimized and evaluated, based on system energy consumption and life-cycle cost (LCC). By combining GD for Architecture with GD for HVAC, two areas of building design can be analyzed and optimized simultaneously, resulting in novel designs with improved energy performance. This paper presents GD for HVAC, a Matlab script developed to create improved zone level mechanical systems for improved energy efficiency. Through experiments, GD methodologies are explored and their applicability and effect on building HVAC design is evaluated.
In construction one of the important factors of the effective development of the construction production is improving the methodology of management, as an economic entity. This became especially possible in view of the rapid development of computer technology and logical-linguistic methods. Active application of information and computing technology allows you to select economically feasible methods of management based on reliably grounded methods of artificial intelligence. The ways of solving the tasks in these studies were developed on the basis of the extension of mathematical and symbolic logic. Given the characteristics of the condition of application of methods of artificial intelligence in the field of organizational solutions for the construction. Stages of formation of organizational and technological solutions using rule-based expert systems, artificial neural networks, genetic algorithms. The proposed method is effective implementation of organizational and technological solutions with the use of its information models and systems of making management decisions on the results of monitoring the construction process. The basic information based on models of knowledge representation (semantic, frames, production rules and regulations, precedents), and concepts on the application of artificial neural networks in construction and information about the methods of extraction of knowledge and formation of knowledge bases. Examines the structure and functions of expert systems and decision support systems solutions.
Buildings are responsible for 40% of global energy use and contribute towards 30% of the total CO2 emissions. The drive to reduce energy use and associated greenhouse gas emissions from buildings has acted as a catalyst in the development of advanced computational methods for energy efficient design, management and control of buildings and systems. Heating, ventilation and air-conditioning (HVAC) systems are the major source of energy consumption in buildings and an ideal candidate for substantial reductions in energy demand. Significant advances have been made in the past decades on the application of computational intelligence (CI) techniques for HVAC design, control, management, optimization, and fault detection and diagnosis. This article presents a comprehensive and critical review on the theory and applications of CI techniques for prediction, optimization, control and diagnosis of HVAC systems.The analysis of trends reveals the minimization of energy consumption was the key optimization objective in the reviewed research, closely followed by the optimization of thermal comfort, indoor air quality and occupant preferences. Hardcoded Matlab program was the most widely used simulation tool, followed by TRNSYS, EnergyPlus, DOE–2, HVACSim+ and ESP–r. Metaheuristic algorithms were the preferred CI method for solving HVAC related problems and in particular genetic algorithms were applied in most of the studies. Despite the low number of studies focussing on MAS, as compared to the other CI techniques, interest in the technique is increasing due to their ability of dividing and conquering an HVAC optimization problem with enhanced overall performance. The paper also identifies prospective future advancements and research directions.
Electrical load forecasting plays a vital role in order to achieve the concept of next generation power system such as smart grid, efficient energy management and better power system planning. As a result, high forecast accuracy is required for multiple time horizons that are associated with regulation, dispatching, scheduling and unit commitment of power grid. Artificial Intelligence (AI) based techniques are being developed and deployed worldwide in on Varity of applications, because of its superior capability to handle the complex input and output relationship. This paper provides the comprehensive and systematic literature review of Artificial Intelligence based short term load forecasting techniques. The major objective of this study is to review, identify, evaluate and analyze the performance of Artificial Intelligence (AI) based load forecast models and research gaps. The accuracy of ANN based forecast model is found to be dependent on number of parameters such as forecast model architecture, input combination, activation functions and training algorithm of the network and other exogenous variables affecting on forecast model inputs. Published literature presented in this paper show the potential of AI techniques for effective load forecasting in order to achieve the concept of smart grid and buildings.
When heat conduction is considered, assessment of simplified RC (resistances and capacities) model parameters is a great challenge due to thermal mass. This article presents a new method for identifying the parameters of equivalent thermal RC network models by illustrating the case of 1D conduction in a wall. The parameters of the RC model are identified by the optimization method of genetic algorithms. The performance criterion is based on the sum of errors related to surface temperature responses, due to steps imposed on the ambient temperatures on either side of the wall. The originality of the method is to consider a constant time step on the logarithmic time. This is particularly important to ensure the validity of the RC model for high pulses when the frequency analysis is considered.
The simultaneous optimization of building's fabric construction, the size of heating, ventilating and air conditioning (HVAC) system, and the HVAC system supervisory control strategy, would auto- matically account for the thermal coupling be- tween these building design elements. This paper describes the formulation of such an optimization problem for a single air conditioned zone. The problem formulation is described in terms of the optimization problem variables, the design con- straints, and the design objective function. The optimization problem has been solved using a Genetic Algorithm (GA) search method. It is concluded that the GA is able to find a feasible so- lution and it exhibits an exponential convergence on a solution. The solutions obtained are near- optimal, the lack of final convergence being re- lated to variables having a secondary eect on the energy cost objective function. Further research is required to investigate methods for improving the handling of equality constraints and to reduce the number of control variables (which will also improve the robustness of the algorithm).
A chiller model is developed to simulate the operational performance of centrifugal chillers based on the performance data from site measurement. The model is developed on the basis of basic mechanistic principles of chillers to ensure the reliability and accuracy of the model in a wide working range using the parameters identified using limited performance data. A pre-processor identifies the parameters of chillers using chiller performance data to ensure convenience of application. This paper presents the model, the strategy for model parameter identification, and the applications and validation of the model on an existing multiple chilling system in a building and on a chiller using manufacturer's catalogue data.
In the present day environment, smart buildings require optimization of energy consumption through monitoring, consumption prediction and making policy decisions accordingly. Attributes related to building design and structure play a vital role in heating load(HL) and cooling load(CL) of the building which directly affects the energy performance of the buildings. For prediction of HL and CL, emerging machine learning approaches can help in improving accuracy and efficiency in real time. This paper provides improvements in energy load assessment of the buildings. It is the first is the in-depth study and analysis of design and structural attributes and their correlation with HL and CL, the novel methods based on ELM and its variants online sequential ELM(OSELM) to predict HL and CL. This study also proposes OSELM based online/real-time prediction when data is coming in stream The total 24 models have been developed including 12 models based on ELM and 12 models based on OSELM with different feature sets and activation functions. Models have been compared on the basis of accuracy, computational performance and efficiency with few existing models. The experimental results show that the proposed models learn better and outperform other popular machine learning approaches such as the artificial neural network(ANNs), support vector machine(SVM), radial basis function network(RBFN), random forest(RF) and existing work in the energy and building domain.
Membrane fouling in membrane bioreactors (MBR) normally depends on the microbial cell density and microbial population structure. A nitrifying-enriched activated sludge (NAS) was obtained in this study through particular ammonium feeding of conventional activated sludge (CAS). Next, the dominance of autotrophic nitrifier population in NAS system was checked and compared with CAS system by its high nitrification efficiency (100% vs. 43%) and low COD removal (9% vs. 65%). Furthermore, the maximum amount of N-NO3- produced from similar concentrations of ammonium in CAS and NAS systems were 6.6 mg/L and 37.5 mg/L, respectively. A filterability test also was done in the cross-flow and simple dead-end filtration systems proportionally employing different amount of NAS and CAS with a constant MLSS concentration of 2000 mg/L. NAS was twice as filterable compared to CAS. Soluble microbial products (SMP) and extracellular polymeric substance (EPS) in CAS were significantly higher than NAS system (2 and 6 mg/L for NAS vs.100 and 36 mg/L for CAS). By increasing the proportion of nitrifying bacteria, the permeation was enhanced remarkably about 2.5 folds and the operation time of MBR was approximately doubled (6 to 11.5 hours). The results indicated that an appropriate C/N ratio can control the microbial population and help the nitrifier significantly mitigate fouling in MBR.
Two types of sampling plans are examined as alternatives to simple random sampling in Monte Carlo studies. These plans are shown to be improvements over simple random sampling with respect to variance for a class of estimators which includes the sample mean and the empirical distribution function.
Recent Artificial Intelligence studies have achieved substantial improvements in practical tasks by using extensive amounts of data. We assume that a substantial part of the data to guide artificial design technologies resides in existing design examples. Developing ways to use this data may enable improvements in intelligent design tools, with the hope that these may provide more effective design workflows and more productive design practices. Such improvements may result in more in-depth evaluations of potentials and alternatives for design situations; hence better planning for the spatial environment.
Various approaches have been developed to use representations of architectural examples for artificially tackling architectural design tasks. This study presents a review of the historical development of these approaches, with an overall aim to investigate where and how design examples have been used for practical computational design applications.
The review encompasses traditional and recent Shape Grammar and Procedural Modeling studies, Case-Based Design, Similarity-Based Evaluation and Design, and recent studies on the architectural uses of Machine Vision, Semantic Modeling, Machine Learning, and Classification. The emphasis of the review is on the studies that aim at designing or generating new design examples, particularly for building layouts, façades, envelopes, and massing.
For a comparative evaluation of the current capabilities of the examined lineages of studies, we propose a minimum set of design capabilities, and assess each study through this framework. This reveals the overall patterns of already covered requirements.
The review shows that initial hand-operated SGs gave way to automatic generation, which in turn developed into automated SG extraction, through increasing levels of computational capabilities. Case-Based Design has been neglected; however, it can be reinvigorated through novel AI techniques. On the other hand, Similarity-Based Evaluation may complement and balance the orientation towards technical performance. Machine Learning and Computer Vision appear as potential intermediaries for connecting these threads.
There are example-based studies towards almost all aspects of artificial design; yet, these have not been tackled adequately or definitively. In particular, dynamic process control is still only a future potential. On the other hand, the examined research lineages have the potential to assume complementary roles for more capable and multifaceted design systems. As an overall result, example-based research perspectives raise important possibilities for intelligent design systems.
Configuring the number and size of chillers in a multiple-chiller plant properly is an efficient way to improve the plant energy efficiency. At the design stage, the optimal configuration can be achieved through matching the capacity to load as closely as possible across the full-load profile. However, in spite of the fact that current literature offers practical recommendations, a systematic method to optimize the configuration of multiple-chiller plants is lacking. Due to the lack of accurate information at the design stage and only limited knowledge of the eventual realization it is hard to predict the building’s cooling load. Moreover, there is no operational data to predict the system performance. Both explain the existence of uncertainty in the HVAC plant design process. This paper, therefore, proposes a strategy to optimize the configuration of multiple-chiller plants, which takes account of the load side uncertainty as well as the COP uncertainty and selects the optimal configuration through a life-cycle analysis. Both the load side uncertainty and the COP uncertainty are quantified using statistical distributions. To facilitate applications, the distributions of the cooling load profile of different types of buildings under different weather conditions are investigated and are classified into four categories, and the optimal configuration schemes under each type of cooling load distribution are analyzed and summarized in a tabulated form.
Heating load and cooling forecasting are essential for estimating energy consumption, and consequently, helping engineers in improving the energy performance right from the design phase of buildings. The capacity of heating ventilation and air-conditioning system of the building contribute to the operation cost. Moreover, building being one of the sectors with heavy energy use, it is required to develop an accurate model for energy forecasting of building and constructing energy-efficient buildings. This paper explores different machine learning techniques for predicting the heating load and cooling load of residential buildings. Among these methods, we focus on advanced techniques like Multivariate Adaptive Regression Splines (MARS), Extreme Learning Machine (ELM) and a hybrid model of MARS and ELM along with a comparison of the results with those of more conventional methods like linear regression, neural network, Gaussian processes and Radial Basis Function Network. The MARS model is a non-parametric regression model that splits the data and fits each interval into a basis function and ELM is similar to a Single Layer Feed-forward Neural Network except that in ELM randomly assigned input weights are not updated. As an improvement, we have tried a hybrid model that uses MARS to evaluate the importance of every parameter in the prediction and these important parameters have been fed to the ELM to build hybrid model and it can be seen that this boosts the ELM performance to match up to the accuracy of MARS with lesser computation time. Finally, a comparative study examines the performances of the different techniques by measuring different performance metrics.
Finding the optimal balance between electricity demand and production constrained to economic and comfort variables requires intelligent decision and control. This article addresses the formulation of three models that optimize control of a heating, ventilation and air conditioning (HVAC) system in an experimental room, which are coupled with two thermal models of the indoor temperature. Electricity is supplied by the grid and a photovoltaic system with batteries. The primary objective is to maximize users comfort while minimizing cost constrained to: thermal comfort; variable electricity price; and available electricity in batteries that are charged by a PV system. Three models are developed: (i) dynamic programming with simplified thermal model (STM), (ii) genetic algorithm with STM, and (iii) genetic algorithm with EnergyPlus.
The genetic algorithm model that uses EnergyPlus to simulate indoor temperature generally achieves higher convergence to the optimal value, which also is the one that uses more electricity from the PV system to operate the HVAC. The dynamic programming performs better than the genetic algorithm (both coupled with STM). However, it is limited by the fact that uses STM, which is a less accurate model to simulate indoor temperature especially because it is not considering thermal inertia.
A novel fault detection algorithm based on machine learning is introduced in this paper, that is applied to the detection of faults in heater, ventilation and air conditioning (HVAC) systems. The algorithm is based on the use of a set of nonlinear regressors intended to estimate the response of the HVAC to the external variables. The regression algorithm is the well known Gaussian Process Regression, which, through a Gaussian modeling of the parameter priors and the conditional likelihood of the observations, is able to produce a probabilistic model of the prediction. We use the prediction error and its estimated variance as an input to a Support Vector Machine novelty detector that, in an unsupervised way, is able to detect the faults of the HVAC. This algorithm improves the standard novelty detection, as it can be seen in the experiments.
Opportunities to save energy in the design and operation of Heating, ventilation and air conditioning (HVAC) systems have come into sharp focus. Lighting intensity, electric equipment load and occupant density have great impact on the peak cooling load of building, which is the basis of the chiller sizing. There is inevitable uncertainty in the determination of the values of these three scenario parameters at the detailed HVAC design stage and thus these uncertainties can considerably impact chiller sizing choice. However, the conventional chiller sizing method does not deal well with these uncertainties. A new chiller sizing method is proposed in this paper taking into account the scenario parameter uncertainty, the discrete spectrum of nominal cooling capacity of available chillers, the difference of chiller cooling capacity under nominal condition and peak cooling load condition. The new approach quantifies the impacts of scenario uncertainty on peak cooling load, chiller life cycle cost (LCC), and also on annual set point unmet hours based on the Monte Carlo method and dynamic simulation. For the case study, compared with conventional sizing method, the new sizing method can greatly reduce chiller nominal cooling capacity and the minimum reduction is 22.51%. The new method can help the HVAC designer to determine the optimal chiller size with the consideration of scenario uncertainty and the balance of chiller LCC and indoor thermal comfort of HVAC system.
In most heating, ventilation and air conditioning systems, the ductwork layout, i.e., the network structure of the ducts, as well as the number and locations of the fans, is an important determinant of the installation's cost and performance. Nevertheless, the layout is not explicitly taken into account in existing duct design methods. Most methods assume the layout of the air distribution system to be predetermined and focus solely on the sizing of each fan and duct in the network. This paper aims to outline the current state-of-the-art in air distribution system design and highlights the main shortcomings. Additionally, previous research is extended by presenting a novel problem formulation that integrates the layout decisions into the optimization problem. In this problem, called the air distribution network design optimization problem, the optimal air distribution system configuration, i.e., the optimal ductwork layout, is determined jointly with the duct and fan sizes, thereby minimizing the total cost of the system. This novel combinatorial optimization problem is characterized by discrete decision variables, and non-linear constraints. This paper also motivates the need for benchmark instances to evaluate the performance of existing or new developed optimization methods and advance future research in the field of air distribution system design optimization. A test network generator is developed in this research to generate such a set of instances.
Dynamic modeling of HVAC system is extremely important for control analysis toward building energy saving. In order to provide researchers a simulation platform to analyze different control strategies, this paper introduces a simulation platform with customized Simulink block library based on dynamic HVAC component model. As an initiating effort, the current simulation platform is composed of basic modular HVAC components, including conduit, damper/valve, fan/pump, flow merge, flow split, heating coil, cooling coil, and zone. These modules are developed into Simulink customized blocks. The simulation platform is capable to calculate the flow rates of fresh air, exhaust air, and return air based on system characteristic and fan curve with customizable basic/advanced control strategies. A case study is proposed using single-zone, constant volume system by comparing the uncontrolled, fixed temperature and damper position controlled, and schedule based reset controlled cases. The simulation result proves that schedule based temperature and damper position reset has a significant impact on energy saving for both heating and cooling seasons. This simulation platform can be especially useful for analyzing the dynamic performance of different HVAC control strategies.
Electricity load forecasting is an important tool which can be utilized to enable effective control of commercial building electricity loads. Accurate forecasts of commercial building electricity loads can bring significant environmental and economic benefits by reducing electricity use and peak demand and the corresponding GHG emissions. This paper presents a review of different electricity load forecasting models with a particular focus on regression models, discussing different applications, most commonly used regression variables and methods to improve the performance and accuracy of the models. A comparison between the models is then presented for forecasting day ahead hourly electricity loads using real building and Campus data obtained from the Kensington Campus and Tyree Energy Technologies Building (TETB) at the University of New South Wales (UNSW). The results reveal that Artificial Neural Networks with Bayesian Regulation Backpropagation have the best overall root mean squared and mean absolute percentage error performance and almost all the models performed better predicting the overall Campus load than the single building load. The models were also tested on forecasting daily peak electricity demand. For each model, the obtained error for daily peak demand forecasts was higher than the average day ahead hourly forecasts. The regression models which were the main focus of the study performed fairly well in comparison to other more advanced machine learning models.
A previous paper articulated the current problems faced by HVAC&R professionals involved in the conceptual design process and proposed a solution based on a knowledge-based expert system (KBES) approach, which can automatically synthesize all the feasible secondary and primary systems that can then be evaluated using currently available hourly building energy simulation programs. The previous paper described the general framework of such a KBES module called HVAC-KBCD. The module consists of(1) static knowledge (based on heuristics, design practice, and standards) containing assembly and application rules needed to prune or shrink the solution space of all feasible secondary and primary systems and (2) dynamic knowledge containing initiation and matching rules to provide a guided search and further shrink the solution space by imposing restrictions of how to combine secondary systems among themselves (since the building has several zones) and with primary systems. These capabilities, along with specially developed models for first and maintenance costs, have been programmed into a commercial KBES shell. This paper describes the type of knowledge specific to office buildings that needs to be coded into the HVAC-KBCD module and illustrates its capabilities when applied to a case study example. The case study serves to illustrate the relative ease and thoroughness with which this design assistant can generate and evaluate a large number of HVAC&R system design alternatives.
A design methodology which integrates a knowledge based expert system (KBES) that automate the synthesis of various secondary and primary systems of a building HVAC&R system is discussed. The system can be analyzed using building energy simulation programs. The system configurations from the HVAC- knowledge-based conceptual design (KBCD) help to comply with current energy standards. Most of the building energy simulation tools employ similar components, subsystems and methodologies for simulation, making the HVAC-KBCD, a generic approach for integrating building simulation programs.
An HVAC duct system was designed using the dynamic programming method (DPM), which considered system pressure equilibrium and the least life-cycle cost to derive the duct size and fan capacity. A simple example was provided to understand the characteristics of DPM and compare with the conventional duct design methods. Since DPM contains the concept of minimizing the life-cycle cost, the design results not only guarantee each path to share the same pressure but also bear a smaller cost than other methods. The limit on duct diameter or flow velocity was added to the computation process. As a result, the derived outcome satisfied all the requirements of duct diameter or flow velocity. In particular, the computation needed in DPM is simpler than most methods that consider the least life-cycle cost, making DPM a successful method for duct system design.
The move towards a de-carbonised world, driven partly by climate science and partlyby the business opportunities it offers, will need the promotion of environmentallyfriendly alternatives, if an acceptable stabilisation level of atmospheric carbon dioxide isto be achieved. This requires the harnessing and use of natural resources that produce noair pollution or greenhouse gases and provides comfortable coexistence of human,livestock, and plants. This study reviews the energy-using technologies based on naturalresources, which are available to and applicable in the farming industry. Integral conceptfor buildings with both excellent indoor environment control and sustainableenvironmental impact are reported in the present communication. Techniques consideredare hybrid (controlled natural and mechanical) ventilation including night ventilation,thermo-active building mass systems with free cooling in a cooling tower, and air intakevia ground heat exchangers. Special emphasis is put on ventilation concepts utilisingambient energy from air ground and other renewable energy sources, and on theinteraction with heating and cooling. It has been observed that for both residential andoffice buildings, the electricity demand of ventilation systems is related to the overalldemand of the building and the potential of photovoltaic systems and advanced cogenerationunits. The focus of the world's attention on environmental issues in recentyears has stimulated response in many countries, which have led to a closer examinationof energy conservation strategies for conventional fossil fuels. One way of reducingbuilding energy consumption is to design buildings, which are more economical in theiruse of energy for heating, lighting, cooling, ventilation and hot water supply. Passivemeasures, particularly natural or hybrid ventilation rather than air-conditioning, candramatically reduce primary energy consumption. However, exploitation of renewableenergy in buildings and agricultural greenhouses can, also, significantly contributetowards reducing dependency on fossil fuels. This article describes various designs oflow energy buildings. It also, outlines the effect of dense urban building nature on energyconsumption, and its contribution to climate change. Measures, which would help to saveenergy in buildings, are also presented.
Dynamic models of heat pumps are useful in developing feedback controllers and fault-detection-diagnostic (FDD) studies. Several system models have been documented in the literature, but few are for centrifugal chillers. Existing publications focus on model development and validation, providing minimal detail on the numerical aspects of the solution. The solution of the PDEs that are obtained to model the heat exchangers is critical in terms of accuracy and execution speed. This paper presents the development of a centrifugal chiller system model, using the finite-volume (FV) approach for shell-and-tube heat exchangers and aspects such as mesh dependence, integration order, and step size. Sufficient and necessary mesh sizes for accurate steady-state prediction are determined for the heat exchangers. Execution speeds with integration algorithms of the first, second, and fourth order are compared for equivalent accuracy. The model is based on firs principles, allowing it to be used over a wide range of operating conditions and transients. The model is validated using data from a 90-ton R-134a centrifugal chiller.
The leading introductory book on data mining, fully updated and revised!When Berry and Linoff wrote the first edition of Data Mining Techniques in the late 1990s, data mining was just starting to move out of the lab and into the office and has since grown to become an indispensable tool of modern business. This new edition—more than 50% new and revised— is a significant update from the previous one, and shows you how to harness the newest data mining methods and techniques to solve common business problems. The duo of unparalleled authors share invaluable advice for improving response rates to direct marketing campaigns, identifying new customer segments, and estimating credit risk. In addition, they cover more advanced topics such as preparing data for analysis and creating the necessary infrastructure for data mining at your company. Features significant updates since the previous edition and updates you on best practices for using data mining methods and techniques for solving common business problemsCovers a new data mining technique in every chapter along with clear, concise explanations on how to apply each technique immediatelyTouches on core data mining techniques, including decision trees, neural networks, collaborative filtering, association rules, link analysis, survival analysis, and moreProvides best practices for performing data mining using simple tools such as ExcelData Mining Techniques, Third Edition covers a new data mining technique with each successive chapter and then demonstrates how you can apply that technique for improved marketing, sales, and customer support to get immediate results.
The several parameters affect the heat load of a building; geometry, construction, layout, climate and the users. These parameters are complex and interrelated. Comprehensive models are needed to understand relationships among the parameters that can handle non-linearities. The aim of this study is to predict heat load of existing buildings benefiting from width/length ratio, wall overall heat transfer coefficient, area/volume ratio, total external surface area, total window area/total external surface area ratio by using artificial neural networks and compare the results with a building energy simulation tool called KEP-IYTE-ESS developed by Izmir Institute of Technology. A back propagation neural network algorithm has been preferred and both simulation tools were applied to 148 residential buildings selected from 3 municipalities of Izmir-Turkey. Under the given conditions, a good coherence was observed between artificial neural network and building energy simulation tool results with a mean absolute percentage error of 5.06% and successful prediction rate of 0.977. The advantages of ANN model over the energy simulation software are observed as the simplicity, the speed of calculation and learning from the limited data sets.
In current practice, HVAC systems are sized based on standardized procedures that were mostly developed by ASHRAE. The standard approach only implicitly deals with uncertainty in peak system demand through the selection of an appropriate design day and the choice of a safety factor. Although this method works satisfactorily in most cases, it offers no support to a system designer who wants to track the risk associated with an undersized system. The opposite, i.e. avoiding that the system is needlessly oversized deserves even more attention given the fact that current practice of “defensive sizing” leads to oversized systems which leads to wasted capital investment and systems that operate far away from the optimum efficiency loads. This paper explores a new framework to guide the use of uncertainty analysis (UA) and sensitivity analysis (SA) in HVAC system sizing. UA will replace the safety factor with quantified margins based on comprehensive quantification of different sources of uncertainty. A probabilistic-based SA is then used to identify the important individual factors or groups of factors that contribute to uncertainty, providing means of risk management by applying better quality assurance methods or negotiating performance contracts.
The purpose of this study is to propose an optimal design method for the HVAC system in apartment using a genetic algorithm and to examine the possibility for the energy conservation of a designed HVAC system. The energy demand for cooling and heating in apartment house is determined by using TRNSYS. By a modified genetic algorithm called multi-island genetic algorithm, the optimal running pattern of HVAC systems is decided to minimize the energy consumption. An optimal design method for the HVAC system of the apartment house was proposed using both genetic algorithm and data of cooling/heating demand load simulated by TRNSYS. It has been confirmed that energy for equipment systems in apartment house can be saved by using operation plan of HVAC systems. The results show that this proposed method is significantly capable of determining optimal system design for saving energy in apartment house. We will perform the design of HVAC system considering an initial cost, a running cost and emission of CO2, and so on in the future.
An optimization model is presented for maximizing the generation of wind power while minimizing its variability. In the optimization model, data-driven approaches are used to model the wind-power generation process based on industrial data. A new constraint is developed for governing the data-driven wind-power generation model based on physics and statistical process control theory. Since the wind-power model is nonparametric, computational intelligence algorithms are utilized to solve the optimization model. Computer experiments are designed to compare the performance of computational intelligence algorithms. The improvement in the generated wind power and its variability is demonstrated with the computational results.
A data-mining approach is proposed to model a pumping system in a wastewater treatment plant. Two parameters, energy consumption and wastewater flow rate after the pumping system, are used to evaluate the performance of 27 scenarios while the pump was operating. Five data-mining algorithms are applied to identify the relationships between the outputs (energy consumption and wastewater flow rate) and the inputs (elevation level of the wet well and the speed of the pumps). The accuracy of the flow rate and energy consumption models exceeded 90%. The derived models are deployed to optimize the pump system. The computational results obtained with the proposed models are discussed.
This paper presents a new modeling approach to quantify building energy performance in early design stages. Building simulation models can accurately quantify building energy loads, but are not amenable to the early design stages when architects need an assessment tool that can provide rapid feedback based on changes to high level design parameters. We utilize EnergyPlus, an existing whole building energy simulation program, within a Monte Carlo framework to develop a multivariate linear regression model based on 27 building parameters relevant to the early design stages. Because energy performance is sensitive to building size, geometry, and location, we model a medium-sized, rectangular office building and perform the regression in four different cities—Miami, Winston-Salem, Albuquerque, and Minneapolis—each representing a different climate zone. With the exception of heating in Miami, all R2 values obtained from the multivariate regressions exceeded 96%, which indicates an excellent fit to the EnergyPlus simulation results. The analysis suggests that a linear regression model can serve as the basis for an effective decision support tool in place of energy simulation models during early design stages. In addition, we present standardized regression coefficients to quantify the sensitivity of heating, cooling, and total energy loads to building design parameters across the four climate zones. The standardized regression coefficients can be used directly by designers to target building design parameters in early design that drive energy performance.
In this study, we introduce the Modified T-Method which can be applied to the prediction of the flow distribution in multi-fan duct systems. Introduction of individual flow routes enables the division of a complicated multi-fan ductwork into a set of simultaneous single-fan subsystems. The present method was tested on the single-fan duct system presented in the ASHRAE handbook. The test system consisted of several branches and a fan. Our results showed good agreement with those published. Also, we tried another approach, in which the symmetry of a two-fan duct system was investigated. Verification was accomplished successfully by the analysis of a two-fan duct system that yielded physical consistency. In order to evaluate the utility and capability of the title method, a bathroom ventilation system in a 20-storey residential building was selected as an example of an application. Under the typical design conditions, the air flow rate of each exhaust fan at the balance point was predicted successfully and it was shown that such information can lead to an engineering estimate for the system performance as a whole. While some deficiencies in ventilation were found in bathrooms on the lower floors with 6mmAq-rated exhaust fans, they disappeared over the building as a whole by using fans of enhanced static pressures, 7 and 8mmAq. Finally, the work has demonstrated that this method is likely to be useful for the design of multi-fan duct systems.
A cooling coil unit (CCU) is nonlinear in nature. Existing CCU models for control and optimization are either linear approximations around an operating point or very complex nonlinear ones, resulting in difficulties for real time applications. Therefore, it is of practical importance to develop a simple, yet accurate CCU engineering model that will yield better real time control and optimization of heating, ventilating and air conditioning (HVAC) systems. In this paper, a technique for developing a simple, yet accurate engineering CCU model is presented. The modeling technique is based on an energy balance and heat transfer principles. Commissioning information is then used to estimate, at most, three model parameters by either a linear or nonlinear least squares method. Experiment shows that the method is robust and gives a better match to real performance over the entire operating range when compared to existing methods in the literature. This model is expected to have wide applications in control and optimization of HVAC systems. The modeling methodology can also be extended to other heat exchangers.
This paper is the second paper out of two which present the development of a dynamic model for single-effect LiBr/water absorption chillers. The first part describes the model in detail with respect to the heat and mass balances as well as the dynamic terms. This second part presents a more detailed investigation of the model performance, including performance analysis, sensitivity checks and a comparison to experimental data. General model functionality is demonstrated.A sensitivity analysis gives results which agree very well to fundamental expectations: it shows that an increase in both external and internal thermal mass results in a slower response to the step change but also in smaller heat flow oscillations during the transient period. Also, the thermal mass has been found to influence the heat flow transients more significantly if allocated internally. The time shift in the solution cycle has been found to influence both the time to reach steady-state and the transients and oscillations of the heat flow. A smaller time shift leads to significantly faster response.A comparison with experimental data shows that the dynamic agreement between experiment and simulation is very good with dynamic temperature deviations between 10 and 25s. The total time to achieve a new steady-state in hot water temperature after a 10K input temperature step amounts to approximately 15min. Compared to this, the present dynamic deviations are in the magnitude of approximately 1–3%.
Current duct design methods for variable air volume (VAV) systems are based on the use of peak constant airflow. However, VAV systems operate much of the time at an off-peak load condition and the impact of varying airflow rates to the sizing of duct systems has not been considered. This and a companion paper introduce an optimum duct design proce- dure for VAV systems to investigate the importance of the vary- ing airflows to the system design. Hourly airflow requirements, part-load fan characteristics, and duct static pressure control are incorporated into the problem formulation. Constraints, such as discrete duct sizes and velocity limitations, are incor- porated into the duct design procedure. In part 1, the domain of a VAV optimization problem is analyzed to define the prob- lem characteristics and to suggest an optimization procedure. In part 2, the VAV duct design procedure is fully developed and applied to several VAV duct systems with different parameter values. The results are analyzed to compare duct design meth- ods, and the effect of several factors that influence optimal design are investigated.
Current duct design methods for variable air volume (VAV) systems are based on the use of peak constant airflow. However, VAV systems operate much of the time at an off-peak load condition, and the impact of varying airflow rates to the sizing of duct systems has not been considered. This and a companion paper introduce an optimum duct design proce- dure for VAV systems to investigate the importance of the vary- ing airflows to the system design. Hourly airflow requirements, part-load fan characteristics, and duct static pressure control are incorporated into the problem formulation. Constraints, such as discrete duct sizes and velocity limitations, are incor- porated into the duct design procedure. In part 1, the domain of a VAV optimization problem is analyzed to define the prob- lem characteristics and to suggest an optimization procedure. In part 2, the VAV duct design procedure is fully developed and applied to several VAV duct systems with different parameter values. The results are analyzed to compare duct design meth- ods, and the effects of several factors that influence optimal design are investigated.