Vassilios S. Vassiliadis's scientific contributions

Publications (9)

Article
Full-text available
This article proposes a novel method to optimise the Dynamic Architecture Neural Network (DAN2) adapted for a multi-task learning problem. The multi-task learning neural network adopts a multi-head and serial architecture with DAN2 layers acting as the basic subroutine. Adopting a dynamic architecture, the layers are added consecutively starting fr...
Article
Full-text available
Industry 4.0 aims to transform chemical and biochemical processes into intelligent systems via the integration of digital components with the actual physical units involved. This process can be thought of as the addition of a central nervous system with a sensing and control monitoring of components and regulating the performance of the individual...
Conference Paper
A first contribution of this paper is an overview of the research efforts and contributions over several decades in the area of scheduling maintenance optimization for decaying performance dynamic processes. Following breakthrough ideas and implementation in the area of heat exchanger networks for optimal scheduling of cleaning actions subject to e...
Article
Full-text available
This paper presents a framework for the use of variable pricing to control electricity im-ported/exported to/from both fixed and unfixed residential distributed energy resource (DER) network designs. The framework shows that networks utilizing much of their own energy, and importing little from the national grid, are barely affected by dynamic impo...
Article
This paper presents a novel optimisation method, termed Hessian-free Gradient Flow, for the optimisation of deep neural networks. The algorithm entails the design characteristics of the Truncated Newton, Conjugate Gradient and Gradient Flow method. It employs a finite difference approximation scheme to make the algorithm Hessian-free and makes use...
Article
This paper presents two nonlinear model predictive control based methods for solving closed-loop stochastic dynamic optimisation problems, ensuring both robustness and feasibility with respect to state output constraints. The first one is a new deterministic approach, using the wait-and-see strategy. The key idea is to specifically anticipate viola...
Article
Full-text available
A modification to the mixed-integer nonlinear programming (MINLP) formulation for symbolic regression was proposed with the aim of identification of physical models from noisy experimental data. In the proposed formulation, a binary tree in which equations are represented as directed, acyclic graphs, is fully constructed for a pre-defined number of...
Article
Thermal runaways in exothermic batch reactors present major safety and economic issues for industry. Control systems currently used are not capable of detecting thermal runaway behaviour and achieve nominally safe operation by carrying out the reaction at a low temperature. Recently, improvements in safety and process intensity have been achieved b...
Preprint
p>A mixed-integer nonlinear programming (MINLP) formulation for symbolic regression was proposed to identify physical models from noisy experimental data. The formulation was tested using numerical models and was found to be more efficient than the previous literature example with respect to the number of predictor variables and training data point...

Citations

... Further improvements could consider different import and export prices for the energy exchanged with the grid [45]. Thus, the benefits of the presence of the storage options ...
... e primary concern is data; in natural language processing, it is the corpus [25]. Second, corpus must be carefully screened in terms of quality and data scale to ensure that the results generated have sufficient accuracy [26]. Due to the large capacity of computer storage, authentic text data, and fast and accurate information extraction, linguists can use the electronic corpus to describe language from multiple angles and levels, verify various language theories and hypotheses, and even establish new language models and language concept. ...
... Among these systems, non-linear dynamical systems involving uncertainties and unknown disturbances exist on a very large scale. One can find such systems very commonly in various engineering and industrial control fields such as electrical networks (Aslam & Dai, 2020;Demenko et al., 2015;Mielczarski, 1988;Taylor, 1994), microwave oscillators (Chembo et al., 2008;Hoeye et al., 2004;Vahdati & Abdipour, 2008), turbojet engines (Andoga et al., 2019;Han, 2018), hydraulic systems (Ba et al., 2020;Sirouspour & Salcudean, 2001;Yu et al., 2004;Zhong et al., 2021), robotic systems (Ibrir et al., 2005), nuclear reactors (Dong, 2010;Ibrir et al., 2005;Wahi & Kumawat, 2011), chemical reactors (Arellano-Garcia et al., 2020;Badillo-Hernandez et al., 2019;García-Sandoval et al., 2008), etc. which possess the attributes of such uncertainties and unknown disturbances. Because of being nonlinear and possessing such attributes of uncertainties and unknown disturbances, the process of designing a closedloop control for these systems becomes very difficult. ...
... 48 Recently, an MINLP for symbolic regression successfully recovered the relationship between shear stress and shear rate for both Newtonian and non-Newtonian fluids and chemical reactions kinetic laws. 49 Ansari and colleagues investigated relationships between variables in computational fluid dynamics simulations combining artificial intelligence and symbolic regression using sureindependence screening and sparsifying operators. 50,51 Lastly, linear sparse regression techniques, such as LASSO or elastic nets, can be deployed as an alternative to MI(N)LP formulations. ...
... trajectory control, MPC). [3][4][5][6][7][8][9] In recent years, process industries have invested in machine learning teams, software, and infrastructure due to the promise of data-driven applications in manufacturing. [10][11][12][13][14][15][16][17] Unlike big tech companies -on which online recommendation systems allow quick iterations by trial-and-error-, manufacturing industries must deal with the safety of such recommendations and the inevitable challenges imposed by the physicochemical, engineering, and operation constraints. ...