1. Integrating Multi-Physics Modeling within Multi-Objective Optimization to Enhance the Performance and Efficiency of Permanent Magnet Synchronous Machines : doktorska disertacijaMitja Garmut, 2025, doctoral dissertation Abstract: This Dissertation focuses on the optimization of an Interior Permanent Magnet (IPM) machine for handheld battery-powered tools, aiming to enhance performance and efficiency.
The research integrates multi-physics modeling, including electromagnetic Finite Element Method (FEM) and thermal models, to evaluate machine performance under various operating conditions. The performance is evaluated according to selected Key Performance Indicators (KPIs). Further, different control methods, such as Field Oriented Control and Square-Wave Control, impact the performance significantly and are incorporated into the optimization process.
Due to the computational challenges of FEM-based performance evaluations in Multi-Objective Optimization (MOO), this work utilizes Artificial Neural Network (ANN)-based meta-models, to accelerate the optimization process while preserving accuracy.
The developed meta-models capture nonlinear machine characteristics from the FEM model. These meta-models are then used to evaluate machine performance through a combination of analytical and numerical post-processing methods.
Four MOO scenarios are presented, each aimed at optimizing the cross-sectional design of IPM machines, to enhance performance and efficiency while reducing mass and cost. Additionally, these scenarios modify the machine’s electromagnetic behavior, to ensure better alignment with the selected control method.
By comparing the optimization process of Scenario 1, which uses direct FEM-based evaluation without time reduction measures, to the approach incorporating Artificial Neural Network based meta-models, the total number of individual FEM evaluations decreased from 2.35×10^9 to 2.03×10^5, without almost any loss of accuracy. This reduced the computation time from 297 years to 9.07 days on our standard desktop computer. The obtained ANN-base meta-models can be used further for other optimizations without the need for additional FEM evaluations.
In all four optimization scenarios, the use of meta-models enabled the generation of a Pareto front of the optimal solutions, leading to improved KPIs compared to the reference design. The highest relative improvement occurred in Scenario 1, where the selected optimized machine design achieved a 30% increase in power density compared to the reference design. Keywords: Interior Permanent Magnet (IPM) Machine, Artificial Neural Network (ANN), Meta-Modeling, Multi-Objective Optimization (MOO), Finite Element Method (FEM), Multi-Physics Modeling, Field Oriented Control (FOC), Square-Wave Control (SWC) Published in DKUM: 15.05.2025; Views: 0; Downloads: 37
Full text (17,79 MB) |
2. The MINLP approach to topology, shape and discrete sizing optimization of trussesSimon Šilih, Zdravko Kravanja, Stojan Kravanja, 2022, original scientific article Abstract: The paper presents the Mixed-Integer Non-linear Programming (MINLP) approach to the
synthesis of trusses. The solution of continuous/discrete non-convex and non-linear optimization
problems is discussed with respect to the simultaneous topology, shape and discrete sizing optimization of trusses. A truss MINLP superstructure of different topology and design alternatives
has been generated, and a special MINLP model formulation for trusses has been developed. In the
optimization model, a mass objective function of the structure has been defined and subjected to
design, load and dimensioning constraints. The MINLP problems are solved using the Modified
Outer-Approximation/Equality-Relaxation (OA/ER) algorithm. Multi-level MINLP strategies are introduced to accelerate the convergence of the algorithm. The Modified Two-Phase and the Sequential
Two-Phase MINLP strategies are proposed in order to solve highly combinatorial topology, shape
and discrete sizing optimization problems. The importance of local buckling constraints on topology
optimization is also discussed. Some simple numerical examples are shown at the end of the paper to
demonstrate the suitability and efficiency of the proposed method. Keywords: structural synthesis, topology optimization, discrete sizing optimization, mixed-integer non-linear programming, MINLP, modified OA/ER algorithm, multi-level MINLP strategies, steel structures, trusses Published in DKUM: 11.03.2025; Views: 0; Downloads: 3
Full text (2,49 MB) This document has many files! More... |
3. Multi-objective scenario optimization of the food supply chain – Slovenian case studyJan Drofenik, Bojan Pahor, Zdravko Kravanja, Zorka Novak-Pintarič, 2023, original scientific article Keywords: self-sufficiency, food supply chain, multi-objective optimization, mixed integer linear programming, sustainable development Published in DKUM: 14.02.2025; Views: 0; Downloads: 5
Full text (957,78 KB) |
4. Active BIM system for optimized multi-project ready-mix-concrete deliveryHana Begić, Mario Galić, Uroš Klanšek, 2024, original scientific article Abstract: Purpose – Ready-mix concrete delivery problem (RMCDP), a specific version of the vehicle routing problem (VRP), is a relevant supply-chain engineering task for construction management with various formulations and solving methods. This problem can range from a simple scenario involving one source, one material and one destination to a more challenging and complex case involving multiple sources, multiple materials and multiple destinations. This paper presents an Internet of Things (IoT)-supported active building information modeling (BIM) system for optimized multi-project ready-mix concrete (RMC) delivery. Design/methodology/approach – The presented system is BIM-based, IoT supported, dynamic and automatic input/output exchange to provide an optimal delivery program for multi-project ready-mix-concrete problem. The input parameters are extracted as real-time map-supported IoT data and transferred to the system via an application programming interface (API) into a mixed-integer linear programming (MILP) optimization model developed to perform the optimization. The obtained optimization results are further integrated into BIM by conventional project management tools. To demonstrate the features of the suggested system, an RMCDP example was applied to solve that included four building sites, seven eligible concrete plants and three necessary RMC mixtures. Findings – The system provides the optimum delivery schedule for multiple RMCs to multiple construction sites, as well as the optimum RMC quantities to be delivered, the quantities from each concrete plant that must be supplied, the best delivery routes, the optimum execution times for each construction site, and the total minimal costs, while also assuring the dynamic transfer of the optimized results back into the portfolio of multiple BIM projects. The system can generate as many solutions as needed by updating the real-time input parameters in terms of change of the routes, unit prices and availability of concrete plants. Originality/value – The suggested system allows dynamic adjustments during the optimization process, andis adaptable to changes in input data also considering the real-time input data. The system is based on spreadsheets, which are widely used and common tool that most stakeholders already utilize daily, while also providing the possibility to apply a more specialized tool. Based on this, the RMCDP can be solved using both conventional and advanced optimization software, enabling the system to handle even large-scale tasks as necessary. Keywords: active building information modeling, BIM, internet of things, IoT, multi-project environment, optimization, ready-mix-concrete delivery, RMC Published in DKUM: 11.09.2024; Views: 32; Downloads: 15
Full text (3,94 MB) This document has many files! More... |
5. A multi-objective solution of green vehicle routing problemÖzgür Kabadurmuş, Mehmet Serdar Erdoğan, Yiğitcan Özkan, Mertcan Köseoğlu, 2019, original scientific article Abstract: Distribution is one of the major sources of carbon emissions and this issue has been addressed by Green Vehicle Routing Problem (GVRP). This problem aims to fulfill the demand of a set of customers using a homogeneous fleet of Alternative Fuel Vehicles (AFV) originating from a single depot. The problem also includes a set of Alternative Fuel Stations (AFS) that can serve the AFVs. Since AFVs started to operate very recently, Alternative Fuel Stations servicing them are very few. Therefore, the driving span of the AFVs is very limited. This makes the routing decisions of AFVs more difficult. In this study, we formulated a multi-objective optimization model of Green Vehicle Routing Problem with two conflicting objective functions. While the first objective of our GVRP formulation aims to minimize total CO2 emission, which is proportional to the distance, the second aims to minimize the maximum traveling time of all routes. To solve this multi-objective problem, we used �-constraint method, a multi-objective optimization technique, and found the Pareto optimal solutions. The problem is formulated as a Mixed-Integer Linear Programming (MILP) model in IBM OPL CPLEX. To test our proposed method, we generated two hypothetical but realistic distribution cases in Izmir, Turkey. The first case study focuses on an inner-city distribution in Izmir, and the second case study involves a regional distribution in the Aegean Region of Turkey. We presented the Pareto optimal solutions and showed that there is a tradeoff between the maximum distribution time and carbon emissions. The results showed that routes become shorter, the number of generated routes (and therefore, vehicles) increases and vehicles visit a lower number of fuel stations as the maximum traveling time decreases. We also showed that as maximum traveling time decreases, the solution time significantly decreases. Keywords: green vehicle routing problem, alternative fuel vehicles, epsilon-constraint, multi-objective optimization, Pareto optimality Published in DKUM: 22.08.2024; Views: 44; Downloads: 9
Full text (764,60 KB) This document has many files! More... |
6. Sustainable design of circular reinforced concrete column sections via multi-objective optimizationPrimož Jelušič, Tomaž Žula, 2023, original scientific article Abstract: An optimization model for reinforced concrete circular columns based on the Eurocodes is presented. With the developed optimization model, which takes into account the exact distribution of the steel reinforcement, which is not the case when designing with conventional column design charts, an optimal design for the reinforced concrete cross section is determined. The optimization model uses discrete variables, which makes the results more suitable for actual construction practice and fully exploits the structural capacity of the structure. A parametric study of the applied axial load and bending moment was performed for material cost and CO2 emissions. The results based on a single objective function show that the optimal design of the reinforced concrete column cross section obtained for the material cost objective function contains a larger cross-sectional area of concrete and a smaller area of steel compared with the optimization results when CO2 emissions are determined as the objective function. However, the optimal solution in the case where the material cost was assigned as the objective function has much more reserve in axial load capacity than in the optimal design where CO2 was chosen as the objective function. In addition, the multi-objective optimization was performed to find a set of solutions that provide the best trade-offs between the material cost and CO2 emission objectives. Keywords: reinforced concrete columns, circular cross section, costs, CO2 emissions, multi-objective optimization, genetic algorithm Published in DKUM: 15.04.2024; Views: 338; Downloads: 218
Full text (4,56 MB) This document has many files! More... |
7. A graph pointer network-based multi-objective deep reinforcement learning algorithm for solving the traveling salesman problemJeewaka Perera, Shih-Hsi Liu, Marjan Mernik, Matej Črepinšek, Miha Ravber, 2023, original scientific article Abstract: Traveling Salesman Problems (TSPs) have been a long-lasting interesting challenge to researchers in different areas. The difficulty of such problems scales up further when multiple objectives are considered concurrently. Plenty of work in evolutionary algorithms has been introduced to solve multi-objective TSPs with promising results, and the work in deep learning and reinforcement learning has been surging. This paper introduces a multi-objective deep graph pointer network-based reinforcement learning (MODGRL) algorithm for multi-objective TSPs. The MODGRL improves an earlier multi-objective deep reinforcement learning algorithm, called DRL-MOA, by utilizing a graph pointer network to learn the graphical structures of TSPs. Such improvements allow MODGRL to be trained on a small-scale TSP, but can find optimal solutions for large scale TSPs. NSGA-II, MOEA/D and SPEA2 are selected to compare with MODGRL and DRL-MOA. Hypervolume, spread and coverage over Pareto front (CPF) quality indicators were selected to assess the algorithms’ performance. In terms of the hypervolume indicator that represents the convergence and diversity of Pareto-frontiers, MODGRL outperformed all the competitors on the three well-known benchmark problems. Such findings proved that MODGRL, with the improved graph pointer network, indeed performed better, measured by the hypervolume indicator, than DRL-MOA and the three other evolutionary algorithms. MODGRL and DRL-MOA were comparable in the leading group, measured by the spread indicator. Although MODGRL performed better than DRL-MOA, both of them were just average regarding the evenness and diversity measured by the CPF indicator. Such findings remind that different performance indicators measure Pareto-frontiers from different perspectives. Choosing a well-accepted and suitable performance indicator to one’s experimental design is very critical, and may affect the conclusions. Three evolutionary algorithms were also experimented on with extra iterations, to validate whether extra iterations affected the performance. The results show that NSGA-II and SPEA2 were greatly improved measured by the Spread and CPF indicators. Such findings raise fairness concerns on algorithm comparisons using different fixed stopping criteria for different algorithms, which appeared in the DRL-MOA work and many others. Through these lessons, we concluded that MODGRL indeed performed better than DRL-MOA in terms of hypervolumne, and we also urge researchers on fair experimental designs and comparisons, in order to derive scientifically sound conclusions. Keywords: multi-objective optimization, traveling salesman problems, deep reinforcement learning Published in DKUM: 28.03.2024; Views: 184; Downloads: 32
Full text (7,89 MB) This document has many files! More... |
8. Accuracy is not enough: optimizing for a fault detection delayMatej Šprogar, Domen Verber, 2023, original scientific article Abstract: This paper assesses the fault-detection capabilities of modern deep-learning models. It highlights that a naive deep-learning approach optimized for accuracy is unsuitable for learning fault-detection models from time-series data. Consequently, out-of-the-box deep-learning strategies may yield impressive accuracy results but are ill-equipped for real-world applications. The paper introduces a methodology for estimating fault-detection delays when no oracle information on fault occurrence time is available. Moreover, the paper presents a straightforward approach to implicitly achieve the objective of minimizing fault-detection delays. This approach involves using pseudo-multi-objective deep optimization with data windowing, which enables the utilization of standard deep-learning methods for fault detection and expanding their applicability. However, it does introduce an additional hyperparameter that needs careful tuning. The paper employs the Tennessee Eastman Process dataset as a case study to demonstrate its findings. The results effectively highlight the limitations of standard loss functions and emphasize the importance of incorporating fault-detection delays in evaluating and reporting performance. In our study, the pseudo-multi-objective optimization could reach a fault-detection accuracy of 95% in just a fifth of the time it takes the best naive approach to do so. Keywords: artificial neural networks, deep learning, fault detection, accuracy, multi-objective optimization Published in DKUM: 30.11.2023; Views: 363; Downloads: 29
Full text (478,93 KB) This document has many files! More... |
9. An efficient metaheuristic algorithm for job shop scheduling in a dynamic environmentHankun Zhang, Borut Buchmeister, Xueyan Li, Robert Ojsteršek, 2023, original scientific article Keywords: metaheuristic algorithm, improved multi-phase particle swarm optimization, cellular neighbor network, dynamic job shop scheduling, simulation modelling Published in DKUM: 19.05.2023; Views: 527; Downloads: 49
Full text (5,62 MB) This document has many files! More... |
10. Multi-objective optimization algorithms with the island metaheuristic for effective project management problem solvingChristina Brester, Ivan Ryzhikov, Eugene Semenkin, 2017, original scientific article Abstract: Background and Purpose: In every organization, project management raises many different decision-making problems, a large proportion of which can be efficiently solved using specific decision-making support systems. Yet such kinds of problems are always a challenge since there is no time-efficient or computationally efficient algorithm to solve them as a result of their complexity. In this study, we consider the problem of optimal financial investment. In our solution, we take into account the following organizational resource and project characteristics: profits, costs and risks.
Design/Methodology/Approach: The decision-making problem is reduced to a multi-criteria 0-1 knapsack problem. This implies that we need to find a non-dominated set of alternative solutions, which are a trade-off between maximizing incomes and minimizing risks. At the same time, alternatives must satisfy constraints. This leads to a constrained two-criterion optimization problem in the Boolean space. To cope with the peculiarities and high complexity of the problem, evolution-based algorithms with an island meta-heuristic are applied as an alternative to conventional techniques.
Results: The problem in hand was reduced to a two-criterion unconstrained extreme problem and solved with different evolution-based multi-objective optimization heuristics. Next, we applied a proposed meta-heuristic combining the particular algorithms and causing their interaction in a cooperative and collaborative way. The obtained results showed that the island heuristic outperformed the original ones based on the values of a specific metric, thus showing the representativeness of Pareto front approximations. Having more representative approximations, decision-makers have more alternative project portfolios corresponding to different risk and profit estimations. Since these criteria are conflicting, when choosing an alternative with an estimated high profit, decision-makers follow a strategy with an estimated high risk and vice versa.
Conclusion: In the present paper, the project portfolio decision-making problem was reduced to a 0-1 knapsack constrained multi-objective optimization problem. The algorithm investigation confirms that the use of the island meta-heuristic significantly improves the performance of genetic algorithms, thereby providing an efficient tool for Financial Responsibility Centres Management. Keywords: 0-1 multi-objective constrained knapsack problem, project management portfolio problem, multi-objective evolution-based optimization algorithms, collaborative and cooperative meta-heuristics Published in DKUM: 04.05.2018; Views: 1571; Downloads: 320
Full text (993,98 KB) This document has many files! More... |