Harshalatha, Shantharam Patil, Pradeep G. Kini · Journal of Building Pathology and Rehabilitation · 2024
Abstract Improving the energy performance of buildings is crucial for environmental protection, energy savings, and a better living environment. The growing emphasis on sustainable building practices has led to an increased focus on optimizing space layout design parameters to enhance building energy performance. This review explores the application of simulation-based multi-objective optimization techniques in the context of studying the impact of space layout design on building energy efficiency. The integration of advanced simulation tools with optimization algorithms allows for a comprehensive analysis of multiple conflicting objectives like energy performance, user comfort as well as cost factor. The review begins by outlining the key parameters influencing building energy performance, including spatial configurations, orientation, and space perimeter variables. Subsequently, it delves into the various simulation tools employed to model the complex interactions between these parameters and their effects on energy performance. The integration of energy simulation software is highlighted as a crucial step towards achieving accurate and realistic assessments. In summary, this review delivers a comprehensive overview of the state-of-the-art methods in simulation-based multi-objective optimization for studying space layout design parameters and their impact on building energy performance, offering insights for researchers, practitioners, and policymakers in the field of sustainable architecture. There is a requirement for a comprehensive multi-objective framework for complex structures in the investigation of building energy performance giving more focus on reducing the cooling load and optimization of space layout along with envelope parameters.
Cited by 20Open access
Aejin Lee, Jiwon Gu, Yus Donald Chaniago et al. · Separation and Purification Technology · 2023
• Efficient extractive-dividing wall column for recovery of ultra-high-purity solvents from semiconductor waste solvents. • Optimization of extractive distillation and extractive-dividing wall column by applying multi-objective genetic algorithms. • Improvement of energy, exergy, economic, and environmental performances for extractive-dividing wall column. The waste solvent is frequently generated from the processes that highly rely on solvents. Diethylene glycol monomethyl ether or methyl di glycols (MDG) and N-methyl-2-pyrrolidone (NMP) are representative valuable solvents used broadly and removed as waste solvents during the semiconductor material manufacturing processes. Although waste solvent can be practically retrieved by distillation, azeotropic waste solvent mixture only can be recovered by advanced distillation process. In this study, optimal extractive distillation and extractive-dividing wall column are used to recover waste solvent. The process is optimized by multi-objective optimization using genetic algorithm by linking Aspen Plus® and MATLAB. All optimal cases are compared in terms of energy, exergy, economic and environmental parameter. As a result, the potential energy, total annual cost saving and exergy efficiency for extractive dividing wall column are 26.29%, 24.15% and 21.02%, respectively. Exergy loss that is associated with the number of trays can be significantly reduced by optimization while exergy loss that is associated with remixing only can be significantly reduced by dividing-wall column. Further, multi-objective optimization using a genetic algorithm and a range of population provides various results that determine process selection.
Cited by 12Open access
Jiaheng Tan, Wei Xu, Shuangtong Zhu et al. · Photonics · 2026
Aiming at the structural lightweight design of a 700 mm aperture primary mirror for a space camera, a novel success history-based adaptive multi-objective differential evolution algorithm with dynamic constraint handling is proposed to solve the multi-objective optimization problem of simultaneously minimizing mass and compliance under strict constraints for surface error and first-order modal frequency. Firstly, a surrogate model for the mirror was constructed using the Kriging algorithm based on Optimal Latin Hypercube Sampling, establishing a mapping relationship between input design variables and output responses, thereby replacing computationally expensive finite element simulations. Subsequently, a dynamic constraint adjustment mechanism was introduced into the Success History-based Adaptive Multi-Object Differential Evolution algorithm for the surrogate model, dynamically relaxing and tightening constraint violation requirements during iteration. This allows for utilizing promising yet infeasible solutions for rapid convergence while ensuring the feasibility of the final solutions. Comparisons with 13 advanced constrained multi-objective optimization algorithms demonstrate that the proposed algorithm exhibits excellent convergence, diversity, and consistency. Finally, the optimal solution was selected from the Pareto front obtained by the proposed algorithm, and the design variable values were adjusted according to manufacturing constraints to yield the final optimization result, which was then verified by finite element simulation. The simulation results show that the final mirror structure meets all performance constraints, demonstrating the effectiveness and engineering applicability of the proposed algorithm for the structural lightweight design of space camera mirrors.
Cited by 0Open access
Ken Chen, Wenyao Zhu · International Journal of Advanced Computer Science and Applications · 2024
The growing significance of energy-efficient building management techniques has led to research that combines precise heating demand predictions with sophisticated optimization algorithms. This research seeks a comprehensive solution to enhance building energy efficiency, addressing the growing concern for sustainability and responsible resource use in contemporary research and practice. In this research endeavor, the complex topic of energy optimization within the complex domain of heating, ventilation, and air conditioning (HVAC) systems is being tackled with a combination of creative problem-solving techniques and thorough examination. The significance of accurate heating load forecasts for raising HVAC system efficiency and cutting expenses is emphasized in this study. It introduces innovative methods by combining two advanced optimization algorithms, the Artificial Hummingbird Algorithm (AHA) and the Improved Arithmetic Optimization Algorithm (IAOA), with the Multi-Layer Perceptron (MLP) model. The main objective is to improve heating load forecast accuracy and expedite HVAC system optimization procedures. This study emphasizes how important precise heating load forecasts are to attaining energy efficiency, cost savings, and the ultimate objective of encouraging environmental sustainability in building management. The assessments unequivocally illustrate that the MLAH (Multi-Layer Perceptron with Artificial Hummingbird Algorithm) model in the second layer emerges as the most exceptional predictor. It attains an impressive maximum Coefficient of Determination (R2) value of 0.998 during the testing phase, reflecting a remarkable explanatory capacity and displaying remarkably low Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) values of 0.43 and 0.337, indicating minimal prediction discrepancies compared to alternative models.
Cited by 0Open access
Idan Distelfeld, Shmuel Osovski · Composite Structures · 2025
This paper presents a surrogate model-based approach for multi-objective optimization of composite representative volume elements under thermo-mechanical loading. The RVE architecture, inspired by metallic honeycomb structures with inclined fibers, allows tailoring the anisotropy of thermal and mechanical properties. A parametric model is analyzed using Finite Element Analysis with periodic boundary conditions and homogenization theory. The 10-dimensional design space is sampled using Latin Hypercube Sampling, and simulated to calculate effective elastic moduli and thermal conductivity. This dataset is used to train a shallow neural network (SNN) model, offering computational efficiency and rapid exploration of complex design spaces. The SNN is employed in a multi-objective optimization process using the NSGA-II algorithm, allowing simultaneous optimization of elastic properties, thermal conductivity, and density. This reveals trade-offs between competing objectives, with resulting Pareto frontiers providing crucial information for informed design decisions. The method demonstrates a fast, accurate, and flexible approach for optimizing composite architectures. Combining advanced modeling techniques with efficient optimization algorithms, this work contributes to developing lightweight, multifunctional materials for aerospace, automotive, and other demanding applications. The approach has significant implications for optimizing composite materials in complex structures, advancing the state-of-the-art in composite materials research and providing a powerful tool for high-performance material design. • Honeycomb-inspired RVE with tunable thermal and mechanical anisotropic properties. • Fast surrogate model enables instant exploration of 10-dimensional design space. • Multi-objective optimization using neural network reduces design iteration time. • Enables designers to optimize composite architecture without engineering support. • Designers can visualize property trade-o!s, enabling real-time design iterations.
Cited by 0Open access
Chiweta E. Abunike, Ogbonnaya I. Okoro, Ali Far et al. · Greater South Information System · 2023
Flux switching machines play a crucial role in emerging sectors such as renewable energy, electric vehicles, and the aerospace industry. Recent advancements have underscored their potential for achieving high torque density, effective heat management, and flux-weakening capabilities. This paper provides a comprehensive review of the design optimization of these machines, offering a unique contribution, as no previous attempts have delved into a detailed exploration of various optimization techniques specific to this technology. Special attention is given to state-of-the-art technologies related to multi-objective optimization and the performance analysis of the three variants: permanent magnet, wound-field, and hybrid excited. The review highlights and compares several optimization strategies from recent research, revealing research gaps in integrating advanced control algorithms, thermal, vibroacoustic, and structural models into the optimization design process of these machines. Furthermore, this review identifies essential trends in optimization development, offering valuable insights and future perspectives for these machines across diverse applications.
Cited by 0Open access
Joshbie Love De Guzman, Alliah Cel Villagomez, Edwin R. Arboleda · Preprints.org · 2024
This literature review offers a comprehensive analysis of the design and optimization of microstrip patch antennas for wireless communication systems, drawing insights from a variety of research papers. It examines diverse approaches aimed at enhancing antenna performance and functionality. Key topics explored include optimization of resonant frequency, design of antenna arrays, and bandwidth enhancement. Noteworthy optimization techniques, such as genetic algorithms and artificial neural networks, are discussed, highlighting their effectiveness in achieving superior antenna performance. Application-specific solutions tailored to areas such as military communication and RFID systems are emphasized. The review underscores the significant role played by optimization techniques in achieving desired performance metrics and identifies future research directions, including multi-objective optimization, investigation of robustness and reliability, and integration with advanced technologies. Overall, this literature review provides valuable insights into the state-of-the-art in microstrip patch antenna design and optimization, guiding future research endeavors in this dynamic and rapidly evolving field.
Cited by 0Open access
Chiweta E. Abunike, Ogbonnaya I. Okoro, Ali Far et al. · Greater South Information System · 2023
Flux switching machines play a crucial role in emerging sectors such as renewable energy, electric vehicles, and the aerospace industry. Recent advancements have underscored their potential for achieving high torque density, effective heat management, and flux-weakening capabilities. This paper provides a comprehensive review of the design optimization of these machines, offering a unique contribution, as no previous attempts have delved into a detailed exploration of various optimization techniques specific to this technology. Special attention is given to state-of-the-art technologies related to multi-objective optimization and the performance analysis of the three variants: permanent magnet, wound-field, and hybrid excited. The review highlights and compares several optimization strategies from recent research, revealing research gaps in integrating advanced control algorithms, thermal, vibroacoustic, and structural models into the optimization design process of these machines. Furthermore, this review identifies essential trends in optimization development, offering valuable insights and future perspectives for these machines across diverse applications.
Cited by 0Open access
Yucheng Yang, Zhou, Tianyi, Mykola; id_orcid 0000-0003-4955-0743 Pechenizkiy et al. · TU/e Research Portal · 2025
Practical reinforcement learning (RL) usually requires agents to be optimized for multiple potentially conflicting criteria, e.g. speed vs. safety. Although Multi-Objective RL (MORL) algorithms have been studied in previous works, their trained agents often cover limited Pareto optimal solutions and they lack precise controllability of the delicate trade-off among multiple objectives. Hence, the resulting agent is not versatile in aligning with customized requests from different users. To bridge the gap, we develop the “Preference controllable (PC) RL” framework, which trains a preference-conditioned meta-policy that takes user preference as input controlling the generated trajectories within the preference region on the Pareto frontier. PCRL is compatible with advanced Multi-Objective Optimization (MOO) algorithms that are rarely seen in previous MORL approaches. We also proposed a novel preference-regularized MOO algorithm specifically for PCRL. We provide a comprehensive theoretical analysis to justify its convergence and preference controllability. We evaluate PCRL with different MOO algorithms against state-of-the-art MORL baselines in various challenging environments with up to six objectives. In these experiments, our proposed method exhibits significantly better controllability than existing approaches and can generate Pareto solutions with better diversity and utilities.
Cited by 0Open access
Sanjai Pathak, Ashish Mani, Amlan Chatterjee · 2025
The tuning of Advanced Driver Assistance Systems (ADAS) involves resolving trade-offs among several competing objectives, including operational safety, system responsiveness, energy usage, and passenger comfort. This work introduces a novel optimization framework based on Quantum-Inspired Hybrid Swarm Intelligence (QiHSI), in which quantum-inspired mechanisms are embedded within a multi-objective salp swarm optimization process to strengthen global search capability and preserve population diversity in complex, high-dimensional decision spaces. In addition, a decision-maker-in-the-loop strategy is incorporated to incorporate adaptive expert guidance, enabling the optimization process to respond dynamically to changing design priorities and system constraints. The effectiveness of QiHSI is assessed using established multi-objective benchmark problems as well as a practical ADAS calibration scenario. Experimental comparisons with several state-of-the-art evolutionary and swarm-based algorithms, including MSSA, MOPSO, MOEA/D, SPEA2, NSGA-III, and RVEA, show that the proposed method consistently produces well-distributed Pareto-optimal solutions with faster convergence and improved adaptability. These findings demonstrate that QiHSI offers a reliable and scalable approach for intelligent ADAS calibration, supporting the development of more responsive, efficient, and safety-oriented autonomous driving technologies.
Cited by 0
Sanjai Pathak, Ashish Mani, Amlan Chatterjee · ArXiv.org · 2026
The tuning of Advanced Driver Assistance Systems (ADAS) involves resolving trade-offs among several competing objectives, including operational safety, system responsiveness, energy usage, and passenger comfort. This work introduces a novel optimization framework based on Quantum-Inspired Hybrid Swarm Intelligence (QiHSI), in which quantum-inspired mechanisms are embedded within a multi-objective salp swarm optimization process to strengthen global search capability and preserve population diversity in complex, high-dimensional decision spaces. In addition, a decision-maker-in-the-loop strategy is incorporated to incorporate adaptive expert guidance, enabling the optimization process to respond dynamically to changing design priorities and system constraints. The effectiveness of QiHSI is assessed using established multi-objective benchmark problems as well as a practical ADAS calibration scenario. Experimental comparisons with several state-of-the-art evolutionary and swarm-based algorithms, including MSSA, MOPSO, MOEA/D, SPEA2, NSGA-III, and RVEA, show that the proposed method consistently produces well-distributed Pareto-optimal solutions with faster convergence and improved adaptability. These findings demonstrate that QiHSI offers a reliable and scalable approach for intelligent ADAS calibration, supporting the development of more responsive, efficient, and safety-oriented autonomous driving technologies.
Cited by 0Open access
Sanjai Pathak, Ashish Mani, Amlan Chatterjee · arXiv (Cornell University) · 2026
The tuning of Advanced Driver Assistance Systems (ADAS) involves resolving trade-offs among several competing objectives, including operational safety, system responsiveness, energy usage, and passenger comfort. This work introduces a novel optimization framework based on Quantum-Inspired Hybrid Swarm Intelligence (QiHSI), in which quantum-inspired mechanisms are embedded within a multi-objective salp swarm optimization process to strengthen global search capability and preserve population diversity in complex, high-dimensional decision spaces. In addition, a decision-maker-in-the-loop strategy is incorporated to incorporate adaptive expert guidance, enabling the optimization process to respond dynamically to changing design priorities and system constraints. The effectiveness of QiHSI is assessed using established multi-objective benchmark problems as well as a practical ADAS calibration scenario. Experimental comparisons with several state-of-the-art evolutionary and swarm-based algorithms, including MSSA, MOPSO, MOEA/D, SPEA2, NSGA-III, and RVEA, show that the proposed method consistently produces well-distributed Pareto-optimal solutions with faster convergence and improved adaptability. These findings demonstrate that QiHSI offers a reliable and scalable approach for intelligent ADAS calibration, supporting the development of more responsive, efficient, and safety-oriented autonomous driving technologies.
Cited by 0Open access
Bach Do, Ruda Zhang · arXiv (Cornell University) · 2023
Resided at the intersection of multi-fidelity optimization (MFO) and Bayesian optimization (BO), MF BO has found a niche in solving expensive engineering design optimization problems, thanks to its advantages in incorporating physical and mathematical understandings of the problems, saving resources, addressing exploitation-exploration trade-off, considering uncertainty, and processing parallel computing. The increasing number of works dedicated to MF BO suggests the need for a comprehensive review of this advanced optimization technique. In this paper, we survey recent developments of two essential ingredients of MF BO: Gaussian process (GP) based MF surrogates and acquisition functions. We first categorize the existing MF modeling methods and MFO strategies to locate MF BO in a large family of surrogate-based optimization and MFO algorithms. We then exploit the common properties shared between the methods from each ingredient of MF BO to describe important GP-based MF surrogate models and review various acquisition functions. By doing so, we expect to provide a structured understanding of MF BO. Finally, we attempt to reveal important aspects that require further research for applications of MF BO in solving intricate yet important design optimization problems, including constrained optimization, high-dimensional optimization, optimization under uncertainty, and multi-objective optimization.
Cited by 3Open access
Xinran Liu, Yaqiong Liu, Guochu Shou et al. · IEEE Transactions on Vehicular Technology · 2026
The rapid development of the Internet of Vehicles (IoV) has driven the need for efficient Integration of Communication, Sensing, and Computation (ICSC) to support advanced applications such as autonomous driving and real-time traffic management. However, the dynamic and resource-constrained nature of IoV environment poses significant challenges in achieving optimal performance, particularly in balancing multiple optimization objectives such as latency and energy consumption. Traditional optimization methods, such as Genetic Algorithms (GA), often struggle with inefficiency and unpredictability in dynamic scenarios. In this paper, we first propose an IoV Communication-Sensing-Computation Integration architecture to address the problem that the traditional IoV architectures cannot meet the needs of the ICSC. Then, we formulate the joint optimization of latency and energy consumption in ICSC as a constrained Multi-objective Optimization Problem (MOP). We employ Multi-Agent Deep Reinforcement Learning (MADRL) with a transformer module, combined with a weighted decomposition method and a neighborhood parameter transfer technique, to solve the problem. Simulation results show that our approach outperforms traditional methods in terms of convergence and diversity, and has a lower running time.
Cited by 0
Bierlaire, Michel, Thémans, Michaël · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · 2005
Discrete choice models have received a great amount of attention in the last years.Recent advances have proposed new models in the Generalized Extreme Value (GEV) family, mixed Logit models and even mixed GEV models.Estimating those models becomes more and more problematic.The objective function becomes highly nonlinear and non concave and the complexity of the model forces to impose constraints (maybe nonlinear) on the parameters in order to obtain meaningful values or to overcome model overspecification which can lead to singularity in the objective function.State-of-the-art algorithms can no longer be applied and specific optimization algorithms must be developed for the estimation of advanced discrete choice models.In particular, we need optimization algorithms able to deal with singularity (possibly nonlinear).Our first step toward this goal is to investigate the case of unconstrained optimization when an affine singularity arises in the objective function.First, in order to get a robust method in the presence of a singularity, we propose to perform an eigen-structure analysis on the second derivatives matrix of the objective function which allows us to characterize the subspace in which the singularity lies.Second, once the singularity has been properly identified, we fix this singularity by adding constraints describing it in order to better guide the algorithm toward a local solution of the optimization problem.The difficulty here is that the identification of the singularity is an iterative process taking place within the optimization algorithm.Therefore, those constraints must be included "on-the-fly" in the optimization problem.This second part is achieved using a multidimensional filter which is an algorithmic tool coming from multi-criteria optimization.In our context, we want to maximize the log-likelihood function but also to satisfy the constraints associated with the singularity.In this paper, we present preliminary numerical results with the current version of our algorithm designed to solve singular unconstrained optimization problems.We show that our methods should significantly decrease the model estimation time.We also discuss the future modifications to generalize our algorithm to deal with singular constrained optimization.
Cited by 1Open access
Zaman, Dewan Maisha · Journal of the Arkansas Academy of Science · 2025
This research examines the problem of estimating the survival signature of all-terminal networks using Monte Carlo (MC) simulation. Following a recent similar result for twoterminal networks, we show that the work required within each MC replication corresponds to solving a multi-objective bottleneck spanning tree (MOBST) problem. We implement the resulting MC procedure using a “Blocks” algorithm from the literature to solve the MOBST in each replication by identifying its minimal set of non-dominated points. We compare this implementation against intuitive benchmark procedures for completing the work within an MC replication. We conduct numerical experiments to assess the efficacy of multi-objective optimization algorithms in expediting survival signature estimation for network structures. Our focus is to determine if these advanced computational strategies can significantly reduce the time complexity inherent in large-scale network analyses.
Cited by 0Open access
Chih-Chang Lin, Yuxuan Li, Y. Yang · 2025
Constrained multi-objective optimization problems are common in various real-world applications, and the dual-population framework is preferred for its excellent performance in addressing such issues. In the dual-population algorithms, if the connection between the two populations is too tight, it is not conducive to the iteration of the population; On the contrary, the advantages of the dual-population mechanism cannot be fully developed. This paper proposes a dual-population constrained multi-objective evolutionary algorithm based on a dynamic contact mechanism (DPDC) to address these limitations. The DPDC algorithm introduces a variable (ncse) to regulate the number of individuals from the auxiliary population in the offspring. Initially, the populations exhibit a strong degree of interaction. The auxiliary population promotes the movement of the main population toward the Pareto front, thereby enhancing the convergence of the main population. In later stages, the offspring includes fewer individuals from the auxiliary population. Consequently, the distribution of the main population gradually improves. This mechanism effectively balances convergence and diversity. To prove the algorithm's effectiveness, it is tested on the benchmark test problem set DASCMOP, and the test results verify that the DPDC algorithm proposed in this paper outperforms five advanced constrained multi-objective evolutionary algorithms.
Cited by 0
Kianoosh Sattari · 2024
This dissertation presents a comprehensive exploration of scientific machine learning methodologies applied to various aspects of material science and additive manufacturing. Chapter 2 introduces a scientific machine learning framework tailored to understand the synthesis process of flash graphene. Leveraging advanced algorithms and data-driven approaches, this framework facilitates a deeper comprehension of the intricate mechanisms involved in flash graphene synthesis, thereby offering valuable insights for optimization and enhancement. In Chapter 3, physics-informed machine learning models are developed for the classification of printability and glass transition temperature (Tg) in additive manufacturing processes. By integrating fundamental principles of physics into the machine learning models, this chapter demonstrates improved accuracy and reliability in predicting printability and Tg. Chapter 4 focuses on physics-constrained multi-objective Bayesian optimization techniques to expedite the 3D printing of thermoplastics. Through the utilization of Bayesian optimization algorithms that incorporate physical constraints, this chapter presents a systematic approach to accelerate the optimization process while maintaining the integrity and stability of printed structures. Collectively, these chapters contribute to the advancement of scientific machine learning methodologies in material science and additive manufacturing, offering novel insights, techniques, and tools for enhancing process understanding, optimization, and control.
Cited by 0
Karthick Manikandan, P. Venkatesh, Naga Pratyusha B et al. · International Journal of Electrical and Electronics Engineering · 2025
In recent years, the focus on optimal scheduling methods for Microgrids (MGs) has intensified due to their ability to efficiently manage Distributed Generation (DG) over varying time intervals. The intermittent nature of renewable energy sources like wind and solar creates significant challenges for economic dispatch within MGs, as their unpredictability complicates coordination among energy sources. This paper presents a novel, comprehensive framework for multi-objective optimal dispatch in MGs, considering both ecological and economic factors. The framework integrates various generating units, including Photovoltaic (PV) systems, Wind Turbines (WT), Microturbines (MT), Fuel Cells (FC), and Battery Storage (BT) systems. To address the complexities of balancing multiple objectives and the stochastic nature of renewable energy, the research aims to develop a feasible multi-objective optimal dispatch strategy that improves MG operations by making them faster, more stable, and more efficient in convergence. Advanced optimization techniques, particularly intelligent algorithms, are utilized within a robust simulation environment to tackle the challenges. The study develops and partially tests a mathematical model for multi-objective optimal dispatch, facilitating dynamic parameter calculations to ensure efficient system operations despite renewable energy complexities. This research demonstrates the potential for a highly efficient, sustainable MG dispatch system using intelligent optimization, providing insights for future experimental implementations and refinements.
Cited by 0Open access
Huaiyu Guo, Zhiyu Feng, Xiao Wang · 2023
High stable and high reliable DNA sequences are requisite in DNA computing search field. DNA sequences can be quantified in an array of thermodynamical and logical constraints which play an important role in process of hybridization. Since designing of DNA sequences set regraded as a multi-objective optimization problem due to these constraints, epsilon constraint introduced in this paper which is a thought of reduce object dimensions. These constraints can be classified in single squence constraints and multiple sequences constraints, two meta heuristic algorithms tunicate swarm algorithm and grasshopper optimisation algorithm used in different steps according to their features and applicable scenarios. Result of this progress shows that epsilon constraints collaborate with other algorithms may produce advanced DNA sequences set compare to previous work.
Cited by 1