Multi-objective Optimization Framework Research Articles

Multi-objective design optimization of hermetically sealed core-type distribution transformer considering current harmonics of power grid using NSGA III

In this paper, the design optimization of hermetically sealed oil-immersed distribution transformers is studied. For the first time in the design of this type of transformer, harmonics of the power grid affecting the transformer design, along with all design variables, are included, and their boundaries suitable for the production line are determined. The design probability (search space) for a total of 21 design variables is 1.32×1025. The Brute Force method is not a time-efficient approach to finding the optimal result in this search space. Due to the mixed integer nonlinear programming (MINLP) nature of the transformer design problem, heuristic optimization algorithms are selected because of their shorter solution time and near-optimal results. Considering the principle of the “No Free Lunch” theorem, extensive evaluations were conducted on various heuristic algorithms to identify the most effective ones for addressing the transformer design problem. The most successful algorithms for solving the problem were identified after thorough evaluation, considering three distinct scenarios within a multi-objective optimization framework. These algorithms are namely MOEA/D, MOGWO, MOWOA, and NSGA III. In this study, an optimization tool is developed for the design optimization of hermetically sealed distribution transformers, covering a range from 50 kVA to 3150 kVA and encompassing all design variables. The developed optimization program is fully compatible with the production line and can be easily adjusted to meet all national and international standards, such as IEC and IEEE, as well as customer specifications.

Multi-objective optimization framework for deepwater riser jetting installation parameters using deep reinforcement learning

Optimizing conductor casing jetting installation in offshore oil and gas exploration is essential for underwater wellhead reliability. Traditional models often struggle with the complexities of this process, characterized by jetting drilling and soaking phases. To address this issue, a novel approach using deep reinforcement learning is introduced for comprehensive optimization of jetting parameters, enhancing deep-sea drilling efficiency. The methodology begins with an orthogonal simulated experiment to compile a robust dataset. Following preprocessing, including feature extraction and parameter scaling, a predictive model is developed to correlate installation parameters with jetting and soaking time. A composite action-value Q-learning (CAPQL) based multi-objective reinforcement learning framework is employed, establishing a Markov decision process environment for simultaneous optimization of jetting time and soaking periods. Applying this framework to selected deep-sea wells demonstrates a significant decrease in jetting drilling time (average 46.18%) and soaking time (22.57%), with the predictive model achieving a 99.32% average fit. This approach effectively navigates parameter interdependencies, ensuring optimal outcomes across diverse objectives. These findings present a groundbreaking approach for conductor casing jetting, offering greater precision and efficiency in offshore drilling operations and the potential to redefine industry standards.

Exploring Evolution and Trends: A Bibliometric Analysis and Scientific Mapping of Multiobjective Optimization Applied to Hybrid Microgrid Systems

Hybrid energy systems (HESs) integrate renewable sources, storage, and optionally conventional energies, offering a sustainable alternative to fossil fuels. Microgrids (MGs) bolster this integration, enhancing energy management, resilience, and reliability across different levels. This study, emphasizing the need for refined optimization methods, investigates three themes: renewable energy, microgrid, and multiobjective optimization (MOO), through a bibliometric analysis of 470 Scopus documents from 2010 to 2023, analyzed using SciMAT v1.1.04 software. It segments the research into two periods, 2010–2019 and 2020–2023, revealing a surge in MOO focus, particularly in the latter period, with a 35% increase in MOO-related research. This indicates a shift toward comprehensive energy ecosystem management that balances environmental, technical, and economic elements. The initial focus on MOO, genetic algorithms, and energy management systems has expanded to include smart grids and electric power systems, with MOO remaining a primary theme in the second period. The increased application of artificial intelligence (AI) in optimizing HMGS within the MOO framework signals a move toward more sustainable, intelligent energy solutions. Despite progress, challenges remain, including high battery costs, the need for reliable MOO data, the intermittency of renewable energy sources, and HMGS network scalability issues, highlighting directions for future research.

Evolutionary Multiobjective Molecule Optimization in an Implicit Chemical Space.

Optimization techniques play a pivotal role in advancing drug development, serving as the foundation of numerous generative methods tailored to efficiently design optimized molecules derived from existing lead compounds. However, existing methods often encounter difficulties in generating diverse, novel, and high-property molecules that simultaneously optimize multiple drug properties. To overcome this bottleneck, we propose a multiobjective molecule optimization framework (MOMO). MOMO employs a specially designed Pareto-based multiproperty evaluation strategy at the molecular sequence level to guide the evolutionary search in an implicit chemical space. A comparative analysis of MOMO with five state-of-the-art methods across two benchmark multiproperty molecule optimization tasks reveals that MOMO markedly outperforms them in terms of diversity, novelty, and optimized properties. The practical applicability of MOMO in drug discovery has also been validated on four challenging tasks in the real-world discovery problem. These results suggest that MOMO can provide a useful tool to facilitate molecule optimization problems with multiple properties.

Optimal demand response scheduling and voltage reinforcement in distribution grids incorporating uncertainties of energy resources, placement of energy storages, and aggregated flexible loads

Instead of expanding power plant capacities, which is an extremely expensive investment option, demand response offers an economical solution to the challenges arising from the variability and intermittency of the renewable energy resources and demand variations, particularly during demand peak periods. This paper proposes a multi-objective optimization framework for the optimal power flow problem that integrates a stepwise demand response involving flexible and aggregated loads. The process includes short-term demand forecasting using long short-term memory (LSTM) networks in a smart distribution grid, followed by the optimal allocation of energy storage systems, and load aggregators. By determining the optimal solution point of the multi-objective problem analytically, significant system costs and peak demand can be reduced without compromising system stability. Through numerical studies for a sample study case, a reduction of 22% in system costs, 2% in total voltage variation, and 10% in peak demand is observed for a negligible impact on customers’ convenience.

DOT: a flexible multi-objective optimization framework for transferring features across single-cell and spatial omics

Single-cell transcriptomics and spatially-resolved imaging/sequencing technologies have revolutionized biomedical research. However, they suffer from lack of spatial information and a trade-off of resolution and gene coverage, respectively. We propose DOT, a multi-objective optimization framework for transferring cellular features across these data modalities, thus integrating their complementary information. DOT uses genes beyond those common to the data modalities, exploits the local spatial context, transfers spatial features beyond cell-type information, and infers absolute/relative abundance of cell populations at tissue locations. Thus, DOT bridges single-cell transcriptomics data with both high- and low-resolution spatially-resolved data. Moreover, DOT combines practical aspects related to cell composition, heterogeneity, technical effects, and integration of prior knowledge. Our fast implementation based on the Frank-Wolfe algorithm achieves state-of-the-art or improved performance in localizing cell features in high- and low-resolution spatial data and estimating the expression of unmeasured genes in low-coverage spatial data.

Hierarchical thermal modeling and surrogate-model-based design optimization framework for cold plates used in battery thermal management systems

The continued advancement of battery-powered electric vehicles (EVs) towards higher energy and power densities poses significant thermal challenges for batteries. This work proposes a generalized design optimization framework for battery thermal management systems (BTMS), responding to the need for EVs with enhanced battery thermal performance, safety, and lifetime. This framework combines hierarchical thermal modeling and surrogate-model-based design optimization, explicitly tailored for liquid-cooled cold plates commonly used in EV BTMS. The hierarchical thermal modeling component decouples the battery cell, battery module, and cold plate models to reduce computational modeling costs while preserving the relative thermal performance of different cold plate designs. The design optimization component leverages our pioneer deep encoder–decoder hierarchical (DeepEDH) convolutional neural network surrogate modeling methodology to predict the cold plate’s full pressure, velocity, and temperature fields. Computationally efficient DeepEDH neural networks replace costly transient thermal simulations of cold plates and compute the objectives for evolutionary optimization using the Non-dominated Sorting Genetic Algorithm II (NSGA-II). This work’s novel modeling and optimization framework produces optimal cold plate designs that consider battery-module-specific design features, including distributed battery cell heat generation, packaging materials, and heat-spreading characteristics. The proposed framework is evaluated through a cold plate design for modular BTMS, thoroughly assessing the impact of optimization objectives, flow path constraints, modeling assumptions, and design variable choices. The results demonstrate the effectiveness of the combined hierarchical thermal modeling and design optimization framework, with the optimized cold plates achieving reductions of 6.26K, 6.29K (22.7%), and 16.16Pa (18.7%) for maximum temperature, maximum temperature difference, and pressure loss, respectively. Moreover, compared to other methodologies that apply direct optimization without hierarchical and surrogate models, our approach significantly reduces the computational cost — from approximately 4800h to 13.5h. Our generalized multi-objective optimization framework is an effective and efficient design tool that can be leveraged to advance thermal management innovations for next-generation battery systems.

Enhancing project portfolio selection for construction holding firms: a multi-objective optimization framework with risk analysis

Enhancing project portfolio selection for construction holding firms: a multi-objective optimization framework with risk analysis

Inverse design of nonlinear metasurfaces for sum frequency generation

Abstract Sum frequency generation (SFG) has multiple applications, from optical sources to imaging, where efficient conversion requires either long interaction distances or large field concentrations in a quadratic nonlinear material. Metasurfaces provide an essential avenue to enhanced SFG due to resonance with extreme field enhancements with an integrated ultrathin platform. In this work, we formulate a general theoretical framework for multi-objective topology optimization of nanopatterned metasurfaces that facilitate high-efficiency SFG and simultaneously select the emitted direction and tailor the metasurface polarization response. Based on this framework, we present novel metasurface designs showcasing ultimate flexibility in transforming the outgoing nonlinearly generated light for applications spanning from imaging to polarimetry. For example, one of our metasurfaces produces highly polarized and directional SFG emission with an efficiency of over 0.2 cm2 GW−1 in a 10 nm signal operating bandwidth.

Designing a Multi-Objective Optimization Framework for Global Supply Chains Using Genetic Algorithm and NSGA II

Supply chain optimization is crucial for firms to boost efficiency and competitiveness amidst increasing complexity and risks in global markets. This study provides light into the optimization of a global supply chain network using genetic algorithms, concentrated on avoiding disruptions and decreasing total supply chain costs. The main objectives comprise decreasing total supply chain costs, managing disruptions in product quantity and quality, and optimizing operating parameters and features. The study employs genetic algorithms for optimization, explaining the approach and providing final interpretations to boost supply chain efficiency and resilience against disturbances.

A spatially constrained independent component analysis jointly informed by structural and functional network connectivity.

There are a growing number of neuroimaging studies motivating joint structural and functional brain connectivity. Brain connectivity of different modalities provides insight into brain functional organization by leveraging complementary information, especially for brain disorders such as schizophrenia. In this paper, we propose a multi-modal independent component analysis (ICA) model that utilizes information from both structural and functional brain connectivity guided by spatial maps to estimate intrinsic connectivity networks (ICNs). Structural connectivity is estimated through whole-brain tractography on diffusion-weighted MRI (dMRI), while functional connectivity is derived from resting-state functional MRI (rs-fMRI). The proposed structural-functional connectivity and spatially constrained ICA (sfCICA) model estimates ICNs at the subject level using a multi-objective optimization framework. We evaluated our model using synthetic and real datasets (including dMRI and rs-fMRI from 149 schizophrenia patients and 162 controls). Multi-modal ICNs revealed enhanced functional coupling between ICNs with higher structural connectivity, improved modularity, and network distinction, particularly in schizophrenia. Statistical analysis of group differences showed more significant differences in the proposed model compared to the unimodal model. In summary, the sfCICA model showed benefits from being jointly informed by structural and functional connectivity. These findings suggest advantages in simultaneously learning effectively and enhancing connectivity estimates using structural connectivity.

Research on multi-condition and multi-objective control of pumped storage unit optimized by improved multi-objective particle swarm optimization algorithm

Abstract In order to maintain good control quality of the pumped storage unit regulation system during the switching of operating conditions, a multi-objective optimization framework based on improved multi-objective particle swarm optimization algorithm is proposed, which takes into account the control stability, rapidity and robustness under multiple working conditions. Firstly, the fractional order PID controller with higher applicability and flexibility is introduced to replace the traditional PID controller. Then, an objective function considering the ITAE index of low-head and medium-high-head conditions is constructed. On this basis, the multi-objective particle swarm optimization algorithm with pigeon swarm search operator, adaptive Cauchy mutation strategy and improved initialization method is used to optimize the controller parameters, and the multi-condition and multi-objective fractional order PID optimal control of pumped storage unit regulation system is realized. Through simulation, it is proved that the controller parameters based on the framework optimization proposed in this paper have strong adaptability and robustness under different working conditions, which provides a new idea for the optimal control of pumped storage unit regulation system.

CARIn: Constraint-Aware and Responsive Inference on Heterogeneous Devices for Single- and Multi-DNN Workloads

The relentless expansion of deep learning (DL) applications in recent years has prompted a pivotal shift towards on-device execution, driven by the urgent need for real-time processing, heightened privacy concerns, and reduced latency across diverse domains. This paper addresses the challenges inherent in optimising the execution of deep neural networks (DNNs) on mobile devices, with a focus on device heterogeneity, multi-DNN execution, and dynamic runtime adaptation. We introduce CARIn , a novel framework designed for the optimised deployment of both single- and multi-DNN applications under user-defined service-level objectives (SLOs). Leveraging an expressive multi-objective optimisation (MOO) framework and a runtime-aware sorting and search algorithm ( RASS ) as the MOO solver, CARIn facilitates efficient adaptation to dynamic conditions while addressing resource contention issues associated with multi-DNN execution. Notably, RASS generates a set of configurations, anticipating subsequent runtime adaptation, ensuring rapid, low-overhead adjustments in response to environmental fluctuations. Extensive evaluation across diverse tasks, including text classification, scene recognition, and face analysis, showcases the versatility of CARIn across various model architectures, such as Convolutional Neural Networks (CNNs) and Transformers, and realistic use cases. We observe a substantial enhancement in the fair treatment of the problem’s objectives, reaching 1.92 × when compared to single-model designs, and up to 10.69 × in contrast to the state-of-the-art OODIn framework. Additionally, we achieve a significant gain of up to 4.06 × over hardware-unaware designs in multi-DNN applications. Finally, our framework sustains its performance while effectively eliminating the time overhead associated with identifying the optimal design in response to environmental challenges.

ESG integration in portfolio selection: A robust preference-based multicriteria approach

We present a framework for multi-objective optimization where the classical mean–variance portfolio model is extended to integrate the environmental, social and governance (ESG) criteria on the same playing field as risk and return and, at the same time, to reflect the investors’ preferences in the optimal portfolio allocation. To obtain the three–dimensional Pareto front, we apply an efficient multi-objective genetic algorithm, which is based on the concept of ɛ-dominance. We next address the issue of how to incorporate investors’ preferences to express the relative importance of each objective through a robust weighting scheme in a multicriteria ranking framework. The new proposal has been applied to real data to find optimal portfolios of socially responsible investment funds, and the main conclusion from the empirical tests is that it is possible to provide the investors with a robust solution in the mean–variance–ESG surface according to their preferences.

Multi-objective optimization and energy efficiency improvement for rotor duct in integrated energy recovery-pressure boost device

To reduce the volumetric mixing rate of the integrated energy recovery-pressure boost device, a novel symmetrical duct structure with diversion surface was proposed. The surrogate model-based multi-objective optimization framework was established, and obtained the optimal ducts V1 and V2 combined with computational fluid dynamics simulation. Numerical results indicate that optimized duct structure could promote the formation of plug flow and effectively improve the collision efficiency between high-pressure brine and low-pressure seawater, and the volumetric mixing rate corresponding to the ducts V1 and V2 was reduced by 31.52% and 22.18%, and the duct volume efficiency was increased by 16.04% and 19.56%, respectively. Finally, the performance of optimized device was verified through seawater reverse osmosis desalination system under actual working conditions. Experimental results showed that the mixing rate of optimized ducts had a significant decrease of 30.86% and 20.53% in the applicable range, and the energy recovery efficiency improved at most 5.41% and 7.55%, respectively. The optimized ducts could significantly reduce the specific energy consumption of the desalination system by 26.01% and 18.22%, respectively, and the specific energy consumption reached the lowest level of 3.02 kWh/m3, demonstrating the superiority of the structure and the accuracy and effectiveness of the optimization strategy.

A general multi-objective Bayesian optimization framework for the design of hybrid schemes towards adaptive complex flow simulations

Achieving accuracy with underresolved simulation of complex compressible flows with multiscale flow structures is a challenge. Either the numerical dissipation or the resolution and thereby the numerical cost is impractically high. Also, in the design of numerical solvers, the application of a solver for specific flow classes is balanced by robustness allowing the study of a broad range of flows. In this study, we propose a hybrid fifth-order targeted essentially non-oscillatory (TENO5)-based scheme tailored to optimally simulate compressible flows with underresolved dilatational and vortical multiscale structures. For optimal design, three data-driven objectives are defined. A novel objective that derives from the numerical dissipation rate analysis is a key element to deal with underresolved complex flows in practical applications. The optimization process employs a multi-objective Bayesian optimization framework with an expected hypervolume improvement and three flow configurations representative for a broad range of two- and three-dimensional flows with genuine and non-genuine subgrid scales. The optimized hybrid scheme is validated by comparing with shock-capturing schemes of the weighted essentially non-oscillatory (WENO)- and TENO- families with flows of complex shock interactions, Kelvin-Helmholtz instabilities, shock-vortex interactions, vortical and turbulent flows

Integrating NSGA-II and CFD for enhanced urban airflow prediction: Recalibration of closure coefficients for a nonlinear eddy viscosity model

Accurately predicting urban environmental airflows using the Reynolds-averaged Navier‒Stokes (RANS) approach presents significant challenges due to unsolved closure issues. Previous attempts to address these limitations through coefficient adjustments were constrained by isotropic assumptions, resulting in limited generalizability and reliability, especially for complex building configurations. In this study, we proposed a recalibration approach within a multi-objective optimization framework. Nondominant sorting genetic algorithm-II (NSGA-II) and computational fluid dynamics (CFD) numerical computations were utilized to optimize the closure coefficients of a nonlinear eddy viscosity (NLEV) model, with a focus on improving the accuracy of the mean velocity, turbulent kinetic energy, and airflow characteristic predictions. To ensure the robustness of the model, appropriate objective functions were carefully defined. The proposed approach was evaluated using three-dimensional (3D) building cases related to urban prototypes, and the results demonstrated the necessity of multi-objective optimization. Our findings indicated that the trade-off solution on the Pareto front consistently outperformed the baseline and single-objective optimal versions across different types of urban airflow, demonstrating superior generalizability. The simulation results of this solution exhibited closer agreement with the wind tunnel experimental data and provided more accurate distributions of the time-averaged values and airflow characteristics. The normalized root mean square errors are reduced by about 33 %, 52 %, 35 % and 9 % in velocity, and 31 %, 25 %, 24 % and 11 % in turbulent kinetic energy, respectively. These results underscore the effectiveness of the multi-objective optimization framework in addressing closure issues and improving the prediction accuracy of complex urban environmental airflows.

Enhancing Robustness in Precast Modular Frame Optimization: Integrating NSGA-II, NSGA-III, and RVEA for Sustainable Infrastructure

The advancement toward sustainable infrastructure presents complex multi-objective optimization (MOO) challenges. This paper expands the current understanding of design frameworks that balance cost, environmental impacts, social factors, and structural integrity. Integrating MOO with multi-criteria decision-making (MCDM), the study targets enhancements in life cycle sustainability for complex engineering projects using precast modular road frames. Three advanced evolutionary algorithms—NSGA-II, NSGA-III, and RVEA—are optimized and deployed to address sustainability objectives under performance constraints. The efficacy of these algorithms is gauged through a comparative analysis, and a robust MCDM approach is applied to nine non-dominated solutions, employing SAW, FUCA, TOPSIS, PROMETHEE, and VIKOR decision-making techniques. An entropy theory-based method ensures systematic, unbiased criteria weighting, augmenting the framework’s capacity to pinpoint designs balancing life cycle sustainability. The results reveal that NSGA-III is the algorithm converging towards the most cost-effective solutions, surpassing NSGA-II and RVEA by 21.11% and 10.07%, respectively, while maintaining balanced environmental and social impacts. The RVEA achieves up to 15.94% greater environmental efficiency than its counterparts. The analysis of non-dominated solutions identifies the A4 design, utilizing 35 MPa concrete and B500S steel, as the most sustainable alternative across 80% of decision-making algorithms. The ranking correlation coefficients above 0.94 demonstrate consistency among decision-making techniques, underscoring the robustness of the integrated MOO and MCDM framework. The results in this paper expand the understanding of the applicability of novel techniques for enhancing engineering practices and advocate for a comprehensive strategy that employs advanced MOO algorithms and MCDM to enhance sustainable infrastructure development.

Synergistic Integration of EVs and Renewable DGs in Distribution Micro-Grids

This paper proposes a multi-objective optimization framework for safe, reliable, and economic integration of electric vehicles (EVs) and renewable distributed generators (DGs) in distribution micro-grids. EV and DG coordination optimization with the use of vehicle-to-grid (V2G) technology along with system reconfiguration optimization is developed to provide collective revenues and address integrational complications that may occur by additional system loading due to EV charging and EV-DG energy exchanges. A Genetic Algorithm (GA) optimizes the EV charging/discharging in synergies with renewable DGs to maximize benefits that can be captured by their collaborative participation in electricity market and through renewable energy arbitrage. The developed EV charging/discharging optimization is implemented in a real 134-bus distribution network and is evaluated for its potential operational implications, namely, increased system losses. A system reconfiguration is then proposed to reduce the system losses by optimizing the flow of power through switching on/off the connections within the micro-grid and/or with other distribution systems. Simulation results demonstrate the efficiency of the proposed method in not only providing collective revenues, but also in enhancing the system operation by reducing the losses of the distribution grid. The collective benefits proposed by the developed optimization and validated by the simulation results facilitate transitioning to clean and eco-friendly sources of energy for generation and transportation, which in turn leads to more sustainable development of societies and communities.

A multiobjective beam angle optimization framework for intensity-modulated radiation therapy

Radiation therapy treatment planning is inherently a multiobjective problem, aiming to obtain the best tradeoffs between irradiating the tumor with the prescribed dose and sparing as much as possible the surrounding healthy organs. Many different multiobjective approaches have been proposed for the optimization of radiation intensities for fixed beam irradiation directions. However, multiobjective beam angle optimization is seldom considered. The purpose of this paper is to introduce a new multiobjective optimization framework that explicitly and simultaneously considers the optimization of intensities and also beam directions. Whilst multiobjective optimization of radiation intensities considering a fixed set of beam directions gives rise to a single Pareto front, beam angle optimization gives rise to the appearance of multiple Pareto fronts, each one associated with a given beam angle set. Our framework proposes a beam angle set choice based on the evaluation of non-dominated solutions belonging to different Pareto fronts, using a tree-based approach and a performance indicator to assess the quality of each Pareto front. The proposed approach, illustrated by head-and-neck cancer cases, allows for more flexibility in the calculation of solutions and a better understanding of the existing compromises between different objectives.