Single-objective Optimization Research Articles

Dynamic programming with meta-reinforcement learning: a novel approach for multi-objective optimization

AbstractMulti-objective optimization (MOO) endeavors to identify optimal solutions from a finite array of possibilities. In recent years, deep reinforcement learning (RL) has exhibited promise through its well-crafted heuristics in tackling NP-hard combinatorial optimization (CO) problems. Nonetheless, current methodologies grapple with two key challenges: (1) They primarily concentrate on single-objective optimization quandaries, rendering them less adaptable to the more prevalent MOO scenarios encountered in real-world applications. (2) These approaches furnish an approximate solution by imbibing heuristics, lacking a systematic means to enhance or substantiate optimality. Given these challenges, this study introduces an overarching hybrid strategy, dynamic programming with meta-reinforcement learning (DPML), to resolve MOO predicaments. The approach melds meta-learning into an RL framework, addressing multiple subproblems inherent to MOO. Furthermore, the precision of solutions is elevated by endowing exact dynamic programming with the prowess of meta-graph neural networks. Empirical results substantiate the supremacy of our methodology over previous RL and heuristics approaches, bridging the chasm between theoretical underpinnings and real-world applicability within this domain.

A Scoping Review on Simulation-Based Design Optimization in Marine Engineering: Trends, Best Practices, and Gaps

AbstractThis scoping review assesses the current use of simulation-based design optimization (SBDO) in marine engineering, focusing on identifying research trends, methodologies, and application areas. Analyzing 277 studies from Scopus and Web of Science, the review finds that SBDO is predominantly applied to optimizing marine vessel hulls, including both surface and underwater types, and extends to key components like bows, sterns, propellers, and fins. It also covers marine structures and renewable energy systems. A notable trend is the preference for deterministic single-objective optimization methods, indicating potential growth areas in multi-objective and stochastic approaches. The review points out the necessity of integrating more comprehensive multidisciplinary optimization methods to address the complex challenges in marine environments. Despite the extensive application of SBDO in marine engineering, there remains a need for enhancing the methodologies’ efficiency and robustness. This review offers a critical overview of SBDO’s role in marine engineering and highlights opportunities for future research to advance the field.

Source code obfuscation with genetic algorithms using LLVM code optimizations

Abstract With the advent of the cloud computing model allowing a shared access to massive computing facilities, a surging demand emerges for the protection of the intellectual property tied to the programs executed on these uncontrolled systems. If novel paradigm as confidential computing aims at protecting the data manipulated during the execution, obfuscating techniques (in particular at the source code level) remain a popular solution to conceal the purpose of a program or its logic without altering its functionality, thus preventing reverse-engineering on the program even with the help of computing resources. The many advantages of code obfuscation, together with its low cost, makes it a popular technique. This paper proposes a novel methodology for source code obfuscation that can be used together with other traditional obfuscation techniques, making the code more robust against reverse engineering attacks. Three program complexity metrics are used to define three different single-objective combinatorial optimization versions of the problem, which are solved and analysed. Additionally, three multi-objective problems are defined, those considering each of the selected metrics together with the program execution time, in order to avoid strong obfuscations penalizing the performance. The goal of the defined problems is to find sequences of LLVM optimizations that lead to highly obfuscated versions of the original code. These transformations are applied to the back-end pseudo-assembly code (i.e., LLVM Intermediate Representation), thus avoiding any further optimizations by the compiler. Classical genetic algorithms (GAs) are used to solve the studied problems, namely a basic cellular GA for the single-objective problems and the popular NSGA-II for the multi-objective ones. The promising results show the potential of the proposed technique.

3D Path Planning of the Solar Powered UAV in the Urban-Mountainous Environment with Multi-Objective and Multi-Constraint Based on the Enhanced Sparrow Search Algorithm Incorporating the Levy Flight Strategy

In response to practical application challenges in utilizing solar-powered unmanned aerial vehicle (UAV) for remote sensing, this study presents a three-dimensional path planning method tailored for urban-mountainous environment. Taking into account constraints related to the solar-powered UAV, terrain, and mission objectives, a multi-objective trajectory optimization model is transferred into a single-objective optimization problem with weight factors and multi-constraint and is developed with a focus on three key indicators: minimizing trajectory length, maximizing energy flow efficiency, and minimizing regional risk levels. Additionally, an enhanced sparrow search algorithm incorporating the Levy flight strategy (SSA-Levy) is introduced to address trajectory planning challenges in such complex environments. Through simulation, the proposed algorithm is compared with particle swarm optimization (PSO) and the regular sparrow search algorithm (SSA) across 17 standard test functions and a simplified simulation of urban-mountainous environments. The results of the simulation demonstrate the superior effectiveness of the designed improved SSA based on the Levy flight strategy for solving the established single-objective trajectory optimization model.

A Multi-Objective Pigeon-Inspired Optimization Algorithm for Community Detection in Complex Networks

Community structure is a very interesting attribute and feature in complex networks, which has attracted scholars’ attention and research on community detection. Many single-objective optimization algorithms have been migrated and modified to serve community detection problems. Due to the limitation of resolution, the final algorithm implementation effect is not ideal. In this paper, a multi-objective community detection method based on a pigeon-inspired optimization algorithm, MOPIO-Net, is proposed. Firstly, the PIO algorithm is discretized in terms of the solution space representation, position, and velocity-updating strategies to adapt to discrete community detection scenarios. Secondly, by minimizing the two objective functions of community score and community fitness at the same time, the community structure with a tight interior and sparse exterior is obtained. Finally, for the misclassification caused by boundary nodes, a mutation strategy is added to improve the accuracy of the final community recognition. Experiments on synthetic and real networks verify that the proposed algorithm is more accurate in community recognition compared to 11 benchmark algorithms, confirming the effectiveness of the proposed method.

Multi-objective generalized normal distribution optimization: a novel algorithm for multi-objective problems

AbstractThis study introduces the Multi-objective Generalized Normal Distribution Optimization (MOGNDO) algorithm, an advancement of the Generalized Normal Distribution Optimization (GNDO) algorithm, now adapted for multi-objective optimization tasks. The GNDO algorithm, previously known for its effectiveness in single-objective optimization, has been enhanced with two key features for multi-objective optimization. The first is the addition of an archival mechanism to store non-dominated Pareto optimal solutions, ensuring a detailed record of the best outcomes. The second enhancement is a new leader selection mechanism, designed to strategically identify and select the best solutions from the archive to guide the optimization process. This enhancement positions MOGNDO as a cutting-edge solution in multi-objective optimization, setting a new benchmark for evaluating its performance against leading algorithms in the field. The algorithm's effectiveness is rigorously tested across 35 varied case studies, encompassing both mathematical and engineering challenges, and benchmarked against prominent algorithms like MOPSO, MOGWO, MOHHO, MSSA, MOALO, MOMVO, and MOAOS. Utilizing metrics such as Generational Distance (GD), Inverted Generational Distance (IGD), and Maximum Spread (MS), the study underscores MOGNDO's ability to produce Pareto fronts of high quality, marked by exceptional precision and diversity. The results affirm MOGNDO's superior performance and versatility, not only in theoretical tests but also in addressing complex real-world engineering problems, showcasing its high convergence and coverage capabilities. The source codes of the MOGNDO algorithm are publicly available at https://nimakhodadadi.com/algorithms-%2B-codes.

Solving Convex Multi-Objective Optimization Problems via a Capable Neural Network Scheme

A neural network architecture is constructed to solve Convex Multi-objective Programming Problem (CMPP). By using the weighted sum method, a single-objective optimization problem related to the CMPP is formulated, in which the scalar objective is summation of weighted original objective functions. The Pareto Optimal Solution (POS) is given by using different values of weights. A neural network framework is then designed for solving the obtained convex problem. Based on employing Lyapunov theory, the proposed model is established to be stable in the sense of Lyapunov and it is globally convergent to an exact optimal solution of the convex programming problem with different values of weights. The computer simulations shows the feasibility and the efficiency of the suggested model.

Optimizing pump operations in water distribution networks: Balancing energy efficiency, water quality and operational constraints

Optimizing pump operation in urban water distribution systems represents a critical task for enhancing energy efficiency. Despite extensive research, achieving optimal pump operation remains a challenge due to the presence of conflicting objectives, such as minimizing energy costs and ensuring adequate pressure head availability. The recent emphasis on maintaining distributed water quality further increases the complexity of the problem. Mathematically, the optimization problem manifests as a non-convex and non-continuous search space, necessitating the application of non-derivative optimization techniques, which are highly susceptible to constraints. This work presents a comprehensive analysis of optimal pump operation, considering energy efficiency and water quality as primary objectives. A series of constraints are progressively incorporated into the model to evaluate their impact on the final operational strategy. Initially, single-objective optimization is employed using the Particle Swarm Optimization (PSO) algorithm, a robust and widely adopted approach in water resources research. Subsequently, a multi-objective algorithm, the Non-dominated Sorting Genetic Algorithm-II (NSGA-II), is utilized to identify the Pareto front, encompassing both energy consumption and water quality indicators. The results demonstrate that constraints exert a significant influence on the final energy consumption and the convergence of the optimization process. Moreover, by simultaneously considering both objectives, this study proves the feasibility of identifying an optimal operational scheme that minimizes energy consumption while maintaining acceptable water quality.

Prediction and decision of free-piston linear generator on starting process for multi-fuel adaptability

Free-piston linear generators, due to their variable compression ratio, are considered to possess excellent multi-fuel adaptability. Further, fuel-flexible is a weighty research topic for addressing energy source limitations, meeting powertrain generalization requirements, and achieving energy savings and emission reductions. The initial concern to solve is the rapid starting of various fuels within the same free-piston linear generator. This study aims to introduce a mechanical resonance starting performance prediction model based on an artificial neural network and provides a control decision method based on single-objective optimization with soft constraints to achieve flexible starting of a variety of fuels. This paper also assessed the starting performance using the free-piston linear generator prototype and conducted a comprehensive parameterized analysis to investigate the influence of parameters and control variables on starting performance. Using the starting decision method proposed, the control variable values for ethanol starting were derived on the prototype, resulting in driving current at 3.5A, intake air mass flow at 95 standard liters per minute, and intake air temperature at 335 K. This decision value effectively met the predefined requirements, successfully igniting and starting the prototype with ethanol. This outcome serves as a robust validation of the fuel-flexible capabilities within free-piston linear generator systems.

A short-term wind-hydrothermal operational framework in the presence of pumped-hydro storage

Short-term operation is one of the most significant aspects of power systems operation and planning and it mainly includes resource scheduling including renewable and non-renewable energy sources. The mentioned problem is modeled as an optimization problem with one or more objectives depending upon the priorities of the decision maker (system operator) and several equality and inequality constraints. In this respect, this paper proposes a short-term operational model for the hydrothermal power systems in the presence of wind power generation and pumped-hydro storage (PHS) units. The problem is formulated as a single-objective optimization problem that aims to minimize the total operating cost including the fuel cost of thermal units. The problem has been comprehensively investigated through five case studies where the first case study is used for the sake of validation. Then, the other case studies assessed the impacts of wind power generation, PHS unit, and uncertainty on the problem. The five case studies are as follows: Case 1: Short-term hydrothermal (STH) scheduling which aims to minimize the total cost. Case 2: Deterministic short-term hydrothermal-wind (STHTW) scheduling which aims to minimize the total cost. Case 3: Stochastic wind-hydro-thermal scheduling that seeks to minimize the total cost taking into consideration the uncertainty of the wind power generation. Case 4: Deterministic wind-hydro-thermal scheduling in the presence of the PHS unit with the total cost minimization as the objective function. Case 5: Stochastic wind-hydro-thermal scheduling in the presence of the PHS unit that is used to evaluate the impact of uncertainty of wind power generation with the total cost minimization as the objective function. It is noteworthy that the first three case studies have been formulated using a non-linear programming (NLP) model and solved using the CONOPT solver in GAMS. Case studies 4 and 5 have been presented using a mixed-integer non-linear programming (MINLP) model and solved using the DICOPT solver in GAMS. A sensitivity analysis has also been carried out to assess the system's performance with different loading conditions. In this respect, the total cost and PHS operation with different amounts of load demand were evaluated and the results have been presented and discussed.

Multi-objective design optimization of polymer spur gears using a hybrid approach

Polymer gears are used in applications requiring small to moderate loads to effectively transmit power and use the limited place available as possible. Various commercial standards have been provided designers with the rating criteria and acquaintance of different polymer material properties for the process of design. However, the result was unsatisfactory in terms of economy, time, and the optimality of the product. Thus, classic and stochastic algorithms have been embraced to reach the best design of polymer gears. Taking advantage of the former and latter algorithm’ methods, optimal design of gears could be attained with an increased gear life span and decreased failure modes. In this study, polyoxymethylene (POM) spur gear set has been optimized combining the mathematical model from plastic standards and hybrid optimization approach of multi-objective genetic algorithm (MOGA) and sequential quadratic programming (SQP). Weight and power loss were the objective functions. Five design variables were optimized with the satisfaction of bending and contact stresses, temperature, wear, and deformation as constraints. Solutions of the problem were formulated as Pareto optimal set. The results of multi-objective were compared with previously published single-objective optimization. The results favored multi-objective optimization (82.67%, 31.39% reduction in weight and power loss respectively) as it gave the best applicable solution fitting in real life situations. The results also went hand in hand with literature confirming the efficiency of the proposed algorithm. With the variation of operating parameters, various optimal designs could be obtained where the designers can choose the design that is suitable for a particular application.

A deep reinforcement learning algorithm framework for solving multi-objective traveling salesman problem based on feature transformation

As a special type of multi-objective combinatorial optimization problems (MOCOPs), the multi-objective traveling salesman problem (MOTSP) plays an important role in practical fields such as transportation and robot control. However, due to the complexity of its solution space and the conflicts between different objectives, it is difficult to obtain satisfactory solutions in a short time. This paper proposes an end-to-end algorithm framework for solving MOTSP based on deep reinforcement learning (DRL). By decomposing strategies, solving MOTSP is transformed into solving multiple single-objective optimization subproblems. Through linear transformation, the features of the MOTSP are combined with the weights of the objective function. Subsequently, a modified graph pointer network (GPN) model is used to solve the decomposed subproblems. Compared with the previous DRL model, the proposed algorithm can solve all the subproblems using only one model without adding weight information as input features. Furthermore, our algorithm can output a corresponding solution for each weight, which increases the diversity of solutions. In order to verify the performance of our proposed algorithm, it is compared with four classical evolutionary algorithms and two DRL algorithms on several MOTSP instances. The comparison shows that our proposed algorithm outperforms the compared algorithms both in terms of training time and the quality of the resulting solutions.

Low–carbon regulation method for greenhouse light environment based on multi–objective optimization

Controlled environment agriculture is uniquely positioned to contribute to the decarbonization of cities due to its low carbon emissions and high energy efficiency in the food supply process. The energy consumed by the horticultural industry to suppplemental lighting is enormous, but improvements in the light environment of greenhouse plants are essential. To make light regulation compatible with low carbon and environmentally friendly while improving photosynthetic production, experiments on tomato plant measurements with nested light intensity, spectra, carbon dioxide concentration and temperature were carried out in a solar greenhouse based on an infrared gas analyzer and machine learning. Based on the photosynthetic response to the photosynthetic photon flux density and red–blue ratio under multi–environmental conditions, the photosynthesis prediction model was developed using the least squares support vector regression, and model hyperparameter optimization was achieved by genetic algorithm nested cross–validation. A calculation method of carbon emissions from suppplemental lighting was proposed, and the low–carbon light regulation was abstracted into multi–objective optimization for the solution. Based on the comprehensive objective evaluation method, low–carbon targets for light regulation were obtained through decision–making. The results showed that the developed prediction model with root mean square error of 1.069 and determination coefficient of 0.976, which is both accurate and fast, realizes the description and expression of the net photosynthetic rate in response to the environments. The optimal targets contain two–dimensional information of light intensity and spectra, and reduces carbon emissions by a maximum of 14.85% compared to the light saturation objective based on single–objective optimization at 95.49 % of the net photosynthetic rate. The method is energy efficient and suitable for supporting multi–spectral light environment regulation applications and guiding the greenhouse environment management under the demand of carbon reduction.

Optimal design of mixed dielectric coaxial-annular TSV using GWO algorithm based on artificial neural network

The single-objective and single-parameter optimization method is commonly used in the structure optimization of TSV to improve the transmission characteristics, for which a structure design scheme that simultaneously satisfies multiple target requirements is difficult to obtain. Moreover, the method cannot simultaneously optimize different design parameters. Aiming at the above problems, a global optimization method based on the gray wolf optimization (GWO) algorithm and artificial neural network (ANN) model is proposed. With the presented mixed dielectric coaxial-annular TSV model, firstly six key design parameters A-F are selected as optimization variables by the control variable method. The L25(56) orthogonal experiment is designed for Taguchi analysis and analysis of variance (ANOVA). Then, three prediction models, ANN, support vector machine (SVM), and extreme learning machine (ELM), are developed with the extended orthogonal data as the training sets. It is found that the ANN model performed best. To search for the global optimal solution, the genetic algorithm (GA) and GWO algorithm, combined with the ANN model are applied, respectively. The results show that the GWO algorithm is more successful in solving the problem of falling into the local optimum than GA, and the convergence speed is faster and more stable. After GWO-ANN optimization, the performance of each S-parameter index is greatly improved, S11 reduces by 14.05 dB, S21 increases by 0.33 dB, and S31 reduces by 12.50 dB at 30 GHz.

Multi-objective (Carbon-Exergy) optimization study of CO2 cascade systems using multiple refrigerant pairs

In order to guide the operation of the actual cascade system, this paper takes the cascade system as the research object and compares the use of six refrigerant pairs. Unlike the previous single-objective optimization of refrigeration system carbon emissions and other indicators, this research uses a multi-objective optimization algorithm, that is, we consider the cascade system's exergy efficiency and carbon emissions at the same time. We reduce carbon emissions while improving the system's exergy efficiency, and finally find out the specific operating conditions under which the system can realize low-carbon and high-efficiency operation, which further promotes the process of achieving energy saving and emission reduction in factories. The optimization results show that among the six refrigerant pairs, the use of R744-R134a and R744-R290 refrigerant pairs can make the system operate better, but considering the high global warming potential (GWP) of R134a refrigerant, the use of R744-R290 combination is more recommended.

Multi-strategy learning-based particle swarm optimization algorithm for COVID-19 threshold segmentation

With advancements in science and technology, the depth of human research on COVID-19 is increasing, making the investigation of medical images a focal point. Image segmentation, a crucial step preceding image processing, holds significance in the realm of medical image analysis. Traditional threshold image segmentation proves to be less efficient, posing challenges in selecting an appropriate threshold value. In response to these issues, this paper introduces Inner-based multi-strategy particle swarm optimization (IPSOsono) for conducting numerical experiments and enhancing threshold image segmentation in COVID-19 medical images. A novel dynamic oscillatory weight, derived from the PSO variant for single-objective numerical optimization (PSOsono) is incorporated. Simultaneously, the historical optimal positions of individuals in the particle swarm undergo random updates, diminishing the likelihood of algorithm stagnation and local optima. Moreover, an inner selection learning mechanism is proposed in the update of optimal positions, dynamically refining the global optimal solution. In the CEC 2013 benchmark test, PSOsono demonstrates a certain advantage in optimization capability compared to algorithms proposed in recent years, proving the effectiveness and feasibility of PSOsono. In the Minimum Cross Entropy threshold segmentation experiments for COVID-19, PSOsono exhibits a more prominent segmentation capability compared to other algorithms, showing good generalization across 6 CT images and further validating the practicality of the algorithm.

Estimation single output with a hybrid of ANFIS and MOPSO_HS

In the field of soft computing, the Adaptive Neuro-Fuzzy Inference System (ANFIS) has been more well-liked in recent years for its predictive capabilities. Appropriate ANFIS parameter adjusting is critical, which creates a gap in its predictive integration with traditional optimization techniques. Although some academics have concentrated on incorporating single-objective optimization, they frequently encounter issues with reliability and stability when striving to solve problems. In this work, an innovative multi-objective optimization technique that integrates ANFIS with MOPSO_HS is introduced. The model has consistency in problem solving and shows accurate predictions for both odd and even interval input models. In addition, three actual datasets are used to demonstrate the effectiveness of the suggested model's integration. A comparison is made between the suggested integrated model and established algorithms after 20 runs of analysis. The algorithm's accuracy, stability, and dependability in resolving integration problems are demonstrated by the results, which also show how superior it is to alternative approaches.

Operation optimization and performance evaluation of a ventilated heating floor in a nearly-zero-energy building with mixed power

With the proportion of renewable energy in the construction sector increasing and the structure of energy systems becoming increasingly complex, it is crucial to optimize the operation strategy of building energy systems with thermal energy storage for improving multi-performance. However, conventional heuristic optimization algorithm exhibits low searching speed and converges slowly, which is hard for operation optimization of such systems. In this study, with a binary particle swarm optimization (BPSO) algorithm coupled with experience-based searching guiding strategy, the operation of a ventilated floor heating system in a nearly-zero-energy building was optimized . System performance under different optimization objectives and combinations, including reducing operation cost and carbon emission fee, meanwhile increasing wind power consumption and thermal comfort time, was investigated. Results show that with the experience-based accelerating strategy, the BPSO algorithm converged fast and found more reasonable solutions. In contrast, multi-objective optimization yielded better and diverse feasible solutions were found than single-objective optimization. By optimizing operating costs, wind power consumption, and environmental costs, the best comprehensive solution was obtained. The cost range of energy consumption and carbon emissions within 10 days is 19.24–22.17 yuan and 17.45–20.11 yuan, respectively. The proportion of thermal comfort time is 95.39–95.82%. However, due to high thermal insulation level, thermal comfort time under different optimization objectives showed little difference.

Improved Finite Element Model Updating of a Highway Viaduct Using Acceleration and Strain Data.

Most finite element model updating (FEMU) studies on bridges are acceleration-based due to their lower cost and ease of use compared to strain- or displacement-based methods, which entail costly experiments and traffic disruptions. This leads to a scarcity of comprehensive studies incorporating strain measurements. This study employed the strain- and acceleration-based FEMU analyses performed on a more than 50-year-old multi-span concrete highway viaduct. Mid-span strains under heavy vehicles were considered for the strain-based FEMU, and frequencies and mode shapes for the acceleration-based FEMU. The analyses were performed separately for up to three variables, representing Young's modulus adjustment factors for different groups of structural elements. FEMU studies considered residual minimisation and the error-domain model falsification (EDMF) methodology. The residual minimisation utilised four different single-objective optimisations focusing on strains, frequencies, and mode shapes. Strain- and frequency-based FEMU analyses resulted in an approximately 20% increase in the overall superstructure's design stiffness. This study shows the benefits of the intuitive EDMF over residual minimisation for FEMU, where information gained from the strain data, in addition to the acceleration data, manifests more sensible updated variables. EDMF finally resulted in a 25-50% overestimated design stiffness of internal main girders.

An adaptive differential evolution algorithm with multi-strategy for solving complex optimization problem

In this paper, an adaptive differential evolution algorithm with multi-strategy, namely ESADE is proposed to solve the premature convergence and high time complexity for complex optimization problem. In the ESADE, the population is divided into several sub-populations after the fitness value of each individual is sorted. Then different mutation strategies are proposed for different populations to balance the global exploration and local optimization. Next, a new self-adaptive strategy is designed adjust parameters to avoid falling into local optimum while the convergence accuracy has reached its maximum value. And a complex airport gate allocation multi-objective optimization model with the maximum flight allocation rate, the maximum near gate allocation rate, and the maximum passenger rate at near gate is constructed, which is divided into several single-objective optimization model. Finally, the ESADE is applied solve airport gate allocation optimization model. The experiment results show that the proposed ESADE algorithm can effectively solve the complex airport gate allocation problem and achieve ideal airport gate allocation results by comparing with the current common heuristic optimization algorithms.