检索结果-内蒙古大学图书馆

A differential evolution algorithm with diversity dynamic adjustment and two-phase constraint handling strategy for solving a pension fund investment problem under market uncertainty

引用

APPLIED SOFT COMPUTING 2025年 172卷

作者： Wu, Xiang Meng, Haozheng Yuan, Xiaolan Zheng, Qunxian Lin, Jinxing Zhang, Kanjian Guizhou Normal Univ Sch Math Sci Guiyang 550001 Peoples R China Southeast Univ Sch Elect Engn Nanjing 210096 Peoples R China Anhui Polytech Univ Sch Elect Engn Wuhu 241000 Peoples R China Nanjing Univ Posts & Telecommun Coll Automat Nanjing 210023 Peoples R China Southeast Univ Sch Automat Nanjing 210096 Peoples R China Southeast Univ Key Lab Measurement & Control CSE Minist Educ Nanjing 210096 Peoples R China

This paper considers the optimization of investment strategies (ISs) for pension fund under market uncertainty. The target for this problem is to maximize the overall desired wealth (ODW) at the closed of the planning period while considering the risk of market uncertainty. To begin with, this problem is modeled as a discrete- time stochastic dynamic optimization problem (DTSDOP) with hard constraints (HCs) and uncertain constraints (UCs). Different from existing models, the main goal is to maximize ODW, rather than maximizing the wealth for individual investors, and to make sure that the pension fund is able to fulfill all obligations with a desired probability, this model introduces UCs instead of constraints on the variance. Additionally, this model takes into account the risk of market uncertainty, which helps to compare the optimal ISs with and without market uncertainty. Then, a deterministic transformation method is proposed for converting DTSDOP to a deterministic problem with HCs by fully utilizing the information of DTSDOP. Following that, a diversity dynamic adjustment and two-phase constraint handling strategy-based differential evolution algorithm (DDA-TPCHS-DEA) is proposed for attaining a global optimum for the discrete-time deterministic dynamic optimization problem (DTDDOP), in which constraints and the cost function are balanced by a diversity dynamic adjustment strategy, and a two-phase constraint handling strategy is utilized for finding the boundary knowledge for feasible and infeasible domains. Finally, the superiority of the proposed method is illustrated utilizing 21 test functions, an optimal design problem for robot grippers, and a pension fund investment problem under market uncertainty. The source code of the proposed method and its supplementary material are available in the following GitHub repository: https://***/xwu-gznu/DDA-TPRHS-DEA

关键词： Pension fund Investment strategies Stochastic dynamic optimization Global optimum differential evolution algorithm

来源：评论

学校读者我要写书评

暂无评论

Stock portfolio optimization based on factor analysis and second-order memetic differential evolution algorithm

引用

MEMETIC COMPUTING 2024年第1期16卷 29-44页

作者： Han, Ning Chen, Yinnan Ye, Lingjuan Zhao, Xinchao Beijing Univ Posts & Telecommun Sch Sci Beijing Peoples R China

Portfolio optimization will apply the concept of diversification across asset classes, which means investing in a wide variety of asset types and classes for a risk-mitigation strategy. Portfolio optimization is a way to maximize net gains in a portfolio while minimizing risk. A portfolio means investing in a wide variety of asset types and classes for a risk-mitigation strategy by the investor. In this paper, factor analysis and cluster algorithm are used to screen stocks and an improved differential evolution algorithm for solving portfolio optimization model is proposed. By comprehensively analyzing the stock data with factor analysis and k-means clustering algorithm, it has found that important factors have important effect on stock price movement, and finally 10 stocks are selected with investment value. Besides, a Mean-Conditional Value at Risk (CVaR) model is constructed, which takes into account both the cost function and the diversification constraint. Finally, a second-order memetic differential evolution (SOMDE) algorithm is presented for solving the proposed model. The experiments show that the proposed SOMDE algorithm is valid for solving the Mean-CVaR model and that factor analysis for stock selection can benefit portfolio with higher return and less risk greatly.

关键词： Portfolio optimization Factor analysis Mean-CVaR model differential evolution algorithm Memetic algorithm

来源：评论

学校读者我要写书评

暂无评论

Two-Stage Deep Feature Selection Method Using Voting differential evolution algorithm for Pneumonia Detection From Chest X-Ray Images

引用

IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE 2025年第1期9卷 918-932页

作者： Ouyang, Haibin Liu, Dongmei Li, Steven Ding, Weiping Zhan, Zhi-Hui Guangzhou Univ Sch Mech & Elect Engn Guangzhou 510006 Peoples R China RMIT Univ Sch Business & Law Melbourne Vic 3000 Australia Nantong Univ Sch Artificial Intelligence & Comp Sci Nantong 226019 Peoples R China South China Univ Technol Sch Comp Sci & Engn Guangzhou 510006 Peoples R China

Chest X-ray images play a crucial role in pneumonia diagnosis, with deep transfer learning being a widely adopted method for pneumonia detection. However, effectively handling feature data extracted from deep models without succumbing to the challenges of feature dimensionality remains a formidable task. In response to this complex issue, we propose a novel two-stage deep feature selection (FS) method utilizing the voting differential evolution (VDE) algorithm. In this approach, a dimension adaptive search strategy is meticulously devised to ensure robust feature selection while concurrently reducing the dimension. To expedite the optimization process, we devise a CR adaptive adjustment method to enhance the efficiency of the algorithm. Notably, an important aspect of our approach is the introduction of a novel DE algorithm that integrates a voting mechanism. This synergistic fusion allows a comprehensive analysis of crucial feature relationships to mitigate the risk of algorithmic entrapment in local optima. Additionally, we propose a dynamic feature evaluation function to avert the oversight of feature sets with optimal classification accuracy during later stages of the algorithm, thereby preserving discriminative features. The method is verified on an open Chest X-Ray Images dataset, achieving 99.04% average precision, 98.67% average accuracy, 99.13% average recall, and 19.93% average feature dimension reduction ratio. The experimental findings reveal that the presented method outperform prevailing state-of-the-art algorithms.

关键词： Feature extraction Pneumonia X-ray imaging Heuristic algorithms Convolutional neural networks Accuracy Training differential evolution algorithm convolutional neural network optimization speed feature selection

来源：评论

学校读者我要写书评

暂无评论

Optimizing mixed-model assembly line efficiency under uncertain demand: A Q-Learning-Inspired differential evolution algorithm

引用

COMPUTERS & INDUSTRIAL ENGINEERING 2025年 200卷

作者： Meng, Kai Li, Shujuan Han, Zhoupeng Xian Univ Technol Sch Mech & Precis Instrument Engn Xian 710048 Shaanxi Peoples R China

Modern manufacturing heavily relies on mixed-model assembly lines to streamline production processes for various product configurations. However, most existing research in this area primarily focuses on deterministic demand scenarios, leaving the challenges posed by uncertain demand relatively unexplored. Such uncertainty can significantly impact assembly line efficiency, resource utilization, and throughput rates. This paper explores the complexities of balancing and sequencing in mixed-model assembly lines, particularly under conditions of uncertain demand. The proposed approach includes a robust mixed-integer linear programming model formulated to optimize production efficiency across diverse scenarios characterized by uncertain demand. To address this complex problem, a novel Q-Learning-Inspired differential evolution algorithm (QL-DE) has been developed. This algorithm utilizes a population-based evolutionary operator, an intra-population crossover operator, six task-centric and three product-centric neighborhood exploration operators, along with a Q-learning-inspired strategy. These components collectively enable the QL-DE algorithm to adaptively handle uncertain demand while optimizing assembly line processes. Finally, through a comparative analysis with five variants and five evolutionary algorithms, the QL-DE approach demonstrates its superior capability in efficiently addressing uncertain demand scenarios and optimizing the performance of mixed-model assembly lines.

关键词： Mixed-model assembly lines Uncertain demand Balancing and sequencing differential evolution algorithm Q -learning inspired strategy

来源：评论

学校读者我要写书评

暂无评论

Design optimization of sandwich composite cylinder using modified differential evolution algorithm

引用

MECHANICS BASED DESIGN OF STRUCTURES AND MACHINES 2025年

作者： Bhadra, Manash Kumar Vinod, G. Jain, Atul Vikram Sarabhai Space Ctr Space Transportat Syst Thiruvananthapuram India Indian Inst Technol Kharagpur Mech Engn Dept Kharagpur India

This study presents a modified differential evolution algorithm for optimizing sandwich composite cylinders under hydrostatic pressure. The algorithm enforces balanced laminates using a while loop to enforce balanced laminates in both the initial population and trial vectors, reducing computational demands and accelerating convergence. Unlike previous studies, it demonstrates another novelty by including the number of plies and core thickness as design variables, alongside ply angles and layup sequences, enabling practical, customizable optimization. The approach increases failure load by 35.4% while reducing mass and demonstrates robustness by tolerating stochastic layup errors, resulting in resilient designs with enhanced performance.

关键词： Sandwich structures differential evolution algorithm finite element analysis hydrostatic pressure composite

来源：评论

学校读者我要写书评

暂无评论

Three-stage multi-modal multi-objective differential evolution algorithm for vehicle routing problem with time windows

引用

INTELLIGENT DATA ANALYSIS 2024年第2期28卷 485-506页

作者： Zhang, Hai-Fei Ge, Hong-Wei Li, Ting Su, ShuZhi Tong, YuBing Jiangnan Univ Sch Artificial Intelligence & Comp Sci Wuxi 214122 Jiangsu Peoples R China Jiangnan Univ Minist Educ Key Lab Adv Proc Control Light Ind Wuxi Jiangsu Peoples R China Anhui Univ Sci Technol Sch Comp Sci & Engn Huainan Anhui Peoples R China Univ Penn Dept Radiol Philadelphia PA USA

In this paper, the mathematical model of Vehicle Routing Problem with Time Windows (VRPTW) is established based on the directed graph, and a 3-stage multi-modal multi-objective differential evolution algorithm (3S-MMDEA) is proposed. In the first stage, in order to expand the range of individuals to be selected, a generalized opposition-based learning (GOBL) strategy is used to generate a reverse population. In the second stage, a search strategy of reachable distribution area is proposed, which divides the population with the selected individual as the center point to improve the convergence of the solution set. In the third stage, an improved individual variation strategy is proposed to legalize the mutant individuals, so that the individual after variation still falls within the range of the population, further improving the diversity of individuals to ensure the diversity of the solution set. Based on the synergy of the above three stages of strategies, the diversity of individuals is ensured, so as to improve the diversity of solution sets, and multiple equivalent optimal paths are obtained to meet the planning needs of different decision-makers. Finally, the performance of the proposed method is evaluated on the standard benchmark datasets of the problem. The experimental results show that the proposed 3S-MMDEA can improve the efficiency of logistics distribution and obtain multiple equivalent optimal paths. The method achieves good performance, superior to the most advanced VRPTW solution methods, and has great potential in practical projects.

关键词： Logistics distribution vehicle routing problem multi-modal multi-objective optimization three-stage strategy differential evolution algorithm

来源：评论

学校读者我要写书评

暂无评论

Improvement of the Approximate Method for Determining the Average Vertical Stress Increase Below the Rectangular Foundation Using differential evolution algorithm

引用

ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING 2025年 1-20页

作者： Dagdeviren, Ugur Kutahya Dumlupinar Univ Fac Engn Dept Civil Engn Kutahya Turkiye

External loads transferred from the structure's foundations to the soil induce stress increases in the soil stratum. Since stress increases within the soil mass vary with depth and across the plane at a given depth, approaches that estimate the average stress increase under foundations can be advantageous for effective foundation design. This study aims to develop optimization-based approximate methods for calculating average vertical stress increases with higher accuracy than the conventional 2V:1H method for rectangular foundations with different L/B ratios. For this purpose, vertical stress increases within the foundation projection at 120 different depths for 12 different L/B ratios were numerically calculated using Boussinesq's stress expressions. The model parameters of the proposed approximate models, such as expansion slopes (k or k1, k2) and normalized critical depth (zcr/B), for each L/B ratio were optimized using the differential evolution algorithm. The proposed three-parameter approximate method achieved the highest accuracy, reducing the RMSE values by an average of 53% compared to the conventional method, while the one-parameter model reduced the RMSE by 9%. The maximum absolute errors in the three-parameter model remained between 0.0217 and 0.0283, with R2 values greater than 0.9972. Building upon and improving the conventional method, this study presents a practical and novel three-parameter method that provides a more reliable and accurate estimation of the average vertical stress increase under flexible rectangular foundations, significantly reducing errors. This study contributes to geotechnical engineering by improving the accuracy of stress increase prediction models, potentially leading to more economical and safer foundation designs.

关键词： Vertical stress increase Influence factor Rectangular foundation Approximate method differential evolution algorithm

来源：评论

学校读者我要写书评

暂无评论

Fitness and Collaborative Information-Driven differential evolution algorithm for Bound-Constrained and Real-World Application Problems

引用

ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING 2025年 1-35页

作者： Gupta, Shubham Singh, Shitu Gao, Shangce Motilal Nehru Natl Inst Technol Allahabad Dept Math Prayagraj 211004 India United Univ Dept Appl Sci Prayagraj 211012 Uttar Pradesh India Univ Toyama Fac Engn Toyama 9308555 Japan

The effectiveness of differential evolution (DE) is significantly impacted by the selection of the mutation operator and the setup of control parameters. However, their unique selection might not ensure the optimized search procedure that enables the algorithm to explore for a global optimal solution for the given optimization problem. Further, with this selection approach, the algorithm might suffer from the issues of getting stuck at local optima and premature convergence. To address these challenges, this paper proposes a new framework of the DE called fitness and collaborative information-driven DE (COLDE). In the COLDE, a novel mutation operator is proposed to strengthen the collaboration among elite and non-elite candidate solutions so that more promising offspring vectors can be generated. The scale factor parameters are adjusted according to the evolutionary state of candidate solutions engaged in the mutation operator, while the crossover operator is tuned based on the success rates of crossover parameters determined in the past evolutionary stage. Moreover, the population size is also reduced over the generations to discard the unfavorable candidate solutions. The validation of the proposed COLDE is conducted on the standard set of benchmark problems provided by the IEEE CEC2017 of real-parameter single-objective problems and eight real-world engineering optimization problems. A comparison of COLDE with other evolutionary algorithms using diverse performance metrics verifies its promising and competitive search efficiency against the compared algorithms.

关键词： Global optimization problems differential evolution algorithm Mutation operator Scale factor Crossover rate

来源：评论

学校读者我要写书评

暂无评论

System Identification of Dual-rate Hammerstein CARARMA Systems by Multi-innovation Improved differential evolution algorithm

引用

INTERNATIONAL JOURNAL OF CONTROL AUTOMATION AND SYSTEMS 2024年第12期22卷 3792-3800页

作者： Zong, Tiancheng Li, Junhong Lu, Guoping Nantong Univ Sch Elect Engn & Automat Nantong 226019 Peoples R China

This study focuses on the parameter estimation of the dual-rate Hammerstein-controlled autoregressive moving average (CARARMA) system. Because dual-rate systems have variables that cannot be directly measured, we use the auxiliary model to obtain a transformed model which can be directly identified. Aiming at the dual-rate Hammerstein CARARMA system, a multi-innovation improved differential evolution identification method is proposed, and its effectiveness is experimentally verified. By comparing with the improved differential evolution algorithm, it can be found that the proposed algorithm can give more accurate parameter estimates.

关键词： Auxiliary model differential evolution algorithm dual-rate sampling Hammerstein systems multi-innovation identification system identification

来源：评论

学校读者我要写书评

暂无评论

Automatic Optimal Design Method for Circuit Sizing Based on CNN Surrogate Model Assisted differential evolution algorithm

引用

IEEE ACCESS 2024年 12卷 136238-136247页

作者： Tang, Chaoying Chen, Xiaofei Luo, Yanshen Zeng, Yanhan Guangzhou Univ Sch Elect & Commun Engn Guangzhou 510006 Peoples R China

The physical characteristics of analog ICs are intricate, leading to prolonged design and development cycles. Currently, the automatic optimization of analog integrated circuit size heavily relies on simulations, which unfortunately results in a significant waste of simulation resources during the optimization process. To address this issue, this paper introduces a novel optimization framework named "MODE-CNN", which integrates Multi-Objective differential evolution (MODE) algorithm and Convolutional Neural Network (CNN) surrogate models. The MODE-CNN framework is designed to reduce unnecessary consumption of simulation resources while maintaining high optimization accuracy. Within the MODE-CNN framework, we first enhance the predictive capabilities of the surrogate models using Latin Hypercube Sampling and Quantile Transformation. Subsequently, we employ CNN surrogate models to evaluate circuit performance during the optimization process, effectively screening out promising designs and substantially reducing the number of simulations required. Moreover, we propose an improved fast non-dominated sorting method that further enhances the global optimization performance of the MODE algorithm. Through testing on three circuit design cases and comparing MODE-CNN with existing mainstream methods and expert manual designs, we find that MODE-CNN not only significantly reduces the number of simulations (by up to 63%), but also surpasses traditional methods in optimization outcomes, with performance improvements of up to 95.09% for TC and 58.75% for LS. The experimental results validate the effectiveness and advancement of the MODE-CNN framework in the optimization of analog integrated circuits.

关键词： Analog circuit design automated circuit design convolutional neural network differential evolution algorithm differential evolution algorithm simulation-based optimization simulation-based optimization differential evolution algorithm simulation-based optimization

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：