检索结果-内蒙古大学图书馆

9th International Conference on Natural Computation (ICNC)

作者： Wang, Xiaohong Su, Qinghua Miao, Yongfei Wuhan Business Univ Wuhan 430074 Hubei Peoples R China Hubei Engn Univ Sch Math & Stat Xiaogan 432000 Peoples R China Wuhan Univ Technol Wuhan Peoples R China

ISBN: (纸本)9781467347143

A geometric program (GP) is a type of optimization problem characterized by objective and constraint functions that have a special form. By combining with a simple constraint-handling mechanism based on feasibility rules, a new approach based on a differential evolution algorithm is proposed for solving the GP problem in this paper. Numerical simulation experiments show the validity of the method, and comparative studies expose the proposed algorithm as a competitive approach for geometric programming problems.

关键词： constraint-handling mechanism differential evolution algorithm geometric programming

来源：评论

学校读者我要写书评

暂无评论

DE-MHAIPs: Identification of SARS-CoV-2 phosphorylation sites based on differential evolution multi-feature learning and multi-head attention mechanism

引用

COMPUTERS IN BIOLOGY AND MEDICINE 2023年第1期160卷 106935-106935页

作者： Wang, Minghui Yan, Lu Jia, Jihua Lai, Jiali Zhou, Hongyan Yu, Bin Qingdao Univ Sci & Technol Coll Math & Phys Qingdao 266061 Peoples R China Qingdao Univ Sci & Technol Coll Informat Sci & Technol Sch Data Sci Qingdao 266061 Peoples R China Univ Sci & Technol China Sch Data Sci Hefei 230027 Peoples R China

The rapid spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) around the world affects the normal lives of people all over the world. The computational methods can be used to accurately identify SARS-CoV-2 phosphorylation sites. In this paper, a new prediction model of SARS-CoV-2 phosphorylation sites, called DE-MHAIPs, is proposed. First, we use six feature extraction methods to extract protein sequence information from different perspectives. For the first time, we use a differential evolution (DE) algorithm to learn individual feature weights and fuse multi-information in a weighted combination. Next, Group LASSO is used to select a subset of good features. Then, the important protein information is given higher weight through multi-head attention. After that, the processed data is fed into long short-term memory network (LSTM) to further enhance model's ability to learn features. Finally, the data from LSTM are input into fully connected neural network (FCN) to predict SARS-CoV-2 phosphorylation sites. The AUC values of the S/T and Y datasets under 5 -fold cross-validation reach 91.98% and 98.32%, respectively. The AUC values of the two datasets on the inde-pendent test set reach 91.72% and 97.78%, respectively. The experimental results show that the DE-MHAIPs method exhibits excellent predictive ability compared with other methods.

关键词： SARS-CoV-2 phosphorylation sites Multi -view feature fusion differential evolution algorithm Group LASSO Multi -head attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Dynamic Pricing and System Optimization of Integrated Energy System in Industrial Parks Based on Carbon Tax 20

Dynamic Pricing and System Optimization of Integrated Energy...

引用

20th International Conference on the European Energy Market (EEM)

作者： Deng, Yuanyuan Mae, Masahiro Matsuhashi, Ryuji Univ Tokyo Grad Sch Engn Dept Elect Engn & Informat Syst Tokyo Japan

ISBN: (纸本)9798350381757;9798350381740

Aiming to improve the efficiency of the industry park's integrated energy system and reduce carbon emissions, this paper adopts dynamic pricing and carbon tax mechanisms. Utilizing game theory, a leader-follower game framework is constructed with the energy system operator (ESO) as the leader and building users as followers. Through the integration of dynamic pricing and carbon tax model, the paper demonstrates notable improvements in economic and low-carbon performance within the energy system. Consequently, in this paper, it provides valuable support for sustainable energy development and economic operations in a low-carbon economy. The findings underscore the potential of dynamic pricing and carbon tax mechanisms as integral contributors to achieving a more sustainable and economically efficient energy landscape.

关键词： Carbon Tax Demand Response differential evolution algorithm Dynamic Electricity Prices Leader-follower Game

来源：评论

学校读者我要写书评

暂无评论

Extended belief rule-based method for the assessment of matching between rescue tasks and volunteers

引用

KYBERNETES 2023年第12期52卷 6521-6541页

作者： Chen, Li Chen, Sheng-Qun Yang, Long-Hao Fujian Business Univ Sch Informat Engn Fuzhou Peoples R China Fuzhou Univ Decis Sci Inst Fuzhou Peoples R China

Purpose This paper aims to solve the major assessment problem in matching the satisfaction of psychological gratification and mission accomplishment pertaining to volunteers with the disaster rescue and recovery tasks. Design/methodology/approach An extended belief rule-based (EBRB) method is applied with the method's input and output parameters classified based on expert knowledge and data from literature. These parameters include volunteer self-satisfaction, experience, peer-recognition, and cooperation. First, the model parameters are set;then, the parameters are optimized through data envelopment analysis (DEA) and differential evolution (DE) algorithm. Finally, a numerical mountain rescue example and comparative analysis between with-DEA and without-DEA are presented to demonstrate the efficiency of the proposed method. The proposed model is suitable for a two-way matching evaluation between rescue tasks and volunteers. Findings Disasters are unexpected events in which emergency rescue is crucial to human survival. When a disaster occurs, volunteers provide crucial assistance to official rescue teams. This paper finds that decision-makers have a better understanding of two-sided match objects through bilateral feedback over time. With the changing of the matching preference information between rescue tasks and volunteers, the satisfaction of volunteer's psychological gratification and mission accomplishment are also constantly changing. Therefore, considering matching preference information and satisfaction at two-sided match objects simultaneously is necessary to get reasonable target values of matching results for rescue tasks and volunteers. Originality/value Based on the authors' novel EBRB method, a matching assessment model is constructed, with two-sided matching of volunteers to rescue tasks. This method will provide matching suggestions in the field of emergency dispatch and contribute to the assessment of emergency plans around the world.

关键词： Data envelopment analysis differential evolution algorithm Extended belief rule-based method Matching assessment

来源：评论

学校读者我要写书评

暂无评论

RF Chain Selection Using Hybrid Optimization with Precoding in mm-Wave Massive MIMO Systems

引用

WIRELESS PERSONAL COMMUNICATIONS 2023年第3期131卷 1997-2017页

作者： Srinivas, Ch V. V. S. Borugadda, Somasekhar Andhra Univ Coll Engn AUCE A ECE Dept Visakhapatnam 530003 Andhra Pradesh India Vishnu Inst Technol ECE Dept Bhimavaram 534202 Andhra Pradesh India Anil Neerukonda Inst Technol & Sci ECE Dept Visakhapatnam 531162 Andhra Pradesh India

Energy efficiency is considered the prominent metric in data communication systems and has a greater impact on designing Millimetre-wave (mm-wave) multiple input multiple output (MIMO) systems. As the recent mm-wave system relies upon the number of radiofrequency (RF) chains, effective RF chains are preferred to maximize overall system performance, energy and spectral efficiency. Degraded system efficiency and increased path loss are the major limitations faced in MIMO systems. This harms the data rate transmission between transmitter and receiver antennas. Hence, an effectual mm-wave with a massive MIMO model is presented in this research by adopting standard techniques. The chief objective of this research is to maximize energy and spectral efficiency with an appropriate selection of RF chains. The optimal RF chains are selected through a hybrid optimization model called the differential evolution firefly selection model (DEA + FOA). After selecting optimal RF chains, an appropriate precoding scheme is used to route the transmitted data over the preferred direction by overcoming the path-loss issues. In the precoding step, the successive interference cancellation based technique is employed to optimize the RF chain's sub-capacity and increase the system efficiency. The performance of a proposed model is analyzed by evaluating the energy and spectral efficiency by varying the number of RF chains at low and high signal-to-noise ratio rates. The proposed optimization and precoding techniques are compared with existing approaches to analyze the efficacy of a proposed MIMO system. The proposed model has proven extremely effective in transforming data from transmitter to receiver with increased system efficiency.

关键词： mm-wave communication RF chain selection Digital to analog converter Firefly optimization algorithm SIC precoding differential evolution algorithm Energy-spectral efficiency

来源：评论

学校读者我要写书评

暂无评论

Routing and Scheduling in Time-Sensitive Networking by evolutionary algorithms

引用

BIOMIMETICS 2025年第5期10卷 333-333页

作者： Wang, Zengkai Liao, Weizhi Xia, Xiaoyun Wang, Zijia Duan, Yaolong Jiaxing Univ Sch Artificial Intelligence Jiaxing 314001 Peoples R China Xuelong Grp Co Ltd Technol Res & Dev Ctr Ningbo 315899 Peoples R China Guangzhou Univ Sch Comp Sci & Cyber Engn Guangzhou 510006 Peoples R China

Routing and scheduling in Time-Sensitive Networking (TSN) is an NP-hard problem. In this paper, we propose a novel routing and scheduling approach for TSN based on evolutionary algorithm. Specifically, we introduce a flow grouping method that leverages the greatest common divisor to optimize flow aggregation. On this basis, we develop a flow routing strategy that employs a genetic algorithm, where the evaluation function considers not only flow combinability but also path length and network load. By exploiting the non-combinable properties of flows, we effectively reduce the search space for the genetic algorithm. Furthermore, we design a scheduling method based on differential evolution algorithms tailored to TSN's requirements of zero jitter and no frame loss. We propose a gene coding method and rigorously prove its correctness, which significantly reduces the search space of the differential evolution algorithm. The experimental results demonstrate that our approach enables more flows to traverse along the shortest path compared to both k-shortest path methods and integer linear programming approaches, while achieving a faster execution time in large-scale scheduling scenarios.

关键词： Time-Sensitive Networking routing and scheduling evolutionary algorithm differential evolution algorithm

来源：评论

学校读者我要写书评

暂无评论

An Improved differential evolution algorithm Based on the Sector Window Floating Zoom Immune Random Search Operator

An Improved Differential Evolution Algorithm Based on the Se...

引用

9th International Conference on Natural Computation (ICNC)

作者： Hu, Shaohua Zhong, Sheng Jiangxi Radio & TV Univ Nanchang Branch Nanchang 330000 Jiangxi Peoples R China Hainan Univ Coll Informat Sci & Technol Haikou 570228 Hainan Peoples R China

ISBN: (纸本)9781467347143

In this paper, an improved differential evolution(DE) algorithm is proposed. The Sector Window Floating Zoom Immune Random Search (SWFZIRS) operator has been utilized in the fitness calculation. The position of the individual of best fitness is recorded. During the selecting operation, each individual will move toward the position of best fitness individual if the new individual's fitness is not as good as the fitness of the individual before crossover. Experiments show that the algorithm can increase the speed of convergence and increase the probability of obtaining the global optimal solution.

关键词： differential evolution algorithm SWFZIRS operator evolutionary algorithm

来源：评论

学校读者我要写书评

暂无评论

Penalized variable selection in multi-parameter regression survival modeling

引用

STATISTICAL METHODS IN MEDICAL RESEARCH 2023年第12期32卷 2455-2471页

作者： Jaouimaa, Fatima-Zahra Do Ha, Il Burke, Kevin Univ Limerick Dept Math & Stat Limerick Ireland Pukyong Natl Univ Dept Stat Pusan South Korea Univ Limerick Dept Math & Stat Limerick V94T9PX Ireland

Standard survival models such as the proportional hazards model contain a single regression component, corresponding to the scale of the hazard. In contrast, we consider the so-called "multi-parameter regression" approach whereby covariates enter the model through multiple distributional parameters simultaneously, for example, scale and shape parameters. This approach has previously been shown to achieve flexibility with relatively low model complexity. However, beyond a stepwise type selection method, variable selection methods are underdeveloped in the multi-parameter regression survival modeling setting. Therefore, we propose penalized multi-parameter regression estimation procedures using the following penalties: least absolute shrinkage and selection operator, smoothly clipped absolute deviation, and adaptive least absolute shrinkage and selection operator. We compare these procedures using extensive simulation studies and an application to data from an observational lung cancer study;the Weibull multi-parameter regression model is used throughout as a running example.

关键词： Variable selection multi-parameter regression Weibull penalized maximum likelihood differential evolution algorithm

来源：评论

学校读者我要写书评

暂无评论

Developing hybrid metaheuristics for the three-dimensional pallet loading considering multiple-size pallet and heterogeneous box

引用

INTERNATIONAL JOURNAL OF SYSTEMS SCIENCE-OPERATIONS & LOGISTICS 2025年第1期12卷

作者： Tsao, Yu-Chung Chen, Ting-Jie Vu, Thuy-Linh Liao, Lu-Wen Natl Taiwan Univ Sci & Technol Dept Ind Management 43Sec 4Keelung Rd Taipei City 106335 Taiwan Natl Taiwan Univ Sci & Technol Artificial Intelligence Operat Management Res Ctr 43Sec 4Keelung Rd Taipei City 106335 Taiwan Taiwan Semicond Mfg Co TSMC Hsinchu Taiwan Feng Chia Univ Dept Business Adm Taichung Taiwan Natl Taichung Univ Sci & Technol Dept Intelligent Prod Engn Taichung Taiwan

Intelligent operations in logistics warehousing play a crucial role in optimising efficiency and enhancing productivity. As logistics operations become increasingly complex and scaled, automation has been widely adopted in streamlined processes such as automatic picking, resulting in improved efficiency and order accuracy. However, the stacking operation in warehousing still relies heavily on manual labour and experience in loading boxes onto the pallet. By integrating metaheuristic algorithms and automated equipment such as robotic arms, the intelligence of pallet loading can be realised. This study focuses on the three-dimensional distributor pallet loading problem (3D-DPLP) with multiple-sized pallets and practical constraints. A two-phase approach is proposed to solve the problem while minimising the number of pallets required. Firstly, Fast Loading Heuristic is developed to generate an initial solution. Subsequently, a simulated annealing-based algorithm is used to refine the solution by searching for loading patterns. Furthermore, we propose improved and hybrid metaheuristic algorithms that integrate simulated annealing and differential evaluation to accelerate convergence and enhance exploration capability.

关键词： Logistics differential evolution algorithm distributor's pallet loading problem multi-size pallet simulate annealing

来源：评论

学校读者我要写书评

暂无评论

New Greedy Randomized Adaptive Search Procedure based on differential evolution algorithm for solving no-wait flowshop scheduling problem

New Greedy Randomized Adaptive Search Procedure based on Dif...

引用

International Conference on Advanced Logistics and Transport (ICALT)

作者： Akrout, Hanen Jarboui, Bassem Rebai, Abdelwaheb Siarry, Patrick MODILS Fac Sci Econ & Gest Sfax Tunisia

ISBN: (纸本)9781479903139;9781479903146

Scheduling in a production workshop consists of assigning a number of different tasks to the existing machines with respect to the constraints of the problem in order to optimise the objective function. In this paper, we focus on the No-Wait Flow Shop Scheduling Problem (NWFSSP) with the makespan objective. We present a new hybrid algorithm NEW-GRASP-DE that combines Greedy Randomised Adaptive Search Procedure (GRASP) with differential evolution algorithm (DEA). Furthermore, we used an Iterative Local Search (ILS) as an improvement phase in the GRASP method. Our algorithm is tested on instances that have been proposed in the literature. We compared our results with those of different authors. The experimental results show the effectiveness of our algorithm.

关键词： no-wait flowshop greedy randomized adaptive search procedure differential evolution algorithm makespan

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：