The differential evolution algorithm is an evolutionary algorithm for global optimization and the un-capacitated facility location problem (UFL) is one of the classic NP-Hard problems. In this paper, combined with the...The differential evolution algorithm is an evolutionary algorithm for global optimization and the un-capacitated facility location problem (UFL) is one of the classic NP-Hard problems. In this paper, combined with the specific characteristics of the UFL problem, we introduce the activation function to the algorithm for solving UFL problem and name it improved adaptive differential evolution algorithm (IADEA). Next, to improve the efficiency of the algorithm and to alleviate the problem of being stuck in a local optimum, an adaptive operator was added. To test the improvement of our algorithm, we compare the IADEA with the basic differential evolution algorithm by solving typical instances of UFL problem respectively. Moreover, to compare with other heuristic algorithm, we use the hybrid ant colony algorithm to solve the same instances. The computational results show that IADEA improves the performance of the basic DE and it outperforms the hybrid ant colony algorithm.展开更多
This article mainly investigates the fuzzy optimization robust control issue for nonlinear networked systems characterized by the interval type-2(IT2)fuzzy technique under a differential evolution algorithm.To provide...This article mainly investigates the fuzzy optimization robust control issue for nonlinear networked systems characterized by the interval type-2(IT2)fuzzy technique under a differential evolution algorithm.To provide a more reasonable utilization of the constrained communication channel,a novel adaptive memory event-triggered(AMET)mechanism is developed,where two event-triggered thresholds can be dynamically adjusted in the light of the current system information and the transmitted historical data.Sufficient conditions with less conservative design of the fuzzy imperfect premise matching(IPM)controller are presented by introducing the Wirtinger-based integral inequality,the information of membership functions(MFs)and slack matrices.Subsequently,under the IPM policy,a new MFs intelligent optimization technique that takes advantage of the differential evolution algorithm is first provided for IT2 TakagiSugeno(T-S)fuzzy systems to update the fuzzy controller MFs in real-time and achieve a better system control effect.Finally,simulation results demonstrate that the proposed control scheme can obtain better system performance in the case of using fewer communication resources.展开更多
Map building by multi-robot is very important to accomplish autonomous navigation,and one of the basic problems and research hotspots is how to merge the maps into a single one in the field of multi-robot map building...Map building by multi-robot is very important to accomplish autonomous navigation,and one of the basic problems and research hotspots is how to merge the maps into a single one in the field of multi-robot map building.A novel approach is put forward based on adaptive differential evolution to map building for the multi-robot system.The multi-robot mapping-building system adopts the methods of decentralized exploration and concentrated mapping.The adaptive differential evolution algorithm is used to search in the space of possible transformation,and the iterative search is performed with the goal of maximizing overlapping regions.The map is translated and rotated so that the two maps can be overlapped and merged into a single global one successfully.This approach for map building can be realized without any knowledge of their relative positions.Experimental results show that the approach is effective and feasibile.展开更多
To implement self-adaptive control parameters, a hybrid differential evolution algorithm integrated with particle swarm optimization （PSODE） is proposed. In the PSODE, control parameters are encoded to be a symbioti...To implement self-adaptive control parameters, a hybrid differential evolution algorithm integrated with particle swarm optimization （PSODE） is proposed. In the PSODE, control parameters are encoded to be a symbiotic individual of original individual, and each original individual has its own symbiotic individual. Differential evolution （ DE） operators are used to evolve the original population. And, particle swarm optimization （PSO） is applied to co-evolving the symbiotic population. Thus, with the evolution of the original population in PSODE, the symbiotic population is dynamically and self-adaptively adjusted and the realtime optimum control parameters are obtained. The proposed algorithm is compared with some DE variants on nine functious. The results show that the average performance of PSODE is the best.展开更多
Differential evolution algorithm based on the covariance matrix learning can adjust the coordinate system according to the characteristics of the population, which make<span style="font-family:Verdana;"&g...Differential evolution algorithm based on the covariance matrix learning can adjust the coordinate system according to the characteristics of the population, which make<span style="font-family:Verdana;">s</span><span style="font-family:Verdana;"> the search move in a more favorable direction. In order to obtain more accurate information about the function shape, this paper propose</span><span style="font-family:Verdana;">s</span><span style="font-family:;" "=""> <span style="font-family:Verdana;">covariance</span><span style="font-family:Verdana;"> matrix learning differential evolution algorithm based on correlation (denoted as RCLDE)</span></span><span style="font-family:;" "=""> </span><span style="font-family:Verdana;">to improve the search efficiency of the algorithm. First, a hybrid mutation strategy is designed to balance the diversity and convergence of the population;secondly, the covariance learning matrix is constructed by selecting the individual with the less correlation;then, a comprehensive learning mechanism is comprehensively designed by two covariance matrix learning mechanisms based on the principle of probability. Finally,</span><span style="font-family:;" "=""> </span><span style="font-family:;" "=""><span style="font-family:Verdana;">the algorithm is tested on the CEC2005, and the experimental results are compared with other effective differential evolution algorithms. The experimental results show that the algorithm proposed in this paper is </span><span style="font-family:Verdana;">an effective algorithm</span><span style="font-family:Verdana;">.</span></span>展开更多
DV-Hop localization algorithm has greater localization error which estimates distance from an unknown node to the different anchor nodes by using estimated average size of a hop to achieve the location of the unknown ...DV-Hop localization algorithm has greater localization error which estimates distance from an unknown node to the different anchor nodes by using estimated average size of a hop to achieve the location of the unknown node.So an improved DV-Hop localization algorithm based on correctional average size of a hop,HDCDV-Hop algorithm,is proposed.The improved algorithm corrects the estimated distance between the unknown node and different anchor nodes based on fractional hop count information and relatively accurate coordinates of the anchor nodes information,and it uses the improved Differential Evolution algorithm to get the estimate location of unknown nodes so as to further reduce the localization error.Simulation results show that our proposed algorithm have lower localization error and higher localization accuracy compared with the original DV-Hop algorithm and other classical improved algorithms.展开更多
To tackle the path planning problem,this study introduced a novel algorithm called two-stage parameter adjustment-based differential evolution(TPADE).This algorithm draws inspiration from group behavior to implement a...To tackle the path planning problem,this study introduced a novel algorithm called two-stage parameter adjustment-based differential evolution(TPADE).This algorithm draws inspiration from group behavior to implement a two-stage scaling factor variation strategy.In the initial phase,it adapts according to environmental complexity.In the following phase,it combines individual and global experiences to fine-tune the orientation factor,effectively improving its global search capability.Furthermore,this study developed a new population update method,ensuring that well-adapted individuals are retained,which enhances population diversity.In benchmark function tests across different dimensions,the proposed algorithm consistently demonstrates superior convergence accuracy and speed.This study also tested the TPADE algorithm in path planning simulations.The experimental results reveal that the TPADE algorithm outperforms existing algorithms by achieving path lengths of 28.527138 and 31.963990 in simple and complex map environments,respectively.These findings indicate that the proposed algorithm is more adaptive and efficient in path planning.展开更多
为进一步提高分割精度并加快分割速度,提出了一种基于邻域搜索可选外部存档自适应差分进行算法(简称为JADE-GL)的二维Otsu多阈值图像分割方案。首先,针对原始JADE算法精英突变策略收敛速度慢、容易陷入局部最优等问题,提出了基于邻域半...为进一步提高分割精度并加快分割速度,提出了一种基于邻域搜索可选外部存档自适应差分进行算法(简称为JADE-GL)的二维Otsu多阈值图像分割方案。首先,针对原始JADE算法精英突变策略收敛速度慢、容易陷入局部最优等问题,提出了基于邻域半径搜索的改进突变策略,以提升算法的全局探索和开发能力。然后,将所提算法与现有分割方法以及其他JADE变种算法进行二维Otsu多阈值分割对比实验。最后,通过函数收敛曲线、分割距离测度、峰值信噪比(peak signal to noise ratio,PSNR)等指标定量分析算法的性能。实验结果表明,随着阈值数增加,提出的算法无论在收敛速度、分割精度还是分割图像效果上都有显著提升。展开更多
针对同步热传递搜索(Simultaneous Heat Transfer Search,SHTS)算法存在的收敛速度慢、优化性能不足及易陷入局部最优等缺陷,提出了一种融合差分进化机制的自适应同步热传递搜索算法(DE-SHTS)。该算法以热传递的传导、对流、辐射三阶段...针对同步热传递搜索(Simultaneous Heat Transfer Search,SHTS)算法存在的收敛速度慢、优化性能不足及易陷入局部最优等缺陷,提出了一种融合差分进化机制的自适应同步热传递搜索算法(DE-SHTS)。该算法以热传递的传导、对流、辐射三阶段为基础框架,嵌入差分进化中的变异-交叉机制,并提出自适应参数调整策略,从而实现了快速收敛、优良的优化性能,并有效平衡了全局探索与局部开采的动态需求。为验证该算法的性能,在CEC2017测试集上与传统SHTS算法进行对比,从最优适应值、平均适应值和收敛速度等方面进行分析,进一步验证了DE-SHTS算法的有效性。展开更多
The search efficiency of differential evolution （DE） algorithm is greatly impacted by its control parameters. Although many adaptation/self-adaptation techniques can automatically find suitable control parameters fo...The search efficiency of differential evolution （DE） algorithm is greatly impacted by its control parameters. Although many adaptation/self-adaptation techniques can automatically find suitable control parameters for the DE, most techniques are based on pop- ulation information which may be misleading in solving complex optimization problems. Therefore, a self-adaptive DE （i.e., JADE） using two-phase parameter control scheme （TPC-JADE） is proposed to enhance the performance of DE in the current study. In the TPC-JADE, an adaptation technique is utilized to generate the control parameters in the early population evolution, and a well-known empirical guideline is used to update the control parameters in the later evolution stages. The TPC-JADE is compared with four state-of-the-art DE variants on two famous test suites （i.e., IEEE CEC2005 and IEEE CEC2015）. Results indicate that the overall performance of the TPC-JADE is better than that of the other compared algorithms. In addition, the proposed algorithm is utilized to obtain optimal nutrient and inducer feeding for the Lee-Ramirez bioreactor. Experimental results show that the TPC-JADE can perform well on an actual dynamic optimization problem.展开更多
文摘The differential evolution algorithm is an evolutionary algorithm for global optimization and the un-capacitated facility location problem (UFL) is one of the classic NP-Hard problems. In this paper, combined with the specific characteristics of the UFL problem, we introduce the activation function to the algorithm for solving UFL problem and name it improved adaptive differential evolution algorithm (IADEA). Next, to improve the efficiency of the algorithm and to alleviate the problem of being stuck in a local optimum, an adaptive operator was added. To test the improvement of our algorithm, we compare the IADEA with the basic differential evolution algorithm by solving typical instances of UFL problem respectively. Moreover, to compare with other heuristic algorithm, we use the hybrid ant colony algorithm to solve the same instances. The computational results show that IADEA improves the performance of the basic DE and it outperforms the hybrid ant colony algorithm.
基金supported by the National Natural Science Foundation of China(61973105,62373137)。
文摘This article mainly investigates the fuzzy optimization robust control issue for nonlinear networked systems characterized by the interval type-2(IT2)fuzzy technique under a differential evolution algorithm.To provide a more reasonable utilization of the constrained communication channel,a novel adaptive memory event-triggered(AMET)mechanism is developed,where two event-triggered thresholds can be dynamically adjusted in the light of the current system information and the transmitted historical data.Sufficient conditions with less conservative design of the fuzzy imperfect premise matching(IPM)controller are presented by introducing the Wirtinger-based integral inequality,the information of membership functions(MFs)and slack matrices.Subsequently,under the IPM policy,a new MFs intelligent optimization technique that takes advantage of the differential evolution algorithm is first provided for IT2 TakagiSugeno(T-S)fuzzy systems to update the fuzzy controller MFs in real-time and achieve a better system control effect.Finally,simulation results demonstrate that the proposed control scheme can obtain better system performance in the case of using fewer communication resources.
基金Supported by the National Natural Science Foundation of China(No.90820302,60805027)the Provincial Natural Science Foundation of Hunan(No.12JJ3064)+1 种基金the Construct Program of the Key Discipline in Hunan Province(No.201176)the Planned Science and Technology Project of Hunan Province(No.2011SK3135,2012FJ3059)
文摘Map building by multi-robot is very important to accomplish autonomous navigation,and one of the basic problems and research hotspots is how to merge the maps into a single one in the field of multi-robot map building.A novel approach is put forward based on adaptive differential evolution to map building for the multi-robot system.The multi-robot mapping-building system adopts the methods of decentralized exploration and concentrated mapping.The adaptive differential evolution algorithm is used to search in the space of possible transformation,and the iterative search is performed with the goal of maximizing overlapping regions.The map is translated and rotated so that the two maps can be overlapped and merged into a single global one successfully.This approach for map building can be realized without any knowledge of their relative positions.Experimental results show that the approach is effective and feasibile.
基金National Key Basic Research Project of China(973 program)(No.2013CB733600)National Natural Science Foundation of China(No.21176073)+1 种基金Program for New Century Excellent Talents in University,China(No.NCET-09-0346)the Fundamental Research Funds for the Central Universities,China
文摘To implement self-adaptive control parameters, a hybrid differential evolution algorithm integrated with particle swarm optimization （PSODE） is proposed. In the PSODE, control parameters are encoded to be a symbiotic individual of original individual, and each original individual has its own symbiotic individual. Differential evolution （ DE） operators are used to evolve the original population. And, particle swarm optimization （PSO） is applied to co-evolving the symbiotic population. Thus, with the evolution of the original population in PSODE, the symbiotic population is dynamically and self-adaptively adjusted and the realtime optimum control parameters are obtained. The proposed algorithm is compared with some DE variants on nine functious. The results show that the average performance of PSODE is the best.
文摘Differential evolution algorithm based on the covariance matrix learning can adjust the coordinate system according to the characteristics of the population, which make<span style="font-family:Verdana;">s</span><span style="font-family:Verdana;"> the search move in a more favorable direction. In order to obtain more accurate information about the function shape, this paper propose</span><span style="font-family:Verdana;">s</span><span style="font-family:;" "=""> <span style="font-family:Verdana;">covariance</span><span style="font-family:Verdana;"> matrix learning differential evolution algorithm based on correlation (denoted as RCLDE)</span></span><span style="font-family:;" "=""> </span><span style="font-family:Verdana;">to improve the search efficiency of the algorithm. First, a hybrid mutation strategy is designed to balance the diversity and convergence of the population;secondly, the covariance learning matrix is constructed by selecting the individual with the less correlation;then, a comprehensive learning mechanism is comprehensively designed by two covariance matrix learning mechanisms based on the principle of probability. Finally,</span><span style="font-family:;" "=""> </span><span style="font-family:;" "=""><span style="font-family:Verdana;">the algorithm is tested on the CEC2005, and the experimental results are compared with other effective differential evolution algorithms. The experimental results show that the algorithm proposed in this paper is </span><span style="font-family:Verdana;">an effective algorithm</span><span style="font-family:Verdana;">.</span></span>
基金supported by Fundamental Research Funds of Jilin University(No.SXGJQY2017-9,No.2017TD-19)the National Natural Science Foundation of China(No.61771219)
文摘DV-Hop localization algorithm has greater localization error which estimates distance from an unknown node to the different anchor nodes by using estimated average size of a hop to achieve the location of the unknown node.So an improved DV-Hop localization algorithm based on correctional average size of a hop,HDCDV-Hop algorithm,is proposed.The improved algorithm corrects the estimated distance between the unknown node and different anchor nodes based on fractional hop count information and relatively accurate coordinates of the anchor nodes information,and it uses the improved Differential Evolution algorithm to get the estimate location of unknown nodes so as to further reduce the localization error.Simulation results show that our proposed algorithm have lower localization error and higher localization accuracy compared with the original DV-Hop algorithm and other classical improved algorithms.
基金The National Natural Science Foundation of China(No.62272239,62303214)Jiangsu Agricultural Science and Tech-nology Independent Innovation Fund(No.SJ222051).
文摘To tackle the path planning problem,this study introduced a novel algorithm called two-stage parameter adjustment-based differential evolution(TPADE).This algorithm draws inspiration from group behavior to implement a two-stage scaling factor variation strategy.In the initial phase,it adapts according to environmental complexity.In the following phase,it combines individual and global experiences to fine-tune the orientation factor,effectively improving its global search capability.Furthermore,this study developed a new population update method,ensuring that well-adapted individuals are retained,which enhances population diversity.In benchmark function tests across different dimensions,the proposed algorithm consistently demonstrates superior convergence accuracy and speed.This study also tested the TPADE algorithm in path planning simulations.The experimental results reveal that the TPADE algorithm outperforms existing algorithms by achieving path lengths of 28.527138 and 31.963990 in simple and complex map environments,respectively.These findings indicate that the proposed algorithm is more adaptive and efficient in path planning.
文摘为进一步提高分割精度并加快分割速度,提出了一种基于邻域搜索可选外部存档自适应差分进行算法(简称为JADE-GL)的二维Otsu多阈值图像分割方案。首先,针对原始JADE算法精英突变策略收敛速度慢、容易陷入局部最优等问题,提出了基于邻域半径搜索的改进突变策略,以提升算法的全局探索和开发能力。然后,将所提算法与现有分割方法以及其他JADE变种算法进行二维Otsu多阈值分割对比实验。最后,通过函数收敛曲线、分割距离测度、峰值信噪比(peak signal to noise ratio,PSNR)等指标定量分析算法的性能。实验结果表明,随着阈值数增加,提出的算法无论在收敛速度、分割精度还是分割图像效果上都有显著提升。
文摘针对同步热传递搜索(Simultaneous Heat Transfer Search,SHTS)算法存在的收敛速度慢、优化性能不足及易陷入局部最优等缺陷,提出了一种融合差分进化机制的自适应同步热传递搜索算法(DE-SHTS)。该算法以热传递的传导、对流、辐射三阶段为基础框架,嵌入差分进化中的变异-交叉机制,并提出自适应参数调整策略,从而实现了快速收敛、优良的优化性能,并有效平衡了全局探索与局部开采的动态需求。为验证该算法的性能,在CEC2017测试集上与传统SHTS算法进行对比,从最优适应值、平均适应值和收敛速度等方面进行分析,进一步验证了DE-SHTS算法的有效性。
基金supported by National Natural Science Foundation of China(Nos.61603244 and 41505001)Fundamental Research Funds for the Central Universities(No.222201717006)
文摘The search efficiency of differential evolution （DE） algorithm is greatly impacted by its control parameters. Although many adaptation/self-adaptation techniques can automatically find suitable control parameters for the DE, most techniques are based on pop- ulation information which may be misleading in solving complex optimization problems. Therefore, a self-adaptive DE （i.e., JADE） using two-phase parameter control scheme （TPC-JADE） is proposed to enhance the performance of DE in the current study. In the TPC-JADE, an adaptation technique is utilized to generate the control parameters in the early population evolution, and a well-known empirical guideline is used to update the control parameters in the later evolution stages. The TPC-JADE is compared with four state-of-the-art DE variants on two famous test suites （i.e., IEEE CEC2005 and IEEE CEC2015）. Results indicate that the overall performance of the TPC-JADE is better than that of the other compared algorithms. In addition, the proposed algorithm is utilized to obtain optimal nutrient and inducer feeding for the Lee-Ramirez bioreactor. Experimental results show that the TPC-JADE can perform well on an actual dynamic optimization problem.
