论文部分内容阅读
基于进化过程中父代个体和子代个体在种群中的适应度梯度,提出进化方向的概念,并对其进行定性分析,在此基础上提出最优进化方向.通过最优进化方向的指导可加速进化过程,提高进化算法的收敛效率.基于进化方向和最优进化方向的描述,设计并实现基于个体适应度梯度的定向进化算法,并针对该算法给出2种不同的个体繁殖策略.对算法的收敛性和复杂度进行理论分析.最后通过仿真实验说明该算法具有精度高、收敛速度快等优点,在一定程度上克服目前进化算法的搜索低效问题.
Based on the gradient of fitness of parents and offspring in the population during the evolutionary process, the concept of evolutionary direction is put forward and its qualitative analysis is made. Based on this, the optimal evolutionary direction is proposed, which can be accelerated through the guidance of the optimal evolutionary direction Evolutionary process to improve the convergence efficiency of evolutionary algorithm.Based on the description of evolutionary direction and optimal evolutionary direction, a directed evolutionary algorithm based on individual fitness gradient is designed and implemented, and two different strategies of individual reproduction are given for this algorithm.The algorithm The convergence and complexity of the proposed algorithm are analyzed theoretically.Finally, the simulation results show that the proposed algorithm has the advantages of high precision and fast convergence speed, and overcomes the search inefficiency of current evolutionary algorithms to a certain extent.