基于模拟退火的自适应正余弦算法

A self-adaptive sine and cosine algorithm based on simulated annealing

为克服正余弦算法寻优精度低、收敛速度慢、易陷入局部最优等缺陷,提出一种基于模拟退火的自适应正余弦算法。设置自适应参数r1,根据粒子自适应值的状态,动态调整参数r1,以增强算法的局部搜索能力;在简化的正余弦算法的位置更新公式中引入对数递减的惯性权重,更好的平衡算法局部搜索与全局搜索的能力;为增加种群多样性,对当前最优解添加高斯扰动,并根据模拟退火中的Metropolis准则接受新解,以避免算法后期陷入局部最优。10个标准测试函数仿真结果表明,改进后算法在收敛速度、寻优精度上更具优势。

In order to overcome the shortcomings of the sine and cosine algorithm,such as low precision,slow convergence speed and being easy to fall into local optimum,a self-adaptive sine and cosine algorithm based on simulated annealing is proposed.Firstly,the self-adaptive parameters r1 are set and the parameters r1 are dynamically adjusted according to the state of the particle self-adaptive value to enhance the local search ability of the algorithm.Secondly,logarithm is introduced into the simplified position update formula of the sine and cosine algorithm.Finally,in order to increase the diversity of population,Gaussian perturbation is added to the current optimal solution,and a new solution is accepted according to Metropolis criterion in simulated annealing,so as to avoid the algorithm′s falling into local optimum.The simulation results of 10 standard test functions show that the improved algorithm has more advantages in convergence speed and optimization accuracy.