一种改进麻雀搜索算法的收敛性分析及应用

Convergence analysis and application of an improved sparrow search algorithm

针对麻雀搜索算法易陷入局部最优、收敛速度慢等问题,提出一种改进的麻雀搜索算法.首先,采用佳点集策略对麻雀种群初始化,增加种群多样性,提高算法的收敛速度和精度;其次,采用黄金正弦策略优化发现者位置更新过程,进一步平衡算法的全局探索与局部开发能力;最后,采用Levy飞行策略优化跟随者位置更新过程,扩大其搜索空间,改善易陷入局部最优的问题.通过建立马尔科夫链模型从理论角度证明改进算法的收敛性,并选取5个标准测试函数与其他经典群智能优化算法从仿真实验角度验证改进算法的有效性.利用改进算法对变分模态分解参数和回声状态网络参数进行优化,搭建ISSA-VMD-ESN模型并应用到短期电价预测中,通过仿真实验进一步验证了改进算法的优越性.

Aiming at the problems of local optimization and slow convergence speed of sparrow search algorithms,an improved sparrow search algorithm(ISSA)is proposed.Firstly,a good point set method is used to initialize the sparrow population,which increases the population diversity and improves the convergence speed and accuracy of the algorithm.Then,the golden sine algorithm is used to optimize the founder’s position update process to further balance the global exploration and local development capabilities of the algorithm.Finally,the Levy flight algorithm is used to optimize the follower’s position update process,expand its search space,and improve the problem that it is easy to fall into local optimization.By establishing the Markov chain model,the convergence of the improved algorithm is proved from the theoretical perspective.Five standard test functions and other classical swarm intelligent optimization algorithms are selected to verify the effectiveness of the improved algorithm from the perspective of simulation experiments.The improved algorithm is used to optimize the variational mode decomposition(VMD)parameters and echo state network(ESN)parameters.The ISSA-VMD-ESN model is constructed and applied to short-term electricity price prediction,and the superiority of the improved algorithm is further verified by simulation experiments.