面向光伏MPPT控制策略的改进果蝇算法

局部遮光会降低光伏发电系统的效率。在局部遮光条件下,光伏系统的输出功率特性曲线会产生多个峰值,传统的最大功率跟踪方法不具有全局搜索的能力,其在进行多峰值最大功率跟踪时会失效。果蝇算法(Fruit Fly Optimization Algorithm,FOA)具有全局寻优能力,但是在求解过程中存在收敛速度慢、收敛精度低及容易收敛于局部最优值的问题。文中对果蝇算法进行改进,提出结合自适应lévy飞行步长的Lévy-FOA算法,该算法充分利用Lévy飞行不均匀随机游走的特性,引入自适应步长调整因子,改进了原有算法的位置更新方式,提高了算法的收敛速度以及收敛精度,避免了算法陷入局部极值。文中利用3个标准函数对自适应Lévy-FOA算法的收敛性进行分析,并与普通FOA算法、自适应改进学习因子粒子群算法(Adaptive Particle Swarm Optimization,APSO)进行对比。结果表明,与FOA算法和APSO算法相比,自适应Lévy-FOA算法的平均跟踪时间有较大幅度的减少,平均收敛精度提高了4个数量级。最后,将自适应Lévy-FOA算法应用于光伏最大功率跟踪中。仿真结果显示,在不同的光照条件下,自适应Lévy-FOA算法能够经过较少的迭代实现最大功率跟踪,并且在第一次迭代后就能达到最大功率的90%以上,与其他算法的跟踪效果对比,自适应Lévy-FOA算法具有较短的跟踪时间和较高的跟踪精度,实际寻优能力优越,能够提高光伏系统的输出效率。

Improved Fruit Fly Algorithm for Photovoltaic MPPT Control Strategy

Local shading will reduce the efficiency of photovoltaic power generation system.Under local shading conditions,the output power characteristic curve of photovoltaic system will produce multiple peaks.The traditional maximum power tracking method does not have the ability of global search,and will fail in multi-peak maximum power tracking.Fruit fly optimization algorithm(FOA)has the ability of global optimization,but in the solution process,there are some problems such as slow convergence speed,low convergence accuracy and easy convergence to local optimum.This paper improves fruit fly algorithm,and proposes a Lévy-FOA algorithm combined with adaptive Lévy flight step size.This algorithm makes full use of the characteristics of Lévy flight non-uniform random walk and introduces adaptive step adjustment factor.It improves the position updating method of the original algorithm,improves the convergence speed and accuracy of the algorithm,and avoids the algorithm falling into local extremum.In this paper,three standard functions are used to analyze the convergence of adaptive Lévy-FOA algorithm,and the results are compared with those of conventional FOA algorithm and adaptive improved learning factor particle swarm optimization(APSO).The comparison results show that compared with FOA algorithm and APSO algorithm,the average tracking time of adaptive Lévy-FOA algorithm is significantly reduced,and the average convergence accuracy is improved by four orders of magnitude.Finally,the adaptive Lévy-FOA algorithm is applied to the maximum power tracking of photovoltaic system.The simulation results show that under different illumination conditions,the adaptive Lévy-FOA algorithm can achieve maximum power tracking with fewer iterations,and can achieve more than 90%of the maximum power after the first iteration.Compared with other algorithms,the adaptive Lévy-FOA algorithm has shorter tracking time and higher tracking accuracy,and has superior practical optimization ability,which can improve the output efficiency of photovoltaic system.