一种基于并行混沌和SFLA的风电电网无功优化模型设计

针对已有方法在解决电网无功优化时,由于系统的无功不足和电网电压的不稳定,容易过早收敛到局部最优解的缺点,设计了一种基于并行混沌和混合蛙跳算法 (Shuffle Frog Leaping Algorithm, SFLA)的电网无功优化模型。首先,建立了最小化有功网损、最大化静态电压稳定裕度和最大化无功补偿单位投资收益的多目标数学优化模型,然后,对经典的SFLA进行改进,通过引入精英协同进化机制和划分种群的方式实现并行寻优,从而增加个体的多样性和加快最优解的求取速度,在不同种群中设计不同的适应度函数和个体更新进化方法。为了使得算法的初始解分布更为均匀,引入用混沌机制来对种群进行初始化,最后,对基于并行混沌和SFLA的总体算法进行了设计和分析。在Matlab环境下进行实验,实验结果表明文中方法得到的优化结果具有电网有功损耗小、单位投资收益高和静态电压稳定裕度大的优点,具有较强的可行性和适应性。

A Method for Design of Reactive Voltage Optimization for Wind Power Grid Based on Parallel Chaos and Shuffled Frog Leaping Algorithm

Aiming at the existing methods have the defects of the reactiveness insufficiency and instability of power voltage, a reactive optimal method based on parallel chaos and Shuffled Frog leaping algorithm (SFLA) is proposed. Firstly, the optimal model with multiple goals such as minimizing wind power grid loss, maximizing return of investment and stability of static voltage is presented. Then, the classical algorithm of SFLA is improved via introducing the elite coevolution mechanism and dividing populations to accomplish parallel optimization. In order to add the individual diversity and accelerate the convergence of algorithm, the different fitness function and renew mechanism for sub-population are designed. In order to make the initial solution more evenly, the chaos mechanism is used to get the initial solution for the algorithm. Finally, the whole algorithm based on parallel chaos and SFLA is proposed. The experiment is implemented in the matlab environment, the result shows the method in this paper has the advantage of litter wind power grid loss, high return of investment and bigger stability of static voltage. Thus it has strong feasibility and applicability.