基于改良和声搜索的电力系统动态经济调度

Dynamic Economic Dispatching of Power System Based on Reformed Harmony Search

针对电力系统动态经济调度(Dynamic economic dispatch,DED)问题提出了一种改良和声搜索算法(Reformed harmony search,RHS)求解。RHS在原始和声搜索算法的基础上进行了四方面改进:第一,以100%概率执行和声记忆步骤,排除了随机化步骤。第二,使用基于混沌序列的随机数对基音调整率进行扰动,使其在整个迭代过程中始终处于动态变化的状态。第三,根据当前和声记忆库获得每个变量的动态上下限,以此确定基音调整步长。第四,引入了一种全局搜索步骤,该步骤根据全局最优和声向量对当前和声向量进行更新。上述四个步骤有利于提高RHS算法的收敛性、种群多样性以及对问题的自适应性。实验表明,RHS在求解三种DED问题时能获得比其他两种算法更低的发电成本,因此,它具有更好的优化性能,是求解DED问题的一种有效方法。

A Reformed harmony search(RHS)algorithm was proposed to solve the Dynamic economic dispatch(DED)problem in power systems.RHS improves on the original harmonic search algorithm in four aspects:First,the harmonic memory step is performed with 100%probability,excluding the randomization step.Secondly,random numbers based on chaotic sequences are used to disturb the pitch adjustment rate so that it is always in a dynamic state during the whole iteration process.Third,the dynamic upper and lower limits of each variable are obtained according to the current harmonic memory bank to determine the pitch adjustment step size.Fourthly,a global search step is introduced,which updates the current harmonic vector according to the global optimal harmonic vector.The above four steps are conducive to improving the convergence,population diversity and self-adaptability of RHS algorithm.The experiment shows that RHS can obtain lower power generation cost than the other two algorithms when solving the 3 kinds of DED problems.Therefore,it has better optimization performance and is an effective method for solving DED problems.