基于改进神经网络的电力系统短期负荷预测

Short-term load forecasting of power system based on Improved Neural Network

电力系统的短期负荷预测是电力系统保证稳定运行的重要环节,也是区域配电调度的重要依据。为了提高短期负荷预测的精准度与稳定性,本文提出一种基于粒子群算法与遗传算法共同优化BP神经网络的负荷预测方法。针对BP神经网络存在的权值取值不确定、收敛速度慢等问题,将粒子的维度空间与人工神经网络的权值建立映射关系,使得神经网络的均方误差作为粒子群的适应函数,同时,引入遗传算法对其迭代过程进行优化,利用遗传算法全局搜索能力对极值进行搜索,并对粒子的适应度进行分类。最后通过实例分析,证明了该方法的有效性。

Short-term load forecasting of power system is an important link to ensure stable operation of power system,and also an important basi s for region al distribution dispatch. In order to improve the accuracy and stability of short-term load forecasting,a load forecasting method based on particle swarm optimization and genetic algorithm is proposed. Aiming at the uncertain weights and slow convergence of BP neural network,the mapping relationship between the dimension space of particles and the weights of artificial neural network is established,so that the mean square error of neural network is taken as the adaptive function of particle swarm optimization. At the same time,genetic algorithm is introduced to optimize the iteration process of BP neural network,and genetic algorithm is used to optimize the iteration process. The global search capability se arches the extremum and classifies the fitness of particles. Finally,an example is given to demonstrate the effectiveness of the proposed method.