基于相空间重构小波神经网络径流式水电站发电量预测

Electric Power Production Forecasting Model for Runoff-hydro Power Plants Based on Wavelet Neural Networks of Phase Space Reconstruction

预测径流式水电站发量。针对径流式水电站发电量变化的随机性及水力发电系统的复杂非线性,提出一种基于相空间重构小波神经网络的径流式水电站发电量预测模型,并借助具有全局搜索能力的改进粒子群优化算法对小波神经网络特征参数进行优化,提高模型的精度和泛化能力。结果表明:相空间重构小波神经网络模型的预测平均相对误差约为8.7%,相关系数达到0.81,误差分析指标优于传统神经网络预测模型,在多步预测情形下模型的收敛性和稳定性得到较为明显增强,能够准确进行发电量预测。

According to the random and complex nonlinear characteristics of the power generation of runoff hydropower stations, this paper presents a new model of runoff-hydro station based on phase-space reconstruction wavelet neural network model. In order to enhance the global search ability of the model, parameters of the wavelet neural network are optimized by using the improved particle swarm optimization algorithm to overcome the randomness of the model parameter selection. Results show that the relative error of the improved particle swarm optimization wavelet neural network model is about 8.7%, the correlation coefficient is 0.81, and the other error analysis indexes are better than those of the traditional BP neural network forecasting model. The model proposed in this paper is obviously more accurate. It can accurately forecast the generating capacity.