基于量子粒子群-径向基神经网络模型的风速预测

Wind Speed Forecasting Based on Quantum-behaved Particle Swarm Optimization and Radial Basis Function Neural Network Model

风速预测对风电场和电力系统的运行都具有重要意义.为了提高风速预测的精度,提出了一种基于量子粒子群-径向基神经网络模型,在确定网络隐含层节点数后,将RBF网络的参数编码成优化算法中的粒子个体进行优化,在全局空间搜索最优适应值的参数.用优化后的神经网络进行风速预测,实例结果表明该算法在预测速度和精度上都得到了提高.

Forecasting of wind speed is very important to the operation of wind power plants and power systems. To improve the wind speed forecasting accuracy,a model based on quantum--behaved particle swarm optimization and radial basis function neural network algorithm is proposed. After the number of nodes in bidden layer is confirmed and all parameters of RBF nets are coded to individual particles to optimize learning algorithm, the parameter of optimal--adaptive values can he searched in global space. Using the optimized neural network to forecast wind speed, and some cal- culation examples were abtained. The results showed that the new method can improve the speed and accuracy of prediction.