基于模拟退火粒子群算法的风电场短期风速预测

Short-term wind speed forecast based on simulation annealing particle swarm optimization algorithm

针对风电场短期风速预测存在精度较低的问题,引入了一种基于灰色模型与模拟退火粒子群优化BP神经网络(SAPSO-BP)相互嵌入而成的预测模型。该方法在SAPSO-BP网络的输入层前增加一个灰化层,在网络输出层后增加一个白化层,以改进网络的拓扑结构,提高模型的容错能力。仿真试验结果表明,该预测模型具有较好的鲁棒性,其平均绝对误差及均方误差分别为18.7%和5.11%,可用于风电场短期风速的预测。

Aiming at low prediction accuracy of short-term wind speed, a forecasting method of grey model and simulated annealing particle swarm optimization BP neural network （SAPSO-BP） embedded in each other was proposed. The method is to optimize the topological structure of the SAPSO-BP network by adding a grey layer before the input layer and a white layer after the output layer so as to enhance the fault-tolerant ability. The simulation results show that the mean absolute error, mean squared error of forecasting are 18.7% and 5.11%; the model has better robustness,and can be used for short-term wind speed prediction.