作　　者： ;

机构地区： 广东水利电力职业技术学院电力工程系

出　　处： 《可再生能源》 2012年第2期32-36,共5页

摘　　要： 风速预测对于风力发电并网调度至关重要。基于BP神经网络建立了风速预测模型,并从BP算法及遗传算法自身特点出发,针对BP网络结构确定困难、收敛速度慢等问题,提出创建多种群遗传算法,实现对BP神经网络的结构和权值初始值的同步优化。通过具体算例表明,经优化后的BP算法的收敛步数和计算时间明显减少,预测精度更高,网络整体性能有了显著提高。 Wind speed forecastion is very important to dispatch wind power grid connected. A wind speed prediction model based on BP neural network is established; and from the characteristics of BP algorithm and genetic algorithm ,considered network structures hard to certain and slow convergence speed ,a multi-population genetic algorithm was proposed to optimize the structure and the initial weights synchronously of BP networks, Through practical examples shows that the convergence steps and computing time of optimized BP algorithm has significantly decreased , and a better prediction accuracy. The overall performance of the network has been remarkably improved.

关 键 词： 神经网络 遗传算法 多种群优化 风速预测

领　　域： [动力工程及工程热物理]