基于 IWOA-SVM 的风电功率预测

针对支持向量机预测精度低、收敛速度慢等问题,提出一种改进鲸鱼算法优化支持向量机的风电功率预测模型。将 Tent 混沌映射引入鲸鱼算法中,使初始种群的分布更加均匀;由于随机抽取猎物具有盲目性,不能充分结合迭代经验对种群进行更新,采用轮盘赌法寻找目标猎物来加快鲸鱼算法的收敛速度,得到改进鲸鱼算法优化支持向量机的风电功率预测模型。将该模型应用到我国东北某处风电场进行风电功率预测,并与其他常用的功率预测模型进行对比分析,仿真结果表明,该模型具有更高的预测精度。

Wind Power Forecast Based on IWOA-SVM

Aiming at the problems of low prediction accuracy and slow convergence speed of support vector machine , a wind power prediction model based on improved whale algorithm and optimization support vector machine is proposed.The Tent chaotic map is introduced into the whale algorithm to make the distribution of the initial population more uniform ; due to the blindness of randomly extracting prey , the population cannot be fully updated with the iterative experience.The roulette method is used to find the target prey to speed up the convergence speed of the whale algorithm.The wind power prediction model based on the improved whale algorithm optimization support vector machine is obtained.The model is applied to a wind farm in Northeast China to predict wind power , and the results are compared with other commonly used power prediction models.The simulation results show that the model proposed in this paper has higher prediction accuracy.