基于改进AdaBoost.RT和KELM的风功率预测方法研究

发布时间：2018-01-18 03:11

本文关键词：基于改进AdaBoost.RT和KELM的风功率预测方法研究　出处：《电网技术》2017年02期 　论文类型：期刊论文

【摘要】：为了提高风功率预测精度及预测模型的泛化能力,提出基于改进Ada Boost.RT算法的风功率预测方法,可以有效提高弱学习算法的性能。首先建立核极限学习机(kernel extreme learning machine,KELM)模型,并用改进蝙蝠算法对其参数进行优化,通过引入局部搜索和莱维飞行使算法具有更好的搜索能力和跳出局部最优的能力。在此基础上进一步通过Ada Boost.RT算法生成多个KELM个体(即基学习器),在训练过程中不断调整每个基学习器的权重及训练集中每个样本的权重。最后用训练好的基学习器来对测试样本进行预测,并集成得到最终结果。从不同时间尺度应用不同月份的风电场数据进行仿真测试,同时与前馈(back propagation,BP)神经网络、支持向量机、极限学习机等预测模型对比,仿真结果表明所提方法具有较好的预测精度及泛化性能。
[Abstract]:In order to improve the precision of wind power prediction and the generalization ability of prediction model, a wind power prediction method based on improved Ada Boost.RT algorithm is proposed. It can effectively improve the performance of weak learning algorithm. Firstly, kernel extreme learning machine model is established. The improved bat algorithm is used to optimize the parameters. By introducing local search and Levy flight, the algorithm has better searching ability and the ability of jumping out of local optimum. On this basis, the algorithm is further adopted through Ada. The Boost.RT algorithm generates multiple KELM individuals. That is, the base learner. In the process of training, we constantly adjust the weight of each basic learner and the weight of each sample in the training set. Finally, we use the trained base learner to predict the test samples. Finally, the final results are obtained. The wind farm data in different months are used for simulation test from different time scales, and at the same time, the neural network is combined with the feedforward propagation (BPN) neural network. Compared with support vector machine (SVM) and extreme learning machine (LLM), the simulation results show that the proposed method has good prediction accuracy and generalization performance.
【作者单位】：武汉大学电气工程学院;
【分类号】：TM614;TP18
【正文快照】： 0引言高精度的风功率预测可以减小风电的波动性和间歇性给电力系统的稳定运行带来的影响。当预测达到足够精度时,可以减少系统用来平衡其功率波动的备用容量,从而减少电网运行费用,并且有利于系统实现大规模风电的有效消纳。文献[1-2]利用神经网络对风电进行预测,取得了较

【相似文献】