基于IAFSA-BPNN的短期风电功率预测

发布时间：2018-03-09 18:15

  本文选题：短期风电功率预测　切入点：人工鱼群算法　出处：《电力系统保护与控制》2017年07期 　论文类型：期刊论文

【摘要】：为提高短期风电功率预测精度,提出一种基于IAFSA-BPNN的短期风电功率预测方法。该方法通过改进的人工鱼群算法来优化BP神经网络的权值和阈值,从而提高BP神经网络的收敛速度和泛化能力。利用2014年上海某风场实测数据对新算法进行检验。试验结果表明,改进的人工鱼群算法一定程度上克服了原算法后期搜索的盲目性较大,收敛速度减慢,搜索精度变低的缺陷。IAFSA-BPNN混合算法在预测的稳定性和精度、收敛速度等方面优于BPNN、AFSA-BPNN算法。IAFSA-BPNN算法不仅能提高短期风电功率预测的精度,而且改善了预测结果稳定性。
[Abstract]:In order to improve the accuracy of short-term wind power prediction, a short-term wind power prediction method based on IAFSA-BPNN is proposed, which optimizes the weights and thresholds of BP neural network through an improved artificial fish swarm algorithm. In order to improve the convergence speed and generalization ability of BP neural network, the new algorithm is tested with wind field data measured in Shanghai on 2014. The experimental results show that, To some extent, the improved artificial fish swarm algorithm overcomes the shortcomings of the original algorithm, such as large blindness, slow convergence speed and low searching precision. IAFSA-BPNN hybrid algorithm is stable and accurate in prediction. The convergence rate is better than that of BPNN-AFSA-BPNN. IAFSA-BPNN can not only improve the accuracy of short-term wind power prediction, but also improve the stability of prediction results.
【作者单位】： 南京信息工程大学信息与控制学院;南京信息工程大学气象灾害预报预警与评估协同创新中心;
【基金】：国家自然科学基金项目(41675156) 江苏省高校优势学科建设工程资助项目(PAPD) 江苏省六大人才高峰项(WLW-021)共同资助~~
【分类号】：TM614;TP18
，

本文编号：1589678