孙驷洲,付敬奇,朱峰.CGAQPSO优化LSSVM短期风电预测[J].电子测量与仪器学报,2016,30(11):1718-1725
CGAQPSO优化LSSVM短期风电预测
Short term wind power prediction based on LSSVM optimized by chaos Gauss attractor quantum behaved particle swarm optimization
  
DOI:10.13382/j.jemi.2016.11.013
中文关键词:  量子粒子群  混沌高斯局部吸引点量子粒子群  短期风电预测  最小二乘支持向量机
英文关键词:QPSO  CGAQPSO  short term wind power prediction  LSSVM
基金项目:安徽省自然资金(1408085ME105,1608085ME106)、安徽省高校自然科学基金重点项目(KJ2015A063)、安徽工程大学安徽检测技术与节能装置省级实验室开放研究基金资助项目
作者单位
孙驷洲 1. 安徽工程大学电气工程学院芜湖241000;2. 上海大学机电工程与自动化学院上海200072 
付敬奇 上海大学机电工程与自动化学院上海200072 
朱峰 上海大学机电工程与自动化学院上海200072 
AuthorInstitution
Sun Sizhou 1.School of Electrical Engineering, Anhui Polytechnic University, Wuhu 241000, China;2. School of Mechatronics Engineering and Automation, Shanghai University, Shanghai 200072, China 
Fu Jingqi School of Mechatronics Engineering and Automation, Shanghai University, Shanghai 200072, China 
Zhu Feng School of Mechatronics Engineering and Automation, Shanghai University, Shanghai 200072, China 
摘要点击次数: 133
全文下载次数: 144
中文摘要:
      提出一种基于混沌高斯局部吸引点量子粒子群(CGAQPSO)优化最小二乘支持向量机(LSSVM)的短期风电功率预测模型。首先,混沌算法初始化粒子种群,提高初始粒子在搜寻空间遍历性,将局部吸引点改进为高斯分布局部吸引点,增强粒子全局搜索能力,从而得到混沌高斯局部吸引点量子粒子群优化算法。对基于不同类型核函数(Linear、POLY、Sigmoid及RBF)进行比较,选择RBF核函数来构建LSSVM风电预测模型。最后,以安徽某风电场实测风电、温度及湿度的历史数据作为CGAQPSO LSSVM (RBF)模型的训练数据。实验表明,与GA、PSO和QPSO优化LSSVM预测模型相比,所提出的CGAQPSO LSSVM模型能够有效提高风电功率预测精确度。
英文摘要:
      Chaos Gauss attractor quantum behaved particle swarm optimization (CGAQPSO) is proposed to optimize the parameters combination by adding chaos algorithm, Gauss attractor and dynamic expansion contraction coefficient in QPSO algorithm. As the kernel function and its parameter have a great influence on the performance of the LSSVM model. The paper establishes LSSVM wind power prediction model based on different kernel functions, including Linear, Poly, RBF and Sigmoid kernel function, and RBF kernel function is selected as its optimal performance. To verify the proposed hybrid prediction model, the seven days actual data recorded in a wind farm located in Anhui of China are utilized to train the proposed model to forecast subsequent 24 h wind power. The results show that the proposed hybrid model achieves higher prediction accuracy compared with other models mentioned in the paper.
查看全文  查看/发表评论  下载PDF阅读器