基于改进萤火虫算法的LSSVM脉动风速预测
|
摘要
由于风速有较大的随机性,最小二乘支持向量机(LSSVM)预测模型在风速预测方面有极为重要的应用。而新兴的萤火虫算法具有设置参数少、易实现、收敛精度高等优点,但同时存在对优秀个体依赖程度高,易产生震荡现象的缺点。针对萤火虫算法所存在的问题提出改进,并结合实测风速数据对风速进行预测,通过与基于基本萤火算法和粒子群算法优化的LSSVM风速预测结果进行了比较分析。实验结果表明,改进萤火虫算法有更好的性能。
Due to the large randomness of wind speed,the least squares support vector machines(LSSVM)prediction model is of great importance in wind speed prediction.The new firefly algorithm has the advantages of less parameters,easy to implement and high convergence accuracy,but there are some disadvantages of high dependence and volatility.According to the existing problems,improvement of firefly algorithm is proposed and combined with the actual wind speed data to predict wind speed.The results are also compared with the basic firefly algorithm and particle swarm algorithm to optimize the LSSVM wind speed based on the forecast results.The experimental results show that the improved firefly algorithm has better performance.
Due to the large randomness of wind speed,the least squares support vector machines(LSSVM)prediction model is of great importance in wind speed prediction.The new firefly algorithm has the advantages of less parameters,easy to implement and high convergence accuracy,but there are some disadvantages of high dependence and volatility.According to the existing problems,improvement of firefly algorithm is proposed and combined with the actual wind speed data to predict wind speed.The results are also compared with the basic firefly algorithm and particle swarm algorithm to optimize the LSSVM wind speed based on the forecast results.The experimental results show that the improved firefly algorithm has better performance.
引文
[1]李亮,李国强,陈军武.某钢结构气象塔在时程风荷载作用下考虑P-Δ效应的动力时程分析[J].结构工程师,2009,25(4):91-95.Li Liang,Li Guoqiang,Chen Junwu.Dynamic time history analysis of a steel structure meteorological tower considering P-delta effect under time wind load[J].Structural Engineer,2009,25(4):91-95.(in Chinese)
[2]戴靠山,盛超.风力发电机台风荷载响应分析[J].结构工程师,2015,31(6):98-106.Dai Kaoshan,Sheng Chao.Analysis of typhoon load response of wind turbine[J].Structural Engineer,2015,31(6):98-106.(in Chinese)
[3]王世民,林友新,李宏男.风雨荷载作用下输电塔动力响应分析[J].结构工程师,2012,28(6):40-47.Wang Shiming,Lin Youxin,Li Hongnan.Dynamic response analysis of transmission tower under wind and rain load[J].Structural Engineer,2012,28(6):40-47.(in Chinese)
[4]关吉平,陈伟,朱乐东.建筑群风环境干扰效应研究[J].结构工程师,2010,26(3):82-88.Guan Jiping,Chen Wei,Zhu Ledong.Study on the interference effect of wind environment in building groups[J].Structural Engineer,2010,26(3):82-88.(in Chinese)
[5]迟恩楠,李春祥.基于优化组合核和Morlet小波核的LSSVM脉动风速预测方法[J].振动与冲击,2016,35(18):52-57.Chi Ennan,Li Chunxiang.Prediction method of LSS-VM fluctuating wind speed based on optimized combination kernel and Morlet wavelet kernel[J].Vibration and Shock,2016,35(18):52-57.(in Chinese)
[6]Yang X S.Nature-Inspired Metaheuristic Algorithms[M].Luniver Press,2010.
[7]符强,童楠,赵一鸣.一种基于多种群学习机制的萤火虫优化算法[J].计算机应用研究,2013,30(12):3600-3602.Fu Qiang,Tong Nan,Zhao Yiming.A firefly optimization algorithm based on multi group learning mechanism[J].Computer Application Research,2013,30(12):3600-3602.(in Chinese)
[8]刘丹,张腾飞.基于改进萤火虫算法的小波神经网络短期负荷预测方法[J/OL].微型机与应用,2016,35(23):56-58.Liu Dan,Zhang Tengfei.Wavelet neural network shortterm load forecasting method based on improved firefly algorithm[J/OL].Microcomputer and Its Application,2016,35(23):56-58.(in Chinese)
[9]郑晓芬,徐畅,李春祥.基于萤火虫算法的风速预测研究现状及发展前景[J].土木工程,2017,6(4):417-424.Zheng Xiaofen,Xu Chang,Li Chunxiang.Research status and development prospect of wind speed prediction based on firefly algorithm[J].Hans Journal of Civil Engineering,2017,6(4):417-424.(in Chinese)
[2]戴靠山,盛超.风力发电机台风荷载响应分析[J].结构工程师,2015,31(6):98-106.Dai Kaoshan,Sheng Chao.Analysis of typhoon load response of wind turbine[J].Structural Engineer,2015,31(6):98-106.(in Chinese)
[3]王世民,林友新,李宏男.风雨荷载作用下输电塔动力响应分析[J].结构工程师,2012,28(6):40-47.Wang Shiming,Lin Youxin,Li Hongnan.Dynamic response analysis of transmission tower under wind and rain load[J].Structural Engineer,2012,28(6):40-47.(in Chinese)
[4]关吉平,陈伟,朱乐东.建筑群风环境干扰效应研究[J].结构工程师,2010,26(3):82-88.Guan Jiping,Chen Wei,Zhu Ledong.Study on the interference effect of wind environment in building groups[J].Structural Engineer,2010,26(3):82-88.(in Chinese)
[5]迟恩楠,李春祥.基于优化组合核和Morlet小波核的LSSVM脉动风速预测方法[J].振动与冲击,2016,35(18):52-57.Chi Ennan,Li Chunxiang.Prediction method of LSS-VM fluctuating wind speed based on optimized combination kernel and Morlet wavelet kernel[J].Vibration and Shock,2016,35(18):52-57.(in Chinese)
[6]Yang X S.Nature-Inspired Metaheuristic Algorithms[M].Luniver Press,2010.
[7]符强,童楠,赵一鸣.一种基于多种群学习机制的萤火虫优化算法[J].计算机应用研究,2013,30(12):3600-3602.Fu Qiang,Tong Nan,Zhao Yiming.A firefly optimization algorithm based on multi group learning mechanism[J].Computer Application Research,2013,30(12):3600-3602.(in Chinese)
[8]刘丹,张腾飞.基于改进萤火虫算法的小波神经网络短期负荷预测方法[J/OL].微型机与应用,2016,35(23):56-58.Liu Dan,Zhang Tengfei.Wavelet neural network shortterm load forecasting method based on improved firefly algorithm[J/OL].Microcomputer and Its Application,2016,35(23):56-58.(in Chinese)
[9]郑晓芬,徐畅,李春祥.基于萤火虫算法的风速预测研究现状及发展前景[J].土木工程,2017,6(4):417-424.Zheng Xiaofen,Xu Chang,Li Chunxiang.Research status and development prospect of wind speed prediction based on firefly algorithm[J].Hans Journal of Civil Engineering,2017,6(4):417-424.(in Chinese)