考虑人工蜂群（ABC）和人工鱼群（AFS）算法的各自优势，提出混合智能算法（ABC+AFS）优化选择最小二乘支持向量机（LSSVM）参数的方法，以提高其脉动风速预测模型的性能。AFS算法有较强的全局寻优能力，混合智能算法以AFS算法中的人工鱼寻优方式代替ABC算法中的引领蜂寻优方式，克服ABC算法易陷入局部最优的问题。同时，ABC算法中的正负反馈机制可以克服AFS算法的后期盲目寻优、收敛速度下降的问题。运用基于混合ABC、AFS优化的LSSVM对脉动风速进行了预测，并与基于ABC、AFS和粒子群（PSO）算法优化的LSSVM脉动风速预测结果进行了比较。数值结果表明，基于混合ABC+AFS优化的LSSVM脉动风速预测模型有更好性能，具有工程应用前景。
Abstract
By resorting to the respective advantages of the artificial bee colony（ABC）and artificial fish swarm (AFS) algorithms, this work has proposed a hybrid optimization algorithm, referred as to ABC+AFS algorithm, used to optimize the parameters of least square support vector machine (LSSVM) in order to enhance its performance of predicting fluctuating wind velocity. Considering that the AFS algorithm has better ability of skipping over the local optimum, substituting the optimizing process of the employed bees in the ABC algorithm with that of the artificial fishes in the AFS algorithm is thus implemented to cope with the problems that the ABC algorithm easily traps into local optima. Concurrently, adopting the positive and negative feedback mechanism of the ABC algorithm comes over the problem of aimless optimization in the late period of the AFS algorithm. LSSVM for the prediction of fluctuating wind velocity has been established making use of the ABC+AFS algorithm to select its parameters. For a comparison, the results of fluctuating wind velocity forecasting are also taken into consideration using the ABC, AFS, and particle swarm optimization (PSO) algorithm based LSSVM. The numerical results show that the fluctuating wind velocity prediction method of the ABC+AFS algorithm based LSSVM has better prediction performance and a bright prospect of engineering application.
关键词
人工蜂群算法 /
人工鱼群算法 /
混合智能优化 /
最小二乘支持向量机 /
脉动风速 /
预测性能
{{custom_keyword}} /
Key words
Artificial bee colony /
Artificial fish swarm /
Hybrid intelligent optimization /
Least square support sector machine /
Fluctuating wind velocity /
Prediction performance
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Monfared M, Rastegar H, Kojabadi HM. A new strategy for wind speed forecasting using artificial intellingent methods [J]. Renewable Energy, 2002, 27(2):163-174.
[2] Suykens JAK, Vandewalle J. Least squares support vector machine classifiers [J]. Neural Processing, 1999, 9(3): 293-300.
[3] Ying DU, Lu JP, Li Q. Short-term wind speed forecasting of wind farm based on least square-support vector machine [J]. Power System Technology, 2008, 32(15):62-66.
[4] 李春祥, 迟恩楠. 基于优化组合核和Morlet小波核的LSSVM脉动风速预测方法. 振动与冲击, 2016年发表.
Li Chunxiang, Chi Ennan. Forecasting fluctuating wind velocity using optimized combination kernel and Morlet wavelet kernel based LSSVM. Journal of Vibration and Shock, to be published, 2016.
[5] 李春祥, 迟恩楠, 何亮, 李正农. 基于时变ARMA和EMD-PSO-LSSVM算法的非平稳下击暴流风速预测. 振动与冲击, 2016年发表.
Li Chunxiang, Chi En-nan, He Liang, Li Zhengnong. Prediction of nonstationary downburst wind velocity based on time-varying ARMA and EMD- PSO-LSSVM algorithms. Journal of Vibration and Shock, to be published, 2016.
[6] 张弦, 王宏力. 基于粒子群优化的最小二乘支持向量机在时间序列预测中的应用[J].中国机械工程, 2011, 22(21): 2572-2576.
Zhang Xian, Wang Hong-li. The application of least squares support vector machine based on Particle Swarm Optimization in the forecasting of time series [J]. China Mechanical Engineering, 2011, 22(21): 2572-2576.
[7] Mehdi Neshat, GhodratSepidnam, Mehdi Sargolzaei, Adel NajaranToosi. Artificial fish swarm algorithm:a survey of the state-of-the-art, hybridization, combinatorial and indicative applications [J]. Artificial Intelligence Review, 2014, 42: 965-997.
[8] Karaboga D, Akay B. A comparative studay of artificial bee colony algorithm [J]. Applied mathematics and computation, 2009, 214(14): 108-132.
[9] 王联国,洪毅,赵付清,余冬梅.一种简化的人工鱼群算法 [J]. 小型微型计算机系统, 2009, 30(8): 1663-1667.
Wang Lianguo, Hong Yi, Zhao Fuqing, Yu Dongmei.Simplified artificial fish swarm algorithm [J]. Journal of Chinese computer systems, 2009, 30(8): 1663-1667.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(1707 KB)
