基于混合智能算法优化LSSVM的短期风压预测
|
摘要
利用最小二乘支持向量机(least square support vector machine, LSSVM)预测建筑表面的风压.为提高LSSVM对风压预测的泛化能力与精度,提出了基于混合蚁群(ant colony optimization, ACO)和粒子群(particle swarm optimization, PSO)优化LSSVM的预测方法(ACO+PSO-LSSVM),对参数进行搜索寻优,该方法避免了ACO和PSO二者的缺点并实现优势互补.利用ACO+PSO-LSSVM算法对建筑表面风压进行单点和空间点预测,并与ACO-LSSVM和PSO-LSSVM算法作比较.结果表明,基于混合蚁群优化和粒子群优化的LSSVM预测模型预测精度高、鲁棒性强,具有较高的工程应用前景.
Least squares support vector machine(LSSVM)is used to predict wind pressure on building surfaces. To enhance the generalization performance and prediction accuracy of LSSVM for wind pressure, LSSVM based on combination of ant colony optimization(ACO)and particle swarm optimization(PSO) is proposed to find optimal parameters. The combination avoids shortcomings in ACO and PSO, and achieves complementary focus of both.Using LSSVM based on ACO+PSO, wind pressure is forecast. It is compared with ACObased LSSVM and PSO-based LSSVM, respectively. The numerical analysis shows that the proposed method can improve prediction accuracy and robustness of LSSVM, and has good prospects in engineering applications.
Least squares support vector machine(LSSVM)is used to predict wind pressure on building surfaces. To enhance the generalization performance and prediction accuracy of LSSVM for wind pressure, LSSVM based on combination of ant colony optimization(ACO)and particle swarm optimization(PSO) is proposed to find optimal parameters. The combination avoids shortcomings in ACO and PSO, and achieves complementary focus of both.Using LSSVM based on ACO+PSO, wind pressure is forecast. It is compared with ACObased LSSVM and PSO-based LSSVM, respectively. The numerical analysis shows that the proposed method can improve prediction accuracy and robustness of LSSVM, and has good prospects in engineering applications.
引文
[1]黄本才.结构抗风分析原理及应用[M].2版.上海:同济大学出版社,2008:51-59.
[2]顾明.土木结构抗风研究进展及基础科学问题[C]//第七届全国风工程和工业空气动力学学术会议.2006:67-83.
[3]孙斌,姚海涛.基于PSO优化LSSVM的短期风速预测[J].电力系统保护与控制,2012,40(5):85-89.
[4]曾杰,张华.基于蚁群优化的最小二乘支持向量机风速预测模型研究[J].太阳学报,2011,32(3):296-300.
[5]Suykens J A K,Vandewalle J.Least squares support vector machine classifiers[J].Neural Processing,1999,9(3):293-300.
[6]龚松建,袁宇浩,王莉,等.基于PSO优化LS-SVM的短期风速预测[J].可再生能源,2011,29(2):22-27.
[7]Dorigo M,St¨utzle T.The ant colony optimization metaheuristic:algorithms,applications,and advances[M]//Glover F,Kichenberger G A.Handbook of metaheuristics.Boston:Springer,2003:250-285.
[8]Zlochin M,Birattari M,Meuleau N,et al.Model-based search for combinatorial optimization:a critical survey[J].Annats of Operations Research,2004,131:373-395.
[9]Shi Y H,Eberhart R.A modified particle swarm optimizer[C]//Proceedings of IEEE International Conference on Evolutionary Computation.1998:69-73.
[10]李春祥,丁晓达,叶继红.基于混合蚁群和粒子群优化LSSVM的脉动风速预测[J].振动与冲击,2016,35(21):131-136.
[11]李锦华,吴春鹏,陈水生.矩形结构非高斯风荷载特性研究[J].振动.测试与诊断,2014,34(5):951-959.
[2]顾明.土木结构抗风研究进展及基础科学问题[C]//第七届全国风工程和工业空气动力学学术会议.2006:67-83.
[3]孙斌,姚海涛.基于PSO优化LSSVM的短期风速预测[J].电力系统保护与控制,2012,40(5):85-89.
[4]曾杰,张华.基于蚁群优化的最小二乘支持向量机风速预测模型研究[J].太阳学报,2011,32(3):296-300.
[5]Suykens J A K,Vandewalle J.Least squares support vector machine classifiers[J].Neural Processing,1999,9(3):293-300.
[6]龚松建,袁宇浩,王莉,等.基于PSO优化LS-SVM的短期风速预测[J].可再生能源,2011,29(2):22-27.
[7]Dorigo M,St¨utzle T.The ant colony optimization metaheuristic:algorithms,applications,and advances[M]//Glover F,Kichenberger G A.Handbook of metaheuristics.Boston:Springer,2003:250-285.
[8]Zlochin M,Birattari M,Meuleau N,et al.Model-based search for combinatorial optimization:a critical survey[J].Annats of Operations Research,2004,131:373-395.
[9]Shi Y H,Eberhart R.A modified particle swarm optimizer[C]//Proceedings of IEEE International Conference on Evolutionary Computation.1998:69-73.
[10]李春祥,丁晓达,叶继红.基于混合蚁群和粒子群优化LSSVM的脉动风速预测[J].振动与冲击,2016,35(21):131-136.
[11]李锦华,吴春鹏,陈水生.矩形结构非高斯风荷载特性研究[J].振动.测试与诊断,2014,34(5):951-959.