基于改进磷虾群优化支持向量机的短期负荷预测
|
摘要
针对支持向量机在短期负荷预测时,因受参数影响较大而导致预测精度不足和速度较慢的问题,提出了基于改进磷虾群算法对支持向量机进行参数选择的算法。阐述了支持向量机的原理与其参数的影响;分析了磷虾群算法并对其改进:加入模拟退火思想降低了算法陷入局部最优的概率,采用自适应迭代步长方法提高算法优化精度。建立预测模型并进行算例分析,实验结果表明改进磷虾群算法对支持向量机参数有较好的优化效果,可以有效地提高负荷预测精度和速度。
When using the traditional support vector machine for short-term load forecasting, the accuracy is insufficient and the speed is slow due to the large influence of the parameters. Aiming at this problem, a method to optimize the parameters of support vector machine by using improved krill herd algorithm is proposed. The principle of support vector machine and influence of its parameter are expounded. The krill herd algorithm is analyzed and improved. The idea of adding simulated annealing reduces the probability that the algorithm falls into local optimum, and the adaptive iterative step method is used to improve the algorithm optimization accuracy. The prediction model is established and the simulation experiment is carried out. The experimental results show that the improved krill herd algorithm has better optimization effect on the support vector machine parameters, which can effectively improve the load prediction accuracy and speed.
When using the traditional support vector machine for short-term load forecasting, the accuracy is insufficient and the speed is slow due to the large influence of the parameters. Aiming at this problem, a method to optimize the parameters of support vector machine by using improved krill herd algorithm is proposed. The principle of support vector machine and influence of its parameter are expounded. The krill herd algorithm is analyzed and improved. The idea of adding simulated annealing reduces the probability that the algorithm falls into local optimum, and the adaptive iterative step method is used to improve the algorithm optimization accuracy. The prediction model is established and the simulation experiment is carried out. The experimental results show that the improved krill herd algorithm has better optimization effect on the support vector machine parameters, which can effectively improve the load prediction accuracy and speed.
引文
[1] 苏学能,刘天琪,曹鸿谦,等.基于Hadoop架构的多重分布式BP神经网络的短期负荷预测方法[J].中国电机工程学报,2017,37(17):4966-4973.
[2] 李婉华,陈宏,郭昆,等.基于随机森林算法的用电负荷预测研究[J].计算机工程与应用,2016,52(23):236-243.
[3] 章治邦. 基于改进网格搜索法的SVM边坡变形预测研究[D].赣州:江西理工大学, 2018.
[4] 李瑾,刘金朋,王建军.采用支持向量机和模拟退火算法的中长期负荷预测方法[J].中国电机工程学报,2011,31(16):63-66.
[5] 陆宁,武本令,刘颖.基于自适应粒子群优化的SVM模型在负荷预测中的应用[J].电力系统保护与控制,2011,39(15):43-46.
[6] 杨小明,崔雪,周斌,等.基于粒子群优化支持向量机的短期负荷预测[J].武汉大学学报:工学版,2018,51(8):715-720.
[7] 孟凡喜,屈鸿,侯孟书.基于GA和SVM的电力负荷预测方法研究[J].计算机科学,2014,41(S1):91-93.
[8] FERREIRA J P,CRISOSTOMO M M,COIMBRA A P.SVR versus neural-fuzzy network controllers for thesagittal balance of a biped robot[J].IEEE Transactions on Neural Networks,2009,20(12):1885-1897.
[9] GANDOMI A H, ALAVI A H.Krill herd: a new bio-inspired optimization algorithm[J]. Communications in Nonlinear Science and Numerical Simulation, 2012,17(12):4831-4845.
[10] LI J P, TANG Y G, HUA C C, et al. An improved krill herd algorithm: Krill herd with linear decreasing step[J]. Applied Mathematics and Computation, 2014(234):356-367.
[11] 郭伟,高岳林,刘沛.一种自适应惯性权重的改进磷虾群算法[J].太原理工大学学报,2016,47(5):651-657.
[2] 李婉华,陈宏,郭昆,等.基于随机森林算法的用电负荷预测研究[J].计算机工程与应用,2016,52(23):236-243.
[3] 章治邦. 基于改进网格搜索法的SVM边坡变形预测研究[D].赣州:江西理工大学, 2018.
[4] 李瑾,刘金朋,王建军.采用支持向量机和模拟退火算法的中长期负荷预测方法[J].中国电机工程学报,2011,31(16):63-66.
[5] 陆宁,武本令,刘颖.基于自适应粒子群优化的SVM模型在负荷预测中的应用[J].电力系统保护与控制,2011,39(15):43-46.
[6] 杨小明,崔雪,周斌,等.基于粒子群优化支持向量机的短期负荷预测[J].武汉大学学报:工学版,2018,51(8):715-720.
[7] 孟凡喜,屈鸿,侯孟书.基于GA和SVM的电力负荷预测方法研究[J].计算机科学,2014,41(S1):91-93.
[8] FERREIRA J P,CRISOSTOMO M M,COIMBRA A P.SVR versus neural-fuzzy network controllers for thesagittal balance of a biped robot[J].IEEE Transactions on Neural Networks,2009,20(12):1885-1897.
[9] GANDOMI A H, ALAVI A H.Krill herd: a new bio-inspired optimization algorithm[J]. Communications in Nonlinear Science and Numerical Simulation, 2012,17(12):4831-4845.
[10] LI J P, TANG Y G, HUA C C, et al. An improved krill herd algorithm: Krill herd with linear decreasing step[J]. Applied Mathematics and Computation, 2014(234):356-367.
[11] 郭伟,高岳林,刘沛.一种自适应惯性权重的改进磷虾群算法[J].太原理工大学学报,2016,47(5):651-657.