摘要
分析了电力系统负荷预测目前采用的方法的不足;在已有研究成果的基础上,根据电网负荷的特点进一步完善了基于模糊支持向量的核回归方法;与目前已有的方法,如神经网络、卡尔曼滤波、最小绝对值参数估计、结合遗传算法的支持向量机、结合模糊小波技术的支持向量机等进行对比实验,实验结果展示了几种方法的性能对比,为该领域的研究提供了参考.
In view of the disadvantages of current methods for load forecasting in power system, a new algorithm named fuzzy support vector kernel regression method （F-SVKR） is proposed to deal with the problem. Comparison with some conventional methods, such as the artificial neural network, Kalman filtering algorithm, the method of minimizing absolute parameter estimation, the support vector machine and so on, have been performed. Expertmental results show their performance differences and provide some reference information for the further research in this domain.
出处
《控制理论与应用》
EI
CAS
CSCD
北大核心
2007年第6期986-990,共5页
Control Theory & Applications
基金
国家自然科学基金资助项目(10576027).
关键词
电力系统
负荷预测
模糊逻辑
支持向量机
核函数
power system
load forcasting
fuzzy logic
support vector machine
kernel function
