电测与仪表2012，Vol.49Issue(6)：5-9,5.
基于主成分与粒子群算法的LS-SVM短期负荷预测
LS-SVM Short-term Load Forecasting Based on Principal Component Analysis and Improved Particle Swarm Optimization
摘要
Abstract
Short-term load forecasting is of great significance for power system economic operation and development of national economy. Least squares support vector machines (LSSVM) has been successfully applied in load forecasting, which has many unique advantages in the performance of solving the small sample, nonlinear problems. This paper presents a principal component analysis based on support vector machine model, using the principal component analysis to extract the principal components of historical data and eliminate a lot of noise and redundancy, then data extraction from the processed LSSVM training samples, and using improved particle swarm optimization which regards parameters in LSSVM as particles to improve the training speed and prediction accuracy. Finally, the model is applied to short term load forecasting, and has better generalization performance and prediction accuracy compared to SVM and BP neural network.关键词
负荷预测/主成分分析/粒子群优化/最小二乘支持向量机Key words
load forecasting/principal component analysis/particle swarm optimization/least squares support vector machine分类
信息技术与安全科学引用本文复制引用
代鑫波,崔勇,周德祥,陈湘华..基于主成分与粒子群算法的LS-SVM短期负荷预测[J].电测与仪表,2012,49(6):5-9,5.基金项目
中央高校基本科研业务费专项资金资助(11QX80) （11QX80）
