基于混沌類電磁算法優化支持向量機的短期負荷預測

王茜 李皓然 王新娜 張媛媛

摘? ?要：短期負荷預測準確性對于電網態勢感知和電網策略具有十分重要的意義。提出一種基于混沌類電磁學（CEM）優化支持向量機的短期負荷預測方法，該方法利用聚類思想判斷數據質量并進行相關數據預處理工作。建立支持向量機的短期負荷預測模型，針對傳統支持向量機參數選擇困難問題，引入混沌類電磁學算法優化參數，提高算法收斂效率和尋優能力。仿真結果表明：所提算法較傳統支持向量機算法和粒子群-支持向量機算法（PSO-SVM）收斂速度更快，尋優能力更強，適用于短期負荷預測。

關鍵詞：負荷預測;類電磁學;支持向量機

中圖分類號：TM796 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?文獻標識碼：A

Short Term Load Forecasting Based on SVM

and Chaos Electromagnetic Algorithm

WANG Qian，LI Hao-ran？覮，WANG Xin-na，ZHANG Yuan-yuan

（Skills Training Center of State Grid Jibei Electric Power Company Limited

（Baoding Electric Power Voc.&Tech. College），Baoding，Hebei 071000，China）

Abstract：The accuracy of short-term load forecasting is very important for power grid situation awareness and power grid strategy. A short-term load forecasting method based on chaotic electromagnetics （CEM） optimization support vector machine （SVM） is proposed. This method uses clustering idea to judge the data quality and preprocess the related data. A short-term load forecasting model of SVM is established. Aiming at the difficult problem of parameter selection of traditional SVM，a new method is introduced. Chaotic electromagnetism algorithm optimizes parameters，and improves the convergence efficiency and optimization ability of the algorithm. Simulation results show that the proposed algorithm has faster convergence speed and stronger optimization ability than the traditional support vector machine algorithm and particle swarm optimization support vector machine （PSO-SVM），and is suitable for short-term load forecasting.

Key words：load forecasting;electromagnetics;support vector machine