基于支持向量机的家庭负荷控制策略研究

Research on family load control strategy based on support vector machine

针对家庭用户多样性的特点,提出基于支持向量机预测的家庭用电策略优化方案。根据不同家庭用户的历史用电数据,结合与电器使用情况密切相关的天气数据,基于支持向量机预测算法,对于不同用户在不同情况下的用电行为进行预测;在此基础上对用户的用电策略进行优化。仿真分析结果表明,基于支持向量机的预测能够较为准确的预测不同用户在不同情况下的用电情况,而优化策略也能够在兼顾用户用电习惯的同时达到降低用户电费支出、改善用户负荷曲线的目的。

According to the diversity characteristics of household,the civil electricity optimization strategy based on support vector machine（ SVM） was proposed in this paper. According to the electricity utilization history data and weather data,the behaviors of household appliances utilization for different users under different situation can be predicted based on support vector machine（ SVM）. And family power strategy was optimized on the basis of prediction. Simulation analysis results show that the prediction based on SVM can realize accurately forecast the behaviors of different users under different situations,and the two stage optimization strategy can also reduce the user＇s electricity expenses and improve user＇s load curve without change the user＇s habits largely.