电力市场中多日负荷曲线的预测

Multi-day load curve forecasting in electricity market

鉴于电力市场对负荷曲线预测提出的新要求 ,提出了连续多日负荷曲线预测 ,并构造了严格的优化模型 ,可以用成熟的二次规划方法求解。为加快计算速度提出了根据样本数据调整预测的思路 ,并探索了 2种不同的调整途径。一种通过对样本曲线的等比调整使其满足要求 ,另一种则通过构造分段映射进行预测。算例分析表明 ,这 2个方法能够在很短的时间内取得较好的预测结果 ,符合电力系统实际工程对预测精度和计算时间的要求。

Along with the development of electricity market,new demand for load curve forecasting arises.A new concept of continuous multi day load curve forecasting is proposed and a mathematical optimization model is constituted strictly in order to solve the quadratic programming with the mature methods.An idea based on sample data adjust is presented to speed up the computation.Two different ways of sample data adjust are proposed,one with proportional adjust and the other with subsection mapping.Case study shows that both ways can get better forecasting results in short time,meeting the requirements of power system engineering for precision and speed.