期刊论文

Fluctuation Prediction Method for Equivalent Load of Bus Based on Information Granulation and Support Vector Machine

中文

电力系统及其自动化学报

1003-8930

2019

31

4

25-31

10.3969/j.issn.1003-8930.2019.04.004

51777079:国家自然科学基金 U1766213:国家电网智能电网联合基金资助项目 ZBKJXM20180104:南方电网科技项目

本校为第一机构

随着接入母线多源功率的不断增加,使电网更加合理安排调度计划有了较大的难度。首先,提出了构建一种母线等效负荷模型,将接入母线的不可调发电功率等效为负的负荷功率,使不可调发电功率和母线负荷功率等效为母线的等效负荷功率。然后,获取等效负荷的历史数据,作为母线等效负荷预测模型的输入。最后,基于模糊信息粒化和支持向量机进行母线等效负荷波动预测。实例验证表明,等效负荷预测值相比单独预测不可调多源功率及母线负荷之后的等效值,精确度有所提高。同时预测结果可以更加清楚地了解各母线不可调等效负荷的波动范围,有利于地调系统更好地计划可调小容量发电的出力,并为省调更合理地安排新能...

The constant increase of multi-source power connected to bus causes a larger difficulty to the more rational scheduling of power grid. First,an equivalent load model of bus is constructed by making an equivalence of nonadjust?able generation power to negative load power. Then,the acquired history data of equivalent load are used as input of the prediction model for the equivalent load of bus. Finally,the equivalent load fluctuation of bus is predicted based on fuzzy information granulation(FIG)and support vector machine. The results of a case study show that the accuracy of the predicted value...