光伏影响下考虑气象负荷分解和LSSVM的负荷预测
投稿时间:2019-09-27  修订日期:2019-10-30  点此下载全文
引用本文:
摘要点击次数: 207
全文下载次数: 0
作者单位邮编
蔡冬阳* 国网江苏省电力有限公司 210024
中文摘要:随着光伏电源并网规模的不断扩大,光伏电源出力的波动性使得负荷预测难度加大。另外,气象因素对电力系统负荷有显著的影响作用,所以,本文首先考虑剥离光伏电源对电网负荷预测的影响后,研究实时气象因素对电力系统净负荷的影响,然后将净负荷分解为基础负荷和气象敏感负荷,采用灰色模型GM(1,1)和最小二乘支持向量机(least squares support vector machine, LSSVM)算法分别对二者进行预测。通过剥离光伏影响得到系统净负荷,之后采用主导气象因素辨识方法分析影响净负荷的主要气象因素,合理选取预测模型的输入向量,实现了考虑光伏影响与气象敏感负荷分解的LSSVM负荷预测。实际应用验证了本文所提出的模型能够明显提高负荷预测的准确度。
中文关键词:光伏  实时气象因素  气象敏感负荷分解  最小二乘支持向量机  负荷预测
 
Load forecasting research based on weather sensitive load decomposition and LSSVM under the influence of photovoltaic
Abstract:With the continuous expansion of grid-connected photovoltaic power supply scale, the fluctuation of photovoltaic power output makes load forecasting more difficult. In addition, meteorological factors have a significant impact on power system load. Therefore, after considering the impact of stripping off photovoltaic power supply on power system load forecasting, this paper studies the impact of real-time meteorological factors on power system net load, and then decomposes the net load into basic load and meteorological sensitive load, using grey model GM (1,1). The least squares support vector machine (LSSVM) algorithm and least squares support vector machine (LSSVM) algorithm are used to predict them respectively. The net load of the system is obtained by stripping off the influence of photovoltaic. Then the main meteorological factors affecting the net load are analyzed by using the identification method of dominant meteorological factors. The input vectors of the prediction model are reasonably selected, and the LSVM Load Forecasting Considering photovoltaic effect and weather sensitive load decomposition is realized. Practical application shows that the proposed model can significantly improve the accuracy of load forecasting.
keywords:Photovoltaic, hourly weather factors, weather sensitive load decomposition, least squares support vector machine, load forecasting
查看全文   查看/发表评论   下载pdf阅读器