负荷密度非参数核密度估计算法应用

Application of Non-parametric Kernel Density Estimation Algorithm for Load Density

饱和负荷是确定电网发展最终规模的重要指标,不同用地性质的饱和负荷密度指标是影响负荷预测精度的关键因素。为了提高饱和负荷预测精度,对西宁地区8类用地性质的负荷发展数据进行统计、分析和研究,采用非参数核密度估计算法,得到8类负荷密度指标。通过验证显示通过该方法得到指标大大减小了负荷计算误差,提升了负荷预测精准度,可用于确定西宁地区最终负荷发展规模,指导目标网架规划,提高电网建设改造的合理性、可靠性与经济性。

Saturated load is an important indicator to determine the final scale of grid development,the saturated load density index of different power properties is a key factor affecting the accuracy of load forecasting.In order to improve the accuracy of saturation load prediction,the load development data of 8 types of electrical properties in xining area were collected,analyzed and studied.Using a nonparametric kernel density estimation algorithm,eight load density indexes were obtained.Through the verification,which shown that the index obtained by this method greatly reduces the load calculation error,improved accuracy of load forecasting,It can be used to determine the ultimate load development scale in xining area,guiding the target grid planning to improve the rationality,reliability and economy of the grid construction and transformation.