基于模糊C均值聚类算法的电-热互联综合能源系统负荷预测

Load Prediction of Power-thermal Interconnection Integrated Energy System Based on Fuzzy C-means Clustering Algorithm

电-热互联综合能源系统负荷预测参数相似度较高,预测收敛性较差。提出基于模糊C均值聚类算法的电-热互联综合能源系统负荷预测方法。联合能源转换器、负荷预测芯片两个初级应用设备,完成电-热互联综合能源系统的特性负荷元件研究。按照能源系统电-热数据密度估计条件,设定实际均值聚类参数,通过预测系数归一化处理实现基于模糊C均值聚类算法电-热互联综合能源系统负荷预测。实验结果表明,与传统基于RBF电量调试的能源系统负荷预测方法相比,所提方法应用过程中,UDC指标(单位时间内电压输出水平)、IDC指标(单位时间内电流输出水平)均出现上升趋势,有效降低了能源系统特性元件负载电量过高情况发生概率。

The load prediction parameters of the comprehensive energy system with electric-thermal interconnection have high similarity and poor convergence.A load forecasting method for integrated energy system based on fuzzy c-means clustering algorithm is proposed.Two primary application devices,energy converter and load prediction chip,are used to study the characteristic load elements of the integrated energy system.According to the estimation condition of the electrical and thermal data density of the energy system,the actual mean clustering parameters are set,and the load prediction of the integrated energy system based on the fuzzy c-means clustering algorithm is realized by the normalization of the prediction coefficient.Experimental results show that with the traditional power system load forecasting method based on RBF electricity debugging,the proposed method is applied in the process,UDC index(level voltage output per unit time),IDC(per unit time current output level)are rising,effectively reduces the energy load system feature element power too high probability of occurring.