基于数据挖掘技术和聚类分析算法的台区线损分析模型研究

Research on Line Loss Analysis Model of Station Area Based on Data Mining Technology and Clustering Analysis Algorithm

台区线损的分析,能够有效保障电网的电能质量。利用模糊C均值聚类的数据挖掘方法,构建模型对台区的线损进行预测,从实验结果可证明该模型具有很好的预测效果,模型的训练均方误差结果为3.0675,而预测的均方误差结果为0.0322。分析考虑了预测模型的影响因素,从结果上可以看到,增加输入变量个数和样本点个数,能有效提高模型的训练效果,预测精度也趋于稳定。在模型中综合考虑了售电量、功率因数、季节等其他影响因素,预测精度尽管稍微有所下降,但还是能满足预测的要求,证明了该模型的可行性。

The analysis of line loss in substation area can effectively ensure the power quality of power grid.Using the data mining method of fuzzy c-means clustering,a model is constructed to predict the line loss in the station area.The experimental results show that the model has a good prediction effect.The training mean square error of the model is 3.0675,while the prediction mean square error is 0.0322.The analysis considers the influencing factors of the prediction model.From the results,increasing the number of input variables and sample points can effectively improve the training effect of the model,and the prediction accuracy tends to be stable.In the model,other influencing factors such as electricity sales,power factor and season are comprehensively considered.Although the prediction accuracy decreases slightly,it can still meet the prediction requirements,which proves the feasibility of the model.