基于多维指标的园区用户生命周期评估模型

Life Cycle Assessment Model of Park User Based on Multidimensional Metrics

随着智能电网的进步和电力物联网的发展,依托园区电力大数据分析用户用能特点及与供电企业互动行为并评估其生命周期阶段,有助于深入洞察客户需求,为其提供差异化、多样化的增值服务。首先根据园区用户用电数据提取生命周期评估指标集,其次为消除传统熵权法在熵值接近1时带来的巨大误差,使用改进熵权法求取基础指标对于中级指标的权值;再使用TOPSIS和灰色关联度分析组合方法计算各中级指标与最优方案之间的接近水平;最后由专家打分确定中级指标相对高级指标权重并得到各用户每月发展水平,拟合每月发展水平得到能够预测未来发展情况的生命周期曲线,进而划分园区用户生命周期阶段并准确把握各个阶段用户用电行为及互动行为变化趋势。实例分析表明该模型能有效且准确地划分园区用户生命周期阶段,与以往算法相比提高了生命周期阶段划分的准确度。

With the progress of smart grid and the development of power Internet of Things,relying on the park power big data to analyze user energy consumption characteristics and the interactive behavior of the user and power supply enterprise and evaluate its life cycle stage,helps to gain insight into user needs,and provide them with differentiated and diversified value-added services.Firstly,the evaluation index set of life cycle is extracted according to the park user’s electricity consumption. Secondly,to eliminate the big error brought about by traditional entropy weight method when the entropy value is close to 1,improved entropy weight method is used to obtain the weights of the basic indices for intermediate indices,and TOPSIS-gray correlation analysis combination method is also used to calculate the close level between the intermediate indices and the optimal scheme. Finally,the weights of the intermediate indices relative to the advanced indices are determined with expert scoring method and the monthly development level of each user is obtained. The life cycle curve of future development can be predicted through fitting the monthly development levels. Moreover,the life cycle stage of the park users is divided,and the trends of change are obtained in the user power consumption behavior and interactive behavior at each stage. The example analysis shows that the model can effectively and accurately divide the life cycle stage of the park users,improving the division accuracy compared with the previous algorithm.