基于集群辨识的电力用户需求响应潜力评估

Evaluation of Power User Demand Response Potential Based on Cluster Identification

针对电力市场参与需求响应(DR)用户的特性不一、响应能力差异大,导致DR潜力评估难度大的问题,提出了一种自底向上的负荷DR潜力评估方法。首先,对用户的用电行为进行分析,将每个用户的负荷变化分别用基于概率分布和隐马尔科夫模型的聚类方法进行聚类,得到适合价格和激励的DR用户分类。然后,将用户负荷大小用K-means算法进行聚类,得到每个用户所属的行业类别。在对每个用户的多标签分类和电力DR潜力进行定性评估后,通过计算区域内各用户负荷与总负荷之间的皮尔逊系数,得到用户的调峰潜力。最后,通过具体算例分析某区域用户的DR潜力和调峰潜力,验证了所提方法的有效性。

Due to the different characteristics and large differences in demand response capabilities of participating interactive users,it is more difficult to evaluate the demand response potential.In this regard,a bottom-up method is proposed to evaluate the load demand response potential.First,the load changes of each user are clustered using the clustering method based on probability distribution and the clustering method based on hidden Markov model to obtain DR user classification suitable for price and incentives.Then the user load is clustered using the K-means algorithm to obtain the industry category to which each user belongs.In this way,the multi-label classification of each user and the qualitative evaluation of the response potential of each user's power demand are completed,and the peak shaving potential of the user can be obtained by calculating the Pearson coefficient between the load of each user and the total load in the area.Finally,in order to verify the effectiveness of the proposed method,the evaluation results of the demand response potential of users in a certain area and the peak shaving potential are obtained through the analysis of examples.