Dual Layer Evaluation Method for User Adjustable Potential of FCM Combined with GMM

To address the challenge of current user electricity consumption behavior analysis in meeting the demands of modern power systems, this paper proposes a two-layer evaluation method for user adjustable potential based on Fuzzy C-Means (FCM) combined with Gaussian Mixture Model (GMM). This method employs a two-layer clustering approach: first, the FCM algorithm is used to perform initial clustering on user load data, identifying typical electricity consumption patterns; second, the Binary Grey Wolf Optimizer (BGWO) algorithm is utilized to screen load characteristics of power users; finally, the GMM algorithm is applied to further analyze the probability distribution and refine the classification of these clustering results. Experimental data is sourced from the 2009–2010 smart meter electricity consumption behavior experiment conducted by the Commission for Energy Regulation in Ireland, which includes typical daily load curves of 146 users. By comparing the comprehensive clustering index DSC, the effectiveness of the FCM -GMM two-layer clustering method is validated. The results demonstrate that this method can effectively evaluate user adjustable potential, providing a reliable basis for optimized scheduling and refined management of power systems.