residential electricity customers classification using multilayer perceptron neural network

This paper proposes a novel approach to analyzing and managing electricity consumption using a clustering algorithm and a high-accuracy classifier for smart meter data. the proposed method utilizes a multilayer perceptron neural network classifier optimized by an imperialist competitive algorithm (ica) called ica-optimized mlp, and a cd index based on fuzzy c-means to optimally determine representative load curves. a case study involving a real dataset of residential smart meters is conducted to validate the effectiveness of the proposed method, and the results demonstrate that the ica-optimized mlp method achieves an accuracy of 98.62%, outperforming other classification methods. this approach has the potential to improve energy efficiency and reduce costs in the power system, making it a promising solution for analyzing and managing electricity consumption.