Electric energy metering box electricity load data analysis method based on big data analysis

Abstract

This invention discloses a method for analyzing electricity load data of electricity metering boxes based on big data analytics, belonging to the field of big data analytics technology. The method includes: collecting electricity metering operation sequence data and generating standardized metering time-series data; establishing a time-series dynamic correlation graph and performing graph homotopy mapping to generate a periodically modulated input current sequence; constructing an improved Izhikevich neuron model to generate a load excitation response sequence; calculating the load behavior correlation characteristics between electricity metering boxes and constructing an electricity metering box load correlation graph; updating the connection edge weights based on Ricci curvature to generate a load correlation structure evolution sequence; constructing a collaborative identification feature vector and outputting the electricity load analysis results. This invention, by introducing an improved Izhikevich neuron model and using the Ricci curvature method, achieves the functions of identifying sudden electricity load behavior and dynamically analyzing and locating abnormal propagation paths in electricity metering boxes.

Images