A mine power distribution cabinet state intelligent monitoring control method and system

Abstract

The application relates to the technical field of mine power distribution cabinets, and discloses a mine power distribution cabinet state intelligent monitoring control method and system, which collects electrical parameter data, mechanical vibration data and thermal imaging data of the power distribution cabinet to obtain multidimensional sensing data; the collected multidimensional sensing data is dynamically calibrated and noise filtered to obtain preprocessed data; the preprocessed data is input into a Transform anomaly detection network to output a fault probability score and a fault type, the electrical coupling relationship between the power distribution cabinets is cooperatively learned through a graph neural network, and a cooperative anomaly score of each power distribution cabinet and a fault propagation path prediction are output; the output result of the Transform anomaly detection network and the output result of the graph neural network are input into a double-layer Bayesian fusion framework to generate final state monitoring results and decision results, and a hierarchical early warning and control instruction is triggered based on the state monitoring results; the application significantly improves the accuracy of state monitoring.

Images