Research on Intelligent Recognition Method for Electrical Equipment and Meter

In order to improve the recognition speed of the BP neural network for digital display instruments in electrical equipment and meters and improve the training speed of the BP neural network, principal component analysis (PCA) is introduced to select the number of nodes in the hidden perception layer in the BP neural network. Get the number of optimal solutions under the premise of ensuring the recognition accuracy, the optimal solution of the number of nodes is obtained, and finally it is simulated and verified by MATLAB. The simulation results prove that the PCA improved BP neural network algorithm has fewer training iterations, and the training speed of the BP neural network has been improved. In the recognition of digital meters, the improvement of the training speed also improves the recognition speed. For the identification of 900 test instrument images, the entire identification process has been improved from 1506ms to 856ms, and the recognition accuracy has also been improved, from 96.9% to 98.6%. Therefore, the improvement of BP neural network algorithm based on PCA in this paper basically meets the design requirements, and can perform intelligent identification of electrical equipment meters and digital display meters.