Intelligent Electric Meter Reliability Assessment Based on Improved SVM

The existing smart meter status evaluation methods have some problems, such as high work intensity and insufficient experience of maintenance personnel. If the smart meter fails to be eliminated in time, it will seriously threaten the security and stability of the power grid. In order to improve the accuracy of smart meter status evaluation, a smart meter status evaluation method based on entropy weight method and improved artificial bee swarm optimization support vector machine is proposed. The random initialization of the traditional artificial bee colony algorithm is prone to lead to unstable optimization results and local final problems in the later period. Chaotic search is adopted to initialize the population, and the locally optimal bees in the late period of bee colony convergence are processed into chaos, so as to improve the population diversity and jump out of the local optimal solution. IABC is adopted to optimize SVM, and the classification accuracy of traditional SVM is improved. Aiming at the problem of smart meter status evaluation, firstly, the index system set of smart meter status evaluation is established, and the index weight is calculated by entropy weight method. The status evaluation index of the smart meter is taken as the input of the IABC-SVM, and the status evaluation result of the smart meter is output. The simulation results show that IABC-SVM algorithm has higher convergence speed and accuracy than SVM and PSO-SVM algorithm. Finally, the proposed method is applied to the status evaluation of smart meters.