Optimization Research on the Circuit Breaker Instantaneous Characteristic Experiment Based on Neural Network

摘要

The circuit breaker's instantaneous characteristic test equipment controlled by computer is introduced in this paper. There are still some problems in the design. To address those issues, compensation coefficient is presented in order to improve the accuracy of instantaneous current. Based on analyzing the influencing factors of instantaneous current and establishing equivalent model, neural network control method is adopted. Under LabVIEW software environment, neural network which has been trained by MATLAB script node is called to compute the compensation coefficient. In order to enhance system adaptability, the neural network online adjustment is achieved by changing training samples. The accuracy of instantaneous current is further improved.