Research on the Method for Improving the Adjustment-Free Rate of Relays Based on GA-BP

Hermetically sealed electromagnetic relays (HSER) are widely used in aerospace, rail transit and other fields due to the high reliability, excellent performance, and compact size. However, the automation level of the product assembly process is relatively low, causing a large amount of labor and time to debug the product during assembly process, which increases the cost of the product. Therefore, a Backpropagation Neural Network optimized by Genetic Algorithm (GA-BP) is proposed to predict the bending angle of static reed, and then use automatic debugging equipment to debug the contact system before final assembly, in order to improve the adjustment-free rate in the final assembly process. This paper takes the bending angle of static reed of a certain type of relay as an example to illustrate the feasibility of applying this method. Compared with the Back-Propagation neural network (BPNN), this method has higher accuracy.