Contact pressure of high-voltage DC power relay change and life prediction and structure optimization. (February 2021)

• Record Type: Journal Article
• Authors: Yu, Guo-Fei; Chiu, Yi-Jui; Zheng, Xiang; Yuan, Zhong-Lin; Wang, Zhi-Xin
• Is Part Of: Advances in mechanical engineering. Volume 13:Number 2(2021)
• Publication Date: 2021-02
• Subjects: High-voltage DC relay -- life prediction -- failure mechanism of contact parallel combination spring -- adaptive neural network system -- gray prediction model; Mechanical engineering -- Periodicals; 621.05
• DOI: 10.1177/1687814021991666
• Languages: English
• ISSNs: 1687-8132

Abstract

To effectively predict the life of high voltage dc power relay, the authors design the structure of the relay to extending its life. The contact failure mechanism of relay products is analyzed by observing the appearance of several contact failures. The electromagnetic-thermal field coupling model is established and the state of heat flow and distribution of temperature increase between electrical and thermal coupling contacts are analyzed through simulations and experiments. An adaptive neural network control system is established to conduct sample training, obtain the best stiffness coefficient of the main spring, and optimize the design of the main spring structure composed of composite springs. The action voltage, release voltage, contact resistance, contact pressure, action time, and release time as characteristic parameters, and the relative error, correlation degree, variance ratio, and small error probability of each characteristic parameter, are calculated. The precision grade index of contact resistance and action time is defined as level 1. The two characteristic parameters are selected as the prediction variables to establish the life-prediction model based on gray theory. The relative error between the predicted and test life is 5.31%–5.6%, indicating that the life-prediction model has high accuracy. The test results show that the combined main spring structure can extend the life of a relay by 18.15%.