The stable availability of power supply equipment is a necessary condition to ensure the safe and stable operation of the freeway electromechanical system. The accurate estimation of equipment remaining useful life (RUL) is of great significance to ensure the continuous availability of the power equipment. This paper proposes a RUL estimation method for freeway electromechanical power supply equipment. Based on the operating status monitoring data of the equipment, we make a multi-angle analysis on the degradation by constructing a multi-dimensional parameter set and selecting sensitive parameters from the set. We estimate the power supply equipment RUL by combining multi-variable gray model (MGM (1, n)) and radial basis neural network (RBFNN). MGM(1, n) is utilized to predict the trends of sensitive parameters and RBFNN is utilized to establish a mapping relationship between sensitive parameters and RUL. Experiments performed on monitoring data of UPS power supply battery pack shows a low test root mean square error of 0.046 , which verifies the effectiveness and accuracy of the method.

CICTP