Remaining Useful Life Prediction of Smart Electricity Meters Based on Particle Filter and a Four-Parameter Degradation Model

Abstract

The smart electricity meter (SEM) is a critical element of the smart grid, so power supply company and customers pay many attention to its service life. However, the existing reliability estimation methods can only provide an overall distribution of the SEM lifetime at certain conditions. To realize real-time prognostics and health management (PHM) of SEMs, this paper proposed a four-parameter model-based particle filter method. Firstly, an accelerated degradation testing is carried out to analyze the aging state of SEMs. Then, a four-parameter degradation model is proposed, and the fitting effectiveness is evaluated. Finally, particle filter is utilized to identify the model parameter and further obtain the RUL distribution of SEMs in service.