The Study of State Prediction Method for Electronic System Based on Modified Grey Theory

摘要

The state prediction of electronic system usually makes full use of historical information to estimate its future state and tendency aiming at avoiding disastrous faults, which is very significant to the development of condition based maintenance. This paper put an analog filter circuit as an example and the state prediction technology based on grey theory was studied through analyzing the characteristic of the key testing signals, where the metabolism method was presented to make the model parameters change on line and particle swarm optimization algorithm was used to obtain the best prediction dimension. Compared with the ARAM model, the experiment results show that the improved model is fit for the state prediction of electronic system due to its strengths of good precision and performance.