Storage Life Prediction for Accelerometer based on Nonlinear Degradation Model

For the accelerometers which have nonlinear degradation characteristics, the existing linear regression model is difficult to accurately evaluate the storage life. In order to solve this problem, under the condition of the accelerated degradation test, a storage life evaluating method based on nonlinear regression model is advanced. Analyzed the relationship between the characteristic parameters of the model and the stress level, and deduced the life distribution of probability which is restricred by acceleration model, and proposed a statistical analysis method which combines maximum likelihood estimation and parameter identification method. Finally, evaluated a certain type of accelerometer of constant stress accelerated degradation experiment data, and it proved that the validity of the analysis and the method is effective.