Tri-parameter power function model linearization method for accelerated degradation of accelerometer

Abstract

The invention provides a tri-parameter power function model linearization method of accelerated degradation of an accelerometer. In the method, assuming that zero offset and scale factor degradation of an accelerometer conform to a tri-parameter power function model and a corrected parameter is unrelated to the temperature; raising the highest temperature mean square related coefficient maximum principle determined by the corrected parameter under the condition that the experimental time under different temperatures are basically the same; and realizing the linearization of the tri-parameter power function model by fixed corrected parameters. The method comprises the following specific steps: 1, building a degradation track model related coefficient function under the highest temperature; 2, building a degradation track model mean square related coefficient function under the highest temperature; 3, determining a degradation track model corrected parameter; and 4, performing linearization and parameter identification for the degradation track model. The method of the invention can avoid the initial value sensitive and convergent problems of non-linear fitting, and ensure significance of the whole fitting effect of the model and the integrity of the degradation track characteristics by the high-temperature mean square related coefficient.

Description

technical field [0001] The invention provides a three-parameter power function model linearization method for accelerated degradation of an accelerometer, which belongs to the technical field of reliability evaluation of accelerated degradation tests. Background technique [0002] The accelerometer is one of the key components of the inertial navigation system. It is used to measure the linear acceleration of the carrier, and then obtain the trajectory (speed and distance) of the carrier through integration. It has a wide range of inertial measurement and guidance in aviation, aerospace, and ships. application. [0003] Bias and scale factor are the key parameters to determine the accuracy of accelerometer measurement. However, a large amount of observation data shows that with the increase of storage time, the zero bias and scale factor will change slowly, which will have a bad influence on the measurement accuracy of the accelerometer. The time during which the accelerom...

AI Technical Summary

Problems solved by technology

However, nonlinear fitting inevitably has initial value sensitivity and convergence problems, which makes model parameter estimation often not unique, and even leads to unreasonable understanding of pseudo-stable period estimation.

Images