Fiber Optic Gyro Temperature Error Compensation Based on Multi- parameter Model

In order to fit the non-linear relationship between zero bias of FOG and temperature changes, a multi-parameter fitting model is established to compensate for zero offset, and the bias stability of FOG is significantly improved. The effect of temperature on fiber optic gyro (FOG) constitutes of several factors. Noise and drift are two different characteristic factors of the temperature's effect to FOG. First, the noise part of FOG output is removed by using the Ensemble Empirical Mode Decomposition (EEMD). Then combined with the correlation analysis, the multi-parameter model is established to fit and compensate bias. Multiple regression analysis (MRA) method is used to obtain the model parameters, and temperature error compensation of FOG is carried out by using the established model. The model of temperature rise and fall process arc established respectively, and then the uniformity of model is verified. Experiment results show that the model can effectively compensate temperature error of FOG and is very adaptable. After compensation, the error ensemble average and variance are reduced by three orders of magnitude, the max bias is reduced by 60 percent.