Simulation of thermal drift of fiber-optic gyroscope taking into account piezo-optical effects

The prediction and compensation of the fiber-optic gyro errors caused by the influence of the environmental factors has been the subject of many studies for a long time, but still remains an urgent problem. In the past few years an interest in this problem has been increased due to a need in improving the precision of instruments incorporated into sensitive equipment. The theoretical aspects of thermally induced phase non-reciprocity of fiber-optic gyroscope phase (FOG) are considered in detail by the authors. The operating principle of the fiber-optic gyros is outlined and the main objective of the study is formulated. The paper presents the basic classification of the FOG drifts based on the drift root causes. The main constitutive relations for piezo-optical effects occurring in silica fiber are derived and the procedure of evaluating the design value of the FOG thermal drift is described. The key idea of the computational procedure is the development of solution to the differential equation of motion by the finite difference method. The peizo-optical relations are used to elaborate the iterative scheme to calculate the time of the ray traveling through the FOG optical loop (optical path length). The paper also presents a detailed (step-wise) procedure for the temperature drift computation, which employs both the third-party programs and authors’ codes. The interrelation of computational steps and the precedence of operations to be made for obtaining the required estimates have been determined. Thus, the problem of FOG thermoelasticity was solved based on the STAR-CCM+ applied package, whereas the drift computation was done separately in the MATLAB environment. A qualitative picture of the FOG drift obtained by direct numerical simulation without using the algorithms for signal analysis has been obtained and analyzed. The advantage of the externally connected processor is multiple processing of the initial data obtained from thermoelastic computations by any of the methods and algorithms which have been programmed by users in the processors.