Temperature compensation model of MEMS multi-ring disk solid wave gyroscope based on RIME-VMD and PSO-DBN multi fusion algorithm

Abstract

In order to address the problem of temperature-induced excessive output drift in MEMS gyroscopes, a zero-bias temperature compensation model based on RIME-VMD-PSO-DBN multiple fusion algorithm is developed in this study. The model utilizes RIME to optimize the input parameters in the VMD algorithm and decomposes the temperature drift of MEMS multi-ring disk solid wave gyroscope through RIME-VMD. The temperature drift signal is decomposed into multiple intrinsic mode functions (IMFs), and is classified and processed according to the frequency. The high-frequency noise term, intermediate frequency mixed noise term, and temperature feature term are processed by PSO-DBN and CWT, respectively. Finally, the processed signal is reconstructed by ELM. Experimental results show that under the conditions of −40 ℃ ∼ 60 ℃, the zero bias instability decreases from 78.48°/h to 0.81 °/h, and the angle random walk reduces from 36.2$°/\sqrt{h}$ to 0.15$°/\sqrt{h}$. Under the conditions of 60 ℃ −40 ℃, the zero bias instability declines from 49.32 °/h to 0.98 °/h, and the angle random walk reduces from 47.53$°/\sqrt{h}$to 0.42$°/\sqrt{h}$.