Search

MEMS accelerometers are significantly impacted by temperature and noise, leading to a considerable compromise in their accuracy. In response to this challenge, we propose a parallel denoising and temperature compensation fusion algorithm for MEMS accelerometers based on RLMD-SE-TFPF and GRU-attention. Firstly, we utilize robust local mean decomposition (RLMD) to decompose the output signal of the accelerometer into a series of product function (PF) signals and a residual signal. Secondly, we employ sample entropy (SE) to classify the decomposed signals, categorizing them into noise segments, mixed segments, and temperature drift segments. Next, we utilize the time-frequency peak filtering (TFPF) algorithm with varying window lengths to separately denoise the noise and mixed signal segments, enabling subsequent signal reconstruction and training. Considering the strong inertia of the temperature signal, we innovatively introduce the accelerometer's output time series as the model input when training the temperature compensation model. We incorporate gated recurrent unit (GRU) and attention modules, proposing a novel GRU-MLP-attention model (GMAN) architecture. Simulation experiments demonstrate the effectiveness of our proposed fusion algorithm. After processing the accelerometer output signal through the RLMD-SE-TFPF denoising algorithm and the GMAN temperature drift compensation model, the acceleration random walk is reduced by 96.11 % , with values of 0.23032 g/h / Hz for the original accelerometer output signal and 0.00895695 g/h / Hz for the processed signal. [ABSTRACT FROM AUTHOR]

This article suggested two methods to compensate for the temperature drift of the micro-electro-mechanical system gyroscopes, which are support vector machine method and C-means support vector machine. The output of X axis which was ranged from −40°C to 60°C based on the micro-electro-mechanical system gyroscope is reduced and analyzed in this article. The results showed the correctness of the two methods. The final results indicate that when the temperature is ranged from −40°C to 60°C, the factor of B is reduced from 0.424 (° h) to 0.02194 (° h) , and when the temperature is ranged from 60°C to −40°C, the factor of B is reduced from 0.1056 (° h) to 0.0329 (° h) , and the temperature drift trend and noise characteristics are improved clearly. [ABSTRACT FROM AUTHOR]

This paper analyzes different noise components in MEMS gyroscope silicon structure, including mechanical–thermal noise (MTN), electronic-thermal noise (ETN), flicker noise (FN) and Coriolis signal in-phase noise (IPN). The structure equivalent electronic model is established, and the improved differential interface is proposed based on weak signal detection technology, after that, the noise components in silicon structure are introduced and analyzed in sense open loop. The quadrature error (QE) signal automatically cancellation loop is proposed, and the results of the experiment indicate that the equivalent angular rates of QE and IPN are 46°/s and 4.55°/s respectively. The interfaces contrast experiments show that the DC noise and the useful signal amplitudes of differential and single-side detection interfaces are −49.8 dBmV, −16.8 dBmV and −39.8 dBmV (−42.1 dBmV), −22.1 dBmV (−22.2 dBmV), which confirms the differential interface has better SNR. The carrier experiments also illustrate that higher carrier frequency (from 500 kHz to 10 MHz) can restrain DC noise (from −19.8 dBmV to −54.2 dBmV) better, which demonstrate the FN is the dominant noise component of the silicon structure under normal temperature. The temperature experiments show the DC noise enhances from −48.5 dBmV to −14.6 dBmV over the range 20 °C to 60°C while the useful signal amplitude remains around -16.6dBmV, and this phenomenon indicates the MTN and ETN become the dominant structure noise components gradually with temperature rising. [ABSTRACT FROM AUTHOR]

A temperature error processing method for a dual-mass micro-electromechanical system (MEMS) gyroscope based on multi-scale parallel model is proposed. At first, a sample entropy based bounded ensemble empirical mode decomposition (SE-BEEMD) is proposed to decompose the original signal into different scales, include noise-only scale, mixed scale and drift scale; then forward linear prediction (FLP) is employed to eliminate the noise at mixed scale and extreme learning machine (ELM) based model is employed to compensate the drift at drift scale, the two steps are carried out paralleled; at last the final results can be obtained after reconstruction. Experimental results show that: (1) compared to tradition serial model, the proposed parallel model can eliminate the temperature errors more effectively, and each parameter of Allan analysis is improved. Specially, the quantification noise reduced from 0.035 μrad to 9.93e4 μrad, angle random walk reduced from 2.13e-5/s 1/2 to 7.94e-6/s 1/2 , bias instability reduced from 5.28e-4/s to 4.79e-4/s, rate random walk from 0.012/s 3/2 to 0.092/s 3/2 and angular rate ramp reduced from 0.013/s 2 to 0.011/s 2 ; (2) compared to traditional time consuming neural networks, the ELM has the best modeling accurate and shortest training time, which would be valuable for online temperature drift modeling and compensation. [ABSTRACT FROM AUTHOR]