Temperature compensation for MEMS resonant accelerometer based on genetic algorithm optimized backpropagation neural network

• Published in: Sensors and actuators. A. Physical. 2020-12, Vol.316, p.112393, Article 112393
• Main Authors: Wang, Shudong, Zhu, Weilong, Shen, Yajing, Ren, Juan, Gu, Hairong, Wei, Xueyong
• Format: Article
• Language: English
• Subjects: Accelerometers; Back propagation; Back propagation networks; Backpropagation neural network; Error analysis; Genetic algorithm; Genetic algorithms; MEMS resonant accelerometer; Neural networks; Polynomials; Temperature; Temperature compensation; Temperature drift compensation; Temperature effects; Temperature measurement
• ISSN: 0924-4247; 1873-3069
• DOI: 10.1016/j.sna.2020.112393

A backpropagation neural network was utilized for temperature compensation of a vacuum-sealed MEMS resonant accelerometer. A large amount of sensor data was acquired through the temperature calibration experiments and then used to find the optimal solution of the neural network model. Through the analysis on the predicted error, it was found that our neural network approach can greatly reduce the compensation error by 2∼3 orders of magnitude compared to the traditional polynomial fitting method. [Display omitted] •Scale factor nonlinearity of a vacuum-sealed MEMS resonant accelerometer caused by ambient temperature was studied.•The polynomial fitting model and the BP neural network model based compensation algorithms were investigated.•Prediction error of the BP neural network was found 2∼3 orders of magnitude smaller than that of the traditional polynomial fitting method. Temperature compensation with high accuracy is crucial for improving the performance of MEMS resonant accelerometers. In this paper, we propose an effective temperature compensation method based on the backpropagation neural network (BP-NN). First, we analyzed the relationship among the input acceleration, the environmental temperature, the output frequencies, and the scale factor of a MEMS resonant accelerometer through the traditional polynomial fitting method. After that, we introduced the BP-NN improved by genetic algorithm (GA). Numerous experiments were performed to train the BP-NN model and establish the relationships between the input layer and the output layer. Comparison between single-beam working mode and symmetrical double-beam working mode of the MEMS resonant accelerometer proved that the latter had a better temperature compensation effect due to its minimized error caused by temperature measurement. Experimental results show that the maximum error of our approach is 0.017 % over the whole temperature range from -10°C to 80°C, which is 173-times better than the traditional polynomial fitting method.