Temperature drift modeling of MEMS gyroscope based on genetic-Elman neural network. (May 2016)

• Record Type: Journal Article
• Authors: Chong, Shen; Rui, Song; Jie, Li; Xiaoming, Zhang; Jun, Tang; Yunbo, Shi; Jun, Liu; Huiliang, Cao
• Is Part Of: Mechanical systems and signal processing. Volume 72/73(2016)
• Page Start: 897
• Page End: 905
• Publication Date: 2016-05
• Subjects: MEMS gyroscope -- Temperature drift -- Elman neural network -- Genetic algorithm; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2015.11.004
• Languages: English
• ISSNs: 0888-3270

Abstract

Abstract: In order to improve the temperature drift modeling precision of a tuning fork micro-electromechanical system (MEMS) gyroscope, a novel multiple inputs/single output model based on genetic algorithm (GA) and Elman neural network (Elman NN) is proposed. First, the temperature experiment of MEMS gyroscope is carried out and the outputs of MEMS gyroscope and temperature sensors are collected; then the temperature drift model based on temperature, temperature variation rate and the coupling term is proposed, and the Elman NN is employed to guarantee the generalization ability of the model; at last the genetic algorithm is used to tune the parameters of Elman NN in order to improve the modeling precision. The Allan analysis results validate that, compared to traditional single input/single output model, the novel multiple inputs/single output model can guarantee high accurate fitting ability because the proposed model can provide more plentiful controllable information. By the way, the generalization ability of the Elman neural network can be improved significantly due to the parameters are optimized by genetic algorithm. Highlights: The temperature variation experiment based on temperature change rate is carried out. Temperature drift model based on T, temperature change rate and △ T is proposed; The genetic-Elman NN is proposed to establish the temperature drift model.