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

Wang, Shudong; Zhu, Weilong; Shen, Yajing; Ren, Juan; Gu, Hairong; Wei, Xueyong

doi:10.1016/j.sna.2020.112393

Cited by 38 publications

(25 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, the resonant frequency shift output metric for MEMS accelerometers is also strongly affected by the environmental variations including temperature and pressure [ 5 , 6 ]. This requires additional error compensation techniques for the stable operation of such resonant MEMS accelerometers [ 7 , 8 ].…”

Section: Introductionmentioning

confidence: 99%

A Low-g MEMS Accelerometer with High Sensitivity, Low Nonlinearity and Large Dynamic Range Based on Mode-Localization of 3-DoF Weakly Coupled Resonators

et al. 2021

View full text Add to dashboard Cite

This paper presents a new design of microelectromechanical systems (MEMS) based low-g accelerometer utilizing mode-localization effect in the three degree-of-freedom (3-DoF) weakly coupled MEMS resonators. Two sets of the 3-DoF mechanically coupled resonators are used on either side of the single proof mass and difference in the amplitude ratio of two resonator sets is considered as an output metric for the input acceleration measurement. The proof mass is electrostatically coupled to the perturbation resonators and for the sensitivity and input dynamic range tuning of MEMS accelerometer, electrostatic electrodes are used with each resonator in two sets of 3-DoF coupled resonators. The MEMS accelerometer is designed considering the foundry process constraints of silicon-on-insulator multi-user MEMS processes (SOIMUMPs). The performance of the MEMS accelerometer is analyzed through finite-element-method (FEM) based simulations. The sensitivity of the MEMS accelerometer in terms of amplitude ratio difference is obtained as 10.61/g for an input acceleration range of ±2 g with thermomechanical noise based resolution of 0.22 and nonlinearity less than 0.5%.

show abstract

Section: Introductionmentioning

confidence: 99%

A Low-g MEMS Accelerometer with High Sensitivity, Low Nonlinearity and Large Dynamic Range Based on Mode-Localization of 3-DoF Weakly Coupled Resonators

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The architecture of backpropagation neural networks still becomes the preferable topic in neural networks, such as optimization in the number of hidden layers [16] [17], and another research in sensors, modeling backpropagation based on GA by specifying the number of hidden layer neurons [18]. Despite its slow training, a backpropagation neural network is easy to use and design depending on the input characteristics, whether univariate or multivariate inputs [19].…”

Section: Introductionmentioning

confidence: 99%

“…Purnawansyah et al / Knowledge Engineering and Data Science 2021, 4 (1):[14][15][16][17][18][19][20][21][22][23][24][25][26][27][28] …”

unclassified

Backpropagation Neural Network with Combination of Activation Functions for Inbound Traffic Prediction

et al. 2021

View full text Add to dashboard Cite

Predicting network traffic is crucial for preventing congestion and gaining superior quality of network services. This research aims to use backpropagation to predict the inbound level to understand and determine internet usage. The architecture consists of one input layer, two hidden layers, and one output layer. The study compares three activation functions: sigmoid, rectified linear unit (ReLU), and hyperbolic Tangent (tanh). Three learning rates: 0.1, 0.5, and 0.9 represent low, moderate, and high rates, respectively. Based on the result, in terms of a single form of activation function, although sigmoid provides the least RMSE and MSE values, the ReLu function is more superior in learning the high traffic pattern with a learning rate of 0.9. In addition, Re-LU is more powerful to be used in the first order in terms of combination. Hence, combining a high learning rate and pure ReLU, ReLu-sigmoid, or ReLu-Tanh is more suitable and recommended to predict upper traffic utilization

show abstract

“…A resonator is the basis for combining many devices, such as radio frequency filters and resonant sensors [ 8 , 9 , 10 ]. In a resonant accelerometer, the resonator is the core sensitive element [ 11 , 12 ]. Thus, the mode and frequency of the resonator affect the performance of the resonant accelerometer greatly [ 13 , 14 ].…”

Section: Introductionmentioning

confidence: 99%

Design and Optimization of the Resonator in a Resonant Accelerometer Based on Mode and Frequency Analysis

Jin

Zhao

et al. 2021

Micromachines

View full text Add to dashboard Cite

This study aims to develop methods to design and optimize the resonator in a resonant accelerometer based on mode and frequency analysis. First, according to the working principle of a resonant accelerometer, the resonator is divided into three parts: beam I, beam II, and beam III. Using Hamilton’s principle, the undamped dynamic control equation and the ordinary differential dynamic equation of the resonant beam are obtained. Moreover, the structural parameters of the accelerometer are designed and optimized by using resonator mode and frequency analysis, then using finite element simulation to verify it. Finally, 1 g acceleration tumbling experiments are built to verify the feasibility of the proposed design and optimization method. The experimental results demonstrate that the proposed accelerometer has a sensitivity of 98 Hz/g, a resolution of 0.917 mg, and a bias stability of 1.323 mg/h. The research findings suggest that according to the resonator mode and frequency analysis, the values of the resonator structural parameters are determined so that the working mode of the resonator is far away from the interference mode and avoids resonance points effectively. The research results are expected to be beneficial for a practical resonant sensor design.

show abstract

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

Cited by 38 publications

References 30 publications

A Low-g MEMS Accelerometer with High Sensitivity, Low Nonlinearity and Large Dynamic Range Based on Mode-Localization of 3-DoF Weakly Coupled Resonators

A Low-g MEMS Accelerometer with High Sensitivity, Low Nonlinearity and Large Dynamic Range Based on Mode-Localization of 3-DoF Weakly Coupled Resonators

Backpropagation Neural Network with Combination of Activation Functions for Inbound Traffic Prediction

Design and Optimization of the Resonator in a Resonant Accelerometer Based on Mode and Frequency Analysis

Contact Info

Product

Resources

About