Nonlinear Decoupling Study of Piezoelectric 6-Degree-of-Freedom Accelerometer.

Saved in:

Bibliographic Details
Title:	Nonlinear Decoupling Study of Piezoelectric 6-Degree-of-Freedom Accelerometer.
Authors:	Min Li¹ limin780815@cqu.edu.cn, Jianhang Yang² 1447993258@qq.com, Ke Jian² 20160813078@cqu.edu.cn, Lan Qin³ qinlan@cqu.edu.cn, Jingcheng Liu³ jingchengliu@cqu.edu.cn, Jun Liu³ junliu@cqu.edu.cn
Source:	Engineering Letters. Apr2025, Vol. 33 Issue 4, p958-970. 13p.
Subjects:	Angular acceleration, Linear acceleration, Accelerometers, Calibration, Signals & signaling, Mathematical decoupling
Abstract:	Theoretically, a piezoelectric 6-degree-of-freedom accelerometer can measure six-dimensional acceleration values through linear operations. However, due to the influence of calibration equipment, signal conditioning devices, and materials, the output often exhibits nonlinear characteristics. In addition, the current fixed, non-feedback solution method cannot meet the sensor's measurement requirements, necessitating research into high-precision and efficient decoupling methods. First, the measurement principles of the piezoelectric 6-degree-of-freedom accelerometer are analyzed. Then, to enable adaptive decoupling based on nonlinear compensation, the linear decoupling model is adjusted using the sensitivity curve obtained from nonlinear fitting, with iterative updates to the solution matrix. Finally, to achieve high accuracy and efficiency, the number of iterations in the nonlinear compensation adaptive decoupling model is analyzed and optimized. The linear decoupling model is compared and evaluated against the existing nonlinear decoupling method. The errors from linear decoupling, neural network decoupling, and nonlinear compensation decoupling are analyzed. Experimental results show that, compared with linear decoupling and neural network decoupling, the nonlinear compensation decoupling model reduces the average solution error of linear acceleration from 0.1170% and 0.0690% to 0.0067%, and the average solution error of angular acceleration from 6.0185% and 2.2899% to 0.8989%. The time for linear decoupling is 0.000004 seconds, for neural network decoupling 0.02458 seconds, and for nonlinear compensation decoupling 0.00047 seconds, demonstrating the effectiveness and practicality of the adaptive decoupling algorithm with nonlinear compensation. [ABSTRACT FROM AUTHOR]
	Copyright of Engineering Letters is the property of International Association of Engineers (IAENG) and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	Theoretically, a piezoelectric 6-degree-of-freedom accelerometer can measure six-dimensional acceleration values through linear operations. However, due to the influence of calibration equipment, signal conditioning devices, and materials, the output often exhibits nonlinear characteristics. In addition, the current fixed, non-feedback solution method cannot meet the sensor's measurement requirements, necessitating research into high-precision and efficient decoupling methods. First, the measurement principles of the piezoelectric 6-degree-of-freedom accelerometer are analyzed. Then, to enable adaptive decoupling based on nonlinear compensation, the linear decoupling model is adjusted using the sensitivity curve obtained from nonlinear fitting, with iterative updates to the solution matrix. Finally, to achieve high accuracy and efficiency, the number of iterations in the nonlinear compensation adaptive decoupling model is analyzed and optimized. The linear decoupling model is compared and evaluated against the existing nonlinear decoupling method. The errors from linear decoupling, neural network decoupling, and nonlinear compensation decoupling are analyzed. Experimental results show that, compared with linear decoupling and neural network decoupling, the nonlinear compensation decoupling model reduces the average solution error of linear acceleration from 0.1170% and 0.0690% to 0.0067%, and the average solution error of angular acceleration from 6.0185% and 2.2899% to 0.8989%. The time for linear decoupling is 0.000004 seconds, for neural network decoupling 0.02458 seconds, and for nonlinear compensation decoupling 0.00047 seconds, demonstrating the effectiveness and practicality of the adaptive decoupling algorithm with nonlinear compensation. [ABSTRACT FROM AUTHOR]
ISSN:	1816093X