A novel pressure sensor calibration system based on a neural network

According to the specific input-output characteristics of a pressure sensor, a novel calibration algorithm is presented and a calibration system is developed to correct the nonlinear error caused by temperature. In contrast to the routine BP and RBF, curve fitting based on RBF is first used to get t...

Full description

Saved in:
Bibliographic Details
Published in:Journal of semiconductors 2015-09, Vol.36 (9), p.121-124, Article 095004
Main Author: 彭晓钧 杨坤涛 元秀华
Format: Article
Language:English
Subjects:
Algorithms
Calibration
Convergence
Curve fitting
Neural networks
Pressure sensors
RBF网络
Robustness
Semiconductors
压力传感器
标定算法
校准系统
神经网络
误差平方和
输入输出特性
非线性误差
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:According to the specific input-output characteristics of a pressure sensor, a novel calibration algorithm is presented and a calibration system is developed to correct the nonlinear error caused by temperature. In contrast to the routine BP and RBF, curve fitting based on RBF is first used to get the slope and intercept, and then the voltage-pressure curve is described. Test results show that the algorithm features fast convergence speed, strong robustness and minimum SSE （sum of squares for error）. It is proven by practical applications that this calibration system works well and the measurement precision is better than the design demands. Furthermore, this calibration system has a good real-time capability.
ISSN:1674-4926
DOI:10.1088/1674-4926/36/9/095004