.png){kind=icon}
.png){kind=icon}
Return
Compact Temperature and Static Pressure Compensation System for Differential Pressure Sensor Based on Neural Network and Improved Wolf Algorithm
.png){kind=icon}
PRE
AI
Z
P
S
W
Y
Y
L
X
DOI：10.1109/tie.2026.3677631.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
For piezoresistive pressure sensor, sensing performance is typically ensured by compensating primarily for the temperature-induced drift. However, for high-precision differential pressure measurements in high static pressure scenarios, the measurement deviation caused by the static pressure cannot be ignored. Therefore, this study proposed a compact two-parameter compensation system for differential pressure sensors. First, a novel structure was proposed for synchronous acquisition of static pressure, temperature, and differential pressure. In addition, a two-layer microcontroller unit (MCU) compensation hardware was designed considering the nonlinear errors caused by temperature and static pressure, which ensures acquisition accuracy through 16-bit analog-to-digital (AD) and digital-to-analog (DA) converter circuits. Furthermore, a back propagation (BP) neural network based on the improved grey wolf optimal algorithm was employed to model the coupling relationship between temperature, static pressure, and difference pressure, thereby overcoming the limitations of polynomial fitting and achieving high-precision difference pressure sensing under multitemperature and high-static pressure. Calibration experiments were conducted to validate the proposed system, and the results compared with those of typical compensation methods. The maximum error of the compensated differential pressure sensor within the temperature range of −40 to 80 <inline-formula content-type="math/tex" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX" version="MathJax">$ {}^{\boldsymbol{\circ}} $</tex-math></inline-formula>C and static pressure range of 0-30 MPa was within 0.28%FS (% of full scale), demonstrating that the proposed system significantly reduced the error influence caused by static pressure.
Keywords:
Compensation
differential pressure sensor
grey wolf optimization algorithm
MCU
neural network
Journal
IF:
7.2
Papers:
1.8W
Citations:
9.8W
Organization
Cited Papers