ZHANG Jiahong，YANG Min，GU Fang，LI Min，GE Yixian，ZHANG Yuexiang，SHEN Lei，BAO Zhiwei（1.Jiangsu Collaborative Innovation Center on Atmospheric Environment and Equipment Technology，Nanjing University of Information Science and Technology，Nanjing 210044，China；2.Jiangsu Key Laboratory of Meteorological Observation and Signal Processing，Nanjing University of Information Science and Technology，Nanjing 210044，China；.School of Electronic and Information Engineering，Nanjing University of Information Science and Technology，Nanjing 210044，China；4.School of Physics and Optoelectronic Engineering，Nanjing University of Information Science and Technology，Nanjing 210044，China）



Development of a High-Precision Silicon Piezoresistive Pressure Transmitter Based on Array Measurement and Data Fusion*

ZHANG Jiahong1，2，3*，YANG Min1，2，3，GU Fang4*，LI Min2，3，GE Yixian1，2，3，ZHANG Yuexiang3，SHEN Lei3，BAO Zhiwei3
（1.Jiangsu Collaborative Innovation Center on Atmospheric Environment and Equipment Technology，Nanjing University of Information Science and Technology，Nanjing 210044，China；2.Jiangsu Key Laboratory of Meteorological Observation and Signal Processing，Nanjing University of Information Science and Technology，Nanjing 210044，China；3.School of Electronic and Information Engineering，Nanjing University of Information Science and Technology，Nanjing 210044，China；4.School of Physics and Optoelectronic Engineering，Nanjing University of Information Science and Technology，Nanjing 210044，China）

The atmospheric pressure is a critical parameter for evaluating atmospheric environmental change.In or⁃der to improve the accuracy of conventional silicon-based piezoresistive pressure measurement system，a high-preci⁃sion digital pressure transmitter has been developed through MEMS pressure sensor array and data fusion with the back propagation（BP）neural network based on particle swarm optimization（PSO）algorithm.The corresponding de⁃signs of hardware and software have been elucidated，and then the transplantation and display optimization of μC/ OS-II with μC/GUI on the STM32 platform have been achieved.Combining with experimental data，the application performance of PSO-BP neural network algorithm in silicon piezoresistive pressure sensors has been finally investi⁃gated and analyzed in terms of performance parameters，such as nonlinear error，hysteresis error，repeatability error and the overall accuracy.The results show that the overall accuracy of the low-cost piezoresistive pressure transmit⁃ter proposed by this paper can reach nearly±0.095%FS in the range of-20℃～60℃，which can basically meet the requirements of meteorological applications.

pressure transmitter；high precision；array measurement；particle swarm optimization algorithm（PSO）；back propagation（BP）neural network；μC/OS-II and μC/GUI

大气压力是评价大气环境变化的关键性参数，其测量精度直接影响气象预报的准确性，因此气象测量中多采用高精度电容式或谐振式MEMS气压传感器，但这些高端产品多为进口，价格高昂。相比较而言，传统硅压阻式MEMS压力传感器具有成本低廉可大批量生产的优势。尽管如此，严重的温漂效应导致其在不同环境温度下的输出曲线不同，加上迟滞和重复性误差，因而在较宽温度范围内此类传感器的总精度通常很难满足气象行业高端应用的要求［1，2］。为提升硅压阻式气压传感器的测量精度以便广泛应用于气象测量领域，如何有效地抑制其对温度的交叉敏感性是亟待解决的关键问题［3］。

目前国内外抑制压阻式传感器温漂的方法主要包括硬件补偿法和软件补偿法［4-5］。硬件补偿方法主要采用硬件补偿电路来消除温度影响，但难以做到完全补偿；软件补偿方法主要将传感器与主控微处理器相结合，利用诸如多元回归法、最小二阶乘法、人工神经网络以及支持向量机等算法对测量参数进行数据融合与修正［6-9］。BP神经网络占用资源少、计算简单、拟合精度较高，因而成为了压阻式传感器温度补偿的首选［3-5，7］。而近些年来，针对BP神经网络算法计算效率低、收敛速度慢、容易陷入局部最小值的问题，各类改进的BP神经网络已被用来提升压阻式压力传感器的测量精度［10-13］。然而相关实验主要采用Matlab仿真进行，不涉及传感器硬件优化设计，实现高精度压阻式气压变送器研制的报道较少［14］。……