Research On Temperature Compensation Method Of Silicon Piezoresistive Pressure Sensor Based On Artificial Neural Network

With the continuous development of industrial production measurement and control systems,more and more information collection requirements in production and life are increasing,and sensors have already become one of the essential components in modern industrial and agricultural production.Among them,pressure sensors have been widely used.The piezoresistive sensor's resistivity will change when the material deforms.Many materials have the piezoresistive effect,but only those materials with high sensitivity are suitable for sensors,such as semiconductor materials and some elastic materials.Silicon is the most popular piezoresistive semiconductor.Intelligent pressure sensors based on diffused silicon piezoresistive technology have the characteristics of good stability,high sensitivity,strong overvoltage capability,and small hysteresis,it is being favored by more and more people.There are some problems with silicon piezoresistive pressure sensors.The main problem is that the effect of temperature on the silicon pressure sensor chip will cause temperature drift and sensitivity drift,which greatly affects the accuracy and reliability of the sensor.There are many reasons for the temperature drift,but the root cause is that in a Wheatstone bridge composed of four resistance bars,the surface doping concentration of the resistance bars and the width of the diffusion resistance bars are difficult or impossible to be consistent during the manufacturing process.As a result,the temperature coefficients are not equal and the phenomenon of zero drift occurs.The phenomenon of temperature drift is mainly manifested in that when the input pressure is zero,but the actual output of the bridge is not zero,and the output continuously changes with temperature to generate temperature drift,which affects the output stability and accuracy of the bridge.The temperature characteristics of semiconductors cause the piezoresistance coefficient to change with temperature,which make the pressure sensitivity also drifts with temperature.Inaddition,some quality control problems in the subsequent process are also one of the important reasons affecting the sensor,such as the cavity design of the chip and glass,the firmness of the bonding,the specifications of the electrostatic package,the selection and use of silicone oil,and so on.Based on the above factors,compensating the packaged pressure sensor is very important to improve the stability and reliability of the sensor,including the sensitivity temperature drift compensation,the zero temperature drift compensation,and the zero drift calibration,etc.A temperature compensation method to improve the accuracy and reliability of the sensor is presented in this paper when faced with many problems of current silicon piezoresistive pressure sensors.The STC15 series chip is used as the main control,and the 26PCCFA6 G series air pressure sensor produced by HoneyWell is used to design the power module in the construction of the hardware platform.Power module,adder module,LCD module and amplifier module are designed in this paper.The signal is processed by amplification,filtering,and then transmitted to the mobile phone wirelessly via Bluetooth to collect the data.In the process of data calibration,in order to adapt the temperature range to the requirements of industrial production,multiple pressure points were selected for calibration under the temperature condition of-40? to 60?.The experimental results are compensated by artificial neural networks or other methods of networks,and the final results are obtained after comparative analysis.