A Wireless Pressure Monitoring System With Temperature Compensation And Low Power Consumption

In order to solve the problems of the pressure acquisition terminal of the pressure monitoring system such as high cost,difficult deployment and maintenance when wired connection,wireless terminals powered by batteries are commonly used today.Low-power consumption technology has become the key to ensure battery life.Otherwise,piezoresistive pressure sensors used in the pressure acquisition terminal have the characteristics of temperature drift and zero drift,which will bring serious errors to the measurement results.In the case of a large number of pressure acquisition terminals,the individual differences of the sensors should be taken into account for temperature compensation.Therefore,the temperature compensation algorithm become the focus of research.In order to solve these problems,a wireless pressure monitoring system with temperature compensation and low-power consumption is proposed.It combines Io T technology and machine learning technology to better meet current and future industry needs.The wireless pressure monitoring system includes the pressure acquisition terminal and the data monitoring platform.The pressure acquisition terminal is composed of six modules,which are the main control module,NB-Io T module,sensor module,display module,485 module and power module.The power module uses battery power.In order to extend the battery life as much as possible,low-power technology is used for the terminal.The power consumption of the NB-Io T module and the main control module is analyzed.Moreover,combined with other modules and required functions,the best project is formulated to maximize the reduction of power consumption when the terminal equipment works.The function of the data monitoring platform is data processing,display and control,etc.In terms of data processing,a DE-SVM temperature compensation algorithm model is proposed for the characteristics of piezoresistive pressure sensors.Support vector machine(SVM)is a classic algorithm in the field of machine learning.It has outstanding advantages in nonlinear problems and can better solve the random difference between sensors.Differential evolution(DE)algorithm can solve the problem of SVM parameter selection.The combination of the two can quickly find the optimal parameters to ensure the algorithm model Accuracy.At the end,according to characteristics of the temperature compensation and low power consumption of the wireless pressure monitoring system,experiments are designed and tested respectively.The result from the test of single sensor shows that the maximum error and mean square error are reduced by 93.87% and 99.89% respectively after its pressure value is corrected by DE-SVM model.In the case of multi-sensor composed of seven silicon-based piezoresistive pressure sensors,the maximum error and the mean square error are reduced by 93.17% and 99.27%,respectively.The average relative error decreased from 14.06% to 0.45%.Moreover,the temperature points not included in the training data are tested by the established model.The model still performs well.The power consumption measurement experiment of the pressure acquisition terminal shows that the average current value in sleep mode is only 4.16?A.In the case of collecting and uploading every 10 minutes as a cycle,the average current is only 0.5813 m A.When 14500 m Ah battery is used for power supply,the battery life is about 2.84 years.The cycle is increased to half an hour,the theoretical value of battery life can reach more than 8 years.Therefore,the system meets the design requirements of low power consumption.