Development Of A Temperature Measurement System Based On FPGA And Acoustic Techniques

Temperature measurement plays a very important role in daily life,medical and healthcare,industrial production and agricultural sectors.With the continuous development of society,the requirements for temperature measurement accuracy become increasingly more demanding,and the temperature measurement environment is becoming more and more complex.Contact methods of temperature measurement can achieve high accuracy,but they are time-consuming and only temperature at a single point can be obtained.Besides,it is easy to damage the sensor in harsh environments,among other shortcomings.In contrast,traditional non-contact temperature measurement methods,such as the radiation method,do not require the sensing elements to directly contact the target object and provide higher accuracy.However,the high cost,vulnerability in dusty environments and the limited ability to measure internal temperature restrict the applicability of such methods in a range of scenarios.Acoustic temperature measurement,as a new non-contact method,has the advantages of low cost,high accuracy,large range,broad application environments and remarkable real-time performance,attracting intensive attention in recent years.The current research is mainly focused on the reconstruction of the two-dimensional temperature field.Due to insufficient sampling rate of the acoustic signals,the accuracy of temperature measurement along a single path is low.Therefore,in this paper an acoustic temperature measurement system based on FPGA is developed.The system exploits the parallel processing capability of FPGA to increase the sampling rate,and transmits multiple acoustic signals to the computer via USB for visualization and storage.A cross-correlation algorithm is used to calculate the transit time and the speed of sound in order to derive the average temperature along the acoustic path.The main work completed in this paper is as follows:1.The principles,advantages and disadvantages of existing temperature measurement techniques are reviewed.The characteristics and advantages of the acoustic temperature measurement method is discussed.The state-of-the-art of acoustic temperature measurement is summarized.Finally,the principle of single-path acoustic temperature measurement is presented and the key factors influencing the accuracy of the method are discussed.2.Design and implement acoustic temperature measurement system.FPGA was selected as the core computing device because of its flexibility and high-speed parallel processing capability.The hardware system includes a FPGA,high-speed ADC and DAC,and a USB 2.0 interface.The digital logic design of FPGA includes the realization of DDS(Direct Digital Synthesis)that generates a sine signal with controllable amplitude and frequency,FIFO(First Input First Output)to buffer the multi-channel ADC data in a ping-pong mode,and control logic of the USB interface for data exchange with the host computer.A piece of software on the host computer was developed to config the USB interface,display multi-channel signals and perform cross-correlation calculations.A multi-threaded method was used to enable the host computer to perform other operations while displaying signals and performing cross-correlation calculations,which facilitates human-computer interaction.3.A test rig for acoustic temperature measurement was developed,an experiment programmed was formulated.A PT100 platinum temperature sensor was used to calibrate the acoustic temperature measurement system.The performance of the measurement was assessed under different sampling rates conditions,and the real-time performance of the system under different conditions was evaluated.Experimental results show that higher sampling rate leads to higher temperature resolution and accuracy,the error of the system is within±0.4℃ at a sampling rate of 12.5 million samples per second,and the resolution is 0.02℃;The real-time performance of the system can be improved by simplifying the signal display and reducing the number of data for cross-correlation calculation,without affecting the temperature measurement accuracy.The system takes about 3 seconds to complete a temperature measurement.