High-precision And Fast Reconstruction Method For Temperature Field Based On Acoustic Tomography

Temperature field reconstruction technology based on acoustic tomography is a widely used non-invasive temperature measurement technology.The basic principle is to estimate the temperature of the measured area by the flight time of the sound wave.Once the acoustic flight time between all transmitting and receiving points is obtained,the temperature distribution of the measured region can be estimated by means of a suitable reconstruction algorithm.This temperature distribution is essential for the safe operation and efficient control of all types of thermo-mechanical systems.Industrial boiler temperature distribution monitoring is a typical application of this technology.In addition,studies have begun using acoustic tomography to monitor the temperature distribution of deep-sea hydrothermal vents,atmospheric temperature distribution,storage grain temperature distribution and storage material temperature distribution.The researchers mainly focus on the composition of the measured gas,the measurement of the time of flight of the sound wave,the reconstruction algorithm of the temperature distribution,the layout of the acoustic sensor,and the bending effect of the acoustic wave propagation path.This dissertation focuses on the key problems affecting the accuracy and real-time performance of acoustic temperature field reconstruction technology,such as ultrasonic sensor layout optimization,temperature field reconstruction algorithm and ultrasonic TOF solution system.The research results are as follows:1)The sensor position optimization index and evaluation function of square layout are proposed,and the evaluation function is optimized by adaptive escape particle swarm optimization algorithm,the optimal spatial position of 8 sensors is determined,and the corresponding optimal layout is obtained.The same reconstruction algorithm is used to reconstruct the unimodal symmetric and unimodal skew temperature fields by using uniform layout,random layout and optimal layout.By calculating the maximum absolute error,average error and root mean square error,it is found that all errors of the optimized layout are the smallest.The results show that the reconstruction accuracy of temperature field can be higher by optimizing the layout.2)A Tikhonov regularization reconstruction method based on finite element sum and improved singular value decomposition is proposed.The mathematical model of temperature field reconstruction is established by the finite element method,and the model is solved by the improved singular value decomposition Tikhonov regularization method,which alleviates the problem of underqualitative matrix equation and reduces the reconstruction error.The simulation results show that the algorithm has good stability and improves the accuracy of temperature field reconstruction.The experimental platform was built to reconstruct the unimodal symmetric and unimodal skew temperature fields,and the reconstruction errors were less than 5%.3)The common radial basis functions are analyzed and compared.The logarithmic quadratic radial basis function is proposed,which has better function approximation ability and can be used to reconstruct the measured temperature field better.The temperature fields of single peak symmetry,single peak skew,double peak skew and three peak skew are reconstructed by using the proposed optimal layout.The results show that the reconstruction algorithm has remarkable effect and high reconstruction precision.The meshing method of the circular temperature field is studied,and the influence of the number of meshing on the reconstruction accuracy is investigated.4)The main methods of acoustic time of flight(TOF)measurement are studied in depth,mainly from the time domain,frequency domain and mixed domain.The time domain method mainly focuses on cross-correlation method and waveform fitting method.Frequency domain method mainly discusses the TOF estimation methods of single frequency signal,multi-frequency signal and linear frequency modulation signal.Finally,the ultrasonic TOF transceiver calculation circuit is designed,the ultrasonic TOF measurement system is constructed by FPGA and ARM heterogeneous system,and the acoustic TOF measurement is realized by cross-correlation method.The real-time performance of temperature field reconstruction is improved.