Research On Acoustic Temperature Measurement And Leakage Source Location Method Of Non-uniform Temperature Field In Furnace

In large-scale coal-fired power plant boilers,obtaining a high-quality measurement of the temperature distribution is crucial to achieve safe and economical operation of the boiler and low pollution emissions,and provide accurate information on the distribution of important temperature fields to locate the water-cooling wall leakage of power plant boilers.The acoustic method for temperature field measurement and leakage sound source location has become the most promising method in power station boiler measurement and monitoring technology.Based on the acoustics theory,this paper studies the acoustic temperature measurement in the furnace and the sound source localization method of the water-cooling wall leakage in the non-uniform temperature field based on the sound location fingerprint.The key technologies are studied,including the propagation of sound waves in the non-uniform temperature field,high-precision acoustic temperature measurement algorithms,time delay algorithms and leakage source location method in non-uniform temperature fields,in order to provide theoretical support and technical guidance for the practical application of the acoustic method.This article mainly completed the following work:(1)Due to the three-dimensional non-uniform temperature distribution in the furnace,the propagation characteristics of sound waves in the non-uniform temperature field are studied.Using the Fermat principle and the variational method,the mathematical model of the sound propagation path in the three-dimensional non-uniform temperature field is given.The shooting method is used to numerically simulate the propagation process of the sound ray in the two-dimensional and three-dimensional non-uniform temperature field.The improved Powell algorithm is used to construct the multi-dimensional search algorithm to obtain the optimal sound wave exit angle direction.The results show that acoustic waves bend from the high temperature area to the low temperature area;The three-dimensional single-peak symmetrical temperature field has a considerable influence on the sound propagation path as a spherical lens,the "lens effect" of sound waves occurs,but the refractive index changes spherically symmetrically in the radius of the sphere,and when the sound ray is bent,the propagation time of the sound ray is shorter than that of the straight line.(2)The high-precision acoustic measurement method of complex temperature field is studied.Based on the mathematical model of the sound propagation path,the influence of the coupling of the sound field and the temperature field on the reconstruction of the temperature field is taken into consideration,and an acoustic temperature measurement system is established.Using a typical 2D/3D boiler temperature field model,the simulation reconstruction results are compared with and without considering the refraction effect of sound waves under different noise levels.Numerical simulation results show that the reconstruction method based on radial basis function approximation with polynomial reproduction(RBF-PR)and the improved regularization reconstruction algorithm can achieve higher measurement accuracy compared with the classic reconstruction method;When considering the effect of refraction,the method can reconstruct more excellent reconstruction performance than others,which do not take into account the refraction effect of sound wave paths.,In addition,the algorithm has the advantages of strong noise resistance and good applicability.(3)The acoustic temperature measurement experimental system can realize the acoustic measurement of the two-dimensional/three-dimensional temperature field under laboratory conditions.The time delay estimation algorithm in the experimental system is very important for the measurement of the sound wave propagation time.The measurement noise interferes with the accuracy of the time delay estimation.A generalized cross-correlation with second correlation is applied to estimate the the sound wave propagation time to effectively suppress interference from noise.The acoustic experimental results of the two-dimensional thermal temperature field show that the proposed method based on radial basis function approximation with polynomial reproduction and truncated singular value decomposition(RBF-PR-TSVD)can effectively improve the reconstruction performance of the temperature field,when considering the refraction effect in the non-uniform temperature field,this reconstruction method can accurately describe the temperature field distribution and reconstruct the location of the hot spot.At the same time,this method can detect the movement of the temperature field hot spot,which can be used to diagnose the deflection of the flame center in the furnace.The acoustic experimental results of the three-dimensional thermal temperature field show that the improved acoustic measurement method considering the refraction effect has the advantages of high temperature measurement accuracy and fast convergence speed.(4)The temperature field information in furnace obtained by acoustic temperature measurement and the propagation characteristics of sound waves in this non-uniform temperature field provide important acoustic information to locate the leakage of the water-cooling wall pipeline.Radial basis function artificial neural network based on sound location fingerprint is used to locate the leakage souce.Numerical simulations and laboratory experiments verify the feasibility and positioning performance of the positioning algorithm.The effects of the experimental environment,time delay algorithm,number of samples,and mesh size on the positioning results are studied and analyzed.The results show that the radial basis function neural network positioning algorithm based on the sound location fingerprint can obtain the coordinates of the leakage sound source in a non-uniform temperature field,and realize high-precision leakage positioning.