| As a new kind of non-contact temperature measurement technology,acoustic pyrometry has many benefits when used in the real-time monitoring of temperature in the power plant boiler,such as good adaptability under severe environment,continuity,wide range,high resolution of temperature measurement and low maintenance.Hence it has a broad application prospects in power plants.The measurement of acoustic wave transit time in the furnace is the key technology of acoustic pyrometry,whose precision will directly affect the accuracy of the temperature field reconstruction.However,modern power station boilers are complex in structure and large in size.Acoustic temperature measurement signals will be significantly attenuated when they propagate in high-temperature flue gas for a long time.In addition,there are a lot of hot-state background noises in boilers which can strongly interfere with the measurement of acoustic wave transit time.Besides,because of the limitation of the technology level,the measurement accuracy of acoustic wave transit time cannot be improved by simply increasing the power of the sounding device.Hence,how to accurately measure the acoustic wave transit time when faced with the problem of acoustic attenuation and a large amount of strong background noise becomes the key to acoustic pyrometry for obtaining the actual flue gas temperature in the furnace and the problem that acoustic temperature measurement technology needs to focus on when it is used in power station boilers.In order to solve this problem,on the one hand,it is necessary to find a time delay estimation algorithm suitable for the acoustical temperature measurement of a power station boiler.On the other hand,it is necessary to select a suitable sound source signal,which is more conducive to the time delay estimation algorithm to effectively distinguish it from a large amount of strong background noise,so as to obtain an accurate acoustic wave transit time.This article is based on the above two aspects of research work,and the main research contents include:1.Based on the basic theory of high-order cumulants and related knowledge of adaptive time delay estimation algorithm,fourth-order cumulant and ETDGE algorithm are selected as signal processing methods for the specific environment of acoustic temperature measurement in power station boilers.The adaptive acoustic delay estimation algorithm based on higher-order cumulants is attempted to improve,and the FOC-ETDGE algorithm is proposed.Through MATLAB simulation experiments,this algorithm is compared with the widely applied cross-correlation delay estimation algorithm in the field under different SNR environments,and the superiority of the proposed algorithm is preliminarily verified.2.Based on the related theory of pseudo-random sequence,according to the special environment of acoustic temperature measurement for power station boilers,several sound source signals with obvious autocorrelation characteristics and excellent anti-noise performance are designed,and related parameters are optimized.Through MATLAB simulation experiments,the autocorrelation characteristics and anti-noise performance of the above-mentioned sound source signal and the sweep signal widely used in the field are compared,and the potential advantages of the m-sequences,the filtered m-sequences,and the BPSK signals that are designed in this paper are preliminarily verified.3.The above time delay estimation algorithms and sound source signals were further optimized and screened through laboratory experiments of cold state and thermal experiments in an on-site acoustic temperature measurement system.The experimental results verify the superiority of FOC-ETDGE algorithm and BPSK signal proposed in this paper under the stable working conditions and variable working conditions of the boiler,which lays a solid foundation for further mature application of acoustic temperature measurement system for power station boilers. |
PDF Full Text Request