The research of the explicit time delay and gain estimation algorithm based on fourth-order cumulants in acoustic pyrometry in the power plant boiler

摘要

In order to overcome the difficulty of estimating the accurate time delay in the furnace with low and time-varying SNR, the FOC-ETDGE algorithm is studied. The algorithm separates the adaptive processes of SNR and time delay, decomposes the single adaptive filter in the traditional model into the two sub-adaptive units, one of which is used to track the time delay estimate, and the other is used to adapt to the variation of SNR, and eventually converge to the delay true value through the adaptive iteration. The method is compared with the conventional cross correlation function method by using the data collected from the acoustic pyrometry system which is put into operation in a domestic owned power plant boiler. The results show that the algorithm is superior to the time delay estimation in the acoustic pyrometry in the power plant boiler.