In the automatic alarm system for leakage of boiler tubes based on acoustic signals

to solve the problems of large data size

high transmission bandwidth and large storage occurred during the acquisition

transmission

processing and storage of acoustic signals

an adaptive compressed sensing method for the boiler tube acoustic signals based on wavelet packet is proposed. Firstly

the wavelet packet analysis method is used to decompose the acoustic signals of the boiler tube in multiple scales

and the mathematical expectation of the wavelet packet coefficient under each scale is calculated as the threshold value. The wavelet packet coefficient is zero processed

and the optimal wavelet packet decomposition layer number is selected adaptively according to the sparse degree values under each decomposition scale. Then

each wavelet packet coefficient block under the optimal decomposition layer is classified according to its mathematical expectation and information entropy

and each wavelet packet coefficient block after classification is processed by different methods. The method in this paper is applied to the historical data of the acoustic wave automatic alarm system of a boiler tube

the results show that

the proposed method can effectively reduce the observed data of acoustic signals in furnace tubes

increase the signal compression rate

reduce the transmission bandwidth and storage capacity of signals at the same observation times

and improve the processing speed and reconstruction accuracy of signals.