The Thermal Signal Multi-resolution Analysis And Application Based On The Wavelet Transform

Boiler combustion state is good or bad had a direct impact on the economy and security of the thermal power units. In this paper we used SIS database data of the power plant, and focused on the study of the thermal signals multi-resolution correlation. For traditional signal analysis method is a portrait of the signals’overall trend, not able to take into account the local fluctuations of the signal related issues. The wavelet transform does not depend on the mechanism analysis, start the study from the signal volatility information so that we can overcome this problem Therefore, this paper presented the wavelet transform combined with the signal correlation analysis techniques, that is the signal multi-resolution correlation analysis method based on wavelet transform. Signal decomposed by wavelet transform to a different frequency range, and then reconstructed the signal components in different frequency bands. Based on signal fluctuations similarity on the same frequency, we calculated the correlation coefficient between them in order to describe the degree of correlation on the different signal scales. By using multi-resolution analysis method, we study the correlation of multiple signals at different scales to verify the effectiveness of the method. And also take frequency-dependent signal for example, we discussed the selection problem of wavelet basis and decomposition level on wavelet transform. An example analysis based on a power plant show that this method can effectively excavate correlated signals of combustion disturbance, and can distinguish the relative strength. That this method can not only quantitative analysis of the correlation strength of the signal low-frequency trends, but also be able to dig out the high-frequency fluctuations of the signal, in order to expand the scope of the excavation of the thermal signal.