An Improved Algorithm, for the Measurement of Flame Oscillation Frequency

The oscillation frequency of a combustion flame measured from the spectrum of a flame signal often depends on the level of white noise in the signal as well as the signal sampling rate. In order to solve this problem, a new computing algorithm has been developed by making two improvements to the original direct calculation method. One of the improvements is the wavelet-based pre-filtering of the original flame signal. The other is an adaptive truncation of the spectrum of the filtered signal. It is found that the white noise can be largely cancelled by wavelet filtering. To decrease the smearing effect of the remaining noise a spectrum truncation scheme is utilised. The effectiveness of the improved algorithm is evaluated by conducting a series of experimental tests on an industrial scale combustion test facility. Results obtained are reported and discussed.