A Preprocessing Method Based on the Wavelet Threshold De-Noising Technique for the Key Variables of Pulverized-Coal Concentration

The accurate measurement of the pulverized-coal concentration in the powder-transmission pipe of the coal pulverizing system of thermal power plant can effectively reflect the operation efficiency of the system, and the preprocessing of several key variables, that are adopted to represent the pulverized-coal concentration, is an important step. In this paper, we propose a preprocessing method based on the wavelet threshold de-nosing technique for the key variables of pulverized-coal concentration. The method uses wavelet transform on noises firstly. Then according to threshold functions and threshold, we do threshold processing on the wavelet coefficients to get the new wavelet coefficients. Finally we reconstruct signals by the new wavelet coefficients to get the de-noised signals. This method can eliminate the noise and remove the random error with maintaining the characteristic information of the original signals. The results of the simulation verify the effectiveness of the proposed method.