A Extraction Method Of Combustion Feature Signal Based On Furnace Pressure Analysis

Influenced by the unit involved in the primary frequency modulation and the coal quality variation, the furnace pressure of the large generating units is volatile,while a large number of experience of the operation shows that many faults will cause the furnace pressure changes, and furnace pressure is one of the important features of the signal reflecting the state of combustion. Utilizing mechanism analysis and statistical analysis methods, a non-linear dynamic model of furnace pressure, fuel and air quantity was constructed. The transfer function of combustion disturbance versus furnace pressure was obtained after model linearization under different load conditions, which were analyzed by means of frequency domain characteristics. The results shows that the frequency domain characteristics of the model have band-pass filter characteristics, for the major frequency of combustion disturbance is within the pass band of the object, so it is difficult for the control system to effectively conquer the combustion disturbance. It is find that the furnace pressure is very correlative with the total fuel amount, through the on-site signal data analysis . The furnace pressure signal can be used for combustion optimization control and fault diagnosis.