FLAME MONITORING AND OPTIMIZATION OF COAL-FIRED BURNERS BASED ON FREQUENCY SPECTRA ANALYSIS

THIS PROJECT WILL EXPLORE THE DEVELOPMENT OF A NEW METHOD FOR CHARACTERIZING AND OPTIMIZING PERFORMANCE OF COAL BURNERS, BY ANALYZING FREQUENCY SPECTRA OF COAL FLAMES OBTAINED FROM THE EXISTING FLAME SCANNERS. MOST INDUSTRIAL AND UTILITY BOILERS ARE PRESENTLY EQUIPPED WITH FLAME MONITORING DEVICES (FLAME SCANNERS) FOR INDIVIDUAL BURNERS. THE SOLE FUNCTION OF THESE SCANNERS IS TO DETECT THE PRESENCE OF FLAME FOR BURNER MANAGEMENT SYSTEMS. THE INPUT TO A FLAME SCANNER (FLAME SIGNAL) CONSISTS OF TWO COMPONENTS, INTENSITY AND TEMPORAL FLUCTUATIONS;ONE OR BOTH OF THESE COMPONENTS CAN BE USED FOR FLAME DETECTION. OVER THE YEARS, NUMEROUS DATA AND OBSERVATIONS HAVE BEEN ACCUMULATED SUGGESTING THAT THE TEMPORAL FREQUENCY COMPONENT OF THE FLAME SIGNAL (OFTEN CALLED "FLAME SIGNATURE") CAN BE UTILIZED TO DEVELOP A POWERFUL DIAGNOSTIC TOOL TO CHARACTERIZE FLAME PERFORMANCE. THE OVERALL OBJECTIVE OF THIS PROJECT IS TO IMPROVE THE PERFORMANCE OF CONVENTIONAL AND ADVANCED COMBUSTORS BY RELATING FLAME FREQUENCY SIGNATURES OF INDIVIDUAL BURNER FLAMES TO FLAME STABILITY, EFFICIENCY, AND EMISSIONS. THIS INNOVATION WILL ALLOW NOX EMISSIONS AND OPTIMIZE OPERATION OF INDIVIDUAL BURNERS. IN PHASE I, A NEW ALGORITHM WILL BE DEVELOPED AND TESTED FOR PROCESSING FLAME FREQUENCY SPECTRA FROM CONVENTIONAL FLAME SCANNERS FOR COAL-FIRED, LOW-NOX BURNERS. THIS NEW ALGORITHM WILL BE BASED ON COMPUTER PROCESSING OF RELATIVE AMPLITUDES OF TEMPORAL SPECTRA AT DIFFERENT FREQUENCY BANDS. THE ALGORITHM IS EXPECTED TO SHOW STRONG CORRELATION WITH FLAME CHARACTERISTICS AND COULD BE USED TO EVALUATE FLAME STABILITY, FUEL-TO-AIR RATIO, UNBURNED COMBUSTIBLES, AND NOX EMISSIONS. IN PHASE II THE RESULTS WILL BE TESTED ON A LARGE SCALE AT SEVERAL INDUSTRIAL AND UTILITY BOILERS EQUIPPED WITH DIFFERENT TYPES OF BURNERS. THE RESULTS WILL BE COMMERCIALIZED IN THE FORM OF NEW PRODUCTS (SOFTWARE PACKAGES AND NEW FLAME MONITORING SYSTEMS).