Research On Ash Fouling Monitoring And Soot Blowing Fuzzy Control Of Boiler Heating Surfaces

With the rapid development of the Internet and big data technology,the intelligent operation optimization of boilers has become a hot topic for smart power plant.Among them,the research on ash slagging in boiler heating surfaces is an important foundation for ensuring the safe and economic operation of units.Ash and slagging on the heating surface will reduce the boiler’s operating efficiency,increase the exhaust temperature,and even corrode the pipe wall in severe cases,causing pipe explosion accidents.Based on the actual problems of ash and slagging during the operation of domestic cooperative power plants,aiming at the serious waste of steam and tube wall wear in the manual fixed soot-blowing operation mode,this paper firstly conducted simulation research on the ash fouling and slagging in the combustion furnace.The effect of coal powder size on the ash slagging in the heating surfaces of the furnace was analyzed.Then the ash thermal resistance monitoring mechanism model of the low-temperature superheater based on the heat balance principle was established,and on this basis,the ash thermal resistance online prediction agent model based on wavelet analysis and SVR was proposed.Finally,the soot-blowing strategy was formulated using a fuzzy control system based on the ash thermal resistance online prediction agent model.The simulation results showed that the boiler heating surfaces ash monitoring and sootblowing fuzzy control achieved good validation.The specific research contents of this paper are as follows:(1)Based on the CFD-DPM method,a simulation model of the ash fouling in the ultra-supercritical boiler furnace was constructed to quantify the regularity of the ash slagging characteristics of the heating surfaces under different pulverized coal particle sizes,and studying the particle movement trajectory,the number of particles deposited on each heating surface,the outlet flue gas temperature of furnace and the maximum combustion temperature of pulverized coal with different particle sizes.The optimal pulverized coal combustion particle size range of the 1000 MW ultrasupercritical boiler was found to be around 50 um.(2)According to the heat balance principle of the flue gas side and the steam side and the heat transfer model of the multi-layer circular tube wall,a monitoring mechanism model of the convection heating surface with the ash thermal resistance as the cleaning index was established,which was validated through the operation data of the power plant.The visualization results exposed the shortcomings of the fixed sootblowing cycle mode.(3)Combining the advantages of wavelet threshold denoising algorithm and SVR method,an ash thermal resistance online prediction agent model based on wavelet analysis and support vector regression was proposed to realize the prediction of the low-temperature superheater’s ash fouling behavior.The Visu Shrink soft threshold denoising method was performed to denoising the thermal resistance data of the heating surface.The results showed that the four-layer wavelet decomposition has the best denoising performance,and the corresponding SNR and RMSE are 31.8 and 0.000865,respectively.At the same time,the mapping relationship between the 20 input features and the ash thermal resistance after denoising was established using the SVR model.The simulation results showed that the prediction accuracy of the SVR model is above 98%.(4)Comprehensively considering the three influencing factors of ash thermal resistance,main steam flow,and exhaust temperature,a soot-blowing control model based on Mamdani fuzzy control rules was established.The soot blowing scheme of the fuzzy control simulation model was verified in MATLAB/Simulink,and the test results showed that it could give accurate soot-blowing operation suggestions.