Research On Modeling And Control Of Combustion System On Large-Scale Thermal Power Unit

As the energy power source of large-scale thermal power unit,the boiler combustion system directly determines the safety and economy of thermal power unit operation.In order to improve the control quality of the combustion system,we need to establish a more accurate mathematical model based on the understanding its thermal characteristics.Due to the characteristics of combustion system such as multi-parameters,strong coupling,time-varying and nonlinear,it is difficult to recognize the thermal model accurately by the traditional method.For its global optimization performance,versatility,suitable for parallel processing,etc.,Swarm intelligence optimization algorithm helps to get a more reliable model of thermal objects.In this paper,artificial bee colony algorithm is used as a tool to study the modeling and optimization control of 1000 MW ultra-supercritical unit combustion system.The main work is as follows:(1)Using artificial bee colony algorithm to identify the combustion system based on field historical data,After preprocessing the data selected in a certain working condition,using the artificial bee colony algorithm to calculate the fuel quantity and the main steam pressure,the sending / drawing air volume and the furnace negative pressure,the fuel quantity / air volume and the flue gas oxygen The relationship between the amount of identification.The model is validated by different operating data in the same condition,which proves the effectiveness of the proposed method.(2)Using artificial bee colony algorithm to adjust PID controller parametersAccording to the model obtained by identification,the parameters of PID controller are set by artificial bee colony algorithm and expert experience formula respectively,and to observe the control effect after adding some disturbance.Keeping the input signal and disturbance signal unchanged,the model parameters are modified to verify the robustness of the system under uncertain parameters.The simulation results show that the system control quality is better after the controller parameters are optimized by artificial bee colony algorithm.