Modeling And Intelligent Control Method For Furnace Temperature In Municipal Solid Waste Incineration Process

Municipal solid waste incineration(MSWI)technology has been widely used in the world with its powerful advantages of harmless,reduction and resource-based solid waste treatment.The effective control of furnace temperature is the key to improve MSW treatment efficiency,inhibit pollutant emission and ensure the safe and stable operation of MSWI process process.The fluctuation of furnace temperature will cause the change of burn out rate,which will lead to excessive ash burn-off rate and affect the steam output and reduce the economy of incinerator operation.However,it is necessary to detect the furnace temperature in real time to realize the accurate control of furnace temperature.Although the existing thermocouple detection method can detect and display the furnace temperature trend,the occurrence of furnace coking during incineration will damage the performance of thermocouple and lead to detection failure.Moreover,due to the high temperature and serious corrosion in the furnace,thermocouples and other devices are easy to be damaged and need to be replaced frequently,which reduces the economy of incineration operation.In addition,the MSWI process is a complex combustion reaction process with strong nonlinearity.At present,the furnace temperature control in the process of urban solid waste incineration in my country is still dominated by manual control,which is subject to subjectivity and uncertainty,and it is difficult to achieve accurate furnace temperature control.Meanwhile,the furnace temperature is easily affected by the solid waste composition and climate,and presents a certain periodic change with seasonal changes,which increases the difficulty of furnace temperature control.Therefore,the control accuracy,adaptability and stability of furnace temperature are poor,which is difficult to ensure the treatment effect of MSW.In addition,the mechanism of MSWI in China is complex,and it is difficult to establish the mechanism model of furnace temperature.Moreover,the MSWI process has the characteristics of strong non-stationarity,uncertainty and disturbance diversity.Furnace temperature control has become the main bottleneck for the operation and development of solid waste incineration plants in China.Therefore,based on the in-depth analysis of the mechanism and characteristics of MSWI process,this subject plans to study from the following contents:(1)From the perspective of data-driven,a furnace temperature prediction model based on self-organizing TS fuzzy neural network is established to solve the problem that the furnace temperature is difficult to detect in real time.(2)The data-driven TS fuzzy neural network furnace temperature model of MSWI process is established to solve the problem that it is difficult to establish the furnace temperature mechanism model,which lays a good foundation for furnace temperature control.(3)The furnace temperature control method based on ET-RBF-PID is designed to solve the problems of low control accuracy,difficult on-line adjustment of parameters and frequent controller updates in the process of furnace temperature control,so as to ensure the stability in the control process.(4)The event-triggered maximum correntropy self-organizing fuzzy neural network(ET-CSTFNN)furnace temperature control method is designed to solve the problems of low furnace temperature control accuracy caused by uncertainty in the incineration process,and the controller structure is difficult to adapt to the changes of the incineration process and frequent controller updates.(5)The anti-interference control method of furnace temperature is designed to solve the problem of accurate control of furnace temperature under interference conditions and further improve the control accuracy.The work and innovation of thesis paper are as follows:(1)Prediction of furnace temperature in MSWI process based on SOTSFNN-IGAAiming at the furnace temperature prediction problem,a self-organizing TS fuzzy neural network algorithm based on improved gradient descent algorithm(SOTSFNN-IGA)is proposed.Firstly,in order to obtain a suitable network structure and efficient computing ability,the fuzzy rules of the SOTSFNN-IGA algorithm are automatically grown and deleted using the error criterion and activity intensity.Secondly,the SOTSFNN-IGA network parameters are updated using an improved gradient descent algorithm.Meanwhile,Lyapunov theory is used to analyze the convergence of SOTSFNN-IGA algorithm.Then,to understand the influence of each variable on the furnace temperature,a new variable importance measure is used for importance analysis.Finally,the SOTSFNN-IGA algorithm is verified by experiments based on nonlinear dynamic system modeling,Lorentz time series prediction and furnace temperature prediction.Experimental results show that the proposed SOTSFNN-IGA algorithm has high prediction accuracy,which verifies the effectiveness of the algorithm.(2)Research on furnace temperature modeling in urban solid waste incineration processTo address the problem that furnace temperature is difficult to establish the mechanism model,a data-driven TS fuzzy neural network-based furnace temperature model for MSWI process is proposed.Firstly,combined with the combustion characteristics of furnace temperature and field expert knowledge,the important operating variables affecting furnace temperature are excavated as the input variables of neural network modeling.Secondly,the data-driven furnace temperature model of MSWI process is established by using the nonlinear mapping ability and learning ability of TS fuzzy neural network.Meanwhile,the gradient descent algorithm is used to adjust the network parameters online,and the network is trained through the learning algorithm to ensure the accuracy of the furnace temperature model.Finally,the experiment is carried out with the field actual data.The simulation results show that the output data and field operation data can fit well with high modeling accuracy,which proves the effectiveness of TS fuzzy neural network modeling and avoids the requirements for the complex mechanism model of furnace temperature in the MSWI process,and lays a good foundation for the subsequent furnace temperature control.(3)ET-RBF-PID-based control study for furnace temperature of MSWI processTo solve the problem of low accuracy and frequent controller updates in the furnace temperature control of MSWI process,a novel approach of furnace temperature control method based on the event-triggered mechanism RBF-PID(ET-RBF-PID)is proposed.Firstly,the RBF-PID controller is constructed,the network parameters are updated by gradient descent algorithm and recursive least squares algorithm to further improve the network convergence and ensure the real-time performance of parameter update.Meanwhile,the square of momentum factor and the momentum term of parameters are introduced to update the controller parameters,to realize the on-line effective adjustment of controller parameters,ensure the real-time and stability of ET-RBF-PID control parameters,and improve the control accuracy.Secondly,a fixed threshold event-triggered condition is designed as the controller update condition to reduce the mechanical wear caused by frequent update of the controller and reduce energy consumption.Finally,the furnace temperature control simulation experiments are conducted based on the actual data of a MSWI plant in Beijing.The experimental results show that the designed ET-RBF-PID controller can accurately track the furnace temperature and has high control accuracy and stability.Compared with the traditional time-triggered PID controller and RBF-PID controller,the ET-RBF-PID controller can accurately track the furnace temperature,and reduce the number of controller updates significantly while achieving accurate furnace temperature control,and alleviate the mechanical wear and energy loss caused by frequent controller updates.(4)ET-CSTFNN-based control study for furnace temperature of MSWI processDue to the large uncertainty in the municipal solid waste incineration(MSWI)process,the furnace temperature of the MSWI process is difficult to control and the controller is updated frequently.To improve the accuracy and reduce the number of controller updates,a novel event-triggered control method based correntropy self-organizing TS fuzzy neural network(ET-CSTSFNN)is proposed.Firstly,the neurons of the rule layer are grown or pruned adaptively based on activation intensity and control error to meet the dynamic change of the actual operating condition.Meanwhile,the performance index is designed based on the correntropy of tracking errors,and the parameters of the controller are adjusted by gradient descent algorithm.Secondly,a fixed threshold event-triggered condition is designed to determine whether the current controller is updated or not.The stability of the control system is proved based on the Lyapunov stability theory.Finally,the furnace temperature control experiments are conducted based on the actual data of a municipal solid waste incineration plant in Beijing.The experimental results show that the proposed ET-CSTSFNN controller shows a better control performance,which can reduce the number of the controller update significantly while achieving accurate furnace temperature control compared with other traditional control methods.(5)Research on anti-disturbance control of furnace temperature in MSWI processDue to there are various external disturbances in the MSWI process,in order to realize the accurate control of furnace temperature under disturbance conditions,a model-free sliding mode controller is introduced on the basis of CSTSFNN controller.Thus a furnace temperature hybrid controller is designed to solve the problem of poor immunity in incineration process.Firstly,a compact form model-free adaptive control method is used to realize the dynamic linearization of the system.At the same time,the discrete sliding mode convergence control strategy with exponential convergence law is designed.Secondly,the weights of the hybrid controller are designed,and the current comprehensive control quantity is calculated to further improve the robustness and anti-interference of the hybrid controller.Thirdly,event-triggered conditions based on performance indicators are designed to reduce the number of controller updates under disturbance conditions.Finally,the hybrid controller is applied to furnace temperature control for experimental verification and compared with PID controller,RBF-PID controller and MFASMC controller.The results show that the proposed hybrid controller can accurately track the furnace temperature with good control performance and anti-interference ability.The simulation results verify the effectiveness of the controller.