Research On Modeling And Optimal Control Of Feed Water System Of Supercritical Unit

With the rapid development of China’s power industry,thermal power generating units plays a main power supply task role.How to ensure a stable,safe and efficient operation of thermal power units has become a key point and main difficulty for the power industry research.The supercritical power generation technology,which is relatively mature and has the advantages of high thermal efficiency,high reliability,energy saving and environmental protection,composite variable pressure operation and good peak shaving performance,is the main development direction of China’s current thermal power generation technology,and gradually becoming the main units of national grid.The feed water control system is a very important subsystem in the supercritical unit control system.Its control task is to use the intermediate point temperature or enthalpy value as the feedback correction signal to ensure that the water-fuel ratio is constant,and then control the steam’ s temperature to meet the requirements for feed water flow under different loads.Although the traditional PID controller has simple structure,good stability,low application cost and wide application in power plants,the complexity and uncertainty of the feed water system control process make it difficult to achieve a satisfactory control effect through PID controller during actual operation.In this paper,the 350 MW supercritical unit feed water system of a power plant is taken as the research object.Because of its characteristics of complexity and variability,large inertia and large delay,the advanced control algorithms,such as fuzzy internal control(Fuzzy-IMC)and dynamic matrix control(DMC),are introduced to the correction circuit of the feed water system,and adjust the controller parameters by applying to the ant colony optimization algorithm.Firstly,the structural characteristics of the supercritical once-through boiler are analyzed,and the process flow and control strategy of the feed water system are introduced.Secondly collect the historical data of the plant operation to preprocess the data and select the appropriate identification data,and use the ant colony optimization algorithm to establish the transfer function model of the feed water system.Then the fuzzy internal model control and dynamic matrix control are combined with PID control.The controllers of fuzzy IMC-PID and DMC-PID cascade feed water control systems are designed respectively,and the controller parameters are optimized.Finally,the MATLAB simulation comparison experiment was carried out with the conventionalPID cascade control.The results show that both of fuzzy IMC-PID cascade control and DMC-PID cascade control have better stability and faster adjustment speed,and have stronger robustness and anti-disturbance capability.