Research On Linear Active Disturbance Rejection Control Method And Pressurized Water Reactor Power Control

With the increase in proportion of the use of domestic nuclear power and new energy access,in the case of insufficient power consumption,when the load fluctuates,in areas with limited high-quality peaking power sources,if the nuclear power plants do not participate in the peaking,the system will face severe peaking pressure.The vast majority of nuclear power plants in China use pressurized water reactor,which is a nonlinear and time-varying controlled object characterized by greater uncertainty and a variety of disturbances.The power control system of the pressurized water reactor is the key system for the peaking operation of the nuclear power plant.However,the existing pressurized water reactor makes it difficult to achieve rapid and smooth regulation at different power levels and fuel consumption.Therefore,exploring the advanced control methods that can achieve good performance efficiently is necessary.Linear active disturbance rejection control,which can handle systems with wide range and complex structural uncertainties,has the advantages of simple structure,fewer setting parameters,strong disturbance rejection ability,and good robustness and has achieved many successful applications.However,for nonlinear and time-varying systems,the quality of conventional LADRC needs to be improved when the working conditions change greatly.In this paper,for the above problems,utilizing nonlinear structure model information of the controlled object of the pressurized water reactor fully,the linear extended state observer and PD control law design of LADRC are improved and the model-assisted LADRC parameter setting method is given.Then,combined with the partial feedback linearization and T-S fuzzy theory,the characteristics of nonlinearity and parameter time-varying of the controlled object of the pressurized water reactor are solved systematically,which poses a challenge to LADRC design.The main work in this paper is as follows:1.Given the power control problem of the pressurized water reactor,the design and parameter setting method of the model-assisted LADRC for the power control of the pressurized water reactor are established.The parameter correspondence relationship between the model-assisted LADRC and PID is analyzed from the angle of the equivalent block diagram and controller gain approximation.The advantages of model-assisted LADRC over PID control in tracking,disturbance rejection,measurement noise suppression,and robustness are verified by typical controlled objects.2.For the dynamic characteristics of the pressurized water reactor and the measurable characteristic of the power output,LADRC based on the reduced-order extended state observer is designed.Frequency domain analysis shows that under the same bandwidth,the observation speed can be accelerated,phase margin can be improved,and the parameter setting is easier.First,the relative power second-order model with "total disturbance" item for controller design is derived from the nonlinear model of the pressurized water reactor.Then,in combined with the variable substitution method in the controller design,the derivative term of power output is avoided as the input of RESO.The parameter relationship between RESO-based LADRC and PID is discussed further.Finally,the nonlinear and time-varying model of the pressurized water reactor is used to verify the performance of the designed controller.The simulation results show that RESO has good estimation ability for all kinds of uncertainties and disturbances and that the RESO-based LADRC has a faster response and stronger disturbance rejection than LQG/LTR.3.LADRC design method based on partial feedback linearization theory is proposed for the nonlinear characteristics of PWR.First,using the PFL theory,the linear second-order model with the "total disturbance" term designed for LADRC is derived from the nonlinear model of PWR.Then,for this model,LADRC with model information is designed,and the stability of the closed-loop system is proved based on Barbalat lemma.Finally,the simulation analysis shows the proposed PEL-LADRC composite controller has good tracking,disturbance rejection,and robustness at all power levels.4.For the limitations of the existing PWR T-S fuzzy model construction method,the modeling method of T-S fuzzy machine model and the parameter identification method of T-S fuzzy model based on operating data are proposed,and the design scheme of the feedforward-LADRC composite controller of T-S fuzzy model is given,which systematically solves the challenges of nonlinearity,time-varying parameters,coupling effect of the coolant temperature and reactor power level,and external disturbances of the controlled objects of PWR brought to LADRC design.The simulation results show that the feedforward-LADRC based on the fuzzy model with relative power variation and rod speed as prerequisite variables has good control performance under the whole operating conditions and has low LESO bandwidth for T-S fuzzy models derived from the mechanic model and that identified from the operating data.