| From the nuclear accident happened in Chernobyl (Russian) and Three Mile Island(US A) several years ago to the grade-7nuclear leakage accident just happened in Fukushima nuclear power plant(Japan) in2011, the world paid more attention on operation safety and stability of nuclear power plants and proposed a higher and more strict requirement of control during its daily running.As a new type of advanced nuclear power plant with the features of Generation IV nuclear energy system, Pebble bed modular high temperature gas-cooled reactor has received much concern by the world’s nuclear power industry due to its inherent security feature and modular design idea. Institute of nuclear and new energy of Tsinghua designed HTR-PM Demonstration nuclear power plant system based on MHGTR and HTR-10. HTR-PM is much more different from the existing pressurized water reactor and boiling water reactor power plant on fuel structure, equipment as well as heat transfer medium because of its particular "Dual Modules with One Turbine" structure mode and the MHTGR. Furthermore, existing control strategy can’t effectively deal with the Complex dynamics and parameter varying feature during its dynamic process and make the nuclear unsafe and unstable. In this paper, several control strategies based on model predictive control method are studied for the control of HTR-PM’s load changing process in order to achieve fast and smooth operation. A dynamic simulation and advanced control simulation platform was built for simulation and validation of the effectiveness of the control strategies of load changing processes of HTR-PM.The main research contents and contributions are as follows:(1) A semi-physical dynamic simulation and advanced control simulation platform of HTR-PM is built in this paper. Then a dynamic simulation of HTR-PM’s mechanical model is implemented using gPROMS. A gPROMS’s OPC communication module is developed to implement the real time communication and closed-loop simulation between simulation platform and a third-part advanced control software. By using HTR-PM simulation platform,we studied the nonlinearity of HTR-PM during load changing process and analyzed existing PID control strategy. at last, a model predictive scheme which can be applicable to the load changing process control. (2) In this paper, a model predictive control strategy of HTR-PM based on operating trajectory LPV model is proposed to solve the nonlinearity and parameter varying of HTR-PM during load changing process. In this strategy, several linear models on different power levels is selected to identify a operating trajectory LPV model in HTR-PM’s operating range according to the operating trajectory of the power load variable. Then, a common linear model predictive control strategy and a model predictive control strategy based on operating trajectory LPV model are simulated by using HTR-PM dynamic simulation and control platform. The simulation results show that the operating trajectory LPV model can accurately describe the nonlinearity of HTR-PM during load changing process and is better than the linear model predictive control strategy.(3) In this paper, a operating space dividing method based on a nonlinearity measure tool of gap metric is proposed to solve the benchmark linear model set determining problem in operating trajectory LPV model identification. This determingation was always determined by experience. This method can divide the operating space automatically to determine the number of subspaces and their corresponding linear models in order to make up the minimum linear model set of operating trajectory LPV model. The simulation results of operating space dividing of HTR-PM load changing process show the effectiveness of this method. |
PDF Full Text Request