Research On Intelligent Coordinated Control Of CEFR OTSG

Chinese Experimental Fast Reactor (CEFR) is Chinese first pool-type fast neutron experimental reactor .It is cooled by sodium and mainly composed of three loops.Once-through steam generator is the heat transfer equipment between water and sodium in the second and third loop. The control of pressure and temperature of once-through steam generator outlet steam influences steam's quality and safety. During actual control,operators manual control and automatic control difficult to meet the demand of changed power. So it is significant to conduct thorough research on the control problem of once-through steam generator.In this paper, the modeling of OTSG and related systems using JTopmeret graphical modeling software,and after successful debugging,the OTSG and related systems are jointly debugged with core and first?second loop systems, comprise the CEFR simulation system,and design the OTSG coordinated control system on this basis. The program mainly targets feed-regulating valve?feed water pump and turbine pressure controlled valve, and applying the PID in the coordinate control system, it will compare with intelligent coordinated controller which will come forward in the following text.On the basis of the control system, this paper introduces the idea of intelligent control method. The choices of this paper are fuzzy neural network and artificial immune system.Fuzzy Neural Network (FNN) combines fuzzy control with neural network and plays both the advantages of fuzzy control but also the specialty of nonlinear tracking of neural network. So this paper studies the control method that combines neural network with fuzzy control and designs water flow controller of once-through steam generator that is based on BP fuzzy neural network.Artificial immune system is a new control system which is developed from simulating biological immune system. The basic immune response controller is easy to implement, but it doesn't have learning ability and it is difficult to adjust control parameters manually. So this paper introduces the learning algorithm of immune control. Combined with error back propagation algorithm and Hill function, this paper designs a self-learning immune controller.And it is applied to intelligent coordination controller of OTSGSimulation experiment on the designed OTSG intelligent coordinated control system under different operating conditions, and contrastive analysis of the simulation result is given.The simulations reveal that the designed coordinated program can satisfy the demand of CEFR. Results show that the latter intelligent methods are better than the former PID method.Compared with AIS control, the FNN control has obvious advantage which has small overshoot and short stabilization period in the dynamic process. The paper will have some useful reference to CEFR control system.