Research On Steam Generator Water Level Modeling And Fractional Order Control Algorithms

The natural circulation steam generator(NCSG)is an important heat transfer equipment for current PWR nuclear power plants(NPP)to connect the primary loop and the secondary loop.SG water level is a key factor affecting nuclear safety.The safety and reliability of the NPP operation can be ensured only by controlling the water level of the steam generator in the allowable range.E-Irving model is widely used in the research of water level control of SG.The model has some applicability because it makes studying the control algorithm more conveniently.However,it can't rounded reflect the operation state of the SG of the NPP for it does not contain the thermal hydraulic process.In order to better control the water level of SG,we can use the operation mechanism of SG to model.However,the complexity of the whole control system increases exponentially at the same time although it is more accurate.In the actual operation of SG,the water level is usually disturbed by some factors.The bubble volume at the bottom of the water level varies with local pressure under disturbance and then the phenomenon of"false water level"appears.The appearance of“false water level”makes the traditional control method fail to achieve good control effect under some cases.Therefore,the dynamic modeling of SG water level and the advanced and effective control methods are the key points of this paper.According to the Linear Parameter Varying(LPV)theory,a multi cell LPV water level model of SG based on E-Irving model is proposed in this paper.According to the parameters of the SG in Dayawan NPP,the four orders dynamic water level mechanism model of NCSG is established based on the lumped parameter method which can't only reflect the certain thermal characteristics but also be easy to design the control system.The positive step of steam flow and the positive step of heat transfer are carried out under the conditions of 100%,70%,50%and 30%.The dynamic simulation analysis of the four power level conditions of the simulation model is also carried out to verify the validity of the model.The simulation results show that the main parameters of the steam generator show a reasonable change trend,and the"false water level"phenomenon is accurately simulated when disturbance occurs.The stability of the fractional order PI~?D~?control system is analyzed.On the basis of the E-Irving multi cell LPV water level model,the fractional operator based on Grunwald-Letnikov(G-L)is added to the main-loop,vice-loop and main vice loop of the SG water level control system respectively.The short memory fractional order PI~?D~?controllers are also designed.A cascade fractional order level controller based on the improved algorithm of Oustaloup indirect discretization method is designed.As the tuning of controller parameters is still a problem at this stage,this paper uses fuzzy adaptive and chaotic particle swarm two different types of intelligent control algorithms to automatically adjust the parameters of the controller in order to realize the automatic tuning of the parameters of the water level controller of SG.The simulation results show that the designed fractional order controller and the intelligent parameter tuning algorithm are all effective.The optimal control effect is achieved by the improved Oustaloup indirect discretization of cascade water level controller among them.Although both of the two intelligent optimization methods have achieved good results,the chaotic particle swarm optimization algorithm is not dependent on the experience or the rule and is more suitable for engineering model.According to the obtained results,the improved Oustaloup fractional order water level control system based on chaotic particle swarm is designed for the established dynamic mechanism model of SG water level.The simulation results show that the intelligent control system can reduce the"false water level"effectively and realize the automatic control of the water level for the NCSG on-line.