Research On Steam Temperature Control Of 1000MW Double Reheat Ultra-supercritical Units

While ultra-supercritical double reheat unit has the advantages of high efficiency and energy saving, the requirement of its control accuracy is becoming stricter and stricter because of its high operating parameters. In order to ensure the safe and efficient operation of the unit, it is very important to control the main steam temperature, the first reheat steam temperature and the second reheat steam temperature.In this thesis, based on the analysis of existing domestic and foreign control methods of double reheat steam temperature, combining the characteristics and control scheme of 1000 MW ultra-supercritical double reheat unit in Taizhou, in view of the disadvantages of the traditional PID controller, such as poor regulatory performance caused by serious internal coupling of the controlled object and poor control effect caused by imprecise transfer function etc., the author designed a control system which can make the main steam temperature, the first reheat steam temperature and the second reheat steam temperature meet the control requirements while using the PID neural network decoupling control method. Since the control algorithm has the function of online learning to adjust parameters, there is no need to get the exact model of transfer function, it can offset the effects on the controlled object come from the internal model or other disturbance, and ensure the steam temperature meet the control requirements more quickly and accurately.This thesis analyses the influencing factors on the steam temperature, and records data of the step disturbance experiments on the unit simulation, gets the model of steam temperature system by PSO identification algorithm, and then builds the PIDNN control system on the Matlab/simulink platform to make the simulation experiment and compares with the effect of traditional PID control at last. The simulation results show that the PIDNN control algorithm has good control effect on this kind of multi-variable system which has large inertia and serious coupling. The PIDNN has simple structure, and its parameters are adjusted as the foundation of the existing PID parameters, so it is easier to be realized in DCS. The thesis provides a new solution for the ultra-supercritical double reheat steam temperature control of the engineering application.