Research On Active Disturbance Rejection Control And Its Application In Thermal Process Of Power Plant

Exploring more efficient modeling method and high performance robust control algorithm has always been a hot and difficult point in the research of automatic control of thermal process for coal-fired power units that have the characteristics of high-order inertia,time delay,non-linearity,multi-source disturbances,loop coupling and uncertainties.Especially in order to balance the system fluctuations caused by the high proportion of intermittent new energy power connected to the grid,thermal power units currently need to adopt deep peak regulation technical measures,which makes the reliability and robustness of the automatic control system of thermal process face more severe challenges.In addition,a large amount of process data due to disturbances or uncertainties are stored in the history database of DCS.It can be considered to make full use of these process disturbance data,and then enhance the model information of the controller to improve the control performance of the control system,such as set point tracking,disturbance rejection and robustness.Therefore,this paper takes linear extended state observer as the main line,and carries out theoretical research,numerical simulation and engineering implementation of phase analysis of generalized integral series system,disturbance response data-driven intelligent identification method of extended state observer model parameters,and design of reduced order ADRC based on phase compensation.The main work of the thesis includes:1)The characteristics of transfer function of the generalized integral series between the virtual control variable and the system output and the estimated output are analyzed in detail in the frequency domain after the total disturbance is estimated and the feedforward compensated by linear extended state observer.When the full-order extended state observer is used,the influencing factors of the generalized integral series approximation to the standard integrator series are simulated and analyzed;when the low order extended state observer is used,in order to ensure the approximate synchronization of the.generalized extended state observer and the generalized controlled plant in the phase,the design method of adding part of the model information to the extended state observer for phase compensation is proposed.The simulation of the example verifies the effectiveness of this method.2)Aiming at the problem that the accuracy of closed-loop data-driven modeling is reduced due to the existence of external disturbance in the output signal under zero initial condition,a new method of parameter identification of process data-driven extended state observer is proposed,which uses the characteristic of dynamic transition to steady state caused by disturbance in the control loop.That is,the initial state value estimation at the end point of the disturbance action in the process data is combined with the deterministic model information estimation in the total disturbance,and the model parameters are optimized and clustered through the swarm intelligence algorithm to obtain the best identification parameters.Numerical example simulation verifies the effectiveness and accuracy of the proposed modeling method.3)In order to improve the set point tracking performance and disturbance rejection ability for thermal process plant with the characteristics of large inertia and time delay,a reduced-order active disturbance rejection control design method based on phase compensation is proposed,and stability analysis is completed.Considering the problem that it is difficult to determine the gain of control variables when using low order ADRC,a new parameter tuning method is proposed.For the multivariable system of thermal process,the phase compensation-based decentralized reduced-order active disturbance rejection control strategy is adopted,and the control system is transformed equivalently in the frequency domain to reveal the inverse decoupler essence under the framework of active disturbance rejection control technology.At the same time,in order to enhance its reverse decoupling ability,a design method of phase compensation for multivariable system is derived.Simulation results show the superiority of the proposed control algorithm.4)The specific problems in engineering design such as SAMA configuration of reduced order ADRC algorithm based on phase compensation,anti-windup of reduced order extended state observer and bumpless transfer are studied and solved,and then simulation and implementation of proposed control strategy on the excitation simulator is carried out.And then it is applied to the main steam temperature system and turbine-boiler load system of thermal power units in service.The implementation results show that the improved ADRC algorithm is feasible,effective and superior,and also shows the good industrial application prospects of the algorithm.