Research On Adaptive Optimal Control Of Supercritical Coal-Fired Units In Deep Peak Shaving Conditions

With the increase of the proportion of clean energy in the power grid,in order to ensure the security and stability of the power grid,higher requirements are put forward for the flexibility,peak load regulation and frequency regulation of thermal power units.However,the controlled object of the coordinated system of supercritical units has the characteristics of large inertia,large delay,nonlinearity and time-varying.When the unit changes load in a wide range,the dynamic characteristics of the controlled object of the coordinated control system change significantly,and it is difficult for the traditional coordinated control strategy to perform efficient effective variable load control.In this paper,the nonlinear modeling scheme and adaptive optimization control strategy of unit coordination system are studied for supercritical units in thermal power plants.The main research results include:(1)A Levy Flights Beetle Swarm Optimization(LFBSO)algorithm is proposed.By introducing Levy Flights Non-Gaussian random process and adaptive step size control strategy,the search rule of BAS algorithm is improved.The group location update strategy of PSO algorithm is used for reference,combined with historical best solutions to search near the best individual.The benchmark test function is selected to test the performance of different optimization algorithms,and the effectiveness and superiority of the proposed algorithm are verified by comparative experiments.(2)The nonlinear control model of the controlled object in the coordination system of supercritical once-through boiler unit is established.The grey box method combining mechanism modeling with parameter identification based on field test data is adopted.The nonlinear control model structure is determined by mechanism analysis of dynamic characteristics of coordinated control system.Based on field test data,the model parameters are fitted and identified.And the nonlinear global model applicable to the full-working conditions of the coordinated controlled object of supercritical units is obtained.The step response simulation tests are carried out on the three load points of high,medium and low load,and the results are compared with the actual test data.It is verified that the model has high accuracy and is suitable for the design of coordinated control system.(3)An adaptive optimization control strategy for coordinated system based on fuzzy gain scheduling predictive control is proposed.The nonlinear control model of coordinated system is linearized locally at several working conditions.The global adaptive fuzzy model is established by combining gain scheduling and fuzzy logic reasoning.The predictive control technology is used to ensure the global optimization quality of the control system,and the problem of constraint handling is solved.The variable load simulation tests are carried out in the full range of normal working conditions and deep peak shaving respectively.The results show that the adaptive optimization control strategy proposed in this paper can quickly track the load changes in the large working conditions,and the coordinated control system has fast response,small overshoot,good stability,good adaptability and robustness.It is suitable for the coordinated control of thermal power units with frequent changes in working conditions.