Research on distributed intelligent predictive control for flexible load tracking of thermal power units

In order to stabilize the fluctuation of renewable energy power generation in the power grid, thermal power units often need large-scale variable load operation. To reduce parameters fluctuations and guarantee operation safety, this paper proposed a distributed intelligent predictive control strategy for flexible variable load operation of thermal power units. The innovation of the proposed distributed intelligent control strategy mainly includes the following two aspects. On one hand, aiming at describing the nonlinear dynamic characteristics of thermal power units in the process of variable load operation, the strategy constructs a nonlinear predictive control model through data-driven machine learning method. On the other hand, the computational complexity of a nonlinear controller is reduced by using a distributed method in this strategy. It divides the thermal power unit into two parallel computing sub-control systems, which are boiler subsystem and steam turbine subsystem, respectively. Moreover, it realizes the information interaction between the two sub-control systems and achieves the global optimization decision of the thermal power unit based on Nash equilibrium game method. The simulation results show that the proposed control method in this paper can not only ensure the tracking accuracy of large-scale variable load operation, but also reduce the fluctuations of operation parameters and guarantee the safe, economic and stable operation of thermal power units.