Distributed system control method of mode prediction control applying multilayer steady-state target computation

The invention relates to a distributed system control method of mode prediction control applying multilayer steady-state target computation. Steady-state target computation is performed by using an upper steady-state target computation (USSTC) module, an internal steady-state target computation module in each distributed subsystem is used for computing a steady-state optimization value of the subsystem, after ach internal steady-state target computation is accomplished, a dynamic control module in each subsystem is used for performing actual control. The problems of large calculated quantity and controlled volume coupling in the distributed system control by the model prediction control can be solved, the inaccuracy caused by performing prediction control by directly taking the steady-state optimization value obtained by single steady-state computation module as a dynamic control module setting value and ignoring the influence on the current subsystem by other subsystems, is reduced, and the influence on the