Distributed Predictive Load Frequency Control For Deregulated Power System

With the rapid development of economy and society,the scale of electric power system is becoming larger than ever brfore.Since entering the new century,thanks to the fast-developing smart grid technology and internet technology,multi-area interconnection and deregulation are the new features of the power system.The past signal-area grids have been connected by the tie-lines,different grids exchange the control information through tie-line to realize new forms of networking operation.Meanwhile,the power supply of many regions are at the expense of the burning of fossil fuels and the damage of environment.Many counties are learning to change the existing power structures.Compare with fossil fuels,the renewable and clean energy represented by wind power has attracted wide attention,owing to its preponderance,such as: clean,huge reserves,technologyis relatively mature and so on.In the meantime,the rapidly development of the wind power brings new technical challenges to the power system.More and more electric power companies request the wind power to supply ancillary service like traditional thermal power generation.This paper studies the new control strategy of the load frequency control on the basis of a three-area interconnected power system,which with and without wind power generation system participation based on the state space model.The DISCOs Participate Matrix(DPM)and Area Participation Matrix(APM)which react the pattern of power market have been taken into account.The distributed model predictive control(DMPC)is designed and applied to that power system.Meanwhile,nonlinearities constraints containing the valve position limit,generator rate constraint(GRC)and the control input constraint of the wind turbine are referred as well.The distributed model predictive algorithm is applied to the deregulated power system to research on the balance of the active power and the fluctuation of the frequency.Centralized model predictive control(cent-MPC),decentralized model predictive control(decent-MPC)are also designed for comparison.The simulation results for a deregulated three-area interconnected power system show possible improvements on closed-loop performance,while respecting the physical hard constraints.