| Load Frequency Control(LFC)plays a key role in the operation and design of modern electric power systems.Its major functions are to match the total generation with the total load demand and to maintain the frequency at the scheduled value.Given the implementation of global energy strategy,large amount of wind power have increased gradually.Due to the intermittency and randomness of wind power,the active power balance of power system is greatly affected by the wind power.To ease the transition from the conventional power systems to modern complex power systems with large-scale wind power,the frequency dnamics and stability will be studied on the power system with wind farm.Based on research on the danamic frequency characteristics of interconnected power system with wind farm,load frequency control of power system are disscussed in theis study with the distributed model predictive control(DMPC)algorithm and mumerical simulation.The main research work and innovative results of this thesis are as follows.(1)According to the scale of the wind farm,two different LFC model of power system with wind farm are discussed in this paper.The first one is the wind turbine and conventional power plant placed in a single area,which is connected by tie-line with power plants in another area.Based on the model of dynamical components such as wind turbine model,generator model,a generalized representation of LFC is established.In this structure,wind turbine is ahieve to the maxmiun power output through regulating pitch angle,the conventional power plant will tracking the load change.(2)The second one is large-scale wind power plant placed in a single area,which is connected by tie-line with power plants in another area.The LFC model of power system with large-scale wind farm is established in this paper.The wind farm like to the tranditional power plants,tracking the load change simultaneously.(3)Based on the first LFC strucutre of power system,distributed model predictive control(DMPC)scheme is presented in this thesis.The generator rate constrants(GRCs)and the limit of the governor are also considered in the design of control scheme.Comparisons of response to step load change and computational burden have been made between DMPC and centralized MPC.The results confirm the superiority of the proposed DMPC technique.(4)Based on the second LFC strucutre of power system,distributed model predictive control(DMPC)scheme based on coodination is presented in this thesis.Different objective functions in different power plants are defined due to various wind speed.The generator rate constrants(GRCs)and the limit of the governor in conventional power plants and the wind speed constraints in wind farm are considered in the design of control scheme.Comparisons of response to variable wind speed,step load change,computational burden have been made between DMPC and the centralized MPC.The simulation demonstrates the DMPC'S abilities in depicting the stabilized operation of the wind farm contained LFC.(5)Considering internal uncertainties and external disturbances due to changes in the system parameters,large power load changes and variation in the operating points,a parameter adjustable constrained robust distributed model predictive control(RDMPC)based on linear matrix inequality is implemented in an LFC problem for tranditional interconnected power system and modern interconnected power system with large-scale wind farm.The RDMPC algorithm solves convex local optimization problems to minimize an attractive range of robust performance objectives by using time-varying state-feedback control for each control area.The constraint handling ability of the MPC is used to address the critical LFC nonlinearities,e.g.,the practical limits of the valve position of the governor,the GRCs and wind speed constraints.In particular,with coupled input and state variables between the neighboring control areas,the proposed approach still guarantees the existence of a stable Nash equilibrium and the identification of the Pareto-optimal solution. |
PDF Full Text Request