The Active/Reactive Power Optimization Control That Considering Distributed Wind Farms Interconnection Into The Grid

With the development of wind power generation technology, distributed wind farms interconnected into the grid have a great influence of frequency and voltage stability of power system. Therefore, the optimization of wind farms on power control is particularly important. In this thesis, Modeling the cage type asynchronous motor and doubly-fed wind turbine, and making analysis the influence of the random output of the two kinds of wind turbine interconnected into the grid IEEE 14 nodes system.Then, wind farm is interconnected with steam turbine as a load and the ACE model that considering wind is established. Load and reserve capacity model that considering wind because of that the wind will affect the power generation program of the system is established. Online rolling optimization is the characteristics of predictive control, then predictive control is adopted for controlling the frequency deviation of the two regional systems and the tie line power deviation caused by the load and the wind power fluctuations, and is compared with the PID control system.In order to give full play to the frequency modulation of wind turbines, makes some studies on the control strategy of doubly-fed generator participate in the system frequency modulation, the mode selection controller cooperated with conventional generators is put forward.Wind turbines can choose different operation modes cooperate with conventional units according to the different operation mode of the system and the load demands of system. In this thesis, four kinds of modes are set. If the load disturbance is small, and the frequency of the system change is also very small or the wind motor in low power operation, wind turbines will do not participate in the system’s frequency modulation. If the load disturbance increased at the time when wind turbines have the ability to participate in the system frequency modulation, which can adjust pitch Angle at the same time of the wind turbines in order to obtain greater catching air volume as well as the generator rotor speed to release the energy; When the frequency of the system rises quickly, at this time, a part of the wind turbines should stop running. Wind turbines automatically choose different modes and decide whether to participate in frequency modulation as well as the contribution of the active power can make full use of wind power output and save energy. Simulations for different modes of each unit output curves.Finally, study the wind farm reactive power optimization control.The objective function of the system is the minimum of the active network loss, and the penalty function method is introduced into the objective function to do the correction. In order to solve the problem of random packet loss of networked control, a redundant-state adjustment matrix that obeys the Bernoulli distribution is proposed to simulate random packet loss in the system, and wipe out the effect of unpredictable or uncontrolled quantity’s influence on the system. At the same time, a delay matrix that satisfied the random of Bernoulli distribution is established. The classic genetic algorithm is improved and used to control the optimization calculation. Making simulation and analysis for the wind farm integrated in IEEE-14 nodes system the results show that the dispose of random packet loss can significantly speed up the convergence speed, reduce the network loss.