Frequency Controlof An Islanded Microgrid Using Model Predictive Control (MPC)

Abstract

The smart grid technology is going through a transformation process characterized by complexity and variety regarding its operation and controlling with respect to frequency response and generation. So for maintaining and stabilizing the frequency of the grid a technique and methodology is need that best stabilize the frequency in least and smoother way. So an innovation is made to a MPC model that best suit for these scenario for regulation of the grid frequency in islanded mode. The innovation is that MPC is upgraded to FOMPC that work on the finite sets instead of infinite set results in stabilization and accuracy in least time, requires the less process time and control the shoot in all aspects .In this model cost function is also improved in order for the betterment of the model .Wind Energy system are key devices that can holdup LFC in operation. This thesis is concerned with the development of model predictive load frequency control (MPLFC) strategies incorporating Wind turbine systems. A model consisting of wind turbine, its electrical generators which includes all dynamic subsystem .Here the frequency regulation control methodology depend on some turbine parameters that correlate with the generation load and frequency relation. This relation is best defined by maintaining some parameter like wind speed, pitch angle and generated power. For independent smooth controller design decentralized control is often required. So domestic mpc controllers are fromed for the Wind turbine and its frequency regulation for stabilizing of its grid. However, this requires advance prediction model, because power generator is derived from complex interconnected components. In addition, the design process should have stability and smoothness guarantees despite the dynamics ignored in the prediction models without compromising control performance. A multi time scale. MPC technique is developed where the challenge for model predictive control in multi time scale systems is selecting a suitable sampling time that ensures dynamics responses at each time state. Hence, multi rate sampling is employed to account for 2 timescales pattern. In this thesis different states and conditions are applied on the system considering the different condition that may be dependent and independent of the system behavior. Simulation results are carried out for observing the response of the FO-MPC with respect to the time, load and time horizon .Simulations show that the generation and load frequency control objective of controlling the frequency variation to zero is achieved by the proposed methods in less time.