Data-Driven Model Predictive Control for Load-Frequency Control in Islanded Microgrids

One of the basic issues of microgrids is frequency and power control in a separate state from the power grid. In this article, the data-based predictive control method is presented for the first time in a microgrid in order to control the load frequency. By applying the control signal to the sources and positioning the proposed controller in the secondary loop of frequency control, frequency disturbances are detected after microgrid power changes. In fact, the designed controller first provides a model of the system based on the Subspace System Identification method, and then, based on the obtained model, based on the appropriate dynamic logic, it directs the control system in alignment with the goals of the system. The proposed method was implemented for the first time in a standard microgrid without considering the dynamic model associated with the targeted system, and in addition to achieving the control goals, i.e., load frequency control, it was also able to repel disturbances well.