Design and analysis of UPQC in a microgrid using model reference adaptive control ensemble with back-stepping controller.

In recent years, the power sector has shifted to decentralized power generation, exemplified by microgrids that combine renewable and traditional power sources. With the introduction of renewable energy resources and distributed generators, novel strategies are required to improve reliability and quality of power (PQ). In our proposed system, a model consisting of photovoltaics, wind energy, and fuel cells has been designed to share a network, bolstered by the integration of UPQC to rectify PQ issues. Notably, our model introduces a Back-stepping controller method featuring Model Reference Adaptive Control (MRAC) with online parameter tuning, offering superior adaptability and responsiveness. This approach not only ensures optimal grid management but also enhances efficiency and stability. Furthermore, the proposed model demands minimal additional infrastructure, leveraging existing resources to streamline implementation and maintenance, thereby promoting sustainability and cost-effectiveness. The research culminates in a comparative analysis between the MRAC-Back-stepping controller, Adaptive Neuro-Fuzzy Inference System (ANFIS), and Fuzzy controller, highlighting the efficacy and versatility of our proposed model in microgrid operations. A Matlab model has been designed along with a hardware setup to demonstrate the robustness of the model.

Performance analysis of DFIG support microgrid using GA optimized restricted Boltzmann Machine algorithm.

Voltage and reactive power regulation in a deregulated microgrid can be achieved by strategically placing the Static Synchronous Compensator (STATCOM) in coordination with other renewable energy sources, thus ensuring high-end stability and independent control. STATCOM plays a crucial role in effectively addressing power quality issues such as voltage fluctuation and reactive power imbalances caused by the intermittent nature of wind energy conversion systems. To successfully integrate STATCOM into the existing system, it is essential that the control system employed for STATCOM coordination aligns with the Doubly-Fed Induction Generator (DFIG) controller within the microgrid. Therefore, an efficient control algorithm is required in the microgrid, capable of coordinating with the DFIG controller while maintaining system stability. The utilization of a Genetic Algorithm (GA) in calibrating the Restricted Boltzmannn Machine (RBM) can streamline the process of determining optimal hyperparameters for specific tasks, eliminating the need for computationally intensive and time-consuming grid searches or manual tuning. This approach is particularly advantageous when dealing with large datasets within short time durations. In this research, a Simulink model comprising a DFIG-based microgrid and STATCOM has been developed to demonstrate the effectiveness of the proposed control system using RBM in managing STATCOM and facilitating microgrid operations.