Control of Hybrid AC-DC Microgrid Using Variable Weighing Factor Algorithm

Abstract

The electricity system is facing a structural transformation due to initiatives such as an increasing penetration of renewable distributed generation (RDG) units, widespread use of different converter topologies for storage, generation & loads, increased utilization of DC generation & load, led to inception of hybrid AC-DC network. This posted significant challenge in terms of its control and optimization. Moreover, Bidirectional Interlinking Converters (BICs) are used for flexible power interaction between AC and DC networks in hybrid AC-DC; both of them thus form a hybrid AC-DC network. Control of the network, especially the control of BIC, must ensure the vital coordination between AC and DC networks to achieve the key objective of appropriate power allocation amongst all the converters. Researchers have shown considerable interest in this promising area due to the non-triviality of the control design. Researchers are using multiple BICs with coordinated control or single BIC with multiple layer Proportional Integral Derivative (PID) control schemes for BIC to act as grid supporting grid forming (GSGFM) or grid-supporting grid feeding (GSGFE) unit. Moreover, fixed roles has been assigned to RDG and BESS based bidirectional DC-DC converter (BDDC). In this thesis, there are two major contributions related to the control of hybrid network: first being is to present Model Predictive Control (MPC) based control for hybrid network. A simple model for the BIC has been used through which it can act as grid supporting GSGFM unit to regulate either AC-DC voltage or GSGFE unit to regulate ACDC power sharing. The efficacy of proposed controller is simulated with realistic considerations and show improved steady state and transient performance with more accurate power allocation amongst all the converters along with detailed comparison with traditional PI based dual-droop control. The second part of our work considers converter-based DC microgrids comprising of RDG and battery energy storage systems (BESS), which are being integrated into power systems infrastructure at rapid pace. For better performance and reliable operations, it is envisioned that RDG, BESS along with loads will form microgrids which can strengthen grid resilience, operate autonomously while the main grid is down, help mitigate grid disturbances as well as function as a grid resource for faster system response and recovery. For autonomous operation power converters will be assigned the role of Grid Supporting Grid Forming (GSGFM) units for voltage regulation and Grid Supporting Grid Feeding (GSGFE) units for current regulation. The architecture of a consensus-based energy management system (EMS) is presented with MPC based variable weighing factor algorithm for power converters of RDG and BESS to act as GSGFM DG or GSGFE DG for both islanded and grid-connected mode. Taking existing control schemes, their voltage regulation capability (GSGFM), current regulation capability (GSGFE) and mode changing capability is discussed to maximize the usage and improve the performance of power converters. The basic design principal of our proposed EMS is to maximize the usage of controllers of all the DG units of microgrid by assigning them the role of GSGFM DG or GSGFE DG as per their power handling capacity and grid loading conditions; thus, achieving the good steady-state performance with THD less than 1.1% and quick dynamic response with settling time less than 0.06 sec. These results will help in achieving the reliable and autonomous operation of hybrid MG with multiple DGs. Pakistan is the sixth largest nation in the world with around 51 million people (20% of the population) living off grid with no access to electricity. The proposed algorithm will aid in offering the offgrid solution to those areas, which are not connected to the grid.