Candidate: Mahmoud Ahmed Allam Sayed Alsanbawy
Title: Steady-State Analysis and Optimal Power Routing of Standalone Unbalanced Hybrid AC/DC Microgrids
Date: July 12, 2018
Time: 3:00 PM
Place: EIT 3145
Supervisor(s): Kazerani, Mehrdad
Distributed generation (DG) allows the connection of small-scale renewable energy resources (RER) as well as energy storage systems (ESS) – through power electronic converters – to existing ac distribution networks. In the early 2000s, the concept of droop-controlled ac microgrids was introduced to integrate DGs as well as controllable and uncontrollable loads within one entity that can operate autonomously or connected to a utility grid. More recently, droop-controlled dc microgrids have received increasing attention as a potential solution to deliver power from DGs to modern dc loads with minimum number of conversion stages for reduced cost and improved efficiency. Moreover, hybrid ac/dc microgrids have been introduced as a paradigm combining the benefits of ac and dc microgrids by interconnecting them through interlinking converters (ICs). Furthermore, autonomous control schemes have been proposed to coordinate the power flow between the ac and dc subgrids without relying on communication links.
Steady-state analysis is essential for planning and operation studies of electrical power systems. However, conventional analysis approaches cannot be applied to hybrid ac/dc microgrids due to their distinctive features, such as droop characteristics that lead to load-dependent system variables, lack of a slack bus, and coupling between the ac and dc variables. Additionally, the unbalanced nature of ac microgrids adds to the modeling and analysis complexity of such networks. Therefore, this thesis is focused on developing steady-state modeling and analysis approaches for standalone unbalanced hybrid ac/dc microgrids. Furthermore, the thesis employs the developed tools to identify challenges pertaining to planning and operation aspects of these microgrids. Subsequently, the thesis proposes novel solutions to some of the challenges and sheds light on potential research directions to tackle the unsolved ones.
First, a steady-state analysis tool for unbalanced hybrid ac/dc microgrids is developed. The unbalanced ac subgrid's components as well as the ac side of ICs are modeled in phase coordinates. Furthermore, the dc subgrid's components are modeled and the coupling between the ac and dc variables is formulated. The models of the various system elements are incorporated into a unified power-flow formulation. The power-flow problem is solved using a Newton-Trust Region (NTR) method. The proposed models and power-flow algorithm are verified through comparisons with detailed time domain simulations, in PSCAD/EMTDC, of test microgrids. The analysis tool is further used to analyze a large-scale hybrid ac/dc microgrid through case studies. The case studies shed light on some challenges of these microgrids, namely, imposed limitations on microgrid loadability due to unbalanced ac subgrid's loading, effect of IC settings on microgrid operation, and trade-off between proportional loading of the ac and dc subgrids and proportional power-transfer sharing among ICs.
Second, based on the above-identified microgrid loadability limitation of unbalanced microgrids, a novel dynamic power routing (DPR) scheme is proposed to maximize the microgrid loadability. The proposed scheme allows independent control of active and reactive powers flowing through IC phases, so that the power can be routed between the ac subgrid's phases. Accordingly, loading can be diverted from heavily loaded phases to lightly loaded ones. The DPR scheme is integrated into an optimal power flow (OPF) formulation with the objective of minimizing load shedding, thus maximizing microgrid loadability. A supervisory controller with low communication requirements is proposed to solve the OPF problem using an interior point method. The supervisory controller adjusts the DG and IC settings to achieve the OPF objective. Several case studies are conducted to highlight the problems associated with autonomous microgrid operation under heavy loading, to show the ineffectiveness of conventional supervisory controllers in solving the loadability issue, and to verify the success of the proposed controller in solving the problem.
In the first stage of the thesis, the developed unified power-flow analysis approach – based on phase-coordinate models of the ac subgrid's components – shows sufficient accuracy for steady-state analysis. Nonetheless, improvements could be achieved by following other approaches that demonstrate more generic and precise representations of different DG types (i.e., electronically interfaced DGs and synchronous generator-based DGs), transformers, and loads. Accordingly, the third part of the thesis proposes modeling the ac microgrid's components in symmetrical sequence components, thereby allowing more accurate modeling of the ac microgrid components. Furthermore, this approach breaks down the system model into smaller subsystems (i.e., positive-, negative-, and zero-sequence) that can be solved in parallel for enhanced performance. A Newton-Raphson (NR) method is adopted to solve the positive-sequence power-flow, while the complex negative- and zero-sequence voltages are obtained by solving linear complex equations. The accuracy of the proposed approach is verified through comparisons with time-domain simulations conducted in MATLAB/Simulink. Moreover, case studies are introduced to verify the effectiveness of the algorithm for large-scale ac microgrids. In addition, the algorithm is utilized to investigate the operation of droop-controlled DGs in isochronous unbalanced ac microgrids and to examine its limit-enforcement abilities at the same time. The algorithm demonstrates significant improvements in terms of accuracy and convergence time when compared against the conventional NTR-based approach in phase coordinates.
Finally, the power-flow approach developed in the third part is extended to include the IC's and dc subgrid's models so that it can solve for hybrid ac/dc microgrids. A sequential, rather than unified, power-flow algorithm is proposed, which solves the ac and dc power-flows independently in a sequential manner. Nevertheless, the correlation between the ac and dc variables is maintained by updating some dc power-flow variables based on the ac power-flow solution results. The proposed sequential method divides the system model into two smaller ac and dc subsystems, while the sequence component technique further splits the ac subsystem's model into three smaller ones that can be solved in parallel. These multiple reductions in the model size offers substantial reduction in the overall computational and memory requirements of the power-flow algorithm. The algorithm is verified through comparisons with time-domain models of test hybrid microgrids in MATLAB/Simulink. Case studies – incorporating a large-scale hybrid ac/dc microgrid – are introduced to test the algorithm's effectiveness in enforcing the DG and IC limits in the power-flow solution under various conditions. The algorithm also shows superior performance – in terms of accuracy and execution time – as compared to the tool developed in the first stage of the thesis.