PLANNING AND ANALYSIS OF DISTRIBUTED ENERGY STORAGE SYSTEMS IN ACTIVE DISTRIBUTION NETWORKS

Abstract

The traditional power system is transforming into a smart and intelligent grid to meet the increasing demand and promote sustainability. Among its various components, the distribution network is undergoing significant transformations to enhance power quality and promote decarbonization. This involves accommodating distributed energy resources like wind, solar PV (photovoltaic), etc. These resources cater to the local demand by providing environmentally friendly electrical energy. However, the intermittent nature of these resources contributes to fluctuations in the power being injected and results in continuous variations in measured quantities like current, voltage, and frequency. These variations lead to a mismatch between supply and demand, influencing the operational dynamics of power systems. Further, the proliferation of renewable energy resources poses problems like bidirectional power flow, reverse power flow, and voltage and frequency fluctuations. The integration of distributed energy storage is proposed to address the above challenges. However, the widespread adoption of energy storage in any distribution network requires optimal planning for both investment and system operation. This necessitates the application of optimization techniques to determine the optimal solution for the planning and operation of distributed energy storage in an active distribution network. This research work is motivated by the distribution system’s dynamic nature, particularly in its pursuit of enhanced power quality and sustainable development through the integration of renewable energy resources. This dissertation presents various optimal planning and operation methods to facilitate a profitable integration of distributed battery energy storage systems (BESS) in balanced and unbalanced distribution networks. Firstly, a two-stage mixed-integer linear programming problem is developed to determine BESS capacity and discharge cycles, considering battery life cycle, load and PV output uncertainty, and system islanded operation. This approach employs probabilistic analysis and timeperiod clustering to address the uncertainty and variability of PV and load demand. The method is validated on standard balanced distribution networks with PV penetration. Recognizing the impact of reconfiguration on power flow, especially evident in the IITR distribution network model. The optimization problem is extended to include the reconfiguration. This extension allows for the observation of changes in the proposed sites and sizes of BESS. However, most low-voltage distribution i networks are unbalanced; a detailed analysis of the modeling of a low-voltage distribution network in an Open distribution system simulator (OpenDSS) is presented. The proposed BESS sizes are validated in OpenDSS using a built-in storage controller. To further enhance the practicality, additional constraints are modeled in the MATLAB COM interface to prevent early end-of-life battery energy storage. The method is validated on a practical Indian distribution network at IITR, showcasing its applicability in real-world scenarios. Further, the increasing presence of renewable energy resources in active distribution networks (ADNs) yields benefits such as low carbon emissions and access to free energy. However, it also introduces challenges like voltage fluctuations and increased losses during BESS operation. To address these challenges, an energy management and optimal dispatch approach is proposed for BESSs in unbalanced distribution networks. This approach uses mixed-integer linear programming formulation for optimal power flow and a rule-based method using load flow in unbalanced networks. The proposed methodology demonstrates effectiveness in minimizing operation and maintenance costs, reducing losses, and improving overall system performance. Moreover, the research delves into an energy management approach utilizing distributed energy storage in an unbalanced distribution network, modeled in OpenDSS with MATLAB-COM interface driving the analysis. Furthermore, the thesis introduces the planning of BESS in an unbalanced distribution network using second-order cone programming. The total problem includes ZIP load models and uncertainty of renewables and various distribution networks components like voltage regulators and capacitor banks. The results are validated in real-time using the RSCAD model of the distribution network. A novel utilization factor is introduced for BESS to determine the usage of BESS w.r.t gird input. The IEEE 13-bus and IEEE 123-bus networks are used for unbalanced network problem validation, while balanced network problem validation utilizes the IEEE 33-bus and IEEE 141-bus distribution networks. A thorough modeling and testing of methods is also carried out on the practical IITR distribution network with rooftop PV penetration. The optimization problems are solved in MATLAB R2019a interfaced with YALMIP version ‘20200930’ and the GUROBI 9.1.1 solver. Furthermore, microgrids (MGs) and ADNs are innovations related to the grid modernization initiative. Both offer the potential for sustainable energy management, reducing carbon emissions and energy-efficient power supply. Therefore, an optimization problem is formulated to determine the size and select the specific chemistry of a BESS. Also, a case study for multi-microgrid coordination with distribution network operation is solved using the consensus-based alternate direction method of multipliers (ADMM). The IEEE 33-bus system is integrated with multi-microgrids to test and validate the proposed optimization scheme. The research presented in this thesis is poised to make substantial contributions to the planning and operation of energy storage systems in ADNs and MGs. The methodologies developed in this dissertation demonstrate significant potential for effectively utilizing distributed BESSs. These strategies aimed to optimize the operation of distribution networks and minimize energy costs. The dissertation concludes by offering future research suggestions, observations, and gaps in this domain for prospective researchers.