Viability and Planning Aspects of Distributed Generation
Ashish Prakash Agalgaonkar, Ph.D, 06
Supervisor(s): S. V. Kulkarni
Many developing countries have emphasis on Distributed Generation (DG) technology for their generation expansion planning. The planning considerations and judicious choice of attributes are dictated by prevailing conditions. The optimal placement of DG using Genetic Algorithm (GA) is done with the objective of minimizing T&D losses and improvement in voltage profile at load buses. Similarly, optimal penetration level is worked out that consists of OPF formulation with the objective of minimizing the total cost of operation. The DG is also of significant importance under Locational Marginal Price (LMP) based market. If LMPs are available at all nodes, these price signals can be used effectively for siting of the DG. This concept has been applied for sizing and siting of the DG under Standard Market Design (SMD) framework (commonly adopted in US) for MATPOWER 30 bus and IEEE 118 bus systems. Developing countries, like India, are also encouraging the DG to reduce the demand-supply gap. Studies on three real life practical systems have been carried out to demonstrate viability of the DG under different considerations. The first one proposes the viability of the DG under intra-State Availability Based Tariff (ABT) regime, where ABT is a frequency linked energy pricing mechanism, adopted in India. The second one quantifies the improvement in the steady state system performance after inclusion of wind generation at a site in the State of Maharashtra, India. The third and the last one demonstrates the reliability enhancement of a distribution system (from Mumbai, India) after inclusion of the DG. Although the conventional planning issues and the system studies justify viability of the DG technologies, mere consideration of these aspects may not be sufficient for typical Multi Attribute Decision Making (MADM) problems. Hence, preliminary feasibility studies and some of the advanced planning techniques have been employed to arrive at a solution with comprehensive perspective of the planning problem. Central to all the advance planning techniques is building of a common platform on which utility function can be evaluated using normalized values of all the attributes. The attributes considered in the MADM techniques are capital costs, energy not served per annum, and profits from injecting power into grid at peak load period, all of which are important for a developing country. The uncertain futures considered are three possible loading conditions which can be low, medium and high. Different scenarios (plans) are generated by various combinations of configurations. The DGs can be configured in stand-alone mode, hybrid operation mode, or micro-grid formation mode with or without grid connection. With the increased complexities in DG planning options along with the multiple attributes to be accounted, more sophisticated techniques other than conventional economic analysis are needed to arrive at correct decisions by decision makers. Analytical Hierarchy Process (AHP) is used for obtaining relative weights in an objective way. Further, the statistical method like the interval based MADM technique with tradeoff analysis is used for shortlisting feasible plans and identifying the most appropriate plan. It is proposed to use the weights obtained from the AHP for finding the performance efficiencies in the Data Envelopment Analysis (DEA) for evaluating the plans. A sample system is derived with reference to a rural electrification scheme in India. The assessment of plans is presented and discussed. The comparative strengths and weaknesses of the methods are reported on the basis of the results obtained.