Rainfall optimization based algorithms for enhancing operational performances of distribution systems

Abstract

The Distribution Networks (DNs) continue to witness an exponential growth in energy demand, owing allegiance to extensive automated needs of both domestic and industrial utilities. It is in tune with the growth of the nation that the extent of sophistication appears to evince a sense of reality in the present-day world. Though technological revelation and the associated emergence of state-of-the-art methodologies facilitate the ordeal, still precise requirements keep emerging and there is considerable scope for improvement. newlineThe serious hassle that poses a threat to the security of DNs is the inadequate VAR support and the imminent problems of voltage instability. The other usual issues are large Active Power Loss (APL) and poor Node Voltage Profile (NVP) and Voltage Stability (VS). The APL of DNs is higher than those of high voltage transmission networks and is nearly 10 13% of the total generated power. Several strategies have been suggested to reduce the network APL, and enhance NVP and VS of the DNs. The popularly adopted methods are Network Reconfiguration (NWR), Capacitor Placement, Distributed Generation Placement (DGP), Flexible AC Transmission Systems (FACTS) Device placement and Load Shedding (LS). The DGP problem also includes the Total Operating Cost (TOC) as another objective. newlineThe Rainfall Optimization (RFO), imitated from the behaviour of the falling of raindrops, (Kaboli et al. 2017) has a few shortcomings such as elimination of inactive raindrops that deteriorates the exploration, and chances of selecting the same raindrop as inactive during the solution process. This thesis suggests modifications in the RFO with phases of Local Fallquot and quotGlobal Throwquot to overcome its drawbacks. This thesis also attempts to formulate NWR and DGP problems as optimization problems and apply a New Enhanced RFO (ERFO) algorithm for solving the formulated problems. It then combines both NWR and DGP problems for tailoring the composite problem, and suggest ERFO based method for solving the composite problem. It presents the results of the proposed methods on two standard DNs with different test cases with a view of exhibiting the superior performances newline newline