| PhD Viva

Name of the Speaker: Ms. Naina P M (EE15D024)
Guide: Prof. Swarup K.S
Online meeting link: https://meet.google.com/piu-nkrq-amo
Date/Time: 1st February 2024, 3:00 PM
Title: Distributed Optimization Algorithms for Virtual Power Plants

Abstract

Virtual Power Plants (VPPs) are becoming increasingly pivotal in the context of modern energy grids, facilitating the integration of renewable and distributed energy resources. VPP enables the market participation of small distributed generators, including renewable sources. The proliferation of distributed generators (DGs) fundamentally reshapes the power grid into a decentralized network. The implementation of Energy Management in VPP faces technical as well as economic challenges. This thesis delves into the realm of Energy Management within VPPs.

Energy Management Operations employing conventional centralized algorithms may not be suitable, effective, and accurate in the presence of large number of DGs. Consensus-based distributed algorithms for energy management are presented in this work. Since the consensus protocol is a communication-based approach, noise and communication delay are unavoidable. Two different variants of consensus algorithms are proposed, which effectively reduce the adverse effects of noise and communication delay with time-varying communication networks. The algorithms are successfully tested on an IEEE 39 bus system and a 15-node VPP.

This work also tackles a crucial challenge: effectively managing a Virtual Power Plant (VPP) by optimizing the scheduling of Distributed Energy Resources (DERs) to maximize profits while minimizing risks. The concept of VPP becomes particularly valuable when faced with uncertainties like variable renewable energy generation and fluctuating electricity prices. Our research explored scenario-based stochastic scheduling techniques to address uncertainties arising from variables such as wind power generation and electricity prices. The VPP is modeled on a modified IEEE 33-node distribution system, and we aimed to maximize its expected profit while considering the network parameters.