Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.786694
Title: Fair load shedding solutions for developing countries
Author: Oluwasuji, Olabambo Ifeoluwa
ISNI:       0000 0004 7972 1352
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
In order to remain in operation by maintaining a balance between demand and supply, grid operators in many developing countries often resort to disconnecting the load on parts of the grid from supply. This measure, known as load shedding, is often necessitated by shortages in their supply capacity. A consequence of load shedding is that households in disconnected parts are left without supply. In addition, existing load shedding schemes do not take fairness into consideration at the household level, meaning that some homes bear the brunt of load shedding. Against this background, we present a number of fair household-level load shedding solutions in this thesis. We first simulate a representative dataset for formulating and evaluating our solutions from a Pecan Street Inc. dataset. Thereafter, we model homes as agents and, in so doing, create a vector of values (i.e., the comfort vector) to embody their electricity needs. Thereupon, we develop a first set of solutions which result in homes being connected to electricity for even durations. Following this, we develop a second set of solutions which make up for the limitations of the first, in that they factor the comfort values of agents into consideration. In developing the second set of solutions, we establish agent utilities in terms of the number of hours they are connected to supply, the comfort they derive from supply, and the level at which their demand is satisfied. Then, we model the solutions as Mixed Integer Programming (MIP) problems, with objectives and constraints that maximize the groupwise and individual utilities of agents and minimize the pairwise differences between their utilities. Using a number of experiments, we show how the MIP solutions outperform the heuristics, by producing results which outperform and Pareto dominate those of the heuristics in terms of all utilities. When taken together, this thesis establishes a set of benchmarks for fair load shedding schemes. In addition, it provides insights for designing fair allocation solutions for other scarce resources.
Supervisor: Ramchurn, Sarvapali Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.786694  DOI: Not available
Share: