Athens University Economics and Business (AUEB)
Economic modelling and optimisation of smart grids using demand response techniques
Abstract
dc:descriptionThe energy sector is experiencing major challenges that are dictated by the deregulation of the electricity market, technological advances, the integration of renewable and distributed energy resources, economic factors and the ever-increasing consumer demand in the so-called smart energy grid. The deployment and exploitation of Demand Response (DR) and Automated Demand Response (ADR) techniques constitute an interesting and promising approach to dealing with the supply and demand mismatches by means of shaping or curtailing of consumer loads. DR programs aim to alleviate the peak demand problem and provide higher system stability by modifying user demand in response to power grid’s supply and economic conditions. Designing a successful DR program is challenging, as it depends on a multitude of factors (e.g. accuracy of user profiles) but most importantly on the load curtailment actually attained by the consumers. This implies that DR programs should be designed in a way that consumers are appropriately triggered –by means of either price signals or incentives – to actively participate in the program by modifying their consumption pattern so that the power demand during the peak hours is reduced. In this way, the energy providers and retailers are able to limit their capacity up to the level that they can serve without activating costly and environmentally non-friendly peak demand generators.The main objective of the present doctoral thesis is to contribute to the empowerment of the (currently limited) adoption of DR programs by designing efficient DR incentives and associated contracts, hence contributing to the creation of a socio-economically sustainable ecosystem where all players benefit and the economic and environmental effects of the mismatch between supply and demand (such as CO2 emissions) are reduced. To this end, we propose a complete, practically applicable and scientifically justified approach to design efficient incentive-based DR and ADR programs and validate them by considering consumers' individual characteristics and context, in order to limit their discomfort from the modifications in their consumption patterns. Thus, consumers’ active participation can be attained by means of proper contracts and consumption schedules (acquired by means of either smart meters or load disaggregation), and therefore ensure the ADR programs’ success. In particular, a novel methodology is proposed that utilizes appliance level measurements to develop a utility function that represents the consumption behaviour of a single consumer and his/her associated comfort zone. The utility function introduced is exploited to define efficient incentives and targeting policies for ADR contracts that energy players such as providers and retailers can use in order to engage customers to ADR programs in an economically mutually beneficial manner. Two general cases regarding the available information are considered in order to evaluate the theoretical models presented in this thesis: (i) full information availability and (ii) partial information availability. 3Economic Modelling and Optimisation of Smart Grids using Demand Response TechniquesMarilena Minou |July 2019Moreover, focusing on users’ characteristics and how they can be leveraged to become of greater value for the society (through the success of the DR programs), specific behavioral characteristics delineating users’ behavior (e.g. altruism) are incorporated in users’ utility functions to study the impact on providers’ goals and the proposed theoretical models. In particular, different blends of population are considered, i.e. either only rational users or a mix of rational and altruistic users, with the latter being modelled by utilizing principles from behavioural economics. Building on users’ utility models, an innovative net benefit-based maximization approach of incentives calculation is proposed accompanied with two policies that restrict in a different way the discomfort due to the demand reduction of users in DR, as well as a selection algorithm to target users for DR in the context of offering ADR contracts and on the basis thereof. The net benefit-based maximization approach of incentives calculation and the selection algorithm can be applied both in the full and partial information cases. In order to tackle real life situations, where only partial information is available, we utilize a straightforward and well-known in the literature technique of clustering and classification to obtain detailed information about users and their preferences, i.e. users’ utility functions. The latter are then employed to apply our net benefit-based maximization approach of incentives calculation together with the two policies. Users’ active participation is further encouraged by offering different scenarios of contract selection from the available offered contracts. The proposed methodologies are validated by means of simulations utilizing real consumption measurements from datasets from India and Sweden pilot sites, expanded by means of synthetic data. One of the main results is that our innovative policy constraining directly the reduction in utility of a DR participant, rather than the reduction in energy consumption (as usually done in the literature) results in lower DR total incentives are being offered by the provider and higher social welfare after ADR. This implies that even with a very small percentage permissible reduction of utility, i.e. with very low discomfort, the desired target can be met in an economically effective way when this policy is applied, as the most flexible users in terms of consumption are selected. Moreover, as demonstrated in the thesis, the application of data mining techniques for clustering users in the case of partial information proves to be a very good approach and can lead to similar results as in the full information case. Finally, based on the findings of this thesis, we compile a set of guidelines for the providers to design economically sustainable DR and ADR programs.
Degree
thesis:*- Grantor dc:publisher
- Athens University Economics and Business (AUEB)
- Year dc:date
- 2019
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Minou, Marilena
- Μήνου, Μαριλένα
Subjects
dc:subject × 8Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/46753