Effective DR gathering and deployment for intensive renewable integration using aggregation and machine learning

Funder

Organizational Unit

Publications

Rating the participation in Demand Response events with a contextual approach to improve accuracy of aggregated schedule

Publication . Silva, Cátia; Faria, Pedro; Vale, Zita; Terras, José M.; Albuquerque, Susete

The flexibility provided by the demand side will be crucial to take a step forward to increase the penetration of renewable energy resources in the system. The proposed methodology provides the aggregator with information about the most reliable consumers, attributing a trustworthy rate that characterizes their performance on Demand Response (DR) events. The innovation relies on applying rates and evaluating the context in which the event is triggered and the factors that influence such rates. The authors find that context is essential to understand which participants are available for the event and achieve the reduction target successfully. Also, the proposed methodology focuses on the performance and the proper motivation for continuous participation, reducing the uncertainty of the response in DR events by giving higher economic compensation to the active consumers with better results. Distributed generation is also optimally managed by the aggregator. Findings prove the feasibility of the proposed methodology supporting the Aggregator in communities and smart cities management.

2022Journal article

Open access

KPI for Managing and Controlling a Demand Response System: A Testing Framework for End Users

Publication . São José, Débora de; Faria, Pedro; Silva, Cátia; Vale, Zita

Considering the increase of distributed generation and the complexity in power electricity management, demand response programs can be a way to reduce stress and strengthen power grids. However, as demand response involves end users changing their consumption patterns and adapting to on time different scenarios, some decision-making support tools are necessary. The present paper proposes and tests an energy management and controlling framework to assist electricity end users in decision making in a demand response scenario while using a set of key performance indicators. The tool was tested using a group of 20 end users and showed a consistent result throughout all the elements in the consumers group, total consumption presented a small decrease due to reduce of comfort, especially during weekdays.

2020Conference object

Open access

Demand response performance and uncertainty: A systematic literature review

Publication . Silva, Cátia; Faria, Pedro; Vale, Zita; Corchado, J.M.

The present review has been carried out, resorting to the PRISMA methodology, analyzing 218 published articles. A comprehensive analysis has been conducted regarding the consumer's role in the energy market. Moreover, the methods used to address demand response uncertainty and the strategies used to enhance performance and motivate participation have been reviewed. The authors find that participants will be willing to change their consumption pattern and behavior given that they have a complete awareness of the market environment, seeking the optimal decision. The authors also find that a contextual solution, giving the right signals according to the different behaviors and to the different types of participants in the DR event, can improve the performance of consumers' participation, providing a reliable response. DR is a mean of demand-side management, so both these concepts are addressed in the present paper. Finally, the pathways for future research are discussed.

2022Journal article

Open access

Clustering Direct Load Control Appliances in the Context of Demand Response Programs in Energy Communities

Publication . Barreto, Rúben; Faria, Pedro; Silva, Cátia; Vale, Zita

The demand response program explained in this article is designed to be implemented in communities seeking to achieve a self-sustaining system, namely through renewable energy such as photovoltaic energy. This article, through concepts such as prosumer and clustering, aims to make the most efficient management of the resources provided by the energy community. The developed demand response clusters the different consumers who have the same type of consumption throughout the day. That is, it brings together those whose behavior of the respective loads resemble each other and can be viewed from the perspective of an individual load or even clustered with one or more loads. The study comprises three villages with different numbers of consumers and charges, where, through their participation, it is estimated that there are reductions in electricity bills and, for those who collaborated for the study, it is attributed a remuneration according to their performance.

2020-07Conference object

Open access

Rating the Participation in Demand Response Programs for a More Accurate Aggregated Schedule of Consumers after Enrolment Period

Publication . Silva, Cátia; Faria, Pedro; Vale, Zita

Aggregation of small size consumers and Distributed Generation (DG) units have a considerable impact to catch the full flexibility potential, in the context of Demand Response programs. New incentive mechanisms are needed to remunerate consumers adequately and to recognize the ones that have more reliable participation. The authors propose an innovative approach to be used in the operation phase, to deal with the uncertainty to Demand Response events, where a certain target is requested for an energy community managed by the Aggregator. The innovative content deals with assigning and updating a Reliability Rate to each consumer according to the actual response in a reduction request. Three distinct methods have been implemented and compared. The initial rates assigned according to participation in the Demand Response events after one month of the enrolment period and the ones with higher reliability follow scheduling, performed using linear optimization. The results prove that using the proposed approach, the energy community manager finds the more reliable consumers in each period, and the reduction target achieved in DR events. A clustering algorithm is implemented to determine the final consumer rate for one month considering the centroid value

2020Journal article

Open access