Authors:
Paula Medina Maçaira
;
Margarete Afonso de Sousa
;
Reinaldo Castro Souza
and
Fernando Luiz Cyrino Oliveira
Affiliation:
Industrial Engineering Department, Pontifícia Universidade Católica do Rio de Janeiro, Rua Marquês de São Vicente, 225, Rio de Janeiro and Brazil
Keyword(s):
Forecasting, Time Series, Hydroelectric Power Generation, Distributed Generation, Small Hydropower Plant, Exogeneous Variables.
Related
Ontology
Subjects/Areas/Topics:
Agents
;
Applications
;
Artificial Intelligence
;
Bioinformatics
;
Biomedical Engineering
;
Decision Analysis
;
Energy and Environment
;
Enterprise Information Systems
;
Forecasting
;
Industrial Engineering
;
Information Systems Analysis and Specification
;
Methodologies and Technologies
;
Operational Research
;
Pattern Recognition
;
Simulation
;
Software Engineering
;
Stochastic Processes
Abstract:
Assertiveness in generation forecast is an important issue for utilities when they are planning their operation. Hydropower Generation forecast has a strong stochastic component and thinking about small hydropower plants (SHP) is even more complex. In recent years, many SHP was installed in Brazil due to a Government incentive and the distributed generation penetration has an impact in technical losses’ estimation. The objective of this study is to propose a methodology to generate synthetic scenarios of distributed generation for hydro sources. A case study was carried on with historical generation data from SHP located in Minas Gerais. The periodic regression model was considered the best model for forecast hydropower generation. Three distributed generation scenarios are obtained using Conditional Value at Risk analysis after combining multiple scenarios from inflow forecasting generated with the periodic regression model.