Despacho de energía y reserva en microredes considerando criterios de seguridad probabilísticos

First, this PhD thesis presents and discusses the various types of forecasting techniques that may be issued as input to decision-making problems. A particular emphasis on predictions was placed, since markets ought to be cleared a fair amount of time before actual operation, while market participants shall then make decisions even before that. On the other hand, it was described a stochastic methodology that models and forecasts the behavior of the load, the wind power generation, the solar power generation and the failure of units, which are the main random variables affecting the normal operation of microgrids. Furthermore, this methodology lets to construct and evaluate both the net load forecast error scenarios and the unit outage scenarios throughout the next scheduling horizon. Statistical techniques were formulated in order to reduce the number of scenarios and, therefore, make the problem more computationally tractable. Then, an energy-reserve market clearing model for microgrids considering probabilistic security criteria was proposed using mixed integer nonlinear programming methods. The probabilistic security criteria include pre-selected scenarios associated to unreliability of generators and uncertainties caused by the stochastic behavior of loads and renewable units. In contrast to traditional deterministic reserve-constrained market clearing models, this approach determines the optimal amount of reserve as the point at which the sum of its operating costs and the expected cost of load shed reach a minimum. The proposed model was formulated as a two-stage stochastic programming problem. The first stage involves the hour-ahead market that evaluates the unit commitment and the energy-reserve scheduling before the realization of scenarios in the microgrid. The second stage considers the balancing market that investigates the security assurance in the pre-selected scenarios of the microgrid. Furthermore, the constraints pertaining to both the hour-ahead and the balancing market were formulated, which describe the operational rules of dispatchable units and flexible demands. After that, the microgrid ATENEA located at the installations of the National Renewable Energy Centre of Spain (CENER) is described. The technical, economical and operational characteristics of this grid are presented and evaluated. On the other hand, an energy management methodology was developed using JAVA in order to assess the market clearing formulation on the real microgrid ATENEA. This methodology was implemented under three operational configurations of the microgrid. The expected costs, the energy-reserve scheduling prior the revelation of scenarios, and the balancing actions to maintain the system security under the real-time scenarios were optimized and evaluated for each operational configuration. Finally, the market clearing solutions were assessed under different technical and economical conditions of the microgrid ATENEA. It was examined the effects of a) the demand-side valuation of energy not served, and b) the penetration of renewable generation technologies. Moreover, c) the scheduling results of the proposed stochastic formulation were compared with those obtained with a purely deterministic formulation, d) the total cost probability distribution of the microgrid operation was formulated and assessed using Monte Carlo techniques, and e) the computational issues related to the stochastic programming problem were discussed.

• Autores: Luna Ramírez, Luis Ernesto
• Tipo de recurso: Doctoral thesis
• Fecha de publicación: 2016
• Institución: Universidad Nacional de Colombia
• Repositorio: Universidad Nacional de Colombia
• Idioma: spa
• OAI Identifier: oai:repositorio.unal.edu.co:unal/55582
• Acceso en línea: https://repositorio.unal.edu.co/handle/unal/55582; http://bdigital.unal.edu.co/51010/
• Palabra clave: 62 Ingeniería y operaciones afines / Engineering; Forecasting; Market clearing; Microgrids; Mixed integer nonlinear programming; Probabilistic security; Reserve; Stochastic programming; Pronóstico; Despacho, microredes; Programación no lineal entera mixta; Seguridad probabilística; Reserva; Programación estocástica
• Rights: openAccess
• License: Atribución-NoComercial 4.0 Internacional