EFECTO DEL MÍNIMO TÉCNICO DE UNIDADES GENERADORAS EN EL CÁLCULO DEL COSTO MARGINAL DE CORTO PLAZO SOBRE LA SEÑAL DE INVERSIÓN DE LARGO PLAZO
VERGARA BRANJE, CARLOS EDUARDO
MetadataShow full item record
The present work will study long-term capacity expansion planning when restrictionsin the short-term operation of the system are considered focusing on minimum stable level(MSL) constraints.The relevance of the analysis is linked to the revenues of current conventional powerplants that are operating at MSL conditions. This condition implies that these units mayoperate above the marginal cost of the electrical power system. From the optimal economicpoint of view, this strategy is inecient due to the place that they have in the order meritlist associated with the dispatch process.Certainly, the situation described above will aect the cost of the system and thereforea speci c entity will have to pay for the de cit to ensure payments for the operation ofthe units.Including MSL constraints of thermal generators in GCEP adds a number of integervariables to the mathematical formulation, wich can slow signi cantly its resolution. Thus,most GCEP models tend to ignore them. Rising penetration of variable renewable generationcan make this omission unadvisable. In order to analyze the investment signals withand without modeling technical minima of fossil-fuel generators (MSL), three simulationcases are considered.The rst set of simulations aims to nd dierences in the amount of installed capacityby technology, with and without considering MSL constraints.The purpose of this simulationis testing the idea that pure marginal theory is valid only for economically adaptedsystems. Thus, investment decisions will be continuous and simulations will be conductedin a single-node small test system with three types of generator and with a decision horizonof only one year and 6 load blocks per year.The second set of simulations will try to identify dierences for the investment timingfor a portfolio of available technology, considering an economically adapted system withand without considering MSL constraints. As the system will be economically adapted,unlike the rst set of simulations all investments will be forced. Thus, the dierences willbe given only by the year in which the investments occur and the fact that the investmentdecisiones are discrete. Simulations will be conducted for a single-node test system withthree dierent types of generators and a 15-year horizon.Finally, the third case will analyze the distortion of investment signals, consideringboth MSL and minimum operation time constraints. These simulations will be conductedon a modi ed SING system, considering transmission and a 15-year horizon. Two modelswill be simulated in this case. A classical model of capacity expansion planning consideringLDC decomposition will be compared against a chronological one.