Author(s): Rolando Munoz Ortiz; Maritza Liliana Arganis Juarez; Ramon Dominguez Mora; Rosalva Mendoza Ramirez; Eliseo Carrizosa Elizondo; Eduardo Juan Diego

Linked Author(s): Maritza Liliana Arganis Juárez, Ramón Domínguez, Rosalva Mendoza Ramírez

Keywords: Genetic Algorithm; Synthetic Records; Optimization; Cutzamala; Dam

Abstract: The objective of this work is to optimize the joint operation of the dams of the Cutzamala system, in order to reduce the spilled volumes or possible deficits presented during the simulation of historical and synthetic records of the input volumes; To do this, a genetic algorithm will be used to establish a monthly extraction policy that efficiently distributes the annual volume assigned to meet the needs of drinking water supply. The Cutzamala System uses the waters of the upper basin of the Cutzamala River, coming from the Tuxpan and El Bosque dams, in the State of Michoacán; Colorines, Ixtapan del Oro, Valle de Bravo and Villa Victoria in the State of México, as well as the Chilesdo dam (which was built to take advantage of the waters of the Malacatepec River), and leads them to the Atarasquillo tunnel that later delivers to the Dolores tank for its distribution. To simulate the operation of a reservoir, the continuity equation is used, in the volumes of entry of the dams only the runoff that occurred in the basin was considered, deduced by the National Water Commission (CONAGUA) in its historical report; while for the output volumes, extractions of water volume for irrigation and for drinking water supply were considered, as well as a net evaporation that considered the input volumes due to rain and outputs due to evaporation. The total annual extraction volume is inferred from a graph that relates annual extractions against initial storages, which will be called: "Z curve". The way in which the graph is determined is by obtaining the total historical extractions for each year, the same happens for the initial storages, considering that the year begins on November 1; subsequently, the maximum and minimum values of each variable are selected, obtaining the pairs of points that make up the Z curve. The slope of the curve determines the extraction corresponding to each storage and with it the simulation can begin to optimize the extractions, (generally) forcing the results to fall within the slope; but for this case, it seeks to propose a new Z curve that meets the reservoir restrictions. Using the genetic algorithm and establishing as search variables the coordinates of points “a” and “b” of the Z curve, and the percentages of monthly distribution of the annual extraction for water supply; New Z curves and a preliminary distribution of volumes destined for drinking water supply were obtained.

DOI: https://doi.org/10.3850/IAHR-39WC252171192022508

Year: 2022