Modelo de simulação qualiquantitativo multiobjetivo para o planejamento integrado dos sistemas de recursos hídricos.

Abstract

The use of mathematical models in the analysis of water resources systems has been widely discussed in the literature, although all of them present simplifications regarding or to the mathematical representation of its dynamic processes or numerical approximations of the nonlinearities that describe their behavior. Depending upon the model applicability, such simplifications may compromise a decison making. All simulation models available for riverbasin water resources system planning, although being very versatile, are not able to include certain complexities due to some mathematical limitations of their algorithms. Based on this context, to deal with these problems, a new simulation model has been proposed, which has, as main characteristics, the development of an integrated qualiquantitative multiuse surface water systems model, the inclusion of nonlinearities of hydraulic and operational processes and a multiobjective function that otimizes, according to priorities criteria, the fulfillment of water demands for the different uses and the targeted water quality levels, according to the CONAMA norms as well as of other operational targets. The considered water quality parameters were the Biochemical Oxygen Demand, the Dissolved Oxygen, the Total Phosphorus, the Total Nitrogen, the Chlorophyll-a and the Faecal Coliform. The operational and hydraulic limitations were included in the model through linear constraints (water balance, target volumes, minimum volumes, component capacities, etc.) and nonlinear ones (maximum reservoir spillage and discharge flows, reservoir surface water, etc.). The nonlinear constraints were approximated by linear segments (linearization artifices) and solved in an iterative process (sequential linear programming) until a desired tolerance for the objective function relative error was reached. The mass balance for water quality parameters, which requires a multiplication of decison variables, were linearized by the Linear Approximation Method and interacted, in the same algorithm, with the water balance equations. To evaluate the performance of the fulfillment of water demand requirements, some performance indicators, such as reliability, resilience, vulnerability and sustainability, were included in the model. The model's applicability were tested in a systemic analyis of two reservoirs, Engenheiro Avidos and São Gonçalo ones, and their water quali-quantitative demands, which are located in Alto Piranhas riverbasin, PB. The analysis of this system was accomplished through the idealization of 4 operation scenarios, which made use of different priorities and the inclusion of a Waste Water Treatment Plant (ETE) in the last scenario. The simulation time horizon was 360 months. The fulfillment of water demand requirements and of the CONAMA 375/05 Class II targeted concentration level for water quality parameters were analyzed. The results have shwon the all constraints were satisfied and the reliability of water demands requirement fulfillment were different for each of the simulated scenario. The ETE inclusion scenario helped to improve the fulfillment of water demand uses requirement and to improve the system water quality. The new model proved to be efficient regarding to the computer time requirement to achieve a solution as well as to satisfy all constraints of the system. It also presented a small numerical error.