Optimal Operation Modeling of Reservoirs using Mixed Integer Linear Programming (MILP)

One of the best ways to deal with water resources management problems and the unwanted temporal and spatial distribution of water is the optimal operation of reservoirs. Optimization techniques have become increasingly important over the last three decades in the management and operation of complex reservoir systems. In recent years the water resource experts have benefited from many hardware and software advances and used different kinds of tools and methods for optimization. In this research a software is developed based on the Mixed Integer Linear Programming (NIILP) for optimal use of the multiobjective multireservoir system of the Tehran-Karaj plane. The MILP method is selected due to the characteristics of the reservoir system, the availability of data, and the objectives and constraints. The topology of the system is designed based on the network flow optimization approach. The topology includes the water resources (Lar, Latian, and Karaj reservoirs), the water consumers (i.e. domestic, agriculture, industrial), and the hydraulic connections. The objective function, constraints, and piecewise linear of nonlinear relationships were formulated and the penalties and the priorities in each branch and bound were introduced. The software includes GUI, a database, a solver, spreadsheets and graphical output analysis. The results of this research indicated that the developed software is a suitable model for deriving operational rules in the system under study. Results also show that using this model for management and operation of the water resources system in the Tehran-Karaj plain result in a better allocation of water to the consumer compared to the LP model and historical operations.