Non-linear programming model for multi-purpose multi-reservoir optimization

In the present study, a detailed non-linear programming (NLP) model for a multi-purpose reservoir has been formulated and applied for a multi-reservoir system in upper Krishna river basin in Karnataka, India. The main objective is maximizing the irrigation, municipal and industrial release, and power production. The developed non-linear programming model is solved by using GAMS/MINOS. The system performances such as the irrigation deficit, frequency of irrigation deficit and power production are analyzed and compared with that of the results of a classical linear programming model (LP). It is found that the system could be managed with less irrigation deficit along with comparable power production in longer run using NLP model. It is also found that the LP model results in more irrigation deficit than NLP model