Optimal reservoir operation policy determination for uncertainty conditions

Abstract

In recent years, optimising reservoir operations has emerged as a hot topic in the field of water resources management. Heuristic approaches to reservoir operation that incorporate rule curves and, to some extent, operator discretion have been the norm in the past. With so many stakeholders involved in water management, it can be difficult to strike a happy medium between everyone's needs and wants. The dissertation proposes a method for transforming traditional reservoir operation into optimal strategies, allowing users to take advantage of the rapid development of computational techniques. This research creates and applies a Multi Objective Fuzzy Linear Programming (MOFLP) model to the monthly operating policies of the stage-I Jayakwadi reservoir located on the Godavari, the largest river in the Indian state of Maharashtra. In order to formulate the problem, we use two objective functions—maximizing irrigation releases and maximising power production releases. The constraints of the study are considered, including turbine release, irrigation demand, reservoir storage capacity, and continuity of reservoir storage. Utilizing linear membership functions, the objective functions are fuzzily defined. All other model parameters except for the goals are assumed to be hard and fast rules. Maximum Happiness Operating Policy (MOFLP) was used to determine the best course of action.