Meta-heuristic Approach for Flood Control in Reservoir Operation

Besides irrigation and power generation, reservoirs in India also play a vital role in protecting the country from floods. During flood, reservoirs store and release the excess water according to its capacity and future requirement. It is imperative that the reservoir has to be operated effectively to conserve water so that competing goals can be accomplished. This requires an extensive investigation and the development of new system analysis approaches using meta-heuristic techniques of soft computing. The Omkareshwar Sagar Project (OSP) Reservoir in India has been designed to satisfy annual demands, protect against flooding, and produce maximum power generation, which is used for analysis in the current study. The present study incorporates one conventional optimization method the nonlinear programming (NLP), a semi- conventional method the genetic algorithm (GA) and a meta-heuristic technique the teaching–learning-based optimization (TLBO) to solve the problem of optimum operation in a flood scenario. A flood-release protocol has to be developed keeping the objective of maximum power output and reach the specified storage target at the end of the operation. Present study has been a successful attempt in getting the desired results, and the adoptability of the TLBO for complex problem is also proven.