Neural Network-Based Simulation-Optimization Model for Reservoir Operation

Abstract

There have been several attempts to use combined simulation-optimization models to solve reservoir operation problems efficiently. In many cases, complex simulation models are available, but direct incorporation of them into an optimization framework is computationally prohibitive. To overcome this problem, in this study, a backpropagation neural network is trained to approximate the simulation model developed for the Chennai city water supply problem. The neural network is used as a submodel in a Hooke and Jeeves nonlinear programming model to find “near optimal policies.” The results are further refined using the conventional simulation-optimization model.