Neural network and regression spline value function approximations for stochastic dynamic programming

2007 ◽  
Vol 34 (1) ◽  
pp. 70-90 ◽  
Author(s):  
Cristiano Cervellera ◽  
Aihong Wen ◽  
Victoria C.P. Chen
Keyword(s):  
Neural Network ◽  
Dynamic Programming ◽  
Stochastic Dynamic Programming ◽  
Value Function ◽  
Stochastic Dynamic ◽  
Regression Spline ◽  
Function Approximations

An integrated neural network stochastic dynamic programming model for optimizing the operation policy of Aswan High Dam

2011 ◽  
Vol 42 (1) ◽  
pp. 50-67 ◽  
Author(s):  
A. H. El-Shafie ◽  
M. S. El-Manadely
Keyword(s):  
Neural Network ◽  
Decision Making ◽  
Dynamic Programming ◽  
Decision Maker ◽  
Stochastic Dynamic Programming ◽  
Programming Model ◽  
Stochastic Dynamic ◽  
Release Policy ◽  
Aswan High Dam ◽  
Decision Making Model

Developing optimal release policies of multipurpose reservoirs is very complex, especially for reservoirs within a stochastic environment. Existing techniques are limited in their ability to represent risks associated with deciding a release policy. The risk aspect of the decisions affects the design and operation of reservoirs. A decision-making model is presented that is capable of replicating the manner in which risks associated with reservoir release decisions are perceived, interpreted and compared by a decision-maker. The model is based on Neural Network (NN) theory. This decision-making model can be used with a Stochastic Dynamic Programming (SDP) approach to produce a NN-SDP model. The resulting integrated model allows the attitudes towards risk of a decision-maker to be considered explicitly in defining the optimal release policy. Clear differences in the policies generated from the basic SDP and the NN-SDP models are observed when examining the operation of Aswan High Dam (AHD). The NN-SDP model yields policies that are more reliable and resilient and less vulnerable than those obtained using the SDP model.

scholarly journals Improving Dam and Reservoir Operation Rules Using Stochastic Dynamic Programming and Artificial Neural Network Integration Model

2019 ◽  
Vol 11 (19) ◽  
pp. 5367 ◽  
Author(s):  
Fayaed ◽  
Fiyadh ◽  
Khai ◽  
Ahmed ◽  
Afan ◽  
...  
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Dynamic Programming ◽  
Reservoir Operation ◽  
Stochastic Dynamic Programming ◽  
Ar Model ◽  
Stochastic Dynamic ◽  
Ann Model ◽  
Water Release ◽  
Reservoir Inflow

The simulation elevation-surface area-storage interrelationship of a reservoir is a crucial task in developing ideal water release policies for reservoir and dam operations. In this study, an inclusive (stochastic dynamic programming-artificial neural network (SDP-ANN)) model was established and applied to obtain an ideal reservoir operation strategy for Sg. Langat reservoir in Malaysia. The problems associated with the management of water resources mostly relate to uncertainty and the stochastic nature of the reservoir inflow, and the SDP-ANN model is meant to consider uncertainty in the input parameters such as reservoir inflow and reservoir evaporation losses. The performance of the SDP-ANN model was compared to that of the stochastic dynamic programming-autoregression (AR) model. The primary aim of the model is to decrease the squared deviation from the desired water release, which we determined by comparing the SDP-AR and SDP-ANN model performances. The results indicate that the SDP-ANN model demonstrated greater resilience and reliability with a lower supply deficit. Consequently, the case study results confirm that the SDP-ANN model performs better than the SDP-AR model in obtaining the best parameters for the reservoir operation. Specifically, a comparison of the models shows that the proposed Model 2 increased the reliability and resilience of the system by 7.5% and 6.3%, respectively.

scholarly journals Combined Aggregated Sampling Stochastic Dynamic Programming and Simulation-Optimization to Derive Operation Rules for Large-Scale Hydropower System

Energies ◽  
2021 ◽  
Vol 14 (3) ◽  
pp. 625
Author(s):  
Xinyu Wu ◽  
Rui Guo ◽  
Xilong Cheng ◽  
Chuntian Cheng
Keyword(s):  
Dynamic Programming ◽  
Stochastic Dynamic Programming ◽  
Large Scale ◽  
Simulation Optimization ◽  
Optimal Operation ◽  
Stochastic Dynamic ◽  
Operation Model ◽  
Two Stage ◽  
Operation Rules ◽  
Reservoir Systems

Simulation-optimization methods are often used to derive operation rules for large-scale hydropower reservoir systems. The solution of the simulation-optimization models is complex and time-consuming, for many interconnected variables need to be optimized, and the objective functions need to be computed through simulation in many periods. Since global solutions are seldom obtained, the initial solutions are important to the solution quality. In this paper, a two-stage method is proposed to derive operation rules for large-scale hydropower systems. In the first stage, the optimal operation model is simplified and solved using sampling stochastic dynamic programming (SSDP). In the second stage, the optimal operation model is solved by using a genetic algorithm, taking the SSDP solution as an individual in the initial population. The proposed method is applied to a hydropower system in Southwest China, composed of cascaded reservoir systems of Hongshui River, Lancang River, and Wu River. The numerical result shows that the two-stage method can significantly improve the solution in an acceptable solution time.

Sign in / Sign up

Export Citation Format

Share Document