The Need for Recurrent Learning Neural Network and Combine Pareto Differential Evolution Algorithm for Multi-objective Real-Time Reservoir Operation Optimization

Document Type: Regular Article


1 Department of Civil and Environmental Engineering, Seattle Campus, University of Washington.

2 University of North Dakota, United States.

3 Department of Mechanical Engineering, The Polytechnic Ibadan, Nigeria.


Reservoir operations need computational models that can attend to both its real time data analytics and multi-objective optimization. This is now increasingly necessary due to the growing complexities of reservoir’s hydrological structures, ever-increasing its operational data, and conflicting conditions in optimizing the its operations. Past related studies have mostly attended to either real time data analytics, or multi-objective optimization of reservoir operations. This review study, based on systematic literature analysis, presents the suitability of Recurrent Learning Neural Network (RLNN) and Combine Pareto Multi-objective Differential Evolution (CPMDE) algorithms for real time data analytics and multi-objective optimization of reservoir operations, respectively. It also presents the need for a hybrid RLNN-CPMDE, with the use of CPMDE in the development of RLNN learning data, for reservoir operation optimization in a multi-objective and real time environment. This review is necessary as a reference for researchers in multi-objective optimization and reservoir real time operations. The gaps in research reported in this review would be areas of further studies in real time multi-objective studies in reservoir operation.


Main Subjects