Fred Otieno

A Combined Pareto Differential Evolution Approach for Multi-objective Optimization

In recent years, methods of multi-objective evolutionary algorithms (MOEAs) have been developed to solve problems involving the satisfaction of multiple objectives within the limits of certain constraints, yet there still exists some uncertainty about finding a generally trustworthy method that can consistently find solutions which are really close to desired objectives in all situations. In this study, a combined Pareto multi-objective differential evolution (CPMDE) algorithm is presented. The algorithm combines methods of Pareto ranking and Pareto dominance selections to implement a novel selection scheme at each generation. The ability of CPMDE in solving unconstrained, constrained and real-world optimization problems was demonstrated. Competitive results obtained from benchmarking CPMDE suggest that it is a good alternative for solving real multi-objective optimization problems.

Real-time optimal water allocation for daily hydropower generation from the Vanderkloof dam, South Africa

Display Omitted The adverse effects of power shortages resulting from escalating energy demands, due to rapid global urbanization and industrial developments, have driven efforts worldwide in search for improved techniques for sustainable reservoir operations and optimization for hydropower generation. Recent studies have shown that, combining accurate reservoir inflow forecasting procedures with efficient optimization techniques can produce more efficient and balanced solutions, for operation of multipurpose reservoir systems to improve on the economy of hydropower production. This study presents the coupling of a data driven artificial neural network (ANN) model and a novel combined Pareto multi-objective differential evolution (CPMDE), for hydrological simulation and multi-objective numerical optimization of hydropower production, from the Vanderkloof dam in real-time. Results from the application of the real-time strategy, indicate a significant improvement in performance over the current practice. Therefore, the hybrid ANN-CPMDE real-time reservoir operation model suggested herein provides a low cost solution methodology, suitable for sustainable operation of the Vanderkloof reservoir in South Africa.

Impact of Regional Climate Change on Freshwater Resources and Operation of the Vanderkloof Dam System in South Africa

© 2012 Olofintoye et al., licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Impact of Regional Climate Change on Freshwater Resources and Operation of the Vanderkloof Dam System in South Africa

Evolutionary Algorithms and Water Resources Optimization

Heuristic optimization models with varying degree of complexity have been widely applied for resolving water resources optimization and allocation problems. Nevertheless, there still exist uncertainties about finding a generally consistent and trustworthy method that can find solutions, which are really close to the global optimum in all circumstances. This paper makes a review of some of the numerous evolutionary optimization algorithms available to water resources planners and managers. Evolutionary algorithms have been found propitious and useful in facilitating critical water management decisions and are becoming promising global optimization tools for major real world applications. Further research aimed at developing optimization models for water resources planning, management and optimization is therefore necessary.