We study the optimisation of a biomass waste to energy conversion system using an adapted Tabu Search heuristic. It corresponds to a non-linear and non-convex optimisation problem whose solution involves several optimisation sub-problems, including three with differential equations. In solving this complex optimisation problem, four contributions have been made to the adaptation of Tabu Search for use in the optimisation of biomass energy conversion systems. These are: multi-period and diversification strategies that lead to an effective search of the solution space, handling of constraints by development of different strategies for searching feasible regions, with some incursions into infeasible regions to find a shortcut towards feasible regions, and evaluation of a multi-objective function exploiting an approximation of the Pareto front. The results of the experiments show that the resulting Tabu Search heuristic, gives better solutions for this type of optimisation problem, compared to the basic Tabu Search. The developed Tabu Search was used to maximize revenue from biomass waste to energy conversion systems for two types of livestock (cows and swines). The Tabu Search was also used to identify the minimum herd size required for commercial viability of a biomass waste to energy conversion system. Experiments show that the adapted Tabu Search corresponds to a very useful tool for determination of commercial viability of biomass waste to energy conversion systems.
Paru en août 2012 , 32 pages