An Improved Plant-wide Hierarchical Optimization Method Based on Technical Indices Decomposition

Basing on the minimum material consumption models, the plant-wide hierarchical optimization method has been successful in determining technical indices and manipulated variables by solving a two-level optimization problem, including the procedure-level optimization and the process-level optimization. However, the performance of the method will deteriorate if there are numerous technical indices and coupling relationship between them in the procedure-level optimization problem. To address this problem efficiently, this paper presents an improved plant-wide hierarchical optimization method based on technical indices decomposition. In the method, the procedure-level optimization problem is decomposed into several easier solvable sub-problems based on technical indices grouping. Two metrics are defined to measure whether the technical indices is divided into groups or separate. Then, the global optimum of procedure-level optimization problem is found by successively solving the sub-problems. Finally, the presented optimization method is applied to the hydrometallurgy process of a gold smelter. The results verify the effectiveness of this method.