In the context of freshwater stress and scarcity, industries use and produce large volume of water with high toxicity. The challenge is to find creative ways to minimise freshwater use with emphasis on protecting the environment. One of such creative ways is Water application Planning (WAP), which employs recycle, regeneration, and reuse to minimise freshwater consumption. In the use of mathematical optimisation methods for solving WAP problems, five major challenges were identified. They are lack of single systematic method to handle reuse/recycle of single and multi-contaminant, a hybrid targeting method which can address multiple local optimal solutions, a single model that provides simultaneous optimisation of the interactions of multiple water-using units and regeneration (wastewater treatment technologies), robust flexible approaches that enables faster numerical solutions and handle single and multi-objective cases of WAP problems. This research developed a framework for WAP optimisation that eliminates all the above challenges without complexity. A Stochastic Optimisation, Genetic Algorithm (GA) improved with a hybrid of the deterministic optimisation, Sequential Quadratic Programming (SQP) algorithm strategy has been used for the analysis of WAP problem in the Matlab software environment to facilitates efficient and effective solution for single objective. It was further developed to handle multi-objective optimisation (MOO) using Multi-objective Genetic Algorithm (MOGA) and improved with the hybrid goal attainment method to increase its performance. Both methods of solution were robust in handling single-contaminant and multi-contaminant WAP problems, with and without regeneration. The time of running WAP model using genetic algorithm (GA) with and without hybrid of deterministic optimisation for different number of industrial processes was investigated. The model runs for the minimum, median and maximum time of 3.0, 5.7 and 16.2 seconds respectively, with the GA population size of 10 for the 2 to 10 number of industrial processes. Furthermore, the result of multi-objective genetic algorithm (MOGA) analysis for WAP problem shows that the model can search for a wide-ranging distribution of Pareto optimal solutions and it has small computational time of 5 seconds for 3 industrial processes. Moreover, a case study of Kaduna Refining and Petrochemical Company Limited (KRPC) refinery was considered in the analysis of minimum freshwater requirement. The challenge of the refinery is to minimise the freshwater use and subsequently the wastewater disposal. The result shows that the sum of freshwater flow rates required by six major water using processes in the refinery can be reduced by 11.3% by reusing wastewater and it can be further reduced by 17.3% through regeneration reuse method. The wastewater produced by the refinery also reduced by the above two percentages. These reductions in water consumption and wastewater production exceed those reported in the literature under similar refinery condition. The superstructure was employed in showing the actual network interconnections of different industrial processes for feasible freshwater minimisation analysed. Finally, the recommendation is to combine two stochastic mathematical optimisation methods with two hybrid functions to improve the optimum solution obtained.
|Date of Award||24 Feb 2020|
|Sponsors||Petroleum Technology Development Fund (PTDF)|
|Supervisor||Ruth Falconer (Supervisor) & Joseph Akunna (Supervisor)|