Energy assessment and scheduling for energy optimisation of a hot dip galvanising process

Abstract

The dearth of energy sustainability is posing major challenges both locally and glob- ally. Galvanising furnaces are categorised as dominant consumers of electricity in the overall galvanising industry. Relatively little research has been carried out concerning energy optimisation through sequencing or scheduling algorithms by way of enhancing the performance of galvanising lines. In this regard, the research centres on evaluating overall energy performance in this industry. The research sought to introduce an opti- mal energy optimisation-scheduling algorithm for a hot dip galvanising process. A DMAIC based methodology was presented for the provisioning of a structured prob- lem-solving process for improving energy efficiency in a galvanising process. Its framework embraces an energy sustainability assessment of four batch hot-dip galva- nising plants. Four energy minimisation opportunities were identified and quantifiable energy and cost savings, as well as avoided carbon dioxide emissions were derived from the analysis of one of the plants. Production or zinc used was identified as the main driver for electricity consumption for Plant 1, while the number of dips per month, amount of zinc used, and ambient temperature conditions were identified as the rele- vant variables for developing a regression model for Plant 2. The amount of zinc used and ambient temperature conditions were found to be the relevant variables for Plant 3. The derived regression model for Plant 4 was based on the amount of zinc used and ambient temperature conditions. The energy performance indicators for a galvanising plant were established through a comparison of actual and expected consumption, energy intensity index, cumulative sum, and specific energy consumption. A bi-objective GECOS algorithm was further introduced to reduce the total energy consumption as well as makespan. The simula- tion results revealed that the GECOS algorithm outperforms McNaughton’s algorithm, Shortest Processing Time Algorithm, and Integer Linear Programming algorithms on minimising makespan on parallel processing machines. The key contributions to the body of knowledge from the study include a unique eval- uation of electrical energy consumption by a hot-dip galvanising plant, development of an energy consumption baseline and performance indices, and the developed novel bi-objective GECOS algorithm that considers reducing total energy consumption by the process tanks as well as makespan. Future research work may focus on hybrid genetic algorithm-artificial immune system scheduling tools that would derive synergy from the advantages of both algorithms to improve energy performance.