Contact
Teaching Summary
TA - Machine Learning (COMP3009 UNUK AUT1) (COMP4139 UNUK AUT1) (23-24)
Research Summary
The escalating environmental challenges faced by the industrial sector, particularly concerning environmental sustainability, are principally due to increasing energy demands. Energy efficiency, a… read more
Current Research
The escalating environmental challenges faced by the industrial sector, particularly concerning environmental sustainability, are principally due to increasing energy demands. Energy efficiency, a vital factor in mitigating these challenges, hinges on the ability to predict energy consumption effectively. As of 2016, energy use was responsible for approximately 75% of global CO2 emissions, with the industrial sector contributing around 30% to this total. Specifically, energy usage in the industrial sector constituted 24.2% of the emissions, signifying the most significant source. In this context, businesses can accrue financial benefits from reducing CO2 equivalent emissions, given the high energy consumption. Such reductions necessitate improvements in energy efficiency to circumvent regulatory penalties, lessen environmental impact, and bolster competitiveness. Therefore, industrial enterprises must accurately estimate their energy usage.
Projections by the Food and Agriculture Organization (FAO) indicate that while food production needs to surge by 60% by 2050, energy output will increase by a meagre 33%. This forecast highlights the imminent and widening discrepancy between energy and food production, reinforcing the importance of energy and resource conservation in the food industry. Ensuring food security, meeting governmental emissions reduction targets, and maintaining the sector's profitability all underscore the necessity of boosting energy efficiency in the food industry. To this end, advanced technologies, particularly Machine Learning (ML) - a subfield of Artificial Intelligence (AI) - could be instrumental. ML can elucidate relationships between environmental performance and process characteristics, thereby aiding in monitoring.
Despite the influx of large data volumes from the contemporary industrial environment, most businesses have not harnessed the full potential of this data. The growing availability of data through Industrial Internet of Things Technologies (IIoT) and the evolving processing capabilities of cloud computing promote the use of data-driven models in manufacturing. However, the application of ML, despite its notable problem-solving and dimensionality capabilities, remains underrepresented in research on demand-side energy forecasting.
This research, thus, aims to investigate the use of ML techniques to enhance energy efficiency in food manufacturing systems.
Keywords: Machine Learning, Deep Learning, Energy Efficiency, Energy Consumption Prediction, Food and Drinks Manufacturing Systems, Sustainability.