AI-powered remote monitoring for refrigeration system

• Main Author: Quek, Zhe Yu
• Other Authors: Kong Wai-Kin Adams
• Format: Final Year Project
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering::Computer science and engineering
• Online Access: http://hdl.handle.net/10356/73943
• Institution: Nanyang Technological University

Engineering has been prevalent in the society ever since the industrial revolution, where machines and systems seamlessly become part and parcel of our lives. In this contemporary society, the integration is so widely spread to the extent that advanced engineered systems and technology can be seen almost everywhere. Such systems are intelligently designed and engineered to cater to our ever-changing needs, be it commercial or personal. In business perspective of an engineering firm in the Heating, Ventilation and Air Conditioning (HVAC) industry, the goals would usually be on ensuring reliability, increasing production, functionalities and lowering cost of production. Improving efficiency, on the other hand, is easily undermined in this industry. So long as the capacity requirement can be met, efficiency level is frequently not seen as an important concern. One of the major lack of incentives to improve efficiency level is the limitations in recognising the efficiency level of a system, especially so when there is no standard mechanism of measurement. However, in the current society, as business processes are progressively more developed and organised, production rates are saturated and functional requirements of HVAC systems are mostly met. Hence, businesses are starting to shift their focus to achieving higher level requirements to obtain competitive advantage in the industry. Efficiency level has been increasingly seen as one of the important measures, mainly because it has direct impact on the electricity usage and hence is reflected greatly on the sustainable recurring cost of electricity. With the development of Science and Technology, new and innovative performance measures have been found to measure the efficiency level of systems. Such measures are usually via a single unit of measurement or through Science, with the use of scientific formulas proven by different researches and experiments. Nonetheless, with the limitation of Science and Technology, such measures are costly and inaccurate. They usually act only as a rough estimation of the true efficiency, and are based on many assumptions. However, with the improvements in advanced technologies, specifically artificial intelligence (A.I.), a different approach can be taken to provide a better scoring system. By leveraging on the power of data, it is now possible to estimate the true efficiency more accurately at a much lower cost. This finding is extremely exciting because the incorporation of advanced computer technologies into mechanical systems are rare, due to the complications in both disciplines. However, due to personal expertise and resources available, this rare opportunity is presented, and it provides a perfect platform to research and develop an innovative approach of performance measure in this project. In hope to explore the potential of this project, such performance measures, if successful, can be applied to any mechanical system where data can be collected. It can provide a more effective measure for efficiency level and will be a breakthrough in the search of maximising system efficiency which ultimately reduce wastage and cost.