Benefits of applying data analysis and machine learning to advanced metering infrastructure system | |
| Author | Makara Greadmeta |
| Call Number | AIT Thesis no.ET-21-13 |
| Subject(s) | Electric power distribution--Data processing Electric meters--Technological innovations Machine learning |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Energy |
| Publisher | Asian Institute of Technology |
| Abstract | Nowadays, electricity is one of the most important form of final energies and living without it is very difficult to imagine. For electrical energy market, the correct measurement of electricity consumption is essential for both providers and consumers. Therefore, the Advanced Metering Infrastructure system has been developed. It is an integrated system of meters, communications and data management systems that enables two-way communication between utilities and consumers in near real-time. According to these functionalities, the large and diverse datasets will be collected and stored much more than before. Only storing may not be useful so this study proposes several processes to leverage these consumption datasets for both utility and consumer aspects. Data analysis and machine learning have been selected to address all of the objectives by mainly using Python programming language. The results show that every proposed process can successfully provide benefits to both of them. For utility aspect, the developed distribution transformer monitoring system and the introduced meters phase-swapping algorithms can help reducing technical losses. Moreover, non technical losses can also be reduced by the developed electricity theft detection system based on machine learning classification algorithms. For consumer aspect, the proposed tariff changing evaluation process can provide suggestions to consumers individually in order to help reducing their electricity bills. The insights from the created consumption comparisons can also be provided to consumers for having more comprehension and useful information. Furthermore, the abnormal consumption data points can be identified by the developed anomaly detection system based on outlier detection algorithms. Finally, this study also presents the convenient solution to deliver aforementioned benefits to end-users by developing the Web Application. All of the proposed processes and solution by this study are suitable and can be adapted for any energy provider and consumer. |
| Year | 2021 |
| Type | Thesis |
| School | School of Environment, Resources, and Development (SERD) |
| Department | Department of Energy and Climate Change (Former title: Department of Energy, Environment, and Climate Change (DEECC)) |
| Academic Program/FoS | Energy Technology (ET) |
| Chairperson(s) | Singh, Jai Govind |
| Examination Committee(s) | Salam, P. Abdul;Weerakorn Ongsakul |
| Scholarship Donor(s) | PEA-AIT Education Cooperation Project;Royal Thai Government Fellowship |
| Degree | Thesis (M. Eng.) - Asian Institute of Technology, 2021 |