Power systems are fundamental infrastructure world-wide and their safe and reliable operations are critical. However, faults and failures are inevitable in power systems. Nevertheless, these should be avoided as much as possible.
When a fault or failure occurs, the real-time supervisory control and data acquisition system (SCADA) usually presents this information as alarms or indications at the control centre. Quite often, the resulting alarm data are reviewed manually with the help of some automated system components to identify the contributing factors associated with the event.
Such a process may introduce some inaccurate or incorrect assessment, leading to inappropriate decisions. Our research problem is to automate this process incorporating with machine learning from big data for root cause analysis of power system faults and failures.
Research activities include:
- design and implementation of computing and management platforms of big data for fault analysis
- development of data mining and machine learning methods appropriate for data information available in an industrial environment
- design of suitable algorithms to implement these methods.
Expected outcomes are computing and management platforms of big data for fault analysis, data mining and machine learning methods for fault analysis from big data and algorithms that implement these methods.
Skills and experience
The candidate is assumed to have good understanding of, and some experience in:
- data mining
- machine learning
- big data computing.
Skills in programming for data analysis is essential.
Knowledge in power systems are desirable but not essential.
You may be able to apply for a research scholarship in our annual scholarship round.
Contact the supervisor for more information.