This research area focuses on developing data-driven predictive mathematical models for the early detection of anomalies and degradation in high-voltage oil-filled equipment. The aging of power grid infrastructure is a global challenge, which is exacerbated in Ukraine by the deliberate destruction of energy facilities. The surviving transformers are forced to operate under abnormal conditions, which significantly affects the types and rate of their degradation and increases the likelihood of accidents.
Currently, the diagnostics of oil-filled equipment relies primarily on static, predefined threshold values for insulation parameters. This classification assumes two distinct states—either the equipment is operational or it is not—and it typically identifies failures only after the system has already entered a defective state. The continuous degradation process is not mathematically described until the critical threshold is crossed, which negatively affects the timeliness of identifying problems in the equipment.
Our research will focus on the dynamic analysis of processes in oil-filled equipment, including the determination of the state of physicochemical markers. The main scientific hypothesis is that the onset of a defect or accelerated aging alters the nature of the relationship between performance indicators and operating time long before the absolute values exceed permissible limits.
Key approaches:
- A shift from evaluating threshold values to dynamic modeling of degradation markers.
- Monitoring changes in the time derivatives of insulation parameters and identifying significant systematic components that indicate the initiation of defects.
- Data-driven anomaly detection using a unique dataset to mathematically formalize the dynamics of the transition from a stable to a defective state even before critical limits are exceeded.
- Development of computational models that utilize the identified statistical patterns to accurately predict the remaining service life of high-voltage equipment, taking into account the conditions under which it operates.