Revolutionizing predictive maintenance: Remaining useful life forecasting based on cognitive artificial intelligence using knowledge growing system

Electricity is vital in modern life, and rising demand makes power transformers key parts of energy distribution systems. In many developing areas, old transformers are checked manually and infrequently, speeding up damage and reducing their lifespan. Limited data availability, complex models, and monitoring difficulties further reduce the effectiveness of traditional prediction techniques. According to IEC 60076–7 (2018), transformers typically have a Remaining Useful Life (RUL) of over 20.55 years. This study introduces an innovative approach using the Knowledge Growing System (KGS), a cognitive artificial intelligence (CAI) method that simulates the human brain’s ability to develop knowledge from minimal data, to address these challenges. The findings reveal that the KGS can accurately estimate the RUL of transformers, showing health conditions of 87.5 % and 75 % in Semesters 1 and 2, respectively, which surpasses standard projections. In practice, this research strengthens preventive maintenance, resource management, investment planning, system reliability, and operational risk reduction.