Power Transformer Overload and Vibrational Stress Estimation Using Nonintrusive Magnetic Flux Leakage Measurements

Power transformers, essential to electrical grids, face reliability challenges from overload and vibrational stress. Conventional monitoring methods often involve invasive modification and high costs, hindering real-time assessment. This study proposes a non-intrusive approach using magnetic flux leakage measurements to evaluate transformer condition. By summing over all the product of single-cycle leakage magnetic flux density (LMFD) and sampling time, a multidimensional metric is derived and processed via the least squares method (LSM). Validation is conducted through finite element method (FEM) simulations of a 110 kV oil-immersed transformer. Sensor placement, optimized based on these simulations, enables precise measurements of overload and vibration, with Pearson correlation coefficients of 0.99983 and 0.99625, respectively. Analysis further identifies critical stress zones—clamp bases, core intersections, and top displacement areas—enhancing damage prediction. Employing micro-magnetic sensors, this cost-effective and safe approach surpasses invasive alternatives, enabling real-time grid monitoring. It establishes a basis for predictive maintenance and future sensor network integration, advancing smart grid technologies.