Predictive maintenance in electrical power systems: Thermography and statistical methods for phase synchronization analysis in disconnected substations.

Phase synchronization can aid in understanding the dynamics of underlying systems. In an electric substation, disconnect, failures mostly occur due to overheating. In this work using the Hilbert transform to assess the phase synchrony of a disconnect temperature and its relationship with meteorological variables and disconnect current can be helpful in understanding temperature influences and composition. Statistical methodologies, including entropy analysis, correlation, Kalman filter, Fast Fourier Transform, Augmented Dickey–Fuller test, and Kwiatkowski–Phillips–Schmidt–Shin test, were employed. One of the differentials of this study is the use of the median filter, which was able to access the phase synchrony of the variables with disconnect temperature. An important result is that current is not the most correlated variable with disconnect temperature, instead, air temperature exhibits the highest correlation, and the highest phase synchronization is only obtained when disconnect current and air temperatures are combined. This study contributes to the advancement of power substation operations, advocating for continuous monitoring and analysis to ensure a reliable electricity supply obtaining better predictive maintenance. [Display omitted] • The median filter facilitates access to phase synchronization information. • The combination of air temperature and current demonstrates the highest level of synchronization information. in the disconnect temperature. • Solar incidence decreases synchrony. • The disconnect temperature, disconnect current, and air temperature share the same frequency modules. • Night-time phase synchrony is higher. [ABSTRACT FROM AUTHOR]