Predicting the service life of high-voltage insulators using actual leakage current values

Evaluation method for insulation degradation of power transformer windings based on incomplete internet of things sensing data

“…Degradation evaluation methods relying on traditional indicators [4] Proposed a method based on spectral analysis and pattern recognition using furfural content as an index Cannot be applied to online deterioration evaluation [5] Investigated a method based on the methanol content [6] Explored the correlation between water content in oil data and transformer degradation [7] Developed a model based on feedforward neural networks using the degree of polymerization as an indicator [8] Studied the impact of electrical conductivity on insulation aging Degradation evaluation methods relying on a single type of IoT data [10] Proposed a method based on statistical indices of partial discharge Lack of consideration for multiple degradation factors and data incompleteness [11,12] Proposed two methods to eliminate partial discharge in transformers by preparing nanofluid to absorb gases such as acetylene in oil [13] Proposed a degradation prediction model based on temperature data [14] Investigated the method based on leakage current [15] Studied the electrical damage trend by incrementally increasing the voltage Consider multiple degradation factors but rely on traditional indicators [17] Constructed a dynamic model under the influence of electrical and thermal stress Cannot be applied to online deterioration evaluation [18] Explored the changing trends of indicators under thermal and mechanical stresses [19] Investigated the trends of the indicators under electrical, thermal and mechanical stresses [20] Studied the method based on tensile strength and dielectric constant under thermal-mechanical stresses Data completion for a single type of IoT sensing data [21] Completed voltage data using deep learning and unscented Kalman filtering Lack of consideration of spatiotemporal correlation between multiple IoT sensing data [22] Investigated data filling method in photovoltaic power using recursive long short-term memory network [23] Used the normal distribution method for filling power data of smart meters [24] Proposed a filling method for household load data based on noisy interpolation Abbreviation: IoT, Internet of Things.…”

“…Aizpurua et al developed a probabilistic deterioration evaluation and life prediction model for transformer winding insulation using temperature data [13]. Valeriy et al explored a degradation prediction method based on the leakage current of insulation materials [14]. Qinghong investigated the electrical degradation damage trend of insulating materials by incrementally increasing the voltage [15].…”

“…Valeriy et al. explored a degradation prediction method based on the leakage current of insulation materials [14]. Qinghong investigated the electrical degradation damage trend of insulating materials by incrementally increasing the voltage [15].…”