Data Driven Transformer Thermal Model for Condition Monitoring

Abstract

Condition monitoring of power transformers, which are key components of electrical power systems, is essential to identify incipient faults and avoid catastrophic failures. In this paper machine learning algorithms, i.e. nonlinear autoregressive neural networks and support vector machines, are proposed to model the transformer thermal behavior for the purpose of monitoring. The thermal models are developed based on the historical measurements from nine transformers comprised of two 180-MVA units, four 240-MVA units and three 1000-MVA units. The data consist of load profile, tap position, winding indicator temperature (WTI) measurement, ambient temperature, wind speed and solar radiation. The results are validated against field measurements, and it is clearly demonstrated that the alternative algorithms surpass the IEEE Annex G thermal model. An incipient thermal fault identification algorithm is then proposed and successfully used to identify an issue using measurements taken in the field. This algorithm could be used to alert the operator and plan intervention accordingly.

Big data analysis method based on time series temperature field data of substation transformer

“…Moreover, based on digital twin 3D modeling, the available monitoring distance can be calculated through spatial location, and the mapping association can be performed according to the accurate pose, so as to realize data and equipment physical location registration and matching. The remote sensing data values can be automatically corrected according to other information such as material, etc., and high-accuracy data acquisition and mapping can be performed [5][6] .…”

“…If the operating temperature of the transformer exceeds the allowable range, the insulation of the transformer will be damaged [2][3][4], The rise of temperature reduces the withstand voltage and mechanical strength of insulating materials. According to the IEC354 Guide for Operating Load of Transformer, when the hottest temperature of the transformer reaches 140℃, bubbles will be generated in the oil, which will reduce the insulation or cause flashover and damage the transformer [5] . Due to the large volume of the transformer, there are many reasons for the abnormal temperature of the transformer.…”

Transformer hot spot temperature estimation through adaptive neuro fuzzy inference system approach

“…As the data related to the transformer thermal behavior are progressively accessible, it is possible to formulate a transformer thermal model based on soft computing techniques regardless of thermal attributes. Power transformer estimation and prediction thermal models based on the available data were developed through exploitation of artificial neural network (ANN) and the obtained results were superior to the one obtained through conventional approaches, but well comparable to the practical-measured data [ [22] , [23] , [24] , [25] ]. In addition, the support vector machine (SVM) [ 26 ], fuzzy [ 27 , 28 ] fuzzy-NN [ 29 ], were investigated in estimating the hot-spot temperature for a power transformer.…”

Real‐time monitoring of transformer hot‐spot temperature based on nameplate data

“…Fuzzy intelligence method and neural networks have been employed for thermal modelling and real-time monitoring of transformer. These models rely on the data-driven black-box method that requires a large data set with long time series for both model development and testing [11,12].…”