Malfunction Detection of the Cooling System in Air-Forced Power Transformers Using Online Thermal Monitoring

In order to make the best use of the existing assets, monitoring of power transformers provides the ability to continuously assess the overload capability of transformers. This can be more economically advantageous than the extension of the network by installing a new transformer. During the overloading periods, the cooling system of a power transformer plays a decisive role to transfer the heat generated in the windings to the surrounding ambient. The faulty operation of the cooling system should be detected by an online monitoring system before the transformer is overloaded. In this paper, an online algorithm is presented for malfunction detection of the cooling system based on the calculation of the standardized error in the calculated top-oil temperatures. Moreover, for calculation of the top-oil temperature, a new thermal model is proposed which uses a few design-dependent variables and considers different heat-transfer modes inside a transformer. For validation of the algorithm, the measured data during the normal operation of three transformers are used-two of them experienced failures in a part of their cooling system. The proposed model and algorithm can be easily implemented and integrated into monitoring systems due to their simplicity and good accuracy.