This paper proposes a semi-supervised autoencoder with an auxiliary task (SAAT) to extract a health feature space for power transformer fault diagnosis using dissolved gas analysis (DGA). The health feature space generated by a semi-supervised autoencoder (SSAE) not only identifies normal and thermal/electrical fault types, but also presents the underlying characteristics of DGA. In the proposed approach, by adding an auxiliary task that detects normal and fault states in the loss function of SSAE, the health feature space additionally enables visualization of health degradation properties. The overall procedure of the new approach includes three key steps: 1) preprocessing DGA data, 2) extracting two health features via SAAT, and 3) visualizing the two health features in two-dimensional space. In this paper, we test the proposed approach using massive unlabeled/labeled Korea Electric Power Corporation (KEPCO) databases and IEC TC 10 databases. To demonstrate the effectiveness of the proposed approach, four comparative studies are conducted with these datasets; the studies examined: 1) the effectiveness of an auxiliary detection task, 2) the effectiveness of the visualization method, 3) conventional fault diagnosis methods, and 4) the state-of-the-art, semi-supervised deep learning algorithms. By examining several evaluation metrics, these comparative studies confirm that the proposed approach outperforms SSAE without the auxiliary task, existing methods, and state-of-the-art deep learning algorithms, in terms of defining health degradation performance. We expect that the proposed SAAT-based health feature space approach will be widely applicable to intuitively monitor the health state of power transformers in the real world.
In the 2015 PHM Data Challenge Competition, the goal of the competition problem was to diagnose failure of industrial plant systems using incomplete data. The available data consisted of sensor measurements, control reference signals, and fault logs. A detailed description of the plant system of interest was not revealed, and partial fault logs were eliminated from the dataset. This paper presents a fault log recovery method using a machine-learning-based fault classification approach for failure diagnosis. For optimal performance, it was critical to be able to utilize a set of incomplete data and to select relevant features. First, physical interpretation of the given data was performed to select proper features for a fault classifier. Second, Fisher discriminant analysis (FDA) was employed to minimize the effect of outliers in the incomplete data sets. Finally, the type of the missing fault logs and the duration of the corresponding faults were recovered. The proposed approach, based on the use of an incomplete-data-trained FDA classifier, led to the second-highest score in the 2015 PHM Data Challenge Competition.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.