Great concern regarding energy efficiency has led the research community to develop approaches which enhance the energy awareness by means of insightful representations. An example of intuitive energy representation is the partsbased representation provided by Non-Intrusive Load Monitoring (NILM) techniques which decompose non-measured individual loads from a single total measurement of the installation, resulting in more detailed information about how the energy is spent along the electrical system. Although there are previous works that have achieved important results on NILM, the majority of the NILM systems were only validated in residential buildings, leaving a niche for the study of energy disaggregation in nonresidential buildings, which present a specific behavior. In this paper, we suggest a novel fully-convolutional denoising autoencoder architecture (FCN-dAE) as a convenient NILM system for large non-residential buildings, and it is compared, in terms of particular aspects of large buildings, to previous denoising auto-encoder approaches (dAE) using real electrical consumption from a hospital facility. Furthermore, by means of three use cases, we show that our approach provides extra helpful funcionalities for energy management tasks in large buildings, such as meter replacement, gap filling or novelty detection.

Rotating machines are critical equipment in many processes, and failures in their operation can have serious implications. Consequently, fault detection in rotating machines has been widely investigated. Conventional detection systems include two blocks: feature extraction and classification. These systems are based on manually engineered features (ball pass frequencies, RMS value, kurtosis, crest factor, etc.) and therefore require a high level of human expertise (it is a human who designs and selects the most appropriate set of features to perform the classification). Instead, we propose a system for condition monitoring and fault detection in rotating machines based on a 1-D deep convolutional neural network (1D DCNN), which merges the tasks of feature extraction and classification into a single learning body. The proposed system has been designed for use on a rotating machine with seven possible operating states and it proves to be able to determine the operating condition of the machine almost as accurately as conventional feature-engineered classifiers, but without the need for prior knowledge of the machine. The proposed system has also reported good classification on a bearing fault dataset from another machine, thus demonstrating its capability to monitor the condition of different machines. Finally, the analysis of the features learned by the deep model has revealed valuable and previously unknown machine information, such as the rotational speed of the machine or the number of balls in the bearings. In this way, our results illustrate not only the good performance of CNNs, but also their versatility and the valuable information they could provide about the monitored machine.

No abstract

No abstract

No abstract