Search

An Anomaly Detection (AD) System for Self-diagnosis has been developed for Multiphase Flow Meter (MPFM). The system relies on machine learning algorithms for time series forecasting, historical data have been used to train a model and to predict the behavior of a sensor and, thus, to detect anomalies.

The detection of anomalous behaviours is an emerging need in many applications, particularly in contexts where security and reliability are critical aspects. While the definition of anomaly strictly depends on the domain framework, it is often impractical or too time consuming to obtain a fully labelled dataset. The use of unsupervised models to overcome the lack of labels often fails to catch domain specific anomalies as they rely on general definitions of outlier. This paper suggests a new active learning based approach, ALIF, to solve this problem by reducing the number of required labels and tuning the detector towards the definition of anomaly provided by the user. The proposed approach is particularly appealing in the presence of a Decision Support System (DSS), a case that is increasingly popular in real-world scenarios. While it is common that DSS embedded with anomaly detection capabilities rely on unsupervised models, they don't have a way to improve their performance: ALIF is able to enhance the capabilities of DSS by exploiting the user feedback during common operations. ALIF is a lightweight modification of the popular Isolation Forest that proved superior performances with respect to other state-of-art algorithms in a multitude of real anomaly detection datasets., Comment: 22 pages, 7 figures

Unsupervised anomaly detection tackles the problem of finding anomalies inside datasets without the labels availability; since data tagging is typically hard or expensive to obtain, such approaches have seen huge applicability in recent years. In this context, Isolation Forest is a popular algorithm able to define an anomaly score by means of an ensemble of peculiar trees called isolation trees. These are built using a random partitioning procedure that is extremely fast and cheap to train. However, we find that the standard algorithm might be improved in terms of memory requirements, latency and performances; this is of particular importance in low resources scenarios and in TinyML implementations on ultra-constrained microprocessors. Moreover, Anomaly Detection approaches currently do not take advantage of weak supervisions: being typically consumed in Decision Support Systems, feedback from the users, even if rare, can be a valuable source of information that is currently unexplored. Beside showing iForest training limitations, we propose here TiWS-iForest, an approach that, by leveraging weak supervision is able to reduce Isolation Forest complexity and to enhance detection performances. We showed the effectiveness of TiWS-iForest on real word datasets and we share the code in a public repository to enhance reproducibility.

Large Scale Particle Image Velocimetry (LSPIV) is widely recognized as a reliable method to measure water surface velocity field in open channels and rivers. LSPIV technique is based on a camera view that frames the water surface in a sequence, and image-processing methods to compute water surface displacements between consecutive frames. Using LSPIV, high flow velocities, as for example flood conditions, were accurately measured, whereas determinations of low flow velocities is more challenging, especially in absence of floating seeding transported by the flow velocity. In fact, in unseeded conditions, typical surface features dynamics must be taken into account: besides surface structures convected by the current, capillary-gravity waves travel in all directions, with their own dynamics. Discrimination between all these phenomena is here discussed, providing a new method to distinguish and to correct unseeded LSPIV measurements associated with wavy structures, accounting for their phase speed magnitude and direction. This has been done measuring wavenumber vectors by using the same images exploited for PIV analysis. All measurements are performed without any flow seeding and in total absence of suspended materials, using the specular reflection of the water surface as a key marker. Results obtained at low-flow regimes in a straight concrete-made rectangular-section channel and in a river are satisfying, especially if compared to those obtained from classic LSPIV application without discrimination and correction. Moreover, a novel simple and safe procedure to orthorectify images is here presented.