The model of an anomaly detector for HiLumi LHC magnets based on Recurrent Neural Networks and adaptive quantization

Abstract

A B S T R A C TThis paper focuses on an examination of an applicability of Recurrent Neural Network models for detecting anomalous behavior of the CERN superconducting magnets. In order to conduct the experiments, the authors designed and implemented an adaptive signal quantization algorithm and a custom Gated Recurrent Unit-based detector and developed a method for the detector parameters selection.Three different datasets were used for testing the detector. Two artificially generated datasets were used to assess the raw performance of the system whereas the dataset intended for real-life experiments and model training was composed of the signals acquired from a new type of magnet, to be used during High-Luminosity Large Hadron Collider project. Several different setups of the developed anomaly detection system were evaluated and compared with state-of-the-art One Class Support Vector Machine (OC-SVM) reference model operating on the same data. The OC-SVM model was equipped with a rich set of feature extractors accounting for a range of the input signal properties.It was determined in the course of the experiments that the detector, along with its supporting design methodology, reaches F1 equal or very close to 1 for almost all test sets. Due to the profile of the data, the setup with the lowest maximum false anomaly length of the detector turned out to perform the best among all five tested configuration schemes of the detection system. The quantization parameters have the biggest impact on the overall performance of the detector with the best values of input/output grid equal to 16 and 8, respectively. The proposed solution of the detection significantly outperformed OC-SVM-based detector in most of the cases, with much more stable performance across all the datasets.

Using Neural Networks with data Quantization for time Series Analysis in LHC Superconducting Magnets

“…This work extends the existing investigations (Wielgosz et al, 2017;2018a;2018b;2018c; 2020) using higher resolution data and more diverse models. The importance of the subject grows because the project High Luminosity LHC (HL-LHC) enters its engineering phase (Apollinari et al, 2017).…”

“…In order to use it with more specific applications such as anomaly detection, the module is extended with the analyzer section which operates on top of the model. A more detailed explanation of this architecture was given by Wielgosz et al (2018a) and is out of the scope of this paper. Figure 8 shows the results of three different experiments.…”

Protection of Superconducting Industrial Machinery Using RNN-Based Anomaly Detection for Implementation in Smart Sensor

“…In the authors’ previous works concerning superconducting magnets monitoring Root-Mean-Square Error (RMSE) [ 24 ] and both static [ 23 ] and adaptive data quantization [ 1 , 25 ] approaches were used. Based on experiments conducted and described therein, a conclusion can be drawn that RNNs can be used to model magnets behavior and detect anomalous occurrences.…”

“…In Ref. [ 25 ], an approach based on adaptive data quantization and automatic thresholds selection was introduced. Adaptive data quantization resulted in much better use of bins and consequently significantly improved the accuracy results.…”

“…When detecting anomalies in time series, RNNs are both scalable and adaptable which is critical when it comes to design and implementation of complex anomaly detection systems [ 23 , 24 , 25 ]. RNNs were introduced long ago [ 26 ] and have been successfully applied in many domains [ 8 ], according to the authors’ knowledge, there are no studies on the performance of compressed RNNs in anomaly detection.…”