Search

Vibration-based condition monitoring systems are receiving increasing attention due to their ability to accurately identify different conditions by capturing dynamic features over a broad frequency range. However, there is little research on clustering approaches in vibration data and the resulting solutions are often optimized for a single data set. In this work, we present an extensive comparison of the clustering algorithms K-means clustering, OPTICS, and Gaussian mixture model clustering (GMM) applied to statistical features extracted from the time and frequency domains of vibration data sets. Furthermore, we investigate the influence of feature combinations, feature selection using principal component analysis (PCA), and the specified number of clusters on the performance of the clustering algorithms. We conducted this comparison in terms of a grid search using three different benchmark data sets. Our work showed that averaging (Mean, Median) and variance-based features (Standard Deviation, Interquartile Range) performed significantly better than shape-based features (Skewness, Kurtosis). In addition, K-means outperformed GMM slightly for these data sets, whereas OPTICS performed significantly worse. We were also able to show that feature combinations as well as PCA feature selection did not result in any significant performance improvements. With an increase in the specified number of clusters, clustering algorithms performed better, although there were some specific algorithmic restrictions., Comment: 12 pages, 10 figures, Proceedings of the 5th International Data Science Conference iDSC2023

Data-driven machine learning is playing a crucial role in the advancements of Industry 4.0, specifically in enhancing predictive maintenance and quality inspection. Federated learning (FL) enables multiple participants to develop a machine learning model without compromising the privacy and confidentiality of their data. In this paper, we evaluate the performance of different FL aggregation methods and compare them to central and local training approaches. Our study is based on four datasets with varying data distributions. The results indicate that the performance of FL is highly dependent on the data and its distribution among clients. In some scenarios, FL can be an effective alternative to traditional central or local training methods. Additionally, we introduce a new federated learning dataset from a real-world quality inspection setting.

Anomaly detection in large industrial cooling systems is very challenging due to the high data dimensionality, inconsistent sensor recordings, and lack of labels. The state of the art for automated anomaly detection in these systems typically relies on expert knowledge and thresholds. However, data is viewed isolated and complex, multivariate relationships are neglected. In this work, we present an autoencoder based end-to-end workflow for anomaly detection suitable for multivariate time series data in large industrial cooling systems, including explained fault localization and root cause analysis based on expert knowledge. We identify system failures using a threshold on the total reconstruction error (autoencoder reconstruction error including all sensor signals). For fault localization, we compute the individual reconstruction error (autoencoder reconstruction error for each sensor signal) allowing us to identify the signals that contribute most to the total reconstruction error. Expert knowledge is provided via look-up table enabling root-cause analysis and assignment to the affected subsystem. We demonstrated our findings in a cooling system unit including 34 sensors over a 8-months time period using 4-fold cross validation approaches and automatically created labels based on thresholds provided by domain experts. Using 4-fold cross validation, we reached a F1-score of 0.56, whereas the autoencoder results showed a higher consistency score (CS of 0.92) compared to the automatically created labels (CS of 0.62) -- indicating that the anomaly is recognized in a very stable manner. The main anomaly was found by the autoencoder and automatically created labels and was also recorded in the log files. Further, the explained fault localization highlighted the most affected component for the main anomaly in a very consistent manner., Comment: accepted at phme 2022

We propose a new sampling strategy, called smart active sapling, for quality inspections outside the production line. Based on the principles of active learning a machine learning model decides which samples are sent to quality inspection. On the one hand, this minimizes the production of scrap parts due to earlier detection of quality violations. On the other hand, quality inspection costs are reduced for smooth operation.

Federated Learning (FL) decouples model training from the need for direct access to the data and allows organizations to collaborate with industry partners to reach a satisfying level of performance without sharing vulnerable business information. The performance of a machine learning algorithm is highly sensitive to the choice of its hyperparameters. In an FL setting, hyperparameter optimization poses new challenges. In this work, we investigated the impact of different hyperparameter optimization approaches in an FL system. In an effort to reduce communication costs, a critical bottleneck in FL, we investigated a local hyperparameter optimization approach that -- in contrast to a global hyperparameter optimization approach -- allows every client to have its own hyperparameter configuration. We implemented these approaches based on grid search and Bayesian optimization and evaluated the algorithms on the MNIST data set using an i.i.d. partition and on an Internet of Things (IoT) sensor based industrial data set using a non-i.i.d. partition., Comment: This paper is accepted at the IDSC https://idsc.at/ and will be published by Springer. The Version uploaded is before the peer review process. The link to the final version will be updated as soon as the paper is published. Figure one was corrected on 2021/10/20

The increasing deployment of low-cost IoT sensor platforms in industry boosts the demand for anomaly detection solutions that fulfill two key requirements: minimal configuration effort and easy transferability across equipment. Recent advances in deep learning, especially long-short-term memory (LSTM) and autoencoders, offer promising methods for detecting anomalies in sensor data recordings. We compared autoencoders with various architectures such as deep neural networks (DNN), LSTMs and convolutional neural networks (CNN) using a simple benchmark dataset, which we generated by operating a peristaltic pump under various operating conditions and inducing anomalies manually. Our preliminary results indicate that a single model can detect anomalies under various operating conditions on a four-dimensional data set without any specific feature engineering for each operating condition. We consider this work as being the first step towards a generic anomaly detection method, which is applicable for a wide range of industrial equipment., Comment: This paper is accepted at the Industrial Track IDSC https://idsc.at/. The link to the publication and final version will follow as so the paper is published by Springer

The increasing deployment of low-cost industrial IoT (IIoT) sensor platforms on industrial assets enables great opportunities for anomaly classification in industrial plants. The performance of such a classification model depends highly on the available training data. Models perform well when the training data comes from the same machine. However, as soon as the machine is changed, repaired, or put into operation in a different environment, the prediction often fails. For this reason, we investigate whether it is feasible to have a robust and transferable method for AI based anomaly classification using different models and pre-processing steps on centrifugal pumps which are dismantled and put back into operation in the same as well as in different environments. Further, we investigate the model performance on different pumps from the same type compared to those from the training data., Comment: This paper is accepted at https://idsc.at/. The final authenticated version is available online and this information will be updated

The $^{239}$Pu(n,f)/$^{235}$U(n,f) cross-section ratio has been measured with the fission Time Projection Chamber (fissionTPC) from 100 keV to 100 MeV. The fissionTPC provides three-dimensional reconstruction of fission-fragment ionization profiles, allowing for a precise quantification of measurement uncertainties. The measurement was performed at the Los Alamos Neutron Science Center which provides a pulsed white source of neutrons. The data are recommended to be used as a cross-section ratio shape. A discussion of the status of the absolute normalization and comparisons to ENDF evaluations and previous measurements is included.

We present the results of a measurement of isotopic concentrations and atomic number ratio of a double-sided actinide target with alpha-spectroscopy and mass spectrometry. The double-sided actinide target, with primarily Pu-239 on one side and U-235 on the other, was used in the fission Time Projection Chamber (fissionTPC) for a measurement of the neutron-induced fission cross-section ratio between the two isotopes. The measured atomic number ratio is intended to provide an absolute normalization of the measured fission cross-section ratio. The Pu-239/U-235 atom number ratio was measured with a combination of mass spectrometry and alpha-spectroscopy with a planar silicon detector with uncertainties of less than 1%., Comment: 8 pages, 5 figures, 6 tables

Federated Learning (FL) is a very promising approach for improving decentralized Machine Learning (ML) models by exchanging knowledge between participating clients without revealing private data. Nevertheless, FL is still not tailored to the industrial context as strong data similarity is assumed for all FL tasks. This is rarely the case in industrial machine data with variations in machine type, operational- and environmental conditions. Therefore, we introduce an Industrial Federated Learning (IFL) system supporting knowledge exchange in continuously evaluated and updated FL cohorts of learning tasks with sufficient data similarity. This enables optimal collaboration of business partners in common ML problems, prevents negative knowledge transfer, and ensures resource optimization of involved edge devices., Comment: 12 pages, accepted for https://www.paams.net/workshops/agedai

The Neutron Induced Fission Fragment Tracking Experiment (NIFFTE) collaboration has performed measurements with a fission time projection chamber (fissionTPC) to study the fission process by reconstructing full three-dimensional tracks of fission fragments and other ionizing radiation. The amount of linear momentum imparted to the fissioning nucleus by the incident neutron can be inferred by measuring the opening angle between the fission fragments. Using this measured linear momentum, fission fragment angular distributions can be converted to the center-of-mass frame for anisotropy measurements. Angular anisotropy is an important experimental observable for understanding the quantum mechanical state of the fissioning nucleus and vital to determining detection efficiency for cross section measurements. Neutron linear momentum transfer to fissioning $^{235}$U, $^{238}$U, and $^{239}$Pu and fission fragment angular anisotropy of $^{235}$U and $^{238}$U as a function of neutron energies in the range 130 keV--250 MeV are presented.

We provide a quantitative description of a method to measure neutron-induced fission cross sections in ratio to elastic hydrogen scattering in a white-source neutron beam with the fission Time Projection Chamber. This detector has measured precision fission cross section ratios using actinide references such as $^{235}$U(n,f) and $^{238}$U(n,f). However, by employing a more precise reference such as the H(n,el) cross section there is the potential to further reduce the evaluation uncertainties of the measured cross sections. In principle the fissionTPC could provide a unique measurement by simultaneously measuring both fission fragments and proton recoils over a large solid angle. We investigate one method with a hydrogenous gas target and with the neutron energy determined by the proton recoil kinematics. This method enables the measurement to be performed in a white-source neutron beam and with the current configuration of the fissionTPC. We show that while such a measurement is feasible in the energy range of 0.5 MeV to $\sim$10 MeV, uncertainties on the proton detection efficiency and the neutron energy resolution do not allow us to preform a fission ratio measurement to the desired precision. Utilizing either a direct measurement of the neutron time-of-flight for the recoil proton or a mono-energetic neutron source or some combination of both would provide a path to a sub-percent precision measurement.

Accurate segmentation of medical images is an important step towards analyzing and tracking disease related morphological alterations in the anatomy. Convolutional neural networks (CNNs) have recently emerged as a powerful tool for many segmentation tasks in medical imaging. The performance of CNNs strongly depends on the size of the training data and combining data from different sources is an effective strategy for obtaining larger training datasets. However, this is often challenged by heterogeneous labeling of the datasets. For instance, one of the dataset may be missing labels or a number of labels may have been combined into a super label. In this work we propose a cost function which allows integration of multiple datasets with heterogeneous label subsets into a joint training. We evaluated the performance of this strategy on thigh MR and a cardiac MR datasets in which we artificially merged labels for half of the data. We found the proposed cost function substantially outperforms a naive masking approach, obtaining results very close to using the full annotations., Comment: Accepted for presentation at 9th International Conference on Machine Learning in Medical Imaging (MLMI 2018)

The University of Wisconsin System and the Wisconsin Department of Public Instruction have been involved in a multi-year program since 1998. One goal of the program has been to align curricula within the K-16 system of education in Wisconsin. The purpose of the Curriculum Articulation Project has been to work with educators in the state of Wisconsin to move toward a well articulated educational system that provides students a seamless transition between secondary and post-secondary institutions. The working groups found that it is not the curriculum, but that student expectations, the teacher expectations, and the professor expectations appear to be misaligned. While curricular alignment addresses performance and achievement in terms of content, it was found that study skills, family and friend support, advisement and personal responsibility also play a crucial role in the success of the student. Additionally, for any effort designed to improve student success, not only the curricula and students must be examined, but also the role of the administrators and teaching faculty involved in the process must be addressed. A follow-up study based on criteria previously identified as misaligned was conducted and the results reported. Recommendations for more successful evaluation of curriculum articulation are offered. A data sheet is attached. (Contains 15 references.) (Author/BT)

Discusses the range and limitations of local and regional case studies based on studies of the educational history of Berlin (Germany), Jewish school systems under National Socialism, and women's education. Examines their content-related results as well as their systematic significance for research on the history of education (DSK)

Argues that technical communicators must be prepared to communicate via multiple media, including video. Describes a technical scriptwriting course that develops skills useful in several media. Discusses pedagogical strategies, three major assignments, reading assignments, client assignments, production diary assignment, and new technology. (SR)