Quality Control Procedure Based on Partitioning of NMR Time Series

Staniszewski, Michał; Skorupa, Agnieszka; Boguszewicz, Łukasz; Sokół, Maria; Polański, Andrzej

doi:10.3390/s18030792

Cited by 3 publications

(3 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A large number of approaches have been proposed for application-oriented modelling of time series [7] and quality control [8], [9], including statistical tests, decomposition methods, autoregressive models, neural networks, and probabilistic models. Among them, the Hidden Markov Models (HMMs) [10], [11], [12] have proven to be versatile and effective across many fields, including fault diagnosis [13], [14].…”

Section: Introductionmentioning

confidence: 99%

Continual learning of a time series model using a mixture of HMMs with application to the IoT fuel sensor verification

Głomb,

Cholewa,

Foszner

et al. 2023

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

This paper presents an application of a mixture of Hidden Markov Models (HMMs) as a tool for verification of IoT fuel sensors. The IoT fuel sensors report the level of fuel in tanks of a petrol station, and are a key component for monitoring system reliability (billing), safety (fuel/oil leak detection) and security (theft prevention). We propose an algorithm for learning a mixture of HMMs based on a continual learning principle, i.e. it adapts the model while monitoring a sensor over time, signalling unexpected or anomalous sensor reports. We have tested the proposed approach on a real-life data of 15 fuel tanks being monitored with the FuelPrime system, where it has shown a very good performance (average area under ROC curve of 0.94) of detecting anomalies in the sensor data. Additionally we show that the proposed method can be used for trend monitoring and present qualitative analysis of the short and long term learning performance. The proposed method has promising performance score, the resulting model has a high degree of explainability, limited memory and computation requirements and can be easily generalized to other domains of sensor verification.This work was partially funded by National Center for Research and Development no POIR.01.01.01-00-0376/17-00 'A system for gathering and analysis of streaming data for fuel stations, for optimization of distribution costs and fuel sales as well as on-line monitoring for leakages-related issues' (FuelPrime).

show abstract

Section: Introductionmentioning

confidence: 99%

Continual learning of a time series model using a mixture of HMMs with application to the IoT fuel sensor verification

Głomb,

Cholewa,

Foszner

et al. 2023

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

show abstract

“…A reason to neglect the time course of the axial peak tibial acceleration may be that relevant changes in such a signal are usually not easily discernible by sight. The technique of change-point analysis may be of use to detect event(s) at which the underlying dynamics of a signal changes over time [17][18][19][20][21][22][23]. Several types of control statistics have been used for change-point discovery.…”

Section: Introductionmentioning

confidence: 99%

Change-Point Detection of Peak Tibial Acceleration in Overground Running Retraining

Berghe

Gosseries

Gerlo

et al. 2020

Sensors

View full text Add to dashboard Cite

A method is presented for detecting changes in the axial peak tibial acceleration while adapting to self-discovered lower-impact running. Ten runners with high peak tibial acceleration were equipped with a wearable auditory biofeedback system. They ran on an athletic track without and with real-time auditory biofeedback at the instructed speed of 3.2 m·s −1 . Because inter-subject variation may underline the importance of individualized retraining, a change-point analysis was used for each subject. The tuned change-point application detected major and subtle changes in the time series. No changes were found in the no-biofeedback condition. In the biofeedback condition, a first change in the axial peak tibial acceleration occurred on average after 309 running gait cycles (3 40"). The major change was a mean reduction of 2.45 g which occurred after 699 running gait cycles (8 04") in this group. The time needed to achieve the major reduction varied considerably between subjects. Because of the individualized approach to gait retraining and its relatively quick response due to a strong sensorimotor coupling, we want to highlight the potential of a stand-alone biofeedback system that provides real-time, continuous, and auditory feedback in response to the axial peak tibial acceleration for lower-impact running.

show abstract

“…Thus the change point detection aims at identifying whether the distribution or parameter of the process changes, and the point that causes this abrupt change is the point that needs to be detected [1][2][3]. It is a significant and practical problem in some modern fields, such as quality control [4], sensor signal analysis [5], Network attack detection [6], tracking [7], and so on. In general, the change point detection algorithms assume that it follows one probability distribution before the change point, and it starts to follow another distribution after the change point.…”

Section: Introductionmentioning

confidence: 99%

Change Point Detection with Mean Shift Based on AUC from Symmetric Sliding Windows

et al. 2020

View full text Add to dashboard Cite

Change point detection is widely used in signal detection, industrial engineering, economy, finance, biomedicine and many other fields. The widely used parametric methods require prior knowledge of the noise signal distribution, which are seldom realistic. In practice, when the distribution of noise is not known, it is desirable to design algorithms based on non-parametric statistics, which, in the null case (no change point), are completely distribution free. To this end, we propose to use two symmetric sliding windows to compute the Area Under the receiver operating characteristic Curve (AUC) as a test statistic to measure the difference between the distribution of two samples. In the stage of change point detection, a threshold is designed according to hypothesis test which is based on the null distribution of the test statistics. This threshold is used to detect the potential change points in the signal. To reduce the probability of false alarm detection, a key parameter K is set to distinguish and delete the false alarms in potential change points. Comparative studies showed that our proposed method outperforms the classical Relative unconstrained Least-Squares Importance Fitting (RuLSIF) algorithm and is also better than the Hawkins, Qiu, and Kang (HQK) algorithm when the noise follows non-normal distributions.

show abstract

Quality Control Procedure Based on Partitioning of NMR Time Series

Cited by 3 publications

References 30 publications

Continual learning of a time series model using a mixture of HMMs with application to the IoT fuel sensor verification

Continual learning of a time series model using a mixture of HMMs with application to the IoT fuel sensor verification

Change-Point Detection of Peak Tibial Acceleration in Overground Running Retraining

Change Point Detection with Mean Shift Based on AUC from Symmetric Sliding Windows

Contact Info

Product

Resources

About