Adaptive similarity search for the retrieval of rare events from large time series databases

Schlegl, Thomas; Schlegl, Stefan; Tomaselli, Domenico; West, Nikolai; Deuse, Jochen

doi:10.1016/j.aei.2022.101629

Cited by 8 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Popular feature mining methods include basic statistics (BS) [24], fast Fourier transform (FFT) [25], short-time Fourier transform (STFT) [26], continuous wavelet transform (CWT) [27], DWT [28], and auto-regressive moving average (ARMA) [29]. Popular machine learning models include SVM [30], k-nearest neighbor (KNN) [31,32], gradient boosting decision tree (GBDT) [33,34], Bayesian network (BN) [35], decision tree (DT) [36], random forest (RF) [37], and CNN [38].…”

Section: Introductionmentioning

confidence: 99%

A Novel Air-Door Opening and Closing Identification Algorithm Using a Single Wind-Velocity Sensor

Shang

Deng

Liu

2022

Sensors

View full text Add to dashboard Cite

The air-door is an important device for adjusting the air flow in a mine. It opens and closes within a short time owing to transportation and other factors. Although the switching sensor alone can identify the air-door opening and closing, it cannot relate it to abnormal fluctuations in the wind speed. Large fluctuations in the wind-velocity sensor data during this time can lead to false alarms. To overcome this problem, we propose a method for identifying air-door opening and closing using a single wind-velocity sensor. A multi-scale sliding window (MSSW) is employed to divide the samples. Then, the data global features and fluctuation features are extracted using statistics and the discrete wavelet transform (DWT). In addition, a machine learning model is adopted to classify each sample. Further, the identification results are selected by merging the classification results using the non-maximum suppression method. Finally, considering the safety accidents caused by the air-door opening and closing in an actual production mine, a large number of experiments were carried out to verify the effect of the algorithm using a simulated tunnel model. The results show that the proposed algorithm exhibits superior performance when the gradient boosting decision tree (GBDT) is selected for classification. In the data set composed of air-door opening and closing experimental data, the accuracy, precision, and recall rates of the air-door opening and closing identification are 91.89%, 93.07%, and 91.07%, respectively. In the data set composed of air-door opening and closing and other mine production activity experimental data, the accuracy, precision, and recall rates of the air-door opening and closing identification are 89.61%, 90.31%, and 88.39%, respectively.

show abstract

Section: Introductionmentioning

confidence: 99%

A Novel Air-Door Opening and Closing Identification Algorithm Using a Single Wind-Velocity Sensor

Shang

Deng

Liu

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Data compression, in addition to saving storage capacity, simplifies the data transfer and enhances the performance of time series databases. (7) Compression techniques fall into two main groups: lossless compression and lossy compression. (8) Lossless compression is used to retrieve the original data for various purposes.…”

Section: Introductionmentioning

confidence: 99%

Agile Lossless Compression Algorithm for Big Data of Solar Energy Harvesting Wireless Sensor Network

El-Hageen¹,

Albalawi²,

Alatwi³

et al. 2022

Sensors and Materials

View full text Add to dashboard Cite

Time series data are collected through most of the applications that permeate our lives today. Internet of Things (IoT) sensor data are generated through smart applications and stored in databases. Time series databases require huge storage spaces, as over time they consume a large amount of memory. In this paper, we propose an enhanced compression algorithm for time series data generated by IoT systems that monitor the production of electrical energy by solar panels. The best way to ensure that solar energy systems have high efficiency is to continuously monitor all electrical and environmental factors. However, this requires the collection of enormous quantities of data that can be used to detect defects in the generation of electric energy or in solar panels. As the data must be available for analysis, a lossless compression algorithm is needed. In addition, the compressed data must be in a format that can be queried to perform analysis operations dependent on speed; this means that the decompression of data should not be timeconsuming. Our results showed the high speed of the compression process along with good compression rate (16.6%) after applying the proposed compression algorithm.

show abstract

Potenzialanalyse industrieller Datenanalysen in der Produktion am Beispiel des Kunststoffschweißens

Beiß,

Klapic,

Hardt

et al. 2023

Industrielle Datenanalyse

View full text Add to dashboard Cite

ZusammenfassungDas Kapitel beschreibt die Umsetzung der Datenakquise am Beispiel des Heizelementschweißens von Kunststoffen an einer Laborschweißanlage sowie an einer industriellen Produktionsanlage. Die Identifizierung der relevanten Datenquellen und die Verknüpfung des Expertenwissens über den Prozess sind Kernbestandteile der Realisierung. Dazu wurde zunächst er Schweißprozess in die relevanten Prozessphasen eingeteilt. Innerhalb der Phasen wurden wichtige und an der Maschine verfügbare Prozessgrößen identifiziert und mithilfe eines Datensammlers (ARENDAR) über die gesamte Prozesszeit aufgezeichnet. Im Anschluss wurden die Daten mittels KI-Methoden analysiert. Es zeigte sich u. a., dass der Prozessablauf in der Maschine Fehler enthielt. Auch der Prozessablauf war in den ersten Messungen mit falschen Parametern eingestellt, die über die Auswertungen identifiziert und korrigiert werden konnten. In weiteren Versuchsreihen wurden gezielt mit Fehlern präparierte Probekörper geschweißt und die Auswirkungen auf die erfassten Daten untersucht. Dabei stellte sich heraus, dass der Prozess sehr robust ist und die automatisierte Erkennung von Fehlern im Prozess im Labormaßstab nicht eindeutig möglich war. Weitere Messungen werden in einem industriellen Szenario durchgeführt, bei dem eine größere Anzahl an Bauteilen als im Labor geschweißt werden kann. Zusammenfassend zeigte sich, dass nur im Zusammenspiel von Datenquellen, Expertenwissen und einer geeigneten Interpretation mittels Data Science Methoden eine erfolgreiche KI-gestützte Inline-Prozessanalyse möglich ist.

show abstract

Adaptive similarity search for the retrieval of rare events from large time series databases

Cited by 8 publications

References 19 publications

A Novel Air-Door Opening and Closing Identification Algorithm Using a Single Wind-Velocity Sensor

A Novel Air-Door Opening and Closing Identification Algorithm Using a Single Wind-Velocity Sensor

Agile Lossless Compression Algorithm for Big Data of Solar Energy Harvesting Wireless Sensor Network

Potenzialanalyse industrieller Datenanalysen in der Produktion am Beispiel des Kunststoffschweißens

Contact Info

Product

Resources

About