Assessing neural networks for sensor fault detection

Jäger, Georg; Zug, Sebastian; Brade, Tino; Dietrich, André; Steup, Christoph; Moewes, Christian; Creţu, Ana-Maria

doi:10.1109/civemsa.2014.6841441

Cited by 25 publications

(6 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A review of the literature demonstrates that training can be very demanding for DL models. In [43], a common procedure for fault detection using DL models is proposed, feature extraction and neural network type are regarded as two key elements of the process. In [44] an autoencoder is used with 4 different and selectable behaviors.…”

Section: Discussionmentioning

confidence: 99%

Fault Diagnosis via Neural Ordinary Differential Equations

2021

View full text Add to dashboard Cite

Implementation of model-based fault diagnosis systems can be a difficult task due to the complex dynamics of most systems, an appealing alternative to avoiding modeling is to use machine learning-based techniques for which the implementation is more affordable nowadays. However, the latter approach often requires extensive data processing. In this paper, a hybrid approach using recent developments in neural ordinary differential equations is proposed. This approach enables us to combine a natural deep learning technique with an estimated model of the system, making the training simpler and more efficient. For evaluation of this methodology, a nonlinear benchmark system is used by simulation of faults in actuators, sensors, and process. Simulation results show that the proposed methodology requires less processing for the training in comparison with conventional machine learning approaches since the data-set is directly taken from the measurements and inputs. Furthermore, since the model used in the essay is only a structural approximation of the plant; no advanced modeling is required. This approach can also alleviate some pitfalls of training data-series, such as complicated data augmentation methodologies and the necessity for big amounts of data.

show abstract

Section: Discussionmentioning

confidence: 99%

Fault Diagnosis via Neural Ordinary Differential Equations

2021

View full text Add to dashboard Cite

show abstract

“…Jäger et al [78] introduced a framework to detect four different types of fault: outliers, offset, noise and stuck-at-zero, using a supervised time-delay neural network (TDNN). It is a type of multi-layer feed-forward ANN that allows the mapping between past and present values by analysing the sliding windows of a signal.…”

Section: Anomaly/fault Detectionmentioning

confidence: 99%

Sensor data quality: a systematic review

2020

View full text Add to dashboard Cite

Sensor data quality plays a vital role in Internet of Things (IoT) applications as they are rendered useless if the data quality is bad. This systematic review aims to provide an introduction and guide for researchers who are interested in quality-related issues of physical sensor data. The process and results of the systematic review are presented which aims to answer the following research questions: what are the different types of physical sensor data errors, how to quantify or detect those errors, how to correct them and what domains are the solutions in. Out of 6970 literatures obtained from three databases (ACM Digital Library, IEEE Xplore and ScienceDirect) using the search string refined via topic modelling, 57 publications were selected and examined. Results show that the different types of sensor data errors addressed by those papers are mostly missing data and faults e.g. outliers, bias and drift. The most common solutions for error detection are based on principal component analysis (PCA) and artificial neural network (ANN) which accounts for about 40% of all error detection papers found in the study. Similarly, for fault correction, PCA and ANN are among the most common, along with Bayesian Networks. Missing values on the other hand, are mostly imputed using Association Rule Mining. Other techniques include hybrid solutions that combine several data science methods to detect and correct the errors. Through this systematic review, it is found that the methods proposed to solve physical sensor data errors cannot be directly compared due to the non-uniform evaluation process and the high use of non-publicly available datasets. Bayesian data analysis done on the 57 selected publications also suggests that publications using publicly available datasets for method evaluation have higher citation rates.

show abstract

“…Während langjähriger Bauwerksmessungen kann es zu verschiedenen Messfehlern kommen, welche bspw. in [21, 22] beschrieben sind. Typische Messfehler zeigt Bild 7.…”

Section: Anwendungsfälleunclassified

Prognose von Messdaten beim Bauwerksmonitoring mithilfe von Machine Learning

Wedel¹,

Marx

2020

Bautechnik

View full text Add to dashboard Cite

Prediction of the bridge temperature using monitoring data and machine learning In this paper, the nonlinear or rather transient relationship between the air temperature and the building temperature is simulated by a machine learning model. Based on this modelling, different use cases for the application of machine learning regression methods to monitoring data are presented, which resulted from practical questions. Basic knowledge of neural networks will be given and for the calculations, long-term monitoring data of valley bridges from VDE 8 are used. It is shown, for example, that these methods can be used to detect and compensate measurement errors or to predict the behaviour of structures. The results show that these methods, have a great potential for the evaluation of large amounts of data since no physical models are required. The paper focuses more on the application of AI methods rather than on the mathematical or theoretical background of the methods.

show abstract

Assessing neural networks for sensor fault detection

Cited by 25 publications

References 15 publications

Fault Diagnosis via Neural Ordinary Differential Equations

Fault Diagnosis via Neural Ordinary Differential Equations

Sensor data quality: a systematic review

Prognose von Messdaten beim Bauwerksmonitoring mithilfe von Machine Learning

Contact Info

Product

Resources

About