A novel unsupervised anomaly detection for gas turbine using Isolation Forest

Zhong, Shisheng; Song, Fu; Lin, Lin; Fu, Xuyun; Cui, Zhiquan; Wang, Rui

doi:10.1109/icphm.2019.8819409

Cited by 26 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This defines the number of neighboring data points that the algorithm considers when computing local density, thereby allowing each data point to be understood about its 20 closest neighbors in the dataset. In our visual analyses, n_neighbors=20 showed an optimal balance, offering a clear distinction between outliers and inliers, and ensuring consistent outlier identification across the dataset [13].…”

Section: A Local Outlier Factormentioning

confidence: 95%

Machine Learning Approaches to Advanced Outlier Detection in Psychological Datasets

Al. Abri,

Singh Sidhu

2024

Int. j. electr. comput. eng. syst. (Online)

View full text Add to dashboard Cite

The core aim of this study is to determine the most effective outlier detection methodologies for multivariate psychological datasets, particularly those derived from Omani students. Due to their complex nature, such datasets demand robust analytical methods. To this end, we employed three sophisticated algorithms: local outlier factor (LOF), one-class support vector machine (OCSVM), and isolation forest (IF). Our initial findings showed 155 outliers by both LOF and IF and 147 by OCSVM. A deeper analysis revealed that LOF detected 55 unique outliers based on differences in local density, OCSVM isolated 44 unique outliers utilizing its transformed feature space, and IF identified 76 unique outliers leveraging its tree-based mechanics. Despite these varying results, all methods had a consensus for just 44 outliers. Employing ensemble techniques, both averaging and voting methods identified 155 outliers, whereas the weighted method highlighted 151, with a consensus of 150 outliers across the board. In conclusion, while individual algorithms provide distinct perspectives, ensemble techniques enhance the accuracy and consistency of outlier detection. This underscores the necessity of using multiple algorithms with ensemble techniques in analyzing psychological datasets, facilitating a richer comprehension of inherent data structures.

show abstract

Section: A Local Outlier Factormentioning

confidence: 95%

Machine Learning Approaches to Advanced Outlier Detection in Psychological Datasets

Al. Abri,

Singh Sidhu

2024

Int. j. electr. comput. eng. syst. (Online)

View full text Add to dashboard Cite

show abstract

“…Classical techniques for detecting abnormalities (values that significantly deviate from many observations) can be categorized into several types. These include methods based on distance metrics [8,19], methods employing density calculation [20,21], isolation-based methods [9,22], and strategies based on statistical inference [23,24].…”

Section: Related Workmentioning

confidence: 99%

Abnormal State Detection using Memory-augmented Autoencoder technique in Frequency-Time Domain

2024

KSII TIIS

View full text Add to dashboard Cite

With the advancement of Industry 4.0 and Industrial Internet of Things (IIoT), manufacturing increasingly seeks automation and intelligence. Temperature and vibration monitoring are essential for machinery health. Traditional abnormal state detection methodologies often overlook the intricate frequency characteristics inherent in vibration time series and are susceptible to erroneously reconstructing temperature abnormalities due to the highly similar waveforms. To address these limitations, we introduce synergistic, end-to-end, unsupervised Frequency-Time Domain Memory-Enhanced Autoencoders (FTD-MAE) capable of identifying abnormalities in both temperature and vibration datasets. This model is adept at accommodating time series with variable frequency complexities and mitigates the risk of overgeneralization. Initially, the frequency domain encoder processes the spectrogram generated through Short-Time Fourier Transform (STFT), while the time domain encoder interprets the raw time series. This results in two disparate sets of latent representations. Subsequently, these are subjected to a memory mechanism and a limiting function, which numerically constrain each memory term. These processed terms are then amalgamated to create two unified, novel representations that the decoder leverages to produce reconstructed samples. Furthermore, the model employs Spectral Entropy to dynamically assess the frequency complexity of the time series, which, in turn, calibrates the weightage attributed to the loss functions of the individual branches, thereby generating definitive abnormal scores. Through extensive experiments, FTD-MAE achieved an average ACC and F1 of 0.9826 and 0.9808 on the CMHS and CWRU datasets, respectively. Compared to the best representative model, the ACC increased by 0.2114 and the F1 by 0.1876.

show abstract

“…iForest [Liu et al, 2012] is an algorithm that is widely used to perform anomaly detection on timeseries data [Calheiros et al, 2017, Puggini and McLoone, 2018, Qin and Lou, 2019, Zhong et al, 2019, Li and Jung, 2021. The algorithm is based on the fact that there are data points that are few and very different from the dominant data points, then based on this assumption, it can be explained that anomalies are susceptible to a mechanism called isolation.…”

Section: Anomaly Detectionmentioning

confidence: 99%

On the statistical learning analysis of rain gauge data over the Natuna Islands

Fajary¹,

Irawan²

2021

Preprint

View full text Add to dashboard Cite

This article presents state-of-the-art statistical learning methods for analyzing rain gauge data over the Natuna Islands. By using shape preserving piecewise cubic interpolation, we managed to interpolate 671 null values from the daily precipitation data. Dominant periodicity analysis of daily precipitation signals using Lomb-Scargle Power Spectral Density shows annual, intraseasonal, and interannual precipitation patterns over the Natuna Islands. Unsupervised anomaly analysis using the Isolation Forest algorithm shows there are 146 anomaly daily precipitation data points. We also conducted an experiment to predict the accumulation of monthly precipitation over the Natuna Islands using the Bayesian structural time series algorithm. The results show that the local linear trend with seasonality model is able to model the value of accumulated monthly precipitation for a twelve-month prediction horizon. The work presented here has profound implications for rainfall observations in this area.

show abstract

A novel unsupervised anomaly detection for gas turbine using Isolation Forest

Cited by 26 publications

References 9 publications

Machine Learning Approaches to Advanced Outlier Detection in Psychological Datasets

Machine Learning Approaches to Advanced Outlier Detection in Psychological Datasets

Abnormal State Detection using Memory-augmented Autoencoder technique in Frequency-Time Domain

On the statistical learning analysis of rain gauge data over the Natuna Islands

Contact Info

Product

Resources

About