Edge-Compatible Convolutional Autoencoder Implemented on FPGA for Anomaly Detection in Vibration Condition-Based Monitoring

2023

Sci Rep

Self Cite

Automated fault diagnosis algorithms based on vibration sensor recordings play an important role in determining the state of health of the machines. Data-driven approaches demand a large amount of labelled data to build reliable models. The performance of such lab-trained models degrades when deployed in practical use cases in the presence of distinct distribution target domain datasets. In this work, we present a novel deep transfer learning strategy that fine-tunes the trainable parameters of the lower (convolutional) layers with respect to the changing target domain datasets and transfers the parameters of the deeper (dense) layers from the source domain for efficient domain generalisation and fault classification. The performance of this strategy is evaluated by considering two different target domain datasets and studying the sensitivity of fine-tuning individual layers in the networks using time-frequency representations of the vibration signals (scalograms) as inputs. We observe that the proposed transfer learning strategy yields near-perfect accuracy, even for use cases where low-precision sensors are used for data collection and unlabelled run-to-failure data with a limited number of training samples.

Section: Methodsmentioning

confidence: 99%

Deep transfer learning strategy for efficient domain generalisation in machine fault diagnosis

2023

Sci Rep

Self Cite

“…Various methods are used to represent vibration signals in the time and frequency domain. However, time-frequency imaging methods are efficient at capturing the non-stationary nature of the vibration signals [28][29][30][31] . In this work, we have considered scalograms obtained using continuous wavelet transform (CWT) as a representation of the vibration signals.…”

Section: Methodology Time-frequency Based Representations Of Vibratio...mentioning

confidence: 99%

Deep transfer learning strategy for efficient domain generalisation in machine fault diagnosis

2022

Preprint

Self Cite

Automated fault diagnosis algorithms based on vibration sensor recordings play an important role in determining the state of health of the machines. Data-driven approaches require a large number of supervised and labelled samples to build reliable models. The performance of such lab-trained models when deployed in practical use cases largely depends on their domain generalisation capability towards distinct distribution target domain datasets. These challenges are addressed to some extent by conventional parameter-based transfer learning techniques that fine-tune deeper (dense) layer parameters of the pre-trained networks when lower (convolutional) layers can capture more domain-specific information concerning the changing target domain distribution. In this work, we present a novel deep transfer learning strategy that utilises fine-tuning the parameters of convolutional layers and transfer of knowledge in the dense layers from the source domain for efficient domain generalisation and fault classification. The performance of this strategy is evaluated with two different target domain data sets, by studying the sensitivity of fine-tuning individual layers in the networks using time-frequency representations of the vibration signals (scalograms) as inputs. We observe that this strategy yields near-perfect accuracy, even for use cases where low-precision sensors are used for data collection, and unlabelled run-to-failure data with a limited number of training samples.

“…Feature selection plays an important role in determining model performance in machine fault diagnosis. Frequency and timefrequency-based features extracted from the raw time-series vibration recordings have been shown to result in enhanced model performance due to their ability to capture nonstationary behavior of the machine condition parameters [22,35]. However, substantial computation power and resources are required for processing large models and computing timefrequency features and thus, it is beneficial to represent vibration signals using features computed from time-domain signals for realizing lightweight models for practical industrial applications [16,36].…”

Section: Feature Extractionmentioning

confidence: 99%

“…However, the major limitation of such approaches is that the number of clusters needs to be defined in advance, which may not be viable in practical scenarios, where the number of unique faults or deviations from normal operating conditions for any machine may be hard to estimate a priori. Apart from clustering techniques, many reports are available on using unsupervised deep autoencoders for machine fault diagnosis applications [20][21][22]. Zhu et al proposed a stacked pruning sparse denoising autoencoder for better generalization ability with enhanced feature extraction thus, resulting in higher diagnostic accuracy for bearing fault prediction [23].…”

Section: Introductionmentioning

confidence: 99%

An explainable unsupervised learning framework for scalable machine fault detection in Industry 4.0

2023

Meas. Sci. Technol.

Self Cite

Despite the diverse number of machine learning algorithms reported in the literature for machine fault detection, their implementation is mainly confined to laboratory-scale demonstrations. The complexity and black-box nature of machine learning models, the processing cost involved in appropriate feature extraction, limited access to labeled data, and varying operating conditions are some of the key reasons that curtail their implementation in practical applications. Furthermore, most such models serve as decision support tools, aiding domain experts in root cause analysis, and are not truly autonomous by themselves. To address these challenges, we present a lightweight autoencoder-based unsupervised learning framework to accurately identify machine faults against the changing operating conditions in a real-world scenario. The fault detection strategy is further strengthened by a model agnostic Shapley Additive exPlanations (SHAP)-based method (kernel SHAP) for identifying the most prominent features contributing to fault detection inference, the findings of which are then explored for identifying trends and correlations among prominent features and various types of faults. The framework is validated using two widely used and publicly available datasets for machine condition monitoring, as well as a large industrial dataset comprising 18 machines installed at 3 factories in India, monitored for several months.