Ensembles of novelty detection classifiers for structural health monitoring using guided waves

Dib, Gerges; Karpenko, Oleksii; Koricho, Ermias Gebrekidan; Khomenko, Anton; Haq, Mahmoodul; Удпа, Лалита

doi:10.1088/1361-665x/aa973f

Cited by 19 publications

(16 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To enhance the performance of the outlier detection models, ensemble models are used since each individual model is good at certain characteristic feature. Dib et al [17] applied an ensemble of machine learning models to monitor structural health by detecting damages using guided waves generated by the building sensors. An ensemble of novelty detection models based on a consensus vote is proposed in [50] and applied for the detection of ADL anomalies, while in [36], an ensemble approach based on similarity measures is proposed.…”

Section: Related Workmentioning

confidence: 99%

Detecting Anomaly and Its Sources in Activities of Daily Living

2021

View full text Add to dashboard Cite

To support the independent living and improve the quality of life for the increasing ageing population, system for monitoring their daily routine and detecting anomalies in the routine is required. Existing anomaly detection systems are unable to identify the sources of the abnormalities, thereby hindering the development of adaptive monitoring systems with reduced false prediction rate. In this paper, an approach for identifying the sources of abnormalities in human activities of daily living is proposed. Anomalies are detected by modelling the existing activity data representing the usual behavioural routine of an individual to serve as a baseline model. Subsequent activities deviating from the baseline are then classified as outliers or anomalies. An ensemble of one-class support vector machine, isolation forest, robust covariance estimator and local outlier factor is utilised for the anomaly detection achieving an accuracy of $$98\%$$ 98 % . The proposed approach for identifying anomaly sources is based on the concept of similarity measure using distance functions. Two methods for measuring the pairwise distance of the features of the activity data termed as one vs one similarity measure and one vs all similarity measure are proposed. Experimental evaluation of the proposed approach on activities of daily living datasets has shown the credibility of the proposed approach for utilisation in an in-home monitoring system.

show abstract

Section: Related Workmentioning

confidence: 99%

Detecting Anomaly and Its Sources in Activities of Daily Living

2021

View full text Add to dashboard Cite

show abstract

“…In [33], an approach for creating an ensemble of outlier detectors using similarity measures is proposed while Dib et al [34] applied an ensemble of novelty detection models for damage detection for structural health monitoring.…”

Section: Related Workmentioning

confidence: 99%

A Consensus Novelty Detection Ensemble Approach for Anomaly Detection in Activities of Daily Living

Yahaya

Lotfi

Mahmud

2019

Applied Soft Computing

View full text Add to dashboard Cite

A new approach to creating an ensemble of novelty detection algorithms is proposed in this paper. The novelty detection process identifies new or unknown data by detecting if a test data differs significantly from the data available during training. It is applicable for anomaly detection in a situation where there is sufficiently large training data representing the normal class and little or no training data for the anomalous (abnormal) class. Abnormality in Activities of Daily Living (ADL) is identified as any significant deviation from an individual's usual behavioural routine. Novelty detection is relevant to ADL anomaly detection since abnormalities in ADL are rare and data representing the anomalous cases are not readily available. The proposed Consensus Novelty Detection Ensemble approach is based on the concept of internal and external consensus. The internal consensus is an internal voting scheme within each model in the ensemble while the external consensus is an external voting scheme among the ensemble models. The weight of each model is estimated based on its performance and a score, called "Normality Score". Computed score is used in classifying the data as abnormal (anomalous) based on certain threshold crossing, normal otherwise. Experimental evaluation is conducted to detect abnormalities in ADL data obtained from CASAS repository as well as experimental dataset collected for this research. The obtained results show that the proposed approach is able to identify anomalous instances. The proposed approach offers more flexibility compared with the existing approaches by allowing the Normality Score threshold to be adjusted without retraining the models.

show abstract

“…This information is important from the perspective of autonomous operation and control of microreactors as the control decision will depend on the system state and the perceived ability of SSCs to meet their functional goals. 27 Diagnostics for SSC degradation require signatures 28,29 that are invariant to minor changes in system state and can account for missing or failure data. This requires sensor models and invariance transformations 30 of the data for enhancing signatures and normalizing data to ensure the diagnostic algorithms are able to generalize across data variations.…”

Section: Diagnostic and Prognostic Technologiesmentioning

confidence: 99%

Concepts for Autonomous Operation of Microreactors

Ramuhalli

Çetiner

2019

View full text Add to dashboard Cite

Microreactor concepts have seen increasing interest in recent years given their potential use for meeting energy needs in remote and grid-isolated communities. Because of the small power outputs of microreactors and the potential for remote siting, there is a need to significantly reduce operational staffing levels to improve their economic viability. Autonomous control capabilities emerge as a key enabling technology. Autonomous control should not be perceived as a single feature that when added delivers certain functions. On the contrary, autonomous control capability is a highly complex systems engineering process that requires detailed understanding of system dynamic response as a function of operational actions. These advanced capabilities can be achieved at different levels of autonomy, which would then require different levels of human operator involvement in operational decision making. The division of functions and responsibilities between the human operator and the automatic system is the starting point in the system engineering life cycle process. Once this division of functions and responsibilities is established, several options exist for autonomous control ranging from simple automation of some procedures to fully autonomous operational mode with automatic decision-making and execution without the involvement of a human operator. A critical aspect of autonomous decision-making is the ability to have as complete an awareness of the system state as possible. In addition to monitoring the conventional process data, such as temperatures, pressures, flow, and neutron flux, measurements pertaining to the condition of important components become as important as the operational data. This type of information will allow the autonomous control system to adapt its internal configuration to changing conditions of structures, systems and components of importance through the decision-making process. The key capability is to execute these advanced functions with minimal operational disturbance and, of course, without compromising the public health and safety. Several technical and regulatory advances are needed before widespread use of autonomous control in microreactors. These include sensor and instrumentation technologies that are capable of long-term unattended operation in harsh environments, technologies for inferring the state of the microreactor system or subsystems with confidence, algorithms for predictive decision-making that account for the assessed condition of the microreactor subsystems as well as the potential impact on those components of any operational decisions, and actuators and control system hardware that are resilient in harsh environments. In addition, given the need to remotely monitor the operations, cybersecurity requirements will likely need to be imposed to ensure secure operations.

show abstract

Ensembles of novelty detection classifiers for structural health monitoring using guided waves

Cited by 19 publications

References 25 publications

Detecting Anomaly and Its Sources in Activities of Daily Living

Detecting Anomaly and Its Sources in Activities of Daily Living

A Consensus Novelty Detection Ensemble Approach for Anomaly Detection in Activities of Daily Living

Concepts for Autonomous Operation of Microreactors

Contact Info

Product

Resources

About