{"id":{"repo_id":"rgu","oai_identifier":"oai:rgu-repository.worktribe.com:950903"},"canonical_url":"https://search.dev.ndltd.org/etd/rgu/oai:rgu-repository.worktribe.com:950903","repository":{"repo_id":"rgu","name":"Robert Gordon University","base_url":"https://rgu-repository.worktribe.com/oaiprovider"},"display":{"title":"Automated anomaly recognition in real time data streams for oil and gas industry.","abstract":"There is a growing demand for computer-assisted real-time anomaly detection - from the identification of suspicious activities in cyber security, to the monitoring of engineering data for various applications across the oil and gas, automotive and other engineering industries. To reduce the reliance on field experts' knowledge for identification of these anomalies, this thesis proposes a deep-learning anomaly-detection framework that can help to create an effective real-time condition-monitoring framework. The aim of this research is to develop a real-time and re-trainable generic anomaly-detection framework, which is capable of predicting and identifying anomalies with a high level of accuracy - even when a specific anomalous event has no precedent. Machine-based condition monitoring is preferable in many practical situations where fast data analysis is required, and where there are harsh climates or otherwise life-threatening environments. For example, automated conditional monitoring systems are ideal in deep sea exploration studies, offshore installations and space exploration. This thesis firstly reviews studies about anomaly detection using machine learning. It then adopts the best practices from those studies in order to propose a multi-tiered framework for anomaly detection with heterogeneous input sources, which can deal with unseen anomalies in a real-time dynamic problem environment. The thesis then applies the developed generic multi-tiered framework to two fields of engineering: data analysis and malicious cyber attack detection. Finally, the framework is further refined based on the outcomes of those case studies and is used to develop a secure cross-platform API, capable of re-training and data classification on a real-time data feed.","abstract_html":"There is a growing demand for computer-assisted real-time anomaly detection - from the identification of suspicious activities in cyber security, to the monitoring of engineering data for various applications across the oil and gas, automotive and other engineering industries. To reduce the reliance on field experts' knowledge for identification of these anomalies, this thesis proposes a deep-learning anomaly-detection framework that can help to create an effective real-time condition-monitoring framework. The aim of this research is to develop a real-time and re-trainable generic anomaly-detection framework, which is capable of predicting and identifying anomalies with a high level of accuracy - even when a specific anomalous event has no precedent. Machine-based condition monitoring is preferable in many practical situations where fast data analysis is required, and where there are harsh climates or otherwise life-threatening environments. For example, automated conditional monitoring systems are ideal in deep sea exploration studies, offshore installations and space exploration. This thesis firstly reviews studies about anomaly detection using machine learning. It then adopts the best practices from those studies in order to propose a multi-tiered framework for anomaly detection with heterogeneous input sources, which can deal with unseen anomalies in a real-time dynamic problem environment. The thesis then applies the developed generic multi-tiered framework to two fields of engineering: data analysis and malicious cyber attack detection. Finally, the framework is further refined based on the outcomes of those case studies and is used to develop a secure cross-platform API, capable of re-training and data classification on a real-time data feed.","abstract_has_math":false,"creators":["Majdani Shabestari, Farzan"],"institution":"Robert Gordon University","degree_name":"PhD","degree_level":"Doctoral","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["A. Petrovski"],"committee_chairs":[],"committee_members":[],"year":2020,"date_issued":"2020","date_published":"2020","updated_at":"2026-07-24T04:10:00Z","subjects":["Machine learning","Anomaly detection","Real-time data streams","Deep learning","Data classification"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:rgu-repository.worktribe.com:950903"],"render_values":[{"text":"oai:rgu-repository.worktribe.com:950903","href":null,"code":true}]},{"key":"dc:creator.authoridentifier","label":"Author Identifier","values":["0000-0002-7925-3347"],"render_values":[{"text":"0000-0002-7925-3347","href":"https://orcid.org/0000-0002-7925-3347","code":true}]}]},"links":{"outbound_url":"https://rgu-repository.worktribe.com/950903/1/MAJDANI%20SHABESTARI%202020%20Automated%20anomaly%20recognition%20in%20real%20time","outbound_label":"Repository record","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["A. Petrovski"]},{"key":"dc:contributor.sponsor","label":"Sponsor","values":["RGU Internal Funding"]},{"key":"dc:creator","label":"Author","values":["Majdani Shabestari, Farzan"]},{"key":"dc:creator.authoridentifier","label":"Author Identifier","values":["0000-0002-7925-3347"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2020-06-30"]},{"key":"dc:date.issued","label":"Date","values":["2020"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["Robert Gordon University"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://rgu-repository.worktribe.com/output/950903"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["Doctoral"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["PhD"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Machine learning","Anomaly detection","Real-time data streams","Deep learning","Data classification"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:rgu-repository.worktribe.com:950903"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://rgu-repository.worktribe.com/950903/1/MAJDANI%20SHABESTARI%202020%20Automated%20anomaly%20recognition%20in%20real%20time"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["There is a growing demand for computer-assisted real-time anomaly detection - from the identification of suspicious activities in cyber security, to the monitoring of engineering data for various applications across the oil and gas, automotive and other engineering industries. To reduce the reliance on field experts' knowledge for identification of these anomalies, this thesis proposes a deep-learning anomaly-detection framework that can help to create an effective real-time condition-monitoring framework. The aim of this research is to develop a real-time and re-trainable generic anomaly-detection framework, which is capable of predicting and identifying anomalies with a high level of accuracy - even when a specific anomalous event has no precedent. Machine-based condition monitoring is preferable in many practical situations where fast data analysis is required, and where there are harsh climates or otherwise life-threatening environments. For example, automated conditional monitoring systems are ideal in deep sea exploration studies, offshore installations and space exploration. This thesis firstly reviews studies about anomaly detection using machine learning. It then adopts the best practices from those studies in order to propose a multi-tiered framework for anomaly detection with heterogeneous input sources, which can deal with unseen anomalies in a real-time dynamic problem environment. The thesis then applies the developed generic multi-tiered framework to two fields of engineering: data analysis and malicious cyber attack detection. Finally, the framework is further refined based on the outcomes of those case studies and is used to develop a secure cross-platform API, capable of re-training and data classification on a real-time data feed."]},{"key":"dc:title","label":"Title","values":["Automated anomaly recognition in real time data streams for oil and gas industry."]}]}],"canonical_facts":{"dc:contributor.advisor":["A. Petrovski"],"dc:contributor.sponsor":["RGU Internal Funding"],"dc:creator":["Majdani Shabestari, Farzan"],"dc:creator.authoridentifier":["0000-0002-7925-3347"],"dc:date":["2020-06-30"],"dc:date.issued":["2020"],"dc:description.abstract":["There is a growing demand for computer-assisted real-time anomaly detection - from the identification of suspicious activities in cyber security, to the monitoring of engineering data for various applications across the oil and gas, automotive and other engineering industries. To reduce the reliance on field experts' knowledge for identification of these anomalies, this thesis proposes a deep-learning anomaly-detection framework that can help to create an effective real-time condition-monitoring framework. The aim of this research is to develop a real-time and re-trainable generic anomaly-detection framework, which is capable of predicting and identifying anomalies with a high level of accuracy - even when a specific anomalous event has no precedent. Machine-based condition monitoring is preferable in many practical situations where fast data analysis is required, and where there are harsh climates or otherwise life-threatening environments. For example, automated conditional monitoring systems are ideal in deep sea exploration studies, offshore installations and space exploration. This thesis firstly reviews studies about anomaly detection using machine learning. It then adopts the best practices from those studies in order to propose a multi-tiered framework for anomaly detection with heterogeneous input sources, which can deal with unseen anomalies in a real-time dynamic problem environment. The thesis then applies the developed generic multi-tiered framework to two fields of engineering: data analysis and malicious cyber attack detection. Finally, the framework is further refined based on the outcomes of those case studies and is used to develop a secure cross-platform API, capable of re-training and data classification on a real-time data feed."],"dc:identifier":["oai:rgu-repository.worktribe.com:950903"],"dc:identifier.uri":["https://rgu-repository.worktribe.com/950903/1/MAJDANI%20SHABESTARI%202020%20Automated%20anomaly%20recognition%20in%20real%20time"],"dc:language":["en"],"dc:publisher.institution":["Robert Gordon University"],"dc:relation.isreferencedby":["https://rgu-repository.worktribe.com/output/950903"],"dc:subject":["Machine learning","Anomaly detection","Real-time data streams","Deep learning","Data classification"],"dc:title":["Automated anomaly recognition in real time data streams for oil and gas industry."],"dc:type":["Thesis"],"dc:type.qualificationlevel":["Doctoral"],"dc:type.qualificationname":["PhD"]},"updated_at":"2026-07-24T04:10:00Z"}