{"id":{"repo_id":"rgu","oai_identifier":"oai:rgu-repository.worktribe.com:2807431"},"canonical_url":"https://search.dev.ndltd.org/etd/rgu/oai:rgu-repository.worktribe.com:2807431","repository":{"repo_id":"rgu","name":"Robert Gordon University","base_url":"https://rgu-repository.worktribe.com/oaiprovider"},"display":{"title":"New methods for the protection of embedded generators against the loss of utility networks.","abstract":"The importance of being able to reliably detect Loss of Mains (LOM) for Embedded Generation (EG) has been much documented. In particular it is very difficult for current methods to detect LOM when the EG is \"islanded\" with part of the Utility Load which has a similar capacity to the EG. For these reasons much research is currently being undertaken in an attempt to find new and improved methods of detecting LOM. This thesis presents the results of one such investigation. A thorough literature review confirmed that the current methods of detecting LOM, particularly ROCOF and vector surge, would sometimes fail to operate correctly. From initial theoretical analysis and computer simulation it was found that these problems were increased when the EG was \"islanded\" with a load of similar capacity. To investigate this phenomenon fully a simulation network was developed using the Alternative Transients Program (ATP). Various control and measurement devices were built using the Transient Analysis of Control Systems (TACS) part of ATP. After extensive investigation it was decided that the problem was ideal for an Artificial Neural Network (ANN) solution. Further research indicated that the Multi-Layer Perceptron (MLP) with back-propagation learning had been used in many power systems control and protection applications. Its ability to recognise and classify complex input patterns made it very suitable for the LOM application. Exhaustive investigation revealed that it was possible to build an ANN which gave over 99% accurate classification, based on input from sampled three phase voltages and current. The possibilities for improving the accuracy of the proposed ANN LOM relay are discussed, in particular a possible method for completely eradicating maloperations. Also, the possibilities for a practical development of the relay are considered. Therefore, this thesis proposes a new method for detecting LOM for EG using an ANN approach.","abstract_html":"The importance of being able to reliably detect Loss of Mains (LOM) for Embedded Generation (EG) has been much documented. In particular it is very difficult for current methods to detect LOM when the EG is "islanded" with part of the Utility Load which has a similar capacity to the EG. For these reasons much research is currently being undertaken in an attempt to find new and improved methods of detecting LOM. This thesis presents the results of one such investigation. A thorough literature review confirmed that the current methods of detecting LOM, particularly ROCOF and vector surge, would sometimes fail to operate correctly. From initial theoretical analysis and computer simulation it was found that these problems were increased when the EG was "islanded" with a load of similar capacity. To investigate this phenomenon fully a simulation network was developed using the Alternative Transients Program (ATP). Various control and measurement devices were built using the Transient Analysis of Control Systems (TACS) part of ATP. After extensive investigation it was decided that the problem was ideal for an Artificial Neural Network (ANN) solution. Further research indicated that the Multi-Layer Perceptron (MLP) with back-propagation learning had been used in many power systems control and protection applications. Its ability to recognise and classify complex input patterns made it very suitable for the LOM application. Exhaustive investigation revealed that it was possible to build an ANN which gave over 99% accurate classification, based on input from sampled three phase voltages and current. The possibilities for improving the accuracy of the proposed ANN LOM relay are discussed, in particular a possible method for completely eradicating maloperations. Also, the possibilities for a practical development of the relay are considered. Therefore, this thesis proposes a new method for detecting LOM for EG using an ANN approach.","abstract_has_math":false,"creators":["King, David John"],"institution":"Robert Gordon University","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["S.K. Salman"],"committee_chairs":[],"committee_members":[],"year":1999,"date_issued":"1999","date_published":"1999","updated_at":"2026-07-24T04:10:09Z","subjects":["Loss of mains (LOM)","Embedded generation (EG)","Islanded","Utility load","Rate of change of frequency (ROCOF)","Vector surge","Alternative transients program (ATP)","Transient analysis of control systems (TACS)","Simulation","Artificial neural network (ANN)"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:rgu-repository.worktribe.com:2807431","https://doi.org/10.48526/rgu-wt-2807431"],"render_values":[{"text":"oai:rgu-repository.worktribe.com:2807431","href":null,"code":true},{"text":"https://doi.org/10.48526/rgu-wt-2807431","href":"https://doi.org/10.48526/rgu-wt-2807431","code":true}]}]},"links":{"outbound_url":"https://rgu-repository.worktribe.com/2807431/1/KING%201999%20New%20methods%20for%20the%20protection","outbound_label":"Repository record","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["S.K. Salman"]},{"key":"dc:contributor.sponsor","label":"Sponsor","values":["Engineering & Physical Sciences Research Council","New Funding Organisation"]},{"key":"dc:creator","label":"Author","values":["King, David John"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["1999-12-31"]},{"key":"dc:date.issued","label":"Date","values":["1999"]},{"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/2807431"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Loss of mains (LOM)","Embedded generation (EG)","Islanded","Utility load","Rate of change of frequency (ROCOF)","Vector surge","Alternative transients program (ATP)","Transient analysis of control systems (TACS)","Simulation","Artificial neural network (ANN)"]}]},{"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:2807431","https://doi.org/10.48526/rgu-wt-2807431"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://rgu-repository.worktribe.com/2807431/1/KING%201999%20New%20methods%20for%20the%20protection"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["The importance of being able to reliably detect Loss of Mains (LOM) for Embedded Generation (EG) has been much documented. In particular it is very difficult for current methods to detect LOM when the EG is \"islanded\" with part of the Utility Load which has a similar capacity to the EG. For these reasons much research is currently being undertaken in an attempt to find new and improved methods of detecting LOM. This thesis presents the results of one such investigation. A thorough literature review confirmed that the current methods of detecting LOM, particularly ROCOF and vector surge, would sometimes fail to operate correctly. From initial theoretical analysis and computer simulation it was found that these problems were increased when the EG was \"islanded\" with a load of similar capacity. To investigate this phenomenon fully a simulation network was developed using the Alternative Transients Program (ATP). Various control and measurement devices were built using the Transient Analysis of Control Systems (TACS) part of ATP. After extensive investigation it was decided that the problem was ideal for an Artificial Neural Network (ANN) solution. Further research indicated that the Multi-Layer Perceptron (MLP) with back-propagation learning had been used in many power systems control and protection applications. Its ability to recognise and classify complex input patterns made it very suitable for the LOM application. Exhaustive investigation revealed that it was possible to build an ANN which gave over 99% accurate classification, based on input from sampled three phase voltages and current. The possibilities for improving the accuracy of the proposed ANN LOM relay are discussed, in particular a possible method for completely eradicating maloperations. Also, the possibilities for a practical development of the relay are considered. Therefore, this thesis proposes a new method for detecting LOM for EG using an ANN approach."]},{"key":"dc:title","label":"Title","values":["New methods for the protection of embedded generators against the loss of utility networks."]}]}],"canonical_facts":{"dc:contributor.advisor":["S.K. Salman"],"dc:contributor.sponsor":["Engineering & Physical Sciences Research Council","New Funding Organisation"],"dc:creator":["King, David John"],"dc:date":["1999-12-31"],"dc:date.issued":["1999"],"dc:description.abstract":["The importance of being able to reliably detect Loss of Mains (LOM) for Embedded Generation (EG) has been much documented. In particular it is very difficult for current methods to detect LOM when the EG is \"islanded\" with part of the Utility Load which has a similar capacity to the EG. For these reasons much research is currently being undertaken in an attempt to find new and improved methods of detecting LOM. This thesis presents the results of one such investigation. A thorough literature review confirmed that the current methods of detecting LOM, particularly ROCOF and vector surge, would sometimes fail to operate correctly. From initial theoretical analysis and computer simulation it was found that these problems were increased when the EG was \"islanded\" with a load of similar capacity. To investigate this phenomenon fully a simulation network was developed using the Alternative Transients Program (ATP). Various control and measurement devices were built using the Transient Analysis of Control Systems (TACS) part of ATP. After extensive investigation it was decided that the problem was ideal for an Artificial Neural Network (ANN) solution. Further research indicated that the Multi-Layer Perceptron (MLP) with back-propagation learning had been used in many power systems control and protection applications. Its ability to recognise and classify complex input patterns made it very suitable for the LOM application. Exhaustive investigation revealed that it was possible to build an ANN which gave over 99% accurate classification, based on input from sampled three phase voltages and current. The possibilities for improving the accuracy of the proposed ANN LOM relay are discussed, in particular a possible method for completely eradicating maloperations. Also, the possibilities for a practical development of the relay are considered. Therefore, this thesis proposes a new method for detecting LOM for EG using an ANN approach."],"dc:identifier":["oai:rgu-repository.worktribe.com:2807431","https://doi.org/10.48526/rgu-wt-2807431"],"dc:identifier.uri":["https://rgu-repository.worktribe.com/2807431/1/KING%201999%20New%20methods%20for%20the%20protection"],"dc:language":["en"],"dc:publisher.institution":["Robert Gordon University"],"dc:relation.isreferencedby":["https://rgu-repository.worktribe.com/output/2807431"],"dc:subject":["Loss of mains (LOM)","Embedded generation (EG)","Islanded","Utility load","Rate of change of frequency (ROCOF)","Vector surge","Alternative transients program (ATP)","Transient analysis of control systems (TACS)","Simulation","Artificial neural network (ANN)"],"dc:title":["New methods for the protection of embedded generators against the loss of utility networks."],"dc:type":["Thesis"]},"updated_at":"2026-07-24T04:10:09Z"}