Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Asset Performance Management Using Machine Learning
Utilities have assets spread across diverse geographies which are very difficult to monitor and provide timely attention on daily basis. In addition, unplanned downtime often costs utilities 10× the cost of planned maintenance. Traditional approach for asset reliability leverages the first principles model. Artificial Intelligence (AI) and Machine learning (ML) techniques appear to solve the same problems but differ in areas of human intervention and accuracy of prediction. A widespread integration of machine learning in Asset Performance Management (APM) marks a transition from estimated engineering and statistical models towards measuring patterns of asset behavior. Utilities are adopting the proliferation of advanced technology where they can predict failures before they occur and build proactive maintenance plan, thereby improving reliability and expenditure. Asset Performance Management (APM), with the capability to analyses large data sets from both IT and OT systems from field assets, helps take 3R decisions (Repair Vs Replace Vs Refurbish). This paper presents a case study of an Indian utility on APM using Machine Learning techniques on Distribution Transformers wherein the failure rate was reduced from 8 to 9% by estimated 2–3% thereby providing system reliability and estimated savings of USD 1.31 million per annum (for 20,000 DTs) through a reduction in transformer repair charges. These ML techniques can also be used for cables, switchgears and other critical assets across the power generation, transmission, distribution and renewables value chain thereby transitioning from Time-Based maintenance (TBM) to Condition-Based Maintenance (CBM).
Asset Performance Management Using Machine Learning
Utilities have assets spread across diverse geographies which are very difficult to monitor and provide timely attention on daily basis. In addition, unplanned downtime often costs utilities 10× the cost of planned maintenance. Traditional approach for asset reliability leverages the first principles model. Artificial Intelligence (AI) and Machine learning (ML) techniques appear to solve the same problems but differ in areas of human intervention and accuracy of prediction. A widespread integration of machine learning in Asset Performance Management (APM) marks a transition from estimated engineering and statistical models towards measuring patterns of asset behavior. Utilities are adopting the proliferation of advanced technology where they can predict failures before they occur and build proactive maintenance plan, thereby improving reliability and expenditure. Asset Performance Management (APM), with the capability to analyses large data sets from both IT and OT systems from field assets, helps take 3R decisions (Repair Vs Replace Vs Refurbish). This paper presents a case study of an Indian utility on APM using Machine Learning techniques on Distribution Transformers wherein the failure rate was reduced from 8 to 9% by estimated 2–3% thereby providing system reliability and estimated savings of USD 1.31 million per annum (for 20,000 DTs) through a reduction in transformer repair charges. These ML techniques can also be used for cables, switchgears and other critical assets across the power generation, transmission, distribution and renewables value chain thereby transitioning from Time-Based maintenance (TBM) to Condition-Based Maintenance (CBM).
Asset Performance Management Using Machine Learning
Lect. Notes Electrical Eng.
Pillai, Reji Kumar (Herausgeber:in) / Singh, B. P. (Herausgeber:in) / Murugesan, N. (Herausgeber:in) / Kumar, Somesh (Autor:in) / Mishra, Vinit (Autor:in) / Singh, Abinash (Autor:in) / Sharma, Amit (Autor:in) / Singh, Vivek (Autor:in) / Thukral, Hem (Autor:in)
28.05.2022
14 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Time-Based maintenance (TBM) , Condition-Based Maintenance (CBM) , Predictive Maintenance , Asset Performance Management (APM) Engineering , Power Electronics, Electrical Machines and Networks , Measurement Science and Instrumentation , Energy Policy, Economics and Management , Cyber-physical systems, IoT , Professional Computing , Circuits and Systems , Energy
Asset Performance Management Using Machine Learning
| TIBKAT | 2022
Asset Management - Corporates entdecken Asset Management
| Online Contents | 2009
Asset management: Learning from other industries
Online Contents | 2004