[2F2] Digital twin opportunities and modelling challenges for condition monitoring
S Ganeriwala
SpectraQuest Inc, USA
The main objective behind the digital twin concept is to create a mathematical model or representation of an asset so that its performance can be monitored remotely in the cloud. The overall idea is to analyse the data in the cloud and combine it with historical performance and other relevant information. The results are then presented to reliability engineering personnel so appropriate action can be taken to prevent machine failure. The requirement for remote monitoring has accelerated due to the interference of the pandemic caused by the global spread of the COVID-19 virus in the past couple of years.
Real-time monitoring of an asset, that is conventionally referred to as online monitoring, is the preferred goal of a world-class condition monitoring programme. However, due to the cost and other factors, this approach is currently being applied to critical and important assets only. Also, product vendors and service providers are somewhat hesitant to adopt new technologies and developments that have occurred in the past 20 years in the academic and research sectors. The reason for this is that they are trying to avoid the disturbance of existing diagnostic methods and their existing machine databases. The rush to adopt digital twin applications provides an opportunity to integrate recent data analysis algorithms, wireless sensors and cloud or edge computing to change the condition monitoring paradigm. However, the application of a digital twin requires state-of-the-art diagnostics techniques and a proper machine model. This work addresses the current condition monitoring practices and challenges that must be overcome to apply the digital twin technologies and derive the benefits it could provide. The results of digital twin model development for misalignment using a physics asset and the finite element model are presented.
Real-time monitoring of an asset, that is conventionally referred to as online monitoring, is the preferred goal of a world-class condition monitoring programme. However, due to the cost and other factors, this approach is currently being applied to critical and important assets only. Also, product vendors and service providers are somewhat hesitant to adopt new technologies and developments that have occurred in the past 20 years in the academic and research sectors. The reason for this is that they are trying to avoid the disturbance of existing diagnostic methods and their existing machine databases. The rush to adopt digital twin applications provides an opportunity to integrate recent data analysis algorithms, wireless sensors and cloud or edge computing to change the condition monitoring paradigm. However, the application of a digital twin requires state-of-the-art diagnostics techniques and a proper machine model. This work addresses the current condition monitoring practices and challenges that must be overcome to apply the digital twin technologies and derive the benefits it could provide. The results of digital twin model development for misalignment using a physics asset and the finite element model are presented.
