Predictive Maintenance for Domain-Specific Pumps through Intelligent Asset Farms Management and Monitoring
Predictive maintenance (PdM) with the industrial internet of things (IIoT) for digital assets has generated significant interest in the oil and gas industry (OGI) like many others. The OGIs are interested in predicting the field-specific specialized equipment pump (asset) failures well advance in time to enhance the operations and reduce the pump’s downtime. One of the ways to tackle such a problem is to continuously monitor the pump’s health and conditions using electrical and mechanical metrics through the IIoT sensors. These IIoT sensors transmit data to a centralized server where a predictive analytics platform resides. Such types of telemetry data in conjunction with advanced machine learning (ML) algorithms attempt to avoid failure by scheduling the predictive/preventive maintenance in a data-driven manner. The main aim of the system is a reduction in production outage, prevent unexpected equipment failure and reduce the monetary costs. This work describes the analytical platform for generating early warning systems (EWS) which predicts impending failures, condition-based monitoring, root cause analysis of failures and characteristics patterns, equipment management, monitoring and visualization in an easy manner.