(268b) A New Paradigm in Real Time Asset Monitoring and Fault Diagnosis

Asset-intensive industries, like Air Products, continuously operate plants and equipment, pushing assets to the limits of their design envelope in order to drive productivity and efficiency.  Operating in this manner can often lead to unexpected process behavior and faults affecting the overall life of the asset.  A solution is needed that provides foresight and insight for resolving an issue before it becomes an adverse situation.  With the wealth of data available in real-time, there is a tremendous opportunity to build signature of a process and an equipment based on the expected engineering performance and history of operating parameters. These signatures can provide tremendous insights and predictive capabilities to enable real-time, intelligent monitoring  and diagnostics of plants and process systems.

At the foundation of this technological revolution are predictive models built uniquely for individual plants or equipments.  Both traditional SQC and state-of-the-art statistical multivariate technology are utilized to capture the correlations amongst variables, transforming huge plant datasets into actionable information.  These data-driven auto-adaptive models provide advanced alerts of subtle abnormalities and deviations from expected behavior across a whole plant or its specific components. The predictive analytics focuses on diagnostics to enable proactive intervention.   This enhanced level of automated monitoring and fault diagnosis results in improved plant and equipment reliability, increased awareness of impending failures, and bringing additional value to our customers. 

In this presentation we will also highlight some of the challenges and lessons learned from a commercially successful technology project. Monitoring and fault diagnostics of both Continuous and Batch Processess will be discussed with case studies.