DCS or PLC feeds your PI System and tags are organized around processes and assets in the PI Asset Framework. The team has created a variety of analytics to monitor the process through a combination of notifications and visualization.
Current PI implementation follows best practices. Areas of improvement include:
- Certain conditions affecting reliability, efficiency, productivity, and profitability are not identified
- Engineers would need to spend a lot of time and apply expertise to generate insights
- Many false positives get created – resolving these is time consuming
- Detecting some conditions earlier will result in substantial benefits in quality, productivity, and cost avoidance
Falkonry enables PI AF users to quickly apply complex pattern recognition to their PI Points. The results of pattern recognition are available as attributes on the same Asset Templates where it can be used to alert workers, trigger workflows, and be rendered in dashboard displays.
Falkonry for PI Asset Framework automatically reveals early onset of undesirable and unexpected condition. Falkonry’s pattern recognition applies powerful artificial intelligence that produces more relevant and accurate results than Anomaly Detection, Statistical Process Control, or calculated thresholds on individual tags as it takes into account the shape of multivariate PI Points.
Falkonry-PI Integrator can be deployed on-site for secure condition monitoring and is integrated with PI System and PI Asset Framework.
These five steps can be performed by existing people who manage the PI system.
Step 1: Download the Falkonry Server and install within the operations network running PI System.
Step 2: Respond to prompts generated by Falkonry PI AF Integrator (which runs in the Falkonry Service). These will be questions about the PI AF and Data Servers to establish connectivity between PI and Falkonry.
Step 3: Use Falkonry UI to automatically generate models for the patterns in PI System data. Optionally, give relevant condition names to patterns discovered by Falkonry.
Step 4: Turn on live monitoring in Falkonry UI and observe pattern-based condition assessments flowing continuously into PI System.
Step 5: Enhance existing PI Coresight dashboards or PI DataLink visualizations to include Falkonry condition assessments. Alternately, use PI Notifications to issue alerts for undesirable conditions based on Falkonry assessments.