Using OPC-UA to Extract IIoT Time Series Data from PLC and SCADA Systems
0 likes801 views
The document discusses a webinar hosted by Factry on improving operational intelligence through data integration, specifically using OPC-UA to collect IIOT time series data from PLC and SCADA systems. It highlights the challenges faced by Algist Bruggeman in managing fermentation data, the solutions implemented to streamline data collection and analysis, and the unexpected benefits achieved since integrating these systems. Key takeaways include the importance of creating a cohesive data platform and the positive impact of proper data management on business operations.
Using OPC-UA to Extract IIoT Time Series Data from PLC and SCADA Systems
- 1. by factry Operational intelligence made easy by factry Operational intelligence made easy InfluxData webinar Data collection and integration Using OPC-UA to Extract IIoT Time Series Data from PLC and SCADA Systems Confidential
- 2. Improving data integrity, gaining more insight in the production process and making better yeast.
- 3. Agenda Introducing Ivo - Algist Bruggeman Introducing Frederik - Factry Situation and challenge Solution and outcomes Unexpected Benefits Q&A
- 4. ● Project Manager Automation ● Responsible for automation and electrical maintenance ● Project lead for MES & Historian project About Ivo Lemmens
- 6. Algist Bruggeman supplies fresh, liquid and dried yeast to industrial, semi-artisanal and artisanal bakeries, as well as to the wine, beer and pharma industry. Algist Bruggeman is part of Lesaffre Group, a key global player in fermentation for more than a century, with a 2 billion EUR turnover and established on all continents, counting 10.000+ employees and more than 85 nationalities. Its Belgian site employs about 170 people and has an annual turnover of over 120 million euros.
- 7. ● Raw Materials ● Yeast fermentation ● Processing ⇒ Fresh, liquid, dry yeast Production Process
- 10. A bit of history ● https://github.com/coussej/n ode-opcua-logger ● Started with InfluxDB v1.0.x in 2016 ● Begin with temperature logging for food safety and then gradually expanded data collection on site. ● ~50 PLCs, 4000 tags, ~1Hz resolution
- 11. ● Co-founder and business developer ● Bioscience engineering degree ● Developed large parts of Algist Bruggeman MES application About me
- 12. About Factry
- 13. The Digital Factory We look at the world of industrial automation through the eyes of an IT company. Extract process data to IT level as soon as possible. Because at that level, the fastest progress is being made. OT IT DIGITAL FACTORY
- 14. Factry Historian Collect data from production equipment Store it in a time-series database Visualize it with web-based visualization tools
- 15. What problems does this solve? Look back (Near) real-time Use as basis to predict
- 17. The situation and challenge
- 18. The challenge - Why? ● How did this specific fermentation perform? ● Does a specific fermentation perform differently on fermentor X than on fermentor Y? ● How well does a certain fermentation parameter follow its reference curve?
- 19. ● Fermentations are followed up on paper Valuable information ends up in a folder ● Some data is registered, but isolated in a specific system Cumbersome to answer these questions Data is not readily available
- 20. The problem: different data sources ● Recipe management in Excel ● Fermentation progress on paper ● Lab results in LIMS software ● …
- 21. Answering the 3 questions is hard! Why? Data is not linked. The human needs to bring everything together. PLANNING PRODUCTION LOG HISTORIAN LOGBOOK TRACEABILITY LAB DATA SHIFT PLANNING REPORTING AFHAALLIJST OTHER... ERP
- 22. So let’s bring this data together When have we succeeded? ● Human has become a user, not the person linking data ● The 3 questions can be answered in reasonable timeframe
- 23. Take a holistic approach ● Before fermentation ○ Planning ○ Recipe management ● During fermentation ○ Dispatching ○ Process data collection with Factry Historian in InfluxDB ○ Automation of fermentation sheet completion ● After fermentation ○ Linking lab results ○ Reporting
- 24. Step 1: get batch information ● Planning of orders ○ Retrieve data from planning software & ERP via B2MML ■ Business 2 Manufacturing Markup Language ■ https://github.com/factrylabs/go-b2mml ● Information: ○ Start- and end times of upcoming batches ○ Equipment these batches will run on ○ BatchID, recipeID and recipe version
- 25. FACTRY MES ERP DATA Closing the loop
- 26. Step 2: get recipe data ● Recipe data contains all reference values ○ Material feed ○ Critical process parameters ● Used for: ○ Providing set points to the PLCs that will control the production process ○ Providing reference values to compare with the actuals recorded during the production process
- 27. FACTRY MES ERP DATA RECIPE DATA Closing the loop
- 28. Now we know: ● When we will produce ● What we will produce ● How to identify what we’re going to produce ● Where we will produce (fermenter) What’s missing? ● The actuals!
- 29. Step 3: dispatching of batches ● Upcoming batches are synched from planning / ERP system ● Because we know the batchIDs, expected start- and end times, recipes… ● The operators can just press “Start” and the SCADA system and PLCs are loaded with the recipe data. Error-free link between planning and production: No more manual selecting of a recipe or typing in a BatchID
- 30. SCADA FACTRY MES ERP DATA RECIPE DATA Closing the loop OPERATOR INTERFACE PLCs
- 31. Step 4: collecting process data ● High resolution data is gathered from different PLCs ○ Broader than just fermentation ● Dashboarding and process analysis ● Data source for storing what actually happened during production
- 32. ● Supported by newer PLCs directly, or via SCADA system OPC-UA server ● Collecting metrics: ○ Polled (typical for sensors: e.g. every 5 sec, every min) ○ Monitored (typical for states or valves: on-change)
- 34. Collect data from production equipment Store it in a time-series database InfluxDB Visualize it with web-based visualization tools
- 35. Collectors Talk industrial protocol (e.g. OPC-UA) on one end and HTTP on the other Store-and-forward ● Local buffering ● Keep a local copy of configuration
- 36. The importance of naming ● Hierarchical structure ● AREA.EQUIPMENT.SENSORID ● Benefits!
- 37. PLCs FACTRY HISTORIAN SCADA FACTRY MES ERP DATA RECIPE DATA Closing the loop OPERATOR INTERFACE
- 38. Step 5: fermentation sheet completion ● Sampling of process data at regular intervals to replace a paper fermentation sheet. ● Operator is partially relieved of repetitive tasks ● Automatic marking of critical process parameters that deviate too much from the expected values from the recipe
- 39. PLCs FACTRY HISTORIAN SCADA FACTRY MES ERP DATA RECIPE DATA Closing the loop OPERATOR INTERFACE
- 40. Step 6: linking LIMS data ● Lab performs analyses for each Batch. This information is synched with all centrally collected data. ○ Yield ○ Dry matter content ○ ...
- 41. Step 7: answering questions with reporting! ● Give me all batches from February 2021 that followed recipe X and had a dry matter content of at least Y. ● Show me how the reference curve of parameter X evolved for a specific batch. And how does this compare with the reference values from all batches that followed the same recipe? ● Or… show me the raw process data for batch X in Grafana.
- 42. PROCESS ANALYSIS PLCs FACTRY HISTORIAN SCADA FACTRY MES OPERATOR INTERFACE REPORTING TOOLS ERP DATA RECIPE DATA Closing the loop
- 43. Open technologies and standards ● Vendor independent ● Source code available ● Has been running in production for years
- 44. The Digital Factory We look at the world of industrial automation through the eyes of an IT company. Extract process data to IT level as soon as possible. Because at that level, the biggest progress is being made. OT IT DIGITAL FACTORY
- 45. Some unexpected benefits 1. MRP functionality: ○ Prediction of material and air usage 2. Real-time machine monitoring ○ Valve maintenance 3. Debugging of cold-room door (anomaly detection) 4. Monitoring of the MES and Historian
- 46. Some unexpected benefits 1. MRP functionality: ○ Prediction of material and air usage 2. Real-time machine monitoring ○ Valve maintenance 3. Debugging of cold-room door (anomaly detection) 4. Monitoring of the MES and Historian
- 47. MRP We know: ● Upcoming production And therefore: ● Expected material use and compressor load As well as ● Current tank levels
- 48. MRP functionality in Grafana
- 49. Some unexpected benefits 1. MRP functionality: ○ Prediction of material and air usage 2. Real-time machine monitoring ○ Valve maintenance 3. Debugging of cold-room door (anomaly detection) 4. Monitoring of the MES and Historian
- 51. Some unexpected benefits 1. MRP functionality: ○ Prediction of material and air usage 2. Real-time machine monitoring ○ Valve maintenance 3. Debugging of cold-room door (anomaly detection) 4. Monitoring of the MES and Historian
- 53. Some unexpected benefits 1. MRP functionality: ○ Prediction of material and air usage 2. Real-time machine monitoring ○ Valve maintenance 3. Debugging of cold-room door (anomaly detection) 4. Monitoring of the MES and Historian
- 54. Monitoring of the MES system
- 55. And finally, our advice Just give it a try! https://www.factry.io/blog/process-data-integration-open-source-or-proprietary-software/
- 56. Wrapping up ● We are now able to answer the 3 questions in a reasonable amount of time ● The human has become a user of the information flow, not the centerpiece ● Additional benefits because of a common data platform ● All of this with open protocols and open source software
- 57. Takeaways 1. Build a platform for your process data, not a collection of point solutions 2. Think of your naming structure 3. Work iteratively with all stakeholders 4. Data has a clear impact on the business
- 58. Thank you! Questions? Further reading: https://www.factry.io/blog https://medium.com/factry https://www.linkedin.com/company/factry.io IVO LEMMENS - Algist Bruggeman FREDERIK VAN LEECKWYCK - Factry +32 474 88 85 73 frederik.vanleeckwyck@factry.io