IIoT Maturation Coming?

By Grant Gerke, Contributing Editor

February’s inaugural “Industrial Internet of Things” (IIoT) column discussed how the massive move to more sensors and analytics in manufacturing isn’t just a passing fad: It’s transformative. How do companies implement a data strategy with current production systems in place?

Each company needs a starting point in the IIoT journey, but a fully realized data strategy is hard to wrap your arms around today—and was even harder in 2012. That’s when Southern Company, Atlanta, an energy producer and transmission line supplier, decided to tackle the problem. The company is a large energy player in the deep-South region, with 27,000 miles of transmission lines that run through Georgia, Florida, Alabama, and Mississippi, in addition to operating several natural-gas and generation assets.

In a recent manufacturing webinar, Elizabeth Bray, principal engineer at Southern Company, discussed some newly enacted pilot projects involving the corporation’s transmission businesses and the move toward condition-based monitoring for its transformers at more than 3,700 substations.

Before the recent pilot, Southern Company began to add sensors and monitoring capabilities to make a future business case for a centralized program. Southern Company uses the eDNA data historian and PRiSM modeling from Schneider Electric for its transformers. These tools allow operations and maintenance teams to organize data into easy-to-read charts on monitoring screens and identify rates of changes or current deviations for its assets.

One example of success in the recent pilot program alerted a maintenance engineer to capacitor issues with a particular transformer. The eDNA trend tool and PRiSM modeling allowed centralized monitoring teams to identify a rate-of-change alert and allow maintenance to be performed before a peak period could cause downtime.

The eDNA trend tool and PRiSM modeling allowed Southern’s centralized monitoring teams to identify a rate-of-change alert and allow maintenance to be performed before a peak period could cause downtime.

This is a great example of software and platform analytic delivering on a large sensing development. In Maintenance Technology’s “Final Thought” column, guest columnist Rene G. Gonzalez noted that this type of trend is quite pervasive in the energy industry. As an example, he cited a typical refinery as increasing its number of sensors from 20,000 five years ago, to 100,000 today.   

Some industry observers, such as Joe Barkai, former VP of Research at IDC, Framingham, MA, are pushing for standardization of instrumentation and devices to reduce costs for manufacturers. According to Barkai, “There aren’t enough standards for the industrial IoT space, and the robust use of standards is critical to accelerate innovation and scalable IoT ecosystems.”

While Barkai is right, most enterprises need solutions now to visualize trapped machine and system data for maintenance teams. With the increasing number of mergers and acquisitions added to the mix, large manufacturers are now assimilating disparate platforms and control architectures to the current plant-production systems.

John Rinaldi, president of Real Time Automation, Pewaukee, WI, spells out specific problems for manufacturers using older controllers in a recent article, titled, “Mining Manufacturing Data | Leveraging Trapped Data for Results” (automation.com, Aug. 21 2015). “Many controllers,” he wrote, “do not have the software and hardware to communicate data to asset-management and information systems using current computing methods.”

Beside the exceptional computing power of the cloud, industrial networking is another huge component of IIoT. Rinaldi pointed to the advantages of intelligent network gateways, which can “extract information residing in PLCs and communicate data to maintenance-management or asset-management systems.” This allows disparate networks or systems to communicate and even perform math functions on process data and send email alarms to maintenance technicians on changes-of-states.

Operations and maintenance now can measure machine cycles, runtime, and other data to perform predictive maintenance without disrupting control architectures and plant performance. Also, a minimal capital investment solution holds water with management. MT

Grant Gerke is a business writer and content marketer in the manufacturing, power, and renewable-energy space. He has 15 years of experience covering the industrial and field-automation industries.