Digitalization Is Now a Power Sector Imperative: Takeaways from Connected Plant Conference 2023
Digitalization is now so firmly embedded in the power sector that if participants haven’t yet begun their digital journey, they’ve inherently fallen behind. That was a message repeated at the three-day 7th annual Connected Plant Conference (CPC), which took place this week in New Orleans, Louisiana.
POWER magazine and Chemical Engineering magazine’s conference this year focused on the intersection of technology, process, and people. At a high level, the agenda—shaped by a conference committee of digitalization experts from the power and chemical process industries—strove to address practical concerns across the two industries. A key tenet was to facilitate information sharing, including best practices, lessons learned, and predictions for the future, all through the lens of real-world case studies.
Key Takeaway: Modernize or Get Left Behind
Compared to past CPC events, this year’s conference began and ended with the distinctive acknowledgment by many of its participants that digitalization in the power sector is well underway, and it already forms a basic framework for modern power system operations.
Speakers and attendees, industry professionals who tend to be typically well-versed in digital speak, warned peers who hadn’t yet embarked on the digitalization journey that they were lagging—competitively, environmentally, and even socially. And for those already on the journey, “once you start, you’re never done, so just be prepared for that,” said Dr. Brian Fitch, a consultant at Barr Engineering, who kicked off workshops on Sunday afternoon.
Throughout the conference, several speakers illustrated why digitalization was proving imperative for the energy transition. Sam Bainbridge, director of Technology Delivery for SKF USA, a firm that delivers products and technology to original equipment manufacturers, pointed to labor and skill shortages across various industries. “With that, [industry is] facing continued escalated costs, lots of stripping post-COVID by inflation, but we’ve also got budget constraints,” he said.
Compounding that are supply chain concerns, which are affecting the timely replacement of parts, particularly during unplanned downtime events. “There’s a lot of pressure” to become more efficient and embrace new approaches, said Bainbridge. Leveraging technology, such as advanced wireless sensors, could ultimately help companies overcome some of these issues, including prioritizing maintenance efforts, optimizing reliability programs, and driving cultural change, he suggested.
Mike Greene, a retired operations technology lead for Nova Scotia Power, meanwhile, emphasized a need for transformational continuity, pointing to a need to capture knowledge lost by the departure of experienced personnel. “How do you address that? You have to have programs and processes.” Nova Scotia Power is the province’s sole power provider, he noted, and it began its digital journey in 2015 to support its crucial reliability and efficiency imperatives.
While Greene stressed the importance of trusting and contextualizing data, as well as integrating various point solutions and systems within the organization, he also highlighted the need for a comprehensive suite of solutions, choosing reliable vendors, and ensuring seamless integration between systems. “Now the mistake that a lot of people make is you can’t wait for the perfect solution,” he said. “What I’ll tell you now is if you aren’t already on the cloud, if you aren’t piloting advanced pattern recognition, and if you’re not dabbling in artificial intelligence (AI), you’re behind the curve.”
Jared Kolwyck, a plant manager at DTE Energy’s three-year-old Central Energy Plant in Dearborn, Michigan, (a 2020 POWER Top Plant winner), suggested digitalization has been integral in optimizing equipment operation, maintenance, and proactively identifying issues. At Tennessee Valley Authority (TVA), digitalization has helped boost the federal corporation’s multi-pronged priorities as the grid evolves, noted Jason Krupp, senior project manager for grid modernization. Those priorities include reliability, affordability, efficiency, resiliency, and security for all its assets, from generation to transmission and distribution, as well as balancing supply and demand from its 153 local power companies. Efforts, for example, support integrated planning and production standards and requirements, Krupp said.
Pervading Challenges: Data Anarchy, Cybersecurity, Interoperability
Foremost among pervasive digitalization challenges that CPC 2023 speakers discussed was the tendency toward “data anarchy.” Renner Burkle, a technical program manager from Casne Engineering, summarized three increasingly urgent business needs: enabling advanced analytics, centralizing data from multiple sites, and the ability to share data securely and easily. Sean Ely, head of product at TDEngine, an industrial time-series database specialist, in addition, highlighted case studies that tackled three crucial data challenges: efficient data mining, data alignment, and data inertia.
Data inertia, Ely noted, “revolves around the idea that it takes considerable effort to do useful work with large amounts of data, similar to the physics concept of inertia.” One example involves the process to extract, transform, and load (ETL) data, “which can be difficult with very large volumes of data,” he said. “From my experience, trying to build advanced analytics is near impossible without first addressing the processes of data engineering, and bringing either your data to your analytics or your analytics to your data.”
Data sharing can be as cumbersome, owing to its complexity and concerns related to ownership, validation, and cybersecurity, Burkle and Ely suggested. Efficient and scalable solutions that consider future requirements must emerge as an industry best practice, they added.
During a discussion dedicated to cybersecurity—another overarching challenge—panelists highlighted the importance of collaboration to ensure systems are both secure and functioning optimally. The wide-ranging conversation was led by Siemens Energy’s Jonathan Tubb, Southern Co.’s Business Technology Planning and Strategic Initiatives Director James Goosby, and Dwayne Edwards, a security engineer at operational technology (OT) security firm Tenable. Panelists shed light on emerging concerns related to communication, risk management, and vulnerability disclosure. While they highlighted the importance of collaboration and trust among individuals and organizations, they also underscored a need for suppliers to understand the OT environment. Another prominent OT cybersecurity highlight involves the alignment of evolving skillsets required to integrate IT systems into OT environments.
CPC 2023 speakers also shed light on yet another integral challenge facing all industrial entities: interoperability. Interoperability involves the easy and secure exchange between diverse platforms from multiple vendors to allow seamless integration of those platforms without costly, time-consuming software development. An increasingly important effort to implement an interoperability standard is being fostered by the OPC Foundation. As Michael Clark, director of the OPC Foundation’s North American segment, described it, the OPC standard, first released in 1996, is a series of specifications developed by industry vendors, end-users, and software developers. The specifications “define the interface between clients and servers, as well as servers and servers, including access to real-time data, monitoring of alarms and events, access to historical data, and other applications.”
David DeBari, a process control engineer at ExxonMobil Corp.—“a very conservative and very security-sensitive company,” he noted—revealed that to encourage adoption of the standard at his company, his team showed leadership that new technologies, open architectures, and standard-based models actually have better security. “It’s built into expectations, and the way we approach this is, in the second that you power that device, it needs to meet security needs—not waiting for you to load a piece of software or put another buyer’s data file on there,” he said. “Orchestration is another tool. Updating this stuff needs to become more normal and routine,” he added. As an added bonus, the company found that it could “do things with new technology and equipment that it could never do with the old equipment.”
A Realm of New Possibilities
Still, as with other technology conferences, optimism was in no shortage at CPC 2023. Speakers described emerging possibilities for dynamic innovation in the power sector, including AI, cloud capabilities, advancements in drones, wearables, robots, other connected devices, and other new approaches. In a keynote speech kicking off the conference on Sunday night, Jonathan Alexander, manager of Albemarle’s AI and Advanced Analytics group, detailed the specialty chemicals manufacturing firm’s use of AI to fuel growth and productivity.
Significant optimism, notably, also derived from discussions about the speed and allure of generative AI. William Hendricks, vice president of manufacturing for Cognite’s Americas Sales, emphasized a growing market need for simplicity. “Previously, AI would find patterns or identify things, but it didn’t generate anything new. And the development of very, very large language models have enabled machine learning to generate new content.” While the model is essentially a statistical model that makes predictions based on rationale, it can “hallucinate,” meaning it returns inaccurate responses, he noted. “Obviously, if you’re in an engineering environment, if you’re trying to solve operational issues, those aren’t good.”
However, those challenges can be tackled by honing in on smaller data subsets that don’t have built-in syntax to give it context, he said. Cognite has notably been making gains on that front, he noted. Last week, the Austin-based industrial software firm launched an “intuitive, composable, visual workspace” dubbed “Industrial Canvas” where users at all levels of the organization can view all data types, such as engineering data sources, 3D models, piping and instrumentation diagrams, historian data, work order data, systems applications, and products data. The collaborative environment leverages contextualized data and generative AI. “Industrial Canvas makes data ‘speak human,’ as for every one person that can ‘speak code,’ there are 100s who cannot,” explained Geir Engdahl, Cognite’s Chief Technology Officer. It “can now empower everyone utilizing data to inform their daily production optimization, maintenance, safety, and sustainability decisions with simple access to complex industrial data—in their language and on their terms.”
During another keynote speech, Alan Ho, an engineer and entrepreneur who formerly led product management at Google Quantum AI, led attendees through a definitive history of quantum computing, explaining concepts like superposition and entanglement. While error correction still “has a ways to go,” quantum computing’s potential for the power industry is beginning to thrive, Ho said. Advancements could bolster material sciences research and development, especially “around superconductors,” he said. Quantum-inspired physics-based innovations could also potentially speed up solving optimization problems, he said.
“We know that in the power industry, discrete optimization is very useful for the purposes of figuring out what turbines to turn on, which to turn off [aligned with] demand, offloading, all that kind of stuff,” Ho noted. But it could do more, he said. “When you’re doing this kind of optimization, you’re usually trying to find the lowest energy state. And the problem is that when you’re doing optimization, you might get trapped in what is called ‘local minima’ that you can’t get out of. And so you want to try to be able to somehow tunnel through these local minima. Well, it turns out that you can use a lot of research revealed from quantum computing to do this kind of tunneling. The combination of machine learning and these quantum-inspired computing allows you to do that.”
Ho ultimately posited that industry shouldn’t dismiss technical innovation given current practical limitations. “My prediction is that it’s a very exciting age in quantum computing because I think history is going to repeat itself. There’s going to be a few applications—they’re going to come out of nowhere, they’re going to be way more impactful than we think, and it’s going to create a new revolution in computing,” he said.
—Sonal Patel is a POWER senior associate editor (@sonalcpatel, @POWERmagazine).