Do you and your company have stories to tell about digital transformation in the power sector? POWERGEN International’s call for abstracts is now open and seeking sessions prospects for the event December 8-10 in Orlando. Click here to see the categories and to submit an abstract! The theme of POWERGEN 2020 is “Shaping the Future of Generation Together!
— — — — —
If the 2010s had to have an overly simplistic and cliched theme, it would be something along the lines of “disrupt or be disrupted.” From the taxi industry’s upheaval to the fundamental shift in how information is stored, shared, and created, the catalyst of software and technology transformed just about every aspect of our lives.
In this time, the power industry morphed from a sector with a traditionally analog and siloed set of workflows to something substantially more complex and intertwined. Distributed generation resources, renewables and smart grids complicated the traditional transaction, requiring entirely new capabilities and infrastructure to support these capabilities.
The U.S. Department of Energy estimated that between 2008 and 2017, $31.6 billion was invested into smart grid technologies. Indeed, the power industry is well into a period of unparalleled digital transformation that is reshaping the sector as a whole.
At the heart of digital transformation is data. There was a significant increase in the number of IoT devices – both within power generation units and into the grid – and the volume of data that must be collected, managed, analyzed and converted into action.
At the same time, advances in cloud computing created a reality of on-demand hyperscale. Previously unattainable to all but the most well-resourced operations, computer workloads could now be spun up and down to accommodate the storage and management demands of IoT data, in addition to analytic techniques like deep learning or neural networks. Suddenly, a whole new world of opportunities to understand our power operations and networks was at our fingertips.
These forces accelerated over the last few years, and increasingly sophisticated technologies emerged, uncovering opportunities to improve operational efficiencies, experiment with new business models and breathe new life into a traditionally slow industry. The 2010s were an exciting step forward, but there’s much to be explored in the coming decade.
Though it’s difficult to predict precisely where innovation will take us next, as we move into the early 2020s, there are a few key trends that are likely to gain ground.
Greater Emphasis on Turning Mountains of Data into Actionable Content
What was once an undervalued output of power operations has become the focus of its entire transformation. Utilities has always analyzed data, but digital transformation will enable the sector to turn mountains of stored data into a meaningful asset. More than curated data in a platform, data will be infused with domain-specific intelligence to create solutions that are functional off the shelf and presented within the standard run of operations for a utility.
We’ve already seen the impact that sensors have had during weather activity, from identifying vulnerabilities in the grid to sending feedback to utility companies. Having the data on hand, in real-time, enables seasoned operators to respond reactively, but we unlock the real value of data when we’re able to be predictive and prescriptive. Predictive analytics will help move the industry beyond looking for the equivalent of red blinking lights before taking action. We’ll arrive at a place where it will be commonplace to take decades worth of outage, asset, geospatial, distribution, and workflow systems data and overlay it with weather forecasts and other third-party data to better predict the sensitive areas of the grid before an event happens.
This next wave of intelligence will also be prescriptive, enabling us to deploy the right specialists or actions at the exact moment when and where they are needed. The analytic outputs will be instantaneous, thanks in part to 5G, reducing the amount of effort required to identify a potential situation, determine its potential impact, and act accordingly.
Efficient Enablement of the Right Machine Autonomy
As generation and grid operations continue to decentralize, so too should the software and analytics that are providing its support. The days of having all operational workloads execute on premises are over. Solutions of tomorrow will embrace the control of on premise, the computing power of the cloud, and flexibility and speed at the edge.
Though each environment has its strengths, the push to the edge will radically transform some utility operations. In time, transactions that require near-instantaneous diagnosis and mitigation will be executed without any human intervention as more monotonous activities continue to be abstracted. This machine autonomy will allow the workforce to focus on solving problems that the human brain is more adept at – ones requiring creativity and multifaceted problem solving.
Security will continue to be of critical importance and embedded into the design of both the products and systems used by utilities. The desire for a more secure infrastructure will accelerate decentralization and autonomy, where breaches can be immediately identified and quarantined in the same way a fault is located and isolated in today’s grid. By pushing more workloads away from a central command, we will limit the impact bad agents can have on a geographic area, reducing the draw of hacking certain systems.
Critical Need to Prepare the Next Generation Power Workforce
The energy industry is expected to lose a large share of its workforce as millions of experienced professionals become eligible for retirement. An estimated 25 percent of U.S. employees in electric and natural gas utilities will be ready to retire within the next few years – taking decades of institutional knowledge with them. It’s true that newer generations bring modern technical ingenuity to the table, but we risk losing valuable deep domain expertise and specialized skills once seasoned workers leave the workforce.
Is the power generation sector prepared to compete for a new generation of skilled workers? And how do we effectively knowledge-share once they arrive? Institutional programs to pass on know-how are essential, but likely won’t be enough to bridge the gap between theoretical knowledge and instinctual knowledge gained through years of on-the-job experience.
As we face more gaps in the workforce, we’ll turn to digital transformation to not only transform the data that is collected but to codify much of the expertise of longtime workers. Technology won’t diminish the need for a robust workforce. Rather, it will enable the discovery of domain-specific, intuitive knowledge as data is processed. Once this context is embedded in the software, new and early-in-career employees can make quicker, more informed decisions.
Slower, But More Meaningful Paths to Progress for Today’s Hot Technologies
Old habits die hard, and the processes of the past can be difficult and expensive to break. Exciting technologies such as AI and blockchain are attractive but have a significant impact on business processes that may be long-standing. A successful digital transformation initiative must also address operational and cultural transformation. Early adopters learned the hard way that if workflows are redefined and historical processes eliminated, the resistance will be swift and merciless.
Rather than implementing entirely new platforms with all the bells and whistles, it will be essential to have a deep understanding of the software and processes that are already in place. We’ll see more solutions that pair well with existing systems for the sole intent of reducing operational friction. The first change may be as simple as adding new capabilities around predictive analytics embedded into the front-ends of current operational systems – within the framework that workers know and trust.
The workers in the field and the office need to know that the asset performance management tools are first and foremost accurate, and that the tools fit appropriately within their standard workflow processes. As their trust levels grow, employees will begin to see firsthand the return on investment in time savings alone. Thus, dramatically increasing the likelihood and speed of adoption.
AI and ML Will Move Beyond Hype to Reality
Artificial intelligence, machine learning, and predictive analytics technologies have the potential to transform the sector. A potential so promising that it’s tempting to think that these technologies and those still to come will solve all problems. The truth is more measured: AI and ML technologies are arrows in our quiver, but they are not the arrow.
While innovations can certainly take utilities to new heights, technologies are too often adopted based on hype instead of taking a deliberate, well-thought approach. One of the biggest missteps is to implement new technology without first understanding the problem that needs to be solved. The rise and subsequent stall of blockchain taught us that early acclaim means little when proper regulations and infrastructure aren’t in place to support it. While its potential is undeniable, adoption has been sluggish at best and premature at worst.
Looking ahead, the realities of technologies, such as AI and ML, will be evident. As more utilities and their vendors become comfortable with what these technologies can (and cannot) do, they will better understand how to wield them. We’ll see a more holistic approach that keeps safety, reliability, usability, and affordability at the forefront of decision-making, rather than just investing in tech for tech’s sake.
The future is uncertain, so we know there will be surprises along the way. If we learned anything from the past decade, the 2020s promise to be an exciting and transformative era.
About the author: Matt Schnugg is Vice President of Data Science and Analytics at GE Digital.