At more than a century old, the U.S. power grid is creaking under the weight of increased electrification demands, prompting a slow-moving transformation of the aging infrastructure into a digital system powered by state-of-the-art automation technology, Big Data analytics and the Industrial Internet of Things (IIoT).
The legacy grid, based on a central generation, transmission and distribution model, was not architected to support modern-day electricity needs—whether accommodating peak demands for unpredictable renewable energy sources like wind and solar, or driving the burgeoning market for electric vehicles (EVs). Today, the transportation sector accounts for about 20 percent of global energy usage; however, as demand for EVs surges—more than half of all vehicle sales will be electric by 2040, according to the Bloomberg New Energy Finance 2017 report—the sector will require another 1,800 TWh of electricity, and that’s just a snapshot into one high-growth market.
At the same time, there is growing worldwide attention to extending electricity access to a much larger population. Industry estimates confirm there are more than 1 billion people on the planet still without access to electricity, and whole societies on the cusp of electrifying. Based on these and other growth factors, GE is expecting global electricity generation to grow by 2.3 percent annually over the next decade at the same time as global electricity use is rising by 26 percent.
There are other factors at play as well. Technology innovation in areas like miniaturization, digitalization, energy storage and pervasive networking are enabling a shift away from centrally located power generation to a decentralized model. Power generation needs would be distributed and handled much closer to where demand is, at the edge of networks, helping to increase efficiency and optimize energy usage. New offerings in areas like microgrids, distributed management software and IIoT-enabled predictive analytics are leading a sustained and protracted effort to transform the legacy grid into something that is smarter, more resilient and more sustainable in an effort to meet next-generation electricity needs.
“Being able to accommodate energy created in a more decentralized fashion is the story of a more complex grid,” says Mark Feasel, vice president of the electric utility segment and smart grid at Schneider Electric. “Automation is key to unlocking the ability to operate in this world of more complexity.”
GE’s Predix enables machine learning-based analytics that can increase the reliability and efficiency of the electrical grid.
Limitations of today’s grid
The shift to more variable renewable energy sources coupled with the transition away from analog control to digital control and communications is putting pressure on the existing electricity grid to perform in ways for which it was not designed. In a report assessing the potential for modernization, the U.S. Department of Energy describes the traditional architecture as large-scale centralized generation that operates remotely from consumers with limited energy storage, passive loads and no ability for two-way communications. A modern grid, in contrast, requires the ability to dynamically optimize operations and energy sources, rapidly detect and mitigate any disruptions or disturbances, and integrate hybrid energy sources. At the same time, the modern grid should enable consumers to manage their own electricity use so they can play a part in energy optimization and reduction of costs.
“There could be a ton of energy towards the edge of the grid, so the ability to flow that back to a substation and on to a larger transmission grid used in another location is a huge opportunity,” explains Steve Martin, corporate vice president and chief digital officer for GE Power. “But the grid was never designed to do this—it’s all about centralized generation and transmission, never a multi- or bi-directional flow.”
To get the job done, the Electric Power Research Institute (EPRI) and other experts are projecting up to $476 billion of new investment required to fund grid modernization initiatives that encompass intelligent control system architectures, advanced grid modeling applications and distributed generation functionality—all designed from the ground up with built-in cybersecurity protections. Intelligent digital devices—from smart meters to battery storage in EV charging stations and solar arrays—will need to enable a two-way exchange of information and energy so usage can be optimized and overflow can be redirected to where it’s needed. Consider this scenario: Using a combination of data analytics and smart meters, the excess energy generated by solar panels on a commercial building or home could be rerouted back to a local microgrid—a key component in a decentralized grid infrastructure—so the reserves could be harnessed elsewhere.
Schneider Electric designed and deployed two microgrids for Montgomery County’s Public Safety Headquarters and Correctional Facility in Maryland.
The technology pieces
Transformation of the electric grid starts with digitization, retrofitting or replacing components such as transformers or meters with sensors and greater network connectivity to add situational awareness to the asset to better understand its operation. Next is implementing automation technology—everything from predictive analytics and smart controls—to optimize the asset and enable a proactive response to avert potential downtime or maximize performance.
“As the world becomes more digitalized, it has a profound effect on how you think about electricity and creates a layer of knowledge that allows you to think about energy in new ways,” Feasel says. “As you begin to digitize the energy value chain, consumers can think about how much resiliency, sustainability and cost matters to them, and solve for outcomes.”
Microgrids and distribution management systems (DMSs) will play key roles in grid modernization. Microgrids are considered a more efficient way to distribute energy to local sources and can also function as a backup “island mode” source in case of a power outage or disruption on the larger grid. Encompassing an array of technology that includes advanced edge controls and cloud-connected optimization services, microgrids are being implemented by utilities, government facilities and manufacturers to enhance resiliency and gain energy independence and efficiency. For its part, Schneider Electric offers EcoStruxure, an IIoT interoperable framework that supports integration of both existing infrastructure and new technology to be designed into a microgrid architecture.
The DMS acts as an orchestration platform that employs rules to manage and govern the flow of energy. Machine learning capabilities will be an essential part of a DMS given the complexity and scope of the grid and the impact of external factors like the amount of sunlight or the intermittent demand for EV charging.
“It’s not possible to codify in rules all the possible scenarios on a second-by-second basis,” GE’s Martin says. “Managing all the data in real time is increasingly complicated. And, ultimately, machine learning is the only way to deliver models that govern all scenarios effectively.”
IIoT analytics will play a key role in the DMS and for energy orchestration. Let’s say that an EV is parked in a garage and it’s determined—through analysis of past behavior—that it won’t hit the road until 7 a.m. Based on that data, a utility could extend the charge over a longer period of time and spread the load to meet other demands. A similar load-shaping scenario could be applied to turning down air conditioning or heat a degree or two midday, while people are out of the home, and then leveraging the excess for peak demand in the evening hours.
“We can take advantage of the intelligence of devices to make sure consumers are getting what they need, but in a way that maximizes efficiency for the generation and distribution of energy,” Martin explains.
GE’s Predix IIoT platform, used as a backbone for its digital grid offerings and to create digital twins of the physical assets in the grid, can be leveraged across generation technologies and power plants to conduct analytics on asset performance management and operations optimization.
Storage—whether lithium ion, chemical or other battery technologies—is emerging as a crucial piece of infrastructure for responding immediately when there is a rise in peak demand, to store and distribute excess energy as required, and to fuel the energy requirements coming from the influx of EVs.
EVs present both a challenge and opportunity for the storage piece of a modern grid infrastructure, says Jeff Phillips, head of automotive marketing for National Instruments. The fairly random pattern of where EVs get plugged in and when they need an energy source brings both an increased load and a level of unpredictability to the grid. At the same time, he adds, OEMs are designing batteries so when they reach end of life on an EV, the high-density battery packs can be redeployed as additional storage capacity for the grid.
“The biggest implication of the EV trend isn’t that the total capacity will increase beyond limits, but that the peak capacity will,” Phillips explains. “Adding EV batteries en masse to the grid provides an additional source that the grid can push to and pull from to meet demands. Moreover, the intelligence that’s been designed into the grid interface could enable a variety of interfaces to help the grid determine which batteries can delay charging or even contribute power back to the grid.” At the same time, he adds, nearly every EV manufacturer has a plan to use the batteries for mass storage once their performance has degraded past the point of meeting the demands of the vehicle.
Siemens is doing its part to help cities map out a plan for their electric transportation infrastructure needs with its new eMobility calculator, which estimates the infrastructure requirements and potential import of electric transportation in cities over the next few decades. For example, a city like Los Angeles will need to install up to 100 EV charging stations per week, now through 2050, to accommodate demand in its geographic area.
“As the rapid influx of new technology like EVs hit, it’s harder for the system to react,” says John DeBoer, Siemens’ director of eMobility for the Americas. “It’s a tremendous automation opportunity and there needs to be a balanced core investment in the grid to bring it to the next level.”
Given the scope of what’s required, the electricity grid’s makeover is going to be a slow-moving transformation. Beyond the obvious technology challenges, there is the question of where the capital comes from to fund these massive upgrades and retrofits—a hurdle underscoring the importance of public/private partnerships.
“This is not a world where you can rip and replace everything out there,” Schneider Electric’s Feasel says. “The grid will be upgraded in a piecemeal fashion as it is fortified for the future.”
