R-evolution: Powering a sustainable future with Google and Hexagon technology
About R-evolution
R-evolution’s mission is to mobilize tech-enabled solutions that can reverse and reduce the impacts of CO2 emissions, speed global transition to renewable energy sources, protect oceans and other natural habitats, ensure biodiversity, and eliminate waste.
Tell us your challenge. We're here to help.
Contact usR-evolution is leveraging Google BigLake to develop a global blueprint that decreases green premiums and accelerates the renewable energy transition.
Google Cloud results
- Early identification and quick resolution of incidents
- Reducing energy yield losses and decreasing total operational cost
- Enables quick innovation via AI/ML-driven use cases (days instead of months)
- Validation of accuracy of AI/ML model
- Global blueprint with Hexagon's sensors, software and autonomy solutions with Google's BigLake and AI/ML use case enablement
Rapid innovation blueprint for energy forecasts and anomaly detection to scale across thousands of renewable energy sites
Scientists agree that to avoid the worst effects of climate change, we must decrease global carbon emissions. One way forward is to shift electricity generation from fossil fuels to renewable sources, such as wind and solar power. "Getting there entails tackling one of the biggest threats to the planet: the belief that someone will save it on our behalf," says Erik Josefsson, CEO, R-evolution, the sustainable innovation and green-tech investment subsidiary of Hexagon.
R-evolution is rising to the challenge of saving the planet by using data and technology innovation and creating the right incentives for companies to place sustainability front and center of their business strategies. The ability to innovate faster is crucial to the success of technology-driven reductions to carbon emissions and limiting global warming to 1.5 oC. A good place to start and bring more companies aboard this mission is by decreasing renewable energy yield losses and operational costs while increasing efficiency.
To make that possible, R-evolution is applying AI to solar photovoltaic (PV) parks with a goal to accelerate renewable energy transition around the world by reducing green premiums (the difference in cost between a product that involves emitting carbon and an alternative that doesn't). Seeking collaborators that are able to handle and analyze the large amounts of data needed to build the blueprint, R-evolution reached out to Google Cloud.
"We are not in the business of becoming a big utility provider. We are here to find the best cutting edge technology ingredients that, combined, have an efficiency impact on renewables, package them as a recipe/blueprint and scale across thousands of solar parks together with partners. When we met the Google Cloud team, we immediately thought, ‘with these people, we can make big things happen.' Secondly, when you think of data scaling capabilities, Google Cloud immediately comes to mind. So when BigLake launched, we saw it as a potential super-ingredient for our solar PV innovation and anomaly detection recipe," explains Josefsson.
"When you think of data scaling capabilities, Google Cloud immediately comes to mind. So when BigLake launched, we saw it as a potential super ingredient for our solar PV innovation and anomaly detection recipe."
—Erik Josefsson, CEO, R-evolutionImproving renewable energy efficiency with AI/ML and BigLake
R-evolution aims to enable renewable energy managers to access quick and reliable insights about what's happening in their solar park. Although many tools exist to support that, connecting each one individually to the solar park's systems is time consuming and expensive. "Hexagon has tons of data points, systems, sensors, and more than 250 technology products that we can leverage to support green initiatives," explains Javier Garcia-Norro, CPO, R-evolution. "We currently use Hexagon's Smart Digital Reality sensor, software and analytical capabilities. Xalt, Hexagon's technology platform that enables solutions to work seamlessly together, is the integration layer of data that flows from sensors and external sources to present the analyzed data in a single interface. What we were missing was anomaly detection and other advanced AI use cases. For that, we required a data lake capable of handling all data, real-time and historical, and providing the AI/ML tools for quick use case development and innovation."
R-evolution is not stopping there, it wants to enable accurate energy forecasting based on weather and also historical data. By analyzing how the solar park has been performing for the past few months and years, an ML model can be trained to accurately detect anomalies based on what the current production should be, compared to its historical performance under similar circumstances. If that expectation is not met, workers can find and fix the issues causing that anomaly. These parameters can also power predictive analysis, which enables workers to fix and exchange machine parts to prevent downtime and avoid energy cuts when they identify that an incident is bound to happen based on historical patterns.
While Hexagon's advanced sensors, monitor the site and Xalt consolidates all SW systems, BigLake brings in the historical data of that park to power these forecasting and anomaly detection analyses. It was only a matter of weeks until this setup was up and running. "With the right platform we can appreciate the real value of speed of innovation," says Josefsson. "BigLake was the perfect match for our use case because it brings to the table agility, AI/ML capabilities to process large amounts of real-time and historic data, and infinite scalability to power more use cases in the future. Besides, we wanted to collaborate with someone who is as committed to helping the planet as we are."
With BigLake, R-evolution can unlock analytics on distributed data regardless of where and how it's stored. Powered by the same technology that underlies BigQuery, the storage engine provides high query performance while also unifying data warehouses and lakes. This way, connected with Hexagon's sensors and data points from power plants, it can support energy forecasts and anomaly detection analytics in a central place. R-evolution has also integrated BigLake with R-evolution's mobile app (built on Xalt) to display the resulting insights in a single pane of glass for end users such as solar PV park managers. With full visibility over what's happening in real time at their park, and how that fares compared to historical performance, they can make better-informed decisions that lead to higher energy yield and better revenue management. "Instead of monthly reviews, we can now do daily reviews to prevent incidents or resolve them faster," explains Josefsson.
"BigLake was the perfect match for our use case because it brings to the table agility, AI/ML capabilities to process large amounts of real-time and historic data, and infinite scalability to power more use cases in the future. Besides, we wanted to collaborate with someone who is as committed to helping the planet as we are."
—Erik Josefsson, CEO, R-evolutionInnovating at scale to lower the green premium on renewables
Garcia-Norro estimates that, without the combination of Google and Hexagon tech, it would have taken months for a site manager to create a system that enables anomaly detection, or different AI/ML use cases. This would entail integrating the park's systems with different solutions that track and analyze data in real time, plus gathering historical data to establish the ML model parameters for the analysis. R-evolution's solution is an off-the-shelf alternative that is fast to implement and results in outcomes within a matter of days. "The first sign of success is how easy it is to integrate a new system to our solution. BigLake makes it simple to apply new ML models to historical data using standard SQL queries so that customers can customize their analysis based on their goals and needs, while Hexagon's Xalt platform makes the integration with the different systems transparent," says Garcia-Norro.
Additionally, PV anomalies are difficult to identify and usually their root causes are found late, requiring expert intervention and potential site visits. All this adds costs and decreases energy yield. But by having a reliable energy forecasting system, site operators can quickly identify any unexpected deviation between the forecasted value and the real one. "We estimate that asset managers can get three to five percent improvements in cost using our anomaly detection solution alone. These are big numbers in the context of solar PV performance, and even bigger if you consider that the solution can be applied to many other use cases such as alarm correlation and preventive maintenance, which all add up towards greater energy efficiency," explains Garcia-Norro.
Josefsson agrees: "Talking about a one or two percent improvement in energy yield may sound small at first, but if you consider that if this was applied across thousands of solar parks and renewable sites, it can make a big impact. We need to start somewhere, and we're here to collaborate with key partners such as Google to amplify that impact and save this planet."
"With this end-to-end blueprint solution, we can help more people to innovate and improve their renewable energy outcome by leveraging Google capabilities and Hexagon technologies through one easy-to-use interface. Anyone can use this blueprint to reduce energy yield losses and decrease total operational costs. The possibilities are endless."
—Javier Garcia-Norro, CPO, R-evolutionAccelerating the transition to green energy in a profitable way
Forecasting energy from solar power plants has traditionally been challenging, so R-evolution's priority for now is to use this prototype to solve that challenge and make solar energy generation more efficient in a scalable way. Google Cloud is one of many ingredients making this possible for R-evolution so it can act on its mission to accelerate the transition to green energy in a profitable way. The company is also looking to partner with more green-tech companies and planning to scale the blueprint solution so it can be applied to more solar parks as well as energy storage facilities.
Meanwhile, although a growing coalition of countries, cities, businesses, and other institutions are pledging to get to net-zero emissions, commitments made to date fall short of what is required for us to reach that goal by 2050. R-evolution is determined to build a means for renewable assets to become more efficient and have a big impact, faster. "We're improving the flow of bytes to manage, maintain, and digitalize renewable assets at speed. This includes using historic data that is already available in more efficient ways. Google Cloud is framing sustainability as a data challenge, and together we're creating the tools needed to enable more people and companies to act on that challenge," says Josefsson.
"With this end-to-end blueprint solution, we can help more people to innovate and improve their renewable energy outcome by leveraging Google capabilities and Hexagon technologies through one easy-to-use interface," says Garcia-Norro. "Anyone can use this blueprint to reduce energy yield losses and decrease total operational costs. The possibilities are endless."
Tell us your challenge. We're here to help.
Contact us
About R-evolution
R-evolution’s mission is to mobilize tech-enabled solutions that can reverse and reduce the impacts of CO2 emissions, speed global transition to renewable energy sources, protect oceans and other natural habitats, ensure biodiversity, and eliminate waste.