In a sprawling landscape of a windmill farm, how can one possibly stay on top of the performance of each turbine?
This is a business question that has plagued every energy company at the forefront of energy transition.
For energy companies with assets in operation, the biggest challenge is to avoid machine underperformance and possible downtime.
While energy companies grappled with this challenge, Jungle AI, a Dutch-Portuguese AI scaleup, was building technology that improved the operational performance of these capital-intensive assets.
Founded in 2017, Jungle AI saw that industrial machines became increasingly complex but the technology to operate and understand them did not keep up.
The co-founders observed that the discrepancy between human comprehension of the technology and the increasingly complex process led to machines not performing at their full capacities.
A data challenge
Every company and industry faces a data challenge where there is always a gap between what businesses do with their data and what is actually possible.
For the co-founders of Jungle AI, the challenge in the energy industry was essentially a data challenge and one they saw possible to address with data analysis technology.
“Although this problem exists in nearly every industry, Jungle AI focuses on those that are pivotal for a sustainable future,” says Silvio Rodrigues, co-founder and Chief Innovation Officer of Jungle AI.
He says many industries face significant data challenges because of the complexity of their data environment.
Since modern machines have become increasingly sophisticated, Rodrigues says they are also much harder to understand due to more sensors, interaction between components, and higher data resolution.
This limits companies from performing adequate monitoring and advanced analytics. As a result, machines either underperform or are left idle, leading to a lot of value left on the table.
To overcome this data challenge, Rodrigues argues there is a need for specialised expertise and advanced technology solutions that ensure “data collection, processing, and analysis.”
Data challenge meets AI solution
AI is seemingly everywhere. From helping us get to our most used apps easily to smart parking, AI is bringing about a big shift.
It should thus not come as a surprise that it is an AI solution that helps overcome the data challenge faced by industries with assets like wind turbines.
The solution built by Jungle AI uses deep learning models based on the transformer architecture to learn how machines should behave under all thinkable circumstances.
Rodrigues says, “This normal behaviour creates a much needed baseline that operators use to benchmark the actual behaviour of their machines.”
Once the baseline is established, Jungle AI’s customers can monitor the performance and health of their machines in real-time on a web tool called Canopy.
The Dutch-Portuguese scaleup does not stop there. It also provides access to its models via API to customers looking to model its results into their existing workflow.
It also has a team of data scientists that work with customers to help create a better understanding of their assets and solve critical issues.
For the layman, Jungle AI provides companies with predictive maintenance that leads to reduced downtime and maintenance costs.
Its AI models work in real-time to predict equipment failure before it happens by helping customers with prediction of wind turbine gearbox or bearing failures weeks or months before any other systems pick it up.
“This allowed our customer to order spare parts and plan the repair without having to take the turbine out of operation in high wind times,” says Silvio.
Their models also identify, classify and quantify underperformance on wind and solar farms around the world.
AI awareness challenge
Since the debut of OpenAI’s GPT-3, there has been a lot of talk around generative AI but it fails to underscore the impact AI has already had in a number of industries.
Silvio says the latest AI developments including release of generative AI tools has led to an increase in the general public’s attention to the world of AI.
But he fears that the ongoing chatter around generative AI could lead to a one-sided perspective of AI.
How should one see AI then? He says AI is a broad and evolving topic with scope dependent on the problem being solved.
For Jungle AI, the biggest challenge was also making companies aware of the benefits of using AI to optimise their operations.
“It was challenging to convince them of the value of our solution,” Silvio quips.
He says poor quality data available from equipment and machines only made it difficult to develop accurate predictive models.
“We had to develop tools to integrate data from different systems and ensure that the data was properly formatted and cleansed before using it in their analytics models,” he adds.
Another challenge, he says, was finding and hiring the right talent with skills and expertise required to build and deploy AI and ML solutions in industrial settings.
Amidst all the hype around generative AI and its possible impact on jobs, the lack of right talent or awareness is probably being left out of conversation.
Scaling with AWS
Like many fledgling AI scaleups, Jungle AI also relies on Amazon Web Services (AWS) to build and deploy its platform quickly and efficiently.
Silvio says with AWS, they have access to a highly scalable infrastructure that can accommodate their growing customer base and the increasing amount of data they need to process.
“AWS provides robust security features and compliance certifications, which are critical for companies operating in regulated industries,” he says.
Amazon Web Services has also played a crucial role in connecting Jungle AI with leading energy companies around the world.
The scaleup is also part of AWS Clean Energy Accelerator, a programme that aims to accelerate the adoption and development of clean energy technology by individuals and organisations.
“They provide mentoring, credibility and network,” Silvio says before adding that it is the push they need to scale the company.
He further notes that they have been able to co-innovate on improving user autonomy with mentors and AWS experts.
With AWS Summit returning to Amsterdam after a three year absence on June 1, Jungle AI CEO Arnoud Kamerbeek will be there to not only talk about the progress but how the scaleup will further accelerate energy transition.
Want to know more about energy transition with Jungle AI? Come visit their presentation at the Startup Loft Theatre during AWS Summit Amsterdam
AI for good
There is no better way to put it but Jungle AI really stands out as a platform that is spearheading the idea of AI for good.
As a transversal technology, AI was envisioned to transform lives and with its AI tools, Jungle AI is playing a crucial role of making energy transition possible.
By extending the life of renewable energy sources and detecting issues in these assets before they occur, Jungle AI not only guards these capital-intensive resources but also increases their value.
In September last year, the company raised €5M in Series A funding to further accelerate this transformation.
At the time, the startup planned to hire for 12 new positions but Silvio says they only have one position open for a DevOps Engineer.
There are businesses with impact and then there are businesses with social impact. Jungle AI is clearly in the latter where its business value derives from the direct and indirect impact it has on the planet.
If a renewable energy company’s assets are functioning well then there is a good chance that Jungle AI’s models are at work.
AWS Summit Amsterdam 2023
Want to learn more about the benefits of the cloud, or talk to one of their experts? Visit AWS Summit at the RAI in Amsterdam on June 1 to learn about the platform capabilities, get free 1:1 support from AWS experts, and discover more resources. Also, don’t forget to join AWS for pre-summit drinks at the Rooftop on May 31.
Read the orginal article: https://siliconcanals.com/news/startups/jungle-ai-canopy-energy-transition-technology/
