VIA and Enel X: Bridging power company data and AI solutions
“Anyone can get lucky and make something new once. For us, we strive to find a faster, smarter, cleaner way of doing everything, all the time”
Kate Ravanis – COO & co-founder of VIA
Imagine a world where data didn’t exist; this would impact every portion of our daily lives. Without data, we wouldn’t be able to connect through our smartphones, deposit money easily and immediately to our bank accounts, find our way around the city or even have access to our health records.
Data is also the key to everything from digital transformation to the acceleration of the adoption of cleantech in the energy sector. In the case of utilities, companies are under constant pressure to reduce operation and management costs to maintain profitability while ensuring grid performance and improving customer affordability. These types of costs are made more efficient through emerging technologies. According to a report by the Analytics Institute at Deloitte, predictive maintenance increases productivity by 25%, reduces breakdowns by 70% and lowers maintenance costs by 25%. This means more uptime and fewer costs.
Although many utilities are already using data and AI to optimize processes, 90% of this data goes unused due to fear of compliance, risk of competition data privacy, and the time or costs of cleaning and aggregating this data. At Enel, data has always been a huge part of our business, from customer data to machinery and renewable product data. Improving the quality and use of data is a priority for us and collaborations with startups are a fruitful opportunity.
Making data universally digestible for Enel
Our ability to be efficient and accurate is directly related to our capability to process and analyze data across data sources, also known as “data ingestion.” Our services involve many different platforms, sometimes each with their own data format. David Rodriguez, Technical Program Manager of Innovation at Enel X, the world’s largest provider of demand response solutions, pointed out that “one of the biggest data ingestion problems that Enel X has is on the Demand Response (DR) settlement side of the business.”
To clarify the issue further, Rodriguez explains: “80% of the data comes from the largest utilities in our portfolio, which had already automated ingestion capabilities as they use a more standardized meter data format which tends to be consistent over time... [however,] the remaining 20% of Enel X’s DR meter data comes from smaller utilities who use multiple formats which vary over time.”
DR Systems and the Analytics team had to manually process this 20% of data due to their varying formats and this situation quickly rose to the top of our agenda as a priority to solve. Such activity is time-consuming, costly and repetitive. Thankfully, our collaboration with VIA allowed us to experience the benefits of applying cutting-edge artificial intelligence (AI) and machine learning (ML) algorithm to make our teams more efficient.
Data + AI = the future of Enel
We first connected with entrepreneurs and executives of VIA Colin Gounden and Kate Ravanis at the Enel Innovation Hub in Boston.
VIA was founded in 2016 and focuses on using AI to automate data privacy, data integration and data analytics. “We saw the power of AI and analytics to help improve the reliability and cost-effectiveness of clean energy,” explains co-founder Ravanis. To do this, VIA uses its core platform, Trusted Analytics Chain™ (TAC™), to obtain smarter data. They achieve this by allowing utilities to pool their data privately and securely. This generates a more rigorous analysis and crucial insights with benchmarked measurements, failures, condition change predictions and more.
Rodriguez reveals that he “was impressed by their ability to do multi-company analytics and predictive diagnostics in a trusted, secure, and privacy-preserving manner.”
Aside from using TAC™ as a foundation for meeting customer needs, VIA also offers other solutions to solve data analyst pain points. Analysts often transfer raw data in different formats to a database, known as data wrangling. The analyst usually identifies a rule or multiple rules to transfer this data. Sometimes this comes easy, but in many cases, it can be incredibly time-consuming. The big leap in AI is being able to process information without humans writing all the rules. For this reason, Via created its unique Smart Wrangling Engine for Extraction and Transformation™ (SWEET™) technology, which uses AI to take data from any kind of format and transform it into a standardized AI-ready format.
The impact of data and AI
After presenting an initial proposal, VIA had the opportunity to witness various Enel X use cases first-hand. The first project we collaborated on was a model that would show how Behind The Meter (BTM) storage impacts the condition of transformers. “We executed the project over the next year and during that time we started discussing other pain points we were having in Enel X such as ingestion capabilities from the DR team,” explains Rodriguez.
With VIA’s SWEET™ technology, we were able to automate the data cleaning and wrangle thousands of meter data files from Enel X customers. This solution worked by first using machine vision to “see” the data from Enel X, map it out, and delineate headers, blank spaces and any data in either columns or rows. The algorithm then used natural language processing which does another check of all data to confirm which are relevant. Finally, the SWEET™ technology used derived data separation, which helped to automatically recognize and separate columns and rows from the raw data.
“The impact of VIA’s technology was that it enabled significant time and cost savings, greater optimization of the data ingestion process, and overall improvement in data accuracy resulting in the capability to process a greater number of files from various sources,” notes Rodriguez of Enel X. According to Ravanis, “using SWEET™, the Enel X team has been able to process files in just minutes, which may have taken hours for a data scientist to manipulate manually.”
A global VIAble solution
At Enel X, innovation is always on. Together with VIA, we are committed to making technology that is constantly evolving to make the world more sustainable through the secure control of any combination of data sources and their formats. Ravanis sums this up by saying: “Anyone can get lucky and make something new once. For us, we strive to find a faster, smarter, cleaner way of doing everything, all the time.”
Our collaboration with VIA is also growing. We’re discussing ways to optimize our Enel Utility Bill Management Platform, and also considering blockchain-based solutions to enhance the purchase of renewables and new ways to automate other processes. As long as we’ve got the drive to work together to lead the energy transition, the innovation will keep coming.