i GRID LAB Builds AI-Controlled Platform for Excess Power with the Goal of Circulation Green Power Regionally on AWS

i GRID Solutions and its group companies are developing its business centered on Green Transformation (GX) Solutions. It promotes GX in order to accelerate its adoption by making social and economic activities centered on renewable energy and decarbonization a reality

In June 2022, i GRID LAB released the R.E.A.L New Energy Platform and started full-scale deployment to a group company, VPP Japan. The R.E.A.L. New Energy Platform is an energy management platform that achieves highly accurate surplus power prediction through artificial intelligence (AI) and detailed automatic control of Distributed Energy Resources (DER). It collects information on things such as solar power generation volume and power, demand, storage batteries, and EV charging/discharging status, and the data is stored in the cloud. AI performs highly accurate predictive analysis combined with external information such as electricity market prices. AI is also used to determine the optimal timing for EV charging and discharging along with the calculation of excess solar energy and the creation automatic device control scenarios.

VPP Japan owns R.E.A.L. Solar Power, one of the largest distributed solar power plant in Japan, and the first goal is to use this platform to circulate the excess power that is generated by the plant regionally.

Tetsu Iwasaki, director and chief technology officer (CTO) of i GRID LAB, says, "We thought that using AI would be an important key to expand the use of renewable energy." The number of self-consumption solar power systems in places like commercial facilities is increasing, and it is difficult to control excess power. Until now, it has been common to only be able to get the amount of power that a facility could fully consume. “The basic concept is to properly control the surplus power with AI and provide it to the other users. This makes it possible to achieve a circulation model for surplus power and increase the utilization ratio of renewable energy by maximizing the introduction of solar power generation.” 

Iwasaki, who joined i GRID LAB in October 2020, began a full-scale project to build a next-generation energy platform in December. He started development on AWS in January 2021. Meanwhile, he asked for help from supermarket chains that had already installed solar panels at multiple locations under the i GRID Group. A proof of concept was conducted at two stores in Saitama prefecture."Building such a platform is the first initiative in the industry, and by actually creating and running it, rather than studying it on a desk, we can concretely ascertain the effects and problems. It was important for us to be agile and start moving quickly,” says Iwasaki.

A platform for the proof of concept was built in just 6 months from the start of the project. The company adopted AWS IoT Greengrass for the data collection from edge devices and Amazon Timestream for high-speed processing of large amounts of time series data and began proof of concept in June 2021. In July, they clarified the specific effects and problems and announced the details on their website. Additionally, they migrated AI functions, which had been operated on external servers until then, to Amazon SageMaker.

They continued the proof of concept after that and, as a second step, commenced the decarbonization of last mile logistics, which incorporates EV for online supermarkets, in January 2022. At the same time, the i GRID Group began providing this adaptation to around 500 solar facilities. Furthermore, in December 2021, they collaborated with a delivery system for sending power to general households. They collected and analyzed electricity usage and other data from approximately 100,000 households.

And, as mentioned previously, the platform was commercialized in June 2022. The brand names of related services and products have been unified into R.E.A.L., and mutual cooperation between various services provided by the i GRID Group has also been strengthened.