JERA Co., Inc. and Japan Data Science Consortium have partnered to develop a “highly accurate system” that predicts solar power generation based on the weather forecast and the performance of past power generation.

In a statement, JERA said the system was used at its solar power generation facilities. The system was deployed in a serverless configuration and the automation of the operation, monitoring, model learning, and updating tasks cuts the operational load.

The parties plan to further improve the accuracy of the system.

Solar power generation is a “promising short-term option” to expand Japan’s renewable energy adoption in line with the country’s target of decarbonising by 2050 as the technology is established and quick to construct.

“At the same time, since solar power generation performance varies with weather conditions, highly accurate predictions of the amount of electricity generated by solar power are needed in order to balance electricity supply and demand,” the statement read.