High-precision prediction of power generation amount from weather forecast, JERA introduces solar power
Outlook for development capacity through business alliance between JERA and West HD (unit: 10,000 kW)
(Source: Joint release by JERA and West HD)
JERA (Chuo-ku, Tokyo) and JDSC announced on November 24 that they had jointly developed a system that highly accurately predicts the amount of power generated by photovoltaic power generation from weather forecast data. The system will be installed in a solar power plant jointly developed by JERA and West Holdings, and will be put into operation from the end of November.
This system builds a learning model for each solar power generation facility, realizing complete automation of machine learning and shortening the learning period. In addition, we built the system on the cloud and implemented it in a serverless configuration. By automatically executing system operation and monitoring, the system operation load can be greatly reduced.
Developed by combining the expertise of JERA and JDSC in their respective fields of energy and data science. In the future, the two companies will accumulate power generation data and utilize JDSC's advanced machine learning knowledge to further improve the forecasting system.
JERA has formed a business alliance with West HD and plans to develop a solar power plant with a capacity of 1 million kW (1 GW) or more in Japan by the end of FY2025. The system will be introduced one by one in the operation of these solar power plants.
With the demand for decarbonization of energy, solar power generation, which is technically established and has a short construction period, is expected to be a powerful option that can expand the amount of renewable energy introduced in the short term. On the other hand, since the amount of power generated is affected by weather conditions, it is necessary to maintain a balance between power supply and demand. Therefore, photovoltaic power generation companies are required to predict the amount of power generated with high accuracy.
