Alinta Energy is a leading Australian utility with an owned and contracted generation portfolio of up to 1,957 MW, over 1 million combined electricity and gas retail customers, and more than 700 employees across Australia and New Zealand.
To ensure efficient operation of their business, Alinta needs to forecast the energy consumption of their commercial customers at a two week horizon. The Australian energy market requires settlement at 30 minute periods at the ‘national metering identifier’ (NMI) or site level.
Intellify Data Scientists worked with Alinta analytics and business SMEs to understand the forecasting problem. Using anonymised data, machine learning models were developed and trained in Amazon Sagemaker, and delivered to the Alinta team for use in their production forecasting system.
How We Helped
Intellify’s data scientists worked with Alinta’s analytics team to build a forecasting model based on time-series machine learning techniques. A significant improvement in accuracy was achieved compared to the existing system.