作者: Jorjeta Jetcheva , Mostafa Majidpour
DOI:
关键词:
摘要: A method of forecasting a resource load for consumption at site, e.g., an electrical site. The includes receiving historical data and ambient condition that are time series pertaining to site generating additional from the received data. building best sub-model each multiple forecast intervals. clustering in parallel training portions data, possible sub-models using clustered portions, verifying forecasted loads output against verification determining first subset parameters based upon accuracy loads. intervals ensemble sub-models.