A prognostic model for managing consumer electricity demand and smart grid reliability

Achieving high reliability in the smart grid depends on, among other factors, the utility companies' ability to create accurate forecasts on consumer demands for the near and long-term future. Forecasts may be based on time series analysis using historical consumer load data combined with local weather forecasts. Forecasts that predict an increase in consumer demand will enable utility companies to make informed decisions in purchasing additional capacity and/or sending out selective consumer alerts. The paper will discuss theoretical aspects of statistical forecasting and demonstrate its usefulness based upon a case study of actual electrical grid demand sampled at an hourly frequency.