A scaling law for short term load forecasting on varying levels of aggregation. (June 2018)

• Record Type: Journal Article
• Authors: Sevlian, Raffi; Rajagopal, Ram
• Is Part Of: International journal of electrical power & energy systems. Volume 98(2018)
• Page Start: 350
• Page End: 361
• Publication Date: 2018-06
• Subjects: Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2017.10.032
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: "Law of Large Numbers" (LLN) is rule of thumb on load forecasting error on grid scale vs. individual home level. Load forecasting accuracy scales with load aggregation size following LLN, up to a point. Diminishing returns shown to exist beyond a critical load. Large scale numerical study and theoretical analysis explore why such diminishing returns exist. Abstract: This paper proposes a simple empirical scaling law that describes load forecasting accuracy at varying levels of aggregation. We show that for many forecasting methods, aggregating more customers improves the relative forecasting performance up to specific point. Beyond this point, no more improvement in relative performance can be obtained. A benchmarking procedure for applying the scaling law to different forecasting models is presented. The aggregation model is evaluated with year long consumption profiles of over 180 thousand Pacific Gas & Electric customers. A theoretical model based on a bias variance decomposition of the forecast error is used to model the Aggregation Error Curves (AECs) that are empirically explored.