Forecasting Daily Primary Three-Hour Net Load Ramps in the CAISO System

TL;DR

This paper introduces a novel LSTM-based methodology for accurately forecasting daily primary three-hour net load ramps in the CAISO system, aiding system operators in managing flexible capacity amidst increasing renewable energy integration.

Contribution

The paper develops a new LSTM-based approach for forecasting net load ramps, incorporating key influencing factors identified through extensive analysis, and demonstrates its effectiveness on real CAISO data.

Findings

The proposed method outperforms benchmark models in forecasting accuracy.

Accurate ramp magnitude and start time predictions improve system flexibility planning.

The methodology effectively captures factors influencing net load ramps.

Abstract

The deepening penetration of variable energy resources creates unprecedented challenges for system operators (SOs). An issue that merits special attention is the precipitous net load ramps, which require SOs to have flexible capacity at their disposal so as to maintain the supply-demand balance at all times. In the judicious procurement and deployment of flexible capacity, a tool that forecasts net load ramps may be of great assistance to SOs. To this end, we propose a methodology to forecast the magnitude and start time of daily primary three-hour net load ramps. We perform an extensive analysis so as to identify the factors that influence net load and draw on the identified factors to develop a forecasting methodology that harnesses the long short-term memory model. We demonstrate the effectiveness of the proposed methodology on the CAISO system using comparative assessments with selected benchmarks based on various evaluation metrics.