Operational Wind Speed Forecasts for Chile's Electric Power Sector Using a Hybrid ML Model

TL;DR

This paper presents a hybrid machine learning approach combining TiDE and GraphCast models to improve wind speed forecasts in Chile, aiding renewable integration and reducing operational challenges.

Contribution

The study introduces a novel hybrid ML model specifically tailored for Chile's wind speed forecasting, outperforming existing deterministic systems in accuracy.

Findings

Hybrid model improves short-term forecast accuracy by up to 21%.

Medium-term forecasts are improved by up to 23%.

Enhanced forecasts help reduce thermal plant ramping and emissions.

Abstract

As Chile's electric power sector advances toward a future powered by renewable energy, accurate forecasting of renewable generation is essential for managing grid operations. The integration of renewable energy sources is particularly challenging due to the operational difficulties of managing their power generation, which is highly variable compared to fossil fuel sources, delaying the availability of clean energy. To mitigate this, we quantify the impact of increasing intermittent generation from wind and solar on thermal power plants in Chile and introduce a hybrid wind speed forecasting methodology which combines two custom ML models for Chile. The first model is based on TiDE, an MLP-based ML model for short-term forecasts, and the second is based on a graph neural network, GraphCast, for medium-term forecasts up to 10 days. Our hybrid approach outperforms the most accurate operational deterministic systems by 4-21% for short-term forecasts and 5-23% for medium-term forecasts and can directly lower the impact of wind generation on thermal ramping, curtailment, and system-level emissions in Chile.