LineVision’s Dynamic Line Rating Methodology
LineVisions DLR model is trained on the combination of advanced wind modeling, machine learning, sensor measurements, and numerical weather prediction for the most confident line ratings available. Our solution rapidly scales across entire transmission systems with DLR powered by advanced wind models and is made more accurate by adding sensors in critical areas facing congestion, load, or constraints.
Our Dynamic Line Rating methodology is based on three pillars:
3 feet per second of wind can add 40% capacity to transmission lines1
But global weather models user over a 10 kilometer resolution
To ensure the most safe and accurate line ratings, LineVision installs sensors at critical spans and uses computational fluid dynamics (CFD) to model wind at a 30 meter resolution, ensuring that all trees, hills, and valleys are captured.
In the example above, the most wind-limited spans (red) are running through valleys with high tree cover. LineVision's sensors and CFD-enhanced forecasts ensure that these limiting sections operate safely.
What is Computational Fluid Dynamics?
Computational Fluid Dynamics (CFD) is a branch of engineering that uses numerical methods and algorithms to simulate and analyze fluid flow, heat transfer, and other physical processes. We use it to model how wind, temperature, and other atmospheric conditions affect the cooling of transmission lines.
CFD is important because weather models do not take hyper-local terrain and vegetation into account. By simulating airflow around conductors, CFD helps determine real-time line capacity based on actual weather conditions with over a 2x improvement in accuracy.
FAQs
Computational Fluid Dynamics (CFD) is a branch of engineering that uses numerical methods and algorithms to simulate and analyze fluid flow, heat transfer, and other physical processes. We use it to model how wind, temperature, and other atmospheric conditions affect the cooling of transmission lines.
Our sensors provide hyperlocal data which is used to continually assess and improve our trained ratings model. Our models are validated using comprehensive data quality checks, ensuring highest quality ratings.
Sensor density is dependent on the size and complexity of a project. Implementations typically feature one sensor per 2-3 miles.
Wind speed is measured locally using an high accuracy sensor. Numerical windspeed predictions are calculated using CFD-corrected third-party weather data.
LineVision uses IEEE 738 and concepts from CIGRE TB 498 to calculate ratings.
LineVision uses machine learning (a branch of AI) to parse through and simplify large amounts of data received by our sensors, weather data providers, and utility partners. We do not just unleash unchecked AI and allow it to make predictions - we validate the AI outputs with our physics-based processes to ensure they are trustworthy.
