Grid Balancing

Why this use case matters

Supply and demand fluctuations are becoming more extreme and difficult to anticipate. Renewable sources, weather, and consumption patterns are all factors that make accurate projections hard to make.

Techniques

Supervised Learning – demand forecasting / yield forecasting, time-series

Algorithms

Dynamic Linear Models (DLM), Classical Time-Series Forecasts, ARIMAX, Long Short-Term Memory (LSTM), DLMs, Neural-Networks, Prophet, Time Series Classification

Outcome

Using data sources, modeling techniques, and knowledge from outside of traditional utility use cases helps energy operators balance grid frequency and achieve cost savings. For example, the integration of weather data can save utilities valuable resources in matching supply with demand.

Mosaic delves into how businesses can integrate weather effects in their decision making.

Mosaic has compiled our industry expertise into a Machine Learning playbook for Utilities.