Consumption Forecasting
Why this use case matters
Machine learning provides an excellent avenue for predicting future energy consumption. Accurate insights can provide critical insights into variables affecting the demand, providing decision-makers with an opportunity to address these levers. Forecasts also provide a benchmark to identify anomalous behavior, either high/low consumption, and alert managers to faults within the building.
Techniques
Supervised Learning – demand forecasting, time-series
Algorithms
Dynamic Linear Models (DLM), Classical Time-Series Forecasts, ARIMAX, Multilayer Perceptrons (MLP), Long Short-Term Memory (LSTM), NeuralProphet
Outcome
Draw a complete picture of consumption drivers by isolating the effects of multiple relevant variables over time, these insights can fuel other progressive analytics goals, such as improved customer experience, workforce schedules, and maintenance operations.
We lay out a hands-on approach to detecting consumption patterns using DLMs.
Mosaic has compiled our industry expertise into a Machine Learning playbook for Utilities.
