Neural network pricing boosts electricity demand response with context
Learning to Price Electricity for Optimal Demand Response
Machine Learning
Summary
Electricity prices that change over time can help people use energy more when renewable sources like solar and wind produce the most power. But setting these prices right is tricky because they depend on many factors like weather or the time of day. The authors created a smart computer program that learns to set electricity prices using lots of information about context. Their tests in simulated US cities show this method helps make demand response programs more effective.
What this means in practice
- •For energy utilities: Set dynamic electricity prices that adapt to environmental and usage patterns to better balance supply and demand in grids.
- •For smart home system developers: Integrate advanced pricing models that respond to real-world context for smarter, user-friendly energy management in residences.
Authors
Jing Shang, Mohammad Mehrabi, Xinyang Zhou, Mahmoud Saleh, Andrey Bernstein, Stefan Wager
Abstract
There is considerable interest in using time-varying electricity prices to shape consumer demand response, and better align energy demand with renewable production. However, optimal prices generally vary over time in response to complex signals such as weather forecasts, sunrise/sunset times, and day-of-week patterns; and existing methods are not able to make efficient use of such rich contextual information. Here, we propose a neural-network-based algorithm for contextual energy pricing, modeling pricing as a Stackelberg game and leveraging a mean-field solution representation from Mehrabi et al.~(2024). The approach learns constrained mappings from contextual features to feasible price signals. We validate our approach by simulating the energy grid in several US cities, and show that incorporating contextual information can considerably increase the value of the demand response programs.