WattCouncil: Context-Aware Household Energy Scenario Generation With Governed LLMs

2026-07-12

AI Summary

High-resolution residential electricity consumption data are scarce due to privacy concerns, regulatory restrictions, and high acquisition costs, significantly hindering machine learning research in smart grids. To address this challenge, this work proposes the first role-specialized large language model (LLM)-based multi-agent generation framework that synthesizes context-aware and physically plausible household electricity usage scenarios under cultural, temporal, and physical constraints through coordinated simulation, auditing, and validation mechanisms. The approach dynamically models energy consumption behaviors by integrating household composition, temporal patterns, and environmental factors. Evaluated on the CER dataset—comprising annual electricity usage and socioeconomic information from 4,232 households—the generated data demonstrate high realism, and ablation studies confirm the contribution of each component to overall performance.

Abstract

The accelerating shift toward low-carbon power systems, together with the widespread adoption of behind-the-meter technologies such as rooftop solar and electric vehicles, is placing new operational and analytical demands on electricity grids. At the same time, smart-grid research increasingly relies on machine learning (ML), yet progress is constrained by limited access to high-resolution household energy data due to privacy concerns, regulatory barriers, and collection costs. This work presents WattCouncil, a data-generation framework in which household electricity demand is generated by a council of Large Language Model (LLM)-based agents operating in specialized roles to generate, audit, and validate structured energy scenarios under explicit cultural, temporal, and physical constraints. Rather than acting as static predictors, these agents serve as adaptive decision-makers within a governed pipeline. Motivated by studies highlighting the importance of contextual factors in energy use, our framework produces context-sensitive daily routines through a guided reasoning process that incorporates household composition, temporal factors, and environmental conditions. We evaluate the generated profiles against the detailed CER dataset, which contains over a year of load measurements for 4232 households together with survey-based socio-economic information. We further assess the consistency of the framework through ablation studies. Source code is available at https://github.com/Singularity-AI-Lab/wattcouncil

Problem

• household energy data
• privacy constraints
• context-aware generation
• smart-grid research
• energy scenario

Innovation

• context-aware energy modeling
• governed LLM agents
• synthetic household load generation
• energy scenario simulation
• smart grid data synthesis