Uncertainty-Aware Knowledge Transformers for Peer-to-Peer Energy Trading with Multi-Agent Reinforcement Learning

2025-07-22

AI Summary

Existing peer-to-peer (P2P) energy trading frameworks lack rigorous quantification of prediction uncertainty, hindering robust, risk-aware decision-making. Method: This paper proposes the Knowledge-Enhanced Transformer with Uncertainty (KTU) model—a novel integration of heteroscedastic probabilistic forecasting and multi-agent reinforcement learning. KTU employs a domain-knowledge-informed Transformer architecture to deliver high-accuracy, well-calibrated quantile forecasts; the resulting uncertainty intervals are explicitly embedded into a Deep Q-Network (DQN) policy network to enable risk-sensitive trading decisions. Results: In non-P2P settings, KTU reduces electricity procurement costs by 5.7% and increases sales revenue by 6.4%. With P2P trading enabled, improvements reach 3.2% and 44.7%, respectively, while significantly flattening peak grid load. This work establishes an interpretable, deployable paradigm for uncertainty-driven intelligent distributed energy trading.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Sequential Decision Making
• Planning, Routing, and Scheduling: Planning under Uncertainty

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

This paper presents a novel framework for Peer-to-Peer (P2P) energy trading that integrates uncertainty-aware prediction with multi-agent reinforcement learning (MARL), addressing a critical gap in current literature. In contrast to previous works relying on deterministic forecasts, the proposed approach employs a heteroscedastic probabilistic transformer-based prediction model called Knowledge Transformer with Uncertainty (KTU) to explicitly quantify prediction uncertainty, which is essential for robust decision-making in the stochastic environment of P2P energy trading. The KTU model leverages domain-specific features and is trained with a custom loss function that ensures reliable probabilistic forecasts and confidence intervals for each prediction. Integrating these uncertainty-aware forecasts into the MARL framework enables agents to optimize trading strategies with a clear understanding of risk and variability. Experimental results show that the uncertainty-aware Deep Q-Network (DQN) reduces energy purchase costs by up to 5.7% without P2P trading and 3.2% with P2P trading, while increasing electricity sales revenue by 6.4% and 44.7%, respectively. Additionally, peak hour grid demand is reduced by 38.8% without P2P and 45.6% with P2P. These improvements are even more pronounced when P2P trading is enabled, highlighting the synergy between advanced forecasting and market mechanisms for resilient, economically efficient energy communities.

Problem

• Addresses uncertainty in P2P energy trading predictions
• Integrates probabilistic forecasting with multi-agent reinforcement learning
• Optimizes energy trading costs and revenue using uncertainty-aware DQN

Innovation

• Uncertainty-aware transformer model for predictions
• Multi-agent reinforcement learning for trading
• Custom loss function for reliable forecasts