An Agentic AI Pipeline for Appliance-Level Energy Anomaly Detection and LLM-Driven Recommendations

2026-06-26

AI Summary

This study addresses the challenge of ineffective response to noisy energy consumption alerts in office building equipment monitoring by non-expert personnel. The authors propose an end-to-end agent pipeline that integrates hybrid SSA-LSTM time-series forecasting, attention-enhanced LSTM-VAE for variational anomaly detection, and a three-stage LangChain agent framework (Context/Diagnosis/Report). By incorporating RAG with a dynamic retrieval mechanism, the system reduces context sources from six to three–six while maintaining performance and improving inference efficiency. A novel reflective memory layer is introduced to establish a human-in-the-loop feedback cycle. Notably, the approach achieves 100% pass rates across all 16 anomaly scenarios on a local 7B large language model, with the best LLM backend scoring 90.4/100, significantly enhancing alert interpretability and maintenance prioritization capabilities.

Abstract

Appliance-level energy monitoring in office buildings produces noisy alerts that non-expert facility managers struggle to use. This paper proposes an end-to-end agentic pipeline that combines deep time-series forecasting, variational anomaly detection, and LLM-based reasoning to generate prioritized, actionable maintenance recommendations. The system tracks seven office appliances using a hybrid Singular Spectrum Analysis (SSA) and Long Short-Term Memory (LSTM) forecasting model, and applies a per-appliance LSTM Variational Autoencoder (VAE) with attention to flag abnormal daily consumption episodes. A three-stage LangChain pipeline begins with a Context Agent that always retrieves three core RAG sources (model reliability, hourly baseline, and expert knowledge) and conditionally adds up to three more (forecast context, anomaly history, global baseline) based on event characteristics, capped at eight reasoning steps. A Diagnosis Agent converts the evidence into a structured JSON diagnosis, and a Report Agent renders a human-readable narrative. A reflective memory layer incorporates operator feedback. The dashboard shows real-time 30-minute forecasts, intraday consumption, the previous day anomaly report, and a feedback form. We evaluate the forecasting model, anomaly detector with appliance-specific thresholds, and LLM reasoning on a 16-scenario benchmark including sustained and transient spikes, unexpected shutdowns, and systemic events, comparing five LLM backends under static vs. dynamic retrieval. Dynamic retrieval matches full static retrieval across all backends while cutting average context from six to three-six sources per event. The best backend scores 90.4/100 with a 100% pass rate at a 70-point threshold, and a fully local 7B-parameter model passes all 16 scenarios.

Problem

• energy anomaly detection
• appliance-level monitoring
• facility management
• actionable recommendations
• noisy alerts

Innovation

• Agentic AI
• Appliance-level anomaly detection
• LLM-driven reasoning
• Dynamic RAG
• LSTM-VAE with attention