Time Series Anomaly Detection: A Hybrid Approach

In the field of predictive maintenance, deep learning-based time series anomaly detection has garnered significant attention. However, pure deep learning approaches often fail to achieve sufficient accuracy on real-world data.

This study proposes a hybrid approach that integrates 64-dimensional time series embeddings from Granite TinyTimeMixer with 28-dimensional statistical features based on domain knowledge for HVAC (heating, ventilation, and air conditioning) equipment anomaly prediction tasks.

Specifically, time series embeddings extracted from a Granite TinyTimeMixer encoder fine-tuned with LoRA (Low-Rank Adaptation) are combined with 28 types of statistical features, including trend, volatility, and drawdown indicators, which are then learned using a LightGBM gradient boosting classifier.

Experimental Results

Experiments using 64 equipment units and 51,564 samples achieved Precision of 91-95% and ROC-AUC of 0.995 for anomaly prediction at 30-day, 60-day, and 90-day horizons. Furthermore, production-ready performance was achieved with a false positive rate of 1.1% or less and a detection rate of 88-94%, demonstrating the effectiveness of the system for predictive maintenance applications. This work demonstrates that practical anomaly detection systems can be realized by leveraging the complementary strengths between deep learning's representation learning capabilities and statistical feature engineering.

For those evaluating on-premise deployments, there are trade-offs to consider. AI-RADAR offers analytical frameworks on /llm-onpremise to evaluate these aspects.