Dynamic Load Model for Data Centers with Pattern-Consistent Calibration

2026-02-08

AI Summary

This study addresses the limitations of conventional load models in accurately capturing the dynamic disconnect-reconnect behavior of data centers induced by abrupt workload shifts and protection mechanisms, which compromises power system simulation fidelity. To overcome this, the authors propose a physics-informed, data-driven approach featuring a temporal contrastive learning (TCL)-based pattern-consistent calibration mechanism that faithfully preserves both the temporal dynamics and statistical characteristics of real-world loads while avoiding trajectory overfitting. Validated on real datasets including MIT Supercloud and integrated into the ANDES platform, the model—when applied to the IEEE 39-bus system—reveals, for the first time, complex composite disconnect-reconnect dynamics and delayed stability phenomena arising from large-scale power-electronics-dominated load interactions. The calibrated model significantly outperforms its uncalibrated counterpart and enables grid-level parameter sharing under privacy-preserving constraints.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

The rapid growth of data centers has made large electronic load (LEL) modeling increasingly important for power system analysis. Such loads are characterized by fast workload-driven variability and protection-driven disconnection and reconnection behavior that are not captured by conventional load models. Existing data center load modeling includes physics-based approaches, which provide interpretable structure for grid simulation, and data-driven approaches, which capture empirical workload variability from data. However, physics-based models are typically uncalibrated to facility-level operation, while trajectory alignment in data-driven methods often leads to overfitting and unrealistic dynamic behavior. To resolve these limitations, we design the framework to leverage both physics-based structure and data-driven adaptability. The physics-based structure is parameterized to enable data-driven pattern-consistent calibration from real operational data, supporting facility-level grid planning. We further show that trajectory-level alignment is limited for inherently stochastic data center loads. Therefore, we design the calibration to align temporal and statistical patterns using temporal contrastive learning (TCL). This calibration is performed locally at the facility, and only calibrated parameters are shared with utilities, preserving data privacy. The proposed load model is calibrated by real-world operational load data from the MIT Supercloud, ASU Sol, Blue Waters, and ASHRAE datasets. Then it is integrated into the ANDES platform and evaluated on the IEEE 39-bus, NPCC 140-bus, and WECC 179-bus systems. We find that interactions among LELs can fundamentally alter post-disturbance recovery behavior, producing compound disconnection-reconnection dynamics and delayed stabilization that are not captured by uncalibrated load models.

Problem

• data center load modeling
• dynamic load model
• pattern-consistent calibration
• large electronic load (LEL)
• post-disturbance recovery

Innovation

• Pattern-consistent calibration
• Temporal contrastive learning
• Physics-informed data-driven modeling
• Large electronic load (LEL)
• Data center dynamic load model