A White-Box Quantification Framework for Hidden Thermal Losses in

作者： Clark Tsang / 2026-08-11

Project Identity: GZDS-Quantification-Framework Author: Yi Zeng Manuscript PDF: <a href="https://www.researchgate.net/publication/412122748_A_White-Box_Quantification_Framework_for_Hidden_Thermal_Losses_in_110kV_Oil-Immersed_Power_Transformers_Based_on_Green-Zone_Dual_Stress" target="_blank" rel="noreferrer noopener">Read the Full Paper Here</a>(ResearchGate) DOI: <a href="https://doi.org/10.5281/zenodo.21886829" target="_blank" rel="noreferrer noopener">10.5281/zenodo.21886829</a> GITHUB:<a href="https://github.com/Shinar-of-Clark/GZDS-Quantification-Framework" target="_blank" rel="noreferrer noopener">GZDS-Quantification-Framework</a> <h2 class="wp-block-heading">📖 Introduction</h2> This repository contains the academic paper and related instructions for the Hidden Thermal Losses of 110kV Oil-Immersed Power Transformers under Green-Zone Dual Stress. Targeting the long-overlooked issue of transformer lifespan degradation caused by the “SCADA green zone” (i.e., compliant sub-threshold harmonics and periodic minor overloads) in modern grids, this paper proposes a dedicated Dual-Engine Physics-Informed Quantification Framework. By integrating the IEEE Std C57.110 eddy current loss factor and the IEEE Std C57.91 Arrhenius thermal aging equation, this framework precisely quantifies hidden lifespan losses. <h2 class="wp-block-heading">🎯 Motivation & Background</h2> With the high-penetration integration of Inverter-Based Resources (IBRs) and nonlinear loads, the power quality characteristics of the grid have fundamentally changed. Although large-amplitude transient voltage spikes and severe overloads can be effectively intercepted by protection devices, the sub-threshold synergistic stresses within the SCADA Green Zone constitute a critical monitoring blind spot: <ol class="wp-block-list"> <li>Compliant but not Absolutely Safe Green Zone: Continuous minor harmonics (THD ≤ 5%) and permissible periodic minor overloads (load factor K ≤ 105%) exist covertly for extended periods and completely fail to trigger static SCADA alarm thresholds.</li> <li>Limitations of Data-Driven Methods: Existing Health Index (HI) evaluations and data-driven Remaining Useful Life (RUL) prediction models generally classify SCADA green zone data as absolutely “safe”, thereby ignoring the cumulative thermal degradation caused by sub-threshold stresses. Furthermore, data-driven models rely heavily on complete historical ledgers and are prone to initial condition drift when sensors are installed midway through the service life.</li> <li>Computational Bottlenecks of White-Box Physics Models: Traditional Finite Element Analysis (FEA) or dynamic thermal circuit models incur extremely high computational costs, making them difficult to scale and embed into modern lightweight grid SCADA dispatch systems.</li> </ol> <h2 class="wp-block-heading">⚙️ Core Methodology</h2> To resolve the absolute positioning dilemma and historical ledger dependency, this paper innovatively proposes a computationally lightweight Dual-Engine Architecture that strictly adheres to physical laws: <ul class="wp-block-list"> <li>Thermodynamic Engine: Transforms SCADA operational state parameters (including electromagnetic, power quality, and baseline oil temperature) into thermodynamic damage factors. The core chain includes: calculating the winding eddy current heating modification factor based on IEEE Std C57.110, deriving the dynamic Hottest-Spot Temperature (HST) rise, and calculating the Arrhenius thermal aging acceleration factor (AF) in conjunction with IEEE Std C57.91.</li> <li>Hidden Loss Quantification Engine: Responsible for completing damage decoupling and full-lifecycle excess degradation calculation based on a sliding window. It embeds a 1-Year Equivalent Operating Time (EOT) sliding window algorithm to smooth seasonal temperature and load fluctuations, extracting the annualized equivalent acceleration factor, and directly mapping it to the physical calendar lifespan reduction of the equipment.</li> </ul> <h2 class="wp-block-heading">🛡️ Core Problems Solved</h2> <ul class="wp-block-list"> <li>Quantifying Physical Penalties within the SCADA Green Zone: One million Monte Carlo evolutionary simulations confirm that the GZDS synergy induces an average lifespan reduction of 9.83% (based on a 20.55-year baseline life, a loss of 2.02 years). Under the extreme compliance limit (constant 105% load, 5% THD), the lifespan reduction reaches up to 22.72% (4.67 years), completely shattering the traditional assumption of “zero damage” in the SCADA green zone.</li> <li>Eliminating Phase Deviation and Ledger Dependency: Through the 1-year EOT sliding window smoothing mechanism, the framework eliminates short-term thermal stress evaluation error drifts caused by sensors being installed in any arbitrary season (phase), achieving an equivalent evaluation independent of the initial state.</li> </ul> <h2 class="wp-block-heading">🏆 Main Academic Contributions</h2> <ol class="wp-block-list"> <li>GZDS Scenario Definition and White-Box Quantification Framework: Formally defines the “Green-Zone Dual Stress (GZDS)” scenario for the first time and constructs a closed-form white-box quantification framework that directly maps thermal degradation.</li> <li>Nonlinear Super-Additive Damage Decoupling: Quantitatively proves the “Super-additive” penalty effect of multiple minor stresses synergizing under Arrhenius exponential scaling, pointing out that traditional independent linear superposition rules will severely underestimate the true physical thermal degradation.</li> <li>Ultra-Low Error Future Loss Projection: The proposed lifespan equivalent projection algorithm based on a 1-year sliding window achieves an extremely low average relative error of only 0.31% (an absolute deviation of 3.10 days) in predicting future remaining extra lifespan losses across continuous monitoring point traversal tests covering the full lifecycle, demonstrating solid physical self-consistency.</li> </ol> <div data-wp-interactive="core/file" class="wp-block-file"><object data-wp-bind--hidden="!state.hasPdfPreview" hidden class="wp-block-file__embed" data="https://shinarofclark.com/wp-content/uploads/2026/08/A-White-Box-Quantification-Framework-for-Hidden-Thermal-Losses-in.pdf" type="application/pdf" style="width:100%;height:600px" aria-label="嵌入 A White-Box Quantification Framework for Hidden Thermal Losses in"></object><a id="wp-block-file--media-734511a6-2844-4e30-80bb-ae7e055b3459" href="https://shinarofclark.com/wp-content/uploads/2026/08/A-White-Box-Quantification-Framework-for-Hidden-Thermal-Losses-in.pdf">A White-Box Quantification Framework for Hidden Thermal Losses in</a><a href="https://shinarofclark.com/wp-content/uploads/2026/08/A-White-Box-Quantification-Framework-for-Hidden-Thermal-Losses-in.pdf" class="wp-block-file__button wp-element-button" download aria-describedby="wp-block-file--media-734511a6-2844-4e30-80bb-ae7e055b3459">Download</a></div>

Project Identity: GZDS-Quantification-Framework
Author: Yi Zeng
Manuscript PDF: Read the Full Paper Here(ResearchGate)
DOI: 10.5281/zenodo.21886829
GITHUB:GZDS-Quantification-Framework

📖 Introduction

This repository contains the academic paper and related instructions for the Hidden Thermal Losses of 110kV Oil-Immersed Power Transformers under Green-Zone Dual Stress. Targeting the long-overlooked issue of transformer lifespan degradation caused by the “SCADA green zone” (i.e., compliant sub-threshold harmonics and periodic minor overloads) in modern grids, this paper proposes a dedicated Dual-Engine Physics-Informed Quantification Framework. By integrating the IEEE Std C57.110 eddy current loss factor and the IEEE Std C57.91 Arrhenius thermal aging equation, this framework precisely quantifies hidden lifespan losses.

🎯 Motivation & Background

With the high-penetration integration of Inverter-Based Resources (IBRs) and nonlinear loads, the power quality characteristics of the grid have fundamentally changed. Although large-amplitude transient voltage spikes and severe overloads can be effectively intercepted by protection devices, the sub-threshold synergistic stresses within the SCADA Green Zone constitute a critical monitoring blind spot:

Compliant but not Absolutely Safe Green Zone: Continuous minor harmonics (THD ≤ 5%) and permissible periodic minor overloads (load factor K ≤ 105%) exist covertly for extended periods and completely fail to trigger static SCADA alarm thresholds.
Limitations of Data-Driven Methods: Existing Health Index (HI) evaluations and data-driven Remaining Useful Life (RUL) prediction models generally classify SCADA green zone data as absolutely “safe”, thereby ignoring the cumulative thermal degradation caused by sub-threshold stresses. Furthermore, data-driven models rely heavily on complete historical ledgers and are prone to initial condition drift when sensors are installed midway through the service life.
Computational Bottlenecks of White-Box Physics Models: Traditional Finite Element Analysis (FEA) or dynamic thermal circuit models incur extremely high computational costs, making them difficult to scale and embed into modern lightweight grid SCADA dispatch systems.

⚙️ Core Methodology

To resolve the absolute positioning dilemma and historical ledger dependency, this paper innovatively proposes a computationally lightweight Dual-Engine Architecture that strictly adheres to physical laws:

Thermodynamic Engine: Transforms SCADA operational state parameters (including electromagnetic, power quality, and baseline oil temperature) into thermodynamic damage factors. The core chain includes: calculating the winding eddy current heating modification factor based on IEEE Std C57.110, deriving the dynamic Hottest-Spot Temperature (HST) rise, and calculating the Arrhenius thermal aging acceleration factor (AF) in conjunction with IEEE Std C57.91.
Hidden Loss Quantification Engine: Responsible for completing damage decoupling and full-lifecycle excess degradation calculation based on a sliding window. It embeds a 1-Year Equivalent Operating Time (EOT) sliding window algorithm to smooth seasonal temperature and load fluctuations, extracting the annualized equivalent acceleration factor, and directly mapping it to the physical calendar lifespan reduction of the equipment.

🛡️ Core Problems Solved

Quantifying Physical Penalties within the SCADA Green Zone: One million Monte Carlo evolutionary simulations confirm that the GZDS synergy induces an average lifespan reduction of 9.83% (based on a 20.55-year baseline life, a loss of 2.02 years). Under the extreme compliance limit (constant 105% load, 5% THD), the lifespan reduction reaches up to 22.72% (4.67 years), completely shattering the traditional assumption of “zero damage” in the SCADA green zone.
Eliminating Phase Deviation and Ledger Dependency: Through the 1-year EOT sliding window smoothing mechanism, the framework eliminates short-term thermal stress evaluation error drifts caused by sensors being installed in any arbitrary season (phase), achieving an equivalent evaluation independent of the initial state.

🏆 Main Academic Contributions

GZDS Scenario Definition and White-Box Quantification Framework: Formally defines the “Green-Zone Dual Stress (GZDS)” scenario for the first time and constructs a closed-form white-box quantification framework that directly maps thermal degradation.
Nonlinear Super-Additive Damage Decoupling: Quantitatively proves the “Super-additive” penalty effect of multiple minor stresses synergizing under Arrhenius exponential scaling, pointing out that traditional independent linear superposition rules will severely underestimate the true physical thermal degradation.
Ultra-Low Error Future Loss Projection: The proposed lifespan equivalent projection algorithm based on a 1-year sliding window achieves an extremely low average relative error of only 0.31% (an absolute deviation of 3.10 days) in predicting future remaining extra lifespan losses across continuous monitoring point traversal tests covering the full lifecycle, demonstrating solid physical self-consistency.

A White-Box Quantification Framework for Hidden Thermal Losses in Download