Digital Twin for Pipeline Leak Monitoring
Bibliographic record
Abstract
Abstract We demonstrate how a visual digital twin system is used to implement a digital twin for pipeline monitoring. A visual digital twin allows for ingestion of data for calibration and creation of models which enable a digital replica of a real-world physical system for decision support and improving situational awareness. A laboratory experimental pipeline system is used to generate data to drive modelling capabilities in this system. We show how machine learning operations (MLOps) principles are applied in the context of digital twins, forming a sub-study area known as Digital Twin ML Ops (DT MLOps). The intended purpose of the system is to both train operators of the leak detection system in its use and provide high situational awareness and operational readiness to users. We demonstrate how multiple sources of monitoring data from an experimental pipeline setup, as well as simulation can be combined in a visual digital twin system and used for pipeline leak detection and leak plume and flow visual prediction. We show how leak detection and visual leak prediction are visualized in the context of a pipeline twin along with confidence and uncertainty and other explainable elements from machine learning models. We demonstrate how models are tracked using the DT MLOps sub system. The overall system demonstrates a novel combination of real experimental data driving models for both leak detection and the prediction of the associated leak plume and pipeline flow visual assessment imagery. This demonstrates a system which can detect leaks and their locations and also provide operators with assessment of the leak. The system provides provenance through its DT MLOps capabilities. This presented virtual pipeline and leak model, which integrates AI with experimental validation, is not widely available in industry.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.002 | 0.004 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.014 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".