Simulation of Subgouge Sand Deformations Using Robust Machine Learning Algorithms
Bibliographic record
Abstract
Abstract Ice gouging is one of the major menaces to the subsea pipelines crossing the Arctic (e.g., Beaufort Sea) or the non-Arctic (e.g., Caspian Sea) shallow waters. Burial of the sea-bottom-founded infrastructures is regarded as a feasible method for protection of the subsea assets against the ice gouging threat. These pipelines are commonly embedded underneath the deepest ice-scoured records in the area, whereas the pipeline system is still threatened by the ice-induced soil displacement developed into the ice tip owing to the shear resistance of the seabed soil. Determination of the sub-gouge soil displacements is a governing design factor for the subsea structures in the Arctic offshore that commonly need costly laboratory studies and long-running finite element (FE) analyses to guarantee the operational integrity of the subsea pipeline against the ice-gouging event. Thus, the industry is still seeking more cost-effective, reliable, and faster alternative approaches for simulation of the iceberg-seabed-pipeline interaction process to minimize the collision risk of ice keels with the subsea structures. Recently, the application of machine learning (ML) in different fields has witnessed impressive growth since the ML technology is sufficiently precise, quick, reliable, and cost-effective to model various linear and non-linear problems. In this study, three robust ML algorithms comprising the Decision Tree Regression (DTR), Random Forest Regression (RFR), and Extra Tree Regression (ETR) models were used for the first time to simulate the iceberg-seabed interaction process in the sandy seabed. Using the parameters governing the ice-seabed interaction mechanism, a set of the DTR, RFR, and ETR models were developed. To verify the ML models, a comprehensive dataset was constructed and the data was divided into two sub-samples including the training (70% of data) and testing (the remaining 30% of the data) datasets. Subsequently, for the DTR, RFR, and ETR models, several analyses such as sensitivity analysis, error analysis, and uncertainty analysis were performed. The conducted analyses demonstrated that the ETR algorithm had a reasonable performance to simulate both horizontal and vertical sub-gouge soil deformations in the sand. The soil depth ratio (y/W) and the horizontal load factor (Lh/γs.W3) had substantial significance to model the horizontal and vertical deformations in the present study. The presented results provided a good notion of modeling the ice-gouging problem through the ETR algorithm. The outcomes may facilitate proposing new solutions to estimate the sub-gouge soil deformations in the sandy seabed. The present work can also be used for the planning of expensive field, laboratory, and FE simulations and to reduce the expenditures on future studies.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
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".