From augmentation to translation: Data generation by conditional hierarchical variational autoencoder, enhancing monitoring mooring systems in floating offshore wind turbines
Bibliographic record
Abstract
The integrity of mooring systems in floating offshore wind turbines (FOWTs) is crucial, as their degradation alters the platform’s dynamic behavior. A robust machine learning-based health monitoring system that continuously monitors different mooring systems for FOWTs requires data under diverse health, operational, and metocean conditions. To this end, we propose a Conditional Hierarchical Variational Autoencoder (CHVAE) generative model designed for simultaneous data augmentation and domain translation to generate the required data. We train the model to learn the nonlinear relationships between healthy and minority-damaged fairlead tension records from the source mooring system across various sea states. CHVAE generates realistic damaged responses under diverse conditions by leveraging healthy data from the target mooring system. We first assess CHVAE’s ability to augment minority data based on majority distribution, validated on the Modified National Institute of Standards and Technology (MNIST) benchmark dataset. This experiment compares the performance of CHVAE variants with conventional and recent oversampling methods. Second, the open-source software OpenFast simulates the testing and training datasets for simultaneously data augmentation and domain translation on the Offshore Code Comparison Collaboration Continuation (OC4) semi-submersible platform (DeepCwind) FOWT benchmark. OpenFast and CHVAE records are compared through visual, statistical, and behavioral methodologies. Simulations utilize diverse wave seeds to represent excitation randomness and undetected damage severities, assessing CHVAE’s one-to-all capability. Generated records for unobserved sea states and damage severities closely mimic real behavior in downstream binary classification, illustrating the versatility of CHVAE for zero-shot, real-time damage identification. • Novel generator for real-scale time-series synthesis (CHVAE) • One-to-all CHVAE maps healthy data to a specified damage severity • Second CHVAE decoder estimates de-normalization parameters from minority distribution • Compared with oversampling baselines on the MNIST benchmark • CHVAE conducts domain translation to generate unseen damaged tension for OC4 FOWT
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) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".