Applied AI & research software for engineering systems
Digital twins · anomaly detection · sensor-data QA/QC · scientific Python · environmental and structural monitoring
I am a Research Fellow and Data Scientist working at the interface of machine learning, sensor data, digital twins, scientific modelling, and research software engineering.
My work focuses on turning complex engineering and environmental data into reproducible tools for monitoring, validation, anomaly detection, digital-twin-style workflows, and decision support for physical systems.
I am especially interested in applied AI for sensor-rich physical systems, including structural testing, hydraulic systems, environmental monitoring, remote sensing, hydrology, and river morphodynamics.
The best starting points are:
- Digital twins / industrial AI:
synthetic-hydraulic-digital-twin-demo - Engineering data QA/QC:
tdms-sync-checker - Applied ML / anomaly detection:
audio-anomaly-detection-structural-testing - Scientific ML / geomorphology:
meander-morphology-classifier - Scientific modelling / research software:
LDSFL_Meander
Together, these projects show how I approach applied AI beyond model fitting: data quality, reproducibility, documentation, validation, reporting, and safe publication boundaries.
| Project | Area | What it demonstrates |
|---|---|---|
Hydraulic Digital Twin |
Digital twins / industrial AI | Synthetic sensor data, hydraulic energy estimation, anomaly detection, digital-twin state classification, and automated reporting |
TDMS Sync Checker |
Engineering data QA/QC | TDMS timing checks, synchronisation diagnostics, split-file continuity, inactive-channel detection, and report generation |
Structural Audio Anomaly Detection |
Applied ML / anomaly detection | Audio-based anomaly detection for large-scale structural testing, feature extraction, model evaluation, and reproducible research workflows |
Meander Morphology Classifier |
Scientific ML / geomorphology | CWT spectra, autoencoder latent spaces, clustering, Streamlit GUI, Zenodo-linked models, and reproducible meander-bend classification workflows |
LDSFL Meander |
Scientific computing / hydrology | Morphodynamic modelling, reproducible simulations, CLI/GUI workflows, documentation, and citation metadata |
-
Remote sensing / environmental monitoring:
strandings_from_space— collaborative open-source workflow for very-high-resolution satellite-image pre-processing, annotation, and observer-count comparison. My fork is available atsergioald/strandings_from_space. -
Open-source research software / deep learning:
GeoOcean/BlueMath_tk— active contributions to thedeeplearningautoencoder module, including smoke tests, implementation fixes, and improved validation of autoencoder behaviour.
- Applied AI: anomaly detection, classification, time-series and signal features, model validation
- Engineering data: sensor networks, TDMS files, synchronisation diagnostics, data-quality checks
- Environmental data: remote-sensing workflows, hydrology, hydraulic modelling, monitoring pipelines
- Scientific ML: autoencoders, latent spaces, clustering, spectral features, river-morphology classification
- Scientific Python: NumPy, pandas, SciPy, Matplotlib, scikit-learn
- Research software: reproducible workflows, command-line tools, documentation, examples, testing
I try to make repositories useful as engineering and research artefacts, not only as code.
Where possible, projects include clear problem statements, installation and usage instructions, example or synthetic data, visual outputs, assumptions and limitations, reproducible scripts, and citation metadata where relevant.
This is especially important when real industrial or research data cannot be shared publicly.
I am interested in applied AI, research software, digital twins, and engineering-data workflows.
- Portfolio: sergioald.github.io
- GitHub: @sergioald
- LinkedIn: Sergio Lopez Dubon
- Academic profile: University of Edinburgh Research Explorer
- Publications: Google Scholar
- ORCID: 0000-0003-0663-607X