Datasetpublic

sankalpsthakur/forge-pump-digital-twin-synthetic

Forge Pump Digital Twin — Synthetic A deterministic, clean-room tabular baseline for pump surrogate modelling, edge-runtime conformance, and advisory anomaly examples. It contains 40,000 synthetic rows split into 28,000 train, 6,000 validation, and 6,000 test rows. This dataset contains no plant telemetry, customer data, equipment identifiers, CAD, BOMs, nameplates, vendor curves, or values copied from a private repository. Every constant is an illustrative engineering proxy. It… See the full description on the dataset page: https://huggingface.co/datasets/sankalpsthakur/forge-pump-digital-twin-synthetic.

sourceHugging Facecc-by-4.0updated 2mo agoView on Hugging Face

0likes36downloads

Dataset Card

Forge Pump Digital Twin — Synthetic

A deterministic, clean-room tabular baseline for pump surrogate modelling, edge-runtime conformance, and advisory anomaly examples. It contains 40,000 synthetic rows split into 28,000 train, 6,000 validation, and 6,000 test rows.

This dataset contains no plant telemetry, customer data, equipment identifiers, CAD, BOMs, nameplates, vendor curves, or values copied from a private repository. Every constant is an illustrative engineering proxy. It is not suitable for equipment selection, protection settings, or control.

Schema

Group	Columns
Inputs	`speed_fraction`, `static_head_m`, `system_k`, `voltage_fraction`, `ambient_temp_c`, `inlet_pressure_bar`
Regression targets	`flow_m3h`, `total_head_m`, `input_power_kw`, `winding_temp_c`, `npsh_margin_m`, `efficiency_fraction`
Metadata	`row_id`, `advisory_flags`, `source_class`

The frozen input envelope is:

Input	Range
Speed fraction	0.35–1.05
Static head	5–120 m
System coefficient	0.03–0.80
Voltage fraction	0.82–1.08
Ambient temperature	5–50 °C
Inlet pressure	0.15–2.50 bar

Generation model

The generator solves a simple pump/system-curve intersection, then derives hydraulic power, bounded efficiency, a winding-temperature proxy, and an NPSH margin proxy. Generation is deterministic with seed 20260802; the source is included with the companion model.

advisory_flags can contain no_flow_risk, cavitation_risk, thermal_risk, undervoltage, and overload_risk. Thresholds are deliberately labelled as demonstration policy. They are not trip setpoints and must not be transferred to real equipment.

Intended uses

—Tabular multi-output regression experiments.
—PyTorch, ONNX, TensorFlow, and LiteRT/TFLite export tutorials.
—Edge inference and cross-runtime conformance tests.
—Synthetic anomaly-detection and telemetry-pipeline examples.

Prohibited interpretation

This release has no operational write capability. It does not implement a PLC, SIS, ESD, permissive, interlock, ramp controller, or closed-loop command path. Real protection and flow-control decisions require independently engineered, validated, and commissioned control systems.

Reproducibility

manifest.json records split sizes, file sizes, and SHA-256 hashes. SHA256SUMS can be checked with:

bash

sha256sum -c SHA256SUMS

Companion model: `sankalpsthakur/forge-pump-surrogate-multiruntime`

Interactive browser demo: `sankalpsthakur/forge-pump-edge-twin-lab`

License and attribution

Licensed under CC BY 4.0. Cite the dataset name, author, version, and Hugging Face URL. See LICENSE.