3D Printer Vibration Data for Federated Learning Case Study

Cheng-Hao Chou from the University of Michigan created a dataset of acceleration data. It was collected from six low-cost 3D printers of the same model while air-printing cubes at different speeds. The dataset is intended as a case study for federated and personalized learning.

Use Cases

• Train federated learning models based on distributed vibration data from multiple printers.
• Develop personalized predictive maintenance models based on individual printer vibration signatures.
• Benchmark anomaly detection algorithms for 3D printers based on acceleration data.
• Study the effect of printing speed on machine vibration patterns.

Strengths

• Data is collected from six identical 3D printers, enabling controlled comparative studies.
• Acceleration data is captured under varied printing speed conditions, introducing a key experimental variable.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

University of Michigan

Collection Method

Acceleration data collected from sensors on six low-cost 3D printers air-printing cubes.