Sallen-Key Filter Frequency Response Data for Predictive Maintenance Benchmarking

9,000 AC sweep frequency traces simulate 9 distinct circuit states, including normal operation and 8 fault modes, for a Sallen-Key low-pass filter. The dataset incorporates 5% and 10% manufacturing tolerances and provides both raw and preprocessed data with added noise. Author Jianjun Zhong published this benchmark on figshare in April 2026 under a CC-BY-4.0 license.

Use Cases

• Benchmarking few-shot learning algorithms for fault diagnosis based on the 8 distinct soft and hard fault modes.
• Evaluating metric learning models for distinguishing between normal component tolerance and minor faults based on overlapping frequency response features.
• Training robust predictive maintenance models for analog circuits using simulated frequency traces with added 2% AWGN.
• Testing machine learning algorithms in low-sample, noisy environments based on the dataset's stated design purpose.

Strengths

• Contains 9,000 frequency traces across 9 distinct circuit states, providing a substantial simulation volume.
• Models real-world component variation by incorporating 5% and 10% Gaussian manufacturing tolerances.
• Offers both raw and preprocessed data (with 2% AWGN and Min-Max normalization) in CSV and NPY formats for flexibility.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for specific modeling needs.
• Data is simulated via Monte Carlo methods and may not capture all nuances of physical hardware failures.

Provenance

Source

figshare, author Jianjun Zhong.

Collection Method

Monte Carlo simulation of a Sallen-Key second-order active low-pass filter circuit.

Freshness

Last updated 2026-04-20 11:57:42; freshness should be verified.