Training Environment Parameters for Power Quality Disturbance Recognition

5.5 KB of parameters for training a neural network to classify power quality disturbances. The dataset was created by Junqing Zhang and uploaded to figshare on June 3, 2026. It likely contains parameters used in a simulation analysis comparing a proposed feature-image fusion method against traditional recognition systems.

Use Cases

• Training neural networks for power quality disturbance classification based on variational mode decomposition features.
• Benchmarking noise robustness in signal recognition models based on the described simulation analysis.
• Developing signal-to-image transformation methods for time-series data based on the described component matrix.
• Comparing the performance of improved ResNet-18 architectures against other recognition systems.

Strengths

• Dataset is 5.5 KB, indicating a focused and compact set of parameters.
• The description details a specific methodological pipeline involving VMD, CWT, and ResNet-18.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset's small size suggests it contains only training parameters, not the raw disturbance signals or images.

Provenance

Source

figshare

Collection Method

Parameters likely generated from simulation analyses described in the associated research paper.

Freshness

Last updated 2026-06-03 17:44:00; freshness should be verified.