Our paper: click
Res-ACWGAN-GP is an end-to-end GAN for AHU FDD with limited fault data. This code achieves fault diagnosis while enhancing fault data through the improvement of ACGAN. The synthetic data is continuously optimized during the training process based on the fault diagnosis scores.
If you use this, please cite the paper below:
Bi, Jian, Ke Yan, and Yang Du. "End-to-end residual learning embedded ACWGAN for AHU FDD with limited fault data." Building and Environment 270 (2025): 112529.