Hybrid Adversarial Spectral Loss Conditional Generative Adversarial Networks for Signal Data Augmentation in Ultra-precision Machining Surface Roughness Prediction

TL;DR

This paper introduces HAS-CGAN, a novel spectral loss-based conditional GAN for generating high-frequency force signals to augment small datasets, significantly improving surface roughness prediction accuracy in ultra-precision machining.

Contribution

The paper presents a hybrid adversarial spectral loss CGAN tailored for 1D signal generation, enhancing data augmentation for UPM surface roughness prediction.

Findings

HAS-CGAN achieves >0.85 wavelet coherence in generated signals.

Data augmentation reduces prediction error from 31.4% to ~9%.

Synthetic data improves model performance across multiple ML and deep learning models.

Abstract

Accurate surface roughness prediction in ultra-precision machining (UPM) is critical for real-time quality control, but small datasets hinder model performance. We propose HAS-CGAN, a Hybrid Adversarial Spectral Loss CGAN, for effective UPM data augmentation. Among five CGAN variants tested, HAS-CGAN excels in 1D force signal generation, particularly for high-frequency signals, achieving >0.85 wavelet coherence through Fourier-domain optimization. By combining generated signals with machining parameters, prediction accuracy significantly improves. Experiments with traditional ML (SVR, RF, LSTM) and deep learning models (BPNN, 1DCNN, CNN-Transformer) demonstrate that augmenting training data with 520+ synthetic samples reduces prediction error from 31.4% (original 52 samples) to ~9%, effectively addressing data scarcity in UPM roughness prediction."