Multi-channel data fusion and fault diagnosis of forging press based on GAF-ViT

Because the multi-channel pressure data of forging press has space complexity and time sequence dependence, the data characteristics cannot be fully mined. To solve this problem, a fault diagnosis method based on the Gram Angular Field (GAF) and Vision Transformer (ViT) is proposed in this paper. The original pressure data are encoded by the Gram Angle Field to retain the correlation and dependence of the time series. The four channel images are fused by the pixel weighted average method, and input into the ViT network for fault classification, so as to fully mine the data features to obtain better classification effect. The effectiveness of the proposed method is verified by experiments. And the proposed method is compared with other methods. The experimental results show that the proposed method has the highest accuracy of $99.6{{\% }}$ , and can effectively solve the problem of insufficient data feature mining of forging press.