In the precision processing industry, maintaining the accuracy of machine tools for an extensive period is crucial. Machining accuracy is affected by numerous factors, among which spindle thermal elongation caused by an increase in machine temperature is the most common. This paper proposed a key temperature point selection algorithm and thermal error estimation method for spindle displacement in a machine tool. First, highly correlated temperature points were clustered into groups, and the characteristics of small differences within groups and large differences between groups were realized. The optimal number of key temperature points was then determined using the elbow method. Meanwhile, the long short-term memory (LSTM) modeling method was proposed to establish the relationship between the spindle thermal error and changes of the key temperature points. The results show the largest root mean square errors (RMSEs) of the proposed LSTM model and the key temperature point selection algorithm were within 0.6 µm in the spindle thermal displacement experiments with different temperature changes. The results demonstrated that the combined methodology can provide improved accuracy and robustness in predicting the spindle thermal displacement.
The grinding/thinning of silicon carbide (SiC) wafers is a critical process in semiconductor manufacturing. The extremely hard and brittle SiC makes it difficult to ensure overall process efficiency. In this study an acoustic signature strategy for adaptive silicon carbide wheel grinding state diagnosis is proposed. It uses sound signals for online monitoring of the grinding ratio (GR) state during processing. First, fast Fourier transform (FFT) and short-time Fourier transform (STFT) are used to identify key frequency bands. The key frequency band time-domain signals are denoised and features extracted using the proposed adaptive variational mode decomposition (VMD). The main technique of adaptive VMD involves combining the whale optimization algorithm (WOA) to optimize VMD and the intrinsic mode functions (IMFs) influence ranking algorithm (IIR), abbreviated as WOA-VMD-IIR. The WOA is used to optimize the parameters of VMD to preserve the signal feature information in the IMFs decomposition, while the proposed IIR extracts the most influential IMFs for signal reconstruction and calculates time/frequency domain statistical features. Finally, the exploratory data auto-labeling (EDAL) method is proposed based on a hypothesis on the probability of GR state changes during the process. Data visualization analysis and model iterative testing were performed on the statistical feature data to improve the accuracy of feature data labeling. Experimental results showed that the GR state evaluation model in the confusion matrix reached 96%, achieving effective monitoring of the efficiency and state of the SiC wafer grinding process.
22 March 2024 in Asia-Pacific Journal of Operational Research