In manufacturing automation, accurately determining the optimal tool replacement timing is critical yet challenging. Tool condition monitoring (TCM) has been widely studied to address this issue. Cutting force is a key parameter for evaluating tool wear, but conventional force sensors are costly and difficult to implement. This study proposes a cost-effective alternative by estimating cutting forces using spindle vibration data through a long short-term memory (LSTM)-based machine learning model. First, the correlation between cutting force and tool wear is analyzed to emphasize the need for accurate force estimation. Then, vibration data collected from the spindle are used to train an LSTM model, which is effective for time-series data processing. The model is trained with vibration signals from various machining positions, with structured time-series datasets improving performance and generalization. Experimental results show that the developed model accurately estimates cutting forces using short segments of vibration data from a single tool revolution. Additionally, the observed relationship between cutting force and tool wear remains consistent across different machining conditions. This study validates real-time cutting force estimation via spindle vibration monitoring and suggests its potential for tool wear prediction. The proposed method offers a practical, low-cost solution for improving tool condition monitoring in automated machining.

Materials such as titanium alloys, nickel alloys, and stainless steels are difficult to machine due to low thermal conductivity, work hardening, and built-up edge formation, which accelerate tool wear. Frequent tool changes are required, often relying on operator experience, leading to inefficient tool use. While modern machine tools include intelligent tool replacement systems, many legacy machines remain in service, creating a need for practical alternatives. This study proposes a method to autonomously determine tool replacement timing by monitoring machining process signals in real time, enabling automatic tool changes even on conventional machines. Tool wear is evaluated using current and vibration sensors, with the replacement threshold estimated from the maximum current observed in an initial user-defined interval. When real-time signals exceed this threshold, the system updates controller variables to trigger tool changes. Results show vibration data are more sensitive to wear, whereas current data provide greater stability. These findings indicate that a hybrid strategy combining both sensors can enhance accuracy and reliability of tool change decisions, improving machining efficiency for difficult-to-cut materials.

This study details the development of an ultra-precision air-bearing stage that integrates real-time motion error measurement and compensation features. The motion errors addressed include horizontal and vertical straightness errors, as well as roll, pitch, and yaw errors. These errors are measured by an embedded system that incorporates five capacitive sensors and a reference mirror within the stage. A key advantage of this stage is its capability to perform real-time compensation using the internal measurement system and on-stage pneumatic regulators, eliminating the need for external measurement and compensation devices. Experimental results show a significant reduction in motion errors, with horizontal and vertical straightness errors decreasing from 3.09 and 1.95 μm to 0.29 and 0.25 μm, respectively. Additionally, roll, pitch, and yaw errors were reduced from 3.18, 3.45, and 4.93 arcsec to 0.35, 0.41, and 0.49 arcsec, respectively. These results clearly demonstrate the effectiveness of the proposed approach.