AI-based inspection of the axes of machine tools.

Artificial intelligence (AI) encompasses versatile computational tools easily adaptable to various applications. This study delves into diagnosing linear stages, crucial components in machinery such as machine tools and additive manufacturing equipment, essential for precise linear motion. Adding sensors to these stages for precision monitoring proves costly and inconvenient in industrial applications. For sensorless diagnostics of linear stages, this paper introduces an innovative automated machine learning (AutoML) approach. AutoML employs diverse methods to interpret electric motor current signals, estimating the extent of misalignment issues. Support vector machine (SVM), gradient boosting (GB), and auto-multilayer perceptron (AutoMLP) methods classify the data. To boost performance, ensemble learning (EL) combines estimations from each method for a final decision. Motor signals, saved in the user interface's database, drive the table horizontally and vertically. AutoML learns proper classification through the user interface, using data and user interpretations for training and subsequent classifications. Hyperparameter optimization improves classification performance, with experimental studies demonstrating the superior fault detection capability of the ensemble method in motor current signal analysis. The findings suggest that monitoring motor current signals and leveraging AI tools for evaluation of the performance of manufacturing operation and machine diagnostics will be essential in the near future. Machine tool and additive manufacturing system manufacturers should include AI tools to their controllers. [ABSTRACT FROM AUTHOR]