In recent years, there has been a demand for the popularization of electric vehicles and further efficiency improvement of internal combustion engines for environmental measures. The cylindrical shape of motor cases and cylinders is machined by boring, but the problem is that the cylindrical portion is distorted by the fixing force applied during machining and the roundness is reduced. The solution is to perform non-circular machining in anticipation of distortion, but at present there is no machine tool spindle that can achieve this at high speed. In this research, we have developed a machine tool spindle with a mechanism that can control the rotational speed and position of the tip of the tool in the radial direction using a special motor. Radial position control is performed synchronously with rotational speed control by a real-time controller. By controlling precession and nutation during rotation of the tool tip, it can be applied to non-circular machining of motor cases and cylinders. In this paper, we have developed a prototype machine with the proposed spindle structure and verified its basic performance through experiments under no-load conditions.

Additive manufacturing has been used to fabricate molds with intricately curved cooling channels, because it is easy to fabricate complex shapes. However, there is a problem that the surface of cooling channels fabricated by additive manufacturing is relatively rough. Therefore, it is necessary to reduce the surface roughness of the cooling channels. Because conventional smoothing methods usually require a rotating tool, they are unsuitable for handling channels with complex shapes. In this study, the authors proposed the use of electrochemical machining (ECM) to realize surface smoothing because it does not require rotating tools to smooth the channel surface. We designed and fabricated a tool electrode unit with a disk-shaped electrode sandwiched between insulating hemispheres that can move freely along complex channels. The movement and smoothing characteristics of the developed tool-electrode unit were investigated experimentally. It was found the tool unit could move freely through a curved channel, and the surface roughness was decreased by scanning the ECM tool unit through a straight channel. Furthermore, it was found that the larger the machining amount, the better is the surface roughness.

This paper describes the method of data augmentation for achieving accurate visual inspection by machine learning. Recently, the application of machine learning for accurate visual inspection has been expected. Generally, machine learning requires many training data. When applying machine learning to visual inspection, a lot of normal images and a lot of defect images are required for training. However, there is a shortage of defect images for machine learning because it is difficult to obtain many defect images from manufacturing factories. This is a critical issue for achieving accurate visual inspection. We propose a method that extracts normal and defect information from images by the principal component analysis and fuses the information in Eigen space to generate high-reality defect images. Moreover, when using these generated defect images for training classifier, the accuracy of normal/defective product discrimination achieved 96.3% for the capsule dataset. In the case of the pill dataset, the accuracy achieved 91.3%. For the aluminum plate dataset, the accuracy achieved 97.8%. These results show the higher accuracy than those of previous data augmentation methods. This shows generated defect images by the proposed method are effective for training a classifier.