This paper proposes an analytical calculation method of frequency characteristics of DC input impedances of an IPM motor drive system for design and stability analysis by the impedance method. The proposed method is based on an averaged frequency model linearized at an operating point. The developed small-signal model considers control principles and mechanical perturbations. It is an analog and digital mixed model which is controlled by a digital signal processor system. Analysis examples of frequency characteristics for the impedance method by the proposed method are investigated and validated analytically and experimentally.

In recent times, as digital transformation (DX) is rapidly advancing, the cultivation of personnel who are proficient in data analysis through AI and the experience of its results in digital spaces such as VR, AR, and so on is an urgent issue. In new technology fields, it is necessary to polish core technologies and understand the contents. On the other hand, social changes are steadily promoted not only by advanced technology but also by the ideas and execution power of people who use new technology. Therefore, we have been working on digital human resources development for not only researchers and developers but also for users, with the goal of raising human resources. In this paper, we will use the persona method in UX design to generate ideas on the theme of metaverse, extract AI functions that can be solved, and create and evaluate prototypes in the metaverse space. By providing feedback to the ideas, we aim to improve the understanding of AI technology among users. This research will conduct practical training for university students, and the results, evaluations, and considerations will be presented.

In order to realize a digital twin, in addition to future prediction technology using AI, modeling technology for virtualization specialized for physical systems and know-how for identifying data used in modeling are required. However, there are many cases in which the know-how specialized in these physical systems is not sufficient in educational settings. For this reason, we set assumptions about modeling techniques specific to physical systems and data to be collected, and prepare prototypes that realize digital twins based on those assumptions. This prototype provides a foundation for machine learning, AI model tuning, and implementation of trained AI models into digital twins. By experiencing the construction of this prototype, we will cultivate the ability to construct future prediction systems using AI in digital twins. In addition, by setting various hypotheses and carrying out realization and verification with prototypes, the educational effect will be enhanced. This paper describes the proposal of this educational method and its practice and evaluation.

Control systems such as factory production systems and mobility systems require countermeasures to prevent serious physical damage from cyber-attacks. Although there have been reported cases of attacks such as installation of unauthorized devices in internal networks and physical tampering with vehicle networks, sufficient countermeasures have not yet been taken in the control systems currently in operation. In this paper, we propose a method to detect network route tampering by adding calculation and communication functions for network tampering detection to devices and components that make up bus-type networks commonly used in field networks. The proposed method makes it possible to detect anomalies caused by network route tampering without changing the specifications of field devices and with minimal impact on communication performance on the field network.

AI, Artificial Intelligence, is being used in many products and services while still being developed. A policy of human resources development that responds to the times and sustainably applying AI to all industries, regions, and governments was issued in the Cabinet Office’s AI Strategy 2019. In this paper, we suggest an AI education method for AI users to expand the base of AI human resources. This education method teaches AI users about various AI cases, data analysis functions, and their relationships. Then, the method creates ideas that can be solved by AI using UX, User Experience, persona and they improve AI understanding. AI users create ideas that can be realized with AI on familiar themes that everyone has experienced. After that they learn what functions the idea is used. In addition, they improve AI understanding by repeating training in the AI users’ specialized fields. Also, we describe the practical evaluation and result of their understanding for AI users, humanities, science university students.

In this study, we propose a method to clarify the application limits of DNN (Deep Neural Network) based visual inspection systems. A process for determining inspection results is a black box because DNN automatically extracts features. However, visual inspection requires judgment based on specifications. The problem is that the basis for the decision is unclear. To address this problem, we interpret the feature space of DNN using known features. Firstly, it generates data with explicit knowledge characteristics (for example, defect length, area, shading depth etc.) that can be arbitrarily modified. Secondary, the generated data are input to trained DNN models and observed the coordinate changes caused by their explicit knowledge characteristics in the feature space. Preliminary experiments on the MNIST dataset (public dataset) confirm that the DNN feature space is able to represent quantitative feature variation. Experiments using actual inspection images also confirmed the effectiveness of the proposed method.