Reduction of AC losses is important for practical application of HTS (High critical Temperature superconductivity) power cable. In this study, we aimed to reduce AC losses by increasing engineering critical current density for one phase of three-in-one cable of 66 kV-5 kArms class. To increase the engineering critical current density of the cable, we modified the conventional cable design which is developed in Japan. Specifically, we transitioned from a structure with four conductor layers and two shield layers (Cable A) to a design with five conductor layers and two shield layers (Cable B), while maintaining a constant cross-sectional area of the conductor layer. This study employed methods electric circuit model and 3D electromagnetic field analysis for design and analysis. Since HTS cables consist of helically wound HTS tapes, determining the winding direction and winding pitch (two variables are referred to as cable configuration parameters) is critical to cable design. We calculated the AC loss of the designed cable model. Then, we compared this calculation result with measurement result of Cable A obtained by Fujikura Ltd. as well as our calculation result of Cable A. As a result of analysis, measurement result of Cable A is 1.4 W/m, calculation result of Cable A is 1.5 W/m and calculation result of Cable B is 1.13 W/m. From these results, reduction of AC loss is confirmed by increasing the number of conductor layers.

View full abstract

Land-based recirculating aquaculture system (RAS) has gained significant attention as a key technology for the safe and environmental-friendly food production. A land-based RAS have the capabilities to exhaust almost waste water, control water condition and prevent disease transmission. Since a RAS system have not been fully established, cost structure of a RAS remains unclear. The cost structure is a significant to estimate aquaculture business feasibilities, therefore the present study built a land-based RAS management simulator. In the practical land-based RAS, water control condition and fish meal demand can fluctuate with the fish growth. However, traditional RAS models have contained no fish growth model. The present study, assuming aquaculture system for a triploid Atlantic salmon (Salmo salar), developed a RAS simulation model with the growth model: the growth model is based on the von Bertalanffy equation. Applying this simulation model, sensitivity analysis was conducted to estimate the effects of salmon price, fish meal price and government grand on the cash flow on the RAS. From this simulation, the present study clarified the operation condition to return profitability in ten years.

View full abstract

This study aims to enable real-time visualization of various physical phenomena using Digital Twin technology. The concept of the Digital Twin refers to the real-time integration and visualization of a physical entity and its corresponding virtual model constructed within a virtual space. In this paper, we explore the application of Digital Twin to physical models such as experimental devices. As an example, we apply the Digital Twin to an ion beam experimental system. The physical twin and its virtual counterpart in the virtual space were synchronized via an operational interface to establish the Digital Twin.

View full abstract

This study explores interactive 3D visualization systems for physical phenomena using HTC VIVE XR Elite, Microsoft HoloLens 2, and Sony Spatial Reality Display. We developed a multi-platform system that allows users to visualize and interact with 3D data, comparing the advantages and limitations of each device. HTC VIVE XR Elite provides immersive experiences suitable for complex simulations. HoloLens 2 enables seamless integration of virtual models with the real world, ideal for hands- on learning, though it has processing limitations. SRD offers a simple, group-friendly experience but lacks interactive depth. The findings highlight the potential of each device for different educational contexts and suggest that combining these technologies can enhance learning experiences. Future work will focus on improving device performance and interaction capabilities for more effective educational applications.

View full abstract

Automatic summarization technology plays a crucial role in generating summary sentences for materials such as meeting minutes in assemblies and judgments in court cases. In recent years, generative summarization techniques utilizing large-scale language models have been employed, achieving a certain level of accuracy. However, these methods face the issue of numerical information errors. This study proposes a method to enhance the accuracy of numerical data in automatically generated summaries by identifying errors in numerical data and automatically correcting them using the original text and the summary sentences. The first method identifies the target text for summarization containing numerical values and units from the original text and compares the numerical values. The second method identifies the dependency origins of each numerical value and compares them based on that information. The accuracy of the proposed methods is 0.573 for the first method and 0.754 for the second method, demonstrating partial capability to correct numerical values to align with the original text.

View full abstract