The subloading surface model is capable of describing the plastic strain rate by the variation of stress inside the yield surface. It is required to make the similarity-center of the subloading and the yield surfaces, i.e. the elastic-core move with the plastic deformation in order that the closed hysteresis loop is depicted realistically. The rigorous evolution rule of the elastic-core is proposed in this presentation.

The isotropic hardening stagnates for a while after the stress-reversal event. The rigorous formulation of this phenomenon such that the isotropic hardening develops continuously noy only during the monotonic loading process but also during the stress-reversal process is formulated based on the subloading surface concept.

In a Super-smart Society (Society 5.0), there is a need to create a sound field that can be used for high-level communication and amenity creation. Several theories have been researched that lead to the reflection and absorption of sound waves on walls, the concentration of sound within spatial regions, and the control of the direction of sound rays, and there is a strong need for technology that can map these results into real space. However, it is still a difficult task, and we have no choice but to start from where we can. In this paper, we review the theory of acoustic metamaterials and examine the concept of sound field control that is effective for a Super-smart Society through sound field analysis using the finite element method.

The elastoplastic and elasto-viscoplastic deformations not only for the monotonic but also the cyclic loading processes at the general deformation rate ranging from the quasistatic to the impact loadings can be described in a unified manner by the subloading-overstress model, while it is noteworthy that the elastoplastic constitutive equation limited to the description of the quasistatic deformation can be disused. It is desirable to extend the subloading-overstress model to describe the temperature-dependence, since the elasto-viscoplastic deformation behavior is influenced by the temperature in general. Then, the formulation of the subloading-overstress model is extended to incorporate the temperature dependence of Young’s modulus, the isotropic hardening, the kinematic hardening and the elastic-core translation and the thermal strain for metals for the range of temperature in the engineering practice in this article.

The Gurson model is capable of describing the ductile damage of metals. It is extended to describe the cyclic viscoplastic deformation behavior by incorporating the extended subloading-overstress model in this article. It is called the extended subloading-overstress-Gurson model.

In municipal waste final disposal sites, gas venting pipes are installed to release gases such as carbon dioxide and methane generated in the landfill into the atmosphere. Though each flow of air, water or produced gas in the final disposal site including the gas venting pipes has been shown schematically, there is no research which quantitatively confirmed these flow characteristics. In addition, the authors have been continuously investigating the cesium-137 concentration in leachate from several disposal sites in Fukushima Prefecture. In this process, it was considered that, since the area around the gas venting pipes is structurally more susceptible to rainwater flow than the waste layer, when radioactive cesium exists near the area around the gas venting pipes, it is strongly affected by rainwater, resulting in a high concentration in the leachate. Therefore, the flow of rainwater permeating in the final disposal site was analyzed by simply simulating the gas venting pipe and leachate collection/discharge pipe, and flow characteristics were clarified.

While I was working for an automobile company, I realized that in various designs and model analyses, by expanding the values, handled in the calculations from points to a distribution with a range and using that distribution as the calculation object, it became possible to handle the variation precisely, and a high level of performance and reliability could be achieved in various developments. After that, I have released tools that can realize complex design calculations and simulations as distributions. It is clear that calculations based on probability distributions can significantly improve calculation accuracy compared to the standard deviation and Monte Carlo method that have been commonly used.

This paper describes a level-set based topology optimization for bi-linear elasto-plastic problems. We use a topology optimization method that combines the level set method and the phase field method, where the geometric complexity is controlled by the value of the regularization factor τ. This study aims to improve the performance of structural design by applying this method in both weight reduction and productivity to bilinear elasto-plastic models. The focus of this paper is to obtain high performance geometry by simple elasto-plastic analysis, and a bilinear elasto-plastic model with yield stress as a boundary is used. In numerical experiments, the effect of Young's modulus in plastic region on the final topology in bilinear elasto-plastic model was investigated using a 2D three-point bending test model. As a result, different generated geometries were obtained by setting the value of Young's modulus in plastic region in the elasto-plastic analysis. A geometry that supports the load application area was obtained when the Young's modulus of the plastic region in the elasto-plastic analysis was small. However, when the Young's modulus of the plastic region was increased, reduce this tendency. Since Young's modulus after plasticity has a significant effect on the generated geometry, it is necessary to set an appropriate value for the purpose.

In this study, we designed a piezoelectric energy harvester using topology optimization based on the level set method, then fabricated and evaluated the device. The design method incorporated microfabrication constraints, particularly focusing on etchability, to ensure manufacturability. Constraints on cross-sectional shape, substrate dependence, and minimum output voltage were considered. The device was fabricated according to the optimized design, and its performance was evaluated. While the device did not fully meet the predicted performance, this discrepancy was attributed to residual stress introduced during the manufacturing process. These findings suggest that incorporating such factors into the optimization process could enhance the accuracy and reliability of performance predictions. Overall, this study demonstrates the potential of topology optimization in designing piezoelectric energy harvesters, while emphasizing the need to account for practical manufacturing challenges.

The present study proposes a method to optimize structure stiffness with the constraint of even reaction forces with a hyperelastic material under dynamic response. The dynamic variance of the reaction forces at the Dirichlet boundary is introduced to evaluate the uniformity of the reaction forces. Several formulations for optimal topology design are provided, including the constraints of dynamic variance of reaction forces. Multi-objective optimization is proposed considering dynamic compliance minimization and uniform distribution of force transmission. Sensitivity analysis is then derived based on the adjoint variable method and verified by the Finite Differentiation Approximation (FDA). Several numerical examples prove the effectiveness of the proposed method.

In this study, we propose a new multi-material topology optimization method based on reaction forces to improve the trade-off relationship between mass, stiffness, and energy absorption performance of structures such as car bodies. This approach addresses the limitations of conventional average compliance and energy-based metrics, accommodates large deformations, and facilitates experimental evaluation. By maximizing reaction forces, this formulation makes it possible to enhance energy absorption performance under enforced displacements. The hyperelastic multi-material topology optimization model is derived from previous studies. The proposed method is demonstrated and discussed through several numerical examples.

The need for thermal management devices has become crucial as computational performance demands have increased in recent years. However, heat sink design is a challenging issue due to the high complexity of thermal management. To address this issue, topology optimization is emerging as a promising method for heat sink designs because of its high flexibility. This study proposes a topology optimization method for micro heat sinks using nanofluids. The nanofluids are simulated using the Eulerian-Eulerian model based on the finite difference method, and the sensitivities required for optimization are solved through automatic differentiation. We present several numerical examples of optimized nanofluid heat sinks and verify the feasibility of this method.

In order to confirm the relationship between the electrode shape and the excitation vibration mode of the quartz crystal of the acoustic wave device, the main vibration was conventionally in thickness mode, but the sliding mode in the X direction of the crystal axis is the main vibration, and the secondary oscillation in the Z direction is controlled. The electrode shape is rectangular and the mode shape (fundamental mode distribution of the waveguide due to the difference in elastic wave velocity in the Z direction). The difference between the rectangular electrode and the mode electrode was clarified from the resulting frequency characteristics. In other words, in the case of a rectangular electrode, the main vibration in the X direction was in the case of the mode shape, the other sub-oscillations were suppressed. In other words, a waveguide extending in the X direction was formed, and the vibration mode, which is a secondary vibration excited in the Z direction, was determined by the electrode shape. This data is the basis for series expansion and multimode transmission using waveguides.

In this paper, we propose a new optimization procedure for texture design parameters that combines the design of experiments approach and the modified acceleration gradient method. For the design of texture parameters, it is necessary to determine the optimal combination of number, placement angle, spacing and horizontal shape of texture elements. There are many possible combinations of design parameters, so it is necessary to decide select efficiently. In a previous study, we clarified that frictional force can be reduced by varying the texture depth. In this study, we investigate a method of determining texture design parameters based on the design of experiments and the modified acceleration gradient method. Circular and rectangular shapes are generally employed as the horizontal shapes of texture elements, but Maldonado-Cortset et al.’s report that an s-shaped texture more effectively reduces the friction coefficient. In this investigation into the reduction of friction coefficient, numerical experiments were performed that included this s-shaped texture.

This paper describes a prediction method of the salinity concentration at the sea urchin farm in Hichirippu-numa, Hokkaido, using long short-term memory (LSTM). LSTM is one of the machine learning methods and also solves long term time series forecasting tasks. One of the main factors affecting the accuracy of the machine learning model is the data quality, although the observations at remote locations are used for real-time salinity concentration forecasting of sea urchin farm in the Hichirippu-numa. In order to improve the data quality, we focus on a data assimilation flow analysis based on Kalman filter FEM (KF-FEM). KF-FEM can estimate the water elevation at the specific location in computational domain. By using the results of KF- FEM considering precipitation, we generated more reliable training data for LSTM. In this study, we proposed to apply the results of the data assimilation flow analysis into training data of LSTM and confirmed proposed method is better accuracy of salinity concentration forecasting than conventional method.

In this study, the electromagnetic inverse analysis method is developed to identify the short circuit current in a lithium-ion battery nailing test. The method measures the non-stationary magnetic flux density generated by the current and identifies the current inside the battery. The analysis is considered in the frequency domain with the analytical solution of the electric dipole model. The numerical verification was performed, in order to show the basic concept of the method.

Numerical simulations of fruit growth of fruits such as apples were performed considering stresses. Simple growth model is assumed, and the final shape of the fruit was examined for several conditions. In the growth models, the load such as gravity and the inhomogenious growth rate were taken into account. Stress analysis and expansion analysis using finite element method were conducted to numerically calculate the final shape of the fruit, and the results were compared.

In air conditioner compressors, refrigerant and lubricating oil flow in the same space, so oil management within the compressor is important. Oil management requires visualization of the internal multiphase flow within the compressor, but experimental visualization is challenging because of the high temperature and pressure. Therefore, we use data assimilation to visualize and understand the internal flow from some experimental data.

In this study, we report the effectiveness of data assimilation for visualizing the internal flow field using twin experiments with a toy model. Using a toy model that simulates the internal flow of a compressor, we proved that the internal flow can be predicted in a flow field with a maximum flow velocity of 70 m/s by observing 50% of the state variables and assimilating data every 0.001 s. And We also confirmed that data assimilation was successful even when the number of observation data was reduced to 12.5%, and we confirmed the importance of observation data of the interior space.

Compliant morphing airfoils improve aerodynamic performance by actively elastic deforming its airfoil shape during flight. In our previous studies, the internal structure was obtained by topology optimization to deform the external shape of the airfoil to a predefined target shape that provides the desired aerodynamic performance. There, however, the problem was that it was not known in advance whether or not a given shape could be obtained, and deformation during flight could not be reproduced. To solve these problems, this study aims to propose a novel optimal design method for compliant morphing airfoils, based on static aeroelastic deformation, to simultaneously find the superior deformation shape and internal deformation mechanism from both aerodynamic and structural perspectives. This method does not require a pre-defined aerodynamic shape. This paper briefly describes the formulation of a multiobjective topology optimal design problem considering both aerodynamic and structural performance based on a coupled aerostructural analysis that combines an aerodynamic panel method with a structural finite element method. Through numerical examples, In our presentation, we will discuss the possibility that the compliant morphing airfoil, which is obtained by the proposed design method, could take advantage of passive deformation caused by airflow.

This study proposes a data-driven approach to support hypothesis formation aimed at improving the predictive accuracy of digital twin systems. By applying Lasso regression, the research successfully identifies key explanatory variables and clarifies the factors contributing to prediction errors in a methane fermentation plant—a chosen use case. The study demonstrates that the proposed approach can potentially enhance prediction accuracy by up to 17% when the suggested hypotheses are implemented in the digital twin system. Furthermore, these hypotheses undergo validation through a Human-in-the-Loop process, ensuring their practical applicability. The findings highlight the significance of hypothesisdriven methodologies in optimizing industrial processes and suggest broader applicability across various fields, including other industrial and bioengineering domains.

In this study, we propose a method using inverse analysis to identify the load distribution acting on the die cutter roll through the elastic deformation model of the roll, based on the roll deflection curve, which is relatively easy to measure. First, the configuration of this inverse problem is clarified by explaining the observation equation based on the elastic deformation model of the Bernoulli-Euler beam. This inverse problem is ill-conditioned because the roll has high stiffness in this context. The effective rank of this problem is evaluated through the model validation test using simulated observation data to quantitatively clarify the ill-conditioning. An approach to model the cutting load distribution using low-dimensional parametric function is also considered to overcome the ill-condition.

An estimation method for unknown constitutive parameters of elasto-viscoplasticity material models by constructing a database that consists of results of finite element analysis is developed. Recently, a Bayesian approach to estimate accurately material properties from data of a single spherical indentation test has been proposed. In this study, we aim to improve generality of the identification method by increasing the number of parameters to be estimated and verify the validity of the material parameter values estimated using spherical indentation test data. Identification of four material properties was attempted, but the data did not differ in the range of small values for the strain sensitivity index, and the identification could not be performed with sufficiently high accuracy. In selecting the parameters to be estimated, it is necessary to fully verify the sensitivity of the parameters to the data used for Bayesian estimation.

In analyzing flow with an interface between two phases, particle methods are commonly used. In most situations, the flow is considered a free-surface flow and ignores the effect of gas-phase. However, such a simulation cannot deal with bubble behaviors. On the other hand, when the flow is treated as a multiphase flow, the calculation easily becomes unstable due to the discontinuity of the physical property. In this study, a new numerical method for the computation of multiphase flow, especially liquid-gas flow, is proposed and validated through simple test cases.

In fluid machines, oil is injected to prevent gas leakage from the gaps between the rotors and to prevent deterioration due to friction. The oil becomes a thin film due to the collisional jet, which is difficult to simulate accurately with the particle method that solves the three-dimensional Navier-Stokes equations. In this study, an MPS Method for Shallow Water Equation incorporating a particle splitting technique is developed. Simulation of a dam collapse problem was performed, and the result was in good agreement with the experiment. In addition, a thin film flow produced by a collisional jet was simulated.

The “giga-casting”, a technology for casting one-large aluminum product, has been attracted much attention in the automotive industry. For stable production with giga-casting, the quality control of molten aluminum is important. Understanding the behavior of inclusions in the entire molten aluminum is necessary for efficient and reliable control of cleanliness of molten aluminum. Computational fluid dynamics is helpful to understand the behavior of inclusions in the entire molten aluminum and to extend the scale from laboratory scale to the actual scale. In this study, the behavior of inclusion in molten aluminum is simulated with the Moving Particle Hydrodynamics (MPH) method. As a result of conducting the simulation with different values of the effective radius, it was found that the alumina particles tend to settle more easily when the effective radius is larger. In addition, by simulating the experiment with the MPH method, the sedimentation rate of the alumina particle was on average close to the experimental results.

When the finite element method is used in structural analysis, the problem of extremely low analysis accuracy due to poor quality mesh partitioning has been a concern, and mesh-free analysis methods that do not depend on mesh partitioning for calculation accuracy have been proposed. The Fragile Points Method (FPM) is proposed as one of the mesh-free methods suitable for large deformation analysis, offering the advantage of performing analyses without suffering from the deterioration in analysis accuracy due to the degradation of mesh quality associated with large deformations. In this study, the numerical solutions of the FPM for small deformation elasto-plastic problems were obtained, and their accuracy was evaluated.

Recently, the isogeometric boundary element method (IGABEM), which combines isogeometric analysis and the boundary element method, has been proposed. IGABEM is a method that combines the use of the same geometric representation as CAD and Isogeometric Analysis (IGA), with the boundary-only discretization characteristic of the Boundary Element Method. This approach is expected to eliminate the need for mesh generation, allowing CAD models to be directly used as analysis models. However, the implementation of IGABEM requires proper handling computations known as singular integrals. In this study, a method is proposed to eliminate the need for handling singular integrals by placing source points exterior of the domain of geometry. Numerical analysis was performed for a two-dimensional elasticity problem, and the accuracy was evaluated by comparing it with the exact solution of linear elasticity. The results show that by even placing the source points outside the domain, the analysis can be performed accurately.

Additively manufactured particle dampers (AMPDs) are novel damping devices formed by intentionally preserving unfused powder within a structure during laser powder bed fusion (LPBF) manufacturing process. Despite their innovative design, the damping mechanism and effectiveness of AMPDs remain uncertain due to the absence of the corresponding numerical analysis. This study focused on exploring the damping mechanism and performance of AMPDs experimentally and numerically. Several AMPDs with the same cavity ratio but different cavity distributions were fabricated using LPBF with 316L stainless steel powder. Complex power experiments were conducted at a vibration frequency of 200 Hz and an acceleration range of 50–250 m/s². Subsequently, a discrete element model (DEM) with a reasonable number of particles and simulation parameters was proposed and cross-verified with experimental results. Notably, the damping mechanism of AMPD was revealed through numerical analysis. Furthermore, the influence of cavity size on the damping performance of AMPD was discussed through both experimental and simulation methods.

Peridigm, an open source implementation of the fracture analysis theory Peridynamics, requires complicated preparation of input data for particle models, making efficient analysis difficult. To solve this problem, we developed an integrated analysis environment, FRAXST, which generates particle models from standard UNV format data, such as those used in the finite element method, and enables effective setting of various analysis conditions for the governing equations in integral form. This paper reports on this support function and an example of analysis.

The rubber, the main material of power transmission belts, contains short fibers for reinforcement, and its mechanical behavior changes depending on the amount and distribution of the fibers. If both the short fibers and the base rubber were modeled as solid elements to predict this with FEM analysis, the total number of elements would become enormous, increasing the calculation costs. Therefore, the short fibers were modeled as a series of truss elements, and an element for embedding the short fibers in the base rubber of the solid elements was developed and implemented in FrontISTR, an open-source software, significantly reducing the total number of elements. This makes it possible to quickly analyze short fiber-mixed rubber with fewer calculation resources.

The contact analysis of various real-world models was performed using FrontISTR, and it was confirmed that FrontISTR can be applied to complex and large-scale contact problems such as bending machines, and that parallel computation using multiple CPU cores can significantly reduce the computation time. The use of the OSS FrontISTR enables low-cost and high-speed contact analysis.

Computer-aided Engineering (CAE) becomes increasingly important in product development processes. To improve design quality and reduce development period, we aim to utilize large-scale computing environments such as public supercomputers and cloud computing services. In this report, we tested an open-source solver FrontISTR that has no license restrictions and verified its functions with chain tensile model. As a result, FrontISTR can calculate results comparable to that of commercial solvers in less time. We also built a system to optimize design geometry embedded a FrontISTR and using the supercomputer Fugaku. By operating the constructed system, we could find dimensions that reduce the mass of the chain components by 7.2% without increasing the maximum principal stress. On the other hand, since the pre-processing could only be performed sequentially, the process was not accelerated through multi-condition parallel calculations. In addition, FrontISTR calculations, which were stable on systems with Intel or AMD CPUs, became unstable when executed on Fugaku.

FEM_FrontISTR is a tool that calls the analysis functions of FrontISTR, a parallel finite element method structural analysis software, as a workbench of FreeCAD, a CAD application with users in many countries. FEM_FrontISTR can smoothly perform all processes of finite element analysis from model geometry creation to visualization of analysis results on the same screen, which is a significant advantage over the pre and postprocessing tools conventionally used in FrontISTR. In previous developments, support for multiple materials and boundary conditions has been increased, and support environments have been built for analyses with nonlinearity, such as material nonlinearity and contact analysis. This paper focuses on the extension of analysis types and describes the implementation of functions for eigenvalue analysis and time history response analysis by time integration, which are included in the framework of dynamic analysis. With this implementation, about 70 % of the functions available in FrontISTR can now be performed through FEM_FrontISTR. FEM_FrontISTR is useful as an introductory tool for FrontISTR due to its low introduction cost, and in the future, there is a need for more tutorials for beginners.

This paper introduces the use of the Fugaku and HPCI supercomputer systems. The HPCI project includes Fugaku installed at RIKEN and other supercomputer systems at 14 research institutions. They are basically available free of charge. There are two types of calls for proposals, one is regular calls to be done once or twice a year (the second call is applicable to Fugaku only), and the other is calls opening throughout the year. The calls opening throughout the year include both trial access proposals and Fee-based access proposals for highly confidential projects. Fugaku trial access includes first-touch option (1000NH. 3months) available with easy application for use. Several types of user support are available for the use of these supercomputer systems. For example, “Program Tuning support” provides supports for application porting, serial and scalability optimization. “Accompanying support” for industrial use is available when it is difficult for the company to solve the problem by itself. Pre-installed software for Fugaku is available by spack, and please check for the latest information as it is updated from time to time. For more information, please visit https://www.hpci-office.jp/ .

In recent years, ultrasonic speakers, which are used as a foundational technology for three-dimensional acoustics, have the advantage of extremely high directivity due to the use of ultrasonic waves as the carrier wave. However, a challenge arises in the form of sound quality degradation due to waveform distortion during demodulation. Therefore, in this study, we conducted a fundamental examination aimed at establishing a design method for high-quality reproduced sound. This was achieved by connecting linear wave acoustic phenomena and fluid phenomena through a partitioned iterative coupling algorithm, thereby reproducing nonlinear acoustic phenomena in large spaces without solving the memory-intensive Westervelt equation required for large-scale analysis using the finite element method.

A fluid-structure interaction (FSI) analysis model based on the partitioned coupling approach and a dynamic mode decomposition (DMD) approach with a mode sensing approach by the Greedy algorithm were developed for flexible aeroshell, which is one of the innovative atmospheric-entry technologies. This FSI model is composed of open-source software SU2 for the flow field, CalculiX for the structure field, and coupling library preCICE. The present DMD approach reproduced dominant modes from the spatiotemporal data of displacement by the FSI simulation. It was revealed that the deformation of the aeroshell and torus mainly consists of modes combining a swinging motion and membrane deformation of about 160—170 Hz, and translation motion with membrane deformation of 550 Hz. These DMD-reconstructed modes indicated deformations and frequencies are completely different from the natural oscillation modes. This is because the deformations of aeroshell were significantly affected by aerodynamic forces.

In the flow governing equations with embedded object boundaries (IB-NS), the flow solution around the object shape expressed by a level set function is defined without boundary conditions. This makes it possible to analyze the flow simulation of objects of arbitrary shapes represented by digital image data, etc., using only image processing technology. In this report, we introduce an example of the implementation of a flow simulation driven by image data representing two-dimensional and three-dimensional object shapes, using an open source image processing library.

Among various methods to develop surrogate models of high-fidelity models, reduced-order modeling using proper orthogonal decomposition (POD) is one of promising methods. However, the method has a well-known drawback that the computation of Galerkin projection is heavy, which overshadows the reduction of computational cost for simultaneous equations. To speed up the simulation, a hyper-reduction method, which approximately calculates Galarkin projection, is introduced. A hyper-reduction method allows to avoid the full-domain integration. As a result, a great speed-up is realized. Although several hyper-reduction methods have been proposed, an empirical cubature method (ECM) is widely used. The present study proposed a novel ECM based on Gappy-POD, which is one of the sparse sampling techniques and was originally proposed for image reconstruction. In this study, we applied the proposed method to various types of nonlinear analyses. As a result, we confirmed that the proposed method can provide the effective hyper-reduced-order model in significant reduction of computational time while preserving desired accuracy.

In this paper, we construct a surrogate model for radiation dose rate predictions using simulation results and deep learning, specifically for an indoor space containing a square pillar, and verify its accuracy. We also demonstrate, based on the principle of superposition, that the surrogate model can predict the distribution of radiation dose rates in a space with multiple radiation sources. Furthermore, we propose a method to predict the radiation dose rates in a space with multiple square pillars and sources by using a corrected surrogate model. Based on these findings, we assess the feasibility of predicting the radiation dose rates in the reactor building with complex structures in real time and with high accuracy.

This study investigates global sensitivity analysis methods on mathematical models in terms of a tradeoff between computational cost and accuracy. Sobol’ method based on Polynomial Chaos Expansion (PCE-Sobol’), Morris method, and variance analysis with orthogonal tables are applied to a mathematical model of an experimental facility for a space propulsion system. It was found that variance analysis with an orthogonal table is the most efficient for identifying the most sensitive factor with fewer evaluations. Conversely, the PCE-Sobol’ and Morris methods are more suitable for screening sensitive factors with hundreds of evaluations.

Fin-integrated phase change material (PCM) devices embedded in satellite panels are being investigated as devices for passively controlling the heat of high-heat-generating equipment. In this study, we aim to streamline the design process for structurally integrated PCM devices and report on the development of surrogate models for thermal-fluid coupled analysis using physics-informed neural networks (PINNs) and sparse identification of non-linear dynamics (SINDy), as well as their practical applicability.

This study proposes the uncertainty model for the resilient operation of our proposed smart antenna for the radio astronomy mission. The smart antenna recovers the antenna gain by compensating the deformation of the primary reflector using actuators installed in the sub-reflector. We have two important tasks for sufficient antenna gain recovery to achieve the resilient operation. The first is to achieve the recovery gain as much as possible. The second is to predict the recovered antenna gain, because we have no method to know the actual antenna gain except for the successful observation. In addition, the antenna gain recovery is degraded by several uncertainties including the measurement errors, install errors and so on. This manuscript describes the uncertainty modeling method for prediction the antenna gain recovery during the mission.

Toshiba Corporation, Corporate Research and Development Center, Advanced Intelligent systems laboratories The lifetime of a battery varies depending on its operational history, such as charge-discharge waveforms and temperature conditions. For improving the reliability of battery systems, it is important to have probabilistic reliability prediction methods that can consider the impact of uncertainties, such as load history and cooling performance, on lifetime distribution. In this study, we propose a method for rapidly executing probabilistic lifetime predictions by modeling the irregular waveforms of charge-discharge of lithium-ion batteries with an Evolutional spectrum, and utilizing a surrogate model based on Operator Learning, which can learn the operators of differential equations related to electrical circuit analysis and thermal analysis.

CALPHAD database coupling non-equilibrium multi-phase field method is applied to study hot tearing analysis for a practical engineering nickel alloy in the powder bed fusion process. It is confirmed that the probability of liquid film between grains varies with the cooling rate differences. Al solute segregation is also confirmed to be vary with the probability of liquid film

For numerically investigating motion of fine droplet on micro/nano-structured solid surfaces, phase-field model (PFM)-based computational fluid dynamics (CFD) simulations are performed in two and three dimensions. The Cahn–Hilliard (CH) equation or the conservative Allen–Cahn (CAC) equation is adopted to calculate the advection and construction of diffusive interfaces. Lattice Boltzmann Method (LBM) -based in-house code or Finite Element Method (FEM)-based commercial CAE software is used to solve a set of the Navier–Stokes (NS) equations of fluid and the diffuse-interface advection equation for an immiscible incompressible isothermal two-phase fluid system. The PFM-CFD simulation results are evaluated in comparison with available experimental data to optimally design various micro-fluidic devices such as a novel heat exchanger which can enhance dropwise condensation with structured superhydrophobic solid surface.

The growth kinetics of the antiphase domain (APD) in bulk intermetallic have long posed a challenge because of the variations in curvatures in 3D space. For instance, attempts to evaluate the ordering kinetics in D0₃-structured Fe₃Al and control the APD size have been made since the 1970s. However, the determined ordering kinetics to date exhibit significant discrepancies, and no synergistic studies have been conducted. In the present study, the shape coefficient, which correlates the growth rate of APD with 3-dimensional intricate shapes and the shrinking rate of 2-dimensional circular antiphase boundaries (APBs) was assessed through a comparative analysis of PF simulations using 2D- and 3Dmodels. Conversely, the kinetic of APD size evolution in bulk samples was determined using TEM observation. By incorporating the calculated shape coefficient and experimental results, we successfully derived accurate values of mobility for forming the D0₃-type ordered structure. This finding lays the foundation for optimizing heat treatment conditions to regulate APD structure and enhance the superelasticity of Fe₃Al. This methodology can be extended to estimate the ordering mobility of other intermetallic.

In recent years, there has been a lot of research into estimating parameters from microstructure data using data assimilation with the phase-field method. In this method, parameters are determined so as to reproduce experimental data of microstructure, but in order to obtain valid information from microstructure data, it is important to carry out a sufficient certainty evaluation of the estimated parameters. In this study, we propose a method for estimating parameters using automatic differentiation, a technique that is widely used in the field of machine learning in recent years, and a method for evaluating the certainty of parameters based on the Hessian matrix calculation of the objective function using automatic differentiation. The proposed method was verified using twin experiments that used pseudo-experimental data generated using simulations, and it was shown to be effective for evaluating certainty of estimated parameter.

Thermal Interface Material (TIM) development with high thermal conductivity is necessary for power module of automobile. However, the mechanism analysis of heat transfer between filler and resin in TIM is not sufficient. In this study, the three-dimensional heat conduction was visualized, and the heat transfer coefficient of the interface was quantified as an experimental value. The heat transfer coefficient due to phonon conduction was calculated by the thermal molecular dynamics (MD) method. The results were compared and the relationship between the bonding state of molecules at the filler and resin interface and the heat transfer was found.

In various electronic equipment, it is an issue to design solder joints in consideration of vibration fatigue and thermal fatigue. It is known that the fatigue life of solder joints is sometimes much shorter than that predicted by Mines’s rule on the combined vibration and thermal cyclic loading test. In order to elucidate the mechanism of this shortening, a crack propagation simulation method considering the combined loading is proposed in this paper. In this method, critical section is incorporated in the solder part of the FEM structural analysis model, the amount of damage of elements in the critical section under random vibration and thermal cycle loading is calculated, and the elements whose amount of damage exceeds a certain value are deleted to simulate crack propagation. This simulation method is applied to a structure in which a leaded component is soldered to a wiring board. The simulation results show that, crack propagation behavior is different between vibration loading and thermal cycling, and this difference causes a shorter number of cycles to fracture than that estimated by Miner’s rule in the case where thermal cycling is first loaded and vibration is loaded later.

The study proposed an FEM simulation to reproduce the fatigue crack network failure where cracks occur randomly and repeatedly under a uniform equibiaxial stress field. In this method, randomly occurring cracks are assumed to be deviations following a normal distribution, and crack initiation cycles are calculated with the crack initiation resistance set on the FEM elements using normally distributed random numbers. A fatigue crack growth law is then applied to the cracked element to simulate damage development. The proposed method reproduced the logistic curve-type damage development behavior that are characteristic of fatigue crack network failure. Furthermore, the proposed method is applicable to crack propagation damage from the periphery of die-attach joints and damage in power cycling tests with temperature distribution, and provides a basis for establishing a fatigue life prediction method for die-attach joints.