This paper proposes the methods for integration of product model. It is important to comprehend how design parameters interact in product system, since the interaction among design parameters effects on product performance. Use of multi-domain physical system simulation tools spreads widely in recent year for reducing developing period efficiently. However it is still difficult to deduce and clarify the parameter relationships. One product is commonly decomposed to dozen simulation models to describing complex motion. It makes harder to understand the interaction among design parameters by hiding relations among models. The methods proposed in this paper enable integration of several product information models from simulation tools and construct parameter network. It is expected to comprehend parameter interactions easily between several product models by integration of models.

View full abstract

System has progressed by increasing the scale and complexity, and the diverseness of demand. According to it, development time becomes long and that poses a problem. Writers have proposed SDSI-Cubic to it as a framework for performing the optimal design automatically in the system design process. It also enables the designer who does not know optimization to perform the optimal design. However, although we argued to execution of optimization with manual operation, and modeling of system profile for optimization, by the present perfect automatic optimization was not performed. As the reason of this, there has been no technique for deriving a systems design workflow from the defined system model. This paper describes to the method for creating the system design workflow automatically.

View full abstract

System has progressed by increasing the scale and complexity, and the diverseness of demand. According to it, development time becomes long and that poses a problem. Writers have proposed SDSI-Cubic to it as a framework for performing the optimal design automatically in the system design process. It also enables the designer who does not know optimization to perform the optimal design. However, although we argued to execution of optimization with manual operation, and modeling of system profile for optimization, by the present perfect automatic optimization was not performed. As the reason of this, there has been no technique for deriving the optimization problem to optimization workflow in system design workflow from the system profile automatically. This paper describes to the technique for defining the optimization problem automatically.

View full abstract

Conceptual design has been recentely increasing its importance and it has been strongly desired to develop computer support technology for this design process. A major issue is computer support of the generation process of function structures. This process is carried out depending strongly on designer's implicit knowledge which is hard to unravel. Therefore, it is difficult to establish a methodology for supporting this process. One approach to the function structure generation support is exploiting functions and their structures of existing products, and some methods taking this approach have been proposed. However, the methods requires users to analyze the function structures of existing products and build a database of them in advance, which are labor-intensive. It is desired to develop a method which does not require such labor-intensive works by taking another approach where information other than existing products is utilized. This research takes this approach and focuses on linguistic information included in functions which are described in a national language. In this paper, relationships among verbs included in the national language description of functions are discussed and a function structure generation method exploiting the relationships is proposed.

View full abstract

It is said that the model-based design methodology has big advantage in terms of cost, development time and quality comparing with traditional design methodology. Many simulation and analysis tools are used in the method and they require great amount of know-hows about hardware, software, libraries, theory, etc. Traditional education and manual information are not enough to keep knowledge new and well. To solve these problems, we discuss know-how management methods for model-based manufacturing. First, we focus on many kinds of general work matching services on the internet. We analyze the merits and problems of those services. Finally, we discuss the possibility of the use of those services in the future for model based manufacturing industry.

View full abstract

We have developed design awareness support system which could perform a design of high quality in a short time. Features of developed system are following three functions: (1) Design navigation function prevents omission of procedures by visualizing the standard design process (2) Information automatic delivery function prevents defects and other problems from reoccurring by automatically retrieving trouble information that relates to current work and by delivering it to users (3) Design rule check function prevents problems at the manufacturing stage by finding violation portions of 3D CAD shape in the design state. And we confirmed these functions could contribute to efficiency improvement of the development of products.

View full abstract

Fault Tree Analysis is one of effective methods of ensuring security and safety of a product. Creating a fault tree, however, requires a deep knowledge about the product and physical phenomena, and it takes a lot of time and effort to create one. To solve this problem, authors propose a knowledge management method called quantity dimension indexing. Authors also report the efficiency and effectiveness of computer-aided fault tree analysis software (FTAid) which uses this method. Lastly, the authors propose an automatic indexing method using shallow parsing for the usage of quantity dimension indexing in real work places.

View full abstract

This paper describes a method of converting block diagrams into fault tree forms for computer-aided fault tree analysis (FTA). This conversion method aims at achieving high quality and efficiency of FTA for reliability design in manufacturing industry. The method has been implemented as one of the capabilities of computer-aided FTA software "FTAid", which has been developed in a case study to realize a practical software product for the corporate business based on academic research products. The proposed method enables to construct fault tree (FT) automatically from a block diagram of the system to be analyzed. Moreover, the resulted FT includes the additional information to be utilized by the other capabilities of FTAid for supporting FTA. Therefore, the proposed method can contribute to achieving high quality and efficiency of FTA.

View full abstract

In this paper, the authors report the effect of using FTAid and it has difference by education for making FT diagram. FTA is a method of analyzing and visualizing the causes of a fault event by Fault Tree diagram (FT diagram) in problem resolution of product and design processes. However FTA has a problem that is so many working hours to make a FT diagram without any misunderstandings and oversight. To solve this problem, the collaborative research group ; Jatco Ltd, The University of Tokyo and AIST, developed computerized support system named "FTAid". We distributed the FTAid to designers in JATCO, and analyzed FT diagrams made with FTAid and their working hours. We distributed the FTAid to designers in JATCO, and analyzed FT diagrams made with FTAid and working hours. As a result, the effect of using FTAid has difference by education for making FT diagram. Then we suggest FTAid needs education for making FT diagram to use it effectively.

View full abstract

As the rapid growth of technologies and global competition intensifying, quality management of product development becomes difficult and quality problems are increasing. Most of the troubles are caused in the design phase, and they should have been prevented if design knowledge is applied and examined appropriately. We have made a field survey of design and review processes from a standpoint of knowledge management, and discussed the problems extracted from the survey. This paper describes the result of the survey, a proposal of knowledge management platform, and technical or operational requirements of the platform components.

View full abstract

The increase of quality problems is one of the most important management issues for manufacturing business. This originates from decreasing the number of veteran engineers and shortening term of product development cycle. It is known that most of the problems are caused in the design phase. According to our survey, we found the following three main issues about design and review process for prevention of the quality problems: reducing troublesomeness of design validation task, sharing design consideration records, and maintaining reliability of design knowledge. To solve these issues, we proposed a knowledge management platform. Based on the proposal, we have developed a design review support system and applied it to electrical circuit design process. This paper describes the functions of the system and the results of evaluation experiments.

View full abstract

We have developed an XML schema of manufacturing case data and visualizing methods of manufacturing knowledge. The manufacturing case data is composed of material specifications, processing conditions, machine specifications and evaluation results, which are basically represented in text to share and to compare the associated knowledge. The data is including multimedia data partly, but it is difficult to handle directly the visual data in XML for the purpose. In the paper, we proposed an image search method and examined its possibility through the experiments on the appearance images of laser cutting surface and beads of arc welding.

View full abstract

In this paper, the author discusses the importance of time axis in designing products and services. Two examples examined are: systematic classification of approaches to design against disaster based on stress - strength model in reliability engineering, and Expectology as a framework for a systematic and MECE (mutually exclusive collectively exhaustive) classification of consumers' every possible cognitive and emotional response to product and service design prior and posterior to experiencing and purchasing them: positive (e.g., expectation, satisfaction, delight), neutral (e.g., disinterest, detachment) and negative (e.g., anxiety, dissatisfaction, disappointment).

View full abstract

Due to the product diversification and complication, sharing the product information between the product development members has been important in the product development process. Quality Function Deployment (QFD) is one of the effective methods that enables the development members to share the information of the product using the quality matrices that describes the relationship between design elements needed to be considered. This paper improves the quality matrices by introducing the multispace design model and the interpretive structural modeling (ISM) method. The multispace design model is a design theory to express the design activity as the inference using the design elements. Based on the concept of the design model, we changed the table items of the quality matrices. The ISM method is one of the design methods to visually express the complex relationship between design elements by using matrix operation. We introduced the ISM method to figure out the relationships between design elements in each deployment chart. The proposed quality matrices are applied to a disc brake design problem, and their applicability is confirmed.

View full abstract

This paper describes a value growth design methodology and an unsteady and non-equilibrium system design methodology, which is one of a typical means to embody a timeaxis design. The timeaxis design is a new design, which introduced time axis into the theory and methodology of design. The value growth design is a design, which appreciate in value with time progress. On the other hand, an unsteady and non-equilibrium system design is design, which focuses on transition period. Moreover, as a means for realizing of these designs, a bio-inspired technology and service technology are described. Finally, a new possibility by using the timeaxis design is indicated.

View full abstract

This paper addresses an emergent control system for a series hybrid electric vehicle (S-HEV) that automatically creates a control algorithm adapting to driving characteristics based on a timeaxis design. Because driving behaviors of a driver vary by driving state, and/or with time, it will be possible to realize more efficient S-HEV by switching a control algorithm to others dynamically corresponding to variations of driving behavior. An emergent control algorithm using a genetic network programming (GNP) method is proposed. It estimates driving characteristics of a driver and creates a control algorithm improving total drive cost of S-HEV, fuel consumption and battery shelf life, in adapting to driving characteristics. The algorithm was implemented in the control system of a driving simulator, and its function and effectiveness were verified. It was found that the algorithm could create networks of GNP in response to driving characteristics of each driver in each driving state, and that the total driving cost of S-HEV decreased by 11 percent compared to a driving-state adaptive control algorithm.

View full abstract

We developed an appropriate mesh size automatic decision technique that can be detecting the mesh size with sufficient analytical accuracy and short analytical computation time as possible. The mesh size is one of the most important factors affecting the analytical accuracy. However, the appropriate mesh size is not unique depending on a shape of analysis model and/or physical phenomena. In particular, beginners of analysis work take a lot of trial and error to evaluate the appropriate mesh size. So, we developed a mesh size detection tool that can be guaranteed reliability of analytical results. More specifically, relations between the mesh sizes and the analysis results is gathered by executing FEM analysis while decreasing mesh size, and appropriate mesh size is set at the time when the difference of analysis results approach asymptotically to 0%. This technique can be eliminated trial-and-error adjustment of the mesh size. We demonstrated to be reduced analysis work time to 1/50 or less by application to electromagnetic field analysis tool developed in-house.

View full abstract

Recently, the advancement of the laser range scanner has made it possible to provide dense and accurate range data of 3D objects. From these point clouds, one can reconstruct a CAD model of the existing old mechanical parts, which is called reverse engineering. However, if we fit the point cloud without any constraints on geometric characteristics, these features will be easily lost on the reconstructed surface. In this paper we introduce novel methods to impose geometric constraints on the B-spline approximants during the iterative geometric fitting process. The geometric constraints include reflectional symmetry of the surfaces of industrial products and feature curves on the aesthetic surfaces provided by designers.

View full abstract

We present a method for reconstructing complete 3D models from a number of images of an object taken from multiple viewpoints under the circular motions. We employ the shape from silhouette (SFS) method which constructs a voxel-based 3D model from the silhouettes of 2D images. In general, a single camera with an object's circular motion cannot capture the entire object with one orientation. Accordingly, we need to change the orientation of the object at least one time to capture the whole object. In this paper, we discuss a novel method to merge the two incomplete reconstructed voxel models into a single complete voxel model. Complex examples are provided to demonstrate the effectiveness of the proposed algorithms.

View full abstract

Request for power saving of data centers in Japan becomes higher due to the fact that almost all nuclear power plants have been shut down after the Tohoku Earthquake. Target of our research is to establish a control method to save electric power of total datacenter including facilities and ICT instruments. In our first paper on datacenter power saving, we have described an energy simulator which estimates the total electric power of datacenter and evaluates power saving controls of datacenter. In our second paper, we have presented an original real-time thermal fluid simulator and have showed that power consumption of the whole datacenter can be reduced up to 30% and more by workload consolidation and optimal cooling control. In this paper we show that 50% and more power saving is feasible by a coupling control of ICT and cooling facilities for an actual datacenter with 160 servers and an outdoor air cooling system.

View full abstract

This paper proposes an approach for computing CAD models of thin-plate industrial structures, from their volume data. As CAD systems use mid-surfaces for modeling thin-plate structures, a mid-surface computation from its CT image is useful in various reverse engineering applications. Conventionally, skeleton definitions are used for approximating mid-surfaces but it is not possible to obtain a precise representation of corresponding shape by using these skeletons. We propose a method which first extracts one such skeleton i.e. medial-surface and then transforms it into the desired mid-surface. Medial surfaces have some undesirable features like extraneous branches, non-linear junctions etc. Our method removes such flaws and obtains a precise mid-surface representation. The proposed method can generate mid-surface automatically and shows effective results on complex industrial models, even including non-manifolds.

View full abstract

This paper presents a method for computing non-uniform rectilinear grid structure, which is topologically equivalent to a uniform grid with non-uniform pitch size. Our idea is to coarse dense uniform grids created by voxelization methods by removing grid lines iteratively so that topological structure is preserved. In addition, we improve quality of the grids by comparing initial grids. This grid structure is efficient for FDTD (finite difference time domain) method, a popular method for electromagnetic field analysis. We also demonstrate that reasonable result can be computed by the computed grid, although computation time is reduced.

View full abstract

Uniaxial tension and stress reversal experiments have been carried out on 430SS stainless steel clad A1100 aluminum sheet in order to investigate its large strain elasto-plasticity behavior and cyclic stress-strain response including Bauschinger effect. Subsequently, parameter identification method for Yoshida-Uemori kinematic hardening model was proposed based on the inverse analysis with which the discrepancy of stress-strain curves between experiment and numerical simulation was minimized by numerical optimization technique. The proposed method consists of two stages, where parameters for 430SS were determined in the first stage which is followed by the parameter identification of A1100 sheet. In the second stage, parameters for A1100 were determined by using the stress-strain curves for clad sheet and rule-of-mixture. It was found that the calculated stress-strain curves incorporating the set of material parameters obtained by the proposed method well demonstrated the corresponding experimental results. Additionally, the importance of decent search region of material parameters on accuracy of parameter identification was discussed.

View full abstract

Welding the future, it produces deformation at the weld area. The more welding pass there increases, the more influence of welding deformation or bead by deposit metal there is in the future. It is known these influence are suppressed by welding order. The purpose of this work is to decide the optimal order to decrease welding deformation in the future. It is necessary to estimate welding deformation for influence of bead based on numerical analysis in three dimensional problem using FEM. In this study, the inherent distortion method is applied as the analysis method. This study aims for optimization of the welding order to decrease welding deformation in the future. It is necessary to have many analytical calculations. Therefore, to use the inherent distortion method and the combination based on the principle of superposition decrease analysis time and the number of analyses. In this paper, welding deformation for influence of bead has been analyzed. So, it is proposed to decide bead form of shell element that has stiffness similar to solid model. It is also considered what influence the bead form give to the future in the case of welding.

View full abstract

It is necessary to obtain the detailed stress distribution by evaluating a strength based on the FEM in the design phase because body and frame design rule of construction machinery has not been specified in the industrial standard. On the other hand, the structural analysis with shell elements and coarse mesh division for time shortening of design proposal construction is used widely. This research aims to establish the technique of acquiring a shell element model stress value equivalent to a solid element model. The stress value is calculated with a database that combines the simple analysis using the shell element and the detailed analysis using a solid element. The effectiveness of this technique is examined by applying a database to a damp truck model. This method can correspond to any stress state using the principle of linear superposition.

View full abstract

This study intends for some large structure which have no common structure rules on the initial design stage. Therefore, it is necessary to perform a detailed stress evaluation by FEM (finite element method) for each design proposal. However, the creation of a detailed model with fine element size in the initial design stage is difficult. In addition, there is a problem that it takes lot of time to create a design proposal. In short, because the results of FEM depend on the size of the element, it is necessary to evaluate the error in these analyses. Then, this study aims to evaluate the error and recover the stress for FEM at the initial design stage. In this study, the two methods of stress recovery for a model with regular element shape were proposed. First, it is the technique for obtaining stress using the element force of an adjacent element. When this technique was used for a model with large element size, it turned out that stress distribution is equivalent to a detailed model with fine element size except a joint of each structural component. Second, it is the technique for obtaining stress using the element force based on the quantity gap of the element force between adjacent elements. The influence from each structural component in a joint can be taken into consideration by using this technique. When this technique was used for a model with large element size, it turned out that stress at each structural component in a joint is equivalent to a detailed model.

View full abstract

A method is presented for analysis and design of mechanisms consisting of frames with partially rigid connections. Equilibrium equation is formulated with respect to the reduced vector of member-end forces, for which linear conditions are assigned. The modes of self-equilibrium forces and unstable mechanisms are found using singular value decomposition of the matrix consisting of equilibrium and constraints. A link mechanism with one-degree of kinematic indeterminacy can be found using limit analysis for an artificial proportional loading condition. The effectiveness of proposed method is demonstrated through examples of frames with a hexagonal grid.

View full abstract

Recently various measures are made towards realization of low carbon society. Water-power generation is one leading policy. However, development of the small-scale water-power generation to power consumption, such as a plain part, is not fully progressing. Therefore, this research aims to develop the micro water-power generation equipment which used the new type of turbine. For developing new type of turbine called the tornado type, its prototype model experiment and numerical simulation are performed and the various parameters of the turbine are determined and the manufacturing checks are performed. Then, we conducted the profile data of turbine generated by the 3D-CAD, production of the prototype model by using RP equipment, and the check experiments of operation by the produced the prototype model. Moreover, cutting of the actual turbine by using 5-axis processing machine with the tool path data generated by the 3D-CAM was also performed. As a result, we confirmed that the rapid prototyping model is possible to rotate in the water flow. In addition, in the actual cutting experiment of the turbine, we confirmed that it is possible to manufacture of the turbine by 5-axis processing machine.

View full abstract

When major train collisions occur, train structure deforms to reduce the impact effect. However, some passengers are injured in spite of the shock absorption of the structure. Therefore, it is necessary to design train structure to reduce injuries of passengers when accidents occur. In this study, we proposed the design improvements strategies to reduce injuries of the passengers caused by train collisions. We experimented with scale-down models of passengers and vehicles. To evaluate injury levels, we used HIC36 as a standard for head injuries. As a result of these experiments, HIC36 is influenced by seating position and type. We proposed three train structures to reduce injuries of passengers. These are that of using convertible seats, seating passengers with their backs to the front of the train and seating passengers in the rear of the vehicle.

View full abstract

For the stabilization of insulation performance in resin-molded insulators, strong adhesion between the resin and ceramic is required. In this paper, a new design technique for the control of the interfacial strength over multiple scales in resin-molded structures has been proposed. The adhesive strength of the internal interface in a resin-molded structure was estimated as the interfacial fracture energy by using the molecular dynamics method. The interfacial fracture energy was qualitatively in agreement with the adhesive strength index obtained by shear experiments. Based on the strength evaluations, the interfacial strength could be controlled, and it became clear from the results that the interfacial strength between resin and ceramic can be improved up to the strength level of the glaze adhesion.

View full abstract

Carbon supported Au core was synthesized with two-phase protocol using dodecanethiol as a stabilizer and Pt shell was formed on the core surface by immersing the core into K_2PtCl_4 aq. at 303 K. Durability of the Au core/Pt shell structured catalyst was evaluated with rectangular potential cycling in 0.1 M HClO_4 aq. at 333 K. In the early stage of the durability test, electrochemical surface area of the catalyst decreased by 40 % of the initial surface area, which was mainly due to solid solution formation of the Pt shell with the Au core. As-synthesized core/shell catalyst was annealed in air to form the solid solution in advance. The decrease of the electrochemical surface area of the annealed core/shell catalyst was suppressed in the early stage of the durability test.

View full abstract

In the field of electrical equipments such as motor, unsaturated polyester resins and vinyl ester resins are widely used for adhesive and insulator. These resins are generally cured by radical polymerization and have several advantages such as low cost and good insulation and adhesion. Recently, downsizing of electrical equipments have resulted in them generating more heat, and thus resins need to have better thrmostability. In the case of thermosetting resins cured by radical polymerization, thermal decomposition occurre from terminal vinyl groups derived from disproportionation reaction. It is known that the living radical polymerization inhibits the disproportionation reaction. Thus, we attempt to use living radical polymerization to improve thermalstability of resins. A mixture of styrene and vinyl ester, which has a weight ratio of 50:50, was used as a varnish. The thermal decomposition temperature of resins cured by various radical initiators was measured by using thermal gravimetric analysis. The thermal decomposition temperature of resin cured by the borane compound, a living radical initiator, was higher than that cured by non-living radical initiators such as peroxide. Furthermore, the glass-transition temperature of the resin cured by the borane compound was higher than that cured by peroxide. From these results it was revealed that living radical initiator is a very effective reagent to improve thermostability of thermoset resins.

View full abstract

Hydrogen fuel is a zero-emission fuel which uses combustion in internal engines to power vehicles and electrical devises. In electrolysis, electricity is run through water to separate the hydrogen and oxygen atoms. This method can be used by using wind, solar, geothermal, and many other resources. Hydrogen is an energy carrier. Hydrogen embitterment is the process by which various metals, most importantly high-strength steel, become brittle and fracture following exposure to hydrogen. The mechanism starts with lone hydrogen atoms diffusing through the metal. A hydrogen vehicle uses aluminum alloys, which shows high hydrogen embitterment characteristics. This paper makes an experiment of HEAC (Hydrogen Environment Assisted Cracking) characteristics using aluminum alloy. The aluminum alloys is 7075-T6. The fracture surface on repeated load and static load by DNG and hydrogen embitterment environment shows the facilitating mechanism. The surface features indicated that the hydrogen accelerated granular surface fracture.

View full abstract

To explain the mechanism of friction for a rubber block sliding on a road surface with multi-scale asperities, we estimate the friction coefficient for rubber based on Persson's mathematical model. Persson has stated that the major contribution to rubber friction, such as tire/road interaction, is from the hysteresis term for rubber deformation and from the multi-scale roughness of the road surface. Applying Persson's model, we propose a new methodology which determines the friction coefficient from the standpoint of the rubber elastic limit. Comparing the calculated friction coefficient with the friction coefficient measured in the friction testing, we find that there are some differences between the experiment and calculation in the range of measured sliding velocities. We discuss these differences and show that it is necessary to modify this friction model to consider not only the hysteresis term but also adhesion and local temperature terms due to the interaction between the rubber and the road surface.

View full abstract