This paper describes a method which treats compound features by extending the feature modeling technique for primitive features. The compound features, that are user-defined features, are represented using symbolical description for topological and geometrical constraints similarly as that for the product model with Face-based representation. The compound features can be defined, displayed on the graphics, stored in or retrieved from the data base, attached to a product model and modified by changing parameters. Some techniques are demonstrated such as representing the feature model symbolically, preventing duplicated registration of the same type of feature, and retrieving the feature type according to the users' specification.

抄録全体を表示

It is a difficult task to select the process variables and determine the optimum design using trial-and-error procedure in the metals industry. In this paper, an attempt has been made to combine finite element analyses (FEA) and artificial neural network (ANN) to study the effect of process variables on material flow behavior and required forming load in cold forging. The process variables are used as the inputs and the finite element results as the target outputs, a neural network model was established. In order to achieve a proper neural network model which can generalize the training data obtained from FE data well, some crucial training parameters, such as learning algorithms, hidden neurons and hidden layers, number of training data and error goal, have been investigated. Finally, the optimum neural network model was used to predict random unseen data and thus find the optimum process variables according to a certain decision rule.

抄録全体を表示

This paper deals with bi-criteria job shop scheduling for meeting due-dates and minimizing setup times. An efficient search method using the mixture of a genetic algorithm and a priority rule is proposed. In the method, the individuals of the genetic algorithm are ranked using a lexicographic ordering approach. A parameter of the priority rule to be mixed with the genetic algorithm is set to have high performance for minimizing tardiness until the schedule having no tardy job is found. A lot-splitting method is applied to meet due dates if necessary. After the schedule having no tardy jobs is found, the parameter is gradually shifted to have high performance for minimizing setup times. By appropriately tuning the shifting speed of the parameter, the proposed method can efficiently minimize setup times. Numerical experiments show the effectiveness of the proposed method.

抄録全体を表示

This paper presents an Innovative Product Development Process (IPDP) for a product whose function is to convert the condition of a certain object from an input state into a specified output one. This product innovation approach consists of two steps; the first step is the problem setting stage through the newly developed processing point analysis, and the second one is the problem solving stage through the restructured 40 principles of innovation. The processing point analysis first captures the gap between the ideal and the reality of the output condition of the object. Then, it traces each cause of the gap along the causal relationships back to an action which should be given to the processing point of the object from the product, and finally formulates it into a technical problem of the deficiency, excessiveness or harmfulness of the action. Further, the 40 principles of innovation are rearranged from three viewpoints (preventive means against harmful factors, directions of technical evolution, and options of design changes) so that each technical problem identified by the processing point analysis can be dealt with properly. The proposed approach is described using a rice cooker as an illustrative example, and it is shown that some innovative conceptual design solutions are successfully derived through the standardized product innovation approach.

抄録全体を表示

This paper presents a methodology to calibrate kinematic parameters in a Hexapod-type parallel mechanism machine tool. In the conventional calibration process, the influence of the calibration error and that of the deformation of struts due to the gravity cannot be explicitly distinguished on contouring error trajectories. As a result, the gravity-induced contouring error becomes one of major causes for calibration error of kinematic parameters. This paper presents a kinematic calibration methodology that takes gravity-induced errors into explicit consideration. By performing the calibration of kinematic parameters and the modification of the prediction model of gravity-induced errors in an iterative manner, the machine's contouring accuracy can be improved. Experimental comparison shows the machine's contouring accuracy is significantly improved by applying the proposed calibration method.

抄録全体を表示

Strain gauges are frequently used for inspecting the strength and reliability of actual structures. However, the measurement of actual structures by strain gauges needs dangerous works such as pasting strain gauges and cabling them to strain meters located in a remote place. Therefore there is a growing demand for developing non-contact and accurate deformation and stress measuring technique of structures.
The authors developed a non-contact and accurate deformation and stress measuring technique using the sub-pixel digital image processing method. This paper presents the detail of the measuring technique and the method of analyzing the measured data, and the results of performance test of the measurement system. The results of the numerical computation and the experiments demonstrated the effectiveness of our deformation and stress measuring technique using the sub-pixel digital image processing method.

抄録全体を表示

Surface roughness and specular glossiness are generally used as evaluation indexes of processing surfaces. In evaluating a metal surface, it is important to know how external fineness of the processing surface and an ideal specular surface are different. Here, this difference is called a “specular rate”. It is impossible to evaluate specular rate using surface roughness or specular glossiness. Consequently, this paper considers new evaluation indexes for quantitatively evaluating specular rate of a metal surface. First, the method for estimating the impulse response between a certain pattern image and it's reflected image on an evaluated surface is proposed. This impulse response indicates the optical sharpness of a reflected image. Then, the surface's specular rate and directionality indexes are defined by estimated impulse response. The validity of these indexes has been verified by surface evaluation experiments on processed SUS304 stainless steel samples.

抄録全体を表示

This paper describes a technique, based on the shape-from-focus method, to measure the shapes and heights of an array of solder ball bumps through visually inspecting an IC chip-size package (CSP). We devised a focus-measure difference method for detecting the horizontal section of a bump quickly and precisely. This was where we selected bump image pixels that had the same focus measure at two specified focal points along the height axis. The selected pixels corresponded to the contour points of the section at the mid-height between the two focal points. We applied it to measuring the height of an array of 208 bumps that were 260 μm high arranged at intervals of 500 μm. By measuring the area of the horizontal section at a specified height and using it to fit a sphere in the bump, the height of the bump top was efficiently estimated to match the sphere top. This sphere fitting achieved a height measurement accuracy of 8 μm (2σ) at a resolution of 3.4 μm pixels on the bump using a 2/3-inch high resolution CCD camera with 1300 x 1030 pixels. The accuracy was sufficient to inspect the coplanarity of an array of solder ball bumps.

抄録全体を表示

This paper deals with a one-way brake to generate a large braking force using a piezoelectric actuator. In the proposed brake unit, a break plate is pressed on a slider with a small inclination angle perpendicular to the slider. In the case of thrusting the slider, a large braking force in proportion to the thrust is generated between the slider and the brake plate. The braking force does not depend on the generative force of the piezoelectric actuator, which releases the brake plate from the slider. It does not consume the energy to keep the slider position because the piezoelectric actuator is a capacitive load. A prototype was designed based on the preliminary experiments of the inclination angle of the brake plate. The measured inclination angle for braking the slider coincided with the theoretical one by statics and the finite element analysis. The braking force control was demonstrated. After a dither was added to the control signal of the piezoelectric actuator, the braking force became stable.

抄録全体を表示

This paper describes a new type of surface encoder for planar position measurement, which consists of a sinusoidal grid (target sine-grid) and an optical sensor unit. The optical sensor employs a transparent sine-grid instead of a grid pattern film used in a conventional surface encoder. The transparent sine-grid is generated on a glass plate with the same amplitude and wavelength as the target sine-grid. Simulation results have shown that planar displacement between the two sine-grids can be obtained from the output of the sensor unit. Compared with conventional surface encoder, the new type surface encoder is expected to have higher S/N ratio. Experimental results are also presented.

抄録全体を表示

Since organic contaminants, which adhere on the silicon substrate surface, have an adverse impact on the quality of the gate oxide films, it is very highly required for the current semiconductor fabrication processes to take them off the surface before producing the films. Therefore, some kinds of cleaning method for silicon substrate, such as SPM (Sulfuric acid-hydrogen Peroxide Mixture) cleaning, ozone water cleaning and ultraviolet exposure cleaning, have applied to remove organic contaminants from the substrate surface. But these cleaning methods exert a bad influence of the global environment and spend high cost for the semiconductor fabrication. On the contrary, we have developed the new cleaning method for the silicon substrate using the high-speed shear flow of only ultrapure water and composed the new cleaning system. In the result of cleaning the silicon substrata surface contaminated by dioctyl phthalate at 1×10¹⁵atoms/cm², we demonstrated that DOP(dioctyl phthalate) concentration on the substrate reduced at the level of 10¹³atoms/cm² using our cleaning system. Moreover, we showed that our cleaning method possessed oxidizing property to Si substrate surface and supposed that DOP removal mechanism from Si substrate surface and this oxidizing property were related.

抄録全体を表示

In this paper, a refrigerating chuck system is developed to reduce the thermal damages of workpiece in laser cleaving process. The system reduces the surface temperature of table below the freezing point of water, and the work material is fixed on the table by the frozen water between the material and the table. The cooling capability of the system is investigated. The system is applied to the laser cleaving process of silicon wafer with cw-Nd:YAG laser.
The laser cleaving experiment of silicon wafer is conducted with Nd:YAG laser, then the width of thermal affected zone on the irradiated surface, the roughness of fractured surface and the deviation of cleaving trajectory are measured. The silicon oxide is caused on the surface of wafer in the room temperature, but the refrigerating chuck can prevent the thermal damage and improve the reliability of the cleaving process. By use of the chuck, the smooth fractured surface is achieved and the linearity of the cleaving trajectory is also improved.

抄録全体を表示

Milling with small-diameter end mill is sensitive process, since the end mill is weak and flexible. This paper proposes a monitoring technique for the in-process measurement of cutting force and the prediction of tool breakage in the high speed slotting with the small-diameter end mill, which is based on the newly developed visualization system with high-speed CCD camera, image acquisition and image processing. In order to remove the effect of cutting chip and fluctuation of tool rotation on the projection image of tool, the continuous 200 frames of image was used for the image processing of tool shape. As a result, this measuring system enables the precise measurement of tool deflection in the high speed milling process.
It was shown from the comparison between monitoring system and experiment that the tool deflection had good correlation with the cutting force, in which the static compliance and the geometrical moment of inertia were introduced in the prediction of the cutting force. The cutting force predicted by the static compliance was well agreed with the one measured by the dynamometer at the high speed milling of pre-hardened steel and brass in the various cutting conditions. It was also shown that the breakage force was estimated by the principal stress due to the bending moment and was nearly equal to the strength of tool. Since this system is able to measure the tool deflection and to predict the cutting force in process, it is useful for the monitoring of the cutting state in high speed milling with the small-diameter tool.

抄録全体を表示

In order to realize fine processing with a high precision optical system, we have developed a laser processing system with galvanometer scanners and optical lenses using a shrink fitter. The shrink fitter prevents the misalignment of the lenses at the elevated temperature. Therefore fine laser spots can be obtained over the quite wider scanning width, even at elevated temperatures. We examined the possibility of application to laser processing of this lens system using the shrink fitter.
Diameters of the lenses used in this study were 160 mm so that the laser spots could be scanned over the width of 100 mm, in both directions of the X-axis and Y-axis. There was large distortion of the laser beam which was inherent to each lenses and can not be removed by mechanical adjustment So we introduced the formula to correct the beam path using the least square method. Finally, we developed the fine laser processing system using shrink fitter.

抄録全体を表示

The existing method of calculation for the basic dynamic load rating mentioned in the ISO standard is based on the method used for the angular contact ball bearing and does not align with the one used for the linear bearing. The unit for the rating life, i.e., the Unit Running Life (URL) is defined in terms of 10⁶ rev. As a result of this, the effect of linear stroke of the nut or screw shaft, lead, the number of starts, and the number of circulating ball circuits and their turns are not clearly reflected in the load rating formula. In this paper, the basic dynamic load rating and the life formula for the ball screw based on the method used for the linear bearing, i.e., in terms of an unit URL = 100 km is proposed and some representative calculation results for a ball screw are discussed. The results are summarized as follows: (1) Basic dynamic load rating C_a for standard stroke corresponding to the nut raceway length is defined, and a stroke factor is also defined for any stroke of the ball screw operation. (2) The C_a value for the ball screw is proportioned according to the number of ball circuits and their turns to the power of 0.7 as well as the number of starts. But, as the number of starts especially affects the nut raceway length and the number of balls, the C_a value becomes remarkably large.

抄録全体を表示