With the present interest in energy saving, the improvement of gear efficiency is being looked at again. In this paper, differences between Buckingham's and Merritt's gear efficiency equations are described. These equations are of the same calculation form, but the manner in which they are used differs. The equations are valid under the assumption that input energy is calculated as the product of the normal force and displacement along the line of action. The author clarifies that an incorrect manner of calculating input power in the meshing gear may lead to trouble, and that Buckingham's and Merritt's equations are approximations. Moreover, in this paper a problem derived from the treatment of pitch point force is described. From the results of this study, it is concluded that the use of a tangential force of the pitch circle leads to an incorrect theory. The author deals with the locking phenomenon of the planetary gear mechanism as an example.
The performance of a tennis racket in terms of the coefficient of restitution(COR)is closely related to impact phenomena. This paper investigates the effects of frame vibrations on the coefficient of restitution and the contact time during impact of a ball / string system and a simulated frame model, using FEM simulation and modal analysis. The results show that the COR is mainly affected by a rigid motion and a bending vibration with two nodes of racket frame. In addition, the COR increases with an increase of frame rigidity, but then saturates at a certain rigidity depending on the impact velocity. Furthermore, the COR increases as the impact point approaches the center of rotation and the node of racket frame vibration.
In this paper, we propose an optimum design method of humps to reduce the excessive shocks experienced by drivers of heavy vehicles passing over existing circular arc humps with a length of 3.67 m at normal speeds, and to compensate for too few shocks experienced by drivers when light vehicles with semi-active suspensions pass over the same humps at excessive speeds. We take the standpoint of hump design and suspension design alternately, and determine the optimum dimensions of a circular arc hump by optimizing it and the damping coefficient of light vehicle suspension dampers alternately. The maximum acceleration of the driver of the heavy vehicle and that of the light vehicle is set as a multiobjective function for the optimization of the hump, while the maximum acceleration of the driver of the light vehicle is selected as a single objective function for the optimization of the suspension of it. It is found that the optimum hump determined in this study further reduces the excessive shock to drivers of heavy vehicles compared to the existing hump and to the optimum hump obtained in the previous study, in which heavy vehicles were not considered. For the light vehicles, a decrease in the amount of control of vehicle speed was found previously, but the present method is found to overcome this weak point. Speed control effects of the existing hump and the one optimized by the present method on three types of light vehicles with different wheelbases which have two types of conventional passive suspensions are compared using simulation calculations. The optimized hump exhibited less deviation among the above different vehicle types in terms of the speed control effects.
A simulation model for biomechanical analysis was developed in order to calculate the internal loads, such as muscular tension and energy consumption, from measured kinesiological data for various kinds of motion. In this model, the human body is divided into a maximum of nineteen rigid segments connected to each other with ball joints. A maximum of 156 skeletal muscles throughout the body are modeled as tensile force generators which include physiological elements that consume mechanical energy. The component ratio among muscle fiber types, number of body segments and number of degrees of jointfreedom can be changed easily, and according to the alteration of the segment number, the musculoskeletal system is changed automatically. This flexibility is of benefit in application of the method to various motion analyses. Three-dimensional bipedal walking was analyzed using a simplified model, and rowing motions were analyzed using the whole-body model.
A new type of acoustic emission(AE)sensor was developed for detecting head-disk interaction in magnetic disk devices. In order to develop a small sensor, we used a thin-film piezoelectric material and mounted it on the Si slider using micromachining techniques. We conducted a basic experiment and a simulation of sensor output, and confirmed that the simulation used here can predict sensor output characteristics. This paper describes the structure of the sensor and the method of simulation. We also performed several other simulations of sensor output characteristics using this method, and confirmed that the new sensor is applicable to tribological research on head-disk interaction.
In this paper we describe a method of designing a radial-basis-function-network-based(RBFN)controller. The RBFN controller is derived from a neural-network-based(NN)controller using differential operator focusing on the sigmoid function derivative of the NN controller. To overcome the Jacobian problem, the RBFN controller uses a learning algorithm and a neural identifier which uses an adaptive algorithm to estimate the plant Jacobian. A conventional feedback controller is incorporated into the RBFN controller to ensure both robustness and stability at the beginning of the learning process. Simulation results for mathematical plants demonstrate the applicability of the RBFN controller for controlling nonlinear systems and experimental results for 1-degree-of-freedom robots demonstrate its usefulness for controlling practical systems. Application to controlling an ODD positioner and a tunneling machine demonstrates the effectiveness of the RBFN controller for controlling mechanical systems.
In this paper, a neural-network-based adaptive control is presented to solve the output tracking problem for a class of nonlinear continuous-time feedback linearizable MIMO systems with unknown nonlinearities. The adaptive control adopted in this paper ingeniously combines the conventional sliding control technique and the approximation scheme of the radial basis function(RBF)neural networks to perform approximating input-output linearization. The sliding control is used to compensate the network approximation errors and the neural network parameters are updated according to the Lyapunov principle. It is shown that the outputs of the closed-loop system asymptotically track the desired output trajectories while maintaining the boundedness of all signals within the system. The effectiveness and robustness of the proposed control scheme are demonstrated in the case of two-DOF robotic manipulator.
The problem of disturbance attenuation and covariance assignment for an interconnected system using decentralized controls is addressed. We solve the H∞ and variance constraints using the Riccati equation approach. To achieve the specified upper boundaries of disturbance attenuation and state covariance, a condition which ensures the existence of local controllers is derived. The entire set of state covariances which may be assigned to each subsystem by local state feedback is characterized and all corresponding local state feedback controllers are derived. Complete solutions to the decentralized controller design of interconnected systems with H∞ and variance constraints are achieved merely by iteratively solving a set of local algebraic Riccati equations.
Proportional and derivative (PD)control is a simple and effective control technique for a wide class of dynamic systems. System uncertainties due to nonlinearity distortions, parameter variations, or external disturbances can be suppressed by increasing the feedback gains. In practice, however, there are limits to the PD gains without causing instability. The instability is usually attributed to measurement noise or high-frequency unmodeled dynamics excited by the high feedback gains. In this paper we show that the sample-and-hold effect in a digital control system may cause instability in an otherwise stable PD control system. With proper approximations, explicit and practical stability criteria are derived relating the magnitudes of the feedback gains to the sampling period and other system parameters. Basides the sampling period, the stability of the closed-loop system is also significantly affected by the computation time, which is often neglected in theoretical analyses or numerical simulations. Case studies are presented to illustrate the use of the stability criteria in the design of high-gain PD control systems.
A new vibration control mechanism and control method for suppressing horizontal vibration in super-high-speed elevator cars are considered here. The actuator consists of an AC servomotor and a ball screw, and the vibration of the cabin, detected by an acceleration sensor. The acceleration signal is integrated to determine the velocity and the position. The control force is determined based on the above position and velocity, and on the velocity of a motor encoder taking into account the car dynamics. First, the dynamics of an elevator car with the control are formulated, and the effect of the control on vibration suppression is evaluated by numerical simulation. The results indicate good performance of the vibration control system, which suppresses 1st- and 2nd-mode vibrations of the car. Furthermore, the control force and actuator stroke are within the practical range of designed values. Second, a vibration test equipment system using a full-size car is developed in order to test the performance of the vibration control system. Finally, using this system, it is experimentally confirmed that the vibration control mechanism reduces vibration to half that of a control-free car.
High performance robot control sometimes requires direct velocity measurement at the end effector as well as at the joint. However, to date, sensing devices are still not readily available for tip velocity measurement on most industiral manipulators and one often has to rely on approximations to derive this velocity. The limitations of these approximations are analyzed in this paper. In order to overcome these limitations and derive the robot tip velocity, end point acceleration measurement is imployed and a complementary filtering technique is adopted. The filter design approach is described with some implementation results given.
In an articulated robot, coupling forces are generated between the joints. To improve dynamic behavior and reduce the work cycle time, it is necessary to compensate for the force generated between the joints. In this paper, robot dynamics are expressed in a linear form by applying a gain scheduling method and then transformed into a discrete form. A simple control algorithm is proposed which can perform the decoupling and the pole assignment simultaneously by using an observer with a disturbance compensation function. The results, based on experiments with a SCARA-type robot, show that the proposed approach can compensate for the coupling force between joints.
In this paper we describe a new design of a fluid-driven microactuator which comprises a planar piston, a four-way control valve, and a mechanical feedback system that controls the output in accordance with an input displacement. Merits of this design are expected in terms of output density, scalability, and feedback ability realized with no electronic devices. Fabrication processes of the microactuator include bulk micromachining of two sheets of silicon substrate and their assembly by electrostatic bonding to glass plates. The output characteristics were measured while compressed air was supplied. Generated output force showed an almost linear relationship to supplied pressure and to valve displacement. However, the magnitude was asymmetric in terms of driven directions, which resulted from geometrical configuration essentially introduced by an anisotropic etching of the silicon substrate. The position of the output shaft was sufficiently controlled with a designed magnification factor multiplied to the input displacement in both static and dynamic senses.
Sliding mode controllers are known to be capable of achieving robustness against modeling errors, plant disturbances and plant parameter variations. They, however, require upper-bounds of uncertainties in plant parameters to specify feedback gains that realize robustness. These upper-bounds are often chosen conservatively and, as a result, high feedback gains and oscillations, refered to as chattering, occur. Recently, on-line identification of the upper-bounds of uncertainties have been successfully incorporated into sliding mode regulators to alleviate the chattering. Here, we propose to extend this on-line identification technique to two types of controllers, called adaptive robust sliding mode controllers, for trajectory control of robot manipulators. The stability properties of the proposed controllers are demonstrated using Lyapunov functions and are implemented using a PUMA-type manipulator. The performance of these controllers is compared with that of regular sliding mode controllers. The result of comparison reveals excellent robustness and chattering suppression of the proposed controllers.
Algorithms of a non-numerical and non-heuristic nature are developed and implemented in this work for the analysis of motion planning. In particular, the most complete analysis of the motion planning problem is performed for a 2-link planar mechanism and the problem is formulated in purely geometric terms. We describe an algorithm which determines the shortest path between any two points in the joint space. Moreover, the algorithm can be used to choose a path such that the variation of the joint angles along the path guarantees the fastest transference of an arm end-effector between two generic points, i.e., from A to B, in the workspace. Finally, the problem of planning the shortest path for a 2-link arm is considered in an environment containing obstacles in the form of points which the manipulator cannot cross.
The design of variable pitch lead screws with swinging and translating meshing rollers is proposed in this study. Mathematical expressions representing the surface geometry of the screw generated by an endmill cutter are derived for the purpose of calculating the meshing points on the threaded surface of the screw. This mechanism is modeled using a Silicon Graphics workstation for motion simulation. Two examples are given. Our results show that constant thickness of screw threads can be attained using the design proposed. Finally, the machining settings of a 5-axis CNC machine are provided for manufacture of the screws.
The vibration transmission in an experimental gearbox and the structure-borne noise are measured for various positions of bearings in the gearbox. The bearing offset from the center of the housing wall considerably influences the vibration and radiated sound power. The radiated sound power increases with the offset. The sound power is well evaluated by the proposed method of simulation which integrates the FE vibration analysis and BEM-based acoustic analysis. The simulation suggests that the design of the gearbox is the key, and the vibration mode, which reduces the stiffness at the bearing positions, should be isolated for the reduction of sound power.
The curvature properties of ruled surfaces are derived. Several parameters are used to express the geometry of ruled surface explicitly. Using the ruled surfaces as generating surfaces, the equation of meshing is derived for a prescribed input-output relation. Considering the motion types of mechanisms and the particular ruled surfaces, the meshing equation is simplified. The pressure angle of the cam is analyzed as well as that of the follower, and the relation between these pressure angles is obtained using the equation of meshing. Two examples are given.
The primary purpose of the present paper is to investigate the behavior of bubbles all over a porous friction facing in a wet clutch, since this behavior might affect the performance of the wet clutch significantly. There in still much to be learned about the extent and location of occurrence of cavitation phenomenon in actual coupling systems, and it is one of the most important unknowns in the study of the wet clutch engagement mechanism. The observed features of bubbles arising from a oil film were recorded, and provided considerable insight into the possibility of occurrence of cavitation. Moreover, the relationship between the cavitation phenomena and frictional torque was elucidated and is discussed. The effect of friction paper porosity and applied load on the cavitation development were also experimentally studied.
The homology design concept was devised for application to the large telescope structure by S.V.Hoerner. He defined the deformation of a structure as homologous if a given geometrical relation holds for a given number of structural points before, during, and after the deformation. Some researchers have utilized the design theory on structural design with the finite element method. In the present study, a simple method using geometrical equality equations is suggested for performing the homology design. The previous methods are improved for the cost of making simultaneous equations. The homology design technique is expanded to designs for user-specified responses. The basic formulations of the homology design with the optimization concept are described and several practical examples are solved to verify usefulness and validity.
In this study, a high precision V-bending process control system incorporating an database and modified fuzzy model was developed. Information including the punch force-stroke curve measured in the bending process was stored in the database and applied directly to the process control. In the control system, a new evaluation method, which is called the vectorial evaluation method, was proposed to check irregularities of information in the experimental database, and two modified fuzzy models were developed to compensate defects in the experimental database. Simulation of the control system using the experimental results verified that the proposed control system significantly improves the accuracy and reliability of the process.
The machining center is currently positioned as one of the core facilities in a wide range of production systems. With its increasing importance in the manufacturing environment it has become a consequential requirement for producers of machining centers to improve their market competitiveness. This implies a need for establishing a new methodology for market competitiveness analysis based specifically on the conversion of uncertain market requirements into the technological attributes rather than on statistical data. This paper proposes a methodology for systematically and rationally analyzing the market competitiveness of a machining center using multidimensional vectors:in order to quantify the competitiveness factors, user needs and performance specification of the machining center are represented by the vectors in multidimensional coordinates. The validity of the proposed method has been verified through case studies.
A conventional 4-Hi cold rolling mill configuration was modelled and a computer program was developed to analyze the effects of inlet thickness, width, desired exit thickness, symmetric and asymmetric jack forces, screw positions, tensions on the strip, roll geometries, and material properties on strip shape and thickness. For the analysis, the finite difference method was used as the discretization technique. The roll force was obtained by integrating the calculated roll pressure over the arc of contact, and elastic distortion of rolls due to the strip interaction was also considered. For determining inter-roll force, the compatibility equation between work and backup rolls was expressed in terms of work and backup roll axis deflection and work roll surface deformation. The second compatibility equation between work roll and strip was expressed in terms of work roll axis deflections, work roll surface deformation, and estimated strip thickness. The calculated thickness distribution was used for determining shape.