Abstract: This paper proposes a constitution method for an adaptive PID control system that follows a non-stationary system. Because a PID controller has various practical benefits that are easy to implement, unnecessary of controlled model and highly robust, it is the most common control system in industrial world even today. However, its main drawback is that tuning is time consuming because each parameter is determined empirically based on trial-and-error, which is especially noticeable in a multi-input multi-output (MIMO) system composed of multiple PID controllers with interference between control input and controlled output. Other methods have been proposed, including the Ziegler-Nichols ultimate sensitivity method, but it cannot deal with a MIMO system. Additionally, methods using optimization exist, but they cannot provide online tuning for non-stationary systems during operations due to the numerous tuning parameters and repeated computations. In this study, we introduce a computationally efficient optimization method called the Simultaneous Perturbation Stochastic Approximation (SPSA) and investigate its performances when applied to a PID control system. We also propose an online parameter tuning method for the controller by improving the standard SPSA algorithm. The efficiency of proposed method is demonstrated by applying it to a MIMO system, which has some interference. Nomenclature SPSA Simultaneous Perturbation Stochastic Approximation. FDSA Finite Differential Stochastic Approximation. p-dimensional optimization parameters vector. ∗ Optimal parameters vector. Approximation of optimization vector at k h iteration. L, L(∙) Loss function. g( ) Gradient of loss function L to , i.e. ∂L ∂ ⁄ g ( ) Estimated value of gradient g( ). Update gain of optimization parameters vector. Perturbation gain of optimization parameters vector. ith perturbation index vector added to optimization vector on FDSA. ∆ Random perturbation vector added to optimization vector on SPSA.

Abstract: Production planning is basically concerned with determining the production and inventory level over a finite planning horizon to satisfy customer demand in the most efficient way. Such a decision has long been an important issue in manufacturing since it leads higher utilization of resources and customer satisfaction, and eventually, more profit for a company. So far, several mathematical optimization models have been proposed, including linear programming, mixed integer programming, and goal programming (Nam and Logendran, 1992; Pochet and Wolsey, 2006). When researchers and practitioners develop mathematical models, dealing with uncertainty is recognized as a critical concern, since models without consideration of uncertainty can be expected to perform worse than the models accounting for the uncertainty (Mula et al., 2006). Ho (1989) classified several types of uncertainty affecting production processes into two groups: environmental uncertainty beyond the production processes, and system uncertainty within the production processes. Similarly, Ravindran and Warsing (2012) divided the sources of uncertainty into internal and external sources. Examples of system uncertainty are production lead time, yield, and quality. Environmental uncertainty includes demand uncertainty and supply uncertainty. Typically, environmental uncertainty is less manageable than system uncertainty, since it arises outside of a manufacturing system. Therefore, determining a robust production planning and inventory management policy that is less sensitive to environmental uncertainty may be one of the most effective ways to mitigate loss for a company. There have been several approaches to optimization under uncertainty, such as stochastic optimization (Birge and Louveaux, 1997; Shimizu et al, 2011) and robust optimization (RO) (Ben-Tal et al., 2009; Seo and Chung, 2014). In traditional stochastic optimization, it is assumed that an uncertain parameter follows a certain probability distribution, and an optimal solution minimizes the average objective value. In contrast, discrete scenarios or a continuous range is assumed in RO to find an optimal solution that minimizes the worst-case objective value. The use of discrete scenarios Kwang-Kyu SEO*, Jun KIM** and Byung Do CHUNG** *Department of Management Engineering, Sangmyung University 31, Sangmyungdea-gil, Dongnam-gu, Cheonan, Chungnam 330-720, Republic of Korea **Department of Industrial Engineering, Sungkyunkwan University, 2066 Seobu-ro, Jangan-gu, Suwon 440-746, Republic of Korea E-mail: bd.chung@skku.edu

Abstract: We studied the effectiveness of diamond nanoparticles (DNPs) dispersed in water as a lubricant additive between stainless steel plates and sintered tungsten carbide (WC) balls. DNP dispersions with concentrations of 0.01, 0.1 and 1 wt.% were prepared and used as lubricants under a load of 1.88 N, for 240,000 friction cycles. High-friction coefficients of more than 0.3 were observed in an initial period. Then friction coefficients declined and stabilised at values of approximately 0.1. The steady-state friction coefficients were independent of the DNP concentration and lower than that for distilled water. In the initial period, wear of both the plates and ball was obvious. In the steady-state period, additional wear on the plates was a little; however, ball wear scars were clearly observed. The size of the ball wear scars decreased with decreasing the DNP concentration. It is likely that DNPs were embedded mainly in the stainless steel plates, and the embedded DNPs protected the plates and wore the balls in the steady-state period. Compared with the lubrication under distilled water, the friction coefficient and wear of the plate under the lubrication by the 0.01 wt.% DNP dispersion were lower, and the wear of the ball by this lubrication condition was not higher.

Abstract: The invention discloses an efficient improved FSPSO (Particle Swarm Optimization based on prey behavior of Fish Schooling). According to the efficient improved FSPSO, an intelligent behavior is simulated, and current globally-optimal particles search for current globally-superior positions through own optimal position information provided by a minority of other random particles. When fish schooling is attacked by other predators, weak fish which cannot escape quickly is eaten. The behaviors are simulated, weak particles close to current globally-worst particles are replaced with particles which are generated randomly, so that the diversity of the schooling is improved, and a local optimum can be effectively avoided by the FSPSO. The efficient improved FSPSO can be particularly applied to the solving process of complicated optimization problems such as function optimization and knapsack problems, traveling salesman problems, assembly line work problems and graph and image processing problems.

Abstract: Scissor mechanism, a basic element of structure and mechanism, can be used to make lots of deployable structures and functional motion devices. In such structures or devices, input configuration and parameter settings of actuator in scissor mechanism are variable and very important to structural forces and kinematics, however, has not been investigated systematically. Former researches on input vectors only consider a mechanism with decided input configuration and parameter settings. To supply a universal model for design purpose, a CHAracteristic Triangle (CHAT) is herein reported to calculate input vectors. The CHAT can be picked up from each of six input configurations of single group scissor lift mechanism. Based on former works and authors’ simulations, several value rules of input vectors in scissor mechanism are discovered, summarized, and approved via mathematics analysis. One of those rules shows that the maximum input force always occurs in the beginning position of lifting. Which indicates extreme importance of initial lift angle to design. Meanwhile, a further research on typology refines six input configurations into more detailed value cases according to possible parameter settings. Among them, three constant lift force cases, two singular cases and other cases are discussed individually. Typology and value rules are significant to power selection, mechanism configuration design and dimension optimization. Validity of presented equations is approved by four reported samples. Not limit to studied single group mechanisms, CHAT can also be found easily in every linear driven scissor lift mechanism such as multiple group mechanism, parallel mechanism, or nonstandard connected single group mechanism. Thus, presented universal equations are suitable for widespread applications with scissor mechanism.

Abstract: The Pre-stressed Hardening Grinding (PSHG) is presented combining with the advantages of Pre-stressed Grinding and Grinding Hardening. In order to study the variation mechanism of metallographic structure and the residual stress within hardening layer under PSHG, the paper took 45 steel as the experimental object and carried on the PSHG experiment. The surface hardening layer with little residual stress was obtained from it. The surface hardness, the wear loss and residual stress were measured and its metallographic structure was observed. Combined with experimental result and the simulation of grinding temperature field, the paper studied the mechanism which the pre-stress influences the metallographic structure and the residual stress. The study shows that parent-phase-hardening and strain-inducing-phase-changing due to applying pre-stress have comprehensive effect on the martensitic phase transformation in the grinding process. And its result indicates that the pre-stress restrains the phase transformation with grinding hardening layer firstly and then promotes it. And then pre-stress affects the residual stress mainly through affecting phase transformation in the typical PSHG. It has the same variation as phase transformation with the increasing of pre-stress in the process.

Abstract: In the helical milling process, a rotating tool traverses a helical trajectory to generate a hole. In order to investigate the cutting state in helical milling of carbon fiber reinforced polymer(CFRP), experiments were conducted with unidirectional and multidirectional laminates. Cutting forces, tool wear state and hole quality were discussed respectively. The effect of cutting parameters on the cutting forces was first presented, then the influence of different workpiece material on the cutting forces was secondly analyzed. Because of higher abrasion of carbon fiber, tool wear was a major problem without being ignored in the cutting process, so the study on the tool wear state was carried out from the point of wear process, wear form, coated state and change of element etc. Therefore the effect of tool wear on the cutting forces was detailedly analyzed. Next the relation between tool wear and entry delamination of hole-making was discussed. Finally, inner delamination, delamination factor, diameter error, surface roughness and circularity error were also investigated in detail.

Abstract: Nano/micro scale surface structure on precision machined parts dominates the various product functions such as friction, wear resistance, corrosion, fatigue, and wetting. This paper presents a multi-dimensional surface texture assessment using laser speckle pattern analysis. From the observation result of laser speckle in precision machined surface, the probability density distribution of light intensity on laser speckle changes with the surface roughness. In addition, the autocorrelation function of speckle intensity distribution changes with the periodicity of micro surface structure. By analyzing the relation between surface texture and laser speckle pattern, characteristic parameters of laser speckle reflecting surface texture properties are identified; i.e. the mean light intensity, the deviation of the probability density distribution of light intensity, the autocorrelation length of intensity distribution, and the degree of dissimilarity between adjacent speckle patterns. The proposed parameters can evaluate the roughness, the period, and the degree of dissimilarity in periodic micro surface structure. Furthermore, the surface texture distribution and the anisotropy of micro surface structure can be assessed by detecting the variation of the proposed parameters. In consequence, the proposed method can quantitatively assess the nano/micro scale surface texture.

Abstract: The thin liquid film entrained by a continuous immersed solid pulled through a tank is of the main interest in many industrial processes. In order to appraise the state of art, it has been reviewed the experimental and theoretical studies of a solid being withdrawn horizontally from a tank. This review has revealed that the open literature is scarce. The main objective of this paper is to study the above stated problem. To this end, it has been performed a series of experiences of horizontal withdrawal of wires from a tank of water, and an approximate theoretical model has been developed.

Abstract: The generation principle and meshing characteristics of conjugate-curve circular arc gears, which is proposed based on the theory of conjugate curves, are studied in this paper. The generation principle and mathematical model of this gear drive are introduced according to the given spatial screw curve. Tubular meshing surfaces contacting in the orientation of designated contact angle are provided to build up circular arc tooth profiles, which inherit all properties of conjugate curves. Geometrical three-dimensional models are established based on a numerical example and the results of motion simulation show that the generated gear drive meets general meshing and motion conditions. The comparisons between this designed gear drive and conventional circular arc gearing for the generation method and meshing characteristics are carried out. And the transmission efficiency of gear pair manufactured by hobbing cutter tools is also test. Through the analysis, it shows not only the general corresponding relationship between two gears, but also the conjugate meshing essence of circular arc gears. Theoretical and experimental results demonstrate the feasibility and correctness of proposed conjugate curves theory and the application to new types of gear drive with high performance will be carried out.

Abstract: The friction of rolling guideways in the prerolling region displays hysteretic behavior known as nonlinear spring behavior (NSB). NSB deteriorates motion accuracy and causes vibration in the feed direction. Therefore, the influences of NSB on the dynamic characteristics of rolling guideways should be clarified. This paper describes the influence of NSB on the dynamic characteristics of a rolling guideway. A simple friction model is constructed based on the Masing rule. Because the proposed friction model is described with only three parameters, the factor that inherently affects the dynamic characteristics can be clearly identified. To clarify the influence of NSB, the impulse response, frequency response, and steady state motion are analyzed by numerical analysis. According to the results, the dynamic characteristics in the feed direction depend only on the change rate of friction in the prerolling region, which is introduced to the friction model as the shape factor, n. The stiffness and damping are high when the change rate of friction is high in the prerolling region. The frequency response function is forcedependent, and its tendency is varied by n. The frequency response function includes harmonic and super harmonic resonances. When the carriage is excited with a frequency lower than that of the super harmonic resonance, a displacement spike (quadrant glitch) is observed. Additionally, the nonlinearity cannot be ignored when the carriage is excited with a frequency lower than the harmonic resonance. Finally, an experiment with a roller guideway is conducted to prove the validity of the analysis. The resonance frequency and compliance at the harmonic resonance measured by the experiment accurately conform to the analytical results.

Abstract: In order to enhance the load capacity of gears, researchers take notice in the asymmetric tooth gears. Asymmetric gear which has higher pressure angle of back tooth aiming to improve banding strength has been analyzed (Deng, et al., 2003). Brown et al. (Brown, et al., 2010) investigate the asymmetric tooth from both point of view of bending and scuffing. Kruger et al. (Kruger et al., 2013) analysed root stress of asymmetric gear by the boundary element method. Recently in Japan Society of Mechanical Engineers, improvement of contact strength by higher pressure angle is paid attention (Morikawa, 2013). However, they have not mentioned to torque capacity for bending considering the single tooth contact. This paper attempts the design of a tooth profile that has a higher load capacity. The root stress by change of pressure angle is analysed by FEM. Furthermore, the bending fatigue strength of tooth paying attention to the torque capacity is investigated by a simulation method based on inclusion distribution (Masuyama, et al., 2013).

Abstract: The purpose of a vehicle suspension is to maximize the contact between tires and road surface fo r good road holding, to provide steering stability fo r good handling and to ensure comfort of passengers for good ride. Vehicles’ road holding and handling performance main ly depend on the kinematic characteristics of vehicle suspension systems. From a kinemat ics point of view, a suspension system can be defined as a combination of links and joints. Changes in orientation and positions undergone by wheel during bump and rebound are called the kinematic characteristics defined by means of some functional parameters, whose values are specified by the design. A higher performance in the design is achieved as kinemat ic characteristics are improved. The most important kinematic paramet ers of suspension systems that effect handling, road holding, and ride characteristics are camber, caster, kingpin, and toe angle (Reimpell and Stoll, 1998). Coord inates of hard points and/or lengths of linkages are commonly considered as design variables to optimize performance of suspension systems. Due to the complexity of suspension systems, models are usually developed by using some commercial software. Independent suspension is a broad term for any vehicle suspension system that allows each wheel on t he same axle to move independently of each other. Independent suspension typically offers better ride quality and handling characteristics, due to lower unsprung weight and ability of each wheel to address road undisturbed by activities of other wheel on vehicles. The existence of different topologies and configurat ions confirms that there is no unique solution adapting well to all situations (Raghavan, 1991). The Macpherson strut (Habibi, et al., 2008) is a popular choice due to its simplicity and low manufacturing cost, but the design has a few disadvantages. Kinematic analysis shows it can’t allow vert ical movement of the wheel without some degree of either camber ang le change. It is not generally considered to give as good handling as a double wishbone s uspension, because it allows the engineers less freedom in design. A double wishbone suspension is an independent suspension design using two wishbone -shaped arms to locate wheel. Th is type of suspension system is also known as the acronym SLA (short -long arm) due to its

Abstract: The dynamic behaviors of the joint interfaces make a great difference in machine tools, especially in high-end numerical control machines. To provide an assessment of the normal dynamic parameters in contact deformation of elastoplastic solid joint interfaces theoretically, a dynamic model of joint interfaces was proposed and studied. Based on contact fractal theory and contact-mechanics theory, interfacial parameters, including normal damping dissipation factor, stiffness and damping have been investigated, while the influences of the elastic, elastoplastic and fully plastic deformations of contacting asperities are considered. Solution for the force-displacement relationship in the elastoplastic regime is done by the well supported assumptions: the microcontact area and microcontact normal load are enforcing continuity between the elastic and fully plastic regimes. Numerical calculation results reveal the complications of nonlinear relation between normal contact load and normal dynamic parameters, including the normal stiffness, damping dissipation factor and damping over joint interfaces. Furthermore, normal dynamic parameters and fractal parameters, including fractal dimension D and fractal roughness parameter G also present complicated relationships.

Abstract: A new cam-geared mechanism, consisting of a cam-follower and elementary planetary gear train is proposed for exact path generation. The formation and mobility of the proposed mechanism are illustrated in detail, in addition to inverse kinematic analysis of the design and the design procedure. The advantages of the new design include its simple, compact structure, simple design procedure and no numbers limitation of precision point. This new mechanism is suitable for generating diverse continuous curve paths and symmetric curve paths with an intersection. Examples are provided to show the feasibility and effectiveness of this proposed method.

Abstract: In this paper, dynamic reliability models of series mechanical systems in terms of stress parameters and strength parameters are established, in which strength degradation path dependence (SDPD) and failure dependence of components in the system are taken into consideration. Despite the computational convenience for reliability evaluation by using the independent strength distribution at each load application, large errors could be caused due to neglecting the existence of SDPD. In this paper, influences of SDPD on both dynamic system reliability and failure dependence are investigated. Moreover, the impacts of the dispersion of initial strength and the number of components in a system on the influences of SDPD on dynamic system reliability are analyzed. In addition, the clamp band joint system is used as illustrative examples to demonstrate the proposed models. The results show that SDPD have considerable influences, which vary in different operational stage of series mechanical systems, on dynamic system reliability and failure dependence. Besides, the dispersion of initial strength and the number of components have different impacts on the effects of SDPD on system reliability.

Abstract: A new type of gear pair consisting of a helical non-circular gear and a curve-face gear is designed, which can transmit the time-varying motion of intersected axes of rotation. Based on the generation mechanism of non-circular gear, the geometric model of helical curve-face gear is considered and the generating method of pitch curve of helical curve-face gear is discussed. According to the meshing theory, the meshing equation of helical curve-face gear is built and the mathematical model of tooth surface is developed. The method used to calculate its minimum inner radius and maximum outer radius is obtained. Furthermore, the method that the helical curve-face gear is generated by a shaper cutter is put forward and the three-dimensional model of helical curve-face gear is completed by modeling software. A phenomenon of transmission stability is discovered through contrast to the transmission ratio curve of experimental and theoretical value, which can illustrate the correctness of the design method and the practicability of the helical curve-face gear.

Abstract: This paper presents a novel study on the analysis of the “fully compliant” spatial four-bar mechanism. To the best of our knowledge, any research on “fully compliant” spatial four-bar mechanism is not available in the literature. In the previous study performed by the authors, a “partially compliant” version of the spatial four-bar mechanism was introduced. There was a rigid spherical joint in that case, thus there was no torsional loading at flexural hinges. For the fully compliant case, there is no spherical joint in the structure of the mechanism, thereby there is also torsion available at multiple axis flexural hinges. Design of this fully compliant mechanism is different from the partially compliant case. In this study, deflections of the multiple axis flexural hinges are determined separately as bending and twist. Essential angles for manufacturing a mechanism are determined. A prototype is built and results of the mathematical model are verified with experiments. Finally, a fatigue test is performed. After one and a half million cycles it is observed that there is no indication of any failure. Since there are many applications of rigid spatial four bar mechanisms, it is strongly believed that a fully compliant version of such a mechanism may also find applications.

Abstract: This paper deals with axial ultrasonic vibration-assisted machining with workpiece bending. It was proposed as a novel machining method for the reduction of the chippings at the machined holes during micro through-hole drilling of chemically strengthened glass. In micro through-hole drilling of chemically strengthened glass, machining accuracy and efficiency tend to be low because the material’s high hardness and brittleness cause rapid tool wear and large chippings at the inlet and outlet of the machined holes. In order to machine small holes with high accuracy, the reduction of the tensile stress that causes large chippings at the outlet of the machined holes is an issue of primary importance that deserves investigation. In the proposed machining method, the glass plate is bent slightly to be convex upward through the application of a compressive stress at the posterior surface of chemically strengthened glass, with a specially designed jig. Using this proposed method that can reduce the tensile stress, the chipping size at the outlet of the machined holes was successfully reduced with applied compressive stress values of 38.9 MPa. In conclusion, it has been clear that the axial ultrasonic vibration-assisted machining with workpiece bending has the potential for achieving high-precision and high-efficiency machining for chemically strengthened glass.

Abstract: Double cylinder structures (Yamamoto, et al., 2014), (Tamura, et al., 1999) are frequently used as machine elements and tube elements of equipments such as reactors (Fujiwara, et al., 2000), catalysts, heat exchangers (Alkan and Al-Nimr, 1999), and thermal insulators. A clearance exists between double cylinders of annular cross-section. This has engineering benefits if we can predict the acoustic properties such as the sound absorption coefficient of a sound incident on the clearance of such an annular cross-section based on the geometric dimensions of the tubes and the physical properties of the gas. With respect to the propagation of a sound wave in a narrow tubule structures, several studies have been performed on the propagation constant and particle velocity distribution (Tijdeman, 1975), complex density and complex compressibility (Stinson, 1991), complex density and complex bulk modulus (Allard and Atalla, 2009) in a circular hole, in an equilateral-triangular tube, or in between two planes (Stinson and Champou, 1992). Further, analysis using the finite element method of the vibration of the air in a rectangular tube (Beltman, et al., 1998), and on the propagation of a sound wave in the clearance created by the close packing of cylinders has also been conducted (Matsuzawa, 1963). With respect to naturally derived narrow tubes, studies have been performed on the acoustic absorption properties of straw (McGinnes, et al., 2005) and rice straw (Sakamoto, et al., 2011). The authors experimentally clarified that acoustic absorption effect in the clearance of an annular cross-section of a rice straw was useful (Sakamoto, et al., 2-8050 Ikarashi, Nishi-ku, Niigata, 950-2181, Japan

Abstract: Received 18 March 2015 Abstract The purpose of this paper is to achieve machining equal base circle bevel gears with a pot-shaped gear cutter. Based on the theory, the feasibility is analyzed, the coordinate systems of cutting, on both convex and concave surfaces, are established, and the mutual transformation matrices of these cutting coordinate systems are obtained. With the pot-shaped cutter coordinate system, the unit tangential vector and normal vector of the curved surface are calculated. Then, the tool surface point velocity vector under the fixed coordinate system is worked out through solving the instantaneous angular velocity of the points on the wheel billet and vector of the cutter curved surface. Finally, after establishing and solving of the meshing equations on the conjugate contact point between the cutter curved surface and tooth surface, the target machining expression is presented. The three dimensional tooth surface is modeled by Matlab, solving the contact line family at different locations on the cutter, and then comparison between the resulted tooth surface and the theoretical one is conducted. The tooth cutting experiment verified the feasibility of the proposed machining processes, which may lay a solid foundation for the industrial application of this new method.