Manufacturing Science and Engineering I

Abstract: The centrifugal pump of pressurized water reactor (PWR) in nuclear power plant is characterized by its complicated system, small accumulated data and fault samples. HMM has a strong ability to deal with time series modeling for dynamic process, while SVM has excellent generalization ability to solve the learning problems with small samples. This paper develops a state monitoring system based on the hybrid HMM/SVM model. The wavelet analysis techniques are used to extract features and the Hidden Markov Model (HMM) and Support Vector Machine (SVM) are used as the basic modeling and identification tools. The identification results of centrifugal pump show that the hybrid HMM/SVM system is effective and available for the state monitoring of the centrifugal pump of PWR in nuclear power plan.

Abstract: Resistance spot welding process is strongly related to interfacial contact behaviors. The effects of thermal contact is rarely investigated so far and generally ignored in numerical models. In this work, a parametric FE model, which considers the variation of the surface roughness of the electrodes and workpieces, has been developed to investigate the effects of thermal contact on weld nugget formation. With the parametric model, four cases, e.g. ideal smooth surface, minimal roughness surface and maximum roughness surface for steel sheets and electrodes of as-received condition, and highly rough electrode surface, are investigated. Researches show that when the surface roughness of the electrodes exceeds some limit, the thermal contact conductance will substantially affect the weld nugget formation, therefore, must be considered in numerical models to precisely predict welding process.

Abstract: A particle swarm algorithm is proposed to generate optimal assembly sequences for compliant assemblies. Firstly, the liaison graph and the adjacency matrix describe the geometry of the compliant assemblies. An assembly sequence is represented by a character string, whose length is the number of all parts. The conceptual tolerance analysis is used to evaluate feasible sequences. Thereafter, the particle swarm algorithm is presented to generate assembly sequences, in which the elite ratio is applied to improve optimization results. Finally a fender assembly is used to illustrate the algorithm of assembly sequence generation and optimization.

Abstract: In this paper, Monte Carlo (MC) technology with welding experiments and thermal simulation experiments were introduced to simulate the grain growth process in welding heat affected zone(HAZ) of 1060 industrial pure aluminum. First, a kinetic model was defined by both experimental and statistics method. Then the thermal cycle was calculated and at the same time the simulation program was worked out based on MC technology. Thirdly, the grain growth process in HAZ was simulated during different welding heat input, which has great influence on grain growth in HAZ. The result of the simulation demonstrates the grain growth process dynamically and embodies the “thermal pin effect”. Good agreement between MC simulation results and the experimental results was obtained which can provide a reliable evidence for evaluating the welding craft and the weldability.

Abstract: Pre-tensioning forces are, in essence, the selective application of clamping forces applied prior to processess to create a “stress field” envelope that aids the processes of components. There are many potential functions of applying pre-tensioning forces, such as improvement of component rigidity, reduction of machining deflection, and holding of components to counteract the machining forces etc. However, the use of pre-tensioning forces has not been extensively and comprehensively investigated. The aim of this paper is to strengthen the understanding of the impact of applying pre-tensioning forces firstly on simple parts and specifically on the fixture design development by establishing a methodology of using pre-tensioning forces. To investigate the optimised fixture layout and clamping strategy, Finite Element Analyses (FEA) were established to show the effect of applying pre-tensioning forces on machining deflection. Meanwhile, the relevant model validation experiments were applied to verify the FEA models in this study appropriately. Eventually, the results show that the FEA simulations are sufficient and the use of pre-tensioning forces effectively reduces the machining deflection by using optimised clamping strategy.

Abstract: A miniaturized CNC (Computerized Numerical Control) milling machining equipment was mentioned for micro milling processing. The equipment has a highest spindle speed of 90000rpm and the positioning accuracy of submicron. Furthermore, it described the design of CNC servo system of precision stage driven by linear motor. The servo experiments results verified the excellent static and dynamic performance of the system. The machining tests included a plane with a surface roughness of 215nm using a end milling cutter of Φ0.2μm, micro straight slots with a dimension error range of 1-2μm using a ball-end cutter of Φ0.2μm and some complex structures. The analysis results of these tests show that the system is able to fulfill the micro milling machining of micro components.

Abstract: In this paper biomechanical behavior of dental implant and surrounding bone system are investigated under static occlusal loads through 3D nonlinear finite element analysis (FEA), taking into account the interaction of implant-bone and implant-abutment contact interfaces. Stress-based performances of four commercially-available dental implant systems are evaluated in detail, demonstrating that implant and bone stability is strongly affected by implant-abutment connection structure as well as by a number of geometrical parameters. The results also indicate that platform-switching configuration can significantly reduce the crestal bone stress peaks, which contributes to the bone preservation for long-term success.

Abstract: The feasibility of using MR fluids to develop a smart material-structure system and thereby to prevent structure from catastrophic damage is actively carried out in this study. The MR smart fluids are utilized to develop a smart damper with controllable parameters and the mechanical model of MR devices is presented by involving the effects of damper's axial stiffness. The interaction of a smart material-structure system is analyzed through the numerical integral and iteration. The vibration mitigation of the dynamic responses of an engineering structure with the smart coupled system is conducted based on the intelligent fuzzy control for performance evaluation. It is observed that the structural internal forces such as shear forces and bending moments can be remarkably mitigated with the protection of the smart system.

Abstract: A theoretical model is developed to examine the effect of bending rigidity on measuring the adhesion work between a flat punch and a flexible membrane by pull-off test. The validity of this model is certified by the theoretical analysis. It can be used in the future experiments to analyze the experimental data when the bending rigidity of the membrane cannot be ignored. From our quantitative analysis, the effect is remarkable and sensitive to the bending rigidity. The effect of bending rigidity should not be neglected unless the membrane is very soft or the membrane is very thin. The elastic modulus of membrane should be less than 1.6MPa when the thickness of the membrane is less than 10μm. Moreover, it is obviously that the thickness plays a key role in the bending rigidity.

Abstract: To overcome the problem of low convergence speed and sensitivity to local convergence with the traditional Artificial Fish-swarm Algorithm (AFSA) to handle complex functions, a novel compound evolutionary algorithm, called AFS-EMPCEOA, was introduced which is combined Artificial Fish-swarm Algorithm with the Elite Multi-parent Crossover Evolutionary Optimization Algorithm (EMPCEOA) that is GuoTao Algorithm improved by elite multi-parent crossover method. AFSEMPCEOA algorithm program with hybrid discrete variables was also developed. The computing example of mechanical optimization design shows that this algorithm has no special requirements on the characteristics of optimal designing problems, which has a fairly good universal adaptability and a reliable operation of program with a strong ability of overall convergence and high efficiency.