Yan Song Wang

A Particle Swarm Algorithm for Assembly Sequence Planning

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.

Assembling Variation Analysis Based on Power Balance Method for Sheet Metal Parts

Sheet metal parts are widely used in auto body and airplane body. As the method of influence coefficient cannot accurately analyze the relationship between part deviations and assembly variations, this paper proposes a new variation analysis model based on a method of power balance for sheet metal parts. The simple metal planes are loaded to fixtures according to the principle of “4-2-1”; and then variety of part deviations is analyzed during assembly process. The analysis shows that the welding guns do same work during welding process. Based on the above, this work establishes sensitive coefficient matrices between part deviations and assembly variations. Finally, two Z-shape sheet metal parts illustrate the assembling variation analysis process and the results indicate the accuracy of the method of power balance.

Assembly Sequence Generation and Optimization for Compliant Assemblies Based on Genetic Algorithm

This paper proposes a genetic algorithm to generate and optimize assembly sequences for compliant assemblies. An assembly modeling is presented to describe the geometry of the assembly, which includes three sets of parts, relationships and joints among the parts. Based on the assembly modeling, an assembly sequence is denoted as an individual, which is assigned an evaluation function that consists of the fitness and constraint functions. The fitness function is used to evaluate feasible sequences; in addition, the constraint function is employed to evolve unfeasible sequences. The genetic algorithm starts with a randomly initial population of chromosomes, evolves new populations by using reproduction, crossover and mutation operations, and terminates until acceptable sequences output. Finally an auto-body side assembly is used to illustrate the algorithm of assembly sequence generation and optimization.

Tolerance Analysis for Compliant Assemblies with Thickness Deviation

Dimensional tolerance is an important factor for the geometry quality of products. The stiffness matrices of sheet metal components are usually not variational during the process of tolerance analysis; however, material properties of the compliant parts may change in volume production. The variation of the part thickness changes the part stiffness matrix so that the sensitivity matrix is different from that of the assembly with nominal thickness. The paper proposes an improved sensitivity matrix between the incoming part variation and the assembly variation for the assembly with thickness variation. Moreover, the orthogonal design is used to judge the importance of part and thickness variation for the assembly deviation. The proposed method is illustrated through a case study and the results show that the thickness variation influences the relationship between the part variation and the assembly variation. In addition, part variation plays a more important role to assembly variation than the thickness deviation does according to the orthogonal design.

Variation Analysis of Sheet Metal Assemblies with Fixture Configuration

The dimensional quality of auto-body relates to the whole external appearance and wind noise, the effect of closing the door and even driving smoothness of vehicles. The deviation propagation can be analyzed through the method of influence coefficient (MIC). However, fixture deviation usually impacts on the assembly variation than part variation. A variation analysis modeling with fixture configuration is presented to improve the current variation analysis efficiency of sheet metal parts. The variation change process of part, fixture and assembly was analyzed by researching the locating and assembly process of sheet metal parts. The linear relationships among variations of parts, fixtures and assemblies are established with two locating principles of “N-2-1” and “3-2-1”. Moreover, in accordance with the different release modes of the fixture locating points after the assembly, the assembly variation modeling is established in the two modes of over-constrained release and full release. Finally, a case of sheet metal assembly is illustrated to show the effectiveness of the assembly variation analysis modeling.

An Analysis of the Sound Insulation for Compound Steel Panel

Sound insulation equation of steel panel and other insulation materials of compound panel was deduced in normal case, based on the Wave Transferring theory in this paper. Single steel panel and compound steel panel with a certain thickness were predicted theoretically and then measured in the standing wave tube practically. Compared with the two results, the predicted values were matched with the measured values.

A Study on Static Features of a High-Speed Solenoid Valve Used in Diesel Engine

Based on the Maxwell equation and simulation analysis method, this paper investigates and retains the optimized parameters of a high-speed solenoid valve, which is very important for electrical unit pump (EUP) diesels. According to the electromagnetic theory, firstly, a three-dimensional (3D) model of the solenoid valve is established, and working process of the electromagnets is analyzed by using both the theoretical method and Ansys Workbench software. In view of the electromagnetic force, furthermore, several parameters related to the solenoid valve performance are considered and their effects on solenoid valve are analyzed and discussed. Finally, the optimized parameters are proposed for the analyzed solenoid valve and proved by theoretical equations. The studied approach can be extended to any other field related to the electromagnetic device design.

CFD Simulation of Volute of Variable Geometry Turbocharger

To study the Volute of Variable Geometry Turbocharger (VGT) flow field and the possibility of providing the basis theory for control strategy and matching with engine, in this paper, a method is presented. The 3D viscous compressible flow in the model of volute and the vanes is simulated by CFD using FVM (Finite Volume Method). And taking some VGT as an example, the simulation is carried out. The result shows that the method can display the distribution of pressure and velocity in the model clearly. The zone and the reasons resulting in loss will be found after analyzing the results, and then the turbocharger can be optimized and redesigned purposeful to reduce the losses resulted from improper figure. The distribution of pressure and velocity at open and close vanes will be found after analyzing the results, and the basis theory for VGT control strategy and matching with engine can be provided.

A Comparison of Psychoacoustical Evaluation Methods for Vehicle Engine Noises Based on Wavelet Shrinkage Denoising Technique

Based on the measured and denoised vehicle engine noises, the sound quality evaluation (SQE) techniques in common use for A-weighting sound pressure level (SPL) and psychoacoustical parameters, such as loudness, sharpness and roughness, are performed and compared in this paper. The wavelet threshold shrinkage method is adopted for sound signal denoising. The comparisons suggest that the A-weighting SPL is not sufficient and can only be used as a reference in vehicle SQE, the psychoacoustical parameters can provide a complete description and more exactly reflect to the subjective feeling of vehicle noises, should be considered in vehicle designs.

A Standing Wave-Duct System for Acoustical Property Prediction of Multi-Layered Noise Control Materials

Based on a hybrid transfer-matrix method, a new standing wave-duct system with four microphones for acoustical property prediction of multi-layered noise control materials is developed in this paper. In this system, the two-load method (TLM) and two-cavity method (TCM) are used for computing the transfer matrix of each material sample. The transfer matrices saved in a database in the system hardisk may be selected for predicting both the absorption ratio and transmission loss of a multi-layered treatment of materials. Verification results suggest that the newly designed standing wave-duct system is effective for acoustical prediction of multilayer material configurations.