Search results

In today's electronic package development cycle, activities are managed by multiple participants in the supply chain, which might have different quality and reliability impacts to the end product. As a result, the reliability risk is much higher for companies who do not have insight into and/or control over the products received. The purpose of this paper is to show how design‐for‐reliability (DFR) approaches will come into play to manage the risk.

In this paper, DFR approaches for package development will be discussed from the perspective of the original equipment manufacturers (OEMs). DFR practices through the package development cycle will be described based on key development modules. A case study for flip chip ball gris array package development using an advanced Cu/Low‐k silicon technology will be presented. Key measures to help control the quality and improve the reliability will be presented.

The proposed methodology significantly improves component and package reliability through the engagement in design, manufacturing, assessment and system evaluation.

The paper discusses the research results and the proposed DFR methodology will be helpful for fabless design houses, electronics manufacturing service (EMS) partners in the supply chain, and OEMs to manage the reliability of the products.

This paper aims to improve the adaptability and control performance of the controller, a proposed seeker optimization algorithm (SOA) is introduced to optimize the controller parameters of a robot manipulator.

In this paper, a traditional proportional integral derivative (PID) controller and a fuzzy logic controller are integrated to form a fuzzy PID (FPID) controller. The SOA, as a novel algorithm, is used for optimizing the controller parameters offline. There is a performance comparison in terms of FPID optimization about the SOA, the genetic algorithm (GA), particle swarm optimization (PSO) and ant colony optimization (ACO). The DC motor model and the experimental platform are used to test the performance of the optimized controller.

Compared with GA, PSO and ACO, this novel optimization algorithm can enhance the control accuracy of the system. The optimized parameters ensure a system with faster response speed and better robustness.

A simplified FPID controller structure is constructed and a novel SOA method for FPID controller is presented. In this paper, the SOA is applied on the controller of 5-DOF manipulator, and the validation of controllers is tested by experiments.

The purpose of this study is to develop a modified parameter identification method and a novel measurement method to calibrate a 3 degrees-of-freedom (3-DOF) parallel tool head. This parallel tool head is a parallel mechanism module in a five-axes hybrid machine tool. The proposed parameter identification method is named as the Modified Singular Value Decomposition (MSVD) method. It aims to overcome the difficulty of choosing the algorithm parameter in the regularization identification method. The novel measurement method is named as the vector projection (VP) method which is developed to expand the measurement range of self-made measurement implements.

Newton Iterative Algorithm based on Least Square Method is analyzed by using the Singular Value Decomposition method. Based on the analysis result, the MSVD method is proposed. The VP method transforms the angle measurement into the displacement measurement by taking full advantage of the ability that the 3-DOF parallel tool head can move in the X – Y plane.

The kinematic calibration approach is verified by calibration simulations, a Rotation Tool Center Point accuracy test and an experiment of machining an “S”-shaped test specimen.

The kinematic calibration approach with the MSVD method and VP method could be successfully applied to the 3-DOF parallel tool head and other 3-DOF parallel mechanisms.

This study aims to investigate the noise-inducing characteristics during the start-up process of a mixed-flow pump and the impact of different start-up schemes on pump noise.

This study conducted numerical simulations on the mixed-flow pump under different start-up schemes and investigated the flow characteristics and noise distribution under these schemes.

The results reveal that the dipole noise is mainly caused by pressure fluctuations, while the quadrupole noise is mainly generated by the generation, development and breakdown of vortices. Additionally, the noise evolution characteristics during the start-up process of the mixed-flow pump can be divided into the initial stage, stable growth stage, impulse stage and stable operation stage.

The findings of this study can provide a theoretical basis for the selection of start-up schemes for mixed-flow pumps, reducing flow noise and improving the operational stability of mixed-flow pumps.

The purpose of this paper is to build a transient wear prediction model of surface topography of textured work roll, and then to investigate the wear performance of different original textured surfaces. The surface topography of steel sheets is one of the most important surface quality indexes, which is inherited from the textured work rolls in cold rolling. Surface topography of work roll is obviously changing in the cold rolling process. However, surface topography is difficult to measure in the industry production process.

This paper presents a numerical approach to simulate the wear process based on the mixed lubrication model of cold rolling interface developed by Wilson and Sheu (Sheu and Wilson, 1994). It is assumed that wear takes place at locations where the surfaces are in direct contact, and the volume is removed by an abrasive particle which is an abstract concept based on the wear phenomenon of textured work roll. At each simulation cycle, the distribution of the contact pressure is calculated by the lubrication model. The material is removed by an abstract abrasive particle and the surface topography is modified correspondingly. The renewed surface topography is then used for the next cycle.

Through comparative analysis, it can be found that the simulation results possess similar statistical characteristic with the measured data. A set of roughness parameters such as the amplitude, spacing and frequency-domain characteristics are introduced to analyze the wear performance of different textured surfaces. Numerical examples show that the surface topography has a significant effect on the wear performance of work roll in cold rolling.

The proposed model can accurately predict the wear process of the surface topography in the cold rolling process, which provides the foundation for optimization of original surface topography of textured work roll. The model can also be considered as a tool applicable for research on control of the surface topography of steel strip in the cold rolling process.

Automatic guided vehicles (AGVs) are widely used in industrial fields. But most control strategies merely take the lateral force into consideration. This will reduce the accuracy, stability and robustness and will pay additional costs. Therefore, this paper aims to design a control strategy that initially considers lateral force. Thereby, it will improve the accuracy, stability and robustness and reduce the overall cost of AGV.

To achieve the goal of comprehensively improving AGV operating performance, this paper presents a new scheme, combining the dual-wheeled chassis model (DCM) using proportional–integral–differential (PID) control and a supporting quick response (QR) code navigation technology. DCM is the core, which analyzes the deviation caused by lateral force. Then, DCM with PID control by the control law is combined to suppress the errors. Meanwhile, QR code navigation technology provides effective data support for the control strategy.

Most AGV experiments are carried out in a standard environment. However, this study prepares unfavorable scenarios and operating conditions for the experiments that generate detailed data to demonstrate this study’s strategy, which can make an accurate, stable and robust operation process of AGV under various adverse environmental and mechanical factors.

This study proposed DCM, fully considering lateral force and converting the force into velocity. Subsequently, PID controls the speed of two wheels to reduce the error. QR code provides an efficient and low – cost way to obtain information. The three are cleverly combined as a novel industrial AGV control strategy, which can comprehensively improve the operating performance while reducing overall costs.