A novel micro-assembly method based on the mapping between assembly force and position

Assembly Performance of a Robotic Vibratory Wrist

This paper presents a comprehensive methodology to model the assembly process of flip chip on flex interconnections with non-conductive adhesive (NCA). The methodology combines experimental techniques for material characterization, finite element modeling and model validation. The non-conductive adhesive material has been characterized using several techniques. A unique experimental technique has been developed to measure the cure force. A 2-D axisymmetric finite element model is used for analysis of flip chip on flex package with nonconductive adhesive, which takes into account assembly force, cure shrinkage, adhesive modulus buildup, removal of assembly force and cooling down to room temperature. The relationship between the contact pressure obtained from finite element simulation and the measured bump contact resistance has been established through the development of a dedicated experimental setup, which uses a microforce-tester combined with a digital multimeter and nano-voltmeter. The process modeling has been validated by comparing the predicted contact resistance value and the measured contact resistance value after assembly process. The approach developed in this paper can be used to provide guidelines with respect to adhesive material properties, assembly process parameters and good reliability performances.

In order to forecast missile storage reliability better, the paper researched a forecasting method based on neural network which is with the ability of actualizing multi-nonlinear mapping from input to output, and discussed steps of forecasting based on back propagation (BP) network and radial basis function (RBF) network respectively. At last, the storage reliability of one type ship-to-ship missile is forecasted based on BP network and RBF network respectively. The results show that both of the BP and RBF are suitable for Missile storage reliability forecasting, and the precision of the train goal is better by using RBF network. RBF network is more suitable for dealing with this problem.

This paper presents a test environment enabling the study of factors affecting on the success of a robotic precision assembly work cycle. The developed testing environment measures forces and torques occurring during the assembly, and uses a system based on machine vision to measure the repeatability of work piece positioning. The testing environment is capable of producing exactly known artificial positioning errors in four degrees-of-freedom to simulate errors in work-piece positioning accuracy. The testing environment also measures the total duration of the robot work cycle as well as the durations of all essential phases of the work cycle. The testing environment is best suited for light assembly operations and has measurement ranges of ±36 N and ±0.5 Nm and the vision system has a field-of-view mm.The latter part of this paper presents the results of the research done in order to find out how some selected factors affect the assembly forces of robotic assembly. These factors include work piece and process parameters such as work piece material and design (chamfered/straight), positioning tolerances, and robot insertion motion speed.Key wordsAssembly forceassembly process testingwork cycle timepositioning errors

Several self‐assembly structures of anatase TiO2 nanorods were obtained by a two‐step assembly process, and these structures formed different lyotropic liquid crystal in solution. Primary self‐assembly occurred in synthesis process and formed two structures, in the morphology of ribbon and honeycomb, respectively. Secondary‐assembly took place when the products were placed at lower temperature, where unique structures were obtained as the relative amount of ribbon and honeycomb changed with the increase of TiO2 concentration. These structures showed nematic, spherulites, and lamellar phases. The mechanism of the two‐step self‐assembly was clarified. The driving force of primary assembly is deduced to be anisotropic attractive force, for NRs can assemble at any concentrations, while gravity is the driving force of the secondary assembly. It is worth mentioning that this paper is the first report about spherulites composing of anatase TiO2 nanorods. The spherulites obtained were negative or of tangential type, and its structure, growth process, and temperature influences were also investigated. The spherulites may have promising application in temperature microsensor.

It is of clinical interest to estimate pure-tone thresholds from potentially available objective measures, such as stimulus-frequency otoacoustic emissions (SFOAEs). SFOAEs can determine hearing status (normal hearing vs. hearing loss), but few studies have explored their further potential in predicting audiometric thresholds. The current study investigates the ability of SFOAEs to predict hearing thresholds at octave frequencies from 0.5 to 8 kHz. SFOAE input/output functions and pure-tone thresholds were measured from 230 ears with normal hearing and 737 ears with sensorineural hearing loss. Two methods were used to predict hearing thresholds. Method 1 is a linear regression model; Method 2 proposed in this study is a back propagation (BP) network predictor built on the bases of a BP neural network and principal component analysis. In addition, a BP network classifier was built to identify hearing status. Both Methods 1 and 2 were able to predict hearing thresholds from 0.5 to 8 kHz, but Method 2 achieved better performance than Method 1. The BP network classifiers achieved excellent performance in determining the presence or absence of hearing loss at all test frequencies. The results show that SFOAEs are not only able to identify hearing status with great accuracy at all test frequencies but, more importantly, can predict hearing thresholds at octave frequencies from 0.5 to 8 kHz, with best performance at 0.5 to 4 kHz. The BP network predictor is a potential tool for quantitatively predicting hearing thresholds, at least at 0.5 to 4 kHz.

Residual storey drift estimation of the MDOF system with the weak storey under seismic excitations using the BP network

The paper describes a neural network system for forecasting time series and its application to a non-trivial task in forecasting currency exchange rates. The architecture consists of a two-layer backpropagation network with a fixed number of inputs modelling a window moving along the time series in fixed steps to capture the regularities in the underlying data. Several network configurations are described and the results are analysed. The effect of varying the window and step size is also discussed as are the effects of overtraining. The error backpropagation network was trained with currency exchange data for the period 1988-9 on hourly updates. The first 200 trading days were used as the training set and the following three months as the test set. The network is evaluated both for long term forecasting without feedback (i.e. only the forecast prices are used for the remaining trading days) and for short term forecasting with hourly feedback. By careful network design and analysis of the training set, the backpropagation learning procedure is an active way of forecasting time series. The network learns the training set near perfect and shows accurate prediction, making at least 20% profit on the last 60 trading days of 1989.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

The neural network (NN) training algorithm based on gradient optimization can not avoid falling into the local minimum because of the inappropriate initial weight value. The paper applies the genetic algorithm (GA) to training the linkage weights of NN. The training result can be used as the weights of an initial network for a back propagation (BP) training algorithm, then online optimization work can be done by BP training algorithm. It succeeds in avoiding the GA's defect of high calculating cost of every step, and giving full play to GA's advantage of greater probability of global convergence. Thus the online BP training algorithm can be lifted out of local minimum with greater probability and the better training property of the network is gained. The result of simulation shows that the robustness of the control system is improved.

Prediction of laser butt joint welding parameters using back propagation and learning vector quantization networks

Big Data Methods for Precision Assembly

In interference fit assembly, the magnitude and deviation of the assemble force are large so that it is hard to reach high accuracy of position for components of multiple parts stacked up. A force/stiffness compensation method is proposed to control the positioning accuracy in the interference fit assembly for multi-peg-hole components. Based on the force and displacement information measured in the assembly process, the position errors are acquired, and the stiffness of the assembly system under the exerted assemble force is calculated. According to the stiffness, the deviation from the target position is calculated and compensated. An experimental equipment based on this method was developed. As an example, assembly of rings, 6.2 mm in diameter and 0.25 mm in thickness, was carried out to demonstrate the feasibility of the proposed method. The assembly results show that high positioning accuracy of the assembled rings can be achieved with a large variation of assembly force. The presented method provides a simple, feasible, and efficient solution for interference fit assembly for multi-peg-hole components.

The modeling and analysis of the assembly contact problems in the picking and placement of the automobile alternator frame are carried out in the multi-station processing. According to the three-dimensional assembly characteristics of the actual alternator frame and the fixture, the assembly model in the drilling and milling station is simplified. The alternator frame and fixture assembly problem as a whole is simplified to a dual peg-in-hole assembly problem. All possible one-point contact states and two-point contact states during the assembly process are proposed(L-1)∼(R-10). The contact force analysis is carried out on the typical one-contact state L-1 and the typical two-point contact state L-5, and the relationship between the corresponding assembly forces Fx, Fy, Fz and the assembly moments Mx, My, Mz is obtained. According to the relationship between these assembly forces and assembly moments, the corresponding jamming diagrams are obtained. Finally, according to the three-dimensional size of the actual alternator frame, the experimental verification was carried out by using a 6-DOF force/moment sensor. The experimental results confirmed the validity of the analysis. Technical support is provided for the assembly of automobile alternator frames based on force sensing control for the follow-up work.

In the mass production assembly process, the fixture system’s reliability is vital for products’ quality. The failure of the fixture system depends not only on the assembly operation times, but also upon the system degradation which is caused by the original manufacturing accuracy of fixtures, assembly forces, qualities of subassemblies from upstream assembly stations etc. In this paper, we propose a dynamic preventive maintenance policy for the fixture components based on the system reliability model of a multi-station assembly process. The proposed reliability model not only considers the degrading fixture components and other factors in the station, but also the dependence of process factors across the stations. Based on the reliability model and a given cost, an optimization method for the manufacturing tolerances and a dynamic maintenance schedule of locating pins are presented. At last, a body side assembly case is given to illustrate the proposed method.

an Artificial Neural Network classifier is a nonparametric classifier. It does not need any priori knowledge regarding the statistical distribution of the class in a giver selected data Source. While, neural network can be trained to distinguish the criteria used to classify easily in a generalized manner that allows successful classification the newly arrived inputs not used during training. Through this paper it is eastliblished that back propagation neural network works successfully for the purpose of classification. Back propagation suffers from getting stuck into Local Minima. Weight optimization in Back propagation can be optimized using the Genetic Algorithm (GA). The back propagation algorithm is improved by invoking Genetic algorithm, to improve the overall performance of the classifier. The performance of a fitness algorithm using the approach suggested by us is a Hybrid System that is being analyzed in this paper. In this paper the issue of improving the fitness (weight adjustment) of Back propagation algorithm is addressed. Some of the Advantages of Hybrid algorithms are: convergence speed will be increased and the local minima problem can be overcome. The proposed Hybrid Algorithm is to perform learning as a back propagation and optimize weights using GA for classification.