基于注塑机螺杆位置与压力曲线的注射成型过程监测方法

提出基于注塑机螺杆位置与压力信号，采用非线性数据降维方法提取成型特征，建立原材料、模具温度与制品成型质量之间的神经网络模型，实现注射成型过程的故障监测和质量预测。试验结果表明：与未降维的数据和主成分分析等线性降维方法相比，拉普拉斯映射法和扩散系数图法相结合的非线性降维方法可从螺杆位置与压力信号中判断制品原材料、模具温度的异常及预测制品成型质量，在验证集和测试集均获得了高于0.92的回归系数，所提出的方法提取了原始曲线数据内在的高维度、非线性、强耦合的数据关系，有效进行了数据降维，提高了故障监测与质量预测模型的精度与效率。     

Fuzzy Regression Model to Predict the Bead Geometry in the Robotic Welding Process

Recently, there has been a rapid development in computer technology, which has in turn led to develop the fully robotic welding system using artificial intelligence (AI) technology. However, the robotic welding system has not been achieved due to difficulties of the mathematical model and sensor technologies. The possibilities of the fuzzy regression method to predict the bead geometry, such as bead width, bead height, bead penetration and bead area in the robotic GMA (gas metal arc) welding process is presented. The approach, a well-known method to deal with the problems with a high degree of fuzziness, is used to build the relationship between four process variables and the four quality characteristics, respectively. Using these models, the proper prediction of the process variables for obtaining the optimal bead geometry can be determined.     

面向顾客满意度的废旧机床再制造设计参数规划方法

基于目前大多数再制造机床仅注重机床本身性能的恢复,从未无法达到预期市场竞争力的问题,结合顾客需求和废旧机床再制造设计,构建了再制造设计参数决策框架,并充分考虑顾客需求与再制造设计参数之间的不确定性,应用模糊非线性回归法分析顾客需求与再制造设计参数之间以及不同再制造设计参数之间的模糊相关关系。同时,针对基于传统模糊非线性回归法的规划方程不具备处理模糊关系的能力,导致规划结果与实际情况偏离的情况,将模糊度融入规划方程,构建基于改进模糊非线性回归法的废旧机床再制造设计参数规划方程。以某机床再制造企业的废旧CAK6163的再制造设计过程为例,应用改进的再制造设计规划方程对其再制造设计参数进行规划。结果表明,由改进的规划方程得到的再制造设计参数,相对于传统规划方程,能够获得更高的顾客满意度,进而有效提升了再制造机床的市场竞争力。     

A study of the effects of process parameters for injection molding on surface quality of optical lenses

The purpose of this study is to determine the effects of process parameters on optical quality of lenses during injection molding. The quality characteristics chosen are light transmission, surface waviness and surface finish. The Taguchi method is used to perform screening experiments to identify the important significant process parameters affecting quality of lenses. Through empirical and theoretical analysis, the most significant process parameters affecting surface waviness is the melt temperature, followed by mold temperature, injection pressure and packing pressure. On the other hand, injection molding process parameters are found to have little effect on light transmission and surface finish of lenses. Regression approach is then implemented based on surface waviness data from full factorial experiments to formulate the regression models. Three different regression methods are used, namely, linear, exponential and nonlinear regression. Verification experiments are executed to examine the accuracy of the regression model for predicting quality characteristics of lenses. The result showed the highest accuracy prediction of surface waviness was from a nonlinear regression model. An error of only 4.24% suggested the existence of a correlation among injection molding process parameters.     

Linear and Non-linear Modeling of Cement-bonded Moulding Sand System Using Conventional Statistical Regression Analysis

A cement-bonded moulding sand system takes a fairly long time to attain the required strength. Hence, the moulds prepared with cement as a bonding material will have to wait a long time for the metal to be poured. In this work, an accelerator was used to accelerate the process of developing the bonding strength. Regression analysis was carried out on the experimental data collected as per statistical design of experiments (DOE) to establish input-output relationships of the process. The experiments were conducted to measure compression strength and hardness (output parameters) by varying the input variables, namely amount of cement, amount of accelerator, water in the form of cement-to-water ratio, and testing time. A two-level full-factorial design was used for linear regression model, whereas a three-level central composite design (CCD) had been utilized to develop non-linear regression model. Surface plots and main effects plots were used to study the effects of amount of cement, amount of accelerator, water and testing time on compression strength, and mould hardness. It was observed from both the linear as well as non-linear models that amount of cement, accelerator, and testing time have some positive contributions, whereas cement-to-water ratio has negative contribution to both the above responses. Compression strength was found to have linear relationship with the amount of cement and accelerator, and non-linear relationship with the remaining process parameters. Mould hardness was seen to vary linearly with testing time and non-linearly with the other parameters. Analysis of variance (ANOVA) was performed to test statistical adequacy of the models. Twenty random test cases were considered to test and compare their performances. Non-linear regression models were found to perform better than the linear models for both the responses. An attempt was also made to express compression strength of the moulding sand system as a function of mould hardness.     

Thermal modelling of the multi-stage heating system with variable boundary conditions in the wafer based precision glass moulding process

In the precision glass moulding process, the heat transfer and the resulting transient temperature distributions of the molten glass are of great importance because they significantly affect the productivity as well as the thermally induced residual stresses in the final product. Thermal modelling of the heating system in the glass moulding process considering detailed heating mechanisms therefore plays an important part in optimizing the heating system and the subsequent pressing stage in the lens manufacturing process.The current paper deals with three-dimensional transient thermal modelling of the multi-stage heating system in a wafer based glass moulding process. In order to investigate the importance of the radiation from the interior and surface of the glass, a simple finite volume code is developed to model one dimensional radiation–conduction heat transfer in the glass wafer for an extreme case with very high temperature difference considering temperature dependant thermal conductivity and heat capacity. Afterwards, by using three-dimensional FEM modelling along with a predefined experimental test, the equivalent glass–mould interface contact resistance is determined for two different pressures. Finally, the three-dimensional modelling of the multi-stage heating system in the wafer based glass moulding process is simulated with the FEM software ABAQUS for a particular industrial application for mobile phone camera lenses to obtain the temperature distribution in the glass wafer. In the numerical modelling, the interface boundary conditions for each heating stage are changed according to the determining heat transfer mechanism(s). Numerical results are compared with experimental data to show the validity of the numerical modelling. The obtained results show that the right thermal modelling is highly dependent on the proper choice of thermal boundary conditions in different stages according to the real physical phenomena behind the process.     

Evaluation of the viscoelastic behaviour and glass/mould interface friction coefficient in the wafer based precision glass moulding

Recent improvements in the manufacturing process of camera lenses have introduced the use of a new technology involving wafer based precision glass moulding. The utilization of this technology has some important advantages such as cost reduction, supply chain simplification and higher image quality. However, the required accuracy for the final size and shape of the moulded lenses as well as the complexity of this technology call for a high level of process understanding and numerical simulation is a very important part of achieving this goal. The viscoelastic parameters of the optical glass as well as the glass/mould interface friction coefficient play a key role in deformation behaviour and stress distribution of the moulded glass lens. Therefore, a proper evaluation of these parameters is the first important step in numerical modelling of the precision glass moulding process.The current paper deals with characterization of the interfacial glass/mould friction coefficient and viscoelastic behaviour of the L-BAL42 glass material above the glass transition temperature. Several glass rings are pressed at three different temperatures to various thicknesses and the experimental force, displacements, internal diameter and thickness of the rings are measured during the tests. Viscoelastic and structural relaxation behaviour of the glass are implemented into the ABAQUS FEM software through a FORTRAN material subroutine (UMAT) and the FE model is validated with a sandwich seal test. Then, by FE simulation of the ring compression test and comparison of the experimental creep with the simulated one in an iterative procedure, viscoelastic parameters of the glass material are characterized. Finally, interfacial glass/mould friction coefficients at different temperatures are determined through FEM based friction curves combined with experimental data points. The obtained viscoelastic parameters and interfacial friction coefficients can later be employed for prediction of the final shape/size as well as the stress distribution in the glass wafer during a real wafer based precision glass moulding process.     

Analysis of the benefit generated by using fuzzy numbers in a TOPSIS model developed for machine tool selection problems

Selection of the appropriate machine tools for a manufacturing company is a very important but at the same time a complex and difficult problem because of the availability of wide-ranging alternatives and similarities among machine tools. In the literature, various machine tool selection procedures are developed. The developed procedures mainly use Multi Criteria Decision Making (MCDM) methods. In the literature, fuzzy MCDM models, in which fuzzy numbers are used instead of crisp values, are proposed to deal with the vagueness and imprecision inherent in the machine tool selection problem. Although, the available studies in the literature developed various fuzzy models, they do not propose any approaches to measure the benefit generated by incorporating fuzziness in their selection models. This paper aims to fill this gap by trying to quantify the level of benefit provided by employing the fuzzy numbers in the MCDM models. Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) is used as the MCDM approach to rank the machine tools in this paper. In the paper, by increasing the fuzziness level steadily in the fuzzy numbers, the obtained machine tool rankings are compared with the ranking obtained with the crisp values. The statistical significance of the differences between the ranks is calculated using Spearman's rank-correlation coefficient. It can be observed from the results that as the vagueness and imprecison increases, fuzzy numbers instead of crisp numbers should be used. On the other hand, in sitiuations where there is a low level of fuzziness or the average value of the fuzzy number can be guessed, using crisp numbers will be more than adequate.     

基于不确定性云推理的刀具磨损量预测方法

针对刀具磨损存在随机性和模糊性的特点,提出基于不确定性云推理的刀具磨损量预测模型。首先利用逆向云算法计算刀具磨损声发射信号的3个云特征参数,期望、熵和超熵;其次,通过条件云发生器挖掘不同磨损阶段磨损趋势与不同磨损阶段云特征参数之间的关系,并构建基于云条件发生器的云预测规则;最后,在此基础上建立了多条件单规则云发生器的磨损量预测方法。研究结果显示:云推理刀具磨损量预测模型符合刀具磨损规律;对非确定模型进行预测,云推理比模糊推理更符合实际情况。此外,该方法能反映磨损的实时情况,具有较强的实用性。     

Change on temperature at the surface of injection moulded parts

A measuring system and a mathematical model are developed to study thermal transfer within polymers during the injection moulding phase. Temperature measurements within moulded parts reveal several “rise in the temperature” events at the surface of the part during the moulding cycle. These thermal singularities are related to the evolution of the parameters of implementation and depend on the nature of moulded polymer. The modelling of heat exchange makes it possible to integrate the evolution of the thermal contact between polymer and the mould as well as crystallization in the case of PP. The comparison between temperature experimental and theoretical results makes it possible to determine the characteristics of the process at the origin of these heating events.     

Modelling and prediction of cut shape for plasma arc cutting based on artificial neural network

Abstract

An artificial neural network approach for the modelling of plasma arc cutting processes is introduced. Neural network models have been proposed for predicting the cut shape and estimating the special cutting variables. The implementation of artificial neural networks in the modelling of cutting processes is discussed in detail. The performance of the neural networks in modelling is presented and evaluated using actual cutting data. Moreover, prediction applications of the above neural network models are described for various cutting conditions. It is shown that estimated results based on the proposed models agree well with experimental data; the neural network models yield good prediction results over the entire range of cutting process parameters spanned by the training data. The testing and prediction results show the effectiveness and satisfactory prediction accuracy of the artificial neural network modelling. The developed models are applicable to carbon steel.     

基于模糊熵的安全等级隶属度向量的离散化方法

针对矿山系统安全的随机性及模糊性 ,将模糊熵引入系统安全性评价模型的建模之中 ,提出了基于模糊熵和权广义距离之和最小的模糊模式识别模型 ,实现了系统安全等级隶属度向量的离散化 ,给出了模糊熵和权广义距离之间平衡参数的两种确定方法。实例表明 ,该模型优于基于权广义距离平方和最小的原则所建立的模糊模式识别模型 ;平衡参数的两种确定方法对评价结果具有良好的一致性。模糊熵可作为安全等级隶属度向量离散程度的评价指标。     

基于QGA-SVR的工件表面粗糙度预测和分析

在自动化生产中建立难加工材料的表面质量预测模型,是实现可持续制造的基础。提出一种结合量子遗传算法和支持向量回归(Quantum genetic algorithm-Support vector regression,QGA-SVR)的已加工表面粗糙度预测模型,改进了现有寻优方法在搜索支持向量回归的模型参数易陷入局部最优解的问题。在量子门更新的过程中加入交叉和变异的操作,保证了模型全局搜索能力,为了提高支持向量回归的泛化能力,在参数优化过程结合了K-折叠交叉验证。结合干车削304不锈钢的切削试验以及现有的铣削实验数据,对比分析了基于量子遗传算法和遗传算法的支持向量回归模型。结果表明:QGA-SVR具有收敛速度快、预测精度高的优点,基于建立的QGA-SVR模型分析了切削参数对车削表面粗糙度的影响规律。     

非等厚异种钢电阻点焊熔核成形的多元非线性回归模型

利用多元非线性回归正交组合的方法进行试验设计,分析两种厚度不一的异种钢板电阻点焊工艺.试验将表征非等厚异种钢材料电阻点焊熔核形状的熔核直径、熔核偏移,作为考察指标,将焊接脉冲电流、电极压力、焊接时间、热处理脉冲电流4个工艺参数,以及各参数之间的交互作用作为影响指标的因素,得到可预测熔核形状参数的回归数学模型.结果表明,优化的回归数学模型可实现该类非等厚异种钢电阻点焊接头熔核成形的较为有效的预测.在模型的基础上分析各工艺参数及各交互作用对焊点质量的影响规律,可进而对该类材料电阻点焊工艺参数进行优化设计.     

CO2弧焊电流波形的多参数自寻优智能控制

研究并提出了一种CO2气体保护焊电流波形的多参数自寻优智能控制法,电流波形的模糊寻优计及弧焊过程特征参数的综合影响,并在表征短路过渡综合性能的特征综合值作为弧焊过程定量评价和目标实时寻优的函数。在波控参数基值的基础上,基于综合性能寻优过程采用三级模糊调节算法,对电流波形参数分步调节寻优。该法实际上将双输入三输出式的波控参数模糊控制系统化为多路,开关性的双输入单输出式的模糊控制,通过短路及燃弧初始阶段的电流波形控制参数的模糊自寻优,逐步使波控参数和焊接规范参数达到和谐匹配,以实现焊接过程熔滴的稳定可控过渡,改善焊接过程的综合性能。     

镀锌钢板电阻点焊的多元非线性回归模型

为了研究用于家用轿车车身制造的镀锌钢板电阻点焊工艺,采用多元非线性回归正交组合的方法设计试验.试验将电阻点焊熔核形状参数和焊接接头抗剪强度作为考察指标,将焊接电流、电极压力、通电时间、预热电流四个参数,以及各参数之间的交互作用作为影响指标的考察因素,得到可预测熔核形状和焊接接头力学性能的四元二次回归数学模型,并通过方差分析对模型进行优化.结果表明,优化的回归数学模型可实现焊接接头熔核成形及力学性能较为准确的预测.在模型的基础上研究各参数及各交互作用对焊点质量的影响规律,从而可实现电阻点焊工艺参数的优化设计.     

Development of efficient numerical heat transfer model coupled with genetic algorithm based optimisation for prediction of process variables in GTA spot welding

Abstract

Although numerical heat transfer models based on conduction mode of heat transfer have become a strong basis for the quantitative analysis of fusion welding, they still find limited use in actual design for three primary reasons. First, these traditional models consider a volumetric heat source term, which ironically requires a-priori knowledge of the final weld pool dimensions. Second, the numerical models need confident values of a few parameters, e.g. arc efficiency and arc radius, which are usually uncertain and requires many trial and error simulations to realise their suitable values. Third, these models are rarely attempted for the prediction of possible weld conditions for a requisite or target weld dimensions, which is of paramount interest in design for welding. The present work attempts to circumvent these issues by linking a genetic algorithm (GA) based global optimisation scheme with a finite element based three-dimensional numerical heat transfer model. The numerical model includes a volumetric heat source that adapts itself to the computed weld pool geometry at any instant. The GA module identifies the optimum values of a set of uncertain parameters needed for the reliable modelling calculations and next, identifies the suitable values of the process variables, e.g. weld current, for a target weld dimension. In each case, the GA module guides the numerical model to compute weld dimensions for a given set of inputs, traces the sensitivity of the error in prediction on the inputs being optimised, updates them accordingly and reuses the numerical model to finally obtain their optimised values. The complete integrated model is validated with a number of experimental results in gas tungsten arc spot welding processes.     

基于机器学习的滚筒式飞剪剪切质量预测

滚筒式飞剪剪切质量与剪刃重叠量、侧向间隙等参数的取值密切相关,参数匹配不当不仅会影响断面质量,也会使刀具磨损严重,甚至出现剪不断的问题。为了预测选定参数下飞剪的剪切质量以选择合理的工艺参数,通过DEFORM-3D仿真了不同工艺参数下的飞剪剪切过程,运用机器学习算法分别对剪切断面质量和整体剪切效果建立了预测模型。测试集在剪切断面质量的预测模型上准确率接近90%,在整体剪切效果预测模型上的误差小于15%,两类模型为现场选择剪切工艺参数和智能化生产提供了理论指导。     

Sensitivity analysis of a finite element model for the simulation of stainless steel tube extrusion

In this work, a sensitivity analysis has been performed on a finite element model of glass-lubricated extrusion of stainless steel tubes. Fifteen model parameters, including ram speed, billet and tool temperatures, friction coefficients and heat transfer coefficients, were considered. The aim of the study was to determine the parameters that are most important for the response of the extrusion force. The relationship between the model parameters and the responses was analyzed by a calculation of two different regression models: one linear polynomial model and one model that includes interaction terms. Additional simulations were then carried out to validate the regression models. The results show that the initial billet temperature is the factor that has the strongest impact on the extrusion force within the parameter ranges studied in this work. The goodness of prediction and goodness of fit are very good for both regression models.     

Sensitivity analysis for process parameters in GMA welding processes using a factorial design method

Generally, the quality of a weld joint is strongly influenced by process parameters during the welding process. In order to achieve high quality welds, mathematical models that can predict the bead geometry and shape to accomplish the desired mechanical properties of the weldment should be developed. This paper focuses on the development of mathematical models for the selection of process parameters and the prediction of bead geometry (bead width, bead height and penetration) in robotic GMA (Gas Metal Arc) welding. Factorial design can be employed as a guide for optimization of process parameters. Three factors were incorporated into the factorial model: arc current, welding voltage and welding speed. A sensitivity analysis has been conducted and compared the relative impact of three process parameters on bead geometry in order to verify the measurement errors on the values of the uncertainty in estimated parameters. The results obtained show that developed mathematical models can be applied to estimate the effectiveness of process parameters for a given bead geometry, and a change of process parameters affects the bead width and bead height more strongly than penetration relatively.