S形件翻边成形工艺参数优化设计

建立了S形件成形的参数化有限元模型并进行参数化有限元分析(PFEA).结合人工神经网络(ANN)和遗传算法(GA)进行优化设计,研究了S形件翻边成形中的工艺参数优化.然后利用优化的结果值进行验算,获得了良好的成形件,表明基于PFEA、ANN和GA进行成形工艺参数优化是可行的.     

拉深孔成形的工艺参数优化与实验研究

为了探讨拉深孔成形技术对提高板料成形性能的有效性，以圆筒形件为研究对象，研究了拉深孔成形条件下的成形性能。采用数值模拟、人工神经网络和遗传算法进行板材成形工艺参数优化，得到了最优化的压边力和拉深孔相对密度等拉深工艺参数。根据优化后的结果设计并完成了相关的工艺实验，取得了与有限元模拟相一致的结果，证明了拉深孔成形技术是一种提高板材成形性能的行之有效的方法。     

基于神经网络和遗传算法的BK7玻璃精密模压成形多目标优化

残余应力和形貌偏差是玻璃模压成形影响整体质量的重要指标，而影响这些指标的成型条件是一个多因素问题。为了得到玻璃精密模压成形过程中成形透镜的残余应力和形貌偏差的极小值，以模压成形工艺参数为决策对象，以玻璃透镜产品质量为优化目标，对成形工艺参数进行多目标优化分析，优选出最终结果。首先基于仿真与实验结合对玻璃成形过程进行分析验证，然后采用正交实验方法选择合理参数，用有限元技术得到模压成形透镜实验样本的最大残余应力、成形应力、最大形貌偏差，并使用方差统计分析方法降维筛选出主要成分，建立合理的分析主参数，接着利用人工神经网络映射成形工艺参数与优化目标间的关系，建立工艺优化的数学模型，最后利用遗传算法寻找最优解。结果表明提出的基于人工神经网络和遗传算法的BK7玻璃精密模压成形工艺优化方法是有效可靠的。     

有限元逆算法与板料成形工艺的评价

依据理想形变理论，研究开发了冲压成形过程模拟的有限元逆算法，根据变形体的整体塑性功取相对极值的条件，导出了塑算法有限元方程。提出了求逆算法初始解以及求解与给定形状的毛坯相对应的冲压件形状的迭代计算方法。采用有限元塑算法预测了与冲压件形状相对应的冲压件毛坯的展开形状，根据给定的板坯形状计算了冲压件最终构形及应变分布。分析计算实例表明，逆算法可用于对板料成形工艺方案进行快速评价，对冲压工艺参数进行优化。     

材料成形计算机模拟中的参数化有限元法

在分析常规有限元法在材料成形模拟中的不足的基础上，提出了参数化有限元分析的概念，阐述了其基本任务，基本过程，实现方法和进行参数化有限元分析的技术要点。针对方孔翻边问题进行了编程式参数化有限元法研究，其结果与现有文献试验结果基本吻合，针对方盒拉深中的坯料形状优化问题进行了交互式参数化有限元法研究，其结果与现有文献试验结果基本吻合。研究表明，提出的参数化有限法实用，可行，为材料成形计算机模拟探索出一种高效快捷、实用的新方法。     

瑞风商务车托架拉延成形数值模拟及工艺参数优化

基于逆向工程建立瑞风商务车托架零件的几何模型,并基于Dynaform软件平台对不同工艺参数下该零件的拉延成形过程进行数值模拟。在此基础上,以压边力、拉延筋高度和拉延筋圆角半径作为设计变量,以零件不发生破裂为优化目标,以有限元数值模拟结果作为虚拟样本,建立目标函数的人工神经网络预测模型;将人工神经网络预测模型作为优化算法的知识源,采用遗传算法对压边力、拉深筋几何参数等工艺参数进行了优化设计。试验结果表明,数值模拟、神经网络预测和工艺优化是可靠的,从而可为制定金属板料最佳的冲压成形工艺提供一条先进、合理的途径。     

基于多目标遗传算法的板料拉深成形工艺参数优化设计

以多种工艺参数(压边力、摩擦因数等)作为优化变量,多种成形缺陷(起皱、破裂等)作为优化目标,结合多目标遗传算法和数值模拟,建立了板料拉深成形工艺参数的优化设计模型。为了减少数值模拟的次数,利用人工神经网络建立了各种工艺参数和模拟结果之间的映射关系,大大提高了优化的效率。以汽车消声器为例,对其拉深成形工艺参数进行了优化,通过对优化结果进行数值模拟可以看出,该优化参数完全避免了各种缺陷的产生,这说明该优化算法具有较好的优化结果。     

基于有限元逆算法的板料成形模拟拉深筋的灵敏度优化

基于有限元逆算法和灵敏度优化的BFGS （Broyden-Fletcher-Goldfarb-Shanno）算法，提出一种以成形极限曲线和起皱极限曲线作为目标函数，优化拉深成形中的拉深筋位置和大小的拉深筋灵敏度优化算法。优化算法综合考虑冲压件成形过程中起皱和破裂对成形性的影响因素，比只考虑板厚变化更能准确地反映板料的成形性，而且基于逆算法的灵敏度优化算法计算速度很快，可以应用于大型复杂冲压件的工艺参数优化过程模拟。通过两个典型的零件验证了该算法的优化结果符合实际工艺。     

基于微型多目标遗传算法的薄板冲压成形变压边力优化

提出一种高效的薄板冲压成形变压边力多目标优化方法，该方法以减少冲压件的成形缺陷为优化目标，以变压边力曲线的特征参数为优化变量，采用自主开发的微型多目标遗传算法作为优化算法，并在优化过程中引入神经网络近似模型以减少数值模拟的次数，提高优化效率。通过NUMISHEET’93的U形弯曲标准模型和某车型前地板角支撑板冲压成形模型两个变压边力优化实例对该方法进行了验证。结果表明，该方法既能高效率地解决薄板冲压成形变压边力优化问题，又能仅通过一次计算就提供多组方案以满足对冲压件成形质量控制的不同需要。     

基于多目标遗传算法优化板料拉深成形工艺参数

利用人工神经网络构建了板料拉深成形的目标函数模型,建立了板料拉深成形工艺参数和性能评价指标之间的映射关系.以多种工艺参数(压边力、摩擦因数等)作为优化变量,多种成形缺陷(起皱、破裂等)作为优化目标,结合多目标遗传算法和数值模拟,建立了板料拉深成形工艺参数的优化设计模型,大大提高了优化的效率.以油底壳下盖为例,对其拉深成形工艺参数进行了优化,通过对优化结果进行数值模拟可以看出,该优化参数完全避免了各种缺陷的产生,这说明该优化算法具有较好的优化结果.     

Multi-objective optimization of electric-discharge machining process using controlled elitist NSGA-II

Parametric optimization of electric discharge machining (EDM) process is a multi-objective optimization task. In general, no single combination of input parameters can provide the best cutting speed and the best surface finish simultaneously. Genetic algorithm has been proven as one of the most popular multi-objective optimization techniques for the parametric optimization of EDM process. In this work, controlled elitist non-dominated sorting genetic algorithm has been used to optimize the process. Experiments have been carried out on die-sinking EDM by taking Inconel 718 as work piece and copper as tool electrode. Artificial neural network (ANN) with back propagation algorithm has been used to model EDM process. ANN has been trained with the experimental data set. Controlled elitist non-dominated sorting genetic algorithm has been employed in the trained network and a set of pareto-optimal solutions is obtained.

     

Multi-objective optimization of sheet metal forming die using FEA coupled with RSM

Present study describes the approach of applying response surface methodology (RSM) with a Pareto-based multi-objective genetic algorithm to assist engineers in optimization of sheet metal forming. In many studies, finite element analysis and optimization technique have been integrated to solve the optimal process parameters of sheet metal forming by transforming multi objective problem into a single-objective problem. This paper aims to minimize objective functions of fracture and wrinkle simultaneously. Design variables are blank-holding force and draw-bead geometry (length and diameter). Response surface model has been used for design of experiment and finding relationship between variables and objective functions. Forming limit diagram (FLD) has been used to define objective functions. Finite element analysis applied for simulating the process. Proposed approach has been investigated on a cross-shaped cup drawing case and it has been observed that it is more effective and accurate than traditional finite element analysis method and the ‘trial and error’ procedure.     

Non-parametric dimension reduction algorithm approach for neural networks applied to diagnostic systems

Diverse techniques have been developed for dimension reduction, especially to facilitate the implementation of artificial neural networks (ANNs). For ANNs, the training process can become very complex and demand a great deal of hardware resources, making prior dimension reduction very important; accordingly, this research proposes a new algorithm to increase the degree of dimension reduction. A new procedure is applied to extract important and meaningful non-parametric characteristics from the data. The data in this research was obtained from accelerometers installed in a wind power machine and processed using a linear predictive coefficient/cepstrum coefficients procedure. The procedure consists of the extraction of linear predictive coefficients from the signal data, and subsequent extraction of six features from those coefficients, thereby reducing the amount of data to process and enabling the processing of that information using neural networks. The features employed were selected carefully based on the error obtained from a neural network implementation. The results of this implementation reveal to reduce the data shown to reduce the data to only six input variables for the ANN, thereby enabling the ANN to achieve a very low rate of classification error and training time consuming.     

Optimisation of turning parameters by integrating genetic algorithm with support vector regression and artificial neural networks

This paper focuses on optimisation of process parameters of the turning operation, using artificial intelligence techniques such as support vector regression (SVR) and artificial neural networks (ANN) integrated with genetic algorithm (GA). The model is trained using the turning parameters as the input and corresponding surface roughness, tool wear and power required as the output. Data, obtained from conducting experiments is analysed using support vector machine (SVM) and artificial neural network. SVM, a nonlinear model, is learned by linear learning machine by mapping into high-dimensional kernel-induced feature space. The genetic algorithm is integrated with these to find the optimum from the response surface generated. The results are compared with those obtained by integrating GA with traditional models like response surface methodology (RSM) and regression analysis (RA). This paper illustrates the impact that techniques based on artificial intelligence have on optimising processes.     

Hill非二次屈服准则次数m值识别及应变控制研究

正确确定本构模型中的物性参数是金属成形过程准确分析和模拟的基础。以Hill非二次屈服准则为基础,应用人工神经网络(ANN)技术建立了材料在变形过程中不同应力状态下物性参数m值的识别方法,并在MTS试验机上进行了薄壁管拉扭试验,通过试验中各阶段的实际应变增量值与m值识别前后计算所得理论应变增量值的比较,验证了识别所得m值以及根据识别所得m值进行应变控制的正确性。