基于RBF神经网络多步预测的自适应PID控制

提出一种基于RBF神经网络多步预测的自适应PID控制算法．该算法用无局部极小的径向基函数网络对非线性系统进行在线辨识，利用多步预测误差对PID型控制器网络进行训练，从而实现PID参数的在线自适应寻优．通过对典型非线性系统的仿真研究，该控制系统具有较强的适应性和鲁棒性．     

Research on Chaotic Time Series Prediction Based on K-entropy and RBF Neural Networks

In this paper, a method of direct multi-step prediction of chaotic time series is proposed, which is based on Kolmogorov entropy and radial basis functions neural networks. This is done first by reconstructing a phase space using chaotic time series, then using K-entropy as a quantitative parameter to obtain the maximum predictability time of chaotic time series, finally the predicted chaotic time series data can be acquired by using RBFNN. The application considered is Lorenz system. Simulation results for direct multi-step prediction method are compared with recurrence multi-step prediction method. The results indicate that the direct multi-step prediction is more accurate and rapid than the recurrence multi-step prediction within the maximum predictability time of chaotic time series. So, it is convenient to forecast and control with real time using the method of direct multi-step prediction.     

冷冲压模具质量综合评价体系的建立

针对冷冲压模具生产的特点，提出了模具质量综合评价体系和多层次递进修正指标体系；参照模具工艺和设计规范确定了评价标准，采用层次递进和模糊综合法进行了科学评价，实际应用表明该体系能够对模具质量做出客观、公正、公平的评价。     

板材冲裁和翻口工艺特征识别及图形转换的研究

该文针对冲孔、翻孔两类板材冲压成形工艺方法的主要特点，设计了两类成形件的特征识别规则．根据形状特征与冲压加工特征间存在着单映射的关系，可以充分利用模型中的几何信息和拓扑关系来识别零件的特征．因板材零件的厚度与整个零件尺寸比较是小的，可以假设零件厚度为常数．在消除厚度边和厚度边组成的面后，其余边和面分成上、下两组．两组的几何边界和拓扑结构是相同的，上、下两面的法向矢量方向相反．通过对读入的IGES三维线框文件进行预处理，使之满足特征识别的需要，采用基于几何推理的特征识别技术实现了对冲压件主型面特征、冲孔特征、翻孔特征的识别与提取，从而为后期冲压工艺方案的制定做准备．     

捣固焦经济型配煤的应用

结合太化焦化厂炼焦煤现状,科学制定了配煤炼焦实验方案,同时扩大了炼焦煤资源,实现了良好的经济和环保效益。     

Comparative analysis by simulating and testing pole impact for hot stamping front crossbeam of automobile

Hot stamping 22MnB5 steel plate with ultra-high strength has been widely used for body structural members in consideration of automobile safety and lightweight.This paper presents a verification program of simulating and testing pole impact in order to verify if the front crossbeam reinforcement assembly can meet the design requirements,reduce the overall vehicle simulation and test cost and shorten the operation period.In the same condition,the simulation proved conforming to the design requirements;however,the bumper cracked at the impacting point in the course of pole impact test.The analysis of the crack by optical microscope,Vickers and scanning electron microscope indicates that mixture of ferrite,bainite and martensite was produced in the weld heat-affected zone of the 340/590DP tow hook holder and the 22MnB5 front crossbeam;therefore,their hardness and mechanical property were reduced obviously,so that they fractured when impacted.No welding process with continuous weld but spot welding or other bonding method may be employed for the reason that the capacity of the 22MnB5 steel plate in the weld heat-affected zone was reduced obviously.     

地连墙变形的神经网络多步预测研究

结合润扬长江公路大桥南汊北锚碇深基坑工程 ,提出并应用神经网络多步预测方法来研究地连墙施工变形的预测问题。系统介绍了基于时间窗口的神经网络多步滚动预测技术 ,并详细讨论了输入输出层的设计、隐层神经元数以及预测时间步长等一些基本预测技术问题。该预测方法应用于润扬长江公路大桥南汊北锚碇深基坑围护工程 ,取得了较好的工程效果。     

国产轿车用冷轧钢板冲压级别评定

讨论了ＧＢ５２１３－８５在钢板冲压级别评定指标方面的问题,并对钢板的基本成形性和模拟成形的相关性进行了分析,在轿车冲压件成形分类的基础上,提出了新的冲压级别评定方法。     

基于AOS树表达的冲压工艺设计

提出了冲压元的概念,阐述了ＡＯＳ树表示方法,提出利用冲压元与ＡＯＳ树建立冲压工芤模型。首先描述冲压件各特征的冲压工艺ＡＯＳ树表达,然后以此为基础描述特征组合的冲压工艺ＡＯＳ树表达,该描述方式符合工艺设计分阶段、分层次、从局部到全局的规律。最后以一具体的冲压件实例说明ＡＯＳ树表示方法。     

轨道车辆S形截面结构件的排样设计与应用

轨道车辆S形截面结构件冲压成形容易产生回弹变形,回弹变形又会带来制件上的孔洞和局部形状的二次形位误差。以轨道车辆S形截面结构件为对象,参照坯料的排样设计流程,通过制件工艺分析、CAE仿真分析,初步设计了坯料的十工序排样;从成形工艺、模具概要设计角度详细分析了坯料排样各工序的设计目的;依据所设计的坯料排样完成了级进模设计,并在所设计、制造、调试好的级进模上做了实冲试验。结果表明：所设计的十工序排样图适用于轨道车辆S形截面结构件的冲压加工。对该坯料排样设计方法与方案推广应用,具有较好的经济效益和社会效益。