基于BP神经网络算法的柔性制造系统故障诊断

在分析和研究人工神经网络的前提下,将BP神经网络作为故障诊断的技术指导,以本校柔性制造系统中的电源管理模块故障诊断为实例,构建其智能诊断系统。通过3层BP神经网络,将输入端传感器收集到的特征量经中间层信息处理和输出层的进一步处理,使其达到网络性能目标,最终得到诊断数据。比较诊断数据与标准故障集的类型,获得准确的结果。结果表明:利用BP神经网络可以构建一个智能诊断系统,实现故障准确快速自动识别。     

基于触摸屏的柔性生产线物流运输控制系统的设计

基于柔性生产线中立体仓库和运输小车的结构及其可视化控制的要求,设计并实现了功能强大、界面友好的控制系统。控制系统硬件主要基于西门子SIMATICs7200系列可编程逻辑控制器（programmable logic controller, PLC ）,同时引入人机界面友好的触摸屏eview—MT5000;软件分别选用西门子Step7编程软件和EV5000触摸屏编程软件。控制系统经实验室模拟柔性生产线测试,结果表明设计的控制系统运行稳定可靠,实现了对立体仓库系统和运输小车的实时监视和控制。     

P-FMS小型柔性制造系统的研制与开发

本文简要介绍了一种用于教学的小型柔性制造系统 ,简述了控制系统的组成、硬件设备以及软件环境。     

基于FPGA的气动柔性手指的神经网络控制器设计

研究了气动柔性手指智能控制策略的硬件实现，设计了气动柔性手指的神经网络控制器。气动柔性手指神经网络控制器是在已经对气动柔性手指进行运动学和动力学分析的前提下，使用MATLAB对神经网络进行试训，依据训练所得的权值和阈值，在现场可编程门阵列集成环境下，基于超高速集成电路硬件描述语言完成的。该控制器适用于具有复杂智能控制策略且对实时性要求高的多手指控制系统。     

基于PLC回转型自动化控制生产线仿真示教设备的设计与实现

良好的教学效果来自于教学的直观性。复杂的自动化控制生产线教学更需要直观的仿真示教设备。本文设计、介绍一种基于PLC控制的回转型自动化控制生产线仿真示教设备，从其结构、原理到PLC控制的实现作了详细讲析。     

基于神经网络控制算法的气动伺服系统运动分析研究

提出了一种神经网络控制方法并通过对气动伺服系统的无杆气缸运动控制,探究此控制方法的控制精度。由于受空气可压缩性、摩擦力以及启动系统的扰动等非线性因素的影响,气动伺服系统很难去建立精确的数学模型。根据系统的非线性特点及PID控制不足,基于BP神经网络控制,设计神经网络PID控制器,并进行实验。通过实验,对无杆气缸的运动特性分析,表明这种控制策略可以更好控制气动伺服系统的运动精度。     

工艺设计系统的人工神经网络控制

介绍了应用BP网络实现工艺设计的思想及方法。应用表明,工艺设计系统的学习能力强且通用性好。     

网络化协同设计制造中的柔性工作流管理研究

针对典型区域性网络化协同设计制造平台及其虚拟社区的管理机制进行了研究,提出了基于柔性工作流和数据库技术的网络化协同管理方案,结合实例分析,介绍了网络化协同设计制造过程中的柔性工作流定义、数据库操作.对于网络化协同虚拟社区中典型的链状合作型和排斥竞争型企业分别实现了不同的应用模式,建立了基于数据库的协同柔性工作流模板库,并对协同设计制造柔性工作流中的权限管理和冲突协调提出了初步的解决方法,能够有效地自动控制典型的区域性的网络化协同设计制造平台的工作流程和进行资源管理、状态监控,对于提高网络化协同设计制造的工作效率和自动化有一定的应用价值.     

神经网络自适应控制及其在电液伺服系统中的仿真研究

针对非线性系统控制和神经网络控制实时性差的问题,提出一种神经网络在线控制结构,做到了对象模型在线辩识和控制器在线设计。通过仿真研究,证明该方法有效。     

Representing the Material Handling System in Tractable Queuing Network Models of Flexible Manufacturing Systems: A Workload Data Aggregation Approach

This paper deals with the use of queuing network (QN) models for quantitatively evaluating the impact of the material handling system (MHS) on the steady-state performance of a flexible manufacturing system (FMS) at the strategic and tactical decision levels. A direct exploitation of the workload data provided by industrial experts often results in QN models that cannot be analyzed efficiently because of the prohibitive number of customer classes. In this paper, we propose a systematic data aggregation approach for deriving the aggregated characteristics of the service offered by a device of the MHS at steady state. This generic aggregation scheme explicitly captures empty trips on the device, MHS devices that have different motions depending on whether they travel empty or loaded, and enables further, consistent use of the various central server QN models appearing in the literature. The quality of the estimates provided by this automated data aggregation approach is tested on several examples, and their integration into a QN model for performance evaluation is illustrated on an FMS presented in the literature.     

The Effects of Scheduling Rules and Routing Flexibility on the Performance of a Random Flexible Manufacturing System

The increased use of flexible manufacturing systems to efficiently provide customers with diversified products has created a significant set of operational challenges for managers. Many issues concerning procedures and policies for the day-to-day operation of these systems still are unresolved. Previous studies in this area have concentrated on various problems by isolating or simplifying the systems under study. The primary objective of this study is to extend previous research by examining the effects of scheduling rules and routing flexibility on the performance of a constrained, random flexible manufacturing system (FMS). Other experimental factors considered are shop load, shop configuration, and system breakdowns. Within the bounds of this experiment, the results indicate that, in the presence of total routing flexibility, the effects of shop load, system breakdowns, and scheduling rules are significantly dampened. In particular, when total routing flexibility exists, the choice of scheduling rules is not critical. We also show that the behavior of scheduling rules in a more constrained FMS environment (i.e., where system breakdowns occur and material handling capability is limited) is consistent with the findings of previous research conducted under less constrained environments. Finally, results indicate that the shop configuration factor has little or no impact on a system's flow-time performance.     

重新定義系统输出的柔性櫟械臂控制器的設計

通过重新定义系统的输出,将柔性机械臂的模型分解为输入输出和零动态两个子系统;对于输入输出子系统采用模糊控制,对于零动态子系统采用作图的方法分析;找到了特征值和重新定义参数之间的关系,并将其用于控制器的设计,可以使得零动态子系统快速渐近稳定于平衡点,从而保证整个系统渐近稳定。     

基于FMS的立体仓储堆垛机控制系统研究与设计

针对FMS多品种变批量的加工特点,首先阐述了单柱巷道式堆垛机的机构及组成,通过路径规划与库区分配优化、货位分配优化和行驶时间优化,实现堆垛机快速准确移动。针对堆垛机的运行策略,分别设计和实现了堆垛机单机运行、在仓储管理系统中的运行和FMS联机运行。系统运行结果表明,该设计方案合理、可靠、高效。     

华中网络数控系统的开发与应用

网络数控是各种先进制造技术的基本单元,它为各种先进制造环境的研究与实现提供最基本的技术支持。为此,武汉华中数控系统有限公司利用进程的面向对象Ｐｅｔｒｉ网理论对网络环境下的开放式数控系统体系结构进行研究分析,并在ＤＯＳ／ＷＩＮ平台下初步实现了开放式网络化数控系统。该系统能支持网络制造、远程制造、诊断与维护,且利于对已有设备进行快速网络数控改造,是适合我国国情的网络数控系统。     

The Application and Evaluation of Banker's Algorithm for Deadlock-Free Buffer Space Allocation in Flexible Manufacturing Systems

Deadlock-free operation is essential for operating highly automated manufacturing systems. The seminal deadlock avoidance procedure, Banker's algorithm, was developed for computer operating systems, an environment where very little information regarding the future resource requirements of executing processes is known. Manufacturing researchers have tended to dismiss Banker's algorithm as too conservative in the manufacturing environment where future resource requirements are well defined by part routes. In this work, we investigate this issue by developing variants of Banker's algorithm applicable to buffer space allocation in flexible manufacturing. We show that these algorithms are not overly conservative and that, indeed, Banker's approach can provide very good operational flexibility when properly applied to the manufacturing environment.     

基于神经网络的卷烟共线分拣系统自适应控制

针对目前方法自适应控制卷烟共线系统时,由于未能依据Lyapunov函数确定系统的控制规律,导致在实施系统自适应控制时,存在控制效果差、控制误差高和控制性能低的问题,提出基于神经网络的卷烟共线分拣系统自适应控制方法。首先依据Lyapunov函数确定系统的控制规律,建立对象控制模型并使用前馈神经网络训练模型,优化控制器参数,完成控制器的设计;再利用控制器的参数建立线性和非线性2种自适应控制方法;最后通过制定的切换规则,完成自适应算法的平滑转换,实现系统的自适应控制。实验结果表明,运用该方法控制系统时,控制效果好、控制误差低以及控制性能高。     

采用神经网络与模糊控制的制动需求识别

以保证车辆的制动安全性为最终目的,基于制动推杆行程传感器信号,利用神经网络算法与模糊控制算法提出了对制动需求进行识别的方法。首先采用神经网络算法与模糊控制算法完成对驾驶员制动需求的初步判断,然后通过主缸压力反馈信号对制动系统的制动动作完成程度进行预测,进而对制动需求识别结果进行持续反馈跟踪修正,以保证车辆的制动系统能够实现预期的制动效能。通过MATLAB/Simulink仿真试验与台架试验对控制算法进行了对比与验证,结果表明该方法能迅速准确地识别驾驶员的制动需求,从而为车辆的制动安全性提供保障。     

An Analytical Queueing Network Model for Flexible Manufacturing Systems with a Discrete Handling Device and Transfer Blockings

In this paper, we study job shop-like flexible manufacturing systems (FMSs) with a discrete material handling system (MHS). In such FMSs, the MHS is a critical device, the unavailability of which may induce transfer blockings of the machines. The FMS devices therefore are hierarchically structured into primary and secondary devices to manage such blocking and avoid deadlocks in these FMSs.For evaluating the quantitative steady-state performance of such FMSs, we propose an analytical queueing network model that relies on an approximate method proposed for analyzing computer systems with simultaneous possessions of resources. Such a model is obtained using the concept of passive resources and by aggregating the FMS workload data so that models are much more tractable.The analytical results are validated against discrete event simulation and shown to be very encouraging. We also show how to increase their robustness, especially under light workload conditions, by modifying an assumption of the method concerning service time distributions.     

Adaptive Control of Multivariable Process Using Recurrent Neural Networks

Abstract

Industrial processes are naturally multivariable in nature, which also exhibit non-linear behavior and complex dynamic properties. The multivariable four-tank system has attracted recent attention, as it illustrates many concepts in multivariable control, particularly interaction, transmission zero, and non-minimum phase characteristics that emerge from a simple cascade of tanks. So, the multivariable laboratory process of four interconnected water tanks is considered for modeling and control. For processes which show nonlinear and multivariable characteristics, classical control strategies like PIDs have performance limitations. Hence, intelligent approaches like Neural Networks (NN) is an important term in this juncture. The use of Recurrent Neural Network (RNN) is apt for modeling and control of nonlinear dynamic processes as it contains the past information about the process. The objective of the current study is to design and implement an adaptive control system using RNN for a nonlinear multivariable process.

The proposed adaptive design comprises an estimator based on RNN, which adapts online and predicts one step ahead output. A Recursive Least Square (RLS) based back propagation algorithm is used for training the network. The controller used is also a RNN, which minimizes the difference between the predicted output and reference trajectory. The objective function is minimized using a steepest descent algorithm which gives the optimum control input. Desired performance of the system is ensured by the parallel operation of both. The proposed control strategy is implemented in a laboratory scale four tank system. The trajectory tracking and disturbance rejection response obtained are compared with the response obtained by using a well designed decoupled, decentralized IMC controller.     

直流电动机PID神经网络双闭环控制系统

本文在分析传统直流电动机双闭环控制系统性能特点的基础上,利用一种新的神经网络——PID神经网络实施了自学习双闭环控制系统,本文给出了PID神经网络网络的前向和反传算法,给出了实时仿真结果,证明该控制系统具有较好的控制性能。