Dynamic recurrent neural networks for a hybrid intelligent decision support system for the metallurgical industry

Knowledge-based modeling and implementation of the various manufacturing processes represent an intensive research area. It is known that it is difficult to analyze the mechanisms of many industrial production processes and build dynamic models by employing classical methods for intelligent systems in manufacturing. This paper describes how to use dynamic recurrent neural networks to provide the model base of a hybrid intelligent system for the metallurgical industry with a quality control model. The hybrid system extracts the features of image sequences obtained through the vision detection subsystem and employs a dynamic recurrent neural network to assess and predict the product qualities to further coordinate the entire production process.     

模糊神经网络及其在智能控制中的应用

文章介绍了处理不精确信息的模糊逻辑、模糊推理、模糊神经网络构造等方法，给出了一个在智能控制中的实例。     

A neural network approach to freeway network traffic control

The paper investigates the application of a feedforward neural network approach to freeway network control via variable direction recommendations at bifurcation locations. A nonlinear control problem is formulated and solved first by use of computationally expensive nonlinear optimization techniques. A feedforward neural network is then trained by optimally adjusting its weights so as to reproduce the optimal control law for a limited number of traffic scenarios. Generalisation properties of the neural network are investigated and a discussion of advantages and disadvantages compared with alternative control approaches is provided.     

基于神经网络滑模的智能车辆横向控制

以智能车辆为研究对象,针对车辆模型存在高度非线性动态特性、参数不确定性以及行驶时受外部干扰较多导致控制精度不高、鲁棒性差等问题,提出了采用径向基函数(RBF)神经网络滑模控制方法.建立2自由度线性车辆模型和自由度非线性整车模型,在传统2自由度车辆控制模型状态方程的基础上推导出新的状态方程并以此设计了相应控制器.利用李雅普诺夫(Lyapunov)稳定性理论推导出神经网络的权,并证明控制系统的稳定性.仿真结果表明:与传统的滑模控制方法相比,该方法控制精度高,有较强的鲁棒性.     

改进遗传神经网络控制混沌运动的研究

用最大Lyapunov指数构造遗传算法中的适应度函数,通过遗传算法优化神经网络的权系数．根据所得到的适应度函数和权系数来构造遗传神经网络控制器,从而提高神经网络控制效果．对离散系统Logistic映射和连续系统Rossler方程、AFM（原子力显微镜）悬臂梁振动系统的混沌运动分别进行了仿真控制．数值实验结果表明本文改进的遗传神经网络控制方法对离散或者连续的混沌系统都能控制到低周期轨道上去,证明了算法的有效性．     

基于动态神经网络集成的驾驶员行为学习算法*

特定交通环境下的驾驶员行为的变化是一个非线性的复杂系统,传统的驾驶决策模型和车辆行驶模型难以体现驾驶员的感知、判断、决策、动作等一系列心理、生理活动的不确定性和不一致性,而人工神经网络特别适合于因果间不易建立明确联系的问题。提出了基于神经网络集成的驾驶员行为学习算法DNNIA,该算法的有效性通过标准数据集和仿真实验得到了验证,学习到的诸如驾驶员踩踏踏板的习惯行为的仿真结果与采集的样本数据总体趋势较为一致,且实现了系统泛化性能的提高。     

级联RTAC系统动态神经网络辨识与分散镇定控制

针对含不确定关联项的级联RTAC系统的镇定控制问题, 提出了一种基于动态神经网络辨识的分散控制方 案. 应用拉格朗日方程建立起了考虑不确定非线性作用力的级联RTAC系统数学模型, 采用动态神经网络实现级 联RTAC系统中不确定关联项的在线辨识, 通过构造含神经网络权值矩阵迹的Lyapunov函数, 证明了辨识误差的一 致有界性. 通过动态神经网络辨识不确定关联项、补偿系统建模误差, 建立级联RTAC系统分层滑模控制算法, 以实 现级联RTAC系统的高精度分散镇定控制. 数值仿真验证了动态神经网络的引入对级联RTAC系统分散镇定控制系 统瞬态幅值抑制、稳态精度提升的效果.     

Dynamic analysis and intelligent control techniques for flexible manipulators: a review

This paper reviews literature on dynamic analysis and intelligent control techniques for flexible robot manipulators. First, a comparative dynamic analysis of flexible manipulators was presented and then control strategies were categorized and studied. Fuzzy logic, neural network, and genetic algorithm approaches were introduced and a range of contributions of such methods in flexible robot control were presented. A total of 115 papers were surveyed in this research, covering a sufficient depth in assessment of dynamic and control of flexible manipulator systems for the time span of 1970–2013.     

Event-based intelligent control of saturated chemical plant using an endomorphic neural network model

An event-based control system with an endomorphic neural network model is designed and realized to control a saturated non-linear plant. The scheme employed in this system is based on an event-based control paradigm previously proposed to control monotonic plants. However, this scheme is different from the previous one in that it can be used to control plants with saturation property. This new scheme may be viewed as a combined method of a time-based diagnosis mechanism in an event-based control system and a state-based control mechanism in a neural network control system. A chemical plant having strong non-linearity and complicated dynamics is controlled using this realized event-based control system. This paper discusses the structure of an event-based controller, the neural network modelling methodology, some related problems, and the simulation results.     

一类不确定系统的神经网络L2--增益鲁棒控制

对于一类具有三角结构的单输入单输出的不确定非线性系统, 用反步法(backstepping)和动态面控制方法(dynamic surface control technique)设计了一种使用神经网络补偿未知非线性的L2--增益鲁棒控制器. 控制器设计中没有直接解HJI(Hamilton-Jacobi-Isaac)不等式. 合理的选择了L2--增益性能指标, 将被控系统各个状态变量的跟踪误差和神经网络各权值的跟踪误差看作整个控制系统的各个状态变量, 并用Lyapunov定理和HJI不等式证明了使用提出的控制器后, 这些状态变量具有小于等于事先规定的正实数γ的L2--增益. 当系统的扰动信号为零向量时, 提出的控制器在原点是大范围渐近稳定的. 仿真研究结果表明所提出的控制器具有很好的跟踪性能和很强的鲁棒性.     

一类二阶时延神经网络的分岔分析和控制

讨论了一类二阶时延网络系统的非线性特性,应用线性化稳定性和分岔理论,提出了该系统从稳定到分岔的条件.结论指出利用延迟时间可以进行分岔控制、极限环幅值控制等,并给出了仿真的具体实例.     

基于径向基函数神经网络与扩张状态观测器的无人直升机控制

针对具有系统不确定和外部干扰的无人直升机飞行控制问题,提出了一种基于神经网络和扩张状态观测器的控制方法.利用神经网络逼近系统的不确定性,引入扩张状态观测器对神经网络的逼近误差和系统外部干扰进行估计.基于神经网络和扩张状态观测器的输出,对无人直升机的主旋翼挥舞角、姿态角速率、姿态角、速度与位置系统分别进行了控制器设计,以增强系统鲁棒性和抗干扰能力.同时,引入动态面控制方法以避免对虚拟信号进行直接求导,并通过李雅普诺夫方法分析了闭环控制系统的稳定性.最后使用无人直升机数据进行仿真验证,结果表明设计的控制律能使无人直升机有效跟踪控制指令,具有良好的稳定性与鲁棒性.     

考虑能见度影响的公路隧道照明动态优化与智能控制

为解决不同能见度影响下公路隧道实际路面亮度变化过大以及由此引起的行车安全与能源虚耗问题, 本文提出了一种能够改善公路隧道照明环境的动态优化与智能控制方法. 首先, 通过对不同时空条件下的公路隧道进行现场试验和数据分析, 得到了隧道内能见度的变化规律; 其次, 在公路隧道传统照明设计的基础上考虑能见度对照明环境的影响, 建立了基于隧道内能见度、交通量、车速、路面亮度和照明亮度的按需照明与动态优化模型;随后, 以不同地区公路隧道的实测数据为样本, 结合划分出的公路隧道典型照明场景和模糊径向基神经网络算法构建了公路隧道照明智能控制模型, 最后, 通过仿真实验验证了所构建模型的有效性, 其结果表明, 本文所提出的优化控制方法能够在保证隧道照明安全性的前提下兼顾节能性.     

改进粒子群算法的PID神经网络解耦控制

综合减摇控制系统存在非线性、多变量、强耦合等因素,会导致减摇系统达不到最佳控制状态。利用粒子群算法具有对整个空间进行高效搜索以及PID神经网络的自适应特点,提出一种改进粒子群算法,以解决粒子群算法中存在算法精度不高、粒子易陷入局部极小值等问题,并提高PID神经网络训练速度和训练精度,便于参数寻优。仿真结果表明,改进的粒子群算法具有一定优越性,将其运用到综合减摇控制系统解耦控制器设计中,能够有效地减小船舶横摇,达到较好的控制效果。     

非线性互联系统的分散混合自适应智能控制

针对一类高阶互联MIMO非线性系统,利用TS模糊系统和神经网络的通用逼近能力,在神经网络控制器中引入模糊基函数,提出一种分散混合自适应智能控制器设计的新方案．基于等价控制思想,设计分散自适应控制器,无需计算TS模型．通过对不确定项进行自适应估计,取消了其存在已知上界的假设．通过理论分析,证明了闭环智能控制系统所有信号有界,跟踪误差收敛到零．     

一类非线性系统的自适应神经网络控制

针对一类具有非仿射函数和下三角结构的、受干扰未知的非线性系统,提出一种新的自适应神经网络控制方法．它是严格反馈不确定系统和纯反馈系统的更一般化表达．在Backstepping设计思想基础上,证明了闭环信号的半全局最终一致有界性,并很好地处理了控制方向和控制奇异问题．通过仿真验证了该方法的有效性．     

规则推理与神经计算智能控制系统改进及比较

针对生产生活实践中的智能系统在实施控制过程中关键参数的实时在线智能整定与优化问题与需求,实现将不同类型人工智能方法与经典的控制方法对接从而构成多种复合控制（AI-CC）方法,提出改进算法并进行理论分析与仿真对比研究。首先实现了基于规则与模糊推理机制的AI-CC方法,提出了增量式改进算法,进而提出基于小波神经网络的AI-CC方法,进一步对两类智能系统的稳定性进行理论分析,提出稳定性保证算法,最后对比研究不同类型的智能系统在智能程度与性能特征方面的差异。研究成果为该领域研究者提供了多种改进的智能控制算法及其对比参照和理论分析,为该方法在工程实践中低成本地升级并稳定可靠地应用提供可操作方案。     

滑模动态递归模糊神经网络船电推进复合控制

提出了船舶电力吊舱推进系统的复合控制策略,以消除吊舱推进的过冲现象并获得快速平滑的动态响应.复合控制由鲁棒滑模控制和动态递归模糊神经网络控制组成,鲁棒滑模控制利用死区非线性和误差边界厚度法,克服系统的不确定与外界扰动,具有在线自学习算法的动态递归模糊神经网络控制促使系统的跟踪误差趋近于0.建立了基于SIMOTION的半实物仿真Siemens-Schottel推进器系统,仿真与实验结果表明,复合控制具有暂态快速和稳态平滑的动态响应,提高了吊舱推进系统的鲁棒性和运动精度.     

Intelligent scheduling and control of rail-guided vehicles and load/unload operations in a flexible manufacturing system

This paper presents a framework of intelligent manufacturing scheduling and control with specific applications to operations of rail-guided vehicle systems (RGVS). A RGVS control architecture is discussed with a focus on a simulated experiment in operations of the load/unload area of a real industrial flexible manufacturing system (FMS). In the operation stage of a material handling system (MHS), all shop floor data are subject to change as time goes. These data can be collected using a data acquisition device and stored in a dynamic database. The RGVS simulator used in this experimental study is designed to incorporate some possible situations representing existing material handling scenarios in order to evaluate alternative control policies. At the development stage of the controller, all possible combinations of most commonly encountered scenarios such as RGV failures, production schedule changes, machine breakdowns, and rush orders are to be simulated and corresponding results collected. The data are then structured into training data pairs to properly train an artificial neural network. The neural network, trained by using input/output data sets obtained from a number of simulation runs, will then provide control strategy recommendations. At the application stage, whenever an abnormal scenario occurs, a pre-processor will be activated to pre-screen and prepare an input vector for the trained neural network. If such an abnormal scenario falls outside the existing domain of data sets employed to train the neural network, as judged by the MHS supervisory controller, an off-line training module will be activated to eventually update the neural network. The recommended control strategies will be transmitted to the MHS control for real-time execution. If there is no further abnormal event detected, the dynamic data base (DDB) module simply continues to monitor the MHS activities. The proposed MHS control system combines the features of example based neural network technology and simulation modeling for true intelligent, on-line, pseudo real-time control. Not only will the system assure that feasible material handling control actions be taken, but also it will implement better control decisions through continuous learning from experiences captured as the operation time of the MHS accumulates.     

基于神经网络的铝电解混合控制模型研究*

铝电解优化控制生产过程的实质是将可控参数尽量控制在工艺要求的目标值范围内。传统的基于神经网络的控制方法或直接判断氧化铝浓度,缺乏对历史浓度的有效追踪,致使判断准确率下降;或对铝电解生产状况进行宏观识别,并调整,但缺乏实时性、及时性。针对上述问题,将神经网络和关联规则库、专家知识库、控制策略相结合,提出了一种新的氧化铝浓度识别及控制方法,从而将神经网络上升为一种混合控制模型HC-NN （hybrid control- neural network）。该模型以控制参数（自变量）为神经网络输入,和其输出共同作为