Multi-objective optimal design of online PID controllers using model predictive control based on the group method of data handling-type neural networks

In this paper, model predictive control (MPC) is used for optimal selection of proportional-integral-derivative (PID) controller gains. In conventional tuning methods a history of response error of the system under control in the passed time is measured and used to adjust PID parameters in order to improve the performance of the system in proceeding time. But MPC obviates this characteristic of classic PID. In fact MPC tries to tune the controller by predicting the system's behaviour some time steps ahead. In this way, PID parameters are adjusted before any real error occurs in the system's response. For this purpose, polynomial meta-models based on the evolved group method of data handling neural networks are obtained to simply simulate the time response of the dynamic system. Moreover, a non-dominated sorting genetic algorithm has been used in a multi-objective Pareto optimisation to select the parameters of the MPC which are prediction horizon, control horizon and relation of weight of Δ u and error, to minimise simultaneously two objective functions that are control effort and integral time absolute error of the system response. The results mentioned at the end obviously declare that the proposed method surpasses conventional tuning methods for PID controllers, and Pareto optimal selection of predictive parameters also improves the performance of the introduced method.     

Automatic tool state identification in a metal turning operation using MLP neural networks and multivariate process parameters

This paper describes results of the application of feed-forward Multi-Layer Perceptron (MLP) neural networks for cutting tool state identification in a metal turning operation. Test cuts were conducted using P25 carbide inserts with and without wear (i.e. nominally sharp) on EN24 alloy steel. The acquired data were used to train, cross-validate and test the generalisation capabilities of two MLP configurations. Both networks had exactly the same input and output nodes but differing number of nodes in a single middle layer. Training was achieved via back-propagation of error enhanced by the addition of a momentum term and adaptive learning rate. Different error goal targets during training of the MLP were used, and the validation results of the model investigation analysed and presented. Obtained results for successful classification of the tool state with respect to only two classes (worn or sharp) were between 83 and 96%.     

Feed-forward control of fast tool servo for real-time correction of spindle error in diamond turning of flat surfaces

A fast tool servo is designed and tested to obtain sub-micrometre form accuracy in diamond turning of flat surfaces. The thermal growth spindle error is compensated for real time using a fast tool servo driven by a piezoelectric actuator along with a capacitive displacement sensor. To overcome the inherent non-linearity of the piezoelectric actuator, Proportional Integral (PI) feedback control with a notch filter is implemented. Besides, feed-forward control based on a simple feed-forward predictor is added to achieve better tracking performance. Actual machining data are discussed in detail to prove that the proposed fast tool servo is capable of fabricating flat aluminum specimens of 100 mm in diameter to a form accuracy of 0.10 μm in peak-to-valley error value.     

带有摩擦前馈补偿的伺服控制器设计的研究

为了克服系统摩擦给系统带来的稳态误差和低速爬行问题，利用控制策略来补偿摩擦非线性对系统运动的负面影响是降低摩擦非线性负面影响的有效且节俭的途径。首先对交流伺服驱动的工作台进给系统进行了摩擦力测量实验，根据实验数据，建立了用于摩擦补偿控制的简化Stribeek摩擦力模型：由于XY工作台进给传动机构中存在的伺服滞后和摩擦是降低工作台位置跟踪精度两个主要因素，所以一个完整的前馈补偿方案应该包括两个部分：摩擦前馈补偿器和命令前馈补偿器。分别对命令前馈控制器和摩擦前馈补偿控制器进行了理论设计，借助于现有的GT-400运动控制器的功能，添加摩擦力补偿模块提升了工作台的跟踪精度，并和没有前馈补偿的传统控制器进行了对比研究，实验结果表明带有命令前馈和摩擦前馈的控制方案能取得更好的控制性能。     

Information theoretic competitive learning in self-adaptive multi-layered networks

In this paper, we propose self-adaptive multi-layered networks in which information in each processing layer is always maximized. Using these multi-layered networks, we can solve complex problems and discover salient features that single-layered networks fail to extract. In addition, this successive information maximization enables networks gradually to extract important features. We applied the new method to the Iris data problem, the vertical-horizontal lines detection problem, a phonological data analysis problem and a medical data problem. Experimental results confirmed that information can repeatedly be maximized in multi-layered networks and that the networks can extract features that cannot be detected by single-layered networks. In addition, features extracted in successive layers are cumulatively combined to detect more macroscopic features.     

双缸液压系统活塞运动轨迹同步模糊控制研究

针对负载不平衡的双液压缸同步运动控制问题,提出一种活塞运动轨迹同步模糊控制方法。该方法首先采用基于轨迹约束的控制器将两液压缸的同步误差限制在一定范围内,并使活塞杆运动平滑。在模糊控制的基础上,引入两个模糊控制器,分别用于消除液压缸的轨迹跟踪误差和液压缸之间的同步误差。实验结果表明:即使两液压缸存在较大的负载不平衡,该复合控制方法仍能将系统的同步运动误差控制在±10 mm以内,且系统运动具有很好的平滑性。     

龙门双驱系统转速跟踪同步耦合控制

针对龙门双驱系统的转速同步精度问题,提出一种转速跟踪同步耦合控制方法。为实现对外界负载的准确跟踪,构建基于Sigmoid函数的滑模负载观测器;设计结合积分滑模的双电机均值同步差控制器,结合跟踪误差反步控制器,制定转速同步误差与跟踪误差耦合控制策略,完成了两种负载情况下的仿真测试。仿真结果表明：相比PID控制,耦合控制策略的同步误差降低52.4%,响应时间降低75%;相比单一反步控制,耦合控制策略的同步误差降低39.1%,响应时间降低17%。     

基于误差相关性分析的复合伺服控制器设计

由于伺服系统控制精度对超精密机床性能具有很大影响,伺服系统控制器方案需要根据实际应用的需求进行设计;而位置跟踪误差是机床伺服系统的主要评价标准之一.本文根据实测数据,提出了基于相关性分析的位置伺服控制器改进方案,对跟踪误差与速度、加速度等进行相关性分析,设计了具有速度、加速度前馈补偿的复合控制器.通过对补偿前后系统响应的实验数据进行比较,可以看出进给伺服系统的控制精度显著提高,有效减少了跟踪误差.     

基于BP神经网络的铅黄铜超塑性能预测

运用BP神经网络方法建立了铅黄铜超塑性状态下伸长率与变形参数之间的预测模型,采用标准前馈式神经网络原理建立了铅黄铜超塑性拉伸试验参数与其伸长率之间的神经网络模型,以试验数据为样本,对所建模型进行训练,较好的预测了铅黄铜超塑拉伸的伸长率,最大的误差也只有4.81%.实现了不同变形工艺参数与伸长率之间的非线性映射,也为优化铅黄铜轴承保持架的超塑性成形参数提供理论和试验依据.伸长率预测值与试验结果吻合良好.     

Extracting dimensional information from steel reinforcing bars in concrete using neural networks trained on data from an inductive sensor

A new method is presented for extracting dimensional information from steel bars using images generated by an inductive sensor. The technique is based on the application of two feedforward backpropagation neural networks; one to estimate bar depth and the other to estimate bar diameter. Both of the networks have been trained on a set of data that consists of the peak parameters of six different bars scanned at 41 different bar depths. These input and target data must be pre-processed to obtain a good network generalisation. By testing the two networks with a completely different set of data, accurate performance has been obtained. Real, two-dimensional scan data have then been applied to both of the networks and the bar dimensional parameters have been extracted successfully. The advantage of the neural network method for extracting information is that it continues to operate reliably for very deep bars, for which the signal strength is severely attenuated and manifests a poor signal-to-noise ratio. Depth and diameter measurements have been obtained for bars located down to 58 mm, with errors that satisfy the requirements of the BS 1881 standard. At a depth of 40 mm, these measurements yield an error of ±4%, and this decreases as the depth reduces; in other words, the extracted bar diameter is within the requirements of the DIN 488 standard.     

Adaptive two-degree-of-freedom control of feed drive systems

Feed drive systems are widely used in industrial applications, and many efforts for improving their precision control have been made thus far. One of the basic approaches for improving the control accuracy of feed drive systems is to design a controller based on the internal model principle, which states that for a control system to track a reference signal without a steady state error, it needs to include a generator of the reference signal. Feed-forward controllers, such as the zero phase error tracking controller (ZPETC) proposed by Tomizuka, are also employed for improving control performance. However, prior knowledge of plant dynamics and/or reference signal properties is required for both the internal model principle and the feed-forward controller based designs. For precision control, plant dynamics should be identified in real time because feed drive dynamics are affected by varying conditions, such as frictional and thermal effects. This paper presents a new type of adaptive control for arbitrary reference tracking, which requires neither plant dynamics nor reference signal properties for controller design. This type of controller can also reduce the effect of unknown disturbances. The control system is designed using a discrete-time plant model and consists of adaptive feed-forward and feedback controllers. This design is then applied to a feed drive system with a ball screw drive. The effectiveness of the proposed design is demonstrated by simulation and experimental results, which was obtained by applying the proposed control system to an unknown reference signal whose property is varied during control.     

XY平台直线伺服系统的IP变增益交叉耦合控制

XY平台直线伺服系统中,负载扰动、机械时间延迟及各轴响应速度不同会影响轮廓加工精度,为此提出了一种将积分-比例(I-P)控制、速度前馈控制及变增益交叉耦合控制相结合的控制策略。单轴采用由IP控制和前馈控制构成的复合速度控制器,IP控制具有快速响应和抑制扰动的能力,速度前馈控制可增加系统跟踪能力,降低机械时间延迟效应。间接减小轮廓误差。并且,在X、Y轴间加入变增益交叉耦合控制器,直接减小轮廓误差。仿真结果表明该控制方案可增强系统的鲁棒性,提高系统的快速性和轮廓加工精度。     

A Distributed Memory Model of the Associative Boost in Semantic Priming

Evidence from priming studies indicates that both semantic and associative relations between pairs of words facilitate word recognition, and that pairs of words related in both ways (e.g. hammer-nail, cat-dog) produce an additional ‘associative boost’. We argue that while semantic priming may result from overlapping patterns of micro-features in a distributed memory model (e.g. Masson, 1991), associative priming is a result of frequent co-occurrence of words in the language. We describe a simple recurrent network, with distributed phonological and semantic representations, which is sensitive to the sequential occurrence of phonological patterns during training, and which produces associative facilitation of word recognition in a simulation of the priming task.     

Recognition of chatter with neural networks

Chatter deteriorates surface finish, reduces tool life, and damages machine tools. A chatter development prediction procedure is proposed for the cylindrical turning of long slender bars. The procedure uses two synthetically trained neural networks to recognize the harmonic acceleration signals and their frequency, and based on these observations, the future vibration characteristics of the system are estimated. The developed neural networks are capable of identifying 98% of the harmonic signals with over 90% certainty and estimate their frequencies with less than ±5% error from very short data sequences (only 11 sampled points). The accuracy of the neural networks is equivalent to time domain time series method based approaches; however, the proposed procedure can be implemented very quickly by using commercially available neural network hardware and software, and can use the new neural network chips to make the estimations very quickly by using parallel processors. The validity of the chatter prediction procedure is also demonstrated on the experimental data.     

基于ARIMA 和SVR 的滚动轴承状态预测方法研究

滚动轴承作为多种机械设备的关键零件,其运行状态的好坏往往影响着整机设备的运行状况,因此高精度的滚动轴承状态预测对整机设备的运行状态有着重要的意义。针对滚动轴承单一预测模型精度较差的问题,构建一种基于时间序列ARIMA和支持向量回归机SVR理论的组合预测模型。首先针对单一模型进行预测,应用误差平方和倒数法得到两种预测模型的权重结果,最终将该组合模型的预测结果分别与单一预测模型作比对分析。结果表明：该组合预测模型的预测误差均小于单一模型,具有较高的可靠性。     

Process Control Strategies for Dual-Phase Steel Manufacturing Using ANN and ANFIS

In this research, a comprehensive soft computational approach is presented for the analysis of the influencing parameters on manufacturing of dual-phase steels. A set of experimental data have been gathered to obtain the initial database used for the training and testing of both artificial neural networks (ANN) and adaptive neuro-fuzzy inference system (ANFIS). The parameters used in the strategy were intercritical annealing temperature, carbon content, and holding time which gives off martensite percentage as an output. A fraction of the data set was chosen to train both ANN and ANFIS, and the rest was put into practice to authenticate the act of the trained networks while seeing unseen data. To compare the obtained results, coefficient of determination and root mean squared error indexes were chosen. Using artificial intelligence methods, it is not necessary to consider and establish a preliminary mathematical model and formulate its affecting parameters on its definition. In conclusion, the martensite percentages corresponding to the manufacturing parameters can be determined prior to a production using these controlling algorithms. Although the results acquired from both ANN and ANFIS are very encouraging, the proposed ANFIS has enhanced performance over the ANN and takes better effect on cost-reduction profit.     

Thermal error measurement and modelling in machine tools. Part II. Hybrid Bayesian Network—support vector machine model

Prediction accuracy of machine tool thermal error significantly depends on the structure of the error model. Machine tool thermal error varies considerably depending upon the specific operating parameters adopted. Most error models developed thus far generally employ neural networks to map temperature data against thermal error. However, it is very important to account for the specific conditions as well within the model. This paper presents a hybrid Support Vector Machines (SVM)–Bayesian Network (BN) model that seeks to address this issue. The experimental data is first classified using a BN model with a rule-based system. Once the classification has been effected, the error is predicted using a SVM model. The hybrid thermal error model thus predicts the thermal error according to the specific operating conditions. This concept leads to a more generalised prediction model than the conventional method of directly mapping error and temperature irrespective of conditions. Such a model is especially useful in a production environment wherein the machine tools are subject to a variety of operating conditions.     

The Production of Finnish Nouns: A Psycholinguistically Motivated Connectionist Model

Connectionist modelling is a relatively new and valuable technique for the development of language processing theories. Most of language production models thus far have been discussed with respect to the English language. In this study, we present an interactive activation model of noun production for a morphologically complex language, viz. Finnish. Our model follows the psycholinguistic assumptions of the SAID model, an experimentally corroborated production model. The present model is a local connectionist model and it has been evaluated by computer simulation. It has been tested with respect to case and possessive suffixes of nouns. Furthermore, to test the psycholinguistic plausibility of the model, simulations of the most robust phonological frame constraint in speech errors, the initialness effect, were performed. These simulations showed good correspondence to empirical error data. We argue that the present model provides a basis for further computational research in morphologically rich languages like Finnish.     

Distributed computing and reliable communication in sensor networks using multi-agent systems

There is a growing demand for robust distributed computing and systems in sensor networks. Interaction between nodes is required to manage and distribute information. One common interaction model is the mobile agent. An agent approach provides stronger autonomy than a traditional object or remote-procedure-call based approach. Agents can decide for themselves, which actions are performed, and they are capable of flexible behaviour, reacting on the environment and other agents, providing some degree of robustness. The focus of the application scenario lies on sensor networks and low-power, resource-aware single System-On-Chip designs, i.e., for use in sensor-equipped technical structures and materials. We propose and compare two different data processing and communication architectures for the implementation of mobile agents in sensor networks consisting of single microchip low-resource nodes. Furthermore, a reliable smart communication protocol for incomplete and irregular networks are introduced. Two case studies show the suitability of agent-based approaches for distributed computing.     

基于多步预测的无线传感网络自适应采样技术研究

针对现有无线传感网节能算法计算量大、终端节点能耗高以及上位机数据更新实时性低的问题，提出一种基于多步预测的传感网络自适应采样算法。该算法在上位机和终端节点间建立自回归预测模型进行同步预测，同时通过比较预测模型的前向多步预测值与数据变化趋势拟合值，达到自适应改变采样步长的效果。为验证算法的节能性，基于ZigBee的船舶下水气囊气压监测系统平台进行实验。结果表明：所提算法在均方根误差为0.0892的情况下，比固定周期采样节能36.252%，比传统自适应通信算法节能26.912%，具有更好的能耗表现。