Neural network-based face detection

We present a neural network-based upright frontal face detection system. A retinally connected neural network examines small windows of an image and decides whether each window contains a face. The system arbitrates between multiple networks to improve performance over a single network. We present a straightforward procedure for aligning positive face examples for training. To collect negative examples, we use a bootstrap algorithm, which adds false detections into the training set as training progresses. This eliminates the difficult task of manually selecting nonface training examples, which must be chosen to span the entire space of nonface images. Simple heuristics, such as using the fact that faces rarely overlap in images, can further improve the accuracy. Comparisons with several other state-of-the-art face detection systems are presented, showing that our system has comparable performance in terms of detection and false-positive rates     

基于神经网络的链路预测算法

针对当前基于网络拓扑结构相似性的链路预测算法普遍存在精确度较低且适应性不强的问题,研究发现融合算法能够有效改善这些问题。提出了一种基于神经网络的融合链路预测算法,主要通过神经网络对不同链路预测相似性指标进行融合。该算法使用神经网络对不同相似性指标的数值特征进行学习,同时采用标准粒子群算法对神经网络进行了优化,并通过优化学习后的神经网络模型计算出融合指标。多个真实网络数据集上实验表明,该算法的预测精度明显高于融合之前的各项指标,并且优于现有融合方法的精度。     

Neural network-based EKG pattern recognition

The highly nonlinear chaotic nature of electrocardiogram (EKG) data represents a well-suited application of artificial neural networks (ANNs) for the detection of normal and abnormal heartbeats. Digitized EKG data were applied to a two-layer feed-forward neural network trained to distinguish between different types of heartbeat patterns. The Levenberg–Marquardt training algorithm was found to provide the best training results. In our study, the trained ANN correctly distinguished between normal heartbeats and premature ventricular contractions in 92% of the cases presented.     

Neural network-based simulation for response identification of two-storey shear building subject to earthquake motion

This article uses powerful technique of artificial neural network (ANN) models to simulate and estimate structural response of two-storey shear building by training the model for a particular earthquake. The neural network is first trained for a real earthquake data and the numerically generated responses of different floors of two-storey buildings as the training patterns. Trained ANN architecture is then used to simulate and test the structural response of different floors for various intensity earthquake data and it is found that the predicted responses given by ANN model are good for practical purposes. It is worth mentioning that although the simulation is done with numerically generated response data for particular earthquake, the idea may also be used for actual experimental (response) data.     

Modeling and simulation of a SMB chromatographic process designed for enantioseparation

This paper focuses on modeling of a simulated moving bed process (SMB) dedicated to the separation of racemic mixtures. In the first approach, a true moving bed model is derived, which assumes an equivalent counter-current movement of the solid phase. The good agreement between the model and the real system is demonstrated with experimental results. Then, a more rigorous approach is developed, which considers the system as an arrangement of static chromatographic columns and takes into account periodic switching. Attention is focused on model formulation and numerical solution techniques in order to develop efficient dynamic simulation programs.     

基于神经网络库存融合控制技术

神经网络控制系统通常会面临多种选择,如样本的训练方式、神经网络的算法等,不好的选择会降低预测率.BP(Back Propagation)神经网络库存控制系统融合多种库存控制技术,利用BP算法对学习的精度和收敛的速度进行改进,能比较精确地预测库存.讨论了有关BP神经网络的算法及算法改进等问题,以品牌服装库存控制为例,提出用神经网络的多层感知器实现库存融合控制.     

Neural network-based design of cellular manufacturing systems

A neural network based on a competitive learning rule, when trained with the part machine incidence matrix of a large number of parts, classifies the parts and machines into part families and machine cells, respectively. This classification compares well with the classical clustering techniques. The steady state values of the activations and interconnecting strengths enable easier identification of the part families, machine cells, overlapping parts and bottleneck machines. Neural networks are mostly applied by treating them as a blackbox, i.e. the interaction with the environment and the information acquisition and retrieval occurs at the input and the output level of the network. This paper presents an approach where knowledge is extracted from the external and internal structure of the neural network.     

基于Matlab的SMB色谱分离过程计算机仿真研究

针对考虑因素全面的SMB综合速率模型,采用有限元方法和正交配点法分别对柱向和吸附剂颗粒径向模型进行离散化,利用Matlab ODE求解器对SMB过程进行了数值求解,并编制了SMB过程仿真软件。在此基础上进行了一个SMB分离实例仿真,分析了切换时间和流量变化对分离性能的影响,验证了SMB过程分离性能的参数敏感性,提出了对系统实施先进控制的必要性。     

Neural network-based approximation of aircraft attainability boundary

The problem of the approximate dynamic system attainability domain boundary construction is considered. Results of neural network-based methods efficiency research for the highly-maneuverable aircraft attainability domain boundaries approximate construction are presented.     

Neural network-based calibration of electromagnetic tracking systems

Electromagnetic tracking systems are a common component of many virtual reality installations. Their accuracy, however, suffers from the distortions of the electromagnetic field used in calculating the tracker sensor’s position. We have developed a tracker calibration technique based on a neural network that effectively compensates for the errors in both tracked location and orientation. This case study discusses our implementation of the calibration algorithm and compares the results with traditional calibration methods.     

Neural network-based detection of virtual environment anomalies

The increasingly widespread use of large-scale 3D virtual environments has translated into an increasing effort required from designers, developers and testers. While considerable research has been conducted into assisting the design of virtual world content and mechanics, to date, only limited contributions have been made regarding the automatic testing of the underpinning graphics software and hardware. In the work presented in this paper, two novel neural network-based approaches are presented to predict the correct visualization of 3D content. Multilayer perceptrons and self-organizing maps are trained to learn the normal geometric and color appearance of objects from validated frames and then used to detect novel or anomalous renderings in new images. Our approach is general, for the appearance of the object is learned rather than explicitly represented. Experiments were conducted on a game engine to determine the applicability and effectiveness of our algorithms. The results show that the neural network technology can be effectively used to address the problem of automatic and reliable visual testing of 3D virtual environments.     

Neural network-based modelling of subsonic cavity flows

A fundamental problem in the applications involved with aerodynamic flows is the difficulty in finding a suitable dynamical model containing the most significant information pertaining to the physical system. Especially in the design of feedback control systems, a representative model is a necessary tool constraining the applicable forms of control laws. This article addresses the modelling problem by the use of feedforward neural networks (NNs). Shallow cavity flows at different Mach numbers are considered, and a single NN admitting the Mach number as one of the external inputs is demonstrated to be capable of predicting the floor pressures. Simulations and real time experiments have been presented to support the learning and generalization claims introduced by NN-based models.     

Neural network-based text location in color images

This paper proposes neural network-based text locations in complex color images. Texture information extracted on several color bands using neural networks is combined and corresponding text location algorithms are then developed. Text extraction filters can be automatically constructed using neural networks. Comparisons with other text location methods are presented; indicating that the proposed system has a better accuracy.     

Neural network-based model reference adaptive control system

In this paper, an approach to model reference adaptive control based on neural networks is proposed and analyzed for a class of first-order continuous-time nonlinear dynamical systems. The controller structure can employ either a radial basis function network or a feedforward neural network to compensate adaptively the nonlinearities in the plant. A stable controller-parameter adjustment mechanism, which is determined using the Lyapunov theory, is constructed using a sigma-modification-type updating law. The evaluation of control error in terms of the neural network learning error is performed. That is, the control error converges asymptotically to a neighborhood of zero, whose size is evaluated and depends on the approximation error of the neural network. In the design and analysis of neural network-based control systems, it is important to take into account the neural network learning error and its influence on the control error of the plant. Simulation results showing the feasibility and performance of the proposed approach are given.     

Efficient moving horizon state and parameter estimation for SMB processes

In this paper, a moving horizon state and parameter estimation scheme for chromatographic simulated moving bed SMB processes is proposed. The simultaneous state and parameter estimation is based on a high-order nonlinear SMB model which incorporates rigorous models of the chromatographic columns and the discrete shiftings of the inlet and outlet ports. The estimation is performed using sparse measurement information: the concentrations of the components are only measured at the two outlet ports (which are periodically switched from one column to the next) and at one fixed location between two columns. The goal is to reconstruct the full state of the system, i.e. the concentration profiles along all columns, and to identify critical model parameters reliably such that the estimated model can be used in the context of online optimizing control. The state estimation scheme is based upon a deterministic model within the prediction horizon, state noise is only present in the state and the parameters prior to and at the beginning of the horizon. By solving the optimization problem with a multiple-shooting method and applying a real-time iteration scheme, the computation times are such that the scheme can be applied online. Numerical simulations of a validated model for a separation problem with nonlinear isotherms of the Langmuir type demonstrate the efficiency of the algorithm.     

Neural network-based sliding mode control of electronic throttle

A neural network-based sliding mode controller for an electronic throttle of an internal combustion engine is proposed. Electronic throttle is modeled as a linear system with uncertainties and affected by disturbances depending on the states of the system. The disturbances, consisting of an unknown friction and a torque caused by the dual spring mechanism inside the mechanical part of the throttle, are estimated by a neural network whose parameters are adapted on-line. The sliding mode controller and the parameters adaptation scheme are derived in order to achieve a tracking of a smooth reference signal, while preserving boundedness of all signals in the closed-loop system. Experimental results are presented which demonstrate the efficiency and robustness of the proposed control scheme.     

Neural network-based analog fault diagnosis using testability analysis

A fault diagnosis procedure for analog linear circuits is presented. It uses an off-line trained neural network as a classifier. The innovative aspect of the proposed approach is the way the information provided by testability and ambiguity group determination is exploited when choosing the neural network architecture. The effectiveness of the proposed approach is shown by comparing with similar work that has already appeared in the literature.     

SMB色谱模型数值计算的时空守恒元解元方法

为了探寻出一种求解SMB色谱模型的快速数值求解方法,并试图通过比较得出时空守恒元/解元(CE/SE)方法确实是快速数值求解方法,因而采用该方法对SMB色谱模型进行数值求解,并在数值方法的计算效率和精确度两个方面与有限差分法和正交配置有限元法进行了比较,最终得出了CE/SE方法是具有高计算效率和高精确度特性的快速数值求解方法.通过两个实例的模拟仿真,结果表明了该方法在高计算效率和高精确度方面的优越性.     

Neural network-based simulation metamodels for predicting probability distributions

Simulation is an important tool for supporting decision-making under uncertainty, particularly when the system under consideration is too complex to evaluate analytically. The amount of time required to generate large numbers of simulation replications can be prohibitive, however, necessitating the use of a simulation metamodel in order to describe the behavior of the system under new conditions. The purpose of this study is to examine the use of neural network metamodels for representing output distributions from a stochastic simulation model. A series of tests on a well-known simulation problem demonstrate the ability of the neural networks to capture the behavior of the underlying systems and to represent the inherent uncertainty with a reasonable degree of accuracy.