An alternative approach to solve convergence problems in the backpropagation algorithm

The multilayer perceptron network has become one of the most used in the solution of a wide variety of problems. The training process is based on the supervised method where the inputs are presented to the neural network and the output is compared with a desired value. However, the algorithm presents convergence problems when the desired output of the network has small slope in the discrete time samples or the output is a quasi-constant value. The proposal of this paper is presenting an alternative approach to solve this convergence problem with a pre-conditioning method of the desired output data set before the training process and a post-conditioning when the generalization results are obtained. Simulations results are presented in order to validate the proposed approach.

Comparative study between powers of sigmoid functions, MLP-backpropagation and polynomials in function approximation problems

Function approximation is a very important task in environments where the computation has to be based on extracting information from data samples in real world processes. So, the development of new mathematical model is a very important activity to guarantee the evolution of the function approximation area. In this sense, we will present the Polynomials Powers of Sigmoid (PPS) as a linear neural network. In this paper, we will introduce one series of practical results for the Polynomials Powers of Sigmoid, where we will show some advantages of the use of the powers of sigmiod functions in relationship the traditional MLP-Backpropagation and Polynomials in functions approximation problems.

Implementation of a computerized assessment system by using backpropagation neural networks with R and shiny

Las dificultades a las que los estudiantes se enfrentan y su lucha por dominar los temas, podría aumentar como consecuencia de la inadecuada utilización de materiales de evaluación. Generalmente se encuentran en el aula alumnos que hacen buen uso del material de los cursos y de una manera rápida, mientras que otros presentan dificultades con el aprendizaje del material. Esta situación es fácilmente visto en los resultados de los exámenes, un grupo de estudiantes podrían obtener buenas calificaciones animándoles, mientras que otros obtendrían la mala percepción de que los temas son difíciles, y en algunos casos, obligándolos a abandonar el curso o en otros casos a cambiar de carrera. Creemos que mediante el uso de técnicas de aprendizaje automático, y en nuestro caso la utilización de redes neuronales, sería factible crear un entorno de evaluación que podrían ajustarse a las necesidades de cada estudiante. Esto último disminuiría la sensación de insatisfacción de los alumnos y el abandono de los cursos.

A direct adaptive neural command controller design for an unstable helicopter

This paper presents an off-line (finite time interval) and on-line learning direct adaptive neural controller for an unstable helicopter. The neural controller is designed to track pitch rate command signal generated using the reference model. A helicopter having a soft inplane four-bladed hingeless main rotor and a four-bladed tail rotor with conventional mechanical controls is used for the simulation studies. For the simulation study, a linearized helicopter model at different straight and level flight conditions is considered. A neural network with a linear filter architecture trained using backpropagation through time is used to approximate the control law. The controller network parameters are adapted using updated rules Lyapunov synthesis. The off-line trained (for finite time interval) network provides the necessary stability and tracking performance. The on-line learning is used to adapt the network under varying flight conditions. The on-line learning ability is demonstrated through parameter uncertainties. The performance of the proposed direct adaptive neural controller (DANC) is compared with feedback error learning neural controller (FENC).

Filter design using radial basis function neural network and genetic algorithm for improved operational health monitoring

The problem of denoising damage indicator signals for improved operational health monitoring of systems is addressed by applying soft computing methods to design filters. Since measured data in operational settings is contaminated with noise and outliers, pattern recognition algorithms for fault detection and isolation can give false alarms. A direct approach to improving the fault detection and isolation is to remove noise and outliers from time series of measured data or damage indicators before performing fault detection and isolation. Many popular signal-processing approaches do not work well with damage indicator signals, which can contain sudden changes due to abrupt faults and non-Gaussian outliers. Signal-processing algorithms based on radial basis function (RBF) neural network and weighted recursive median (WRM) filters are explored for denoising simulated time series. The RBF neural network filter is developed using a K-means clustering algorithm and is much less computationally expensive to develop than feedforward neural networks trained using backpropagation. The nonlinear multimodal integer-programming problem of selecting optimal integer weights of the WRM filter is solved using genetic algorithm. Numerical results are obtained for helicopter rotor structural damage indicators based on simulated frequencies. Test signals consider low order polynomial growth of damage indicators with time to simulate gradual or incipient faults and step changes in the signal to simulate abrupt faults. Noise and outliers are added to the test signals. The WRM and RBF filters result in a noise reduction of 54 - 71 and 59 - 73% for the test signals considered in this study, respectively. Their performance is much better than the moving average FIR filter, which causes significant feature distortion and has poor outlier removal capabilities and shows the potential of soft computing methods for specific signal-processing applications.

Filter design using radial basis function neural network and genetic algorithm for improved operational health monitoring

The problem of denoising damage indicator signals for improved operational health monitoring of systems is addressed by applying soft computing methods to design filters. Since measured data in operational settings is contaminated with noise and outliers, pattern recognition algorithms for fault detection and isolation can give false alarms. A direct approach to improving the fault detection and isolation is to remove noise and outliers from time series of measured data or damage indicators before performing fault detection and isolation. Many popular signal-processing approaches do not work well with damage indicator signals, which can contain sudden changes due to abrupt faults and non-Gaussian outliers. Signal-processing algorithms based on radial basis function (RBF) neural network and weighted recursive median (WRM) filters are explored for denoising simulated time series. The RBF neural network filter is developed using a K-means clustering algorithm and is much less computationally expensive to develop than feedforward neural networks trained using backpropagation. The nonlinear multimodal integer-programming problem of selecting optimal integer weights of the WRM filter is solved using genetic algorithm. Numerical results are obtained for helicopter rotor structural damage indicators based on simulated frequencies. Test signals consider low order polynomial growth of damage indicators with time to simulate gradual or incipient faults and step changes in the signal to simulate abrupt faults. Noise and outliers are added to the test signals. The WRM and RBF filters result in a noise reduction of 54 - 71 and 59 - 73% for the test signals considered in this study, respectively. Their performance is much better than the moving average FIR filter, which causes significant feature distortion and has poor outlier removal capabilities and shows the potential of soft computing methods for specific signal-processing applications. (C) 2005 Elsevier B. V. All rights reserved.

Machine learning modelling for predicting soil liquefaction susceptibility

This study describes two machine learning techniques applied to predict liquefaction susceptibility of soil based on the standard penetration test (SPT) data from the 1999 Chi-Chi, Taiwan earthquake. The first machine learning technique which uses Artificial Neural Network (ANN) based on multi-layer perceptions (MLP) that are trained with Levenberg-Marquardt backpropagation algorithm. The second machine learning technique uses the Support Vector machine (SVM) that is firmly based on the theory of statistical learning theory, uses classification technique. ANN and SVM have been developed to predict liquefaction susceptibility using corrected SPT (N-1)(60)] and cyclic stress ratio (CSR). Further, an attempt has been made to simplify the models, requiring only the two parameters (N-1)(60) and peck ground acceleration (a(max)/g)], for the prediction of liquefaction susceptibility. The developed ANN and SVM models have also been applied to different case histories available globally. The paper also highlights the capability of the SVM over the ANN models.

Performance evaluation of neural network algorithms for multisensor data fusion in an airborne track while scan radar

This paper deals with the solution to the problem of multisensor data fusion for a single target scenario as detected by an airborne track-while-scan radar. The details of a neural network implementation, various training algorithms based on standard backpropagation, and the results of training and testing the neural network are presented. The promising capabilities of RPROP algorithm for multisensor data fusion for various parameters are shown in comparison to other adaptive techniques

Virtual Sensors for On-line Wheel Wear and Part Roughness Measurement in the Grinding Process

Grinding is an advanced machining process for the manufacturing of valuable complex and accurate parts for high added value sectors such as aerospace, wind generation, etc. Due to the extremely severe conditions inside grinding machines, critical process variables such as part surface finish or grinding wheel wear cannot be easily and cheaply measured on-line. In this paper a virtual sensor for on-line monitoring of those variables is presented. The sensor is based on the modelling ability of Artificial Neural Networks (ANNs) for stochastic and non-linear processes such as grinding; the selected architecture is the Layer-Recurrent neural network. The sensor makes use of the relation between the variables to be measured and power consumption in the wheel spindle, which can be easily measured. A sensor calibration methodology is presented, and the levels of error that can be expected are discussed. Validation of the new sensor is carried out by comparing the sensor's results with actual measurements carried out in an industrial grinding machine. Results show excellent estimation performance for both wheel wear and surface roughness. In the case of wheel wear, the absolute error is within the range of microns (average value 32 mu m). In the case of surface finish, the absolute error is well below R-a 1 mu m (average value 0.32 mu m). The present approach can be easily generalized to other grinding operations.

单相机全向立体视觉研究

折反射全向成像系统是由普通透视相机和反射镜面组成的全向成像装置，可实时获取360°无需拼接的全景图像，近年来已成为研究热点并在视频会议、三维重建和移动机器人导航等领域有着广泛的应用。 本文主要对单相机全向立体视觉系统的设计、标定、匹配以及三维重建展开研究。介绍了一种可实时获取全向三维信息的折反射全向立体视觉光学装置OSVOD（Omnidirectional Stereo Vision Optical Device），OSVOD由两个双曲面镜和一个普通透视相机组成。其中两个双曲镜面上下同轴、间隔一定距离固定在一个玻璃筒内，下镜面中间开有一孔，上镜面通过下镜面的孔在相机像平面上成像，这样空间一点经上下反射镜的反射在像平面上有两个像点，用一个相机实现了立体视觉。两镜面的共同轴和相机镜头的光轴共线，共同焦点和镜头的光心重合，该配置能保证系统满足单一视点约束SVP（Single View-Point）。本结构配置也使系统的外极线呈一系列的放射线，对应点匹配简单。此外两镜面的间隔安装也使得系统的等效基线较长，从而具有较高的精度。 本文第一部分对当前的各种全向成像方法进行了简单介绍，并对各方法的特点做了归纳。第二部分介绍折反射全向视觉的研究现状，就各种反射镜面的成像特点做了对比。 第三部分介绍OSVOD的设计方法，包括机构的设计和镜面的设计，并对设计的结果做了误差分析。 第四部分是OSVOD的标定研究。给出了一种包括OSVOD中相机和镜面位置关系在内的系统参数的标定方法。该方法利用空间坐标已知的标定点在像平面上成的像，结合系统成像模型反算出标定点的空间坐标，再利用标定点的已知空间坐标和反算出的空间坐标建立方程，运用基于Levenberg-Marquardt的反向传播算法(backpropagation)标定相机与反射镜面间的安装偏差。该标定方法可推广到所有的折反射成像系统。 第五部分是基于全向图像的匹配研究。针对系统获取的立体图像对之间成像比例存在较大的差异，首先将图像展开成柱面投影图像，然后就下镜面成像展开的柱面做Canny边缘检测，得到了图像的边缘点；就得到的边缘点在展开的两幅柱面图像上做直接相关匹配。最后将获取的匹配点做一致性校验，并对一致性校验通过的匹配点做三维计算，生产稀疏的三维图像。 最后是结论和将来的工作展望。

基于人工神经网络的PC钢棒力学性能预报模型

提出一种PC钢棒抗拉强度的人工神经网络模型方法,采用4×9×1的三层前向BP网络结构,模型主要因素为淬火温度、回火温度、含碳量和单位长度质量。经1500余次训练后,误差平方和

基于神经网络的激光熔覆高度预测

激光成形过程中,对熔覆高度进行实时检测,从而实现熔覆高度闭环控制是成形高质量零件的保证。激光成形过程是一个多参数耦合的非线性过程,大量激光参数对成形熔覆表面质量具有重要影响。在分析激光参数对熔覆高度影响的基础上,建立利用激光工艺参数预测熔覆高度的误差反向传播(Backpropagation,BP)神经网络模型,完成了网络算法设计。通过激光成形试验采集样本,利用训练样本对所建立的网络进行训练,完成网络输入输出高度映射关系,并利用测试样本对所训练的网络进行检验。仿真试验表明,神经网络熔覆高度预测模型具有很高的精度,验证了该预测模型在理论和实践上的可行性与有效性。神经网络熔覆高度预测模型为实现激光加工过程熔覆高度实时预测与闭环控制打下基础,对提高成形产品质量具有重要意义。

A Theory of Networks for Appxoimation and Learning

Learning an input-output mapping from a set of examples, of the type that many neural networks have been constructed to perform, can be regarded as synthesizing an approximation of a multi-dimensional function, that is solving the problem of hypersurface reconstruction. From this point of view, this form of learning is closely related to classical approximation techniques, such as generalized splines and regularization theory. This paper considers the problems of an exact representation and, in more detail, of the approximation of linear and nolinear mappings in terms of simpler functions of fewer variables. Kolmogorov's theorem concerning the representation of functions of several variables in terms of functions of one variable turns out to be almost irrelevant in the context of networks for learning. We develop a theoretical framework for approximation based on regularization techniques that leads to a class of three-layer networks that we call Generalized Radial Basis Functions (GRBF), since they are mathematically related to the well-known Radial Basis Functions, mainly used for strict interpolation tasks. GRBF networks are not only equivalent to generalized splines, but are also closely related to pattern recognition methods such as Parzen windows and potential functions and to several neural network algorithms, such as Kanerva's associative memory, backpropagation and Kohonen's topology preserving map. They also have an interesting interpretation in terms of prototypes that are synthesized and optimally combined during the learning stage. The paper introduces several extensions and applications of the technique and discusses intriguing analogies with neurobiological data.