管道安全监测定位系统

提出一种可用于管道安全维护的全光纤监测定位系统。系统以多个熔锥型光纤耦合器构造的双Sagnac干涉光路监测扰动信号,利用其复用段上对同一个扰动信号产生两组不同响应的特点消除监测结果的不确定性,将两组信号同时进行差分放大、快速傅里叶变换,在此基础上采用Daubechies3小波去噪的分析方法,解决扰动源特性限制问题,得到表征其位置的频响曲线。在47.401km的环路复用段上施加模拟油气管道所受到的人为侵扰,系统正确确定了这一常见随机扰动源的位置,平均误差小于100m,表明该系统能够对较长距离的管道设施进行实时监测定位。     

分布式光纤管道泄漏检测及预警技术灵敏度分析

针对管道安全存在的问题,提出了一种基于Mach-Zehnder光纤干涉仪原理的分布式光纤管道泄漏检测及预警技术,利用与管道同沟敷设的光缆中的3条单模光纤构成分布式光纤微振动传感器检测管道沿线的振动信号,可以有效地检测管道沿线所发生泄漏和异常事件．分析和研究了检测系统的测试灵敏度影响因素,并采用不同光缆结构和埋设方式对检测系统的检测灵敏度进行测试．实验结果表明,该检测技术可以检测到1m范围内的人工挖掘信号,在气体管道压力不小于0．2MPa的条件下,可以检测到0．4m^3／min的气体管道泄漏．     

周期变化条件下输油管道监测的实用数据处理

面对输油管道监测,分析流量计的周期性特性,提出一种实用数据处理方法,通过对流量信号的数据处理和误差分析,建立相应的数学模型,将实际采集数据与模型数据进行对比,判别管道泄漏发生与否,对泄漏地点进行准确定位.该方法在华北油田长输管道监测系统中应用取得很好的效果,为其他管道监测系统提供了技术借鉴.     

基于一维卷积神经网络的超声导波管道裂纹识别方法

在船舶运输、石油化工等需要广泛使用各类型管道的行业中，管道的结构健康监测(structural health monitoring, SHM)对于工业系统的安全稳定运行意义重大。在基于超声导波的管道裂纹等级识别方面，建立了一个与实际管道基本一致的有限元模型，通过添加噪声的方式合成了更接近实际检测的导波数据。基于包含不同管道裂纹等级的有限元仿真数据库，提出了一种基于多尺度一维卷积神经网络(multi-scale one dimensional convolution neural network, MS-1DCNN)的管道裂纹等级识别模型，该模型以端到端的方法，将原始波形信号直接作为输入，无需专门设计信号降噪及特征提取算法。试验结果表明，该模型相较于传统机器学习方法在噪声环境下对管道裂纹等级的识别具有较高精度，并通过实物管道试验，验证了该模型在管道结构健康监测中的有效性。     

混合干涉型分布式光纤天然气管道泄漏检测及定位方法研究

介绍了一种基于马赫-曾德尔和萨格纳克混合干涉仪原理的分布式光纤天然气管道泄漏检测系统.阐述了检测系统的检测原理及泄漏定位方法,在泄漏点距法拉第旋转镜为6 032m处进行了气体管道泄漏检测实验并重复做了20次试验,测试了传感光纤与泄漏点不同夹角对检测系统性能的影响.实验结果表明,该系统可以检测出管道内气体压力为0.3MPa、泄漏孔径为2.5mm的微小泄漏信号并进行定位,平均相对定位误差为1.29%.在40mm的管道外径下,传感光纤与泄漏点夹角在0～180°范围内,检测系统均能检测出泄漏信号并进行定位.     

利用成像光谱仪识别猪肉和牛肉

利用成像光谱仪采集猪肉和牛肉的光谱,应用主成分分析(PCA)对所获得的原始光谱数据进行降维处理,分别利用KNN判别、人工神经网络(ANN)、支持向量机分类(SVM)三种建模方式,建立判别模型,并对猪肉和牛肉各20个预测样品进行识别.结果显示3种分类模型的正确识别率分别为92.5%、97.5%、100%.表明利用成像光谱仪可以实现对猪肉和牛肉的快速、准确、无损分类检测.     

分布式光纤传感器周界安防入侵信号的多目标识别

针对分布式光纤在周界安防系统中信号种类,即不同的环境下产生的噪声信号干扰和常见的入侵产生的信号。本文基于全光纤马赫—泽德干涉仪的分布式光纤传感模型,提出了一种识别常见的越境信号和消除环境噪声干扰信号的方法,实现了在去除环境干扰的情况下,用BP神经网络对多种入侵信号识别。实验结果证明,该方法能够有效的区分越境信号和不同环境状态产生的噪声信号,极大的提高了整个系统的识别率,降低了其虚警率。     

硅微声光传感特性的实验研究

硅微声光传感器是一种由硅微列阵簧片和光纤组合成的新型传感器，这种传感器在完成声光转换的同时对声频信号进行了并行滤波处理，可用作神经网络的并行输入，是一种光机电一体化的传感系统。本文中叙述通过纤反射强度调制技术检测硅簧片振动信号的方法；通过实验研究，完成了对声光调制信号的探测和预处理，并给出了实测数据及其分析。为实现音频编码和振动信号的实时分析，本文中还给出了采用人工神经网络对阵列传感器的输出信号进     

爆破震动速度峰值预测模型的比较和讨论

传统的爆破震动峰值质点速度(PPV)预测公式已不能适应现代爆破安全的要求。基于人工神经网络(ANN)原理,建立了结构为2-5-1的BP神经网络预测模型,以青藏铁路关角隧道爆破震动测试数据为样本,对人工神经网络模型和几种传统公式的爆破震动速度预测结果进行比较。通过对实测结果与预测结果相关性系数(COD)、平均绝对误差(MAE)和相对误差(RE)的分析,表明:在传统预测模型中USBM预测公式精度最高,而人工神经网络模型较传统模型更接近实际值。     

基于小波能谱和小波信息熵的管道异常振动事件识别方法

提出了基于小波能谱和小波信息熵的油气管道异常振动事件识别方法。基于Mach-Zehnder光纤干涉仪原理的分布式光纤油气管道安全监测系统实时检测管道沿途振动信号,对测量的时间序列进行小波变换,根据小波系数计算小波能谱与小波信息熵,通过小波能谱和小波信息熵值两种测度识别不同的管道安全异常事件。港枣线成品油管道的现场实验结果表明,该方法可以快速有效地识别管道周围发生的泄漏及其他异常情况,其总体识别准确率达到98.5%,有效降低了误报警率,具有较强的在线工况识别能力。     

Rapid analyte recognition in a device based on optical sensors and the olfactory system

We report here the development of a new vapor sensing device that is designed as an array of optically based chemosensors providing input to a pattern recognition system incorporating artificial neural networks. Distributed sensors providing inputs to an integrative circuit is a principle derived from studies of the vertebrate olfactory system. In the present device, primary chemosensing input is provided by an array of fiber-optic sensors. The individual fiber sensors, which are broadly yet differentially responsive, were constructed by immobilizing molecules of the fluorescent indicator dye Nile Red in polymer matrices of varying polarity, hydrophobicity, pore size, elasticity, and swelling tendency, creating unique sensing regions that interact differently with vapor molecules. The fluorescent signals obtained from each fiber sensor in response to 2-s applications of different analyte vapors have unique temporal characteristics. Using signals from the fiber array as inputs, artificial neural networks were trained to identify both single analytes and binary mixtures, as well as relative concentrations. Networks trained with integrated response data from the array or with temporal data from a single fiber made numerous errors in analyte identification across concentrations. However, when trained with temporal information from the fiber array, networks using "name" or "characteristic" output codes performed well in identifying test analytes.     

A Study on the Estimation of Prefabricated Glass Fiber Reinforced Concrete Panel Strength Values with an Artificial Neural Network Model

In this study, artificial neural networks trained with swarm based artificial bee colony optimization algorithm was implemented for prediction of the modulus of rapture values of the fabricated glass fiber reinforced concrete panels. For the application of the ANN models, 143 different four-point bending test results of glass fiber reinforced concrete mixes with the varied parameters of temperature, fiber content and slump values were introduced the artificial bee colony optimization and conventional back propagation algorithms. Training and the testing results of the corresponding models showed that artificial neural networks trained with the artificial bee colony optimization algorithm have remarkable potential for the prediction of modulus of rupture values and this method can be used as a preliminary decision criterion for quality check of the fabricated products.     

Pattern Recognition Techniques Applied to the Detection and Classification of Welding Defects by Magnetic Testing

The importance of nondestructive tests cannot be denied nowadays, and their use in the petroleum industry is extremely important to ensure safe operation of the equipment. The magnetic flux leakage (MFL) technique in instrumented pig is frequently used for pipeline inspection to detect weld defects and internal or external corrosion; however, as every nondestructive technique, it may present report errors, caused by several human factors (like lack of theoretical knowledge or experience in the technique applied, tiredness due to long hours of work, and so on) and by the inspection technique itself. Because pattern recognition techniques are already well known and applied in other technological research areas, such as fuzzy logic and artificial neural networks, the possibility to develop an automatic system to detect and classify defects through some nondestructive techniques aroused, mainly for those techniques that actually operate with signal or image interpretation. In the present work, pattern nonlinear classifiers were applied using artificial neural networks, to check the possibility to detect and classify defects in pipelines inspected through magnetic pigs (MFL technique). Several tests were performed on specimens with defects artificially inserted (internal and external corrosion and lack of penetration).     

An artificial neural network approach to multiphase continua constitutive modeling

Constitutive equations describe intrinsic relationships among sets of material system parameters. This study utilizes artificial neural networks in place of a traditional micromechanical approach to calculate the global (macroscopic) elastic properties of composite materials given the local (microscopic) properties and local geometry. This approach is shown to be more computationally efficient than conventional numerical micromechanical approaches. An eight sub-celled representative volume element is used for the local geometry. Multi target artificial neural networks (MTANNs) and single target artificial neural networks are studied for applicability in predicting the global properties. The best performing MTANN achieves a precision of 9%. The single target artificial neural networks (STANNs) perform best and predicts the global properties within a target error of 5.3%. The computation time is 1.8 s for all six STANNs to predict six global properties for 19,683 different microstructures.     

Eddy Current Signal Deconvolution Technique for the Improvement of Steam Generator Tubing Burst Pressure Predictions

Eddy current techniques are extremely sensitive to the presence of axial cracks in nuclear power plant steam generator tube walls, but they are equally sensitive to the presence of dents, fretting, support structures, corrosion products, and other artifacts. Eddy current signal interpretation is further complicated by cracking geometries more complex than a single axial crack. Although there has been limited success in classifying and sizing defects through artificial neural networks, the ability to predict tubing integrity has, so far, eluded modelers. In large part, this lack of success stems from an inability to distinguish crack signals from those arising from artifacts. We present here a new signal processing technique that deconvolves raw eddy current voltage signals into separate signal contributions from different sources, which allows signals associated with a dominant crack to be identified. The signal deconvolution technique, combined with artificial neural network modeling, significantly improves the prediction of tube burst pressure from bobbin-coil eddy current measurements of steam generator tubing.     

Fatigue life prediction of composite materials using polynomial classifiers and recurrent neural networks

Due to their massively parallel structure and ability to learn by example, artificial neural networks can deal with nonlinear problems for which an accurate analytical solution is difficult to obtain. These networks have been used in modeling the mechanical behavior of fiber-reinforced composite materials. Although promising results were obtained using such networks, more investigation on the appropriate choice of their structure and their performance in the presence of limited and noisy data is needed. On the other hand, polynomials networks have been known to have excellent properties as classifiers and are universal approximators to the optimal Bayes classifier. Not being dependant on various user defined parameters, having less computational requirements makes their use over other methods, such as neural networks, an advantage.

In this work, the fatigue behavior of unidirectional glass fiber/epoxy composite laminae under tension–tension and tension–compression loading is predicted using feedforward and recurrent neural networks. These predictions are compared to those obtained using polynomial classifiers. Experimental data obtained for fiber orientation angles of 0°, 19°, 45°, 71° and 90° under stress ratios of 0.5, 0 and –1 is used.

It is shown that, even when a small number of experimental data points is used to train both polynomial classifiers and neural networks, the predictions obtained are comparable to other current fatigue life-prediction methods. Also, polynomial classifiers are shown to provide accurate modeling between the input parameters (maximum stress, R-ratio, fiber orientation angle) and the number of cycles to failure when compared to neural networks.     

Optical reflectometric gas sensing: classification of hydrocarbon vapours by pattern recognition applied to RIfS sensor signals

Hydrocarbon vapours in air are monitored by a fibre-optical sensor system based on reflectometric interference spectroscopy (RIfS). A short review of the transducer principle and its properties is given. Classification of binary mixtures of hydrocarbon vapours is achieved by pattern recognition methods. The performance of principal component analysis and cluster analysis are compared to supervised and unsupervised artificial neural networks (self-organizing feature maps and back-propagation neural networks) for the evaluation of data from six RIfS sensors with respect to qualitative analysis.     

Statistical approaches to forecast gamma dose rates by using measurements from the atmosphere

In this paper, the results obtained by inter-comparing several statistical techniques for estimating gamma dose rates, such as an exponential moving average model, a seasonal exponential smoothing model and an artificial neural networks model, are reported. Seven years of gamma dose rates data measured in Daejeon City, Korea, were divided into two parts to develop the models and validate the effectiveness of the generated predictions by the techniques mentioned above. Artificial neural networks model shows the best forecasting capability among the three statistical models. The reason why the artificial neural networks model provides a superior prediction to the other models would be its ability for a non-linear approximation. To replace the gamma dose rates when missing data for an environmental monitoring system occurs, the moving average model and the seasonal exponential smoothing model can be better because they are faster and easier for applicability than the artificial neural networks model. These kinds of statistical approaches will be helpful for a real-time control of radio emissions or for an environmental quality assessment.     

基于神经网络的固体火箭发动机比冲性能的预示研究

将神经网络方法引入了固体火箭发动机的比冲性能预测,该方法避开了系统具体规律分析以及相应数学模型建立所带来的困难,直接用神经网络模型来模拟真实的系统关系。采用了一种改进的Ⅱ型RBF神经网络,克服了传统的RBF神经网络径向基函数个数未知的缺陷,并将其预测结果与传统的BP神经网络的预测结果进行了比较。     

Rule Extraction from Trained Artifical Neural Networks

The neural network rule extraction problem is aimed at obtaining rules from an arbitrarily trained artificial neural network. Recently there have been several approaches to rule extraction. Approaches to rule extraction implement a priori knowledge of data or rule requirements into neural networks before the rules are extracted. Although this may lead to a simplified final phase of acquitting the rules from particular type of neural networks, it limits the methodologies for general-purpose use. This article approaches the neural network rule extraction problem in its essential and general form. Preference is given to multilayer perceptron networks (MLP networks) due to their universal approximation capabilities. The article establishes general theoretical grounds for rule extraction from trained artificial neural networks and further focuses on the problem of crisp rule extraction. The problem of crisp rule extraction from trained MLP networks is first approached on theoretical level. Present ed theoretical results state conditions guaranteeing equivalence between classification by an MLP network and crisp logical formalism. Based on the theoretical results an algorithm for crisp rule extraction, independent of training strategy, is proposed. The rule extraction algorithm can be used even in cases where the theoretical conditions are not strictly satisfied; by offering an approximate classification. An introduced rule extraction algorithm is experimentally demonstrated.