Application of data fusion method to fault diagnosis of nuclear power plant

The work condition of nuclear power plant (NPP) is very bad, which makes it has faults easily. In order to diagnose the faults real time, the fusion diagnosis system is built. The data fusion fault diagnosis system adopts data fusion method and divides the fault diagnosis into three levels, which are data fusion level, feature level and decision level. The feature level uses three parallel neural networks whose structures are the same. The purpose of using neural networks is mainly to get basic probability assignment (BPA) of D-S evidence theory, and the neural networks in feature level are used for local diagnosis, D-S evidence theory is adopted to integrate the local diagnosis results in decision level. The reactor coolant system is the study object and we choose 2# steam generator Utubes break of the reactor coolant system as a diagnostic example, The experiments prove that the fusion diagnosis system can satisfy the fault diagnosis requirement of complicated system, and verify that the fusion fault diagnosis system can realize the fault diagnosis of NPP on line timely.     

Multi-agent immune recognition of water mine model

It is necessary for mine countermeasure systems to recognise the model of a water mine before destroying because the destroying measures to be taken must be determined according to mine model. In this paper, an immune neural network （INN） along with water mine model recognition system based on multi-agent system is proposed. A modified clonal selection algorithm for constructing such an INN is presented based on clonal selection principle. The INN is a two-layer Boolean network whose number of outputs is adaptable according to the task and the affinity threshold. Adjusting the affinity threshold can easily control different recognition precision, and the affinity threshold also can control the capability of noise tolerance.     

基于改进OLS算法的RBF神经网络 高速公路事件探测

提出了一种基于模糊聚类技术和RBF神经网络的混合智能高速公路事件自动探测算法,同时改进了用于RBF神经网络训练的Oils(正交最小二乘)选择算法.仿真实验证明,改进的OLS选择算法大大提高了RBF神经网络的训练速度同时具有无须事先确定RBF中心的优点,将之运用于公路事件探测可以获得满意的性能.     

基于神经网络的公路工程承发包价格发展预测

公路工程承发包价格是公路工程造价研究的重要内容。合理确定影响公路工程承发包价格的多种因素,采用基于BP神经网络的公路工程价格预测方法,建立模拟承发包价格与其影响因素间的非线性关系预测模型,经过训练与检验、数据预测,结果表明,该方法实用性强,具有一定的理论与实际应用价值。     

A method to estimate trip O‐D patterns using a neural network approach

This paper proposes a method which identifies the trip origin‐destination (O‐D) matrix when many pairs of values for the right hand side column (B) and the bottom row (A) of the matrix are given. The method considers B and A as the cause (input) and effect (output) of a system, respectively, and that the O‐D matrix represents the relationship between the cause and the effect. The relationship which satisfies all pairs of the cause and the effect data exactly may not be identified, but, should a general pattern of the relationship exist, it should emerge when many data sets of B and A are given. Two steps are involved in the method: the first step examines if a consistent O‐D pattern exists; if a pattern is found to exist, the second step identifies the values of the elements of the O‐D matrix. The first step is based on the shape of the possibility distributions of the values of the matrix elements. The second step uses a simple back‐propagation neural network. The method is useful to problems that require identification of the cause‐effect relationship when many sets of data for the cause and effect are available, for example, the station‐to‐station travel pattern on a rapid transit line when the total entering and exiting passengers are known at each station for many different days. The model can also be applied to other transportation problems which involve input and output relation.     

动态管理模式下路侧停车泊位占有率预测方法

城市停车已逐步实现信息化和动态化管理，本文对动态管理模式下大范围路侧泊位占有率预测方法进行研究.在收集美国旧金山492万条停车交易数据的基础上，利用可同时提取数据空间关联和时序趋势特征的卷积长短时记忆神经网络(Convolutional LSTM Network，ConvLSTM)，分别构建考虑停车费率和时限动态变化的有政策模型，和没有动态管理信息输入的无政策模型.结果显示，有政策模型的训练效率和预测精度会显著提升.在政策平稳阶段，两种模型均能够有效预测泊位占有率；在政策发生变化时段，无政策模型的预测误差出现激增，但有政策模型的预测误差依然保持平稳，表明本文提出的方法能够很好地应对动态管理模式下停车需求的变化.     

PSO-BP神经网络在船舶二级脱硝系统中的应用

随着国际海事组织（IMO）标准Tier III标准的实施，船舶柴油机尾气中的氮氧化物（NOx）的排放受到了更严格的限制。选择性催化还原脱硝技术因其高效、可靠等优势成为船舶柴油机NOx减排的主要选择。本文先根据SCR反应机理在MATLAB中搭建一级SCR脱硝系统和二级SCR脱硝系统模型，通过仿真结果对比，发现加入二级脱硝系统后，脱硝率可以得到提高。鉴于BP网络预测精度较低，本文将粒子群（PSO）算法加入到BP权值训练过程中，利用PSO-BP 神经网络预测 SCR-DeNOX系统出口处NOX浓度。结果表明：PSO-BP 神经网络不但可以预测SCR-DeNOX系统出口处的NOX浓度，而且与传统BP 神经网络相比可以提高预测结果的精度，为有效控制喷氨量、降低氨气逃逸量提供帮助。     

THE ROLE OF COMPOUND BLOCKAGE OF SODIUM CHANNEL ON NEUROPROTECTIVE EFFECT IN STRESSED RATS

ＲｅｐｅａｔｅｄｒｅｓｔｒａｉｎｔｓｔｒｅｓｓｏｒａｃｏｍｂｉｎａｔｉｏｎｏｆｄａｉｌｙｓｔｒｅｓｓｏｒｓｉｎｒａｔｓｉｎｄｕｃｅｄａｔｒｏｐｈｙｏｆｈｉｐｐｏｃａｍｐｕｓＣＡ３ｐｙｒａｍｉｄａｌｎｅｒｕｏｎｓ〔１〕，ｗｈｅｒｅａｓｉｔｈａｄｂｅｅｎｒｅｐｏｒｔｅｄｔｈａｔｔｈｅｅｌｅｖａｔｅｄｃｉｒｃｕｌａｔｉｏｎａｄｒｅｎａｌｓｔｅｒｏｉｄｓｓｅｃｒｅｔｅｄｄｕｒｉｎｇｓｔｒｅｓｓｍａｙｂｅｉｎｖｏｌｖｅｄｉｎｔｒｉｇｇｅｒｉｎｇｔｈｅｍｏｒｐｈｏｌｏｇｉｃａｌａｌｔｅｒａｔｉｏｎｓｓｕｃｈ…     

钢丝绳柔性抽油杆系统故障诊断研究

本文介绍了利用人工神经网络(ANN)技术和计算机图像处理技术对钢丝绳柔性抽油杆抽油装置进行故障诊断的方法。对示功图进行了二值化、滤波和正规化处理,获得了人工神经网络的输入向量,输出结果则表示可能发生的故障形式。采用BP网络,最后给出了一个诊断实例。