基于模糊神经网络的设备故障预测研究

对模糊神经网络技术进行了研究，提出了预测分析的模糊神经网络模型；建立了故障指标评定方法，利用预测算法运用参数历史故障指标对参数指标进行趋势预测，预测得到的参数指标可以根据专家诊断系统判据进行诊断，对未来设备的健康状况进行可信度较高的评估。经仿真结果验证，该算法预测精度较高，预测结果可信.     

基于模糊逻辑的舰船维修经费预测

针对多输入、单输出的复杂非线性系统，基于Takagi-Sugeno模糊规则给出了一种系统预测模型，分别用模糊C- 值聚类算法和线性回归方法导出模型参数。利用该模型对舰船维修经费进行了预测，并进行了精度分析。     

HVAC系统的模糊预测函数控制器设计

针对暖通空调HVAC系统中由于存在高度非线性、时变特征以及扰动和不确定性等因素而难以控制的特点，提出基于Takagi-Sugeno(T-S)模糊模型的预测函数控制器设计方法。该方法通过最小二乘辨识算法建立系统的模糊T-S模型，然后基于模糊全局线性化预测模型，采用预测函数控制算法设计系统控制律。仿真实验结果表明该算法是一种跟踪性能好、鲁棒性强的有效控制方法。与常规的PID控制器相比，该方法具有超调量小、调整时间短等优良的动态性能。     

应用模糊随机理论预测空化状态

本文首先分析了空化初生的模糊随机现象和用模糊集理论进行空化状态预测的可能性，其次，讨论了空化系统的模糊随机变量及其数字特征，在此基础上，提出了空化状态的模糊随机理论预测方法，并给出了相应的算例。     

测量仪器校准间隔的模糊范数融合预测

校准间隔的预测是解决测量仪器不足校准和过剩校准的一种主要方法。现有校准间隔预测模型仅仅关注瞬时值,而随机因素的存在导致瞬时值难以精确预测。针对这一问题,在历史校准数据为小样本条件下,提出一种模糊范数融合模型,进行测量仪器校准间隔预测。首先,通过分析历史校准数据的特征和影响因素,建立了校准数据的数学描述模型。在此基础上,采用模糊范数融合模型,预测校准数据的区间。实验表明,模糊范数融合模型更能反映校准数据序列的变化趋势,克服了其它预测模型仅仅预测瞬时值的缺点,提高了预测的可靠性。     

现有混凝土桥梁结构剩余寿命预测方法研究

在回顾桥梁剩余寿命预测方法的基础上,探讨了利用人工神经网络对现有混凝土桥梁结构剩余寿命进行全桥模糊预测的方法。该方法从全桥的角度综合考虑各种因素影响,利用因果分析图及数学模糊规则,事先归纳出影响混凝土桥梁整体寿命的几种主要因素,并利用自编程序对几种主要因素的相关数据进行神经网络训练。训练结果表明,只要网络结构选择合理,配合正常的检测制度与专门的数据采集系统,其对全桥剩余寿命进行模糊预测的结果是具有实用价值的。     

基于模糊故障树的铁路机车周期预测研究

对当前我国机车检修预测的现况进行分析,在传统的机务检修基础上建立机车检修周期模糊故障树模型,形成一种预测机车检修周期的新方法,解决我国铁路机车科学检修中的机车检修周期预测问题。     

模糊系统方法在混凝土无损检测抗压强度预测中的应用

将模糊系统方法应用于混凝土无损检测技术中,以回弹值、超声值、碳化深度值和含水率作为模型输入变量,抗压强度值作为模型输出变量值,通过对样本数据训练,提取模糊规则,形成完备的模糊规则库。采用单值模糊器、乘积推理机、中心平均解模糊器,建立混凝土无损检测模糊系统模型。实验结果表明:模糊系统模型预测结果的平均相对误差为7.98%,相对标准差为10.53%。该模型的预测精度明显高于目前常用的回归模型预测精度。进而提出一种评定各输入变量重要性的方法,实验结果表明,影响混凝土抗压强度的各因素的重要性评定依次为:回弹值、超声值、碳化深度和含水率,其中回弹值最重要。该方法可以帮助我们选择较重要的变量作为模型的输入变量,实现模型的效率优化。     

规则库自提取和简化的模糊系统

介绍了一种新的规则自提取和简化的模糊模型算法，该算法由3个步骤组成：首先，通过构造一个模糊软划分判别准则对输入输出空间进行模糊划分，得到最佳分类数，从而得到了最佳的规则数；其次，根据最佳规则数和神经网络来构造初始的模糊模型；第三，通过运用模糊相似分析法，可以得到计算2个模糊集相似度的方程，从而根据事先确定的阈值来合并相似的模糊集。这样就得到了一个既满足对精度的要求又简单且具有可解释性的模糊系统。最后，用该算法对我国全要素生产力进行了模糊规则的提取。     

立井爆破效果预测及模糊综合评价

为了预测立井爆破效果以优化设计参数,确定以影响立井爆破效果的10个设计参数及"循环进尺、炮孔利用率、平均单耗、超欠挖量、周边孔眼痕率"5个效果参数作为输入、输出层样本,建立基于BP神经网络的立井爆破效果预测模型。结果表明:各参数的预测误差大多控制在5%以内,模型能够较好地达到预测目的;为了对预测结果进行定量化评价,以"炮眼利用率、炸药单耗、超欠挖量、周边孔眼痕率"作为评价指标,建立模糊综合评价模型并应用于白家宫2#副井中,评价结果与现场实际情况一致,达到了定量化评价的目的。将二者结合应用,对于优化立井爆破设计参数具有重要作用。     

基于模糊神经网络与主成分分析的磨削表面粗糙度在线预测

基于声发射和振动信号提出了一种模糊神经网络和主成分分析的表面粗糙度预测方法,以提高磨削过程中工件表面粗糙度识别的准确性。首先,采集磨削程中声发射与振动信号,提取相关时域特征、频域特征和小波包特征参数,利用主成分分析对特征量进行降维优化;然后,构建表面粗糙度模糊神经网络预测模型,将信号特征量与表面粗糙度作为模糊神经网络的输入和输出;最后,对模型进行训练,并对表面粗糙度预测精度进行验证。实验结果表明:通过主成分分析(PCA)方法对声发射和振动信号特征量进行降维得到5个主成分,以此建立的模糊神经网络表面粗糙度预测模型的效果精度可达到91%以上,与局部线性嵌入和多维标度法降维方法相比,PCA方法降维后的特征所含信息更优,预测准确度更高。     

A Fuzzy Logic-Based Prediction Model for Kerf Width in Laser Beam Machining

In laser beam machining, the main concern is the machining quality as kerf width of the end product. It is essential for industrial applications to cut the workpiece with minimum kerf width. However, it is difficult to develop a precise functional relationship between input and output variables in laser machining. Therefore, an effort has been conducted to build up an intelligent fuzzy expert system (FES) model to predict the kerf width in CO₂ laser cutting. The employed input parameters were assisting gas pressure, laser power, cutting speed, and standoff distance. The fuzzy logic was performed on fuzzy toolbox in MATLAB R2009b by employing Mamdani technique. In total, 81 experiments were carried out and experimental results were used for training and testing of the developed fuzzy model. Relative error and goodness of fit were used to investigate the accuracy of the prediction ability and the values of 3.852% and 0.994, respectively, were found to be satisfactory. This paper will extend knowledge about the prediction of kerf width by using FES model.     

Integrated System Health Management‐based Fuzzy On‐board Condition Prediction for Manned Spacecraft Avionics

The purpose of this study is to develop a prediction methodology for a condition assessment of on‐board integrated health management systems in manned spacecraft avionics. The framework of this study is based on two main premises. The first is the need to examine the problems in the on‐board prediction tools in space avionics integrated system health management, an area that has been rarely focused on. The second is the need to consider the failure correlation coefficients to enable accurate health predictions. To deal with the uncertainty in the prediction process, a fuzzy theory–gray model–support vector machine approach, which uses fuzzy theory combined with a gray model and a support vector machine, is used to make the prediction. An example is given to demonstrate the accuracy and reliability of the model. Copyright © 2014 John Wiley & Sons, Ltd.     

Application of Hybrid Learning Neural Fuzzy Systems in Reliability Prediction

This study presents a hybrid learning neural fuzzy system for accurately predicting system reliability. Neural fuzzy system learning with and without supervision has been successfully applied in control systems and pattern recognition problems. This investigation modifies the hybrid learning fuzzy systems to accept time series data and therefore examines the feasibility of reliability prediction. Two neural network systems are developed for solving different reliability prediction problems. Additionally, a scaled conjugate gradient learning method is applied to accelerate the training in the supervised learning phase. Several existing approaches, including feed‐forward multilayer perceptron (MLP) networks, radial basis function (RBF) neural networks and Box–Jenkins autoregressive integrated moving average (ARIMA) models, are used to compare the performance of the reliability prediction. The numerical results demonstrate that the neural fuzzy systems have higher prediction accuracy than the other methods. Copyright © 2005 John Wiley & Sons, Ltd.     

基于人工智能方法的地下洞室群爆破振动速度预测

特大断面地下洞库爆破开挖工程中涉及到众多的影响因素,为了较准确地预测出爆破振动速度,引入支持向量机理论,建立最小二成支持向量机爆破振动速度预测模型(LS-SVM模型),该模型利用结构风险最小化来提高求解问题的速度和精度。采用该模型对某地下水封LPG洞库工程进行爆破振动速度预测,并与传统的萨道夫斯基回归公式模型(萨氏模型)和模糊神经网络模型(FNN模型)进行对比分析。分析结果表明:LS-SVM模型、FNN模型与萨氏模型的全局均方根相对误差RMSRE分别为4.68%、14.42%与19.33%;LS-SVM模型有14组数据满足预测模型泛化能力误差阀值(6%)的要求,而FNN模型与萨氏模型均不满足要求。因此LS-SVM模型在爆破振动速度预测中的预测性能和泛化能力均优于FNN模型及萨氏模型,可为多因素影响下类似工程爆破振动速度预测提供借鉴经验。     

基于蚁群算法优化BP神经网络的港口吞吐量预测

利用蚁群算法优化反向传播神经网络的初始权值、阈值,建立预测模型,对港口货物吞吐量进行预测。蚁群算法具有全局搜索能力,分布式计算和鲁棒性强等特点,有利于加快反向传播神经网络的收敛速度,避免易陷入局部极值的问题,提高建模精度。在港口吞吐量预测中的应用表明:蚁群算法优化BP神经网络模型、模糊神经网络预测模型、RBF预测模型及BP预测模型的平均绝对百分比误差分别为2.826%、3.734%、4.990%和6.566%;同时,蚁群算法优化BP神经网络模型收敛速度最快。与传统BP神经网络、RBF网络及模糊神经网络相比,蚁群算法优化BP神经网络模型、模糊神经网络预测模型、RBF预测模型及BP预测模型的平均绝对百分比误差分别为2.826%、3.734%、4.990%和6.566%。     

Evaluating the Impact of Prediction Techniques: Software Reliability Perspective

Maintaining software reliability is the key idea for conducting quality research. This can be done by having less complex applications. While developers and other experts have made significant efforts in this context, the level of reliability is not the same as it should be. Therefore, further research into the most detailed mechanisms for evaluating and increasing software reliability is essential. A significant aspect of growing the degree of reliable applications is the quantitative assessment of reliability. There are multiple statistical as well as soft computing methods available in literature for predicting reliability of software. However, none of these mechanisms are useful for all kinds of failure datasets and applications. Hence finding the most optimal model for reliability prediction is an important concern. This paper suggests a novel method to substantially pick the best model of reliability prediction. This method is the combination of analytic hierarchy method (AHP), hesitant fuzzy (HF) sets and technique for order of preference by similarity to ideal solution (TOPSIS). In addition, using the different iterations of the process, procedural sensitivity was also performed to validate the findings. The findings of the software reliability prediction models prioritization will help the developers to estimate reliability prediction based on the software type.     

客户潜在价值预测模型及细分研究

从客户价值的定义和划分入手，详细讨论了客户的潜在价值，建立了一个预测客户潜在价值的多变量概率单位模型；基于该模型进行客户细分，探讨了如何采取不同的客户战略；最后通过一个实际案例证明了客户潜在价值预测模型的有效性。