基于函数链神经网络的深度分类器

目前的宽度学习系统（Broad learning system，BLS）通过所建立的一系列映射节点和增强节点来形成联合节点。因为联合节点与输出层的线性连接，网络权值可以用求解伪逆的方法快速求得，避免了耗时的训练过程，从而成为快速而高效的学习方法。然而在追求高精度结果的过程中，BLS对于增强节点数量的需求过于巨大，容易造成过拟合问题。为此，本文提出了基于函数链神经网络（Functional-link neural network，FLNN）的深度分类器（FLNN based deep classifier，FLNNDC），旨在提供一种更加简单却又不失精度的BLS变体结构。FLNNDC将几个轻量级的BLS子系统堆积成栈式结构，每一个轻量级的BLS子系统随机选择一部分映射节点生成增强节点，而不是全部映射节点。和原宽度结构相比，在几个主流数据集上的实验结果表明本文所提出的FLNNDC分类器具有网络结构更小且学习速度更快的优势。     

基于深度学习的毫米波和亚毫米波成像仪的图像增强技术

风云四号卫星毫米波和亚毫米波成像仪(MMSI)数据根据采样方式分为过采样和非过采样数据。由于采样方式的影响,非过采样数据在采样过程中会有一定的信息损失。为解决采用简单的线性插值方法做精细化处理时提升精度有限问题,采用基于深度学习的方法增强MMSI亮温图像,设计卷积神经网络重建风云四号卫星MMSI的亮温图像和风云三号卫星微波成像仪亮温图像。实验结果显示:相比传统的双三次插值方法,在风云三号卫星微波成像仪亮温图像样本上峰值信噪比提升了1.13dB,结构相似度提升了0.01。实验结果表明:对于非过采样亮温数据,采用基于深度学习的方法增强图像具有更高的精度,同时可在其他微波探测仪数据中使用,具有很强的普适性。     

基于IPSO-Elman神经网络的飞机客舱能耗预测

为了提高飞机客舱使用地面空调制冷时客舱能耗的预测精度，提出了一种改进的粒子群优化（IPSO）Elman神经网络的飞机客舱能耗预测模型。依据对算法中惯性权重与学习因子的收敛域分析，得出了二者合理的取值范围，将粒子到全局最优位置间距离与参数的取值范围相结合，构造了惯性权重与学习因子的动态调节函数，对其进行非线性的动态调节，并引入了变异因子，提出了一种跳出局部最优的策略，防止粒子群优化（PSO）陷入局部最优。将IPSO-Elman应用于Boeing738飞机客舱能耗预测中，与PSO-Elman、Elman算法进行性能比较，仿真结果表明基于IPSO-Elman的客舱能耗预测模型在预测精度和收敛速度方面均有一定的提升。该研究结果为飞机客舱能耗预测模型的建立提供了理论依据，对飞机地面空调的节能与机场电能合理调配提供了支持。     

Rotating machinery fault diagnosis based on convolutional neural network and infrared thermal imaging

Rotating machinery is widely applied in industrial applications. Fault diagnosis of rotating machinery is vital in manufacturing system, which can prevent catastrophic failure and reduce financial losses. Recently, Deep Learning (DL)-based fault diagnosis method becomes a hot topic. Convolutional Neural Network (CNN) is an effective DL method to extract the features of raw data automatically. This paper develops a fault diagnosis method using CNN for InfRared Thermal (IRT) image. First, IRT technique is utilized to capture the IRT images of rotating machinery. Second, the CNN is applied to extract fault features from the IRT images. In the end, the obtained features are fed into the Softmax Regression (SR) classifier for fault pattern identification. The effectiveness of the proposed method is validated using two different experimental data. Results show that the proposed method has a superior performance in identification various faults on rotor and bearings comparing with other deep learning models and traditional vibration-based method.     

Analysis of the dynamic behavior of structures using the high-rate GNSS-PPP method combined with a wavelet-neural model: Numerical simulation and experimental tests

Recently, the high rate global navigation satellite system-precise point positioning (GNSS-PPP) technique has been used to detect the dynamic behavior of structures. This study aimed to increase the accuracy of the extraction oscillation properties of structural movements based on the high-rate (10?Hz) GNSS-PPP monitoring technique. A developmental model based on the combination of wavelet package transformation (WPT) de-noising and neural network prediction (NN) was proposed to improve the dynamic behavior of structures for GNSS-PPP method. A complicated numerical simulation involving highly noisy data and 13 experimental cases with different loads were utilized to confirm the efficiency of the proposed model design and the monitoring technique in detecting the dynamic behavior of structures. The results revealed that, when combined with the proposed model, GNSS-PPP method can be used to accurately detect the dynamic behavior of engineering structures as an alternative to relative GNSS method.     

A new bearing fault diagnosis method based on modified convolutional neural networks

Fault diagnosis is vital in manufacturing system. However, the first step of the traditional fault diagnosis method is to process the signal, extract the features and then put the features into a selected classifier for classification. The process of feature extraction depends on the experimenters’ experience, and the classification rate of the shallow diagnostic model does not achieve satisfactory results. In view of these problems, this paper proposes a method of converting raw signals into two-dimensional images. This method can extract the features of the converted two-dimensional images and eliminate the impact of expert’s experience on the feature extraction process. And it follows by proposing an intelligent diagnosis algorithm based on Convolution Neural Network (CNN), which can automatically accomplish the process of the feature extraction and fault diagnosis. The effect of this method is verified by bearing data. The influence of different sample sizes and different load conditions on the diagnostic capability of this method is analyzed. The results show that the proposed method is effective and can meet the timeliness requirements of fault diagnosis.     

卷积神经网络和峭度在轴承故障诊断中的应用

针对传统智能诊断方法依靠专家知识和人工提取数据特征工作量大的问题,结合深度学习方法在特征提取和处理大数据方面的优势,研究了一种基于卷积神经网络和振动信号峭度指标的滚动轴承故障诊断方法。该方法将深度学习应用于轴承故障诊断,提取滚动轴承正常状态、内圈故障、外圈故障和滚动体故障4种状态的振动信号,将振动信号分段处理得到峭度指标,使用数据到图像的转换方法将峭度指标转换为灰度图,送入卷积神经网络模型完成故障分类。在进行滚动轴承故障诊断的实验时,所提的模型诊断准确率达到99.5%,高于传统支持向量机(SVM)算法的95.8%。      

多航天器协同探测效能影响参数不确定性分析

为解决传统基于蒙特卡洛仿真的探测效能不确定性分析评价方法中时间效率低，参数关系映射单一的问题，提出一种基于神经网络的不确定性分析方法，利用人工神经网络在拟合回归分析上的非线性特性，设计了能够替代复杂系统的神经网络结构，能够通过少量仿真计算结果作为训练样本实现模型的收敛并能有效反映被替代系统的原有特性。选择以一个多航天器的天文观测任务作为典型用例，建立了仿真分析模型，并与蒙特卡洛仿真方法结果进行了对比，验证了此方法的准确性和计算效率。     

基于室内无线模型的穿墙损耗校正

K-M（Keenan-Motley）模型将单墙固定损耗值相加来计算室内无线信号穿透多墙的总损耗值，存在较大误差。针对该问题选取多种室内场景分别进行连续波（Continuous Wave，CW）测试，对无线信号穿墙损耗的影响因素和变化规律进行分析，提出了一种基于人工神经网络的室内无线模型穿墙损耗校正方法，对预处理后的测试数据进行训练并建立预测模型。经验证该预测模型符合校正判别准则，在实际场景下具有良好的预测准确度。     

月球表层采样样品智能确认方法

表层采样是月球采样探测的重要方式，样品智能确认有助于提升工作效率与复杂问题处理能力。结合月球表层采样铲挖工作过程，分析了铲挖过程中臂载相机图像的特点，模仿有人参与识别过程，提出了层次解耦的月球样品智能识别流程，利用深度学习方法构建了一类深度卷积识别网络，完整地描述了图像、特征、标记在网络中的正反传递关系，并在月球表层采样地面试验中进行了验证，结果表明该方法对不同光照、不同背景、不同过程、不同形态的样品，具有较好的泛化识别能力，误识别率优于8.1%，平均单幅识别时间约0.7 s。