一种基于粗糙K-均值的椭球基函数神经网络

改进了一种椭球基函数神经网络,它与经典椭球单元神经网络的结构不同,而与径向基函数神经网络结构类似,即它有一个隐含层,并且隐层单元采用椭球基函数,区别于RBF网络的高斯函数。本文采用粗糙K-均值方法求取椭球函数的中心,并给出了该方法中确定初始阈值的步骤。这种改进方法不但使对输入空间的划分局部作用,而且划分区域封闭有界。因此,改进的神经网络具有较好的函数逼近能力和模式识别能力。仿真实验验证了该椭球基函数神经网络的正确性和有效性。     

双向进气结构脉管制冷机直流形成机理

在脉管制冷机中,由双向进气结构诱发的直流流动,不仅会增加冷端换热器的负载,降低制冷机性能,而且还是引起制冷温度不稳定的重要原因。如何有效地抑制直流流动已成为脉管制冷机能否大规模应用的关键。而对直流形成机理进行全面、深入的研究是发展简单、有效直流抑制手段的前提。针对该研究热点,本文开展了双向进气结构脉管制冷机直流形成机理研究,基于流体网络理论,考察工质物性对直流的影响和作用机理,进一步揭示了直流的形成本质。     

应用神经网络模型评价社保基金运营效果方法研究

社会保障基金是社会成员普遍关注的资金,他的利用关系到全社会成员的公共利益。因此,社会保障基金运用效果的评价问题一直是社会各界争论的焦点,正确评价其运用效果是值得学术界探讨的问题。运用神经网络模型对社会保障基金运用效果进行评价是对神经网络模型的应用及对社会保障基金运用效果的评价进行尝试和探讨。     

一般高速数据总线网络的吞吐量分析

分析了一般高速数据总线网络优先权方案的特性，建立了一般ＨＳＤＢ网络多优先权的平均吞吐量模型，导出了不同优先权的平均吞吐量解析表达式，并通过软件仿真验证了由得到的解析表达式。     

液压余度舵机同步传动研究

几个舵机驱动同一控制对象,通道间产生力的纷争,本文采用均衡网络来减轻力的纷争,而均衡网络的参数则根据解耦控制理论导出。实际系统的实验证明本文所提出的消除余度通道间力纷争保证同步传动的方案是有效的。文章给出了最优均衡网络的参数设计公式。     

情报网络对生产／流通过程中的库存的影响

本文通过建立独立型、垂直型和综合型3种模型,从订货方式和服务水平两个侧面,分析、考察了在生产/流通系统中建立情报网络而对系统中的库存所产生的影响。并据此提出了“失连库存”的新概念,对其进行了定量的分析和评价。本文得到的结论不仅在多阶层库存理论的研究上是必不可少的,而且对我国发展管理信息系统等现代化管理技术都具有指导意义。     

火箭发动机基于神经网络非线性辨识的故障检测

应用神经网络方法，提出了一种液体火箭发动机故障实时检测算法。神经网络采用非线性辨识技术贴近发动机的工作过程，并输出包合发动机故障信息的辨识误差信号。若辨识误差变大超过一定阈值，检测逻辑就预报发动机故障。在发动机启动阶段离线训练神经网络，在发动机稳态过程可以采用离线或在线学习算法。实验研究表明神经网络可以成功地应用于大型泵压式液体火箭发动机的故障检测。     

一种应用于实时深空干涉测量的电离层时延修正方法

针对"嫦娥4号"中继星任务S频段信标信号的高精度实时干涉测量需求,结合深空测控干涉测量系统采用的稀疏标校工作模式,研究验证了一种面向测控模式实时干涉测量的电离层时延修正方法。首先分析了电磁波经电离层传播的延迟机理及特性;基于深空站历史观测数据,通过自相关函数分析验证了天顶向TEC的周日特性;在此基础上,结合深空干涉测量中心数据处理设备软件系统,讨论了电离层时延修正方法;通过任务期间的实测数据处理分析,验证了所提方法可以将实时测量精度提升1～3 ns,对低仰角跟踪弧段,该技术方法优势更为明显。该方法为后续深入推进深空测控干涉测量系统在任务中的实时应用提供了技术储备。     

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.     

Predictive fault-tolerant control of an all-thruster satellite in 6-DOF motion via neural network model updating

The problem of controlling an all-thruster spacecraft in the coupled translational-rotational motion in presence of actuators fault and/or failure is investigated in this paper. The nonlinear model predictive control approach is used because of its ability to predict the future behavior of the system. The fault/failure of the thrusters changes the mapping between the commanded forces to the thrusters and actual force/torque generated by the thruster system. Thus, the basic six degree-of-freedom kinetic equations are separated from this mapping and a set of neural networks are trained off-line to learn the kinetic equations. Then, two neural networks are attached to these trained networks in order to learn the thruster commands to force/torque mappings on-line. Different off-nominal conditions are modeled so that neural networks can detect any failure and fault, including scale factor and misalignment of thrusters. A simple model of the spacecraft relative motion is used in MPC to decrease the computational burden. However, a precise model by the means of orbit propagation including different types of perturbation is utilized to evaluate the usefulness of the proposed approach in actual conditions. The numerical simulation shows that this method can successfully control the all-thruster spacecraft with ON-OFF thrusters in different combinations of thruster fault and/or failure.