水性防腐涂料湿附着力性能的配方优化模型

为提高水性防腐涂料湿附着力,提出基于人工神经网络(ANN)和遗传算法(GA)的水性防腐配方工艺优化模型。采用正交法设计配方工艺,根据实验结果,建立配方(水性环氧乳液、石墨烯、水性环氧固化剂、填料、附着力促进剂)、工艺(干燥温度)和性能(湿附着力)的预测模型,以水性防腐涂料湿附着力预测结果为优化目标,通过GA得出最佳的水性防腐涂料配方和工艺,并对该配方和工艺进行了实验和理论验证。该方法为制备高性能水性防腐涂料提供一定指导依据。     

基于工程图的飞机压延模设计CAD专家系统

基于工程设计ＣＡＤ专家系统特点,建立了ＡＤＤＥＳ专家系统（飞机压延模ＣＡＤ专家系统）。在建立ＤＩＥ编码系统基础上,使用特征编码等方法建立用户模型。系统中机构具有求解部分问题的能力,系统将机构层次地、分布式地组织,综合采用黑板结构、元推理等方法控制系统的问题求解策略,从而解决因知识库庞大而引起的效率低以及多专家系统合作等问题。     

燃气涡轮发动机故障诊断的人工神经网络法

介绍了人工神经网络专家系统在航空涡轮风扇发动机故障诊断上的应用。通过一个高涵道比涡扇发动机故障诊断的实例分析, 验证了三层逆传播网络的可行性。非常令人鼓舞的诊断结果证明:神经网络的模糊联想特点、分布存储和并行处理能力以及对随机误差的抑制作用, 使其成为一个有效的和快速的方法。它有很好的应用前景, 并可广泛用于航空发动机的故障诊断。      

面向问题的稀疏分布式记忆模型

在Kanerva所提出的稀疏分布式记忆(SDM)或存贮模型的基础上,为实现对特定类问题的大维数输入空间的模式识别,如汉字识别,脸谱辩认等,根据问题的具体情况,诸如汉字的频率分布等,提出了一个面向问题的稀疏分布式记忆模型。改进后的模型更符合实际应用,其中的学习规则采用了指数型记忆规则,使模型具有更高的信噪比,存贮容量亦大大提高。计算机模拟表明了这一点。     

模型跟踪和涡流控制在飞翼布局飞机上的应用

为增加小展弦比飞翼类布局飞机的操纵特性,并满足其对隐身性能的要求,将模型跟踪控制器和涡流控制器进行综合使用.模型跟踪是一种提高系统闭环响应并跟踪驾驶员指令的方法,通过此方法可使动态特性较差的飞机对动态特性较好的飞机输出进行跟踪,由此来改善飞机的动态特性.增加涡流控制可以进一步扩展模型跟踪控制的作用效果,涡流控制器使用变步长神经网络进行训练.结果表明,两种方法的综合使用可以有效地提高飞机的动态特性,通过增加涡流控制后,进一步增加了模型跟踪控制的作用效果.     

后掠翼人工诱导横流驻波的升华法研究

改进了添加粗糙带方法,使用丝网印刷方式在后掠机翼前缘添加特定间距的粗糙带.成功地在边界层内引入了特定波长的驻波,提高了人工诱导驻波波长精度.在西北工业大学低湍流度风洞中使用升华法研究了单一波长驻波的发展趋势,其扰动发展趋势与计算结果吻合得很好.     

BP神经网络在结构边界参数识别中的应用

针对建立发动机动力学模型过程中,试车台机架结构边界环境的不确定状况,对神经网络在边界刚度识别中的应用进行了研究。以结构模态频率为网络输入,边界X、Y、Z方向的刚度为输出,通过一种增加训练样本的方法大大提高了网络的映射性能,最终的识别结果达到了预期目标,满足工程需要。     

五种智能算法解决最大割问题分析与比较

最大割问题（Max—eulProblem）是一个典型的NP难组合优化问题。文章采用遗传算法、分布估计算法、Hopfield网络方法、蚁群算法、粒子群算法等5种算法对最大割问题进行求解,并用标准的多个不同规模最大割测试数据进行测试,研究各参数对算法的影响,并比较各种算法的时间复杂度和空间复杂度。测试结果表明该五种算法虽然在执行效率上有差异,但都能较好的解决最大割问题。     

Advanced data acquisition system for SEVAN

Huge magnetic clouds of plasma emitted by the Sun dominate intense geomagnetic storm occurrences and simultaneously they are correlated with variations of spectra of particles and nuclei in the interplanetary space, ranging from subtermal solar wind ions till GeV energy galactic cosmic rays. For a reliable and fast forecast of Space Weather world-wide networks of particle detectors are operated at different latitudes, longitudes, and altitudes. Based on a new type of hybrid particle detector developed in the context of the International Heliophysical Year (IHY 2007) at Aragats Space Environmental Center (ASEC) we start to prepare hardware and software for the first sites of Space Environmental Viewing and Analysis Network (SEVAN). In the paper the architecture of the newly developed data acquisition system for SEVAN is presented. We plan to run the SEVAN network under one-and-the-same data acquisition system, enabling fast integration of data for on-line analysis of Solar Flare Events. An Advanced Data Acquisition System (ADAS) is designed as a distributed network of uniform components connected by Web Services. Its main component is Unified Readout and Control Server (URCS) which controls the underlying electronics by means of detector specific drivers and makes a preliminary analysis of the on-line data. The lower level components of URCS are implemented in C and a fast binary representation is used for the data exchange with electronics. However, after preprocessing, the data are converted to a self-describing hybrid XML/Binary format. To achieve better reliability all URCS are running on embedded computers without disk and fans to avoid the limited lifetime of moving mechanical parts. The data storage is carried out by means of high performance servers working in parallel to provide data security. These servers are periodically inquiring the data from all URCS and storing it in a MySQL database. The implementation of the control interface is based on high level web standards and, therefore, all properties of the system can be remotely managed and monitored by the operators using web browsers. The advanced data acquisition system at ASEC in Armenia was started in November, 2006. The reliability of the multi-client service was proven by continuously monitoring neutral and charged cosmic ray particles. Seven particle monitors are located at 2000 and 3200 m above sea level at a distance of 40 and 60 km from the main data server.     

航天器非平稳振动环境TVAR时频分析(英文)

Predicting the time-varying auto-spectral density of a spacecraft in high-altitude orbits requires an accurate model for the non-stationary random vibration signals with densely spaced modal frequency. The traditional time-varying algorithm limits prediction accuracy, thus affecting a number of operational decisions. To solve this problem, a time-varying auto regressive （TVAR） model based on the process neural network （PNN） and the empirical mode decomposition （EMD） is proposed. The time-varying system is tracked on-line by establishing a time-varying parameter model, and then the relevant parameter spectrum is obtained. Firstly, the EMD method is utilized to decompose the signal into several intrinsic mode functions （IMFs）. Then for each IMF, the PNN is established and the time-varying auto-spectral density is obtained. Finally, the time-frequency distribution of the signals can be reconstructed by linear superposition. The simulation and the analytical results from an example demonstrate that this approach possesses simplicity, effectiveness, and feasibility, as well as higher frequency resolution.