玉米叶片的光谱响应及其氮素含量预测研究

以不同施肥水平下两年玉米田间试验为基础,利用高光谱技术探讨大喇叭口期不同层次玉米叶片光谱响应的敏感区域,并依据叶片氮素含量与原始光谱反射率及其一阶导数的相关性,最终构建了叶片氮素含量的预测模型。结果表明：不同施肥水平下叶片光谱反射率差别明显区域集中在550 nm附近波段、761～1 300 nm波段,不同层次间叶片光谱反射率差别明显区域集中在550 nm附近波段,叶片氮素含量与470～760 nm波段光谱反射率及其一阶导数呈极显著相关。经过对比筛选,以光谱指数DSI(564,681)和DSI(681,707)构建的指数预测模型效果最好,预测精度达93.43%和93.39%,能有效估测叶片氮素含量。     

应用荧光光谱和径向基函数神经网络定量检测三聚氰胺

实验发现三聚氰胺溶液在紫外光激发下产生较强荧光,测得其荧光峰在310～600 nm之间,荧光峰值波长为420 nm左右,荧光相对强度与三聚氰胺溶液浓度呈现复杂的非线性关系。提出了采用径向基函数神经网络结合荧光光谱对三聚氰胺溶液浓度进行测定的方法。对每个样本选取30个发射波长值所对应的荧光强度作为网络数据,训练、建立了径向基函数神经网络。应用训练好的径向基函数神经网络,对5种三聚氰胺溶液的浓度进行预测,结果相对误差分别为0.93%,0.09%,0.31%,1.55%,4.61%。该方法能快捷、准确地测定三聚氰胺在溶液中的含量,为三聚氰胺检测及食品安全监管提供了一种新方法。     

Status of prediction methods for critical heat fluxes in mini and microchannels

Saturated critical heat flux (CHF) is an important issue during flow boiling in mini and microchannels. To determine the best prediction method available in the literature, 2996 data points from 19 different laboratories have been collected since 1958. The database includes nine different fluids (R-134a, R-245fa, R-236fa, R-123, R-32, R-113, nitrogen, CO₂ and water) for a wide range of experimental conditions. This database has been compared to 6 different correlations and 1 theoretically based model. For predicting the non-aqueous fluids, the theoretical model by Revellin and Thome [Revellin, R., Thome, J.R., 2008. A theoretical model for the prediction of the critical heat flux in heated microchannels. Int. J. Heat Mass Transfer 51, 1216–1225] is the best method. It predicts 86% of the CHF data for non-aqueous fluids within a 30% error band. The data for water are best predicted by the correlation by Zhang et al. [Zhang, W., Hibiki, T., Mishima, K., Mi, Y., 2006. Correlation of critical heat flux for flow boiling of water in minichannels. Int. J. Heat Mass Transfer 49, 1058–1072]. This method predicts 83% of the CHF data for water within a 30% error band. Some suggestions have also been proposed in this paper for the future studies.     

石太客运专线特长隧道涌水灾害预测研究

介绍了模糊综合评判方法及其应用步骤,在分析石太客运专线特长隧道地质构造、围岩类别、降雨量、地表环境特征等因素的基础上,建立模糊综合评判矩阵对隧道涌水进行预测。结果表明,模糊综合评判方法能很好地预测隧道涌水量的大小等级。     

Understanding crack growth in fuselage lap joints

The problem of multi-site damage and multiple interacting cracks is one experienced by many aircraft manufacturers and operators. This paper focuses on understanding the phenomena, and on developing a predictive capability that can form the engineering framework for maintaining continued airworthiness. To this end the present paper uses a simple formulation based on the Frost–Dugdale crack growth law to study the problem of cracking at fastener holes in fuselage lap joints and shows that the predicted crack growth history is in good agreement with both experimental results and with fleet data.     

Three-dimensional dynamic model for off-road vehicles using discrete body dynamics

This paper presents a simple, reliable dynamics model of off-road vehicle operation in real-time (RT) on terrain with obstacles. The numerical model was formulated by a new method – DBD (Discrete Body Dynamics). The new method is based on a discrete-element method, where the equations of motion are linear and simple to solve.In this new method, the suspension systems are composed of soft and stiff springs and dampers (instead of suspension arms and joints constrains), to present the kinematics and dynamics of real suspension. Reduction of the number of bodies and avoidance of constraints significantly improves model efficiency and simplicity.The tires–soil interaction was modeled using Brixius prediction. Specific soil properties were obtained from the classification system for each tire–soil interaction, size, and geometric area. The tire–ground contact was determined by topographic surface and adjustment of the forces and direction acting on the tires.The proposed method allows quick and simple definition of a vehicle. The model is written as an independent software infrastructure, enabling easy integration with any other software component.Simulation results were compared with Siemens' VL commercial multibody dynamics program. The performance of the proposed method was very similar to the commercial program (R² > 0.9), with the significant advantage of much higher RT performance.     

Efficient generation of accurate mobility maps using machine learning algorithms

U.S Army’s mission is to develop, integrate, and sustain the right technology solutions for all manned and unmanned ground vehicles, and mobility is a key requirement for all ground vehicles. Mobility focuses on ground vehicles’ capabilities that enable them to be deployable worldwide, operationally mobile in all environments, and protected from symmetrical and asymmetrical threats. In order for military ground vehicles to operate in any combat zone, the planners require a mobility map that gives the maximum predicted speeds on these off-road terrains. In the past, empirical and semi-empirical techniques (Ahlvin and Haley, 1992; Haley et al., 1979) were used to predict vehicle mobility on off-road terrains such as the NATO Reference Mobility Model (NRMM). Because of its empirical nature, the NRMM method cannot be extrapolated to new vehicle designs containing advanced technologies, nor can it be applied to lightweight robotic vehicles.The mobility map is a function of different parameters such as terrain topology and profile, soil type (mud, snow, sand, etc.), vegetation, obstacles, weather conditions, and vehicle type and characteristics.A physics-based method such as the discrete element method (DEM) (Dasch et al., 2016) was identified by the NATO Next Generation NRMM Team as a potential high fidelity method to model the soil. This method allows the capture of the soil deformation as well as its non-linear behavior. Hence it allows the simulation of the vehicle on any off-road terrain and have an accurate mobility map generated. The drawback of the DEM method is the required simulation time. It takes several weeks to generate the mobility map because of the large number of soil particles (millions) even while utilizing high performance computing.One approach to reduce the computational time is to use machine learning algorithms to predict the mobility map. Machine learning (Boutell et al., 2004; Burges, 1998; Barber et al., 1997) can lead to very accurate mobility predictions over a wide range of terrains. Machine learning is divided into two categories: the supervised and the unsupervised learning. Supervised learning requires the training data to be labeled into predetermined classes, while the unsupervised learning does not require the training data to be labeled. Machine learning can help generate mobility maps using trained models created from a minimum number of simulation runs. In this study different supervised machine learning algorithms such as the support vector machine (SVM), the nearest neighbor classifier (k-NN), decision trees, and boosting methods were used to create trained models labeled as 2 classes for the ‘go/no-go’ map, 5 classes for the 5-speed map, and 7 classes for the 7-speed map. The trained models were created from the physics-based simulation runs of a nominal wheeled vehicle traversing on a cohesive soil.     

基于空间填充准则的交叉验证方法及其应用

在统计学与机器学习中,交叉验证被广泛应用于评估模型的好坏.但交叉验证法的表现一般不稳定,因此评估时通常需要进行多次交叉验证并通过求均值以提高交叉验证算法的稳定性.文章提出了一种基于空间填充准则改进的k折交叉验证方法,它的思想是每一次划分的训练集和测试集均具有较好的均匀性.模拟结果表明,文章所提方法在五种分类模型(k近邻,决策树,随机森林,支持向量机和Adaboost)上对预测精度的估计均比普通k折交叉验证的高.将所提方法应用于骨质疏松实际数据分析中,根据对预测精度的估计选择了最优的模型进行骨质疏松患者的分类预测.