基于人工神经网络的台风浪高快速计算方法

采用2010—2017年南海5个浮标波高观测资料和中国气象局热带气旋最佳路径集中的热带气旋参数, 基于前馈型误差反向传播(Forward Feedback Back Propagation, FFBP)神经网络(Artificial Neural Network, ANN)方法, 分别建立了各浮标站的台风浪高快速计算模型。研究显示, 基于热带气旋中心坐标、中心最低气压、近中心最大风速、热带气旋中心与浮标之间的距离和方位4个参数建立的神经网络模型经反复训练后, 模型输出结果可以很好地拟合观测数据, 各浮标有效波高计算值与观测值的均方根误差小于0.3m, 平均相对误差为5.78%~7.23%, 相关系数大于0.9, 属高度相关。独立测试结果显示, “山竹”( 国际编号: 1822)影响期间有效波高最大值的神经网络模型预报结果与观测值基本吻合, 相对误差为-31.06%~0.98%, 但计算的最大值出现时间和观测情况不完全一致。该计算方法可应用于热带气旋影响期间的有效波高最大值计算, 因而在海洋工程领域和海洋预报领域具有应用前景。     

Artificial neural network-based storm surge forecast model: Practical application to Sakai Minato,Japan

The present study describes a novel way of a systematic and objective selection procedure for the development of an Artificial Neural Network-based storm Surge Forecast Model (ANN-SFM) with the 5, 12 and 24 h-lead times and its application to Sakai Minato area on the Tottori coast, Japan. The selection procedure guides how to determine the superiority of the best performing model in terms of the appropriate combination of unit number in the hidden layer and parameter in the input layer. In the application of ANN-SFM to Sakai Minato, it is found that the best 5 and 12 h-forecast ANN-SFMs are established with the most suitable set of 70 units (the number of hidden neurons) and the input components of surge level, sea level pressure, the depression rate of sea level pressure, longitude, latitude, central atmospheric pressure and highest wind speed. The best 24 h-forecast ANN-SFM is determined with 160 units and the input parameters of surge level, sea level pressure, the depression rate of sea level pressure, longitude and latitude. The proposed method of the selection procedure is able to be adaptable to other coastal locations for the development of the artificial neural network-based storm surge forecast model as establishing the superiority of the most relevant set combining unit numbers and input parameters.     

地球科学知识图谱的构建与展望

大数据为地球科学研究带来了新的思路和挑战。但由于存在描述规范不统一、共享机制不明、语义异构等问题,在数据集成、共享与复用等方面存在较大困难,使得大数据的众多优势在地球科学相关研究中难以充分发挥。知识图谱能够准确、清晰地表达概念及其相互之间的复杂语义关系,为机器所理解,是实现语义翻译、数据融合和复用的关键技术。文章对地球科学知识图谱的内涵和特点进行了深入的分析,归纳了地球科学知识图谱的主要构建方法,梳理了数据字典、知识体系和知识图谱之间的关系,对与地球科学知识图谱构建相关的专题数据库和领域本体的建设现状进行了回顾,指出了地球科学知识图谱构建中存在的主要问题,并阐述了地球科学知识图谱的应用前景,以期推动和完善地球科学知识图谱的建设和应用。     

人工智能时代的测绘工程教育改革

人工智能技术发展至今,已掀起了第三次浪潮。人工智能时代的到来必将推动测绘工程教育的改革。本文从人工智能技术的发展历程出发,阐述了人工智能在测绘行业的应用现状和发展趋势,分析了人工智能对教育行业的影响,探讨了人工智能与高等教育相结合的应用前景。结合测绘工程高等教育的现状,分别从培养理念、知识结构、教学方法和创新能力4个方面构建了人工智能时代测绘工程专业的教育改革体系,为探索我国测绘工程高等教育同人工智能的深度融合提供理论参考和经验借鉴。     

Local effects of blue mussels around turbine foundations in an ecosystem model of Nysted off-shore wind farm, Denmark

The development of off-shore wind farms along the coastline of north-west Europe is rapidly increasing; it is therefore important to study how this will affect the marine environment. The present study modelled the growth and feed-backs of blue mussels in natural beds and on turbine foundations in an off-shore wind farm (OWF) located in a shallow coastal ecosystem by coupling a dynamic energy budget (DEB) model to a small-scale 3D hydrodynamic–biogeochemical model. The model results showed that blue mussels located higher up in the water column on turbine pillars achieved a 7–18 times higher biomass than those located on the scour protection because the former experience an enhanced advective food supply. Secondly, the high biomasses of blue mussels on foundations created local ‘hot spots’ of biological activity and changed ecosystem dynamics due to their feed-backs e.g. ingestion of microplankton and copepods, excretion of ammonium and egestion of faecal pellets. The model results were supported by field measurements around foundations of Chl a concentrations and biomasses of the fauna community. Our study emphasised that OWFs seem to be particularly favourable for blue mussels in the western Baltic Sea and that the functioning of the OWFs as artificial reef ecosystems depends upon how the blue mussels interact with their local pelagic and benthic environment.     

A non-iterative and non-singular perturbation solution for transforming Cartesian to geodetic coordinates

The Cartesian-to-Geodetic coordinate transformation is re-cast as a minimization algorithm for the height of the Satellite above the reference Earth surface. Optimal necessary conditions are obtained that fix the satellite ground track vector components in the Earth surface. The introduction of an artificial perturbation variable yields a rapidly converging second order power series solution. The initial starting guess for the solution provides 3–4 digits of precision. Convergence of the perturbation series expansion is accelerated by replacing the series solution with a Padé approximation. For satellites with heights < 30,000 km the second-order expansions yields ~mm satellite geodetic height errors and ~10⁻¹² rad errors for the geodetic latitude. No quartic or cubic solutions are required: the algorithm is both non-iterative and non-singular. Only two square root and two arc-tan calculations are required for the entire transformation. The proposed algorithm has been measured to be ~41% faster than the celebrated Bowring method. Several numerical examples are provided to demonstrate the effectiveness of the new algorithm.     

Advanced processing techniques for remotely sensed imagery

综述了遥感影像信息处理技术的研究进展,主要包括高分辨率影像信息提取技术、影像超分辨率、高光谱影像处理和目标探测,以及遥感影像处理与分类的人工智能方法.对于高分辨率影像处理,从纹理、形状、结构和对象的角度探讨了空间信息提取对于高分辨率影像解译的意义和作用,分析了小波纹理、空间共生纹理、形状特征提取和面向对象分类技术的进展和存在的问题;对于超分辨率技术,文章主要介绍了超分辨率技术的最新进展,及其在遥感影像(sPOT5和MODIS)中的应用;在高光谱数据处理方面,从纯净像元和混合像元两方面介绍了最新的进展.对于纯净像元方法.主要分析了植被指数和统计方法,混合像元方面,则主要分析了像元分解、端元提取的最新技术方法;在智能化信息处理方面,先回顾了神经网络和遗传算法在遥感图像处理中的应用,然后介绍了人工免疫系统对多、高光谱遥感影像分类研究的最新进展.     

An ANN–RTS smoother scheme for accurate INS/GPS integrated attitude determination

Digital mobile mapping, the method that integrates digital imaging with direct geo-referencing, has developed rapidly over the past 15 years. The Kalman filter (KF) is considered an optimal estimation tool for real-time INS/GPS integrated kinematic positioning and orientation determination. However, the accuracy requirements of general mobile mapping applications cannot be easily achieved even when using the KF scheme. Therefore, this study proposes an intelligent scheme combining ANN and RTS backward smoother to overcome the limitations of KF and to enhance the overall accuracy of attitude determination for tactical grade and MEMS INS/GPS integrated systems.

人工神经网络在南海近海面气温反演中的应用研究

基于人工神经网络方法,利用海面水温、海面风速以及海面气压反演南海近海面气温,采用的基础数据集是国际综合海洋-大气数据集(International Comprehensive Ocean-Atmosphere Data Set,2.4 Release,ICOADS2.4)1981—2008年的观测资料,其中1981—2000年的观测资料用来建立模型,2001—2008年的观测资料用来进行模型检验。采用的人工神经网络方法是引入动量因子并采用批处理梯度下降法的BP(Back propagation)算法。试验结果表明,基于人工神经网络建立的近海面气温反演方法明显优于多元线性回归方法,尤其是在春季和冬季,海面水温、海面风速以及海面气压与近海面气温之间存在较强的非线性关系,人工神经网络的优势更加明显。总体而言,人工神经网络在各月的反演效果较均衡,均方根误差介于1.5—1.8℃之间,平均绝对误差为1.1—1.3℃。