Supervised classification of electric power transmission line nominal voltage from high-resolution aerial imagery

For many researchers, government agencies, and emergency responders, access to the geospatial data of US electric power infrastructure is invaluable for analysis, planning, and disaster recovery. Historically, however, access to high quality geospatial energy data has been limited to few agencies because of commercial licenses restrictions, and those resources which are widely accessible have been of poor quality, particularly with respect to reliability. Recent efforts to develop a highly reliable and publicly accessible alternative to the existing datasets were met with numerous challenges – not the least of which was filling the gaps in power transmission line voltage ratings. To address the line voltage rating problem, we developed and tested a basic methodology that fuses knowledge and techniques from power systems, geography, and machine learning domains. Specifically, we identified predictors of nominal voltage that could be extracted from aerial imagery and developed a tree-based classifier to classify nominal line voltage ratings. Overall, we found that line support height, support span, and conductor spacing are the best predictors of voltage ratings, and that the classifier built with these predictors had a reliable predictive accuracy (that is, within one voltage class for four out of the five classes sampled). We applied our approach to a study area in Minnesota.     

Winkler地基梁在温度场中受简谐激励的主共振分析

研究W inkler地基梁在温度场中受简谐激励作用的主共振问题。应用弹性理论和Galerk in方法建立W inkler地基梁在温度场中受简谐激励作用的非线性动力方程。应用非线性振动的多尺度法,求得系统主共振的近似解。分析不同参数对主共振响应曲线的影响。     

钙质土中重力串锚水平承载力特性有限元研究

目前没有规范给出串锚的水平承载力计算公式,为了利用有限元法研究串锚水平承载力特性,采用相关理论计算以及室内试验的方法来验证有限元法模拟计算的可靠性。实践证明,有限元法具有较高的计算精度。利用有限元分析软件ABAQUS建立了串锚-土体模型,模拟串锚在水平荷载作用下破坏过程,研究了串锚在水平荷载作用下破坏机理,从而得到了串锚水平承载力特性。在相同的工况条件下,串锚的水平承载力与其锚链长度有关,在一定范围内,增加锚链的长度可以提高串锚的水平承载力;串锚的水平承载力不是相应的单个锚的水平承载力相加,其水平承载力小于相应的单个锚水平承载力之和,在进行串锚水平承载力设计计算时应给与相应的考虑。     

梅雨降水季节预测的多方法比较

基于1961—2000年逐月降水观测资料和全球大气再分析资料,分析了6—7月长江中下游(108°～123°E,27°～33°N)梅雨的时空分布特征。通过观测诊断和数值试验确定了影响梅雨异常偏多的3个前期因子:4—5月平均的西北太平洋海平面气压正异常;3月至5月北大西洋海平面气压负变压倾向;1月至4月西伯利亚的2 m温度负倾向。利用这3个具有物理意义的影响因子构建了梅雨季节预测模型,该模型在训练期(1961—2000年)和独立预测期(2001—2022年)均具有显著的预测技巧(相关系数分别为0.79和0.77,均方根误差分别为0.59和0.68)。同时,基于相似的潜在预测因子,对比了利用偏最小二乘回归方法和5种机器学习方法(随机森林、轻量级梯度提升机、自适应提升、类别型特征提升、极端梯度提升)建立的预测模型的技巧。虽然训练期(1961—2000年)偏最小二乘回归和机器学习建模拟合效果更高,但在独立预测期(2001—2022年)上述模型的预测技巧显著降低(相关系数均低于0.44,均方根误差均大于0.93),出现了明显的过拟合问题。本研究强调梅雨的短期气候预测应建立在物理机制基础之上,而使用机器学习方法需谨慎。     

Interaction of water waves with vertical cylinders using null-field integral equations

The scattering of water waves by bottom-mounted vertical circular cylinders is solved by using the null-field integral equations in conjunction with degenerate kernels and Fourier series to avoid calculating the Cauchy and Hadamard principal values. In the implementation, the null-field point can be exactly located on the real boundary owing to the introduction of degenerate kernels for fundamental solutions. An adaptive observer system of polar coordinates is considered to fully employ the properties of degenerate kernels. For the hypersingular equation, vector decomposition for the radial and tangential gradients is carefully considered. This method can be seen as a semi-analytical approach since errors attribute from the truncation of Fourier series. Neither hypersingularity in the Burton and Miller approach nor the CHIEF concept was required to deal with the problem of irregular frequencies. Five advantages of free of calculating principal value, well-posed algebraic system, convergence rate of exponential order, meshfree and elimination of boundary-layer effect, are achieved by using the present approach. Numerical results are given for the forces and free-surface elevation around the circular boundaries. Also, the near-trapped behavior arisen from the physical resonance is detected. A general-purpose program for water wave impinging several circular cylinders with arbitrary number, radii, and positions was developed. Several examples of water wave structure interaction by vertical circular cylinders were demonstrated to see the validity of the present formulation.     

Extending self-organizing maps for supervised classification of remotely sensed data

An extended self-organizing map for supervised classification is proposed in this paper. Unlike other traditional SOMs, the model has an input layer, a Kohonen layer, and an output layer. The number of neurons in the input layer depends on the dimensionality of input patterns. The number of neurons in the output layer equals the number of the desired classes. The number of neurons in the Kohonen layer may be a few to several thousands, which depends on the complexity of classification problems and the classification precision. Each training sample is expressed by a pair of vectors ： an input vector and a class codebook vector. When a training sample is input into the model, Kohonen＇s competitive learning rule is applied to selecting the winning neuron from the Kohouen layer and the weight coefficients connecting all the neurons in the input layer with both the winning neuron and its neighbors in the Kohonen layer are modified to be closer to the input vector, and those connecting all the neurons around the winning neuron within a certain diameter in the Kohonen layer with all the neurons in the output layer are adjusted to be closer to the class codebook vector. If the number of training sam- ples is sufficiently large and the learning epochs iterate enough times, the model will be able to serve as a supervised classifier. The model has been tentatively applied to the supervised classification of multispectral remotely sensed data. The author compared the performances of the extended SOM and BPN in remotely sensed data classification. The investigation manifests that the extended SOM is feasible for supervised classification.     

基于迁移学习方法的山东冷云冰晶特征分布研究

为分析冷云中冰晶的分布特征,揭示冰晶增长演变机制,根据冰晶形状和尺度特征分为8类并进行标注,同时标注1类隔断栏进行数据质量控制,将9类标签图像整合并建立图像集,利用迁移学习VGGNet16方法进行识别训练,经训练模型分类准确率达98%。将模型应用到秋季冷云冰晶特征研究中,选取3次积层混合云和3次层状云降水过程,分析冰晶形状在不同温度区间的占比及冰晶谱变化特征,结果表明,温度通过影响冰晶基面与棱面的比值来决定冰晶初始形状分布,相同温度区间积层混合云内球状冰晶和线型冰晶占比高于层状云,低于-12℃后各类冰晶占比相对固定;积层混合云内线型冰晶直径集中在300~800 μm,冰晶谱呈多峰分布,聚合体直径大于600 μm,冰晶谱首尾两端浓度相当,球状冰晶直径集中在120～300 μm,冰晶谱呈单调下降趋势。     

多种模式降水预报的稳定性特征研究

预报的稳定性是指对同一时段在不同时间发布的多时效预报结论的一致性,是模式预报质量的一个重要方面,较大的不稳定性会给使用者造成困扰。为深入了解业务常用模式的稳定性,使用相对标准偏差指标计算不同时效预报的降水量波动大小,并改进了Flip-Flop指数（改进后简称FFnorm）,计算多时效降水量预报变化趋势的翻转程度,衡量预报变化趋势的稳定性,对2种全球模式（ECMWF、NCEP-GFS）、3种区域模式（CMA-MESO、CMA-SH9、HHUPS-ST）,在中国6个气候分区中降水预报的稳定性进行对比分析,分为实况有降水和暴雨及以上降水2种情况进行了讨论。结果表明：实况有降水时,相对区域模式来说,全球模式的多时效降水预报的相对标准偏差较小,即模式降水量预报的波动较小;各模式对西南区的西部、东北区的东部以及华南区的南部预报的波动性相对较小,西北区的西部波动性较大。就多时效降水量预报变化趋势而言,2种情况下均为CMA-MESO、NCEP-GFS和 ECMWF的稳定性较好,其FFnorm指数小于HHUPS-ST和CMA-SH9模式,其中CMA-MESO对西南区、华南部分地区降水量预报变化趋势的稳定性较为突出;CMA-SH9的指数最大,多时效降水量预报变化趋势稳定性较差;各模式对长江中下游地区的FFnorm指数相对较大,多时效预报趋势的稳定性较差。有降水时,CMA-MESO随时效临近的降水量预报变化趋势稳定（单调递增、单调递减或不变）的频次最多,其次是NCEP-GFS,2种降水情况下,该2种模式的降水量预报均为随时效临近单调递增次数大于递减次数,且CMA-MESO单调递增特征尤其显著。以上特征能够为模式调试和预报决策提供参考。     

基于机器学习的格点气温预报订正方法

使用2017年9月至2021年3月国家级业务化运行的智能网格实况分析产品和欧洲中期天气预报中心全球模式(EC)产品,根据湖北省的地理分布特征构建6个分区,采用基于LightGBM机器学习算法建立的气温预报方法,生成湖北省0.05°×0.05°格点气温预报产品。利用2021年4—9月的预报产品和格点实况资料进行检验,结果表明:基于机器学习的气温预报方法(MLT)取得了较好的预报效果,其在0～72 h时效内优于中央气象台下发的气温精细化指导预报(SCMOC)和EC产品;MLT在山区的误差较平原大,但山区的订正幅度大于平原,日最高气温的订正幅度大于日最低气温的订正幅度;4—9月MLT、SCMOC、EC产品的平均绝对误差(MAE)日变化都呈现了白天偏高、夜间偏低、午后凸起的单峰特征,MLT的MAE值较SCMOC和EC产品的更低,并且在转折性天气中仍具有优势;站点检验与格点检验结论一致,基于格点建模的气温预报产品对站点预报同样得到了订正。机器学习在格点气温的模式订正方面可以作为一个行之有效的手段。     

Multiple episodes of tectonothermal processes in the Eastern Ghats granulite belt

A suite of rocks from Borra Carbonate Granulite Complex (BCGC) in the Eastern Ghats granulite belt displays superposed structures and overprinted mineral assemblages that reveal multiple episodes of tectonothermal reworking of the complex under granulite facies condition. Five distinct episodes of deformation (D₁, D₂, D₃, D₄ and D₅) and four phases of metamorphism (M₁, M₂, M₃ and M₄) are recorded. The signature of the earliest tectonothermal event, D₁ is a gneissic foliation (S₁) denned by segregation of peak granulite facies mineral assemblages corresponding to prograde M₁ metamorphism. M₂ metamorphic overprint represents an episode of near-isobaric cooling of the complex under a static condition. D₂ represents an episode of ductile deformation manifested by isoclinal folding (F₂) and associated extensional structures, within a broad framework of coaxial bulk deformation. The present study reveals that D₂ took place subsequent to M₂ - Subsequent deformation, D₃, produced F₃ folds and also deformations of boudins formed during D₂. M₃, which is synchronous with F₃, represents a near isothermal decompression of the BCGC. This was followed by a weak structural readjustment (D₄), producing E-W cross folds. The latter was not, however, associated with any recognizable petrological reworking. In the terminal events, deformation (D₅) and mineral reactions (M₄) were localized along narrow intersecting shear zones. The latter acted as channelways for carbonic and still later hydrous fluid infiltration. The available thermobarometric data from BCGC and other areas of the Eastern Ghats belt reveal that reworking during M₂ and M₃ ensued in a thermally perturbed regime. The high thermal regime might also have persisted during carbonic fluid infiltration related to terminal reworking (M₄).