直流激电测深二维反演的若干问题研究

目前,激电数据的二维反演主要针对高密度激电数据,而对于常规的激电测深则很少涉及.据此,针对常规激电测深二维反演所涉及到的一些技术难点进行探讨.在正演过程中,根据实测数据实现有限元网格自动剖分,并采用双网格系统,用相邻节点电导率双线性连续进行有限元模拟,并对模拟的误差进行校正.在反演过程中,利用互换原理和Broyden的方法计算偏导数矩阵;采用与奇异值分解算法精度相当的共轭梯度算法求解偏导数矩阵方程.最后通过对模拟数据和实测数据进行反演试算,验证本计算方法是可行的.     

岩体分级考虑因素的现状与趋势分析

在对国内外岩体分级方法深入研究的基础上,对岩体分级方法中所考虑的岩体分级因素及对各因素的处理方法进行了系统的归纳和总结。从岩体分级方法的现状来看,虽然目前尚无统一的岩体分级标准,但在岩体分级中应根据岩石的强度、岩体的完整性、地下水条件、地应力状况等多方面因素,进行岩体综合分级上达成了共识,并且国内规范中的岩体分级标准有趋于统一和向国际标准接轨的趋势。     

The ACPI Climate Change Simulations

The Parallel Climate Model (PCM) has been used in the Accelerated ClimatePrediction Initiative (ACPI) Program to simulate the global climateresponse to projected CO₂, sulfate, and other greenhouse gasforcingunder a business-as-usual emissions scenario during the 21st century. In these runs, the oceans were initialized to 1995 conditions by a group from the Scripps Institution of Oceanography and other institutions. An ensemble of three model runs was then carried out to the year 2099 using the projected forcing. Atmospheric data fromthese runs were saved at 6-hourly intervals (hourly for certain criticalfields) to support the ACPI objective of accurately modeling hydrologicalcycles over the western U.S. It is shown that the initialization to1995 conditions partly removes the un-forced oceanic temperature and salinity drifts that occurred in the standard 20th century integration. The ACPI runs show a global surface temperature increase of 3–8 °C over northern high-latitudes by the end of the 21st century, and 1–2 °C over the oceans. This is generally within ±0.1°Cof model runs without the 1995 ocean initialization. The exception is in theAntarctic circumpolar ocean where surface air temperature is cooler in theACPI run; however the ensemble scatter is large in this region. Althoughthe difference in climate at the end of the 21st century is minimalbetween the ACPI runs and traditionally spun up runs, it might be largerfor CGCMs with higher climate sensitivity or larger ocean drifts. Ourresults suggest that the effect of small errors in the oceans (such asthose associated with climate drifts) on CGCM-simulated climate changesfor the next 50–100 years may be negligible.     

Improved methods for estimating equilibrium climate sensitivity from transient warming simulations

Climate Dynamics - Equilibrium climate sensitivity (ECS) refers to the total global warming caused by an instantaneous doubling of atmospheric CO2 from the pre-industrial level in a climate system....     

基于长短期记忆神经网络的风速超短期快速滚动预报技术

利用甘肃省某风电场2017—2020年测风数据,基于长短期记忆神经网络(LSTM)模型,通过评估不同输入数据和模型时间窗口长度下的预报精度,设计一套适用于风电场的风速超短期快速滚动预报方案。结果表明：通过输入不同的特征变量,在风速的超短期(未来4 h内)预报中,风速自身变化起主导作用,模型输入变量中只加入各高度层的风速能得到更好的模拟效果。通过评估LSTM模拟时间窗口长度L对模拟效果的影响,当时间窗口长度L≤24 h时,模拟效果较好,说明超短期风速变化主要和风速自身临近时刻的变化有关;当L>24 h时,模拟效果快速下降,说明过长的L会削弱模拟能力,降低模拟精度。通过分析LSTM在未来4 h内的风速模拟能力,发现随着预报时长的增加,模拟精度逐步下降,但在未来2 h内的风速均方根误差RMSE均小于2 m·s^-1,结果较为理想,且该方法对计算资源要求不高,经济实用性强,在业务中具有较高的应用潜力。     

长短期记忆神经网络(LSTM)模型在低能见度预报中的应用

利用浙江省义乌市2015—2019年逐小时气象观测数据(相对湿度、风速、地气温差、能见度)和空气质量指数(Air Quality Index, AQI)数据, 分析了义乌地区低能见度天气(观测能见度lt; 10 km)的分布特征和气象要素条件。利用长短期记忆神经网络(Long Short Term Memory Neural Network, LSTM)模型对逐小时能见度进行模拟, 分别对比了观测能见度作为输入变量与否的模拟效果; 根据义乌地区低能见度天气条件的特征, 将模拟时段分为三个时期(11月至翌年2月, 3—6月, 7—10月), 对比了分时期模拟的效果; 以及评估了模型的预报步长。结果表明: 高湿、高污染、气温高于地温和低风速是义乌地区低能见度天气的主要特征。LSTM模型对单站能见度有较好的模拟效果, 当输入参数中加入历史观测能见度时, 能大幅提高模拟准确度, 日均能见度模拟结果均方根误差RMSE=0.63 km, 平均绝对误差MAE=0.51 km, 拟合优度R²=0.99;分时期进行模拟能得到更精准的模拟结果。本研究中选用的输入要素在冬季(11月至翌年2月)模拟效果最好, RMSE=2.35 km, MAE=1.46 km, 低能见度均方根误差RMSE_10 km=1.81 km, 低能见度平均绝对误差MAE_10 km=1.13 km, R²=0.83; 3—6月的模拟中, 输入变量中不加AQI模拟效果更好, 这意味着3—6月义乌地区的低能见度天气以雾天气为主导, 加入过多变量并不一定能提高模型准确度; 随着预报步长增大, 模型预报效果变差, 预测步长等于3 h, R²=0.71, 预测结果已不具备实际应用意义。     

ECC方法在中国夏季气温预测研究中的应用

以中国夏季气温为预测对象,选取东亚地区冬季500 h Pa高度场、海平面气压场、地表温度场和850 h Pa温度场为预测因子,采用1951~2009年去趋势处理后的资料,通过变形的典型相关分析(Barnett-Preisendorfer Canonical Correlation Analysis,BP-CCA)方法分别建立单因子预测模型,再利用集合典型相关分析(Ensemble Canonical Correlation,ECC)方法建立集合预测模型,对中国夏季气温进行基于交叉检验方法的预测试验,然后利用2010~2014年的资料对中国夏季气温进行独立样本检验。通过分析BP-CCA模态可知,一对BP-CCA模态的空间型在一定程度上可以反映预报因子场和对象场的遥相关特征。通过基于交叉检验方法的预测试验表明环流场和热力场均能为气温提供预测信息。ECC预测模型综合了各个预报因子的在不同地区的预报技巧,比单因子BP-CCA预测模型有更高、更稳定的预报技巧。独立样本检验表明ECC模型与单因子BP-CCA预测模型相比,对中国夏季气温有更高、更稳定的实际预测能力,对气温季节预测具有参考价值。     

Trends in northern midlatitude atmospheric wave power from 1950 to 2099

In a warming climate, atmospheric wave activity and associated weather patterns may change, although conflicting results have been reported on this topic. Additionally, atmospheric wave changes in a future climate have mainly focused on waves of a specified spatial scale, rather than a particular spatiotemporal scale. Here, changes in the variability of Rossby waves of multiple spatiotemporal scales are analyzed using the wavenumber-frequency power spectrum, a tool commonly applied to analyze atmospheric equatorial waves. Daily 500 hPa geopotential height data over 40°–60°N from historical (1950–2005) and future (2006–2099) simulations from 20 models in the Coupled Model Intercomparison Project Phase 5 (CMIP5) under the RCP8.5 scenario were analyzed. When compared to the historical period, the late 21st century climate projections showed a decline in spectral power for both eastward and westward propagating waves with wavenumbers greater than 8 that spanned over all frequencies in all seasons, but an increase in mean power for eastward propagating waves with wavenumbers 1–7 over all frequencies was shown in winter and spring. This increase in power was accompanied by increased variance, i.e., an increased meridional extent of 500 hPa ridges and troughs, and was the result of increases in the mean number of high amplitude events and duration of activity within this wave band. These results indicate that large-scale (~ 104 km) eastward propagating weather systems may intensify with higher amplitudes for ridges and troughs, while short-scale (102–103 km) weather systems may decrease in their intensity due to reduced variability in the late 21st century under the high emissions scenario. Potential mechanisms for these changes are discussed, including enhanced Arctic warming and midlatitude-tropical interactions.     

雨滴谱仪与风廓线雷达反射率对比试验

针对风廓线雷达估算的反射率数据需要进行验证。开展雨滴谱仪与风廓线雷达反射率对比试验,通过两种不同探测设备观测数据的对比,以验证风廓线反射率数据的可靠性和可用性。结果表明:确定了以风廓线低模360~1440 m采样体积内的反射率与3 min雨滴谱反射率数据对比方法能最大程度的减少时空差异;在雨滴谱仪反射率小于40 dBz时,对应的风廓线雷达反射率数据是可靠和可用的;同时由于风廓线雷达有限的动态范围造成反射率低估的现象,使得风廓线雷达反射率在大气垂直结构以及微物理特性等方面应用受到一定的局限性。     

力觉临场感遥操作机器人(1):技术发展与现状

人机交互式机器人作为最具实用价值的特种机器人已成为当前机器人学研究的前沿和热点.临场感(Telepresence)技术是人机交互的核心.首先,回顾了力觉临场感遥操作机器人技术的产生、发展和现状,介绍了力觉临场感遥操作机器人在核领域、空间探测领域与远程医疗领域的应用情况;其次,对力觉临场感遥操作机器人的4大关键技术:传感技术、力反馈与触觉再现技术、大时延控制技术和虚拟预测环境建模技术等进行了综述;还介绍了东南大学仪器科学与工程学院机器人传感与控制研究所近20年来开展临场感遥操作机器人技术研究,以及在核探测、康复医疗领域应用的情况.通过回顾与分析,指出了力觉临场感遥操作机器人技术今后需要研究的几个重点问题.