预测有效波高的深度学习模型研究

研究基于RNN、LSTM、GRU深度学习模型,针对NOAA浮标数据集中的44013、44014、44017浮标的数据,通过斯皮尔曼相关性分析提高模型预测效果。实验结果表明,在进行相关性分析后,S-RNN、S-LSTM、 S-GRU的预测效果均比原始RNN、LSTM、GRU模型预测效果好。此外,提出一种基于LSTM的LSTM-Attention 波高预测模型,并进行相关实验,量化LSTM-Attention模型的预测效果,实验结果表明LSTM-Attention模型有更好的预测效果。为评估模型的泛化能力,研究还提出了一种采用邻近浮标数据进行学习,预测浮标缺失数据的方 法。实验结果表明,该方法的预测精度可以达到97.93%。本研究为海浪预测提供了新的方法和思路,也为未来深 度学习模型在海浪预测中的应用提供了参考。     

葛洲坝至胭脂坝长江干流段江豚活动规律

长江江豚是我国一级保护野生动物,也是长江生态系统的指示性物种,其种群数量恢复情况及活动规律一直备受关注。本文选择长江宜昌段江豚活动频繁的葛洲坝下游至胭脂坝江段为研究区域,于2021年6月-2022年5月采用定点目测、水上流动监测与无人机监测相结合的方法分区监测长江江豚的游泳行为,分析长江江豚的活动规律,构建了长江江豚出水头次的零膨胀泊松回归模型,识别影响长江江豚出水头次的关键因素,建立长江江豚出水头次与各因素间的相关关系。结果表明:(1)葛洲坝至胭脂坝长江干流段监测到最大江豚出水头次为19头次,长江江豚集群规模以2~3头最为常见,占比达58.1%;长江江豚主要表现出4种行为特征,各行为占比从大到小依次为:玩耍＞休息＞摄食＞抚幼。(2)长江江豚在葛洲坝下游近坝区(A区,葛洲坝至至喜长江大桥)出水头次最多,且在秋冬季节累计出水头次多于春夏季节,冬季累计出水头次最高,达252头次。(3)长江江豚出水头次与电站下泄流量呈显著负相关;浊度增大,长江江豚出水的概率减小。本文研究结果对长江江豚生态保护策略及长江十年禁渔效果评估具有重要参考价值。     

GM4型磁通门磁力仪性能检验方法

目前,磁通门磁力仪广泛应用于观测地球磁场相对变化的仪器。相对于应用,仪器的性能测试不为人们所熟悉。本文针对GM4型磁通门磁力仪的噪声特性,频率特性,线性度,温度特性等关键性能指标的测试依据、测试方法及流程给予详细的阐述。     

基于新生代活动构造背景的地震重点危险区域综合研究——以安徽为例

基于地震的发生与断裂的活动性密切相关,为了科学合理并尽可能准确地判定未来地震重点危险区域,以安徽为例,利用由深浅部对比、宏微观构造解析所给出的新生代构造活动性最新研究成果,结合地震活动性资料研究结果,对一些重点部位的地震危险性进行了综合研究。结果表明,根据长时间尺度的地质构造最新研究揭示的活动性较强区域,与根据近代地震活动性资料研究反映的危险性较大区域,呈现较好的一致性特征,从而为未来地震重点危险区域的判定提供了新的较为可靠的依据和方法。     

Application research of narrow band Internet of things buoy and surface hydrodynamics monitoring

This paper applies the narrow band Internet of things communication technology to develop a wireless network equipment and communication system, which can quickly set up a network with a radius of 100 km on water surface. A disposable micro buoy based on narrow-band Internet of things and Beidou positioning function is also developed and used to collect surface hydrodynamic data online. In addition, a web-based public service platform is designed for the analysis and visualization of the data collected by buoys. Combined with the satellite remote sensing data, the study carries a series of marine experiments and studies such as sediment deposition tracking and garbage floating tracking.     

Function of spa development in countryside for domestic tourism in China

The common problems on tourism of developing countries have often been researched by west researchers until now, which can be concluded into special words: “two sectors”, “enclave construction”, “government’s interference”, “developing scale”, and “the industrial role of tourism” etc. In fact all the researches were done from the areas with the patronage of foreign visitors, how is the situation of domestic tourism was ignored in some points. In this paper, another developing model compared with the one with foreign visitors was presented to the readers on the case studies of countryside spas in China. Through this research, one aspect of Chinese real leisure life can be understood to some extent, and the implying that for bigger developing countries the domestic tourism can match the effect of international tourism should not be ignored at least to China. As the developing of economy, domestic tourism will finally unite with international one together, and the common problems will disappear step by step, therefore domestic tourism and international tourism of developing countries should be researched equally since now. Biography: WANG Yan-ping (1962–), male, a native of Heilongjiang Province, is a doctorate student of Chiba University of Japan. His research insterest is environmental impact assessment.     

关于人才效益的几点思考

阐述人才效益对经济发展的促进作用及提高人才效益的基本途径。     

The precipitation infiltration and runoff recharging experiment observation in the Taklimakan Desert

１　ＩＮＴＲＯＤＵＣＴＩＯＮＴｈｅｌｏｃａｌｐｒｅｃｉｐｉｔａｔｉｏｎｉｓｍｏｓｔｉｍｐｏｒｔａｎｔａｖａｉｌａｂｌｅｆｒｅｓｈｗａｔｅｒｓｏｕｒｃｅｉｎｄｅｓｅｒｔｓ．ＴｈｅＴａｋｌｉｍａｋａｎＤｅｓｅｒｔｂｅｌｏｎｇｓｔｏａｎｅｘｔｒｅｍｅａｒｉｄｂｅｌｔ．Ａｃｃｏｒｄｉｎｇｔｏａｆｅｗｙｅａｒｓ’ｌｉｍｉｔｅｄｏｂｓｅｒｖａｔｉｏｎｄａｔａ,ｉｔｉｓｅｓｔｉｍａｔｅｄｔｈａｔｔｈｅｐｒｅｃｉｐｉｔａｔｉｏｎｓｈｏｕｌｄｂｅｉｎｔｈｅｒａｎｇｅｏｆ３０－５０ｍｍａｎｎｕａｌｌｙ．Ｈｏｗｅｖ…     

地基微波辐射计探测在黄河上游人工增雨中的应用

利用2003、2004年夏秋季在青海省河南县的地基双频段微波辐射计连续观测资料,在实施系统探测实验以进行辐射亮温值(TB(23.87),TB(31.65))与大气汽态总水汽含量(Q)和云中积分液态水含量(L)值反演处理的基础上,分析了黄河上游河曲地区的云水特征,并进行了降水预测及人工增雨作业指标的探讨。结果表明:在黄河上游河曲地区,7—9月纯晴天无云天气条件下,L值基本为0,表明了统计回归反演的显著性。多云条件下Q值和L值分别在3.76~4.75g·cm-2、227.34~859.42g·m-2的范围内。可降水云天气,Q值在5.23~8.65g·cm-2间,L值在421.18~1016.37g·m-2的范围内;积雨云个例分析表明,在降雨开始前近5小时的降雨酝酿期内,Q及L的增加有明显的波动,但在对流云出现并发展时,Q和L总是急剧上升,在降雨前达到一峰值;由河南站和西安站的Q、L值比较差异可以看出,Q、L值受水汽输送、地形等因素的影响,黄河上游河曲地区的人工增雨潜力有显著的优势;所计算出的降水预测阈值,可作为该地区人工增雨作业指标的参考。