To find out the origin of dolomite, the precipitation of primary dolomite, and the formation of pores in dolomite, petrologic and geochemical characteristics of typical samples from Sichuan and Tarim Basin were analyz...To find out the origin of dolomite, the precipitation of primary dolomite, and the formation of pores in dolomite, petrologic and geochemical characteristics of typical samples from Sichuan and Tarim Basin were analyzed based on the previous understandings, and three aspects of results were achieved.(1) A classification of dolomite origins based on petrologic features, forming environment, and time sequence was proposed, which shows clear boundaries of diagenetic and characteristic realms and evolved clues between different types of dolomite.(2) Petrographic and geochemical identification marks for different types of dolomite were presented, revealing that the orderly geochemical variation of different types of dolomite is the response to the change of forming environment of dolomite during continuous time sequence.(3) The contribution of dolomitization to the formation of porosity was re-evaluated, revealing that the porosity in dolomite was mostly attributed to the primary pores and supergene dissolution and burial dissolution, and early dolomitization was conducive to the preservation of primary pores. These understandings are of great theoretical significance for identifying the origins and types of dolomite, and can guide the prediction of dolomite reservoirs.展开更多
随着网络技术的广泛应用,出现了多种多样的网络攻击,其中,分布式拒绝服务(Distributed Denial of Service,DDoS)攻击的危害性较大。将12种DDoS攻击的数据与正常数据流混在一起后难以区分,因此防御DDoS攻击的关键是对其进行有效区分。文...随着网络技术的广泛应用,出现了多种多样的网络攻击,其中,分布式拒绝服务(Distributed Denial of Service,DDoS)攻击的危害性较大。将12种DDoS攻击的数据与正常数据流混在一起后难以区分,因此防御DDoS攻击的关键是对其进行有效区分。文章提出一种基于RNN区分DDoS攻击类型的方法。以循环神经网络(Recurrent Neural Network,RNN)为研究改进对象,运用了模型的模块化研究方法和技术,抽象出3类简单模块组合形成RNN-IDDoS模型,该模型具有5层、3种时间步。在公开数据集上进行实验,实验结果表明该模型的准确率可达99.8%,优于当前其他3种模型,取得了很好的区分效果。展开更多
基金Supported by the China National Science and Technology Major Project(2016ZX05004-002)
文摘To find out the origin of dolomite, the precipitation of primary dolomite, and the formation of pores in dolomite, petrologic and geochemical characteristics of typical samples from Sichuan and Tarim Basin were analyzed based on the previous understandings, and three aspects of results were achieved.(1) A classification of dolomite origins based on petrologic features, forming environment, and time sequence was proposed, which shows clear boundaries of diagenetic and characteristic realms and evolved clues between different types of dolomite.(2) Petrographic and geochemical identification marks for different types of dolomite were presented, revealing that the orderly geochemical variation of different types of dolomite is the response to the change of forming environment of dolomite during continuous time sequence.(3) The contribution of dolomitization to the formation of porosity was re-evaluated, revealing that the porosity in dolomite was mostly attributed to the primary pores and supergene dissolution and burial dissolution, and early dolomitization was conducive to the preservation of primary pores. These understandings are of great theoretical significance for identifying the origins and types of dolomite, and can guide the prediction of dolomite reservoirs.
