Landslide displacement prediction based on optimized empirical mode decomposition and deep bidirectional long short-term memory network

There are two technical challenges in predicting slope deformation.The first one is the random displacement,which could not be decomposed and predicted by numerically resolving the observed accumulated displacement and time series of a landslide.The second one is the dynamic evolution of a landslide,which could not be feasibly simulated simply by traditional prediction models.In this paper,a dynamic model of displacement prediction is introduced for composite landslides based on a combination of empirical mode decomposition with soft screening stop criteria(SSSC-EMD)and deep bidirectional long short-term memory(DBi-LSTM)neural network.In the proposed model,the time series analysis and SSSC-EMD are used to decompose the observed accumulated displacements of a slope into three components,viz.trend displacement,periodic displacement,and random displacement.Then,by analyzing the evolution pattern of a landslide and its key factors triggering landslides,appropriate influencing factors are selected for each displacement component,and DBi-LSTM neural network to carry out multi-datadriven dynamic prediction for each displacement component.An accumulated displacement prediction has been obtained by a summation of each component.For accuracy verification and engineering practicability of the model,field observations from two known landslides in China,the Xintan landslide and the Bazimen landslide were collected for comparison and evaluation.The case study verified that the model proposed in this paper can better characterize the"stepwise"deformation characteristics of a slope.As compared with long short-term memory(LSTM)neural network,support vector machine(SVM),and autoregressive integrated moving average(ARIMA)model,DBi-LSTM neural network has higher accuracy in predicting the periodic displacement of slope deformation,with the mean absolute percentage error reduced by 3.063%,14.913%,and 13.960%respectively,and the root mean square error reduced by 1.951 mm,8.954 mm and 7.790 mm respectively.Conclusively,this model not only has high prediction accuracy but also is more stable,which can provide new insight for practical landslide prevention and control engineering.

Diffraction deep neural network based orbital angular momentum mode recognition scheme in oceanic turbulence

Orbital angular momentum(OAM)has the characteristics of mutual orthogonality between modes,and has been applied to underwater wireless optical communication(UWOC)systems to increase the channel capacity.In this work,we propose a diffractive deep neural network(DDNN)based OAM mode recognition scheme,where the DDNN is trained to capture the features of the intensity distribution of the OAM modes and output the corresponding azimuthal indices and radial indices.The results show that the proposed scheme can recognize the azimuthal indices and radial indices of the OAM modes accurately and quickly.In addition,the proposed scheme can resist weak oceanic turbulence(OT),and exhibit excellent ability to recognize OAM modes in a strong OT environment.The DDNN-based OAM mode recognition scheme has potential applications in UWOC systems.

DeephitTM:医学生存分析的时间相关性深度学习模型

生存分析是医学中经常用到的一种健康预测方法,越来越多的学者开始采用深度学习的方法对生存分析问题进行建模以得到更好的预测结果.目前已有的方法都假设风险和时间的联合概率是无关联的.然而生存分析数据的实际结果中却包含时间因素,这就无法保证不同时刻得到的风险概率是无关联的.本文提出一种带有时间相关性的深度学习模型DeephitTM,该模型对已有的深度学习模型Deephit进行了改进.实验结果表明,在不同的数据集上,改进后的模型的性能相比于原模型能够提升1到3个百分点.

Dynamic impact properties of deep sandstone under thermal-hydraulicmechanical coupling loads

The deep rock mass within coal mines situated in a challenging environment are characterized by high ground stress,high geotemperature,high osmotic water pressure,and dynamic disturbances from mechanical excavation.To investigate the impact of this complex mechanical environment on the dynamic characteristics of roof sandstone in self-formed roadways without coal pillars,standard specimens of deep sandstone from the 2611 upper tunnel working face of the Yongmei Company within the Henan Coal Chemical Industry Group in Henan,China were prepared,and an orthogonal test was designed.Using a self-developed geotechnical dynamic impact mechanics test system,triaxial dynamic impact tests under thermal-hydraulicmechanical coupling conditions were conducted on deep sandstone.The results indicate that under high confining pressure,deep sandstone exhibits pronounced brittle failure at low temperatures,with peak strength gradually decreasing as temperature and osmotic water pressure increase.Conversely,under low confining pressure and low temperature,the brittleness of deep sandstone weakens gradually,while ductility increases.Moreover,sandstone demonstrates higher peak strength at low temperatures under high axial pressure conditions,lower peak strength at high temperatures,and greater strain under low axial pressure and high osmotic water pressure.Increases in impact air pressure and osmotic water pressure have proportionally greater effects on peak stress and peak strain.Approximately 50%of the input strain energy is utilized as effective energy driving the sandstone fracture process.Polar analysis identifies the optimal combination of factors affecting the peak stress and peak strain of sandstone.Under the coupling effect,intergranular and transgranular fractures occur within the sandstone.SEM images illustrate that the damage forms range from minor damage with multiple fissures to extensive fractures and severe fragmentation.This study elucidates the varied dynamic impact mechanical properties of deep sandstones under thermal-hydraulic-mechanical coupling,along with multifactor analysis methods and their optimal factor combinations.

RFFsNet-SEI:a multidimensional balanced-RFFs deep neural network framework for specific emitter identification

Existing specific emitter identification(SEI)methods based on hand-crafted features have drawbacks of losing feature information and involving multiple processing stages,which reduce the identification accuracy of emitters and complicate the procedures of identification.In this paper,we propose a deep SEI approach via multidimensional feature extraction for radio frequency fingerprints(RFFs),namely,RFFsNet-SEI.Particularly,we extract multidimensional physical RFFs from the received signal by virtue of variational mode decomposition(VMD)and Hilbert transform(HT).The physical RFFs and I-Q data are formed into the balanced-RFFs,which are then used to train RFFsNet-SEI.As introducing model-aided RFFs into neural network,the hybrid-driven scheme including physical features and I-Q data is constructed.It improves physical interpretability of RFFsNet-SEI.Meanwhile,since RFFsNet-SEI identifies individual of emitters from received raw data in end-to-end,it accelerates SEI implementation and simplifies procedures of identification.Moreover,as the temporal features and spectral features of the received signal are both extracted by RFFsNet-SEI,identification accuracy is improved.Finally,we compare RFFsNet-SEI with the counterparts in terms of identification accuracy,computational complexity,and prediction speed.Experimental results illustrate that the proposed method outperforms the counterparts on the basis of simulation dataset and real dataset collected in the anechoic chamber.

Method of Multi-Mode Sensor Data Fusion with an Adaptive Deep Coupling Convolutional Auto-Encoder

To address the difficulties in fusing multi-mode sensor data for complex industrial machinery, an adaptive deep coupling convolutional auto-encoder (ADCCAE) fusion method was proposed. First, the multi-mode features extracted synchronously by the CCAE were stacked and fed to the multi-channel convolution layers for fusion. Then, the fused data was passed to all connection layers for compression and fed to the Softmax module for classification. Finally, the coupling loss function coefficients and the network parameters were optimized through an adaptive approach using the gray wolf optimization (GWO) algorithm. Experimental comparisons showed that the proposed ADCCAE fusion model was superior to existing models for multi-mode data fusion.

人工智能赋能英语智慧教学的DEEP模式构建——基于四川开放大学学位英语课程教改的实践

国家开放大学在2022年开展了人工智能赋能英语教学改革项目。作为参与试点工作的分部之一,四川开放大学对学位英语课程开展智慧教学改革,采用学位英语自适应系统、小鱼易连直播系统、自主研发的在线练习题库以及QQ社交媒体为学生开展学术与非学术的支持服务,构建了DEEP教学模式。试点项目团队教师利用前测、后测、访谈、问卷调查等方式对教学效果进行对比和研究。调查结果显示,大部分学生认为直播辅导是学术支持的重要手段,基于社交媒体QQ创建的学习辅导群为学生提供了有效的情感支持和教学支持,学生对自适应系统总体满意。研究显示,基于知识图谱建设自适应学习课程将是开放教育深化教学改革和课程资源建设的有力保障。

Industrial Poverty Alleviation Model in Deep Poverty-stricken Villages in the Dry-hot Valley of Jinsha River: A Case Study of Poverty Alleviation in the Green Prickleyash Planting Industry in Laopingzi Village,Luquan County

Industrial poverty alleviation is the core of poverty alleviation in rural areas of China,and it is the fundamental way for the rural poor to achieve stable income and poverty alleviation. Laopingzi Village,Jiaopingdu Town,Luquan County,Kunming County,Yunnan Province,located in the dry-hot valley area of Jinsha River,has become a typical deep poverty-stricken village due to its special natural conditions.In recent years,in the battle to win the fight against poverty,the people of Laopingzi Village have achieved a virtuous cycle of the ecological environment and an access to get rid of poverty and get rich through vigorously developing green prickleyash planting industry. By the end of 2018,the incidence of poverty in Laopingzi Village Committee dropped from 45. 62% in 2014 to 1. 11%,and the green prickleyash planting industry had achieved remarkable results in poverty alleviation. This article summarizes the specific practices of developing the green prickleyash planting industry in the village,analyzes the main results and successful experiences of the mode and discusses the inspiration of the implementation of green prickleyash cultivation on industrial poverty alleviation,so as to provide an effective practical example for the development and poverty alleviation of poverty-stricken areas.

High common mode rejection ratio InP 90°optical hybrid in ultra-broadband at 60 nm with deep-rigded waveguide based on×4 MMI coupler

An InP optical 90°hybrid based on a×4 MMI coupler with a deep ridged waveguide is designed and fabricated.The working principle of the 90°hybrid is systematically introduced.Three-dimensional beam ropagation method(3D BPM)is used to optimize the structure parameters of the 90°hybrid.The designed compact structure is demonatrated to have a low excess loss less than-0.15 dB,a high common mode rejection ratio better than 40 dB,and a low relative phase deviation less than±2.5°.The designed hybrid is manufactured on a sandwitched structure deposited on an InP substrate.The measured results show that the common mode rejection ratios are larger than 20 dB in a range from 1520 nm to 1580 nm.The phase deviations are less than±5°in a range from 1545 nm to 1560 nm and less than±7°across the C band.The designed 90°optical hybrid is suitable well for realizing miniaturization,high-properties,and high bandwidth of coherent receiver.

Diagenetic Alteration Modes of the Deep Formation of Paleogene of Chezhen Depression,Bohai Bay Basin

The deep buried clastic formation of Paleogene is an important hydrocarbon reservoir in the Chezhen depression.There are two types of diagenetic alteration modes in it.The first mode is weak compaction, strong cementation,fracturing and weak dissolution in the sandstone and conglomerate on the steep slope of the depression.The reservoirs are cemented mainly by carbonate minerals strongly。

Tour Level Mode Choice Analysis of Madison Area in Wisconsin

This study is to investigate what factors and how they affect tours (trip chains) behavior. The key issue is the understanding and definition of tour and tour level mode. Also, these definitions should fit for the data. A semi-home based tour definition is stated, and a competing mode based tour mode is defined. Based on the definition, this study used Madison Area Data from National Household Survey to estimate a MNL structured model. It is found that travel distance could be a positive factor for car mode. Meanwhile, the number of trips is also a positive factor for choosing car.

Hyperparameter Tuning for Deep Neural Networks Based Optimization Algorithm

For training the present Neural Network(NN)models,the standard technique is to utilize decaying Learning Rates(LR).While the majority of these techniques commence with a large LR,they will decay multiple times over time.Decaying has been proved to enhance generalization as well as optimization.Other parameters,such as the network’s size,the number of hidden layers,drop-outs to avoid overfitting,batch size,and so on,are solely based on heuristics.This work has proposed Adaptive Teaching Learning Based(ATLB)Heuristic to identify the optimal hyperparameters for diverse networks.Here we consider three architec-tures Recurrent Neural Networks(RNN),Long Short Term Memory(LSTM),Bidirectional Long Short Term Memory(BiLSTM)of Deep Neural Networks for classification.The evaluation of the proposed ATLB is done through the various learning rate schedulers Cyclical Learning Rate(CLR),Hyperbolic Tangent Decay(HTD),and Toggle between Hyperbolic Tangent Decay and Triangular mode with Restarts(T-HTR)techniques.Experimental results have shown the performance improvement on the 20Newsgroup,Reuters Newswire and IMDB dataset.

A tour-based analysis of travel mode choice accounting for regional transit service

The aim of this work is to explore the impact of regional transit service on tour-based commuter travel behavior by using the Bayesian hierarchical multinomial logit model, accounting for the spatial heterogeneity of the people living in the same area.With two indicators, accessibility and connectivity measured at the zone level, the regional transit service is captured and then related to the travel mode choice behavior. The sample data are selected from Washington-Baltimore Household Travel Survey in 2007,including all the trips from home to workplace in morning hours in Baltimore city. Traditional multinomial logit model using Bayesian approach is also estimated. A comparison of the two different models shows that ignoring the spatial context can lead to a misspecification of the effects of the regional transit service on travel behavior. The results reveal that improving transit service at regional level can be effective in reducing auto use for commuters after controlling for socio-demographics and travel-related factors.This work provides insights for interpreting tour-based commuter travel behavior by using recently developed methodological approaches. The results of this work will be helpful for engineers, urban planners, and transit operators to decide the needs to improve regional transit service and spatial location efficiently.

Comparison of large deformation failure control method in a deep gob-side roadway: A theoretical analysis and field investigation

Under the dual influence of the mining disturbance of the previous working face and the advanced mining of the working face,the roadway is prone to large deformation,failure,and rockburst.Roadway stabilization has always significantly influenced deep mining safety.In this article we used the research background of the large deformation failure roadway of Fa-er Coal Mine in Guizhou Province of China to propose two control methods:bolt-cable-mesh+concrete blocks+directional energy-gathering blasting(BCM-CBDE method)and 1st Generation-Negative Poisson’s Ratio(1G NPR)cable+directional energy-gathering blasting+dynamic pressure stage support(πgirder+single hydraulic prop+retractable U steel)(NPR-DEDP method).Meantime,we compared the validity of the large deformation failure control method in a deep gob-side roadway based on theoretical analysis,numerical simulations,and field experiments.The results show that directional energy-gathering blasting can weaken the pressure acting on the concrete blocks.However,the vertical stress of the surrounding rock of the roadway is still concentrated in the entity coal side and the concrete blocks,showing a’bimodal’distribution.BCM-CBDE method cannot effectively control the stability of the roadway.NPR-DEDP method removed the concrete blocks.It shows using the 1G NPR cable with periodic slipping-sticking characteristics can adapt to repeated mining disturbances.The peak value of the vertical stress of the roadway is reduced and transferred to the deep part of the surrounding rock mass,which promotes the collapse of the gangue in the goaf and fills the goaf.The pressure of the roadway roof is reduced,and the gob-side roadway is fundamentally protected.Meantime,the dynamic pressure stage support method withπgirder+single hydraulic prop+retractable U steel as the core effectively protects the roadway from dynamic pressure impact when the main roof is periodically broken.After the on-site implementation of NPR-DEDP method,the deformation of the roadway is reduced by more than 45%,and the deformation rate is reduced by more than 50%.

深部煤层气游离气饱和度计算模型及其应用

近几年全国深部煤层气基于精细地质研究和水平井多段加砂压裂取得重大突破,部分井日产气量高达十万方,给煤层气产业重新树立了信心。但是,由于深部煤储层处于高地应力、高地温、高孔隙压力、低渗透率的复杂地质环境,不同深度煤储层典型参数和煤层气赋存方式的分布特征以及对储量和产量的影响亟需揭示。基于Langmuir等温吸附式、亨利定律及物质平衡原理,考虑吸附层和溶解气的影响,建立了深部煤层气游离气饱和度计算模型;以国内鄂尔多斯盆地大宁-吉县区块深部煤层气藏为例,分析不同深度深部煤层气赋存方式及分布特征,并评价游离气饱和度对深部煤层气储量、产量与合理配产的影响。研究认为:当煤层埋深大于溶解饱和对应的深度,游离气才会出现,且随着埋深的增加,游离气饱和度先快速增加后缓慢增加,目标区块埋深1875 m处才出现游离气,在埋深2800 m处游离气饱和度高达90%,游离气的占比高达17.3%。游离气饱和度对深部煤层气储量计算、产气特征和合理配产影响很大,随着游离气饱和度的增大,煤层气储量线性增大,累产气量持续上升但后期上升幅度逐渐变缓,深部煤层气井最优配产增加,井底流压下降速度加快,压裂改造区的内外压差降低,未改造区动用程度增加。目标区块主力开发煤层埋深位于2100~2300 m,游离气饱和度介于48%~68%,游离气占比介于10%~13%,建议气井合理配产介于(4~10)×10^(4)m^(3)/d。研究结果可为深部煤层气进一步开发提供理论依据和方法支撑。

考虑分段流变模式的深井井筒压力精确预测模型

超深井、特深井井筒温度和压力分布范围宽,钻井液流变性受超高温超高压影响显著,基于常规流变模式的井筒压力预测误差较大,文章通过开展温度为20~220℃、压力为0.1~200 MPa的水基钻井液和油基钻井液流变性测试实验,提出了不同温度和压力范围内的钻井液分段流变模式优选方法,建立了考虑多因素综合影响的钻井井筒压力精确预测模型。研究结果表明,随着温度和压力的变化,钻井液流变曲线的变化规律不一致,单一流变模式无法完全表征钻井液的流变特性;赫巴流变模式对100℃以下的水基钻井液和140℃以下的油基钻井液的流变性适用性更好,其他温度范围内罗斯流变模式的适用性更好;分段流变模式对井底压力的影响较为明显。将模型的计算结果与实测数据进行对比,发现井底压力预测误差在0.3 MPa以内,立管压力预测误差小于0.6 MPa;相对于油基钻井液,水基钻井液中的井筒压力预测误差更小。研究结果能够为超深井、特深井井筒压力精确预测奠定理论基础。

基于“产学研用”融合的人才培养模式研究与实践——以辽宁工程技术大学机械类专业为例

产教融合、协同育人是应用创新型人才培养战略的重要内容和手段。在实施过程中,紧密围绕高校自身资源和条件,结合研究应用型大学培养目标,深度挖掘基于“产学研用”融合的校企合作人才培养模式和特色,多元化创新其在人才培养中的内涵和外延,促进高校、企业及科研机构的深度融合,发挥企业、大学科技园及孵化器的功能,促进校企双方的共同成长;依托高校教育基金会,探索成立企业-教育基金会“人才培养基金”,通过规范性的项目管理,提升学生科学素养和精准就业能力。通过人才培养模式的研究与实践,使其在振兴东北老工业基地和服务辽西北人才战略中,更好发挥应有作用。

神府区块深部煤层气钻完井关键技术及应用

【目的】深部煤层具有高地应力、中高温度、特低渗透、强压缩性、强非均质性等特点,目前尚未形成成熟的开发技术体系,复杂的地质特征为钻井与完井工程带来了新的技术难题与挑战,亟需开展针对深部煤储层地质特征的钻完井理论与技术攻关,助力油气增储上产,保障国家能源战略安全。【方法】基于鄂尔多斯盆地东缘神府区块深部煤层气开发先导性试验,研发了一套高效钻完井关键技术。【结果和结论】(1)针对深部煤层井壁稳定性差、钻速低、钻井周期长,通过优化二开井身结构、优选钻井液体系与“一趟钻”技术,并结合井眼轨迹精细化控制,实现了深部煤层一体化高效钻进,助力“新优快”井台建设落地。(2)针对深部煤层地质特征复杂、采用常规压裂规模产量低,形成了以“定向射孔+前置酸液降破压+段内多簇密切割+高排量大规模+一体化变黏滑溜水+暂堵转向+多粒径组合支撑剂”为核心的复合极限规模化压裂技术体系。(3)基于“一区一策+全局寻优”的工作理念,设计立体井网工厂化钻完井作业模式,最优水平井距为350 m时,“拉链式”压裂模式效果最佳。(4)“深部煤层气+致密气”协同开采,获得了更高的工业气流,多气合采是提升鄂尔多斯盆地东缘非常规天然气开发效益的重要措施。研究结果有望为鄂尔多斯盆地深部煤层高效钻完井技术提供理论指导与实践经验。

产业化工程型软件工程专业人才培养体系探索与构建

以大数据为引领的电子信息产业迅猛发展,但是现有的人才培养模式并不能完全对接市场需求,人才供需不平衡。为更好地服务区域经济和社会发展,聚焦产业发展,培养适应行业和企业需要的高水平工程应用型人才,与企业深度合作,发挥校企双方各自优势,探索形成六位一体的校企深度融合的产业化、实践型软件工程专业人才培养体系。在此体系下,形成较好的人才培养效果,形成一种可借鉴、可复制的面向本科层次的校企合作人才培养模式。

基于数字孪生和深度学习的结构损伤识别

土木工程实际结构损伤状态的时间跨度通常只占总生命周期的一小部分。为解决传统基于数据驱动的结构损伤识别方法缺乏足够多的损伤训练数据的问题,提出结合数字孪生和深度学习的结构损伤识别方法,并应用于实际工程。该方法利用数值仿真模型和在线监测数据构建结构的数字孪生,以获得不同损伤工况下结构动力响应的“大数据”;为了摆脱对外激励信息的依赖,应用经验模态分解法和传递率函数对得到的数据进行预处理;将预处理后的固有模态传递率函数数据作为深度学习的输入进行训练,实现结构的损伤识别。为验证方法的有效性,对实际结构未经训练的监测数据进行分析,结果表明,该方法泛化能力良好,能够有效识别结构损伤状况。通过数字孪生技术解决了传统方法数据匮乏的问题,不需要任何地震信息,利用固有模态传递率函数数据训练的深度神经网络仍能保持较高的损伤识别准确率,二者结合可以使工程结构健康监测更为主动、可靠、高效。