山区高速铁路深路堑与隧道方案的比较研究

在山区高速铁路建设中,不可避免地会出现深路堑,如果方案选择不合理,则会造成当地环境的破坏,导致地质灾害频发,直接威胁到高铁运营的安全.因此,正确、合理地选用深路堑和隧道方案,显得非常重要.文章主要从地质条件、环境影响、施工难度、综合经济成本、运营安全等五个方面,比较研究了深路堑与隧道选用的适用性,并提出了方案选择的建议.     

Isotopic composition of carbon and nitrogen of suprabenthic fauna in the NW Balearic Islands (western Mediterranean)

Stable isotope (δ¹³C and δ¹⁵N) analyses were performed on suprabenthic fauna collected in the western Mediterranean (NW Balearic Islands), at depths ranging between 350 and 780 m. Samples were collected seasonally at bi-monthly intervals during six cruises performed between August 2003 and June 2004, using a Macer-GIROQ suprabenthic sledge (0.5 mm mesh size). Twenty-four separate species (5 mysids, 12 amphipods, 2 cumaceans, 2 isopods, 1 euphausiid, 1 decapod and 1 fish) and bulk copepods were analyzed on a seasonal basis for stable carbon and nitrogen isotopes. Stable nitrogen isotope ratios (δ¹⁵N) ranged from 2.3‰ (the amphipod Lepechinella manco in September 2003) to 13.0‰ (the amphipod Rhachotropis caeca in August 2003). δ¹³C values ranged from − 24.2 (the cumacean Campylaspis sulcata in June 2004) to − 16.1 (the amphipod Bruzelia typica in November 2006). Both δ¹³C and δ¹⁵N values suggest that there are three trophic levels within the suprabenthic community. However, considering the bathymetric range of the species, the results suggest that the deepest assemblage supported only two trophic levels. The stable isotope ratios of suprabenthic fauna displayed a continuum of values and confirmed a wide spectrum of feeding types (from filter-feeders to predators). In general, and in spite of the poor knowledge about diets available for most suprabenthic species, higher δ¹⁵N were found for carnivorous amphipods (e.g. Rhachotropis spp., Nicippe tumida) consuming copepods. Low overlap for δ¹³C and δ¹⁵N values was observed, though δ¹⁵N values where less variable than δ¹³C, which suggests high resource partitioning in this assemblage. Seasonal variations in isotopic composition for both δ¹³C and δ¹⁵N were low (less than 1‰ and 3‰, respectively) and variable depending on species. Low correlations between δ¹³C and δ¹⁵N of suprabenthic fauna were found for all periods studied, though increasing from February 2004 to June 2004 (after the main peak of primary production in surface). C:N ratio (indicator of lipid content) showed higher values in summer than in winter. This suggests that lipid content may explain the seasonal patterns of δ¹³C variability and, due to the increase of storage products in phytoplankton and zooplankton, it possibly indicates the peak of primary production at the surface.     

深埋大断面隧道土石复合地层塌方机制及处置措施评价

针对黄陵—韩城—侯马铁路如意隧道施工通过土石复合地层时隧道围岩塌方事故,分析塌方的影响因素及机制,提出"管棚+小导管注浆"加固围岩以及增大初期支护和二衬的施工参数等处置措施;通过现场监测和三维数值模拟方法评价处置措施的效果。结果表明:复杂地层、围岩软弱、地下水是引发塌方的主要因素;由于土石复合地层和地下水的影响,使隧道拱部产生了较大的附加荷载,导致拱顶岩体鼓起破坏,进一步发展为掌子面拱部大范围的失稳破坏;对塌方地段隧道处置后,实测的拱顶沉降和水平收敛位移分别为26和30mm,拱顶初期支护与围岩之间的压力为128kPa,数值模拟的拱顶沉降从662mm下降到47mm,洞周的塑性区减小到原设计方案的63%,表明采用该处置措施有效地控制了围岩的压力和变形,降低了施工风险;在隧道施工中应注重掌子面的超前加固。     

浅谈深基坑开挖的控制

对深基抗开挖的施工技术及质量挖制要点进行了详细阐述。     

A simplified life estimation method for the spherical hull of deep manned submersibles

Creep and fatigue are involved in the loading of deep manned submersible, which is a rather complex variable amplitude pattern. Dwell effects resulting in lower life than pure fatigue are observed in the prior experimental research while no proper prediction methods are available to explain the phenomenon. Recently, the authors proposed a modified crack growth rate model to explain the creep effect under the cyclic loading which is validated for the crack growth of some titanium alloys under cyclic creep loading, however, its application is restricted to the condition where lots of parameters have to be determined based on many experiments and it is not very convenient in the material selection and primary design stage of a component. In this paper, a simplified prediction model for the load pattern of constant amplitude cyclic loading is proposed aiming at one of the most applicable materials for the pressure hull of submersibles, Ti–6Al–4V ELI. The method can be easily used to estimate the life of the pressure hull based on two series of basic mechanical properties of the material and validated by the modified crack growth rate model with parameters determined by large amounts of experimental data.     

The dynamics of deep water subsea lifting operations in super-harmonic resonance via the harmonic balance method

The dynamics of deep water subsea lifting operations experiencing super-harmonic resonance is analysed in this study. The harmonic balance method is used to solve the non-dimensional equation of motion of the system and the results are compared with time domain integration and with an equivalent energy dissipation model for typical subsea lifting scenarios. It is demonstrated that 1:3 and 1:5 super-harmonic resonances represent significant features of the response of the system and can lead to large dynamic forces in the cable, which may violate the structural limits of the system in real operations. The harmonic balance method presents results almost as accurate as the time domain integration but up to 25 to 35 times faster, while the equivalent energy dissipation model is not able to represent the super-harmonic resonances. Consequently, taking into account the dynamics introduced by super-harmonic resonances is necessary in the analysis of subsea lifting operations, as it can be the limiting design criterion in certain scenarios, and the harmonic balance method can be used as a fast and accurate method to solve this problem.     

The station-free sharing bike demand forecasting with a deep learning approach and large-scale datasets

The station-free sharing bike is a new sharing traffic mode that has been deployed in a large scale in China in the early 2017. Without docking stations, this system allows the sharing bike to be parked in any proper places. This study aimed to develop a dynamic demand forecasting model for station-free bike sharing using the deep learning approach. The spatial and temporal analyses were first conducted to investigate the mobility pattern of the station-free bike sharing. The result indicates the imbalanced spatial and temporal demand of bike sharing trips. The long short-term memory neural networks (LSTM NNs) were then developed to predict the bike sharing trip production and attraction at TAZ for different time intervals, including the 10-min, 15-min, 20-min and 30-min intervals. The validation results suggested that the developed LSTM NNs have reasonable good prediction accuracy in trip productions and attractions for different time intervals. The statistical models and recently developed machine learning methods were also developed to benchmark the LSTM NN. The comparison results suggested that the LSTM NNs provide better prediction accuracy than both conventional statistical models and advanced machine learning methods for different time intervals. The developed LSTM NNs can be used to predict the gap between the inflow and outflow of the sharing bike trips at a TAZ, which provide useful information for rebalancing the sharing bike in the system.     

Predicting station-level hourly demand in a large-scale bike-sharing network: A graph convolutional neural network approach

This study proposes a novel Graph Convolutional Neural Network with Data-driven Graph Filter (GCNN-DDGF) model that can learn hidden heterogeneous pairwise correlations between stations to predict station-level hourly demand in a large-scale bike-sharing network. Two architectures of the GCNN-DDGF model are explored; GCNN_reg-DDGF is a regular GCNN-DDGF model which contains the convolution and feedforward blocks, and GCNN_rec-DDGF additionally contains a recurrent block from the Long Short-term Memory neural network architecture to capture temporal dependencies in the bike-sharing demand series. Furthermore, four types of GCNN models are proposed whose adjacency matrices are based on various bike-sharing system data, including Spatial Distance matrix (SD), Demand matrix (DE), Average Trip Duration matrix (ATD), and Demand Correlation matrix (DC). These six types of GCNN models and seven other benchmark models are built and compared on a Citi Bike dataset from New York City which includes 272 stations and over 28 million transactions from 2013 to 2016. Results show that the GCNN_rec-DDGF performs the best in terms of the Root Mean Square Error, the Mean Absolute Error and the coefficient of determination (R²), followed by the GCNN_reg-DDGF. They outperform the other models. Through a more detailed graph network analysis based on the learned DDGF, insights are obtained on the “black box” of the GCNN-DDGF model. It is found to capture some information similar to details embedded in the SD, DE and DC matrices. More importantly, it also uncovers hidden heterogeneous pairwise correlations between stations that are not revealed by any of those matrices.     

A novel passenger flow prediction model using deep learning methods

Currently, deep learning has been successfully applied in many fields and achieved amazing results. Meanwhile, big data has revolutionized the transportation industry over the past several years. These two hot topics have inspired us to reconsider the traditional issue of passenger flow prediction. As a special structure of deep neural network (DNN), an autoencoder can deeply and abstractly extract the nonlinear features embedded in the input without any labels. By exploiting its remarkable capabilities, a novel hourly passenger flow prediction model using deep learning methods is proposed in this paper. Temporal features including the day of a week, the hour of a day, and holidays, the scenario features including inbound and outbound, and tickets and cards, and the passenger flow features including the previous average passenger flow and real-time passenger flow, are defined as the input features. These features are combined and trained as different stacked autoencoders (SAE) in the first stage. Then, the pre-trained SAE are further used to initialize the supervised DNN with the real-time passenger flow as the label data in the second stage. The hybrid model (SAE-DNN) is applied and evaluated with a case study of passenger flow prediction for four bus rapid transit (BRT) stations of Xiamen in the third stage. The experimental results show that the proposed method has the capability to provide a more accurate and universal passenger flow prediction model for different BRT stations with different passenger flow profiles.     

基于深度强化学习的分布式电驱动车辆扭矩分配策略

为在保证分布式电驱动车辆制动稳定性的前提下实现经济性的提升,提出了基于深度强化学习的分布式驱动前、后轴扭矩分配策略.在建立分布式电驱动车辆关键部件物理模型的基础上,基于车辆模型及制动稳定性约束,建立了基于深度强化学习的扭矩最优分配控制模型,并对传统固定比值的扭矩分配策略和所提出的策略进行了对比,结果表明:在新欧洲驾驶循...