音频大地电磁测深在某隧道工程中的应用

音频大地电磁(AMT)在隧道的工程地质勘察是一种非常有效的方法,它具有不受高阻屏蔽、轻便、探测深度相对较大等优点,在宏观上可以查明隧道地层岩性分界、地质构造及其赋水性。以某隧道音频大地电磁勘察为例,分别对AMT的采集手段、工频干扰去噪、静态校正、反演方法等关键技术问题进行分析,并结合实际资料对该工作区的地层分布、断层及其破碎带的赋水情况等进行分析和解释,在其应用中取得了理想的效果。     

Effect of ambient wind on pressure wave generated by high-speed train entering a tunnel

Journal of Central South University - Using three-dimensional, unsteady N-S equations and k-ε turbulence model, the effect of ambient wind on the pressure wave generated by a high-speed train...     

An automatic parameter calibration method for the SWAT model in runoff simulation

Runoff simulation is highly significant for hydrological monitoring, flood peak simulation, water resource management, and basin protection. Runoff simulation by distributed hydrological models, such as the soil and water assessment tool (SWAT) model which is the most widely used, is becoming a hotspot for hydrological forecasting research. However, parameter calibration is inefficient and inaccurate for the SWAT model. An automatic parameter calibration (APC) method of the SWAT model was developed by hybrid of the genetic algorithm (GA) and particle swarm optimization (PSO). Multi‐station and multi‐period runoff simulation and accuracy analysis were conducted in the basin of the Zhangjiang River on the basis of this hybrid algorithm. For example, in the Yaoxiaba Station, the calibration results produced an R² of 0.87 and Nash Sutcliffe efficiency (NSE) index of 0.85, while verification results revealed an R² of 0.83 and NSE of 0.83. Results of this study show that the proposed method can effectively improve the efficiency and simulation accuracy of the model parameters. It can be concluded that the feasibility and applicability of GA‐PSO as an APC method for the SWAT model were confirmed via case studies. The proposed method can provide theoretical guidance for many hydrological research fields, such as hydrological simulation, flood prevention, and forecasting.     

基于多层注意力机制—柔性AC算法的机器人路径规划

针对行动者—评论家（AC）算法存在的经验学习样本维度高、策略梯度模型鲁棒性低等问题,依据多代理系统的信息协作优势,构建注意力机制网络并作为代理体,引入多层并行注意力机制网络模型对AC算法进行改进,提出一种基于多层并行注意力机制的柔性AC算法。将其用于解决动态未知环境下的机器人路径规划问题,可增强行动者的策略梯度鲁棒性并降低评论家的回归误差,实现机器人路径规划最优方案的快速收敛。实验结果表明,该算法有效克服机器人路径规划的局部最优,具有计算速度快、稳定收敛的优点。     

全局优化的蝴蝶优化算法

针对基本蝴蝶优化算法中存在的易陷入局部最优值、收敛速度慢等问题,提出一种全局优化的蝴蝶算法,引入limit阈值来限定蝴蝶优化算法陷入局部最优解的次数,从而改变算法易陷入早熟的问题,结合单纯形策略优化迭代后期位置较差的蝴蝶使种群能够较快地找到全局最优解;将正弦余弦算法作为局部算子融入BOA中,改善迭代后期种群多样性下降的缺陷,加快算法跳出局部最优。在仿真模拟实验中与多个算法进行对比,结果表明改进算法的寻优性能更好。     

基于Filter Wrapper模式的特征选择算法*

特征选择是数据挖掘、机器学习和模式识别中始终面临的一个重要问题。针对类和特征分布不均时,传统信息增益在特征选择中存在的选择偏好问题,本文提出了一种基于信息增益率与随机森林的特征选择算法。该算法结合Filter和Wrapper模式的优点,首先从信息相关性和分类能力两个方面对特征进行综合度量,然后采用序列前向选择（Sequential Forward Selection, SFS）策略对特征进行选择,并以分类精度作为评价指标对特征子集进行度量,从而获取最优特征子集。实验结果表明,本文算法不仅能够达到特征空间降维的效果,而且能够有效提高分类算法的分类性能和查全率。     

基于加权网络和局部适应度的蛋白质复合物识别算法

蛋白质复合物识别对分析蛋白质网络的结构特征和模块功能具有重要意义。通常在蛋白质网络中挖掘稠密子图或模块来识别其中的蛋白质复合物,限制了其应用范围和识别的准确性。针对该问题,提出了一种基于加权网络和局部适应度的蛋白质复合物识别算法,该算法综合稠密子图的密度指标和模块性定义了新的局部适应度函数,并基于边聚集系数构建加权的蛋白质网络,根据权值选择边,在加权蛋白质网络中将种子边不断聚类扩展,从而获取具有最大综合适应度的子图作为蛋白质复合物。在酵母蛋白质等多个实际网络中试验表明,该算法能够有效提升蛋白质复合物识别的准确性。     

云存储中加密数据的自适应重复删除方法

为了保护个人隐私,用户倾向于在数据上传至云服务器之前将其加密. 相同的明文数据被加密成不同密文数据,使云服务器无法识别出重复的加密数据. 现存的解决方案多数倚赖可信第三方,且没有划分数据流行度,导致安全性与执行效率较低. 提出一种无需可信第三方的自适应重复删除方法.利用完美散列函数检查数据的流行度,使用口令认证密钥交换协议与同态加密安全传递数据的加密密钥,在保证用户数据隐私的前提下进行安全的重复数据删除. 与现有其它方案相比,安全性与实用性更强. 实验和仿真证明了方案的高效性.     

Super-Planckian thermal radiation enabled by hyperbolic surface phonon polaritons

Excitation of surface resonance modes and presence of resonance-free hyperbolic modes are two common ways to enhance the near-field radiative energy transport, which can find wide applications in noncontact thermal management and energy harvesting. Here, we identify another way to achieve the super-Planckian thermal radiation via hyperbolic surface phonon polaritons (HSPhPs). Based on the fluctuation-dissipation theory, the near-field radiative heat flux between bulk hexagonal boron nitride (hBN) planes with the optical axis perpendicular to the radiative energy flow can be 120 times as large as the blackbody limit for a gap distance of 20 nm. When the film thickness is reduced to 10 nm, the radiative heat flux is found to increase by 26.3%. The underlying mechanism is attributed to the coupling of Type I HSPhPs inside the anisotropic hBN film, which improves the energy transmission coefficient over a broad wavevector space especially for waves with extremely high wavevectors. This work helps to deepen the understanding of near-field radiation between natural hyperbolic materials, and opens a new route to enhance the near-field thermal radiation.     

Aerodynamic performance analysis of trains on slope topography under crosswinds

This work used the computational fluid dynamics method combined with full-scale train tests to analyze the train aerodynamic performance on special slope topography. Results show that with the increment in the slope gradient, the aerodynamic forces and moment increase sharply. Compared with the flat ground condition, the lateral force, lift force, and overturning moment of the train on the first line increase by 153.2%, 53.4% and 124.7%, respectively, under the slope gradient of 20°. However, with the increment of the windward side's depth, the windbreak effect is improved obviously. When the depth is equal to 10 m, compared with the 0 m, the lateral force, lift force and overturning moment of the train on the first line decrease by 70.9%, 77.0% and 70.6%,respectively. Through analyzing the influence of slope parameters on the aerodynamic performance of the train, the relationships among them are established. All these will provide a basic reference for enhancing train aerodynamic performances under different slope conditions and achieve reasonable train speeds for the operation safety in different wind environments.