Abstract The dredge pump is the core equipment in a dredger, which is used to transport dense dredged soil excavated by the dredger. However, the slurry performance of a dredge pump is usually calculated via empirical formula based on the clean-water performance data obtained by numerical simulation or model test, and the conversion accuracy is difficult to evaluate. This paper focuses on reviewing the present technologies of slurry performance test as well as numerical simulation for centrifugal solid-liquid pump. Research progress on special instruments required for testing slurry parameters as well as numerical models for simulating dense solid-liquid flows was summarized. An example was presented to illustrate the scale effect and possible error source in performance conversion between model and prototype pump and then technical difficulties and future research directions in hydraulic performance prediction for dredge pump were put forward.

Abstract In response to the problems of poor stability and low efficiency in the transportation of temporary sludge in Shanghai Laogang, based on indoor experimental data, numerical simulation was used to predict the pressure drop characteristics of pipelines with different solid content, pipe diameters, and flow velocities, the numerical results show that at the same flow rate and pipe diameter, the pressure drop increases with the increase of solid content; The pressure drop gradually increases with the increase of flow rate at the same pipe diameter and solid content, and when the the flow state inside the pipe changes, there is a turning point in the flow rate pressure drop curve; At the same flow rate and solid content, the pressure drop decreases with the increase of pipe diameter. Combined with test for pipe pressure drop, the numerical method was verified, which provides guidance for the optimization and improvement of sludge transportation plans and the selection of transportation pumps.

Abstract Trailing suction hopper dredgers are usually equipped with high-pressure-jet pumps, which provide high pressure water to improve dredging efficiency. In a newly built trailing suction hopper dredger, the high-pressure-jet pump encountered serious vibration problem. In this study, unsteady Reynolds-averaged Navier-Stokes equations were solved with Shear Stress Transport k-ω turbulence model to carry out pump-pipeline matching analysis, and the causes of flow-induced vibration were discussed. The impeller was correspondingly optimized by increasing the blade number and moderately decreasing the blade inlet setting angle and wrap angle. The optimization results show that the high-pressure-jet pump works in the high-efficiency zone both for the high-speed dredging process and low-speed hopper discharging process, and the time-average value and amplitude of the hydraulic radial force drops by more than 30% under the required working condition, which effectively mitigates the flow-induced vibration.

基于欧拉双流体模型对绞吸挖泥船大型泥泵进行固液两相流的非定常数值模拟计算,通过对比分析颗粒粒径、颗粒浓度、颗粒密度对两相流线、固相颗粒浓度分布等流场特性的影响,为提高疏浚泥泵的优化设计提供指导.模拟结果表明,颗粒物理特性对液相流线分布的影响较小,随着颗粒粒径增大,液相对固相的带动作用减弱;随着颗粒密度的增大,液相对固相的带动作用未减弱;随着颗粒浓度增大,液相对固相的带动作用增强;随着粒径和密度增大,颗粒分布不均性加强;随着颗粒浓度的增大,颗粒浓度分布均匀性加强,靠近叶片壁面浓度明显增大.对比了两种泥泵流道的颗粒浓度分布,通过叶轮流道的优化设计,可以减少颗粒在流道表面的集聚,降低泥泵磨损.

绞吸挖泥船泥泵泵轴的断裂事故时有发生,直接影响疏浚工程的进度和效益.以厦门机场大小嶝造地工程为背景,3500 m3/h绞吸挖泥船的泥泵轴系为研究对象,建立了泥泵在流量11000 m3/h、泥砂粒径0.7 mm、混合物体积浓度20％工况下的疲劳寿命预测模型.通过对泥泵运转过程中的多相流场数值分析和对泥泵轴系瞬态动应力响应分析,提取泵轴关键节点处的应力时间历程数据,结合S-N曲线法和线性累积损伤原理,分析泵轴时域疲劳特性,预测相应工况下的泥泵轴系疲劳寿命.以此为基础,为进一步优化施工工艺和提升泵轴运转寿命提供建议.

Deep-learning based methods that aim to extract effective high-level features have steadily improved the performance on the speech emotion recognition. However, low-level features that contain important emotion-related information have not gained much attention. In this paper, we propose a novel low-level feature extraction method based on the Time-Frequency Attention (TFA) module and Time-Frequency Weighting (TFW) module. First, the TFA module is designed to learn notable regions in the detail-rich low-level feature maps produced by the scale-specific convolutional layers. Then, the TFW module is proposed to extract discriminative features from the time and frequency dimensions respectively. Finally, the speech emotion recognition task is completed by the subsequent multi-branch network. Experimental results on the IEMOCAP and RAVDESS datasets demonstrate the importance of low-level features, and show that the proposed method outperforms other state-of-the-art approaches.

消能箱作为泥舱系统的关键设备,直接参与疏浚装舱过程,其出流特性对船舶装舱效果及生产效率有重要影响.以1.50 万m3 耙吸挖泥船的消能箱为研究对象,采用流场数值模拟手段,对消能箱主体尺寸、结构形式等影响出流特性的关键参数进行计算分析.结果表明,在一定范围内,消能箱管径越大,消能效果越好,且出流均匀性得到一定程度改善;等截面积条件下,圆形截面消能箱的装舱效果显著优于方形截面的设计;消能箱中部开口指向舱底的速度分量较大,可通过布置水平挡板进行局部优化.该消能箱设计有助于改善疏浚装舱过程中细粉砂的沉积效果,节省装舱时间,减少溢流损失.

针对挖泥船泥泵轴疲劳断裂问题,研究基于流场非定常数值模拟、轴系瞬态有限元计算、应力-疲劳寿命修正模型下的泥泵轴疲劳寿命预测方法,并据此计算分析泥沙粒径、输送浓度以及叶轮局部磨损对泥泵轴疲劳寿命的影响.计算结果表明,随泥沙粒径与输送浓度的增大,泵轴疲劳寿命均呈对数规律降低,可根据计算结果合理调整泥泵轴系检修周期;叶轮局部磨损会造成泵轴疲劳寿命的显著下降,施工中应注意磨损叶轮的及时修复.

随着人工智能与深度学习的发展,基于深度学习的多通道脑电信号的情绪识别研究逐渐受到关注,但多通道脑电情绪识别信号复杂且各通道重要性一致,并不能高效且有针对性地进行脑电情绪识别.为此,该文提出一种基于缩放卷积层和脑电通道增强模块的情绪识别方法,能直接在脑电物理通道上进行增强学习.首先,通过缩放卷积层提取多通道脑电情绪信号的类时频特征;然后,通过脑电通道增强模块对所有脑电物理通道重新赋予不同的重要性;最后,利用卷积神经网络对情绪进行分类.该方法能够融合多通道脑电信号的时间和频率信息,同时,通过输出各脑电通道的重要性,探究不同情绪维度与脑电通道之间的关系.在DEAP数据集上进行了实验验证,不同脑电通道对情绪识别任务的重要性存在差异,其中,额叶区和枕叶区的C4、P4、P3、P04、F7 5个脑电通道重要性相对较高,该情绪识别方法在愉悦度、唤醒度和支配度3个情绪维度上的识别准确率也均有提升.

In intelligent human-computer interaction systems, speech emotion recognition (SER) is a fundamental task for understanding user intention. One vital challenge for emotion inferring is how to extract discriminative and robust features. In this paper, we propose a novel network based on the Time-Frequency Weighting (TFW) module and the ConvlD enabled Multi-head Element-wise Self-attention (ID-MESA) block to extract discriminative features from three dimensions of time, frequency and channel for improving the performance in SER. The TFW module is designed to capture emotion information along the time and frequency dimensions in the shallow neural network. As the high complexity of the emotion feature, the 1D-MESA block can assist the network to locate the discriminative emotion features in the channel dimension. The proposed architecture outperforms the state-of-the-art methods in the IEMOCAP database, with the absolute increase of 3.98% and 1.58% on unweighted accuracy among four emotion classes and weighted accuracy, respectively.

In order to transiently solve the transient movement of sediment particles in dredge pumps, a modified algorithm is realized based on the discrete phase model in ANSYS Fluent. The granular phase volume fraction solved by using the Eulerian (granular)-Eulerian (liquid) two phases flow method and the Huilin-Gidaspow drag force laws are introduced for tracking particles with the Lagrangian method in dense flow. The solution process that solves particle motion after updating the impeller grid is changed to a process that solves particle relative motion after rotating synchronously the impeller grid with the particles. Under the situations of moving wall, the modified algorithm avoids the calculation error on collision identification and rebounded velocity of the particles. A comparison of numerical results shows that the modified algorithm can significantly improve the accuracy of particle motions in the dredge pump with similar time cost. The erosive wear predicted by the modified algorithm mainly appears on the front edges of the blades and the peak of erosion rate is about 7×10-5 kg/(m2∙s), which is similar to the actual situations, supporting the effectiveness of the modified algorithm.

疏浚工程中,输送管道内壁面受到泥砂浆的持续冲刷,导致管道冲蚀磨损严重.为选择输合理的输送管道材质,以提高疏浚管道的抗冲蚀性能,降低其维修和更换频率,采用冲蚀试验与理论分析的方法,以常见管材Q235为参照对象,对比5种可用于制作耐磨排泥管道的耐磨金属材料的冲蚀性能,包括Cr15铸铁、Cr26铸铁、Fedur? 40合金、中锰钢、信铬钢.根据材料表面扫描电镜(SEM)图像,分析不同冲蚀角度下材料磨损类型.结果表明:冲蚀磨损过程中,各耐磨金属材料同时承受多种磨损作用,合金材料中起支撑作用的软质组分容易因切削、塑性疲劳断裂等因素而被剥离,而较硬的碳化物等组分则在松动后容易被颗粒撞击脱落;除Q235外,其余材料的磨损率均随着冲蚀角度的增加而增大;信铬钢、Fedur? 40合金在中、小冲蚀角度下的耐磨性能表现优秀,若价格与加工性能合适,建议选作疏浚管道金属材料.

In order to study the influence of the rotational speed of the cutter and traverse speed of the dredger on the clay excavation in a clay environment, considering the rheological characteristics of clayslurry,the Herschel-Bulkley non-Newtonian fluid model is introduced and the hydrodynamic analysis software ANSYS CFX is used to numerically predict the flow field and the cutter performancein the clay excavation process, and to obtain the change of cutter clay volume fraction during excavation of clay. The results show that the Herschel-Bulkley non-Newtonian fluid model is more effective in representing the characteristics of clay that is not easily mixed into homogeneous fluid, agglomerated and deposited. The cutter performance under different construction processes is analyzed, and it is found that the cutter has the highest average concentration at the suction port and a volume fraction up to 31.8% at a traverse speed of0.25 m/s and a rotational speed of 30m/s. However, considering the energy consumption factor, the cutter with the construction parameters of a traverse speed of 0.25 m/s and a rotational speed of 25 m/s has the lowest specific energy consumption and is more economical.

In order to ensure the gap between the sliding piece of trailing suction hopper dredger and the hull suction to meet the operation requirements, taking a 4500 m 3 trailing suction hopper dredger and a 12888 m 3 trailing suction hopper dredger as examples, the solid-liquid two-phase flow method was used to simulate and analyze the influence of different gap between sliding piece and the hull suction on the dredging performance. On this basis, the sliding piece installation procedure was improved and successfully applied to a 6500 m 3 trailing suction hopper dredger. The results indicate that the gap has a great influence on the production. The larger the gap, the lower the production and the stronger the non-uniformity of flow field distribution; When the gap is less than 4 mm, the production reduction rate is slow, and the dredging production decreases by about 2% when the gap is 2 mm,. When the gap is more than 4 mm, the dredging production decreases linearly with the increase of the gap. The application effect of the improved sliding piece installation technology is better, and the maximum gap of the sliding piece of the 6500 m 3 trailing suction hopper dredger is only 0.45 mm, which meets the needs of dredging engineering.

使用脑电进行情绪识别已经有了广泛的研究,但由于脑电的低信噪比、不平稳性以及受试者情绪表达方式的不同,不同受试者甚至单个受试者的脑电图情绪特征都会存在差异性,导致脑电样本在特征空间分布不均匀,容易出现模型泛化性能差的问题.为解决这一问题,该文提出了 一种结合提升算法(boost)和梯度下降法(gradient descent)的双策略训练方法交替更新脑电情绪识别模型,梯度下降法在模型推理过程中更新网络参数,使损失最小化,提升算法用于更新脑电样本权重.在DEAP数据集上的实验结果表明,该方法在效价、唤醒和优势度3个维度上的准确率分别为71.25％、71.48％和71.80％,且在跨被试数据集下通过数据驱动的方式有效调整了脑电样本特征的分布,使其分布更均匀,从而提高了情绪识别模型的泛化性能.

Research on emotion recognition based on EEG (electroencephalogram) signals has gradually become a hot spot in the field of artificial intelligence applications. The recognition methods mainly include designing traditional hand-extracted features in machine learning and fully automatic extraction of EEG features in deep learning. However, onefold features cannot represent emotional information perfectly which is contained in EEG signals. Traditional hand-extracted features may lose a lot of hidden information contained in raw signals, and automatically extracted features also do not contain prior knowledge. In this context, a multi-input Y-shape EEG-based emotion recognition neural network is proposed in this paper, which fusing spacial-frequency domain features and data-driven spectrogram-like features. It can effectually extract information in three domains, time, space, and frequency from raw EEG signals. Moreover, this paper also proposes a novel EEG feature mapping method. The experimental results show that the accuracy of EEG emotion recognition has achieved the state-of-the-art result based on the established DEAP benchmark dataset. The average emotion recognition rates are 71.25%, 71.33% and 71.1% in valance, arousal and dominance respectively.

以上航局3000 m3等级耙吸挖泥船建造项目为依托,结合项目中4500 m3耙吸挖泥船耙头研制,深入研究耙臂冲水管系沿程阻力及喷嘴布置对耙头冲水的影响.采用数值模拟方法对4500 m3耙吸挖泥船耙头高压冲水管路及喷嘴流速进行了计算,依据高压冲水泵性能曲线开展工况点的匹配性分析.针对喷嘴附近局部过流面的流场分布,进一步优化耙头内部高压冲水过流区域形状和喷嘴结构形式,有效降低管路阻力,减小了高压冲水沿耙臂管路输送至喷嘴过程的能量损失,显著提高了喷嘴冲水流速,形成了与高压冲水泵工况点相匹配的耙头冲水管系和喷嘴设计方案.

Convolutional Neural Networks (CNNs) have achieved remarkable performance breakthroughs in a variety of tasks. Recently, CNN-based methods that are fed with hand-extracted EEG features have steadily improved their performance on the emotion recognition task. In this paper, we propose a novel convolutional layer, called the Scaling Layer, which can adaptively extract effective data-driven spectrogram-like features from raw EEG signals. Furthermore, it exploits convolutional kernels scaled from one data-driven pattern to exposed a frequency-like dimension to address the shortcomings of prior methods requiring hand-extracted features or their approximations. ScalingNet, the proposed neural network architecture based on the Scaling Layer, has achieved state-of-the-art results across the established DEAP and AMIGOS benchmark datasets.

In recent years, Deep Neural Networks (DNNs) have achieved excellent performance on many tasks, but it is very difficult to train good models from imbalanced datasets. Creating balanced batches either by majority data down-sampling or by minority data up-sampling can solve the problem in certain cases. However, it may lead to learning process instability and overfitting. In this paper, we propose the Batch Balance Wrapper (BBW), a novel framework which can adapt a general DNN to be well trained from extremely imbalanced datasets with few minority samples. In BBW, two extra network layers are added to the start of a DNN. The layers prevent overfitting of minority samples and improve the expressiveness of the sample distribution of minority samples. Furthermore, Batch Balance (BB), a class-based sampling algorithm, is proposed to make sure the samples in each batch are always balanced during the learning process. We test BBW on three well-known extremely imbalanced datasets with few minority samples. The maximum imbalance ratio reaches 1167:1 with only 16 positive samples. Compared with existing approaches, BBW achieves better classification performance. In addition, BBW-wrapped DNNs are 16.39 times faster, relative to unwrapped DNNs. Moreover, BBW does not require data preprocessing or additional hyper-parameter tuning, operations that may require additional processing time. The experiments prove that BBW can be applied to common applications of extremely imbalanced data with few minority samples, such as the classification of EEG signals, medical images and so on.

随着疏浚施工水平的不断提高,单列泥门逐渐被广泛应用于中小型耙吸挖泥船上.基于上航局3000 m3等级耙吸挖泥船建造项目,结合船舶在波浪中的受力情况,研究了适用于单列泥门布置形式的泥门启闭装置结构,采用了自立式支撑机构,创新性地设计了易于拆装的导向装置.此外,针对3000 m3等级项目中的泥门受力分析,采用数值模拟的方法根据项目中4500 m3耙吸挖泥船满载工况,校核了泥门装置主要零部件的强度.结合泥门受力和液压油缸工作参数,总结了一套较为完备的油缸选型方法,为今后泥门启闭装置的优化设计和油缸选型提供借鉴.