Multi-task learning for video anomaly detection

We propose a multi-task learning framework for video anomaly detection based on a novel pipeline. Our model contains two crossing streams, one stream employs the backbone of Attention-R2U-net for future frame prediction, while the other is designed based on an encoder–decoder network to reconstruct the input optical flow maps. In addition, the latent layers of the two streams are merged together and assigned with a Deep SVDD-based loss at each location individually. Through the combination of these three tasks, the two-stream-crossing pipeline can be trained end-to-end to provide a comprehensive evaluation for the anomaly targets. Experimental results on several popular benchmark datasets show that our model outperforms the state-of-the-art competing models, which can be applied to different types of anomalous targets and meanwhile achieves remarkable precision.     

Grain size effect on cyclic deformation behavior and springback prediction of Ni-based superalloy foil

In order to clarify the influence of grain size on cyclic deformation response of superalloy sheets and springback behavior, cyclic loading–unloading and shearing tests were performed on the superalloy foils with 0.2 mm in thickness and diverse grain sizes. The results show that, the decline ratio of elastic modulus is weakened with increasing grain size, and the Bauschinger effect becomes evident with decreasing grain size. Meanwhile, U-bending test results determine that the springback is diminished with increasing grain size. The Chaboche, Anisotropic Nonlinear Kinematic (ANK) and Yoshida-Uemori (Y-U) models were utilized to fit the shear stress–strain curves of specimens. It is found that Y-U model is sufficient of predicting the springback. However, the prediction accuracy is degraded with increasing grain size.     

Efficiency,mechanism and microbial community of Cd(II) removal by mixed bacteria enriched from heavy metals mine soil

Mixed bacteria were enriched from heavy metals mine soil for cadmium (Cd(II))-containing wastewater treatment. Batch adsorption experiment results showed that the optimal pH, temperature, initial Cd(II) concentration, and biomass dosage were 6.0, 30 °C, 20 mg/L, and 1 g/L, respectively. Living biomass exhibited better Cd(II) removal efficiency (91.97%) than autoclaved biomass (79.54%) under optimal conditions. The isotherms and kinetics of living biomass conformed to the Langmuir isotherm model and pseudo-first-order kinetic model, respectively. FTIR results implied that amine groups, hydroxyl groups and phosphoric acid play an important role in the Cd(II) adsorption process, while XRD results showed that crystalline Cd(OH)₂ and CdO were obtained. After Cd(II)-containing wastewater treatment exposure, the dominant bacteria genera included Comamonas (39.94%), unclassified_f__Enterobacteriaceae (34.96%), Ochrobactrum (14.07%), Alcaligenes (4.84%), Bordetella (2.07%), Serratia (1.04%), and Bacillus (1.01%). Function prediction showed that the abundance of metabolic genes changed significantly. This study proposes the potential application of mixed bacteria for Cd(II)-containing wastewater treatment.     

Liquefaction behavior of fiber-reinforced calcareous sands in unidirectional and multidirectional simple shear tests

A study on the liquefaction resistance of calcareous sands reinforced with polypropylene fibers was reported. Stress-controlled cyclic simple shear tests were conducted on specimens prepared at a relative density of 50%, with and without fiber reinforcements. The liquefaction behavior was investigated by considering the effects of fiber contents ranging from 0% to 1%, fiber lengths varying from 3 mm to 12 mm and loading patterns. The results indicated that increasing fiber content and fiber length resulted in a decrease in the deformation, a reduction in pore pressure accumulation rate, and improved the liquefaction resistance of calcareous sands. Additionally, the risk of soil liquefaction could be significantly reduced when the fiber content was greater than 0.8%. The multidirectional loading had a considerable effect in reducing the liquefaction resistance compared to unidirectional loading. Further, the stiffness degradation of calcareous sands decreased with increasing fiber content and fiber length. The pore pressure generated in the cyclic tests was analyzed and was found to be affected by fiber content. A pore pressure prediction model was proposed to obtain the pore pressure characteristics of fiber-reinforced calcareous sands under various fiber content conditions.     

检验THERP合理性及不足的实验研究——以汽车冷媒加注工序为例

为了检验人的失误概率预测技术（technique of human error rate prediction，THERP）的合理性及不足，以汽车内饰冷媒加注工序为例，采用E-prime实验的方法分析了冷媒加注工序的人因失误概率。通过比较发现，汽车冷媒加注工序的4个任务的人因失误概率的实验结果与THERP的理论计算结果基本一致，说明了THERP方法的合理性。并说明了THERP的不足在于计算人因失误时忽略了操作行为中的认知过程。     

基于NB分类器重访概率预测的Web缓存替换策略

Web缓存是用来解决网络访问延迟和网络拥塞问题，缓存替换策略直接影响缓存的命中率。为此，提出一种朴素贝叶斯（NB）分类器重访概率预测的Web缓存替换策略；根据用户之前访问日志，通过分区操作提取多项特征来表示每次访问的对象，并构建特征数据集；训练NB分类器，用来确定缓存中对象被再次访问的概率，为对象分配权重；结合LRU策略来合理删除一些对象。仿真结果表明，提出的策略在保证较高命中率的同时有效降低了执行时间。     

基于物质平衡方程的页岩气井产能预测方法

页岩气井产能受地质和压裂改造参数综合影响,在开发初期,难以确定可靠的地质模型来预测气井产能。为了快速、准确地预测气井产能,在研究页岩气物质平衡方程和产能方程的基础上,考虑吸附气和异常高压气藏岩石弹性能量影响,建立了基于物质平衡方程的页岩气井产能预测方法,该方法需要参数较少,可在开发初期快速对页岩气井产能进行预测。涪陵地区X1井实例计算表明:该方法综合考虑了页岩气吸附气解吸、异常高压气藏岩石弹性能量的影响,产能预测结果准确、可靠;吸附气对气井累产气量影响明显,约占总累产气量17%,对定产生产气井的稳产期影响不大;吸附气的贡献在2.5~4年逐渐体现出来,定压放喷生产初期产量高,地层压力下降快,吸附气的产出贡献比控制产量的定产生产方式要早;考虑岩石弹性能量的影响,气井稳产时间和累产气有所增加,但影响不明显。