A Study on Dual-Load-Zone Model of Overlying Strata and Evolution Law of Mining Stress

The changeable structure and movement law of overlying strata are the main contributor to the change of mining stress. Starting from the relevant theory of key stratum and particularly based on the theory of mine ground pressure and strata control, this research proposed a new solution to mining stress problems by establishing a dual-load-zone stratum structural model. Elastic foundation beam theory was used to solve the stress of overlying strata of the dual-load-zones with superposition method, which revised the traditional calculation method of mining stress. The abnormal increase of lead abutment pressure in the mining area was explained effectively, through which the evolution law of mining stress in the case of hard rock was obtained. The results indicate that mining stress experiences a drastic change within the range of 50 m ahead of the coal wall due to the collapse of main roof; under the influence of main key stratum and inferior key strata, the influence range of lead abutment pressure is extended up to approximately 120 m in the working face; this remarkable increase can be attributed to the excessive length of sagging zone. Results from both the dual-load-zone model experiment and field measurement demonstrate high consistency. The model can predict the influence range of abutment pressure effectively and thus guide the safety production of mining.     

In Situ Investigation and Numerical Simulation of the Failure Depth of an Inclined Coal Seam Floor: A Case Study

Mine Water and the Environment - The 1131 working face of the Yangcheng coal mine was used for a research demonstration to study the failure characteristics of an inclined coal seam floor. A...     

Chemoenzymatic Synthesis of Fragrance Compounds from Stearic Acid

Fatty acids are versatile precursors for fuels, fine chemicals, polymers, perfumes, etc. The properties and applications of fatty acid derivatives depend on chain length and on functional groups and their positions. To tailor fatty acids for desired properties, an engineered P450 monooxygenase has been employed here for enhanced selective hydroxylation of fatty acids. After oxidation of the hydroxy groups to the corresponding ketones, Baeyer–Villiger oxidation could be applied to introduce an oxygen atom into the hydrocarbon chains to form esters, which were finally hydrolyzed to afford either hydroxylated fatty acids or dicarboxylic fatty acids. Using this strategy, we have demonstrated that the high-value-added flavors exaltolide and silvanone supra can be synthesized from stearic acid through a hydroxylation/carbonylation/esterification/hydrolysis/lactonization reaction sequence with isolated yields of about 36 % (for ω−1 hydroxylated stearic acid; 100, 60, 80, 75 % yields for the individual reactions, respectively) or 24 % (for ω−2 hydroxylated stearic acid). Ultimately, we obtained 7.91 mg of exaltolide and 13.71 mg of silvanone supra from 284.48 mg stearic acid.     

计及Spark和属性权重的售电套餐推荐方法

针对电力市场用户群庞大，交易过程中售电套餐选择困难的问题，在Spark环境下设计了一种售电套餐推荐方法，同时也解决了售电套餐推荐过程中在大数据环境下的可扩展性及实时性问题。首先，计算出每个套餐属性的权重值，从而计算得到售电套餐综合相似度。然后，计及用户和套餐两方面提出一种售电套餐推荐方法，实现售电套餐的精准推荐。实验表明，提出的推荐方法能够明显提高推荐的准确度，并且在分布式环境下具有良好的推荐效率和可扩展性。     

针对多种处理痕迹的数字语音取证算法

现有的数字语音取证研究主要集中于对单一的某种操作进行检测，无法对不相关的操作进行判断。针对该问题，提出了一种能够同时检测经过变调、低通滤波、高通滤波和加噪这四种操作的数字语音取证方法。首先，计算语音的归一化梅尔频率倒谱系数（MFCC）统计矩特征；然后通过多个二分类器对特征进行训练，并组合投票得到多分类器；最后使用该多分类器对待测语音进行分类。在TIMIT以及UME语音库上的实验结果表明，归一化MFCC统计矩特征在库内实验中均达到了97%以上的检测率，且在对MP3压缩鲁棒性测试的实验中，检测率仍能保持在96%以上。     

《现代交换技术》教学综合改革与研究

以我校《现代交换技术》课程教学过程中存在的问题进行分析和总结，并结合目前行业发展的需求，对该课程的教学内容设计、教学方法的改进、实践教学内容和设备等各方面给出了相应的改革思路和解决方案。     

液相色谱法快速检测染色梅鱼中日落黄、柠檬黄的含量

目的建立一种快速、低成本检测染色梅鱼中日落黄、柠檬黄的方法。方法将染色梅鱼剥皮,鱼皮经70%甲醇氨水提取液提取,经EclipseXDB-C_(18)色谱柱分离,以甲醇和乙酸铵水溶液为流动相,进行梯度洗脱,流速为1.0 mL/min,柱温35℃,利用紫外检测器检测日落黄、柠檬黄的含量。结果日落黄、柠檬黄在1.25、2.5、12.5、25 mg/kg添加水平的加标回收率为80%～90%,相对标准偏差小于10%(n=6),柠檬黄检出限为0.05 mg/kg,日落黄检出限为0.025 mg/kg。结论该方法快速、准确、成本低,适合染色梅鱼中日落黄、柠檬黄的快速检测。     

Rollover crashworthiness analysis and optimization of bus frame for conceptual design

Journal of Mechanical Science and Technology - At the conceptual design stage, utilizing the detailed model to determine the rollover crashworthiness of bus structure would need long design cycle....     

基于深度学习的彩色以及近红外图像去马赛克

单传感器捕获的彩色-近红外（RGB-NIR）图像存在光谱干扰，从而导致重建出的标准彩色图像（RGB）图像与近红外（NIR）图像存在色彩失真以及细节信息模糊。针对这个问题提出一种基于深度学习的去马赛克方法，通过引入跳远连接与稠密连接解决了梯度消失和梯度弥散问题，使得网络更容易训练，并且提升了网络的拟合能力。首先，用浅层特征提取层提取了马赛克图像的像素相关性以及通道相关性等低级特征；然后，将得到的浅层特征图输入到连续多个的残差稠密块以提取专门针对去马赛克的高级语义特征；其次，为充分利用低级特征与高级特征，将多个残差稠密块提取到的特征进行组合；最后，通过全局跳远连接恢复最终的RGB-NIR图像。在深度学习框架Tensorflow上使用公共的图像与视觉表示组（IVRG）数据集、有植被的户外多光谱图像（OMSIV）数据集和森林（Forest）三个公开数据集进行实验。实验结果表明，所提方法优于基于多级自适应残差插值、基于卷积卷积和神经神经网络以及基于深度残差U型网络的主流的RGB-NIR图像去马赛克方法。     

二次侧非能动余热排出系统流动不稳定性研究

二次侧非能动余热排出（ASP）系统是压水堆核电厂应对全厂断电事故的重要措施之一。为研究ASP系统的运行特性，设计建造了ASP系统试验装置（ASPTF）。在ASPTF上开展了ASP系统运行稳定性影响试验，并对试验结果进行了理论分析。试验研究与理论分析结果表明：低压低功率下ASP系统中出现流动不稳定性；增加蒸汽管线或回水管线阻力系数可抑制ASP系统出现流动不稳定性；降低加热功率时ASP系统更易出现流动不稳定性；主泵运行状态影响ASP系统的输入功率进而影响系统的流动不稳定性。