Effect of heat treatment on microstructures and properties of a novel Al-Zn-Mg-Cu alloy for oil drilling

In view of the special requirements for strength, heat resistance and corrosion resistance of Al-Zn-Mg-Cu alloy for oil drilling, the Al-6.2 Zn-2.5 Mg-1.6 Cu alloy was prepared by increasing Cu content on basis of Russian Series 1953 alloy. The effect of heat treatment on the microstructures and properties of the alloy was characterized by optical microscope(OM), scanning electron microscope(SEM) and transmission electron microscope(TEM), and investigated by tensile test at room temperature, thermal exposure test and corrosion test. The results show that the strength after T6 aging treatment exhibit a decrease trend as an increase of the solution temperature from465 °C to 480 °C. After the solution treated by the rate of 470 °C/1 h, second phases dissolve into the matrix very well and the strength property reaches optimum. The alloy has better comprehensive properties treated by a solution treatment of 470 °C/1 h and then followed by an aging treatment of 120 °C/24 h + 170 °C/1 h + 120 °C/24 h. Under the aging state, the precipitated phases inside the grains are suitable in size, while on the grain boundary distribute discontinuously and the precipitate-free zone is obvious. Besides, the alloy still maintain high tensile properties. The yield strength, tensile strength and elongation are 650 MPa, 686 MPa,12.0%, respectively. The yield strength retention after heat exposure is 92%. The alloy has good corrosion resistance and the exfoliation corrosion degree. The average corrosion rate in the H_2S and CO_2 environment is 0.0024 mm/a, which is far less than the required 0.12 mm/a. It is insensitive to H_2S and CO_2 environments.     

基于改进多目标遗传算法的连铸二冷过程优化

采用一种改进的多目标遗传算法对二冷工艺进行优化.改进的多目标遗传算法应用概率法选取选择算子，根据适应度值来动态计算交叉和变异概率，能够得到更好的全局最优解，提高算法精度和整体性能.在基于凝固传热模型的二冷优化过程中，采用变间距差分法离散求解传热方程，对比粒子群算法、多目标遗传算法，改进的多目标遗传算法搜索效率高，得到的价值函数最小.在实际生产中，采用优化后的二冷工艺，使得总用水量减少约10%，提高了铸坯质量，达到了节能降耗的要求.     

基于新型空间注意力机制和迁移学习的垃圾图像分类算法

随着我国各级政府大力推动垃圾强制分类,分类回收各环节中实现标准化、自动化的垃圾分类识别需要适合云端部署的高准确率、低延时要求的细粒度图像分类模型.本文发挥深度迁移学习的优势建立了一套端到端的迁移学习网络架构GANet (garbage neural network);针对垃圾分类中类别易混淆、背景干扰等挑战,提出一种新型的像素级空间注意力机制PSATT (pixel-level spatial attention).为克服类别多和样本不平衡挑战,提出使用标签平滑正则化损失函数;为改善收敛速度以及模型稳定性与泛化性,提出了阶梯形OneCycle学习率控制方法,并给出了结合Rectified Adam (RAdam)优化方法和权重平滑处理技术的组合使用策略.实验使用了"华为云人工智能大赛.垃圾分类挑战杯"提供的按照深圳市垃圾分类标准标注的训练数据,验证了GANet在垃圾分类问题中的显著效果,获得了全国二等奖(第2名);同时,提出的PSATT机制优于对比方法,且在不同主干网络架构上均得到了提升,具有良好的通用性.本文提出的GANet架构、PSATT机制和训练策略不仅具有重要的工程参考价值,也具有较好的学术价值.     

河流COD污染通量变化规律研究

分析了两种类型河流COD污染通量变化特征。汇水面积较大的河流 ,下游断面汛期COD通量远大于点源COD入河排污量 ,典型河流断面可达其 9～ 2 2倍。这类河流COD污染通量与河水流量呈显著正相关 ,r值在 0 .90 3～ 0 .997范围 ,文中给出线性回归方程。河流点源污染量变异系数一般在 30 %～ 6 0 %,面源可达 12 7%～ 2 0 7%。     

轴对称辐射传递逆问题的三种求解算法比较

针对轴对称光学薄火焰,用图像探测器和滤色片获得火焰的单色辐射图像,根据火焰的单色辐射强度,分别采用最小二乘法、正则化方法和线性规划法对辐射传递逆问题进行了求解.对一个轴对称乙烯扩散火焰进行实验检测,通过对三种求解方法的结果比较,表明对辐射强度较高的火焰中部和上部区域,三种方法均可获得较好的重建结果;而对于辐射强度较低的火焰下部区域,由于线性规划法同时具有光滑性和约束性的优点,可以获得较好重建结果,它对求解轴对称辐射传递逆问题是一种比较理想的算法.     

CFD-ACE＋软件对PCR微芯片传热过程的数值模拟和分析

针对基于MEMS技术的PCR微芯片温度控制难的问题，通过对该芯片的传热过程进行简化并建立物理模型和数学模型，对芯片内部的热传导和温度变化过程进行了理论推导分析。利用CFD-ACE＋软件对芯片的热传导过程进行了数值模拟。模拟结果表明，该芯片具有较高的升、降温速度。芯片在自行构建的温度控制系统下进行了热循环反应，对GUS基因成功实现了扩增。     

Polarized Emission of Molecular Film With Lanthanide (Ⅲ) Complex

1 Results In the coordination system by using complexation with organic ligand, the ff emission of lanthanide(Ⅲ) (Ln(Ⅲ)) is induced the excitation energy transfer form the organic chromophore under the light-irradiation. However, there are not so much number of reports to discuss the energy relaxation mechanism in such complexes with Ln(Ⅲ). Recently, we succeeded firstly to estimate the rate constant of the energy transfer between the ligand and Ln(Ⅲ) in Pr(Ⅲ)-phenanthroline analogs[1]. Here, we will di...     

合金铸铁激光处理的探讨——Ⅱ.激光辐照工艺参数的控制

用热传导方程控制激光辐照温度,建立了一种新回归方程以控制激光辐照硬化深度,选用卡尔丹公式列热传导方程及同归方程联立求解,可得到同时满足一定的硬化深度及表面最高温度双重要求的工艺参数,此种方法可作为激光硬化处理工艺参数的控制或预报,对本合金铸铁的激光硬化处理,其表面温度的下限应大于其熔点温度,即以表面产生微熔的快速冷凝处理为宜,并以选取低功率慢扫描为最佳。     

对摩擦力作功与摩擦生热的量度的认识

本文通过对摩擦力作功与不擦生热的讨论，给出摩擦生热的量度方法．