摘要：This study used the pseudo-spinodal mechanism to obtain the ultrafineαphase for designing highstrength titanium *** multiple experiments were designed to find the composition range of TM-x Mo-y V alloy(TM:Ti-4.5 Al-2 Cr-2.5 Nb-2 Zr-1 Sn)for obtaining ultrafineα*** results confirm that when the alloy composition is located near the intersection of theαandβphase free energy curves,the alloy will undergo pseudo-spinodal transformation and obtain the ultrafineα*** designed TM-6 Mo-3 V alloy has a yield strength of 1411 MPa and an elongation of 6.5%.The strength of the alloy depends on the thickness,fraction of theαphase and the solid solution strengthening effect of the alloying *** deformation mechanism of the alloy is the dislocation slip of theαandβphases and the twin deformation of theα*** large number ofα/βinterfaces produced by the fineαphase is the main reason for limiting the ductility of the *** use of the pseudo-spinodal mechanism combined with diffusion multiple experiments and CALPHAD is an effective method for designing high-strength titanium alloys.

摘要：随着深度学习的快速发展,相关算法被广泛应用于量子化学计算领域以实现高效的分子设计及性质研究,其中,多任务学习方法通过挖掘分子性质之间的关系可以同时预测多个分子属性,然而此类研究目前较为有限.本文采用硬参数共享结构与损失函数加权方法来实现多任务分子性质预测.通过对比单任务基准与各类多任务模型在不同分子属性集上的性能,展示了多属性预测精度强烈依赖于属性间的关系,当关联变复杂时,硬参数共享可以提高预测精度,此外,恰当的损失函数加权方法有利于实现更均衡的多目标优化,使预测更准确.进一步的实验展示了多任务学习模型的计算效率优势及其在训练数据量受限时的预测性能优势.

摘要：Controlling the polar order in ferroelectric materials may enrich the diversity of their property and functionality,offering new opportunities for the design of novel electronic and optoelectronic *** this paper,we report a planar multi-state memory device built upon a twodimensional(2D)van der Waals layered ferroelectric material,2Hα-In_(2)Se_(3).Three(high,median and low)resistance states are demonstrated to be interconvertible in this device with a fast switching speed,excellent endurance and retention performances via the modulation of the polar order of the ferroelectricα-In_(2)Se_(3) layers under an in-plane electric ***,reversible switching between the median-resistance state and the low-resistance state can be achieved by an ultralow electric field of 1-2 orders of magnitude smaller than the reported values in other 2D ferroelectric materialbased memory ***,the three different polar order states are discovered to exhibit distinctive *** results demonstrate great potentials ofα-In_(2)Se_(3)in nonvolatile high-density memory and advanced optoelectronic device applications.

摘要：随着材料基因组计划(MGI)及集成计算材料工程(ICME)的快速发展,材料设计已经从过去的经验试错法转变为基于数据库、计算模型、高通量实验的集成设计。本文回顾了国内外钛合金的发展历史与合金化思路,发现以Al、Zr、Sn、Mo、V、Cr、Fe多元素复合强化的近合金成为高性能合金发展趋势。在此基础上本文对钛合金的设计方法进行了总结,其中包含基于经验/半经验理论的试错法、Mo当量法、e/a法以及团簇-连接原子模型法和基于基础理论计算的Bo-Md方法和CALPHAD方法。最后,文章对结合第一性原理(ab-initio)、计算热力学(Calphad)、晶体塑性有限元模拟(CP-FEM)的集成计算材料设计方法进行了介绍,并对该方法在钛合金设计中的应用前景进行了展望。

摘要：Phase transformation of two-dimensional(2D)nanomaterials can lead to significant changes in electronic and optical properties,which enables the development of novel *** strategies for phase engineering of 2D nanomaterials have drawn considerable attention in recent *** review focuses on the state-of-the-art progress in the phase transformation of 2D nanomaterials and their catalytic ***,the basic concepts of phase transformation and the outstanding electronic and optical properties induced by phase transformation are briefly ***,different strategies for achieving phase transformation are discussed in detail and classified into several types based on their characteristics,including(i)doping,(ii)external fields,(iii)optical irradiation,(iv)strain effect,(v)high-energy particle excitation,and(vi)thermal *** applications of 2D nanomaterials in catalysis based on phase transformation have also been ***,a summary of the technical challenges to phase control in 2D nanomaterials and potential opportunities for developing novel applications is presented.

摘要：本文从人因学角度出发,研究人、民居与环境的关系,通过创新设计民居室内环境,最大程度地满足乡村居民对居住环境的需求,提高乡村居民的幸福感。

摘要：为了实现高效、公平、透明的配电网电力交易,提出了基于区块链的配电网电力交易方法。针对当前含新能源的配电网电力交易的难点,结合区块链的技术特点,提出了配电网电力交易的区块链结构。介绍了基于区块链的配电网电力交易的具体实现步骤和交易过程,深入研究基于XML组织区块链主体内容,并将国产安全算法SM2和SM3引入到区块链应用中。最后设计了基于区块链的配电网电力交易情景,证明了基于区块链的配电网电力交易方法的有效性。

摘要：以准噶尔盆地红山嘴油田红浅1井区火驱先导试验区为例,通过对矿场气体组分监测数据分析,可以快速确定油藏是否被点燃,是否处于高温燃烧状态;在还没有十分有效手段监测火线的情况下,通过对气体组分监测数据的分析,判断空气在油藏中的展布方向,为火驱生产动态调控提供依据,及时对生产井出现的异常状况进行初步判定。通过对H2S在红浅1井区火驱生产过程中的变化规律的重新认识,为火驱工业化扩大地面工程方案设计提供数据。

摘要：针对传统系统太阳辐照波动预警准确率低、误差大和预测效率低的问题,提出并设计了新的并网型分布式光伏发电辐照波动预警系统,采用基于蚁群BP神经网络的辐照波动预测方法获取了预测结果的最优解。预警系统中包含太阳能电池组件、微型逆变器和单机辐照度预警子系统等,采用三端稳压模块LM7803电源模块,保障极端天气正常供电;选择单晶硅太阳电池片的光照传感器,将太阳辐射的光能变换成电能;采用指数平滑法结合趋势移动平均法对太阳辐照进行预测,得到预测时间段范围内的平滑指数,以此作为计算参数,根据蚁群BP神经网络算法,通过构建新传递函数,优化网络结构,改进BP神经网络训练,求出预测结果最优解。试验结果表明:所提系统能在不同天气条件下有效预测并网型分布式光伏发电的辐照波动,且预测误差较小;在不同天气类型下,系统预警的准确率始终高于98%,且预警时长平均用时24.08 s,相比于其他系统用时较短,实现预警系统稳定运行,保证输出电能质量,有力支撑系统优化。

摘要：将基于空间二叉分割理论的无重访机制与基于多目标Pareto最优化思想的第二代非支配排序遗传算法(Non-dominated Sorting Genetic Algorithm-Ⅱ,NSGA-Ⅱ)相结合,构建了无重访NSGA-Ⅱ算法,并应用于求解同时考虑网损降低和供电可靠性提高的多目标配电网络重构问题。所构建的无重访NSGA-Ⅱ算法实现了严格意义上的不重复搜索,避免了重复方案的潮流及可靠性计算,节约了计算资源。IEEE16、IEEE33测试系统的计算结果表明能够在较少的迭代次数下得到每个目标方向上的最优解以及包含若干非支配解的Pareto最优前沿解集。根据网损与可靠性目标之间的关联关系及相应重构方案的拓扑结构分析表明在解空间的全局范围内网损与可靠性目标具有较明显的一致关联性,不论对于网损还是可靠性的优化,网络拓扑都应该接近广度优先树而规避深度优先树。