基于主列筛选的动态测试传感器配置方法研究

动态测试是获取结构真实动态特性的重要手段之一,传感器配置对于保证动态测试数据的信息量有着重要的影响. 尤其是对以试验模态分析为目标的动态测试,传感器配置对于模态分辨率更是决定性因素. 在振型独立性的基础上,提出了一种基于主列筛选的传感器配置方法. 与现有的逐步累积类或逐步消减类方法相比,该方法无需迭代,一次选择所有测试自由度,计算量小. 为了量化比较传感器配置的效果,定义了4个度量指标. 采用卫星天线模型进行了数值仿真,结果验证了基于主列筛选的传感器配置方法的有效性.      

Relation between silico-aluminous fly ash and its coal of origin

Fly ashes are typical complex solids which incorporate at the same time intrinsic properties derived from the layers （various mineralogical and dimensional spectra） and major transformations generated during prior processing. To use fly ashes in various applications, it is necessary to characterise them completely. The first research to date carried out on silico-aluminous fly ashes in order to characterise them physically, morphologically, chemically and mineralogically, resulted in the recognition that they are relatively simple materials. In the present study, a silico-aluminous fly ash coming from the power station of Albi （France） was selected. Heat treatment at 450 and 1200℃ together with coal simulated the treatment undergone by coal in the power station in order to mimic real coal residue. In conclusion, the diversity of the particles contained in fly ash could only be exolained by the relation existing between the fly ash and its coal of origin.     

利用肽库技术筛选成纤维细胞生长因子的短肽亲和配体

成纤维细胞生长因子(FGF)具有促进血管和神经形成的功能^[1],但它在体内作用过度则经常伴随着肿瘤的发生^[2].当前,研制和开发FGF拮抗剂,以有效地抑制FGF与细胞受体的结合,已成为国际性前沿课题.     

Optimal Control of Parabolic Hemivariational Inequalities

In this paper we study the optimal control of systems driven by parabolic hemivariational inequalities. First, we establish the existence of solutions to a parabolic hemivariational inequality which contains nonlinear evolution operator. Introducing a control variable in the second member and in the multivalued term, we prove the upper semicontinuity property of the solution set of the inequality. Then we use this result and the direct method of the calculus of variations to show the existence of optimal admissible state–control pairs.     

基于模拟退火算法的近红外光谱定标模型的简化

对模拟退火算法进行了优化,并以小麦粉样品的光谱为对象,利用优化后的算法在分子含氢基团一倍频区域1400-1860nm(7144-5376cm-1)优选出了分析小麦蛋白质含量的4个特征波长,结合多元线性回归(MLR)建立了定标模型.同时,利用这个谱区的全谱数据,结合偏最小二乘法(PLS)建立了另外一个模型.经过比较,发现这两个模型具有相同的预测效果.说明优化后的模拟退火算法能很好地用于近红外光谱分析中定标波长的选取,进而建立"精而简"的模型.这对于简化定标模型及确定分立波长型仪器的定标波长具有十分重要的价值.同时,对处理其他优化组合问题也有一定的指导意义.     

甘蔗初压汁锤度近红外光谱分析的波段优选

用近红外透射光谱技术和偏最小二乘法建立甘蔗初压汁锤度的定量分析模型。用多元散射校正和Savitzky-Golay平滑化方法对原谱、一阶导数谱和二阶导数谱进行预处理。选取400—1100nm、1100—1400nm、1550—1850nm、2100—2350nm以及全谱（400—2500nm）5个波段,每个波段分别采用原光谱、一阶导数谱、二阶导数谱。同时调整Savitzky-Golay平滑点数和PLS因子数,通过多次PLS数值实验比较,按照预测效果确定每个模型的最优平滑点数、因子数,再从中选优。结果表明,采用二阶导数谱的1550-1850nm波段的定标效果最好,模型的预测相关系数、预测均方根偏差、相对预测均方根偏差分别为0.896,0.406,1.95%。1550—1850nm波段可以代替全谱波段得到好的定标效果,为设计小型专用近红外分析仪器提供依据。     

基于模糊集理论的承包商资格预审模型

招标方在对承包商的资格预审过程中,需要对承包商的财务状况、技术能力、施工经验、社会信誉等方面进行评审,以确定有资格参加投标的承包商,针对有些评审指标难以量化以及评审过程中决策者主观判断影响最终结果的情况,提出了基于模糊集的资格预审模型,并通过算例验证了该模型的可行性。