检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
作 者:孔锤锐 李占锋[1] 武志翔[2] 邓琥[2] 尚丽平[2] KONG Chuirui;LI Zhanfeng;WU Zhixiang;DENG Hu;SHANG Liping(School of Manufacturing Science and Engineering,Southwest University of Science and Technology,Mianyang 621010,China;School of Information Engineering,Southwest University of Science and Technology,Mianyang 621010,China)
机构地区:[1]西南科技大学制造科学与工程学院,四川绵阳621010 [2]西南科技大学信息工程学院,四川绵阳621010
出 处:《兵器装备工程学报》2023年第10期266-271,286,共7页Journal of Ordnance Equipment Engineering
基 金:国家自然科学基金项目(62105271);四川省科技计划项目(2020YJ0160);西南科技大学自然科学基金项目(19zx7160);西安近代化学研究所项目(SYJJ20210411/204-J-2020-2027/SYJJ20210407)。
摘 要:安定剂的检测是推进剂和发射药状态评估的重要环节。为实现安定剂的快速无损检测,研究了紫外-可见漫反射光谱法对安定剂进行定性定量检测的可行性。搭建了一套运用光纤传感的光谱测量系统,获取了3种安定剂样本的光谱。通过支持向量机(SVM)对光谱进行分类识别,运用主成分分析(PCA)可视化样本的聚类趋势;结合化学计量学方法,以N-甲基-4-硝基苯胺(MNA)470~500 nm波段的光谱强度作为特征数据通过偏最小二乘回归(PLSR)和主成分回归(PCR),建立了2个MNA的定量预测模型。结果表明:运用SVM算法可以对具有浓度变化的3种安定剂样本实现分类识别,在测试集中分类的准确率达到100%;2个定量预测模型的决定系数(R^(2))分别为0.9939和0.9946,模型外部验证的均方根误差(RMSE)最大为0.00023,可以对MNA进行定量表征。该方法在安定剂检测领域具有较大的应用潜力,并可扩展至其他检测领域。The stabilizer measurements are an important part of the propellant status assessment.To achieve rapid and non-destructive inspection of stabilizers,the feasibility of qualitative and quantitative analysis of stabilizers by UV-vis diffuse reflectance spectroscopy are investigated.A spectral measurement system using fiber optic sensing is built to acquire the spectra of three stabilizer samples.The spectra are classified and identified by Support Vector Machine(SVM),and Principal Component Analysis(PCA)is applied to visualize the clustering trend of the samples.In combination with chemometrics,two quantitative prediction models are developed by partial least squares regression(PLSR)and principal component regression(PCR)using the spectral intensity of N-methyl-4-nitroaniline(MNA)in the 470~500 nm band as characteristic data.The results show that SVM can classify three types of stabilizer samples,and the accuracy of the classification in the test set reaches 100%.The coefficients of determination(R^(2))of the two quantitative prediction models are 0.9939 and 0.9946,respectively,and the root mean square error(RMSE)of the external validation of the models is 0.00023 at maximum,allowing quantitative analysis of MNA.The method has great potential for application in the field of stabilizer detection and can be extended to other detection fields.
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:3.141.19.32