沥青材料的快速识别与分析方法研究

发布时间：2018-06-20 05:05

本文选题：衰减全反射红外光谱 + 蜡含量　；参考：《北京化工大学》2015年硕士论文

【摘要】：沥青是高速公路的基本建筑材料,其质量的优劣直接影响着其路用性能和公路寿命。一般道路设计寿命为15年,而劣质或掺假沥青产品会导致路面2年后就进入维护期,给高速公路建设和维护带来巨大经济损失。因此,道路沥青生产、流通和道路施工等过程中沥青产品质量的有效监控至关重要。蜡含量、软化点、针入度和SBS改性沥青中SBS含量是沥青重要的工程质量指标。其现行的标准方法分析过程繁冗耗时,无法满足实际过程中沥青质量检测与监管需求。目前也缺乏快速的沥青产品溯源方法。因此研究建立上述沥青质量指标的快速分析技术和产品溯源方法已经成为该领域中亟待解决的重大课题。分子光谱结合化学计量学,具有分析快速和测量方便等优点。本课题旨在运用傅里叶变换红外光谱和化学计量学,建立沥青重要质量指标和油源产地识别快速分析方法。首先,本文提出了一种利用衰减全反射傅里叶红外光谱技术(FTIR-ATR)实现沥青性质快速准确检测的方法。实验共收集了不同公司、不同时间生产的沥青样品220个,其蜡含量、软化点和针入度标准值己依照标准方法SH/T 0425-2003、GB/T 4507-2014、GB/T4509-2010测定。研究了沥青取样、光谱内部扫描次数和重复装样次数对沥青光谱测量精度的影响,确定了最佳光谱采集条件。分别采用相关系数法和CARS选择波长变量,采用偏最小二乘法(PLS)建立了蜡含量、软化点和针入度的红外光谱定量模型,交互验证标准误差(SECV)分别为0.13、0.88、2.84,预测标准偏差(SEP)分别为0.14、1.06、3.02,均不大于标准方法的再现性误差。使用该方法对任意选取不同厂家所生产的3个样品进行重复性测试,精密度不大于标准方法的精密度误差。其次,本文提出了一种利用衰减全反射傅里叶红外光谱技术(FTIR-ATR)实现沥青油源产地快速准确判别的方法。实验共收集了220个包括南美重油、西北局原油、欢喜岭原油和新疆稠油四个不同油源产地的沥青样品。利用PCA和SIMCA方法对沥青油源产地进行判别,PCA模型有效的识别了以欢喜岭原油和新疆稠油为油源的沥青样品油源信息,无法区分以南美重油和西北局原油为油源的沥青样品油源信息。SIMCA方法在PCA方法的基础上实现了以南美重油和西北局原油为油源的沥青样品的油源信息判别,实现了不同油源产地沥青样品100%的识别率。最后,本文制备了SBS含量为4.0%-6.0%一系列浓度梯度的SBS改性沥青样品,通过多元曲线分辨交替最小二乘结合斜投影建立标准曲线,其预测结果的标准偏差为0.096,能够实现改性沥青的质量控制。
[Abstract]:Asphalt is the basic building material of highway, its quality directly affects its road performance and highway life. The general road design life is 15 years, but inferior or adulterated asphalt products will lead to pavement maintenance period after 2 years, which will bring huge economic loss to highway construction and maintenance. Therefore, it is very important to monitor the quality of asphalt products in the process of road asphalt production, circulation and road construction. Wax content, softening point, penetration degree and SBS content in SBS modified asphalt are important engineering quality indexes. The present standard method is time-consuming and time-consuming, and can not meet the requirements of asphalt quality inspection and supervision in the actual process. There is also a lack of rapid traceability of asphalt products. Therefore, it has become an important task to establish the rapid analysis technology and product traceability method of asphalt quality index mentioned above. Molecular spectrum combined with chemometrics has the advantages of fast analysis and convenient measurement. The purpose of this paper is to establish a fast analysis method for asphalt quality index and oil source identification by Fourier transform infrared spectroscopy and chemometrics. Firstly, a method of fast and accurate detection of asphalt properties using attenuated total reflectance Fourier transform infrared spectroscopy (FTIR-ATR) is proposed. A total of 220 asphalt samples were collected from different companies and at different times. Their wax content, softening point and penetration were determined according to the standard method SHP T 0425-2003 GB / T 4507-2014 GB / T 4509-2010. The effects of bitumen sampling, spectral internal scanning times and repeated sample loading times on the accuracy of asphalt spectral measurement were studied, and the optimum spectral acquisition conditions were determined. The infrared spectral quantitative models of wax content, softening point and penetration were established by means of correlation coefficient method and car selection wavelength variable, and partial least square method (PLS). The cross validation standard error (SECV) is 0.13 / 0.88 / 2.84, and the prediction standard deviation (SEP) is 0.14 / 1.06 / 3.02, respectively, which is less than the reproducibility error of the standard method. This method is used to test the repeatability of 3 samples produced by different manufacturers. The precision is less than the precision error of the standard method. Secondly, a fast and accurate method for determining the origin of asphalt oil source by using the attenuated total reflectance Fourier transform infrared spectroscopy (FTIR-ATR) is proposed. A total of 220 bitumen samples from four different oil sources including South American heavy oil, Northwest Bureau crude oil, Huanxiling crude oil and Xinjiang heavy oil were collected. The PCA and Simca methods were used to identify the oil source information of asphalt samples from Huanxiling crude oil and Xinjiang heavy oil. Based on PCA method, the oil source information of bitumen samples with South American heavy oil and Northwest Bureau crude oil as oil source can not be distinguished by SIMCA method, and the oil source information of asphalt samples with South American heavy oil and Northwest Bureau crude oil as oil source can be distinguished by PCA method. The recognition rate of bitumen samples from different oil sources was 100%. Finally, a series of SBS modified asphalt samples with a concentration gradient of 4.0% -6.0% were prepared, and the standard curves were established by multivariate curve resolution alternating least squares and oblique projection. The standard deviation of the predicted result is 0.096, which can realize the quality control of modified asphalt.
【学位授予单位】：北京化工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U414;O657.33

【参考文献】