为了优化沥青老化特征的快速评价方法与效率,基于动态时间规整(dynamic time warping,DTW)算法优化不同老化程度的沥青红外光谱信息评价方法,提出了老化沥青光谱信息相似度SIS评价指标.研究表明,基于DTW算法的最小规整距离可以体现老化沥青中所有化学组成结构累计变化信息量,进而表明沥青的热氧老化程度.沥青相似度(spectral information similarity,SIS)与主要特征峰(亚砜基、羰基)的相关系数不低于0.954 4,与沥青的老化程度具有很强的相关性;并且相对于单一组分特征评价指标,SIS指标的离散系数可以降低92％,显著降低红外光谱规律评价结果的误判概率.说明通过与单一组分的定性评价结合使用,SIS指标可作为评价沥青老化程度的判别依据,且该评价方法也可为光谱信息的相似性鉴别与归类研究提供参考与借鉴.
Similarity Evaluation of Asphalt Aging Infrared Spectrum Based on DTW Algorithm
In order to optimize the rapid evaluation method and efficiency of asphalt aging characteristics,the infrared spectral information evaluation method of asphalt with different aging degrees was optimized based on DTW(dynamic time warping)algorithm,and the spectral information similarity SIS(spectral information similarity)evaluation index of aging asphalt is proposed.The results show that the minimum regularized distance based on DTW algorithm can reflect the amount of information about the cumulative changes in the structure of all chemical compositions in aged asphalt,which in turn indicates the degree of thermo-oxidative aging of asphalt.The correlation coefficient between SIS and the main characteristic peaks(sulfinyl and carbonyl)is not less than 0.954 4,which is strongly correlated with the aging degree of asphalt.The dispersion coefficient of the SIS index can be reduced by 92％compared with the single-component characterization index,which significantly reduces the probability of misclassification of the evaluation results of infrared spectral pattern.It shows that by combining with single-component qualitative evaluation,the SIS index can be used as the basis for evaluating the aging degree of asphalt,and the evaluation method can also be used as a reference for spectral information similarity identification and categorization research.
万方数据
维普
