在役沥青路面结构会经历疲劳累加和损坏扩展两个阶段，现有的弯沉指标和反演模量难以精确表征损坏扩展阶段的结构性能，使得在役沥青路面结构性能演化模型难以准确建立。本项目通过改进落锤式弯沉仪采集方式，获得更为丰富的在役路面的空间结构性能信息，基于模式识别理论研究在役沥青路面结构性能解析和演化模型。首先对疲劳累加阶段和损坏扩展阶段进行科学识别；接着分别通过基于弯沉盆关键节点进行的模量动力反演法和采用模式识别进行的损坏形态解析法对两阶段的结构性能进行合理解析和表征；最后，基于典型高速公路长期监测数据，创建动态模量疲劳衰减模型和损坏形态演化模型并进行实地验证。本项目对改进现有沥青路面结构设计方法，提高道路养护维修设计的科学性有重要意义。

The evolution of pavement structural performance could be divided into a fatigue accumulation stage and a distress propagation stage. The existing deflection parameter indexes and backcalculated modulus could not precisely characterize pavement structural performance at the distress propagation stage. This makes it very difficult to establish a precise structural performance evolution model of in-service asphalt pavement. In this project, abundant spatial structural performance information of in-service asphalt pavement could be collected through the improvement of falling weight deflectometer (FWD). And based on pattern recognition theory, the structural performance analytical method and evolution model for in-service asphalt pavement would be studied. Firstly, fatigue accumulation stage and distress propagation stage would be recognized scientifically. Then, dynamic modulus backcalculation method based on deflection bowl key points and distress mode analytical method based on pattern recognition would be used for the structural performance analysis and characterization of the two stage, respectively. Finally, based on long-term monitoring data of typical expressway, modulus attenuation model and distress mode evolution model would be established and validated. This project is of great significance for supplementing existing asphalt pavement design method and improving the rationality of road maintenance and rehabilitation design.