Abstract: Massive data of bending strain along pipelines can be detected by the Inertial Measurement Unit(IMU), but there is no effective method available for the quantitative evaluation of pipelines with the IMU data so far. To address this issue, an identification method was proposed for thaw settlement segments of pipelines in permafrost zones based on IMU strain data. The method mainly includes the pre-processing method of IMU strain data, i.e., the data alignment method, the identification method of large strain characteristics of the non-thaw settlement pipeline segments, such as bends, dents and tees, and the quick positioning method of geological disaster pipeline segments determined by presetting the identification threshold and the abnormal pipeline segments to be inspected. Thus, the IMU detection data of the First Mohe-Daqing Pipeline in the permafrost region during 2013 to 2018 were identified and analyzed, and the identification results were in good agreement with the geometric inspection data. On that basis, a statistical analysis was made to study the distribution of abnormal pipeline segments in various areas of this permafrost region. The results indicate that the seasonal frozen region has more geological change points than the permafrost region, but the change is small. Generally, quantitative data are provided through this research for the integrity assessment of pipelines in geological disaster areas, which could be used to solve the technical bottleneck of failure to determine the external load of pipelines at present. Hence, it is a significant new direction for the development of pipeline integrity assessment technology.