Abstract: In view of the inadequacy of magnetic memory detection method for easy interference by external factors in the internal detection of oil and gas pipelines, an internal detection method of long-distance oil and gas pipeline stress using dual magnetic field intensity was proposed. Herein, an analytical model was established based on J-A theory for the internal detection of pipeline stress.Meanwhile, analysis was performed on the detection characteristics of stress signals under different magnetic field intensities, and the effectiveness of pipeline stress internal detection using strong and weak magnetic fields was verified by tests. The results show that the pipeline stress has little influence on magnetic signal under strong magnetic field, and the influence gradually decreases with the increasing saturation. However, the pipeline stress can affect the intensity of magnetic signal under the weak magnetic field, and accordingly, the stress concentration zone can be detected. When the external magnetic field intensity is 5 000 A/m, the recognition capability is 50% for 300 MPa stress and 4% for the volume defect of 10 mm length, 10 mm width and 10mm depth. When the external magnetic field intensity is 30 kA/m, the recognition capability is 1.74% for 300 MPa stress and 40% for the same volume defect. Hence, the pipeline damage can be judged according to the signal characteristics under magnetic field of different intensities at the same position during the inline detection with dual magnetic field detector. There is a stress concentration zone in the wall when the signal character is shown for the weak magnetic joint but not for the strong magnetic joint. Volume defects are present in the pipe wall when the signal character is shown for the strong magnetic joint but not for the weak magnetic joint. In addition, a stress concentration area is present at the volume defect on the pipe wall when signal characters are shown for both strong and weak magnetic joints, to which special attention shall be paid. The research results provide theotical guidance for stress detection at defects.