Handheld 3D scanning technology plays a crucial role in wheel wear monitoring of trains. However, without strong feature of the measurement target, generating a precise 3D ground truth to test the handheld 3D scanning algorithm is really hard. To address this issue, this paper proposes a handheld multi-line laser scanning simulation system based on an optical-inertial approach. Leveraging the node-based programming capabilities of the Blender platform, this system accurately simulates the scanning data of multi-line laser image acquisition and 6 degrees of freedom pose and position. By integrating data collected from a real IMU and analyzing the pose transformation between two acquisitions, the real motion process can be simulated, and a virtual multi-line laser scanning image dataset containing complex backgrounds is constructed. In experiments simulating train wheel scanning scenarios, 200 images were generated. The results demonstrate that the proposed system can produce high-quality multi-line laser scanning images of marker-less and texture-less objects while providing effective ground truth pose data. This simulation system provides a validation platform for 3D vision algorithms in handheld scanning techniques. The platform is low-cost, highly flexible, and data-reproducible.