Digital Twins and Data-Driven in Plant Factory: An Online Monitoring Method for Vibration Evaluation and Transplanting Quality Analysis

The plant factory transplanter is a key component of the system. Its operation status directly affects quality and survival rate planted seedlings, which in turn overall yield economic efficiency. To monitor transplanting machine timely manner, primary task to use computerized easy-to-use method units. Inspired by latest developments augmented reality robotics, digital twin model-based data-driven online monitoring for equipment proposed. First, virtual model approach combined construct multi-domain equipment. Then, taking vibration frequency domain signal above manipulator image features seedling tray as input variables, evaluation configuration system are Finally, effect evaluated, cycle can be repeated optimize achieve optimal parameters. results show that effectively sensor data identify mechanical characteristics avoid affecting due resonance. At 3000 plants/h, efficiency maintained at high level X, Y, Z-axis relatively calm. In this case, Combined threshold with traditional Wiener algorithm, identification healthy potted seedlings reach 94.3%. Through comprehensively using 3D block matching filtering algorithm segmentation denoising, recognition has reached over 96.10%. addition, developed predict operational timing detected transplanter, even if environmental not included training. proposed used damage detection effectiveness assessment other structures.