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

• DOI: 10.20944/preprints202305.0619.v1
• Publication Date: 2023-05-10T06:38:16Z
• Authors: Kaikang Chen; Yanwei Yuan; Bo Zhao; Liming Zhou; Kang Niu; Xin Jin; Shengbo Gao; Ruoshi Li; Hao Guo; Yongjun Zheng
• Keywords: Transplanting; Factory (object-oriented programming); Plant factory

Plant Factory Transplanter (PFT) is a key component of the plant factory 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, DT model-based data-driven online monitoring for equipment proposed. First, virtual model approach combined construct multi-domain digital twin equipment. Then, taking vibration frequency domain signal above manipulator image features seedling tray as input variables, evaluation configuration PFT system are Finally, effect transplanter evaluated, cycle can be repeated optimize achieve optimal parameters. The 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.