Xingtao Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xingtao Wang has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Deyong Zhao
College of Mechanical and Transportation Engineering, Southwest Forestry University, Kunming 650224, China
1 shared publication - Dong Wang
Key Laboratory of State Forestry and Grassland Administration on Highly-Efficient Utilization of Forestry Biomass Resources, Southwest Forestry University, Kunming 650224, China
1 shared publication - Yang Dai
Key Laboratory of State Forestry and Grassland Administration on Highly-Efficient Utilization of Forestry Biomass Resources, Southwest Forestry University, Kunming 650224, China
1 shared publication - Yuan Wang
College of Mechanical and Transportation Engineering, Southwest Forestry University, Kunming 650224, China
1 shared publication - Zhichang Guo
Key Laboratory of State Forestry and Grassland Administration on Highly-Efficient Utilization of Forestry Biomass Resources, Southwest Forestry University, Kunming 650224, China
1 shared publication
Latest Publications
- Open Access
Sawing Model and Optimization of Single Pass Crosscut Parameters for Pinus kesiya Based on the Transformer Model
The optimization of the sawing process for Pinus kesiya Royle ex Gordon, an important timber used in construction and furniture, especially through the adjustment of parameters such as wood moisture content, cutting speed, and feed speed, not only helps reduce energy consumption and noise but also improves surface processing quality, thereby promoting green and environmentally friendly production. Therefore, this study selects the wood moisture content, cutting speed, and feed speed during the Single Pass Crosscut process of Pinus kesiya as the preliminary experimental parameters. Based on the Transformer model, a cutting prediction model for Pinus kesiya is established to predict three cutting performance indicators: cutting power consumption, surface roughness, and cutting noise. Meanwhile, Bayesian optimization was used to search for the optimal parameter combination within the specified cutting process parameter ranges that minimizes the objective function for these cutting performance indicators. Finally, experimental verification based on the optimal parameter combination shows that the average coefficient of determination for the cutting performance indicators is 0.937, the average mean squared error is 0.076, and the average mean absolute error is 0.186, indicating good agreement between the predicted and measured values.
Open AccessSawing Model and Optimization of Single Pass Crosscut Parameters for Pinus kesiya Based on the Transformer Model
The optimization of the sawing process for Pinus kesiya Royle ex Gordon, an important timber used in construction and furniture, especially through the adjustment of parameters such as wood moisture content, cutting speed, and feed speed, not only helps reduce energy consumption and noise but also improves surface processing quality, thereby promoting green and environmentally friendly production. Therefore, this study selects the wood moisture content, cutting speed, and feed speed during the Single Pass Crosscut process of Pinus kesiya as the preliminary experimental parameters. Based on the Transformer model, a cutting prediction model for Pinus kesiya is established to predict three cutting performance indicators: cutting power consumption, surface roughness, and cutting noise. Meanwhile, Bayesian optimization was used to search for the optimal parameter combination within the specified cutting process parameter ranges that minimizes the objective function for these cutting performance indicators. Finally, experimental verification based on the optimal parameter combination shows that the average coefficient of determination for the cutting performance indicators is 0.937, the average mean squared error is 0.076, and the average mean absolute error is 0.186, indicating good agreement between the predicted and measured values.