当前位置: X-MOL 学术Wood Sci. Technol. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Prediction of mechanical properties of wood fiber insulation boards as a function of machine and process parameters by random forest
Wood Science and Technology ( IF 3.4 ) Pub Date : 2020-05-01 , DOI: 10.1007/s00226-020-01184-3
M. Schubert , M. Luković , H. Christen

In this case study, machine and process variables were extracted from the process control system (Prod-IQ) and combined with tested mechanical properties of wood fiber insulation boards according to product type and time of manufacture. The boards were taken from the production line (dry process), and the internal bond strength ( σ mt ) and the compressive strength at 10% deformation ( σ 10 ) were determined according to the European Standard EN 826 and 1607. The complete data set was preprocessed and split into training and test sets using k-fold cross-validation. The performance of the random forest algorithm (RF) was evaluated with the correlation coefficient ( R ), the coefficient of determination ( R 2 ), root-mean-square error (RMSE) and mean absolute percentage error (MAPE) and compared with artificial neural networks (ANN) and support vector machines (SVM). Forward feature selection was used to reduce input dimensionality and improve the generalizability of the algorithms. All machine learning algorithms predicted the mechanical properties with high accuracy, but the RF algorithm revealed the best generalization performance ( σ mt : R = 0.960, R 2 = 0.916, RMSE = 4.05, MAPE = 12.11; σ 10 : R = 0.981, R 2 = 0.963, RMSE = 17.19, MAPE = 5.64). This work demonstrates that machine learning can be applied to predict relevant properties of wood fiber boards for an improved quality control in real time.

中文翻译:

通过随机森林预测作为机器和工艺参数的函数的木纤维绝缘板的机械性能

在本案例研究中,机器和过程变量是从过程控制系统 (Prod-IQ) 中提取的,并结合根据产品类型和制造时间测试的木纤维绝缘板的机械性能。板材取自生产线(干法),根据欧洲标准EN 826和1607测定内部结合强度(σ mt )和10%变形时的抗压强度(σ 10 )。 完整数据集被预处理并使用 k 折交叉验证分为训练和测试集。随机森林算法 (RF) 的性能通过相关系数 ( R )、决定系数 ( R 2 )、均方根误差 (RMSE) 和平均绝对百分比误差 (MAPE),并与人工神经网络 (ANN) 和支持向量机 (SVM) 进行比较。前向特征选择用于降低输入维度并提高算法的通用性。所有机器学习算法都以高精度预测机械特性,但 RF 算法显示出最佳的泛化性能(σ mt : R = 0.960, R 2 = 0.916, RMSE = 4.05, MAPE = 12.11; σ 10 : R = 0.981, R 2 = 0.963,RMSE = 17.19,MAPE = 5.64)。这项工作表明,机器学习可用于预测木纤维板的相关特性,以实时改进质量控制。所有机器学习算法都以高精度预测机械特性,但 RF 算法显示出最佳的泛化性能(σ mt : R = 0.960, R 2 = 0.916, RMSE = 4.05, MAPE = 12.11; σ 10 : R = 0.981, R 2 = 0.963,RMSE = 17.19,MAPE = 5.64)。这项工作表明,机器学习可用于预测木纤维板的相关特性,以实时改进质量控制。所有机器学习算法都以高精度预测机械特性,但 RF 算法显示出最佳的泛化性能(σ mt : R = 0.960, R 2 = 0.916, RMSE = 4.05, MAPE = 12.11; σ 10 : R = 0.981, R 2 = 0.963,RMSE = 17.19,MAPE = 5.64)。这项工作表明,机器学习可用于预测木纤维板的相关特性,以实时改进质量控制。
更新日期:2020-05-01
down
wechat
bug