针对增材制造过程质量不稳定的问题,提出一种基于声音识别技术的增材制造过程质量预测(Sound Recognition-based Additive Manufacturing Process Quality Prediction,SRAM-PQP)方法.该方法通过音频信号预处理、声音特征提取、机器学习模型训练,实现增材制造零件缺陷的精准预测.实证结果表明,SRAM-PQP方法的预测准确率达 96.67%,F1 值达96.75%,对不同缺陷类型均展现出良好的预测性能.
Research on Quality Prediction Technology in Additive Manufacturing Process Based on Sound Recognition Technology
Aiming at the problem of quality instability in additive manufacturing process,a Sound Recognition-based Additive Manufacturing Process Quality Prediction(SRAM-PQP)method was proposed.Through audio signal preprocessing,sound feature extraction and machine learning model training,the method can accurately predict the defects of additive manufacturing parts.The empirical results show that the prediction accuracy of SRAM-PQP method is 96.67%,F1 value is 96.75%,showing good prediction performance for different defect types.
NETL
NSTL
万方数据
