Research Initiation: Intelligent Machining Monitoring and Diagnostic System for Quality Assurance Based on Multiple Sensor Integration

研究发起：基于多传感器集成的质量保证智能加工监控与诊断系统

• 批准号: 9111712
• 负责人: Anthony Okafor
• 金额: $ 6万
• 依托单位: Missouri University of Science and Technology
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1991
• 资助国家: 美国
• 起止时间: 1991-06-01 至 1993-11-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9111712&HistoricalAwards=false
• 关键词: Research; Initiation; Intelligent; Machining; Monitoring

The main objective of this research is to develop and demonstrate an intelligent machining monitoring and diagnostic system for quality assurance in end-milling. This will be based on the integration of multiple sensors outputs, such as acoustic emission, vibration, cutting force, and cutting temperature, with cutting conditions via artificial neural networks in conjunction with an expert system. Product quality characteristics such as surface finish and tolerance will be estimated and, therefore, eventually maintained by the system during end-milling. The research team will conduct machining experiments to acquire data and determine the best sensor feature that correlates to product quality, develop multi-sensor integrated neural networks for estimating product quality, develop trend classification neural network for identifying trends and train and evaluate the performance of these networks. Product quality in manufacturing is of paramount importance especially in view of the tremendous competition that we face in the globalized marketplace. Pioneering work of this nature, where the latest developments in artificial intelligence are applied to basic problems in manufacturing and in close cooperation with a United States machine tool manufacturer, will help domestic manufacturing compete successfully.

本研究的主要目的是开发和演示一种智能加工监控和诊断系统，以保证立铣削的质量。这将基于多个传感器输出的集成，如声发射、振动、切削力和切削温度，以及通过人工神经网络与专家系统结合的切削条件。产品质量特征，如表面光洁度和公差将被估计，因此，最终由系统在铣削过程中保持。研究团队将进行加工实验，获取数据并确定与产品质量相关的最佳传感器特征，开发多传感器集成神经网络以估计产品质量；开发趋势分类神经网络，用于识别趋势，并训练和评估这些网络的性能。制造业的产品质量是至关重要的，特别是考虑到我们在全球化市场中面临的巨大竞争。这种开创性的工作，将人工智能的最新发展应用于制造业的基本问题，并与美国机床制造商密切合作，将有助于国内制造业成功竞争。