GOALI: Integrated Signal Processing for Bearing Health Assessment and Life Prediction

这一与行业学术联系机会(GOALI)为开发与滚动轴承的健康评估和故障时间预测密切相关的统一的时间-尺度-频域信号处理技术提供了资金。该项目旨在弥合学术界最先进的信号处理研究与工厂车间当前轴承状况监测实践之间的现有差距，并将重点为有效和高效的算法建立理论基础，以估计有缺陷的轴承的损坏状态和剩余使用寿命。研究将包括多个领域的数值模拟，定制设计的实验验证，以及对该项目工业合作伙伴提供的历史数据的系统分析。开发的技术将以轴承评估固件的形式实施，驻留在数字信号处理器平台上，并安装在适合直接向轴承制造商和最终用户进行行业翻译的手持轴承数据分析仪中。如果这项研究成功，将提高在多个领域同时提取机器故障特征的能力。这种技术对于减少昂贵和意外的机器停机时间以及提高产品质量至关重要。在更广泛的背景下，这项研究将加强现有的制造和机械工程课程，创建以研究为基础的新教材，促进一所主要公立大学与轴承行业之间的合作，并改善PI机构的整体教育基础设施。最终，通过能够更可靠地预测机器系统的健康状态并提高操作安全性和生产率，这项研究有可能对轴承以外的广泛制造工艺产生影响。

This Grant Opportunity for Academic Liaison with Industry (GOALI) provides funding for the development of a unified time-scale-frequency domain signal processing technique that is germane to the health assessment and time-to-failure prediction of rolling bearings. The project is aimed at bridging an existing gap between the state-of-the-art of signal processing research in the academia and current bearing condition monitoring practice on the factory floor, and will focus on establishing the theoretical foundation for an effective and efficient algorithm to estimate a defective bearing's damage status and remaining service life. The research will consists of numerical simulations in multiple domains, custom-designed experimental validation, and systematic analysis of historical data provided by the industrial partner of this project. The developed technique will be implemented in the form of a bearing assessment firmware, residing on a Digital Signal Processor platform, and installed in a handheld bearing data analyzer, suited for direct industry translation to bearing manufacturers and end-users.This research, if successful, will lead to improved capability of concurrent machine fault feature extraction in multiple domains. Such a technique will be essential to reducing costly and unexpected machine downtime, and improving product quality. In a broader context, the research will enhance existing curricula in manufacturing and mechanical engineering, create new, research-based educational materials, promote collaborations between a major public university and the bearing industry, and improve the overall education infrastructure at the PI's institution. Ultimately, by being able to more reliably predict the health status of a machine system and improve operation safety and productivity, the research has the potential to impact a wide range of manufacturing processes other than bearings.