A Cost Optimized Data Acquisition System For Predictive Maintenance

In this paper, we focus on the development and application of a cost optimized sensor platform for predictive maintenance of industrial fans. Typically, condition monitoring systems that include threshold and frequency analysis are expensive and sometimes exceed the costs of the monitored device. With respect to safety issues, supervision of the health state of machinery – such as large industrial radial fans – is eminent. The proposed sensor platform is a cheaper alternative to commercial solutions and provides maximum flexibility concerning sampling rate, mobility, adaptability, and expandability. The prototype is used to collect relevant process data of a real industrial radial fan. In future, raw data shall be used to develop models for predictive maintenance.