Efficiency of diagnosing the condition of rolling bearings in real time

Abstract

One of the control elements of a complex technological system is an information-and-measuring network, a significant part of which is represented by measuring transducers and sensors. Due to the design features, the sensors turn out to be unreliable elements of information-and-measuring systems. Distortions in sensor readings affect the quality of controlling the technical system. It is necessary to account for these distortions when assessing the efficiency of the diagnostics of mechanical components. This fact is especially significant when assessing the efficiency of diagnosing the condition of rolling bearings in real time.
 It is proved that the efficiency of diagnosing the machine parts largely depends on the dynamics of the information-and-measuring system and the prompt detection of changes in the operation of this system. As a result of the experiments, the primary task to diagnose the machine parts was solved on the example of faulty rolling bearings, and the influence of the computing power of the equipment used, the sensitivity of the acceleration sensors, and practical feasibility of the network algorithms for automatic data processing in real time were assessed.