Torque signature signals during the insertion of self-tapping screws insertions can be used to monitor the process. Although it has been shown that artificial neural networks provide an effective means of monitoring screw fastenings, the research to date provides only a binary successful/unsuccessful type of classification. In practice, when a fault occurs, it is useful to know the cause of failure. In this paper a radial basis artificial neural network is used to classify insertion signals, differentiating successful insertions from failed insertions and categorising different types of insertion failures caused by events such as jamming and cross-threading. A normalised representation of the insertion signal is used as the input to the network. It is shown that the approach is a reliable and robust tool for monitoring and failure classification of the screw fastening process. The proposed strategy is experimentally validated and some test results are presented.
展开▼