Automated failure classification for assembly with self-tapping threaded fastenings using artificial neural networks

Abstract

This paper presents a new strategy for the automated monitoring and classification of self-tapping threaded fastenings, based on artificial neural networks. Threaded fastenings represent one of the most common assembly methods making the automation of this task highly desirable. It has been shown that the torque versus insertion depth signature signals measured on-line can be used for monitoring threaded insertions. However, 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 causes leading to it. Extending earlier work by the authors, a radial basis neural network is used to classify insertion signals, differentiating successful insertions from failed insertions and categorizing different types of insertion failures. The neural network is first tested using a computer simulation study based on a mathematical model of the process. The network is then validated using experimental torque signature signals obtained from an electric screwdriver equipped with an optical shaft encoder and a rotary torque sensor. Test results are presented proving that this novel approach allows failure detection and classification in a reliable and robust way. The key advantages of the proposed method, when compared to existing methods, are improved and automated set-up procedures and its generalization capabilities in the presence of noise and component discrepancies due to tolerances.

Failure detection in assembly: Force signature analysis

“…Support Vector Machines have been used by Cho et al [13] and Hsueh and Yang [14] to detect tool breakage in milling operations based on the force signature of the process. Tax et al [15] have proposed a closely related Support Vector method to analyze machine vibration and Althoefer et al [16] have used a neural network to monitor the insertion of self-tapping threaded fasteners using torque signals.…”

A Neural Network Clamping Force Model for Bolt Tightening of Wind Turbine Hubs

“…Various models of bolt tightening have been already reported in the literature: some of them were model-based or model-free controllers [25][26][27], the latter bypassing the need to estimate the parameters of the physical model [28][29][30][31]. For instance, in [25], authors presented the equations of screw insertion torque in function of the screw itself, the hole and the properties of the material; then, a theoretical model was validated by comparing experimental data with predictions of the model, providing basis for computerized monitoring of screw fastenings.…”

Neural networks for prediction of robot failures

“…17 In robotics, this artificial intelligence technique is often applied for control of a mobile robot 18,19 or a robot manipulator. 20,21 For failure problems, the NNs are employed in the assembly tasks, 22 prediction of failure rates of large number of the centrifugal pumps 23 or in the robust scheme for robot manipulators. 24 However, despite various mentioned applications, the robot failure prediction based on the soft computing methods has not been reported in the literature so far.…”