The Fault Diagnosis of Automotive Airbag Assembly Process Based on Self-Organizing Feature Mapping Network SOM

Abstract

Automotive airbag assembly process is complex and nonlinear, and one of its characteristics is that the accuracy of making the threshold comparison for fault diagnosis using field multi-sensor measured value is not high,. In this article, adopt self-organizing feature mapping network SOM to realize the fault diagnosis of automotive airbag assembly process, constitute the field function of SOM through wavelet functions, form sub-excitatory neuron to update weights, avoid SOM local optimum, so improve the accuracy of fault diagnosis of automotive airbag assembly process.

The AAPF fault-tolerant method for small and complex product assembly

“…Du and Xi 13 developed diagnosis methodology to diagnose fixture faults in assembly processes and explored an improved particle swarm algorithm, based on simulated annealing and a selective neural network algorithm, for on-line identification. Zhang et al 14 adopted a self-organizing feature mapping network (SOM) to realize the fault diagnosis of an automotive airbag assembly process. Cojocaru et al 15 employed an efficient image analysis technique to obtain a real-time mechanism for detection of assembly faults.…”