Manufacture process quality control of interferometric fibre optic gyroscope using analyses of multi-type assembly and test data

摘要

Quality control of manufacture process of a type of interferometric fibre optic gyroscope (IFOG) optical path, which uses computer analyses of multi-type assembly and test data, is proposed. Two quality control algorithms are developed. The first is the Monte Carlo (MC) and Support Vector Machine (SVM) iteration algorithm, i.e. the MC_SVM algorithm. The MC is used to generate random values of the assembly correction data, while the SVM is employed to forecast the assembly quality degree of IFOG. Then, the MC_SVM can tune the multi-type assembly errors. The second algorithm is the Series Feature Analysis (SFA) and Analytic Hierarchy Process (AHP) ranking algorithm, i.e. the SFA_AHP algorithm. The SFA is used to calculate the signal features of multi-type data from an IFOG environment adaptability test experiment. The AHP is employed to analyse the importance ranking of potential faults. Then, the SFA_AHP can detect and locate product faults. The above methods can correct errors before the product assembly and locate faults after the assembly. The experiential knowledge of assembly can be inherited and the multidisciplinary restriction of complex manufacturing can be broken. Their analysis results are easy to be understood by users. Many experiments have demonstrated the effectiveness of proposed algorithms.