Study on the Drilling Accident Pre-warning System Based on Multi-source Information Fusion
Lai Xi · Liaoning Chemical Industry · 2012
A diagnosis method of drilling accident based on multi-source information fusion technology was proposed.First,a fault symptom sub-space of the drilling process was created and regarded as the corresponding sub neural network input,and then preliminary diagnosis was carried out.So evidences supporting for different fault mode in drilling fault frames of discernment can be obtained.And by using D-S evidence theory,the evidences formed by sub-net output were fused to obtain confidence interval of fault mode,so drilling accident was diagnosed and identified very well.The drilling accident diagnosis combined with information fusion can take full use of redundant or complementary information of sensor to reduce uncertainty of the sensor state and improve precision and accuracy of accident identification.