Fault diagnosis of rotating system based on multi-sensor data fusion

This paper deals with a multi-sensor system applied to mechanical fault diagnosing fusion pattern based on neural network, the veracity of diagnosis system can be increased greatly. The methods for data fusion architecture and construction of the neural network are given. Data preprocessing and features extraction have been done by using ICA(independent component analysis) method. Multi-sensor information fusion and controller expert fuzzy neural networks are proposed for data relation construct. Applying the redundancy technology and multi-sensor information fusion technology, the synchronized automated data processing and malfunction warning could be realized based on fusion system. Experiments are given showing the effectiveness of the proposed hierarchical intelligent control system.