Research On Detect And Diagnosis Technology Of Phased Array Antenna

Grows increasingly along with the defense military requirements, the application of airborne phased array radar in military is more and more widely, the phased array antenna is the most key part of phased array radar, for its fault diagnosis is very important. Aiming at the problems that the changeable application environment and the complex testing technique of phased array radar antennas,and difficulty to realize the field rapid diagnosis, research on phased array fault diagnosis method based on statistical pattern recognition is conducted. A phased array antenna fault diagnosis system is established. The main work of this paper is as follows:The fault diagnosis method which is composed of linear array scanning and single probe synchronous receiving is proposed. It is free from the constraint of the external environment to the measuring equipment.and it is proved theoretically that the method can distinguish the fault mode.The method of moments is used to verify the feasibility of the method, and the fault database is established, which provides a reliable sample for the following algorithm design.Wavelet modulus maxima algorithm in noise background is used to preprocess and extract feature, reflecting the fault information of the time domain feature vector and wavelet feature vector,and the scatter diagram of fault feature is described and verified that the two groups feature vectors with a strong classifier significantly.Through the projected clustering algorithm to design a classifier combination, first, using the extracted feature vector in time domain to achieve coarse classification, and using k-means algorithm to realize rough classification. Then the extracted Gabor feature vector and the improved SVM are used at a node respectively, to complete the fine classification. Failure mode of hierarchical classification is realized.