Title :
Fault diagnosis of general turret electrically controlled system based on fuzzy neutral network optimized by genetic algorithm
Author :
Zhongting, Su ; Wei, Wang ; Chunlin, Zhang ; Wenhe, Jiang
Author_Institution :
Acad. of Armored Force Eng., Beijing, China
Abstract :
After the fault character parameter of electrically control system was fuzzy processed as the learning swatch of neural network, the paper search their network weights and threshold as the initial weights and initial threshold using 3-layer fuzzy neutral network with the training function of the gradient descent algorithm taking momentum optimized by genetic algorithm, classify the fault pattern of electrically control system using fuzzy algorithm, reform the astringency of diagnosis model. The diagnosis result is exact and credible. The network model optimized by genetic algorithm has preferable robustness and fault tolerant ability.
Keywords :
fault diagnosis; fuzzy neural nets; genetic algorithms; learning (artificial intelligence); military equipment; neurocontrollers; weapons; fault character parameter; fault diagnosis; fuzzy neutral network; general turret electrically controlled system; genetic algorithm; gradient descent algorithm; military application; neural network learning swatch; weapon stabilizer; Artificial neural networks; Circuit faults; Control systems; Fault diagnosis; Fuzzy neural networks; Genetic algorithms; Training; Electrically Controlled System; Fault Diagnosis; Genetic Algorithm; Neutral Network;
Conference_Titel :
Electronic Measurement & Instruments (ICEMI), 2011 10th International Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-8158-3
DOI :
10.1109/ICEMI.2011.6037995
