LIU Bao-jie, YANG Qing-wen, WU Xiang
（Fifth Department, Army Officer Academy of PLA, Hefei 230031, China）
Abstract： According to fault type diversity and fault information uncertainty problem of the hydraulic driven rocket launcher servo system(HDRLSS), the fault diagnosis method based on the evidence theory and neural network ensemble is proposed. In order to overcome the shortcomings of the single neural network, two improved neural network models are set up at the common nodes to simplify the network structure. The initial fault diagnosis is based on the iron spectrum data and the pressure, flow and temperature(PFT) characteristic parameters as the input vectors of the two improved neural network models, and the diagnosis result is taken as the basic probability distribution of the evidence theory. Then the objectivity of assignment is realized. The initial diagnosis results of two improved neural networks are fused by D-S evidence theory. The experimental results show that this method can avoid the misdiagnosis of neural network recognition and improve the accuracy of the fault diagnosis of HDRLSS.
Key words： multi sensor information fusion; fault diagnosis; D-S evidence theory; BP neural network
CLD number： TP181 Document code： A
Article ID： 1674-8042（2016）04-0368-07 doi： 10.3969/j.issn.1674-8042-2016-04-010
References
［1］ZHOU Wen-chi. The research of servo hydraulic control system of the rocket follow up device. Wuhan: Wuhan University of Science and Technology, 2014.
［2］SHEN Mei-jie，ZHAO Long-zhang，ZHOU Bing, et al. Hydraulic pump fault diagnosis of ｒbf network based on pso optimization. Chinese Hydraulics & Pneumatics, 2016, (5): 87-92.
［3］LIU Xiao-ping, GAO Qiang. Fault diagnosis based on bp neural network for hydraulic control system of rotary dump. Chinese Hydraulics & Pneumatics, 2016, (8): 68-73.
［4］CAO Feng-cai, YUE Feng-ying. Fault diagnosis of hydraulic system based on bp neural network. Journal Of North University Of China(Natural Science Edition), 2010, (6): 596-599.
［5］SONG Xing, YANG Yan-qing, JIN Zhen-zhen. Research on the monitoring and fault diagnosis system for hydraulic servo system of Miebach welding machine based on the fuzzy fault diagnosis algorithm. Electric Welding Machine, 2014, (9): 68-71.
［6］YOU Zhang-ping, JIANG Jie, HU Xiao-ping, et al.Fault diagnosis methods for crane hydraulic system based on particle swarm neural network. Chinese Hydraulics & Pneumatics, 2014, (1): 114-118.
［7］SONG Tao，SHU Tao，LEI Rong-qiang, et al.Research on fault diagnosis for launch system based on ant colony neural network.Fire Control & Command Control, 2015, (9): 143-146, 151.
［8］Jousselme A L, Grenier D, Bosse E. A new distance between two bodies of evidence. Information Fusion, 2001, 2(2): 91-101.
［9］ZHANG Xu-jing, DONG Zeng-shou. Modified D-S evidential theory in hydraulic system fault diagnosis. Journal of Taiyuan University of Science and Technology, 2011, 32(5): 370-373.
［10］LIU Xi-liang, CHEN Gui-ming, LI Fang-xi, et al. Multi-sensor information fusion fault diagnosis based on improved evidence theory. China Mechanical Engineering, 2014, 25(10): 1341-1345.
［11］ZHANG Xu-jing, DONG Zeng-shou. Modified D-S evidential theory in hydraulic system fault diagnosis. Procedia Environmental Sciences, 2011, 32(5): 370-373.
神经网络与D-S证据理论融合的液压系统故障诊断方法
刘保杰， 杨清文， 吴翔
（陆军军官学院 五系， 安徽 合肥 230031）
摘要:针对液压驱动火箭炮随动系统故障类型的多样性以及故障信息不确定性等问题， 提出了证据理论与神经网络综合集成的故障诊断方法。 为克服单一神经网络自身的缺点， 在普通节点处建立2个改进神经网络模型来简化网络结构， 分别以铁谱数据和压力、 流量、 温度特征参数作为输入向量进行初始故障诊断， 并将诊断结果作为证据理论的基本概率分配， 从而实现了赋值的客观化。 然后， 利用D-S证据理论对2个改进神经网络的初始诊断结果进行融合。 实验结果表明： 该方法避免了神经网络识别时的误诊， 提高了液压驱动的火箭炮随动系统故障诊断的准确性。
关键词:多传感器信息融合； 故障诊断； D-S证据理论； BP神经网络
引用格式：LIU Bao-jie, YANG Qing-wen, WU Xiang. Fault diagnosis method of hydraulic system based on fusion of neural network and D-S evidence theory. Journal of Measurement Science and Instrumentation, 2016, 7(4)： 368-374. ［doi： 10.3969／j.issn.1674-8042.2016-04-010］
[full text view]
