| 蚁群约简与神经网络用于航空发动机故障识别 |
| |
| 引用本文： | 刘柳,;王君艳,;顾星海.蚁群约简与神经网络用于航空发动机故障识别[J].黑龙江电子技术,2014(6):83-86. |
| |
| 作者姓名： | 刘柳 ;王君艳 ;顾星海 |
| |
| 作者单位： | [1]上海交通大学电子信息与电气工程学院,上海200240; [2]中国东方航空股份有限公司,上海200335 |
| |
| 摘 要： | 为解决航空发动机转子系统故障模式识别这一复杂问题,将蚁群算法与BP神经网络相结合应用于故障模式识别.文中采用蚁群算法对反映发动机运行工况的故障特征参数进行约简,并结合BP神经网络对故障识别过程做了分析,以航空发动机转子系统的故障识别为对象进行了实验验证.结果表明,利用蚁群算法对航空发动机转子系统故障特征参数进行约简,剔除了输入冗余信息,降低了网络数据维数,提高了运算效率和故障识别的正确性.
|
| 关 键 词： | 航空发动机 蚁群算法 BP神经网络 故障模式识别 |
Application of ACA parameter and neural network for aero-engine faults recognition |
| |
| Institution: | LIU Liu , WANG Jun-yan, GU Xing-hai ( 1. School of Electronic Information and Electrical Engineering, Shanghai Jiaotong University, Shanghai 200240, China;2. China Eastern Airlines Corp. , Shanghai 200335, China) |
| |
| Abstract: | The combination of ant colony algorithm and BP neural network is applied to solve the fault pattern recognition of aero-engine rotor. The ant colony algorithm is used to simplify attribute parameter reflecting operating conditions of aero-engine and analyzed the process of the fault pattern recognition. Then this paper does experiments on the experimental target of aero-engine rotor. The result showed that this method is capable of eliminating redundant information of fault symptom parameters, and reduces the dimension of input to BP neural network, raises the operating efficiency and improves the fault pattern recognition accuracy. |
| |
| Keywords: | aero-engine ant colony algorithm BP neural network fault pattern recognition |
| 本文献已被 维普 等数据库收录！ |
|
