| 摘要: |
| 提出了基于径向基函数神经网络（ＲＢＦＮ）的火箭发动机动态过程建模，仿真计算表明：采用ＲＢＦＮ建模，可以达到很好的逼近精度，而且网络训练速度大大加快，可以更好地适应实时状态监控和故障诊断，有实际的工程运用价值。 |
| 关键词: 液体推进剂火箭发动机 人工神经元网络 动态模型 故障诊断 故障仿真 |
| DOI： |
| 分类号:V434.1 |
| 基金项目:哈工大校管航天基金 |
|
| ROCKET ENGINE DYNAMIC PROCESS MODELING BASED ON RBFN |
|
Wang Jianbo, Yu Daren
|
|
Coll.of Energy Sources and Engineering,Harbin Inst.of Technology,Harbin,150001
|
| Abstract: |
| Rocket engine(RE)dynamic process was modeled by radial basis function networks(RBFN).Using RBFN for modeling can gain a good precision and make the learning speed of networks rapid.So the method presented will be beter to adapt to the state online detection and fault diagnosis.It is practical in the actual engineering |
| Key words: Liquid propellant rocket engine Artificial neural network Dynamic model Fault diagnosis Fault simulation |
