| HGA-RBF网络在动力调谐陀螺仪漂移预测中的应用 |
| |
| 引用本文： | 李真真,王宏力,王世方,王新国.HGA-RBF网络在动力调谐陀螺仪漂移预测中的应用[J].战术导弹技术,2009(1):59-62. |
| |
| 作者姓名： | 李真真 王宏力 王世方 王新国 |
| |
| 作者单位： | 第二炮兵工程学院,西安710025 |
| |
| 摘 要： | 由于经典RBF神经网络预测DTG漂移存在参数优化困难和预测精度不理想的问题，将具有可同时优化神经网络参数和隐节点数的递阶遗传算法引入到了对RBF神经网络核函数各项参数的优化中，旨在提高RBF神经网络的建模能力，从而提高DTG漂移的预测精度．首先归一化DTG漂移数据，建立HGA—RBF网络预测模型，然后应用DTG实测漂移数据给出预测模型应用结果，最后利用DTG随机漂移率对预测模型的有效性进行验证．实验结果表明，用这种方法建立的模型能较好地描述动调陀螺仪的漂移特性，其补偿后的漂移量较原始漂移量减小4．5倍．
|
| 关 键 词： | 动力调谐陀螺仪 陀螺仪漂移 递阶遗传算法 RBF神经网络 |
Application of HGA-RBF Neural Network to Drift Prediction of Dynamically Tuned Gyro |
| |
| Li Zhenzhen,Wang Hongli,Wang Shifang,Wang Xinguo.Application of HGA-RBF Neural Network to Drift Prediction of Dynamically Tuned Gyro[J].Tactical Missile Technology,2009(1):59-62. |
| |
| Authors: | Li Zhenzhen Wang Hongli Wang Shifang Wang Xinguo |
| |
| Affiliation: | (The Second Artillery Engineering Institute, Xi＇an 710025, China) |
| |
| Abstract: | The hybrid hierarchy genetic algorithm （HGA） is used to optimize the kernel function parameters of radial basis function neural networks （RBFNN） for overcoming the drawbacks of dynamically tuned gyro （DTG）. The drift data is normalized and the prediction model is established. The application result of prediction model is given out. The model is validated. It is proven in the experiment that the model can be used to describe the drift characteristic of gyroscope and the drift mobility is decreased by 4.5 times after compensated. |
| |
| Keywords: | dynamically tuned gyro gyro drift hierarchical genetic algorithm RBF neural network |
| 本文献已被 维普 等数据库收录！ |
