基于误差预测修正的液体火箭发动机故障预测方法研究

发布时间：2018-03-21 15:11

本文选题：液体火箭发动机　切入点：故障预测　出处：《推进技术》2016年08期 　论文类型：期刊论文

【摘要】：为解决液体火箭发动机故障预测这一难题,提出一种基于误差预测修正的故障预测方法。在历史数据的基础上建立小波过程神经网络故障预测模型,同步计算学习样本的预测误差,根据上述误差建立双并联离散过程神经网络预测模型。预测时,将预测误差值实时补偿到小波过程神经网络预测模型以提高预测精度。通过液体火箭发动机地面试验中的涡轮泵数据对该方法进了验证。结果表明,该方法在预测精度和适应能力上较单一的过程神经网络预测模型有显著提高,进行10步预测时,预测值的标准化均方根误差为0.392,预测平均耗时为76ms,能够用于解决液体火箭发动机故障预测问题。
[Abstract]:In order to solve the problem of liquid rocket engine fault prediction, a fault prediction method based on error prediction correction is proposed. Based on the historical data, the wavelet process neural network fault prediction model is established. The prediction error of learning sample is calculated synchronously. According to the above error, the neural network prediction model of double parallel discrete process is established. The prediction error is compensated to the wavelet process neural network prediction model in real time to improve the prediction accuracy. The method is verified by the turbine pump data in the ground test of liquid rocket engine. The prediction accuracy and adaptability of this method are significantly higher than that of the single process neural network prediction model. The standardized root mean square error (RMS) of the prediction value is 0.392, and the average prediction time is 76ms. it can be used to solve the trouble prediction problem of liquid rocket engine.
【作者单位】：国防科学技术大学航天科学与工程学院;
【基金】：国家自然科学基金(51206181;51506219)
【分类号】：V434

【相似文献】