某型飞机油量传感器修正及其流量特征提取方法研究

Research on Modification of Aircraft Fuel Sensor and Extraction of Fuel Flow Characteristic

摘要:

针对某型飞机油量传感器记录时易受外界环境影响而导致所记录信号存在大量噪声且不具有单调性,同时缺少燃油流量参数的记录问题,提出了一种改进的自适应噪声完备经验模态分解(Complete Ensemble Empirical Mode Decomposition with Adaptive Noise, CEEMDAN)方法,在保证单调性的前提下对传感器数据的记录进行了修正,并实现了对流量特征的提取。首先提取飞参记录器中记录的剩余油量信号,然后使用改进的CEEMDAN方法提取固有模态函数和剩余分量并进行重构,最后将处理之后的剩余油量信号进行求一阶导数得到发动机的燃油消耗率。实验表明,使用改进CEEMDAN去噪方法相比于CEEMDAN去噪方法的信噪比提升了48.6%,均方根误差降低了69.9%。

Abstract:

Aiming at the problem that the fuel flow sensor of a certain type of aircraft is susceptible to the external environment when recording, which results in a large amount of noise and non-monotonicity in the recorded signal, and lacks the recording of fuel flow parameters, an improved Complete Ensemble Empirical Empirical Mode Decomposition (CEEMDAN) method is proposed, which corrected the recording of sensor data and realized the extraction of fuel flow characteristic on the premise of monotonicity. Firstly, extracted the residual fuel volume recorded in the Flight Data Recorder (FDR). Secondly, the improved CEEMDAN method was used to extract the intrinsic mode function (IMF) and residual components and reconstructed them. Finally, the fuel consumption rate of the engine can be obtained by calculating the first derivative of the residual fuel volume. Experiments show that the SNR of the improved CEEMDAN method is 48.6% higher than that of the CEEMDAN method, and the RMSE is reduced by 69.9%.