空中加油系统软管燃油压力测量算法研究

发布时间：2018-05-27 12:34

本文选题：空中加油技术 + 软管燃油压力测量　；参考：《西北大学》2015年硕士论文

【摘要】：随着空中加油技术的迅速发展,空中加油系统的自动化、智能化已成为不可阻挡的发展趋势。加油系统各部分燃油状况的监控是空中加油系统的关键技术之一,它是保证加油过程顺利进行的重要基础。针对某型软式空中加油系统软管燃油压力难以用传感器直接测量的问题,本文提出基于流体力学和基于参数优化SVR的软管燃油压力测量算法,并通过MATLAB仿真实验验证算法的有效性。论文主要包括以下几方面内容：首先,分析空中加油系统内部结构,在理解加油系统软管燃油压力测量原理的基础上,提出软管燃油压力测量的总体研究方案,鉴于实际的试验数据夹杂大量噪声且量纲不同,根据算法需要,酌情选择滤波、剔除、归一化的方法对数据进行预处理,以保证数据的可靠性；其次,提出基于流体力学的软管燃油压力测量算法,结合加油系统的结构参数和燃油参数,建立符合系统的燃油压力测量物理模型,计算软管燃油压力；接着,提出基于参数优化的支持向量回归机(SVR, Support Vector Regression)软管燃油压力测量算法,结合历史加油试验数据建立燃油压力测量数据模型,完成软管燃油压力的计算；最后,通过MATLAB仿真,基于流体力学和基于参数优化SVR的软管燃油压力测量算法的平均相对误差分别为0.9365%、1.8405%,比对两种算法,前一种算法计算精度高,后一种算法计算速度快且建模过程简单,两种算法都满足设计要求,进而提出一种基于神经模糊控制的软管燃油压力控制思想。
[Abstract]:With the rapid development of aerial refueling technology, the automation and intelligence of aerial refueling system has become an irresistible trend. The monitoring of fuel condition in various parts of the refueling system is one of the key technologies of the aerial refueling system, and it is an important foundation to ensure the smooth progress of the refueling process. In order to solve the problem that the fuel pressure of flexible aerial refueling system is difficult to be directly measured by sensors, an algorithm for measuring fuel pressure of hose based on hydrodynamics and parameter optimization SVR is proposed in this paper. The effectiveness of the algorithm is verified by MATLAB simulation. This paper mainly includes the following aspects: firstly, the internal structure of aerial refueling system is analyzed, and the overall research scheme of hose fuel pressure measurement is put forward on the basis of understanding the principle of fuel pressure measurement of hose in refueling system. In view of the large amount of noise and different dimensions of the actual experimental data, the filtering, culling and normalization methods are selected, as appropriate, according to the need of the algorithm to preprocess the data to ensure the reliability of the data. This paper presents an algorithm for measuring fuel pressure of hoses based on hydrodynamics. Combining the structural parameters and fuel parameters of the fuel system, a physical model of fuel pressure measurement in accordance with the system is established to calculate the fuel pressure of the hose. Based on parameter optimization, a support vector regression machine (SVR) and a fuel pressure measurement algorithm for Support Vector Regression) hose are proposed. The model of fuel pressure measurement data is established by combining with the data of historical refueling test to calculate the fuel pressure of the hose. Finally, the calculation of fuel pressure of the hose is accomplished by MATLAB simulation. The average relative error of the fuel pressure measurement algorithm based on hydrodynamics and parameter optimization SVR is 0.9365 and 1.8405, respectively. Compared with the two algorithms, the former algorithm has a high calculation accuracy, the latter has a fast calculation speed and a simple modeling process. The two algorithms meet the design requirements, and a fuel-pressure control idea based on neural fuzzy control is proposed.
【学位授予单位】：西北大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：V228.17

【相似文献】