一种对飞行参数选取的自适应方法

飞参数据的典型选取问题是单机寿命监控以及飞行品质分析中压缩数据的储存空间关键。针对飞参数据的特征,提出了一种基于ELM极限学习机的飞参数据选取的模型。利用极限学习机ELM神经网络、文化基因算法MAS优化的方式,克服了算法中存在早熟收敛和局部极小的问题。实现了对飞机不同部位载荷自适应选取不同飞行参数的效果,有效获得评估出飞参数据的重要度。验证结果表明,优化后的ELM-M模型比传统选取飞参模型的精度得到了极大提高,泛化能力增强,说明了方法的可行性、有效性。

A Self Adaption Method of Flight Parameters Selection

A typical flight data selection problem is the key to the single life quality monitoring and analy- sis of flight data compression storage space. For the characteristics of flight data, flight parameters are pro- posed based on Extreme Learning Machine, according to ELM selected models. The model utilizes the Ex- treme Learning Machine ELM neural networks, genetic algorithms culture MAS optimal way to overcome premature convergence and local minima algorithms problems. Through this model, the realization of dif- ferent parts of the aircraft flight parameters to select different load adaptive effect was to assess the impor- tance of the effective flight data. Verification results show ELM- M model optimized precision than tradi- tional parametric model chosen to fly has been greatly improved, the generalization capacity enhancement, to illustrate the feasibility and effectiveness of this method.