基于樽海鞘群极限学习机的进/发一体化性能寻优控制模型研究

为充分发挥航空推进系统的性能,提高性能寻优控制的实时性,将樽海鞘群算法(SSA)与极限学习机(ELM)相结合,基于进/发一体化部件级模型建立数据集,提出一种基于SSA-ELM的数据驱动模型。将该建模方法与广义回归神经网络(GRNN)、BP神经网络(BPNN)和极限学习机(ELM)比较,结果表明,相比于BPNN,ELM,GRNN,SSA-ELM用于预测可以使安装推力的均方根误差(RMSE)分别降低7.41%,17.01%,72.57%,安装油耗的RMSE分别降低4.32%,19.41%,66.77%,具有更高的预测精度。将基于SSA-ELM的数据驱动模型作为机载模型应用到性能寻优控制,结果表明,该机载模型能够维持理想的寻优效果。针对最大安装推力模式开展实时性分析,该机载模型相比于进/发一体化部件级模型,平均计算时间由184.05 ms缩短至1.357 ms,实时性得到显著改善,大大提高了寻优效率。

基于分裂激盘的进气道/风扇联合流场计算

为对进气道和风扇联合流场一体化进行计算,在常规激盘模型理论的基础上建立了改进的分裂激盘模型,并通过数值模拟方法与原模型进行对比,验证了分裂激盘模型相对常规激盘模型的准确性。通过使用分裂激盘模型对压气机部分模化,分析了包含分裂激盘模型的进气道风扇联合流场数值模拟问题。研究表明:分裂激盘模型可以有效地消除激盘模型假设产生的误差,数值模拟结果更接近全3维模拟结果,为进/发一体化联合流场的数值模拟提供1种可行的准确方法。

The Protection and Suggestion of the East Asian Maritime Cultural Routes Heritage

It is of great practical significance to study and protect the East Asian Maritime Cultural Routes Heritage. This is conducive to a profound understanding of the spirit of the East Asian region, especially traditional Chinese marine culture, and the maintenance of peace and order in East Asia and the world as a whole. But at present, some problems increase the difficulty for the protection of East Asian Maritime Cultural Routes Heritage, for example, improper development and utilization in the process of urbanization, imperfection of the management mechanism, multiplicity of the ownership of the estate. Propose to carry out a survey of the East Asian Maritime Cultural Routes Heritage as soon as possible, build regional integration cooperation mechanism, improve the marine cultural heritage protection technology, perfect the relevant laws and regulations.