论文部分内容阅读
利用DNA分子的复制、储存和传递信息的功能,采用基于DNA编码技术的遗传算法优化BP神经网络,建立了具有更高辨识能力的气液两相流流型辨识模型,并就若干参数对辨识模型计算效率的影响规律进行了研究。研究结果表明,基于DNA编码遗传技术对BP神经网络进行优化处理而形成的气液两相流型辨识模型比已有其它辨识模型具有更高的辨识能力;通过轴承腔气液两相流动数值仿真流型模式的辨识分析,表明这一辨识模型可应用于航空发动机轴承腔气液两相流的研究。
Using the function of DNA molecules to copy, store and transmit information, genetic algorithm based on DNA encoding technology is used to optimize BP neural network to establish a flow identification model of gas-liquid two-phase flow with higher identification ability. The impact of model calculation efficiency has been studied. The results show that the identification model of gas-liquid two-phase flow based on genetic algorithm of DNA coding has better identification ability than other identification models. Through numerical simulation of gas-liquid two-phase flow in bearing cavity The identification of flow pattern shows that this identification model can be applied to the research of gas-liquid two-phase flow in aeroengine bearing cavity.