摘要
为了减小采用基于SPDA的飞机电脉冲除冰供电对变频交流配电网电源品质的影响,针对电脉冲除冰负载控制问题,提出了BP神经网络PID控制算法。该算法充分利用神经网络的自学习能力和逼近函数的能力及PID控制的鲁棒性,有效地提升了电脉冲除冰负载控制能力。使用仿真软件分析初步验证了该控制方法的有效性,并基于实验平台模拟飞机变频发电机(VFG)对发动机短舱进行除冰供电。实验结果显示,提出的基于BP神经网络PID控制方法是有效的。
In order to reduce the influence of aircraft electro-impulse de-icing power based on SPDA on the power quality of variable-frequency AC distribution network,BP neural network PID control method is proposed for the load control of electro-impulse de-icing.The ability of neural network self-learning and approximation function and PID are fully utilized.The robustness of the control effectively improves the control ability of the electro-impulse de-icing load.The control strategy is preliminarily verified with simulation software.The experimental platform is used to simulate the supply of the aircraft variable frequency generator(VFG)to the nacelle de-icing.Experimental results show that the BP neural network PID control method is valid.
作者
牛俊峰
NIU Junfeng(Shanghai Aircraft Design and Research Institute,Shanghai 201210,China)
出处
《电子器件》
CAS
北大核心
2019年第5期1320-1324,共5页
Chinese Journal of Electron Devices
基金
国家自然科学基金项目(51377161)
关键词
二次配电管理装置
电脉冲除冰负载
变频发电机
神经网络
secondary power distribution assembly
electro-impulse de-icing loads
variable frequency generator
neural network