摘要
对传统补偿模糊神经网络(CFNN)的算法进行了改进,提出了动态调整学习步长的方法,避免了较大震荡,同时加快了迭代速度,与定学习步长的方法相比,学习速度和误差精度都有大大提高,最后通过仿真实验证明该方法在地下通风空调系统故障诊断中,具有收敛速度快,诊断精度高,并且适应性强等优点。
The paper establishes a kind of improved compensation fuzzy neural networks, which puts forward a method that can dynamically adjust the learning step, so the sway phenomenon can be minimized and the learning step can be quickly speeded, compared with traditional method the convergence speed and the error precision are improved greatly, finally, experiments show that the program is excellence in ventilation and air-conditioning system of underground engineering, realizes the fimction of high study speed and error precision.
出处
《制冷与空调（四川）》
2013年第2期121-125,共5页
Refrigeration and Air Conditioning
关键词
补偿模糊神经网络
算法改进
故障诊断
通风空调系统
compensation fuzzy neural networks
arithmetic improving
fault diagnosis
ventilation and air-conditioning system
