摘要
针对传统诊断技术的局限性,研究了基于BP模型神经网络的故障诊断方法,建立了基于BP神经网络的空气源热泵机组的故障诊断模型,并用来自模拟实验的征兆实例和领域专家的知识对神经网络进行了训练.诊断结果表明,对于已学习过的样本知识,网络的输出与希望结果充分相符,基于人工神经网络的空气源热泵冷热水机组的故障诊断是行之有效的.
From the traditional diagnosis technology, a fault diagnosis based on BP model neural network is established for the air source heat pump unit, and then it is trained by typical fault samples from simulation experiment and expert knowledge. It needn't set up a complicated mathematical model and a complicated mathematical calculation and data processing. All you need do is to select enough typical fault samples to train the neural network. The simulation results show that once the neural network training has finished, the output of neutral net is well corresponding to expected results, and the fault diagnosis for air source heat pump unit based on neural network is very effective.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
2002年第6期770-772,共3页
Journal of Harbin Institute of Technology
基金
国家自然科学基金资助项目(50278021)
哈尔滨工业大学科学研究基金资助项目(HIT.2000.26).
