摘要
人工神经网络用于复杂过程中故障诊断和状态监测,具有良好的发展前景。由于模糊ARTFAM(FAM)神经网络分类性能受训练样本序影响,提出一种利用改善贝叶斯信念方法并基于模糊ARTMAP集成的模型(BBM),并描述其在电力发电系统中作为故障检测和诊断的智能学习适用性。围绕检测循环水系统冷凝器传热性能展开试验,判断该模型在故障预测检测和诊断任务中的有效性。试验结果证明,采用BBM模型具有智能故障检测和诊断工具的解释能力,并能监测和诊断发电系统故障的复杂过程。
The artificial neural network used in fault diagnosis and condition monitoring problem in the process of the complex has good prospects.Due to the fuzzy ARTFAM(FAM)neural network classification performance is affected by the training sample sequence,this paper puts forward a method of using improved bayesian belief and model based on fuzzy ARTMAP integration(BBM),and describes its in electric power systems as the applicability of the fault detection and diagnosis of intelligent learning.Experimental process is mainly used for detection of circulating water system in the condenser of the heat transfer performance,a series of experiments to determine the model to predict the effectiveness of the detection and diagnosis of task failure.The test results prove that the BBM model has the ability to interpret intelligent fault detection and diagnostic tools,and can monitor and diagnose the complex process of power system faults.
作者
蒋浩
李林
JIANG Hao;LIN Ling(Business School,University of Shanghai for Science and Technology,Shanghai 200093,China)
出处
《软件导刊》
2019年第2期128-131,共4页
Software Guide
