摘要
贝叶斯网络处理不确定性问题的能力可以很好地解决变压器故障诊断中因数据不完整而难以得到可靠结论的问题。为此,将贝叶斯网络分类器和粗糙集约简理论相结合,基于专家知识及统计数据建立贝叶斯网络分类模型,并综合运用色谱数据及电气试验数据作为变压器故障诊断的属性集输入,实现概率推理及对可能故障类型的排序,提高诊断结论的可靠性。此外,利用粗糙集约简理论对贝叶斯网络分类模型进行最小约简,降低网络结构的复杂性,减小模型所依赖的输入量,以更切合实际诊断情况。实验证明,该方法具有处理信息缺失的能力及容错特性,准确率较高,是一种变压器故障诊断的有效方法。
The ahility of Bayesian network to deal with uncertain problems can be a solution to transformer fauh diagnosis when data is incomplete and reliable c.onclusions are difficult to be reached. By eomhining the Bayesian network elassifier and rough set reduction theory, this paper set up a Bayesian network classification model based on expert knowledge and statistical data. The integral use of DGA and electrical test data as the input set of diagnosis realized the probahilistic reasoning and sequencing of the possible failure types, and as a result, improved the reliahility of the diagnosis. Meanwhile the rough set reductiou theory was used to make the minimization reduction of Bayesian network classification model, effectively reduced the complexity of network structure and the input of the model. Experiments proved that this method is an effective transformer fauh-diagnosis method and has the ability to deal with the absence of information, the fauh-tolerant features and high accuracy.
出处
《中国电力》
CSCD
北大核心
2013年第9期75-79,共5页
Electric Power
基金
中国南方电网有限责任公司科技项目(K-SZ2012-070)
关键词
变压器
故障诊断
决策表
贝叶斯网络
粗糙集
知识约简
transformer
failure diagnosis
decision table
Bayesian network： rough sets： infommtion reduction
