粗糙集理论与神经网络在变压器故障诊断中的应用

Rough sets theory and artificial neural networks applied in the transformer fault diagnosis

粗糙集(RS)理论是一门新兴的智能信息处理技术,它对各种不完整数据进行分析、推理,发现数据间的关系,提取有用特征,简化信息处理。人工神经元网络(ANN)具有本质的非线性特性、并行处理能力、以及自组织自学习的能力。但单独使用ANN处理问题时,往往会存在一些缺陷。该文把粗糙集理论与人工神经元网络结合起来,应用于变压器故障诊断,可以充分发挥两种方法的优势,取长补短。粗糙集理论可以有效地对样本集进行约简,从而简化了ANN的网络结构,减少了网络的训练步数,提高了判断准确率。并用仿真实验验证了此方法的有效性。

Rough set theory is a new intelligent information process technology. It can analyse and deduce all kinds of incomplete data, find the relationship between the data, pick up the useful characters and reduce the information process. Artificial neural networks has the essential nonlinear character, parallel processing ability, and the ability of self organization and self-learning. But when only using ANN to solve a problem, it often has some shortcomings. This paper combines rough set theory with artificial neural networks, applying it in the transformer fault diagnosis. It can fully develop the two methods＇advantages, learn from other＇s strong points to offset one＇s weakness. Rough set theory can efficiently process the reduction of stylebook collection, so it simplifies the networks＇ structure, reduces the networks＇ training epochs and improves the judgement accuracy. Simulation experiment verifies the validity of this method.