摘要
为提高现场检修的效率,将多种常用的变压器油气试验仪器集成到一辆汽车上,构建移动试验平台。基于油气试验车上配置的仪器所测得的变压器油气参数和电气参数,提出了BP神经网络变压器故障诊断模型,并采用GA算法对BP神经网络模型进行了优化。利用训练样本数据集对所提出的模型进行了训练,结果表明系统的收敛性好、误差满足实际检测要求。利用测试样本数据集对所提出的模型进行了测试验证,诊断的准确率达到98%,表明GA优化BP神经网络模型可以发挥油气试验车的特长,能够满足变压器油气试验车的要求。现场应用结果表明,采用本文所提出的故障诊断方法,可以快速、准确地定位故障点。
The hardware platform of the transformer oil and gas testing vehicle is constructed, with integrating various commonly used oil and gas test instruments, which can improve the efficiency of on-site maintenance and reduce the time of overhaul. The BP neural network model is presented based on the oil and gas parameters measured by the equipment of the oil and gas test vehicle, and it is optimized by GA algorithm. The proposed model is trained by the training sample data set, and the results show that the system converges well and the error meets the actual testing requirements. Using the diagnostic sample data set to test the proposed model, the accuracy can reach 98%, which shows that BP neural network model optimized by GA can meet the requirements of transformer oil and gas testing vehicle. The on-site application results show that the fault diagnosis method can give play to the strength of the oil and gas testing vehicle and be more accurately identified by using the fault diagnosis method proposed in this paper.
出处
《科技创新导报》
2018年第20期63-68,70,共7页
Science and Technology Innovation Herald
关键词
变压器
油气试验车
BP神经网络
GA算法
故障诊断
Transformer
Oil and gas testing vehicle
BP neural network
GA algorithm
Fault diagnosis
