
基本信息:
- 专利标题: 一种故障识别方法及装置
- 专利标题(英):Fault identification method and device
- 申请号:CN201910360197.2 申请日:2019-04-30
- 公开(公告)号:CN110163263A 公开(公告)日:2019-08-23
- 发明人: 刘秀华 , 汪万根 , 郭亮 , 项秀明 , 李嘉 , 苏海涛 , 吕宝超 , 史冠军 , 张聪
- 申请人: 首钢京唐钢铁联合有限责任公司 , 北京首钢自动化信息技术有限公司
- 申请人地址: 河北省唐山市曹妃甸工业区
- 专利权人: 首钢京唐钢铁联合有限责任公司,北京首钢自动化信息技术有限公司
- 当前专利权人: 首钢京唐钢铁联合有限责任公司,北京首钢自动化信息技术有限公司
- 当前专利权人地址: 河北省唐山市曹妃甸工业区
- 代理机构: 北京华沛德权律师事务所
- 代理人: 马苗苗
- 主分类号: G06K9/62
- IPC分类号: G06K9/62 ; G06N3/04 ; G05B23/02 ; G05B19/042
The embodiment of the invention relates to the technical field of sensor fault diagnosis, in particular to a fault identification method and device. The method comprises the following steps: firstly,determining the number of nodes of an input layer and the number of nodes of an output layer based on obtained experimental data; secondly, calculating the node number of the hidden layer according tothe node number of the input layer and the node number of the output layer; secondly, establishing a multi-layer feedforward neural network according to the number of input layer nodes, the number ofoutput layer nodes and the number of hidden layer nodes, and training the multi-layer feedforward neural network by adopting a training set; and finally, adopting the trained multi-layer feedforwardneural network to carry out fault identification on the prediction set, so that fault diagnosis and identification can be rapidly and accurately carried out on the digital sensor of the electronic platform scale.
IPC结构图谱:
G | 物理 |
--G06 | 计算;推算;计数 |
----G06K | 数据识别;数据表示;记录载体;记录载体的处理 |
------G06K9/00 | 用于阅读或识别印刷或书写字符或者用于识别图形,例如,指纹的方法或装置 |
--------G06K9/62 | .应用电子设备进行识别的方法或装置 |