厌氧塘污水处理系统液压故障在线检测仿真
|
摘要
对液压故障进行在线检测,是解决诊断厌氧塘污水处理系统实际应用的有效途径。当前液压故障检测方法是首先将最严重的故障划分出来,然后采用分层聚类算法对液压故障状态进行检测,通过以上步骤,提高了污水处理系统的检测效率,但检测不出污水处理系统液压设备小幅度突变故障和早期缓变故障,针对上述问题,提出基于动态GRNN模型的厌氧塘污水处理系统液压故障在线检测方法。通过传感器设备将厌氧塘污水处理系统关键部位的信号进行采集,获取系统正常运行状态下的数据,根据这些数据训练神经网络故障观测器模型,用训练好的故障观测器模型来获取厌氧塘污水处理系统残差。引入自适应阈值,通过判断残差平方和与相应阈值对比,可判定故障。实验结果表明,所提方法能够快速、有效地检测出污水处理系统液压设备的小幅度突变故障和早期缓变故障,且具有较好的鲁棒性。
The online detection for hydraulic fault is an effective way to solve the actual application of anaerobic pond sewage disposal system. This paper presents an online detection method for hydraulic fault in anaerobic pond sewage treatment system based on dynamic GRNN model. Firstly,the signals from key parts of anaerobic pond sewage treatment system were collected by the sensor,and then the data under the normal operating state were obtained.Based on these data,the neural network fault observer model was trained,and then the trained fault observer model was used to obtain the residual error of anaerobic pond sewage treatment system. Moreover,the adaptive threshold was introduced. Finally,the failure can be determined by comparing the residual sum of squares with the corresponding threshold. Simulation results prove that the proposed method can quickly and effectively detect the small-amplitude hit fault and early slow-variation fault from hydraulic equipment of sewage treatment system,which has good robustness.
The online detection for hydraulic fault is an effective way to solve the actual application of anaerobic pond sewage disposal system. This paper presents an online detection method for hydraulic fault in anaerobic pond sewage treatment system based on dynamic GRNN model. Firstly,the signals from key parts of anaerobic pond sewage treatment system were collected by the sensor,and then the data under the normal operating state were obtained.Based on these data,the neural network fault observer model was trained,and then the trained fault observer model was used to obtain the residual error of anaerobic pond sewage treatment system. Moreover,the adaptive threshold was introduced. Finally,the failure can be determined by comparing the residual sum of squares with the corresponding threshold. Simulation results prove that the proposed method can quickly and effectively detect the small-amplitude hit fault and early slow-variation fault from hydraulic equipment of sewage treatment system,which has good robustness.
引文
[1]张卫亮,等.故障树和在线监测在高空作业车液压系统故障诊断的应用[J].机床与液压,2016,44(7):168-172.
[2]刘志海,等.无轨胶轮车全液压制动系统故障诊断研究[J].工矿自动化,2016,42(10):30-34.
[3]付一丁,等.基于Volterra级数的水轮机调速器液压系统故障诊断[J].水力发电,2016,42(2):64-69.
[4]张锋.污水源热泵系统设计.[J].科技通报,2017,2(33):205-208.
[5]乔文刚,赵静思,王磊.轧机活套液压控制系统状态监测与故障诊断专家系统研究[J].锻压技术,2017,42(6):122-125.
[6]刘志强,等.基于多特征融合多核学习支持向量机的液压泵故障识别方法[J].中国机械工程,2016,27(24):3355-3361.
[7]李洪儒,等.基于MMSE和ABCSVM的液压泵故障模式识别[J].振动与冲击,2016,35(9):152-158.
[8]经哲,郭利.随机共振液压泵故障特征信号提取[J].中国测试,2016,42(5):107-112.
[9]周博,等.基于GRNN观测器的液压作动器系统自适应故障检测[J].南京理工大学学报,2016,40(2):149-155.
[10]王印松兵,等.基于改进模糊聚类的控制系统故障检测[J].信息与控制,2017,46(1):41-45.
[2]刘志海,等.无轨胶轮车全液压制动系统故障诊断研究[J].工矿自动化,2016,42(10):30-34.
[3]付一丁,等.基于Volterra级数的水轮机调速器液压系统故障诊断[J].水力发电,2016,42(2):64-69.
[4]张锋.污水源热泵系统设计.[J].科技通报,2017,2(33):205-208.
[5]乔文刚,赵静思,王磊.轧机活套液压控制系统状态监测与故障诊断专家系统研究[J].锻压技术,2017,42(6):122-125.
[6]刘志强,等.基于多特征融合多核学习支持向量机的液压泵故障识别方法[J].中国机械工程,2016,27(24):3355-3361.
[7]李洪儒,等.基于MMSE和ABCSVM的液压泵故障模式识别[J].振动与冲击,2016,35(9):152-158.
[8]经哲,郭利.随机共振液压泵故障特征信号提取[J].中国测试,2016,42(5):107-112.
[9]周博,等.基于GRNN观测器的液压作动器系统自适应故障检测[J].南京理工大学学报,2016,40(2):149-155.
[10]王印松兵,等.基于改进模糊聚类的控制系统故障检测[J].信息与控制,2017,46(1):41-45.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |