| Improved K-means Algorithm for Manufacturing Process Anomaly Detection and Recognition |
| |
| 引用本文： | ZHOU Xiaomin~(1,2) PENG Wei~1 SHI Haibo~1 (1.Shenyang Institution of Automation Chinese Academy of Sciences,Shenyang 110016,China,2.Graduate School,Chinese Academy of Sciences,Beijing 100039,China). Improved K-means Algorithm for Manufacturing Process Anomaly Detection and Recognition[J]. 武汉理工大学学报, 2006, 0(Z3) |
| |
| 作者姓名： | ZHOU Xiaomin~(1 2) PENG Wei~1 SHI Haibo~1 (1.Shenyang Institution of Automation Chinese Academy of Sciences Shenyang 110016 China 2.Graduate School Chinese Academy of Sciences Beijing 100039 China) |
| |
| 作者单位： | ZHOU Xiaomin~(1，2) PENG Wei~1 SHI Haibo~1 (1.Shenyang Institution of Automation Chinese Academy of Sciences，Shenyang 110016，China，2.Graduate School，Chinese Academy of Sciences，Beijing 100039，China) |
| |
| 摘 要： | Anomaly detection and recognition are of prime importance in process industries.Faults are usually rare,and, therefore,predicting them is difficult.In this paper,a new greedy initialization method for the K-means algorithm is proposed to improve traditional K-means clustering techniques.The new initialization method tries to choose suitable initial points,which are well separated and have the potential to form high-quality clusters.Based on the clustering result of historical disqualification product data in manufacturing process which generated by the Improved-K-means algorithm,a prediction model which is used to detect and recognize the abnormal trend of the quality problems is constructed.This simple and robust alarm-system architecture for predicting incoming faults realizes the transition of quality problems from diagnosis afterward to prevention beforehand indeed.In the end,the alarm model was applied for prediction and avoidance of gear-wheel assembly faults at a gear-plant.
|
Improved K-means Algorithm for Manufacturing Process Anomaly Detection and Recognition |
| |
| ZHOU Xiaomin. Improved K-means Algorithm for Manufacturing Process Anomaly Detection and Recognition[J]. Journal of Wuhan University of Technology, 2006, 0(Z3) |
| |
| Authors: | ZHOU Xiaomin |
| |
| Affiliation: | ZHOU Xiaomin~ |
| |
| Abstract: | Anomaly detection and recognition are of prime importance in process industries.Faults are usually rare,and, therefore,predicting them is difficult.In this paper,a new greedy initialization method for the K-means algorithm is proposed to improve traditional K-means clustering techniques.The new initialization method tries to choose suitable initial points,which are well separated and have the potential to form high-quality clusters.Based on the clustering result of historical disqualification product data in manufacturing process which generated by the Improved-K-means algorithm,a prediction model which is used to detect and recognize the abnormal trend of the quality problems is constructed.This simple and robust alarm-system architecture for predicting incoming faults realizes the transition of quality problems from diagnosis afterward to prevention beforehand indeed.In the end,the alarm model was applied for prediction and avoidance of gear-wheel assembly faults at a gear-plant. |
| |
| Keywords: | data mining clustering quality management anomaly detection and recognition |
| 本文献已被 CNKI 等数据库收录！ |
|
