Fault Diagnosis of Automobile ECUs with Data Mining Technologies

Abstract:

With the rapid development of automotive ECUs(Electronic Control Unit), the fault diagnosis becomes increasingly complicated. And the link between fault and symptom becomes less obvious. In order to improve the maintenance quality and efficiency, the paper proposes a fault diagnosis approach based on data mining technologies. By making full use of data stream, we firstly extract fault symptom vectors by processing data stream, and then establish a diagnosis decision tree through the ID3 decision tree algorithm, and finally store the link rules between faults and the related symptoms into historical fault database as a foundation for the fault diagnosis. The database provides the basis of trend judgments for a future fault. To verify this approach, an example of diagnosing faults of entertainment ECU is showed. The test result testifies the reliability and validity of this diagnostic method and reduces the cost of ECU diagnosis.

• Periodical: Applied Mechanics and Materials (Volumes 40-41)
• Pages: 156-161
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.40-41.156
• Online since: November 2010
• Authors: Yang Li, Yan Qiang Li, Zhi Xue Wang
• Keywords: Data Mining (DM), Data Preprocessing, Decision Tree, Failure Predication, Fault Diagnosis

References

[1] Ya Di. Data Analysis in Automotive Fault Diagnosis. Beijing: Machinery Industry Press, (2009), pp.2-4.

[2] Han Jiawei，Kambr M. Data Mining: Concepts and Techniques. Bei jing：Higher Education Press(2001), p: 286-316.

[3] Yang jing, Zhang Shaobing. Application of data mining on optimization and fault diagnosis of machine. Mining machiney(2005), p: 146-147.

[4] Benno Stein. Model Compilation and Diagnosability of Technical Systems. In M.H. Hanza, editor, Proceeding of the 3rd IASTED International Conference on Artificial Intelligence and Application(AIA 03), BenalmAqdena, Spain, p: 191-197, ACTA Press(2003).

[5] TAO Yong-jian, DONG De-cun. An improved method for system fault diagnosis using fault tree analysis. Journal of Harbin institute of technology(2010), p: 144-147.

[6] LI Shiqi , DUAN Xueyan , LIU Yan. Application of incremental learning algorithm of decision tree to fault diagnosis. Journal of Huazhong University of Science and Technology: Nature Science (2006), p: 79-81.

[7] L. Breiman J.H. Friedman R.A. Olshen，and C.J. Stone. Classification and Regression Trees. Wadsworth(1984).

[8] Angeli Chr, Chatzinikolaou A. Fault prediction and compensation functions in a diagnostic knowledge-based system for hydraulic systems. Journal of Intelligent and Robotic Systems(1999), p: 153-165.