Application of Machine Learning to Engine Air System Failure Prediction

Author:

Li Yong1,Hui Hongjun1,Shuai Shijin2,Qin Yu1,Shao Peng3,Liu Zemin2

Affiliation:

1. Cummins Inc.

2. Tsinghua University

3. Cummins China Investment Co Ltd

Abstract

<div class="section abstract"><div class="htmlview paragraph">With the capability of avoiding failure in advance, failure prediction model is important not only to end users, but also to the service engineers in vehicle industry. This paper proposes an approach based on anomaly detection algorithms and telematic data to predict the failure of the engine air system with Turbo charger. Firstly, the relationship between air system and all obtained features are analyzed by both physical mechanism and data-wise. Then, the features including altitude, air temperature, engine output power, and charger pressure are selected as the input of the model, with the sampling interval of 1 minute. Based on the selected features, the healthy state for each vehicle is defined by the model as benchmark. Finally, the ‘Medium surface’ is determined for specific vehicle, which is a hyperplane with the medium points of the healthy state located at, to detect the minor weakness symptom (sub-health state). The precisions of our model are 89.1% and 92.9% on training dataset and field test dataset, respectively. The average time gap between prediction and failure is 45 days, which means the model can help our end users and service engineers take actions to avoid sever failure 45 days in advance. There are also two applied cases of the model to real vehicles. In both cases, the model successfully detects the sub-health state and helps service engineers repair the air system in advance. After the repair, the state indicator was recovered to the normal level, which can be a validation of our model. The algorithm could be generalized to failure of other vehicle parts with similar conditions and save money and time for our customers.</div></div>

Publisher

SAE International

Reference22 articles.

1. Czop , P. , Kost , G. , Sławik , D. , and Wszołek , G. Formulation and Identification of First-Principle Data-Driven Models Journal of Achievements in materials and manufacturing Engineering 44 2 2011 179 186

2. Liu , Z.E. , Li , Y. , Tan , G. , Xu , L. et al. Synthesis of Representative Driving Cycle for Heavy Duty Vehicle Based on Markov Chain and Big Data Considering Weight Variation SAE Technical Paper 2023-32-0177 2023

3. Liu , Z.E. , Xu , L. , Li , Y. , Shuai , B. et al. Deep Reinforcement Learning Based Energy Management Strategy for Hybrid Vehicles in Consideration of Engine Start-Up Conditions Proceedings of the International Conference on Applied Energy, ICAE 9 2021

4. Prytz , R. 2014

5. Voronov , S. , Frisk , E. , and Krysander , M. Data-Driven Battery Lifetime Prediction and Confidence Estimation for Heavy-Duty Trucks IEEE Transactions on Reliability 67 2 2018 623 639

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3