DeepEngine: an expert system for prediction of vehicle engine breakdown

DeepEngine：汽车发动机故障预测专家系统

• 批准号: 556148-2020
• 负责人: Armanfard, Narges
• 金额: $ 1.51万
• 依托单位: McGill University
• 依托单位国家: 加拿大
• 项目类别: Alliance Grants
• 财政年份: 2020
• 资助国家: 加拿大
• 起止时间: 2020-01-01 至 2021-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=706153
• 关键词: DeepEngine; expert; system; prediction; vehicle

As the second-largest country on earth, with a population stretched from coast to coast to coast, Canada faces unique transportation challenges. As a result, the transportation industry remains a significant force in the Canadian economy. Therefore, traditional fleet management systems are gradually extended towards digitization, with new features to improve reliability, such as better maintenance planning. This research project aims to develop an Artificial Intelligent based framework that objectively and automatically analyze information recorded by the sensors embedded throughout a truck and predict whether an engine break down is going to happen in the next few weeks, well before it causes downtime. The proposed expert system will significantly impact on vehicle maintenance strategies, increase vehicle up-time, improve technician efficiency and reduce maintenance cost. In addition, the project will significantly contribute to the advancement of knowledge in machine learning research for analyzing large scale time-series data.

作为地球上第二大国家，人口从东海岸延伸到西海岸，加拿大面临着独特的交通挑战。因此，运输业仍然是加拿大经济的重要力量。因此，传统的车队管理系统正逐步向数字化方向延伸，增加新功能以提高可靠性，例如更好的维护计划。该研究项目旨在开发一个基于人工智能的框架，该框架可以客观和自动地分析嵌入卡车的传感器记录的信息，并预测发动机是否会在未来几周内发生故障。建议的专家系统将显着影响车辆维修策略，增加车辆的正常运行时间，提高技术人员的工作效率，降低维修成本。此外，该项目还将为机器学习研究的知识进步做出重大贡献，以分析大规模时间序列数据。