Real time condition monitoring and fault diagnosis of propulsion motors used in electric vehicles

该研究计划旨在实现电动汽车（EV）中使用的推进电机的实时状态监测和故障诊断。 推进电机是电动车辆的关键部件，无论是地面电动车辆如电动汽车、电动公共汽车等，还是海上电动车辆如电动船、渡轮等，还是空中电动车辆如有人驾驶和无人驾驶飞行器。 电动汽车的推进电机发生任何意外故障都会影响电动汽车的使用者和制造商。对于用户来说，推进电机的意外故障可能小到电动汽车无法工作的不便，也可能大到灾难性事故。对于电动汽车制造商来说，频繁和突然的驱动电机故障会严重损害品牌形象，从而赶走潜在客户。此外，修理和更换费用可能相当高。因此，拥有一种能够实时监控推进电机状况并在早期阶段警告故障的技术至关重要。 在推进电机中可能发生许多不同类型的故障。此外，每个故障以不同的速率传播，并在不同程度上影响推进电机的性能。因此，故障检测系统应该能够分离故障类型并估计每个故障的严重性。此外，推进电机具有基于电动运输系统的额定功率、性能、空间、成本、操作条件等的多种设计。因此，状态监测和故障诊断方案应该能够适应这些设计变化，而不损害可靠性。 拟议的研究计划旨在开发这样一个全面的状态监测和故障诊断方案，以满足电力运输系统的全方位需求。研究将采用基于数学模型的方法 (a)为了研究不同类型和严重程度的内部故障对推进发动机性能的影响， (b)分析推进电机设计参数（如气隙长度、转子类型、开槽和绕组布置等）的变化对故障识别和隔离方案的影响，以及 (c)训练和开发基于人工智能的实时状态监测和故障检测系统。 该研究项目的成果将对加速电动交通系统推向市场产生重大影响。此外，该研究计划将产生高技能的HQP，以满足电力运输部门的需求。最后，该提案将允许在加拿大维多利亚大学开发一个世界级的推进电机状态监测和故障诊断项目，这是目前加拿大其他大学所缺乏的。

The proposed research program aims to accomplish real-time condition monitoring and fault diagnosis of propulsion motors used in electric vehicles (EV). Propulsion motor is a critical component of electric vehicle be it ground electric vehicle such as electric car, electric bus etc. or seaborne electric vehicle such as electric boat, ferry etc. or air borne electric vehicle such as manned and unmanned aerial vehicle. Any unexpected failure of propulsion motor in an EV would affect the users as well as the manufacturers of the EV. For the users, an unexpected failure of propulsion motor could be as small as the inconvenience of EV not working or could be as big as a catastrophic accident. For an EV manufacturer, frequent and abrupt failure of propulsion motors would severely dent one's brand image and hence drive away prospective customers. Furthermore, repair and replacement costs might be quite high. Therefore, it is vital to have a technology that would monitor the condition of the propulsion motor in real-time and warn about faults at an incipient stage. There are many different types of faults that can occur in a propulsion motor. Moreover, each fault propagates at different rates and afflicts the performance of a propulsion motor to different extent. Thus, the fault detection system should be able to segregate the fault type and estimate the severity of each fault. Furthermore, propulsion motor has numerous designs based on power rating, performance, space, cost, operating conditions etc. of an electric transportation system. Consequently, the condition monitoring and fault diagnostic scheme should be able to accommodate these design variations without compromising on reliability. The proposed research program intends to develop such a comprehensive condition monitoring and fault diagnostic scheme that would cater to the full spectrum of electric transportation systems. The research would adopt a mathematical model based approach (a) to study the impact of different types and severities of internal faults on the performance of propulsion motors, (b) to analyze the impact of variations in propulsion motor design parameters like air-gap length, rotor type, slotting and winding arrangement etc. on fault identification and segregation schemes, and (c) to train and develop an artificial intelligence based real-time condition monitoring and fault detection system. The research program's outcomes will have a significant bearing on accelerating electric transportation systems to market. Additionally, the research program will generate highly skilled HQPs to meet the demands of electric transportation sector. Finally, the proposal will allow a world class program in condition monitoring and fault diagnosis of propulsion motors to be developed in Canada at the University of Victoria that are currently lacking in other Canadian universities.