Power Electronics Converter's Health Management using Machine Learning

使用机器学习的电力电子转换器的健康管理

• 批准号: 2802267
• 金额: --
• 依托单位: University of Nottingham
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2023
• 资助国家: 英国
• 起止时间: 2023 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2802267
• 关键词: Power; Electronics; Converter; Health; Management

This project will propose new techniques for assessing degradation of power electronics converters especially considering industrial drive applications. The main aim of this project is to develop health management techniques for industrial drives considering the use of wide bandgap power semiconductor devices. Predictive techniques such as machine learning will be utilised for condition monitoring and prognostics of semiconductor devices and DC-link capacitors in industrial motor drives. The target of this research is to ensure better reliability in industrial motor drives without compromising efficiency, cost, and operational limits. Life consumption prediction could enable intelligent performance derating to protect the hardware and reduce chances of a breakdown and plant shutdown.

该项目将提出用于评估电力电子转换器降级的新技术，尤其是考虑工业驱动应用程序。该项目的主要目的是考虑使用广泛的带隙功率半导体设备来开发用于工业驱动器的健康管理技术。预测技术（例如机器学习）将用于工业运动驱动器中半导体设备和直流链接电容器的状况监测和预后。这项研究的目标是确保在不损害效率，成本和运营限制的情况下，在工业电机驱动器上的可靠性更好。生命消耗预测可以使智能绩效降低，以保护硬件并减少崩溃和工厂关闭的机会。