摘要
该方案实现了基于嵌入式AI推断电机运行异常的方法。基于瑞萨电机控制MCU RA6T1,结合瑞萨的e-AI工具,将Google的TensorFlow Lite模型部署在MCU端,结合瑞萨专有的BLDC电机控制程序套件实现AI对电机运行状态的判断,可作为工厂自动化中预测性运维的实践基础。
Abstract
This solution implements motor abnormal detection based on AI inference embedded in Renesas MCU RA6T1,along with Renesas e-AI toolkit,which helps to translate TensorFlow Lite from Google to accelerator on MCU.This is the base for factory automation predictive maintenance where motor monitoring is the major task to resolve.
关键词
瑞萨电机MCU /
RA6T1 /
嵌入式AI /
电机运行监测
Key words
Renesas motor control MCU /
RA6T1 /
embedded AI /
motor predictive maintainer
顾晨宇.
基于RA6T1和AI技术的电机故障智能检测[J]. 集成电路与嵌入式系统. 2022, 22(9): 18-21
Gu Chenyu.
Intelligent Detection of Motor Fault Based on RA6T1 and AI Technology[J]. Integrated Circuits and Embedded Systems. 2022, 22(9): 18-21
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 瑞萨电子.RA6T1 Motor Failure Detection Example by TensorFlow Lite for Microcontroller(R01AN5636EJ0100 Rev.1.00), 2020.
[2] 瑞萨电子.Motor Control Evaluation System for RA Family User's Manual (R12UZ0078EJ0120 Rev 1.20),2011.
[3] 瑞萨电子.Renesas RA6T1 Group User's Manual (R01UH0897 EU0110 Rev.1.10),2020.
[4] 瑞萨电子.E-AI translator V2.2.0用户手册(R20UT4135EJ0900 Rev.9.00),2020.
[5] 瑞萨电子.What is Renesas' e-AI Solution?[EB/OL].[2022-06]. https://www2.renesas.cn/jp/en/application/key-technology/artificial-intelligence/about-e-ai.