期刊文献+

基于B/S结构的多核微处理器实速故障诊断技术

Real speed fault diagnosis technology of multi-core microprocessor based on B/S structure
下载PDF
导出
摘要 针对目前使用的g-超条件诊断度技术、强鲁棒性开路故障诊断策略受到多核微处理器并发不确定性影响,导致诊断效果差的问题,提出了基于B/S结构的多核微处理器实速故障诊断技术。构建基于B/S结构的多核微处理器,并对其故障诊断原理展开详细分析。设置微程序改变目录产生过程,通过监控读写事务的等待时间,判断死锁故障。以单输出响应为例,在输入激励作用下,基于扫描链的故障输入激励与输出响应诊断链路故障。由性能评估结果可知,该技术除了处理管道吞吐项目比标准时间长1 s外,其余所有项目处理时间均在标准时间范围内,具有良好的诊断效果。 In view of the poor diagnosis effect caused by the concurrent uncertainty of multi-core micropr-ocessors,the current g-super conditional diagnosis degree technology and strong robustness open circuit fault diagnosis strategy are affected,a real-speed fault diagnosis technology of multi-core microprocessors based on B/S structure is proposed.A multi-core microprocessor based on B/S structure is constructed,and its fault diagnosis principle is analyzed in detail.Set the microprogram to change the directory generation process,and judge the deadlock fault by monitoring the waiting time of read-write transactions.Taking the single output response as an example,under the action of input excitation,the link fault is diagnosed based on the fault input excitation and output response of scan chain.According to the performance evaluation results,except that the processing time of pipeline throughput projects is 1 s greater than the standard time,the processing time of all other projects is within the standard time range,which has good diagnostic effect.
作者 高吉普 徐长宝 辛明勇 陈军健 刘德宏 GAO Jipu;XU Changbao;XIN Mingyong;CHEN Junjian;LIU Dehong(Electric Power Research Institute of Guizhou Power Grid Co.,Ltd.,Guiyang 550002,China;Digital Grid Research Institute Co.,Ltd.,CSG,Guangzhou 510663,China)
出处 《电子设计工程》 2023年第22期171-175,共5页 Electronic Design Engineering
基金 国家重点研发计划(2020YFB0906000,2020YFB0906001) 贵州电网有限公司科技项目(GZKJXM20200720)。
关键词 B/S结构 多核微处理器 实速故障诊断 死锁故障 动态可配置 B/S structure multi-core microprocessor real speed fault diagnosis deadlock failure dyn-amically configurable
  • 相关文献

参考文献16

二级参考文献137

共引文献224

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部