There are the application scope limits for single differential-mode current injection test method, so in order to carry out injection susceptibility test for two-pieces equipment interconnected with both ends of a cab...There are the application scope limits for single differential-mode current injection test method, so in order to carry out injection susceptibility test for two-pieces equipment interconnected with both ends of a cable simultaneously, a double differential-mode current in- jection test method (DDMCI) is proposed. The method adopted the equivalence source wave theorem and Baum-Liu-Tesche(BLT) equation as its theory foundation. The equivalent corresponding relation between injection voltage and radiation electric field intensity is derived, and the phase relation between the two injection voltage sources is confirmed. The results indicate that the amplitude and phase of the equivalent injection voltage source is closely related to the S parameter of directional coupling device, the transmission line length, and the source vector in BLT equation, but has nothing to do with the reflection coefficient between the two equipment pieces. Therefore, by choosing the right amplitude and phase of the double injection voltage sources, the DDMCI test is equivalent to the radiation test for two interconnected equipment of a system.展开更多
为实现更精确、更实时的智能围岩分级以匹配双护盾TBM高效掘进,基于先进的HRNet高分辨率神经网络,通过分析地质与掘进参数2类指标的强关联性,构建基于该2类指标融合的高分辨率TBM隧道双模态围岩精准智能分级模型,开展实时的围岩分级工...为实现更精确、更实时的智能围岩分级以匹配双护盾TBM高效掘进,基于先进的HRNet高分辨率神经网络,通过分析地质与掘进参数2类指标的强关联性,构建基于该2类指标融合的高分辨率TBM隧道双模态围岩精准智能分级模型,开展实时的围岩分级工作。结果表明:1)推进速度与刀盘转矩2个参数与围岩坚硬度有较强的关联性,通过这2个掘进参数二维组合的线性趋势以及拐点区间可以区分不同的岩石坚硬程度;2)刀盘滚动角和撑靴俯仰角2个参数与岩体完整性有较强的关联性,通过这2个掘进参数的摆动幅度和角度可区分不同的岩体完整性;3)通过将2个模态单一的HRNet高分辨率神经网络进行拼接和全连接,构建的双模态围岩分级模型能够实现离散型数值的地质模态数据与连续型数值的掘进模态数据融合,该模型能够从围岩稳定性与岩机交互等多方面更加精确地反映围岩分级,应用准确率可达95.0%;4)相比仅用掘进指标的单模态模型,双模态模型在测试集上表现效果更好,其Macro F1 Score为0.985。展开更多
基金Project supported by Arm Pre-research Program （51333040101）, National Defense 973 Program （6131380301 ）, National Natural Science Foundation of China （61040003）.
文摘There are the application scope limits for single differential-mode current injection test method, so in order to carry out injection susceptibility test for two-pieces equipment interconnected with both ends of a cable simultaneously, a double differential-mode current in- jection test method (DDMCI) is proposed. The method adopted the equivalence source wave theorem and Baum-Liu-Tesche(BLT) equation as its theory foundation. The equivalent corresponding relation between injection voltage and radiation electric field intensity is derived, and the phase relation between the two injection voltage sources is confirmed. The results indicate that the amplitude and phase of the equivalent injection voltage source is closely related to the S parameter of directional coupling device, the transmission line length, and the source vector in BLT equation, but has nothing to do with the reflection coefficient between the two equipment pieces. Therefore, by choosing the right amplitude and phase of the double injection voltage sources, the DDMCI test is equivalent to the radiation test for two interconnected equipment of a system.
文摘为实现更精确、更实时的智能围岩分级以匹配双护盾TBM高效掘进,基于先进的HRNet高分辨率神经网络,通过分析地质与掘进参数2类指标的强关联性,构建基于该2类指标融合的高分辨率TBM隧道双模态围岩精准智能分级模型,开展实时的围岩分级工作。结果表明:1)推进速度与刀盘转矩2个参数与围岩坚硬度有较强的关联性,通过这2个掘进参数二维组合的线性趋势以及拐点区间可以区分不同的岩石坚硬程度;2)刀盘滚动角和撑靴俯仰角2个参数与岩体完整性有较强的关联性,通过这2个掘进参数的摆动幅度和角度可区分不同的岩体完整性;3)通过将2个模态单一的HRNet高分辨率神经网络进行拼接和全连接,构建的双模态围岩分级模型能够实现离散型数值的地质模态数据与连续型数值的掘进模态数据融合,该模型能够从围岩稳定性与岩机交互等多方面更加精确地反映围岩分级,应用准确率可达95.0%;4)相比仅用掘进指标的单模态模型,双模态模型在测试集上表现效果更好,其Macro F1 Score为0.985。
