石英的微量元素记录了石英生长的物理化学条件。通过微量元素对石英原岩进行分类的研究历史已久,经典工作是在以微量元素为坐标轴的图解上绘制各类型石英的分布范围,以区分石英类型。经典图解包括Rusk(2012)提出用于区分三种矿床类型石...石英的微量元素记录了石英生长的物理化学条件。通过微量元素对石英原岩进行分类的研究历史已久,经典工作是在以微量元素为坐标轴的图解上绘制各类型石英的分布范围,以区分石英类型。经典图解包括Rusk(2012)提出用于区分三种矿床类型石英的Al-Ti二元图解,和Schr9n et al.(1988)提出的用于判别不同岩浆岩类型石英的Ti-Al-Ge三元图解。越来越多的研究表明,上述图解不能满足对更多石英类型进行分类的需求,同时也出现与部分已知产状类型的石英微量元素判别相矛盾的情况。随着石英原位微区测试方法的成熟,高精度石英微量元素数据逐渐丰富为系统开展机器学习提供了大数据基础,为石英微量元素研究提供了新的角度和可能性。本研究运用机器学习分类方法对石英微量元素进行精确数学分析,提出Ti/Ge-P图解为石英成因研究提出新的地球化学指标。本文同时测试了六种经典机器学习分类算法,提高Ti/Ge-P图解在石英成因分类研究上的精度。此Ti/Ge-P图解适用于多种矿床研究,包括但不局限于斑岩型矿床、矽卡岩型矿床、浅成低温热液型矿床、卡林型矿床以及造山型矿床中的石英。这项工作是大数据技术与机器学习技术在地球化学研究中的积极探索。展开更多
Grid-forming(GFM)converters can provide inertia support for power grids through control technology,stabilize voltage and frequency,and improve system stability,unlike traditional grid-following(GFL)converters.Therefor...Grid-forming(GFM)converters can provide inertia support for power grids through control technology,stabilize voltage and frequency,and improve system stability,unlike traditional grid-following(GFL)converters.Therefore,in future“double high”power systems,research on the control technology of GFM converters will become an urgent demand.In this paper,we first introduce the basic principle of GFM control and then present five currently used control strategies for GFM converters:droop control,power synchronization control(PSC),virtual synchronous machine control(VSM),direct power control(DPC),and virtual oscillator control(VOC).These five strategies can independently establish voltage phasors to provide inertia to the system.Among these,droop control is the most widely used strategy.PSC and VSM are strategies that simulate the mechanical characteristics of synchronous generators;thus,they are more accurate than droop control.DPC regulates the active power and reactive power directly,with no inner current controller,and VOC is a novel method under study using an oscillator circuit to realize synchronization.Finally,we highlight key technologies and research directions to be addressed in the future.展开更多
Fe-based metallic glasses(MGs)have shown great commercial values due to their excellent soft magnetic properties.Magnetism prediction with consideration of glass forming ability(GFA)is of great signifi-cance for devel...Fe-based metallic glasses(MGs)have shown great commercial values due to their excellent soft magnetic properties.Magnetism prediction with consideration of glass forming ability(GFA)is of great signifi-cance for developing novel functional Fe-based MGs.However,theories or models established based on condensed matter physics exhibit limited accuracy and some exceptions.In this work,based on 618 Fe-based MGs samples collected from published works,machine learning(ML)models were well trained to predict saturated magnetization(B_(s))of Fe-based MGs.GFA was treated as a feature using the experimental data of the supercooled liquid region(△T_(x)).Three ML algorithms,namely eXtreme gradient boosting(XGBoost),artificial neural networks(ANN)and random forest(RF),were studied.Through feature selection and hyperparameter tuning,XGBoost showed the best predictive performance on the randomly split test dataset with determination coefficient(R^(2))of 0.942,mean absolute percent error(MAPE)of 5.563%,and root mean squared error(RMSE)of 0.078 T.A variety of feature importance rankings derived by XGBoost models showed that T_(x) played an important role in the predictive performance of the models.This work showed the proposed ML method can simultaneously aggregate GFA and other features in ther-modynamics,kinetics and structures to predict the magnetic properties of Fe-based MGs with excellent accuracy.展开更多
文摘石英的微量元素记录了石英生长的物理化学条件。通过微量元素对石英原岩进行分类的研究历史已久,经典工作是在以微量元素为坐标轴的图解上绘制各类型石英的分布范围,以区分石英类型。经典图解包括Rusk(2012)提出用于区分三种矿床类型石英的Al-Ti二元图解,和Schr9n et al.(1988)提出的用于判别不同岩浆岩类型石英的Ti-Al-Ge三元图解。越来越多的研究表明,上述图解不能满足对更多石英类型进行分类的需求,同时也出现与部分已知产状类型的石英微量元素判别相矛盾的情况。随着石英原位微区测试方法的成熟,高精度石英微量元素数据逐渐丰富为系统开展机器学习提供了大数据基础,为石英微量元素研究提供了新的角度和可能性。本研究运用机器学习分类方法对石英微量元素进行精确数学分析,提出Ti/Ge-P图解为石英成因研究提出新的地球化学指标。本文同时测试了六种经典机器学习分类算法,提高Ti/Ge-P图解在石英成因分类研究上的精度。此Ti/Ge-P图解适用于多种矿床研究,包括但不局限于斑岩型矿床、矽卡岩型矿床、浅成低温热液型矿床、卡林型矿床以及造山型矿床中的石英。这项工作是大数据技术与机器学习技术在地球化学研究中的积极探索。
基金supported by the National Natural Science Foundation of China(No.52177122)the“Transformational Technologies for Clean Energy and Demonstration”,Strategic Priority Research Program of the Chinese Academy of Sciences(No.XDA 21050100)the Youth Innovation Promotion Association CAS(No.2018170)。
文摘Grid-forming(GFM)converters can provide inertia support for power grids through control technology,stabilize voltage and frequency,and improve system stability,unlike traditional grid-following(GFL)converters.Therefore,in future“double high”power systems,research on the control technology of GFM converters will become an urgent demand.In this paper,we first introduce the basic principle of GFM control and then present five currently used control strategies for GFM converters:droop control,power synchronization control(PSC),virtual synchronous machine control(VSM),direct power control(DPC),and virtual oscillator control(VOC).These five strategies can independently establish voltage phasors to provide inertia to the system.Among these,droop control is the most widely used strategy.PSC and VSM are strategies that simulate the mechanical characteristics of synchronous generators;thus,they are more accurate than droop control.DPC regulates the active power and reactive power directly,with no inner current controller,and VOC is a novel method under study using an oscillator circuit to realize synchronization.Finally,we highlight key technologies and research directions to be addressed in the future.
基金financially supported by National Natural Science Foundation of China(No.21771017)the Fundamental Research Funds for the Central Universities。
文摘Fe-based metallic glasses(MGs)have shown great commercial values due to their excellent soft magnetic properties.Magnetism prediction with consideration of glass forming ability(GFA)is of great signifi-cance for developing novel functional Fe-based MGs.However,theories or models established based on condensed matter physics exhibit limited accuracy and some exceptions.In this work,based on 618 Fe-based MGs samples collected from published works,machine learning(ML)models were well trained to predict saturated magnetization(B_(s))of Fe-based MGs.GFA was treated as a feature using the experimental data of the supercooled liquid region(△T_(x)).Three ML algorithms,namely eXtreme gradient boosting(XGBoost),artificial neural networks(ANN)and random forest(RF),were studied.Through feature selection and hyperparameter tuning,XGBoost showed the best predictive performance on the randomly split test dataset with determination coefficient(R^(2))of 0.942,mean absolute percent error(MAPE)of 5.563%,and root mean squared error(RMSE)of 0.078 T.A variety of feature importance rankings derived by XGBoost models showed that T_(x) played an important role in the predictive performance of the models.This work showed the proposed ML method can simultaneously aggregate GFA and other features in ther-modynamics,kinetics and structures to predict the magnetic properties of Fe-based MGs with excellent accuracy.
