特征提取和健康状态的辨识是复杂系统健康状态评估中的关键问题。提出一种新的健康状态评估方法,该方法分为3个步骤:首先,采用经验模态分解(empirical model decomposition,EMD)和奇异值分解(singular value decomposition,SVD)来提取...特征提取和健康状态的辨识是复杂系统健康状态评估中的关键问题。提出一种新的健康状态评估方法,该方法分为3个步骤:首先,采用经验模态分解(empirical model decomposition,EMD)和奇异值分解(singular value decomposition,SVD)来提取振动信号的特征变量。然后,运用马田系统(Mahalanobis-Taguchi system,MTS)构造马氏空间,并对其进行优化,从而降低特征变量的维度。最后,提出了一种健康度(health index,HI)的概念,并且用来对复杂系统健康问题进行评估。该方法成功地应用在轴承的健康状态评估中。展开更多
The climate variability induced by the El Nino-Southern Oscillation(ENSO)cycle drives significant changes in the physical state of the tropical Western Pacific,which has important impacts on the upper ocean carbon cyc...The climate variability induced by the El Nino-Southern Oscillation(ENSO)cycle drives significant changes in the physical state of the tropical Western Pacific,which has important impacts on the upper ocean carbon cycle.During 2015-2016,a super El Nino event occurred in the equatorial Pacific.Suspended particulate matter(SPM)data and related environmental observations in the tropical Western Pacific were obtained during two cruses in Dec.2014 and 2015,which coincided with the early and peak stages of this super El Nino event.Compared with the marine environments in the tropical Western Pacific in Dec.2014,an obviously enhanced upwelling occurred in the Mindanao Dome region;the nitrate concentration in the euphotic zone almo st tripled;and the size,mass concentration,and volume concentration of SPM obviously increased in Dec.2015.The enhanced upwelling in the Mindanao Dome region carried cold but eutrophic water upward from the deep ocean to shallow depths,even into the euphotic zone,which disrupted the previously N-limited conditions and induced a remarkable increase in phytoplankton blooms in the euphotic zone.The se results reveal the mechanism of how nutrient-limited ecosystems in the tropical Western Pacific respond to super El Nino events.In the context of the ENSO cycle,if predicted changes in biogenic particles occur,the proportion of carbon storage in the tropical Western Pacific is estimated to be increased by more than 52%,ultimately affecting the regional and possibly even global carbon cycle.This paper highlights the prospect for long-term prediction of the impact of a super El Nino event on the global carbon cycle and has profound implications for understanding El Nino events.展开更多
The mechanism of the locking of the E1 Nino event onset phase to boreal spring （from April to June） in an intermediate coupled ocean-atmosphere model is investigated. The results show that the seasonal variation of ...The mechanism of the locking of the E1 Nino event onset phase to boreal spring （from April to June） in an intermediate coupled ocean-atmosphere model is investigated. The results show that the seasonal variation of the zonal wind anomaly over the equatorial Pacific associated with the seasonal variation of the ITCZ is the mechanism of the locking in the model. From January to March of the E1 Nino year, the western wind anomaly over the western equatorial Pacific can excite the downwelling Kelvin wave that propagates eastward to the eastern and middle Pacific by April to June. From April to December of the year before the E1 Nifio year, the eastern wind anomaly over the equatorial Pacific forces the downwelling Rossby waves that modulate the ENSO cycle. The modulation and the reflection at the western boundary modulate the time of the transition from the cool to the warm phase to September of the year before the E1 Nifio year and cause the strongest downwelling Kelvin wave from the reflected Rossby waves at the western boundary to arrive in the middle and eastern equatorial Pacific by April to June of the E1 Nino year. The superposition of these two kinds of downwelling Kelvin waves causes the El Nino event to tend to occur from April to June.展开更多
In this paper, a Gaussian mixture model (GMM) based classifier is described to tell whether precipitation events will happen on a certain day at a certain time from historical meteorological data. The classifier deals...In this paper, a Gaussian mixture model (GMM) based classifier is described to tell whether precipitation events will happen on a certain day at a certain time from historical meteorological data. The classifier deals with a two-class classification problem where one class represents precipitation events and the other represents non-precipitation events. The concept of ambiguity is introduced to represent cases where weather conditions between the two classes like drizzles, intermittent or overcast are more likely to happen. Six groups of experiments are carried out to evaluate the performance of the classifier using different configurations based on the observation data released by Shanghai Baoshan weather station. Specifically, a typical classification performance of about 75% accuracy, 30% precision and 80% recall is achieved for prediction tasks with a time span of 12 hours.展开更多
The authors demonstrate that the E1 Nifio events in the pre- and post-1976 periods show two ampli- tude-duration relations. One is that the stronger E1 Nifio events have longer durations, which is robust for the moder...The authors demonstrate that the E1 Nifio events in the pre- and post-1976 periods show two ampli- tude-duration relations. One is that the stronger E1 Nifio events have longer durations, which is robust for the moderate E1 Nifio events; the other is that the stronger E1 Nifio events have shorter durations but for strong E1 Nifio events. By estimating the sign and amplitude of the nonlinear dynamical heating （NDH） anomalies, the au- thors illustrate that the NDH anomalies are negligible for moderate E1 Nifio events but large for strong E1 Nifio events. In particular, the large NDH anomalies for strong E1 Nifio events are positive during the growth and mature phases, which favor warmer E1 Nifio events. During the decay phase, however, the negative NDH anomalies start to arise and become increasingly significant with the evolution of the E1 Nifio events, in which the negative NDH anomalies dampen the sea surface temperature anomalies （SSTA） and cause the E1 Nifio events to reach the SST normal state earlier. This pattern suggests that the nonlinearity tends to increase the intensities of strong E1 Nifio events and shorten their duration, which, together with the previous results showing a positive correlation between the strength of E1 Nifio events and the signifi- cance of the effect of nonlinear advection on the events （especially the suppression of nonlinearity on the SSTA during the decay phase）, shows that the strong E1 Nifio events tend to have the amplitude-duration relation of the stronger E1 Nifio events with shorter durations. This result also lends support to the assertion that moderate E1 Nifio events possess the amplitude-duration relation of stronger E1 Nifio events with longer durations.展开更多
文摘特征提取和健康状态的辨识是复杂系统健康状态评估中的关键问题。提出一种新的健康状态评估方法,该方法分为3个步骤:首先,采用经验模态分解(empirical model decomposition,EMD)和奇异值分解(singular value decomposition,SVD)来提取振动信号的特征变量。然后,运用马田系统(Mahalanobis-Taguchi system,MTS)构造马氏空间,并对其进行优化,从而降低特征变量的维度。最后,提出了一种健康度(health index,HI)的概念,并且用来对复杂系统健康问题进行评估。该方法成功地应用在轴承的健康状态评估中。
基金the Strategic Priority Research Program of the Chinese Academy of Sciences(Nos.XDB42010203,XDA19060401,XDA23050503)the Science&Technology Basic Resources Investigation Program of China(No.2017FY100802)+1 种基金the Open Fund for Key Laboratory of Mar.Geol.and Environment,Institute of Oceanology,Chinese Academy of Sciences(No.MGE2019KG03)the Qingdao(Laoshan)Postdoctoral Applied Research Proj ect in 2019(No.Y9KY161)。
文摘The climate variability induced by the El Nino-Southern Oscillation(ENSO)cycle drives significant changes in the physical state of the tropical Western Pacific,which has important impacts on the upper ocean carbon cycle.During 2015-2016,a super El Nino event occurred in the equatorial Pacific.Suspended particulate matter(SPM)data and related environmental observations in the tropical Western Pacific were obtained during two cruses in Dec.2014 and 2015,which coincided with the early and peak stages of this super El Nino event.Compared with the marine environments in the tropical Western Pacific in Dec.2014,an obviously enhanced upwelling occurred in the Mindanao Dome region;the nitrate concentration in the euphotic zone almo st tripled;and the size,mass concentration,and volume concentration of SPM obviously increased in Dec.2015.The enhanced upwelling in the Mindanao Dome region carried cold but eutrophic water upward from the deep ocean to shallow depths,even into the euphotic zone,which disrupted the previously N-limited conditions and induced a remarkable increase in phytoplankton blooms in the euphotic zone.The se results reveal the mechanism of how nutrient-limited ecosystems in the tropical Western Pacific respond to super El Nino events.In the context of the ENSO cycle,if predicted changes in biogenic particles occur,the proportion of carbon storage in the tropical Western Pacific is estimated to be increased by more than 52%,ultimately affecting the regional and possibly even global carbon cycle.This paper highlights the prospect for long-term prediction of the impact of a super El Nino event on the global carbon cycle and has profound implications for understanding El Nino events.
基金This work was supported by The National Key Basic Reserch and Development Project of China(2004CB418303)Project 4023100 of the Major Research Program for Global Change and Regional ResponseNational Natural Science Foundation of China(Grant No.40231005).
文摘The mechanism of the locking of the E1 Nino event onset phase to boreal spring （from April to June） in an intermediate coupled ocean-atmosphere model is investigated. The results show that the seasonal variation of the zonal wind anomaly over the equatorial Pacific associated with the seasonal variation of the ITCZ is the mechanism of the locking in the model. From January to March of the E1 Nino year, the western wind anomaly over the western equatorial Pacific can excite the downwelling Kelvin wave that propagates eastward to the eastern and middle Pacific by April to June. From April to December of the year before the E1 Nifio year, the eastern wind anomaly over the equatorial Pacific forces the downwelling Rossby waves that modulate the ENSO cycle. The modulation and the reflection at the western boundary modulate the time of the transition from the cool to the warm phase to September of the year before the E1 Nifio year and cause the strongest downwelling Kelvin wave from the reflected Rossby waves at the western boundary to arrive in the middle and eastern equatorial Pacific by April to June of the E1 Nino year. The superposition of these two kinds of downwelling Kelvin waves causes the El Nino event to tend to occur from April to June.
文摘In this paper, a Gaussian mixture model (GMM) based classifier is described to tell whether precipitation events will happen on a certain day at a certain time from historical meteorological data. The classifier deals with a two-class classification problem where one class represents precipitation events and the other represents non-precipitation events. The concept of ambiguity is introduced to represent cases where weather conditions between the two classes like drizzles, intermittent or overcast are more likely to happen. Six groups of experiments are carried out to evaluate the performance of the classifier using different configurations based on the observation data released by Shanghai Baoshan weather station. Specifically, a typical classification performance of about 75% accuracy, 30% precision and 80% recall is achieved for prediction tasks with a time span of 12 hours.
基金jointly sponsored by the National Basic Research Program of China (Nos.2010CB950402 and 2012CB955202)the Knowledge Innovation Program of the Chinese Academy of Sciences (No. KZCX2-YW-QN203)the National Natural Science Foundation of China (No.41176013)
文摘The authors demonstrate that the E1 Nifio events in the pre- and post-1976 periods show two ampli- tude-duration relations. One is that the stronger E1 Nifio events have longer durations, which is robust for the moderate E1 Nifio events; the other is that the stronger E1 Nifio events have shorter durations but for strong E1 Nifio events. By estimating the sign and amplitude of the nonlinear dynamical heating （NDH） anomalies, the au- thors illustrate that the NDH anomalies are negligible for moderate E1 Nifio events but large for strong E1 Nifio events. In particular, the large NDH anomalies for strong E1 Nifio events are positive during the growth and mature phases, which favor warmer E1 Nifio events. During the decay phase, however, the negative NDH anomalies start to arise and become increasingly significant with the evolution of the E1 Nifio events, in which the negative NDH anomalies dampen the sea surface temperature anomalies （SSTA） and cause the E1 Nifio events to reach the SST normal state earlier. This pattern suggests that the nonlinearity tends to increase the intensities of strong E1 Nifio events and shorten their duration, which, together with the previous results showing a positive correlation between the strength of E1 Nifio events and the signifi- cance of the effect of nonlinear advection on the events （especially the suppression of nonlinearity on the SSTA during the decay phase）, shows that the strong E1 Nifio events tend to have the amplitude-duration relation of the stronger E1 Nifio events with shorter durations. This result also lends support to the assertion that moderate E1 Nifio events possess the amplitude-duration relation of stronger E1 Nifio events with longer durations.
