Achieving reliable and efficient weather classification for autonomous vehicles is crucial for ensuring safety and operational effectiveness.However,accurately classifying diverse and complex weather conditions remain...Achieving reliable and efficient weather classification for autonomous vehicles is crucial for ensuring safety and operational effectiveness.However,accurately classifying diverse and complex weather conditions remains a significant challenge.While advanced techniques such as Vision Transformers have been developed,they face key limitations,including high computational costs and limited generalization across varying weather conditions.These challenges present a critical research gap,particularly in applications where scalable and efficient solutions are needed to handle weather phenomena’intricate and dynamic nature in real-time.To address this gap,we propose a Multi-level Knowledge Distillation(MLKD)framework,which leverages the complementary strengths of state-of-the-art pre-trained models to enhance classification performance while minimizing computational overhead.Specifically,we employ ResNet50V2 and EfficientNetV2B3 as teacher models,known for their ability to capture complex image features and distil their knowledge into a custom lightweight Convolutional Neural Network(CNN)student model.This framework balances the trade-off between high classification accuracy and efficient resource consumption,ensuring real-time applicability in autonomous systems.Our Response-based Multi-level Knowledge Distillation(R-MLKD)approach effectively transfers rich,high-level feature representations from the teacher models to the student model,allowing the student to perform robustly with significantly fewer parameters and lower computational demands.The proposed method was evaluated on three public datasets(DAWN,BDD100K,and CITS traffic alerts),each containing seven weather classes with 2000 samples per class.The results demonstrate the effectiveness of MLKD,achieving a 97.3%accuracy,which surpasses conventional deep learning models.This work improves classification accuracy and tackles the practical challenges of model complexity,resource consumption,and real-time deployment,offering a scalable solution for weather classification in autonomous driving systems.展开更多
Using the 2015-2018 Hangzhou city PM2.5,PM10,SO2,CO,NO2 and O3 mass concentration data,ERA5 reanalysis data and ground observation data,through the PCT classification method,the objective analysis of the winter air po...Using the 2015-2018 Hangzhou city PM2.5,PM10,SO2,CO,NO2 and O3 mass concentration data,ERA5 reanalysis data and ground observation data,through the PCT classification method,the objective analysis of the winter air pollution weather situation in Hangzhou was obtained.The results showed that the winter air quality concentration in Hangzhou continued to be high from 2015 to 2018,and the air pollution was the most significant.Through objective classification,it is concluded that the main weather conditions affecting the region in winter are divided into 6 types,namely high pressure control,high pressure bottom control equalizing field,L-shaped high pressure control,high pressure front control equalizing field,low pressure control,low pressure front control Equalizing field.Among them,when high pressure control,high pressure bottom control equalizing field,L high pressure control,low pressure control are affected by local sources,the impact of external sources has a greater impact on the air quality in Hangzhou,and air pollution is prone to occur;before low pressure When the pressure equalization field is controlled by the Ministry and the pressure equalization field is controlled by the high pressure front,the local wind and precipitation in Hangzhou are relatively high,which is not conducive to the accumulation of air pollutants.The probability of occurrence of air pollution is small,and air pollution is not easy to occur.展开更多
Objective weather classification methods have been extensively applied to identify dominant ozone-favorable synoptic weather patterns(SWPs),however,the consistency of different classification methods is rarely examine...Objective weather classification methods have been extensively applied to identify dominant ozone-favorable synoptic weather patterns(SWPs),however,the consistency of different classification methods is rarely examined.In this study,we apply two widely-used objective methods,the self-organizing map(SOM)and K-means clustering analysis,to derive ozone-favorable SWPs at four Chinese megacities in 2015-2022.We find that the two algorithms are largely consistent in recognizing dominant ozone-favorable SWPs for four Chinese megacities.In the case of classifying six SWPs,the derived circulation fields are highly similar with a spatial correlation of 0.99 between the two methods,and the difference in themean frequency of each SWP is less than 7%.The six dominant ozone-favorable SWPs in Guangzhou are all characterized by anomaly higher radiation and temperature,lower cloud cover,relative humidity,and wind speed,and stronger subsidence compared to climatology mean.We find that during 2015-2022,the occurrence of ozone-favorable SWPs days increases significantly at a rate of 3.2 days/year,faster than the increases in the ozone exceedance days(3.0 days/year).The interannual variability between the occurrence of ozone-favorable SWPs and ozone exceedance days are generally consistent with a temporal correlation coefficient of 0.6.In particular,the significant increase in ozone-favorable SWPs in 2022,especially the Subtropical High type which typically occurs in September,is consistent with a long-lasting ozone pollution episode in Guangzhou during September 2022.Our results thus reveal that enhanced frequency of ozone-favorable SWPs plays an important role in the observed 2015-2022 ozone increase in Guangzhou.展开更多
Extreme ozone pollution events(EOPEs)are associated with synoptic weather patterns(SWPs)and pose severe health and ecological risks.However,a systematic investigation of themeteorological causes,transport pathways,and...Extreme ozone pollution events(EOPEs)are associated with synoptic weather patterns(SWPs)and pose severe health and ecological risks.However,a systematic investigation of themeteorological causes,transport pathways,and source contributions to historical EOPEs is still lacking.In this paper,the K-means clustering method is applied to identify six dominant SWPs during the warm season in the Yangtze River Delta(YRD)region from 2016 to 2022.It provides an integrated analysis of the meteorological factors affecting ozone pollution in Hefei under different SWPs.Using the WRF-FLEXPART model,the transport pathways(TPPs)and geographical sources of the near-surface air masses in Hefei during EOPEs are investigated.The results reveal that Hefei experienced the highest ozone concentration(134.77±42.82μg/m^(3)),exceedance frequency(46 days(23.23%)),and proportion of EOPEs(21 instances,47.7%)under the control of peripheral subsidence of typhoon(Type 5).Regional southeast winds correlated with the ozone pollution in Hefei.During EOPEs,a high boundary layer height,solar radiation,and temperature;lowhumidity and cloud cover;and pronounced subsidence airflow occurred over Hefei and the broader YRD region.The East-South(E_S)patterns exhibited the highest frequency(28 instances,65.11%).Regarding the TPPs and geographical sources of the near-surface air masses during historical EOPEs.The YRD was the main source for land-originating air masses under E_S patterns(50.28%),with Hefei,southern Anhui,southern Jiangsu,and northern Zhejiang being key contributors.These findings can help improve ozone pollution early warning and control mechanisms at urban and regional scales.展开更多
The RMR system is still very much applied in rock mechanics engineering context. It is based on the evaluation of six weights to obtain a final rating. To obtain the final rating a considerable amount of information i...The RMR system is still very much applied in rock mechanics engineering context. It is based on the evaluation of six weights to obtain a final rating. To obtain the final rating a considerable amount of information is needed concerning the rock mass which can be difficult to obtain in some projects or project stages at least with accuracy. In 2007 an alternative classification scheme based on the RMR, the Hierarchical Rock Mass Rating(HRMR) was presented. The main feature of this system was the adaptation to the level of knowledge existent about the rock mass to obtain the classification of the rock mass since it followed a decision tree approach. However, the HRMR was only valid for hard rock granites with low fracturing degrees. In this work, the database was enlarged with approximately 40% more cases considering other types of granite rock masses including weathered granites and based on this increased database the system was updated. Granite formations existent in the north of Portugal including Porto city are predominantly granites. Some years ago a light rail infrastructure was built in the city of Porto and surrounding municipalities which involved considerable challenges due to the high heterogeneity levels of the granite formations and the difficulties involved in their geomechanical characterization. In this work it is intended to provide also a contribution to improve the characterization of these formations with special emphasis to the weathered horizons. A specific subsystem applicable to the weathered formations was developed. The results of the validation of these systems are presented and show acceptable performances in identifying the correct class using less information than with the RMR system.展开更多
Compensating for photovoltaic(PV)power forecast errors is an important function of energy storage systems.As PV power outputs have strong random fluctuations and uncertainty,it is difficult to satisfy the grid-connect...Compensating for photovoltaic(PV)power forecast errors is an important function of energy storage systems.As PV power outputs have strong random fluctuations and uncertainty,it is difficult to satisfy the grid-connection requirements using fixed energy storage capacity configuration methods.In this paper,a method of configuring energy storage capacity is proposed based on the uncertainty of PV power generation.A k-means clustering algorithm is used to classify weather types based on differences in solar irradiance.The power forecast errors in different weather types are analyzed,and an energy storage system is used to compensate for the errors.The kernel density estimation is used to fit the distributions of the daily maximum power and maximum capacity requirements of the energy storage system;the power and capacity of the energy storage unit are calculated at different confidence levels.The optimized energy storage configuration of a PV plant is presented according to the calculated degrees of power and capacity satisfaction.The proposed method was validated using actual operating data from a PV power station.The results indicated that the required energy storage can be significantly reduced while compensating for power forecast errors.展开更多
文摘Achieving reliable and efficient weather classification for autonomous vehicles is crucial for ensuring safety and operational effectiveness.However,accurately classifying diverse and complex weather conditions remains a significant challenge.While advanced techniques such as Vision Transformers have been developed,they face key limitations,including high computational costs and limited generalization across varying weather conditions.These challenges present a critical research gap,particularly in applications where scalable and efficient solutions are needed to handle weather phenomena’intricate and dynamic nature in real-time.To address this gap,we propose a Multi-level Knowledge Distillation(MLKD)framework,which leverages the complementary strengths of state-of-the-art pre-trained models to enhance classification performance while minimizing computational overhead.Specifically,we employ ResNet50V2 and EfficientNetV2B3 as teacher models,known for their ability to capture complex image features and distil their knowledge into a custom lightweight Convolutional Neural Network(CNN)student model.This framework balances the trade-off between high classification accuracy and efficient resource consumption,ensuring real-time applicability in autonomous systems.Our Response-based Multi-level Knowledge Distillation(R-MLKD)approach effectively transfers rich,high-level feature representations from the teacher models to the student model,allowing the student to perform robustly with significantly fewer parameters and lower computational demands.The proposed method was evaluated on three public datasets(DAWN,BDD100K,and CITS traffic alerts),each containing seven weather classes with 2000 samples per class.The results demonstrate the effectiveness of MLKD,achieving a 97.3%accuracy,which surpasses conventional deep learning models.This work improves classification accuracy and tackles the practical challenges of model complexity,resource consumption,and real-time deployment,offering a scalable solution for weather classification in autonomous driving systems.
基金supported by the National Natural Science Foundation of China(41975011)National key Research and Development Projects(2016YFA0602003)It also partially supported by Fuyang Government Funded Project of Hangzhou(ZJHCCGFY-0808001).
文摘Using the 2015-2018 Hangzhou city PM2.5,PM10,SO2,CO,NO2 and O3 mass concentration data,ERA5 reanalysis data and ground observation data,through the PCT classification method,the objective analysis of the winter air pollution weather situation in Hangzhou was obtained.The results showed that the winter air quality concentration in Hangzhou continued to be high from 2015 to 2018,and the air pollution was the most significant.Through objective classification,it is concluded that the main weather conditions affecting the region in winter are divided into 6 types,namely high pressure control,high pressure bottom control equalizing field,L-shaped high pressure control,high pressure front control equalizing field,low pressure control,low pressure front control Equalizing field.Among them,when high pressure control,high pressure bottom control equalizing field,L high pressure control,low pressure control are affected by local sources,the impact of external sources has a greater impact on the air quality in Hangzhou,and air pollution is prone to occur;before low pressure When the pressure equalization field is controlled by the Ministry and the pressure equalization field is controlled by the high pressure front,the local wind and precipitation in Hangzhou are relatively high,which is not conducive to the accumulation of air pollutants.The probability of occurrence of air pollution is small,and air pollution is not easy to occur.
基金supported by the Guangdong Basic and Applied Basic Research project (No.2020B0301030004)the Key-Area Research and Development Program of Guangdong Province (No.2020B1111360003)+1 种基金the National Natural Science Foundation of China (No.42105103)the Guangdong Basic and Applied Basic Research Foundation (No.2022A1515011554).
文摘Objective weather classification methods have been extensively applied to identify dominant ozone-favorable synoptic weather patterns(SWPs),however,the consistency of different classification methods is rarely examined.In this study,we apply two widely-used objective methods,the self-organizing map(SOM)and K-means clustering analysis,to derive ozone-favorable SWPs at four Chinese megacities in 2015-2022.We find that the two algorithms are largely consistent in recognizing dominant ozone-favorable SWPs for four Chinese megacities.In the case of classifying six SWPs,the derived circulation fields are highly similar with a spatial correlation of 0.99 between the two methods,and the difference in themean frequency of each SWP is less than 7%.The six dominant ozone-favorable SWPs in Guangzhou are all characterized by anomaly higher radiation and temperature,lower cloud cover,relative humidity,and wind speed,and stronger subsidence compared to climatology mean.We find that during 2015-2022,the occurrence of ozone-favorable SWPs days increases significantly at a rate of 3.2 days/year,faster than the increases in the ozone exceedance days(3.0 days/year).The interannual variability between the occurrence of ozone-favorable SWPs and ozone exceedance days are generally consistent with a temporal correlation coefficient of 0.6.In particular,the significant increase in ozone-favorable SWPs in 2022,especially the Subtropical High type which typically occurs in September,is consistent with a long-lasting ozone pollution episode in Guangzhou during September 2022.Our results thus reveal that enhanced frequency of ozone-favorable SWPs plays an important role in the observed 2015-2022 ozone increase in Guangzhou.
基金supported by the National Natural Science Foundation of China(Nos.U19A2044,42105132,42030609,and 41975037)the National Key Research and Development Programof China(No.2022YFC3700303).
文摘Extreme ozone pollution events(EOPEs)are associated with synoptic weather patterns(SWPs)and pose severe health and ecological risks.However,a systematic investigation of themeteorological causes,transport pathways,and source contributions to historical EOPEs is still lacking.In this paper,the K-means clustering method is applied to identify six dominant SWPs during the warm season in the Yangtze River Delta(YRD)region from 2016 to 2022.It provides an integrated analysis of the meteorological factors affecting ozone pollution in Hefei under different SWPs.Using the WRF-FLEXPART model,the transport pathways(TPPs)and geographical sources of the near-surface air masses in Hefei during EOPEs are investigated.The results reveal that Hefei experienced the highest ozone concentration(134.77±42.82μg/m^(3)),exceedance frequency(46 days(23.23%)),and proportion of EOPEs(21 instances,47.7%)under the control of peripheral subsidence of typhoon(Type 5).Regional southeast winds correlated with the ozone pollution in Hefei.During EOPEs,a high boundary layer height,solar radiation,and temperature;lowhumidity and cloud cover;and pronounced subsidence airflow occurred over Hefei and the broader YRD region.The East-South(E_S)patterns exhibited the highest frequency(28 instances,65.11%).Regarding the TPPs and geographical sources of the near-surface air masses during historical EOPEs.The YRD was the main source for land-originating air masses under E_S patterns(50.28%),with Hefei,southern Anhui,southern Jiangsu,and northern Zhejiang being key contributors.These findings can help improve ozone pollution early warning and control mechanisms at urban and regional scales.
文摘The RMR system is still very much applied in rock mechanics engineering context. It is based on the evaluation of six weights to obtain a final rating. To obtain the final rating a considerable amount of information is needed concerning the rock mass which can be difficult to obtain in some projects or project stages at least with accuracy. In 2007 an alternative classification scheme based on the RMR, the Hierarchical Rock Mass Rating(HRMR) was presented. The main feature of this system was the adaptation to the level of knowledge existent about the rock mass to obtain the classification of the rock mass since it followed a decision tree approach. However, the HRMR was only valid for hard rock granites with low fracturing degrees. In this work, the database was enlarged with approximately 40% more cases considering other types of granite rock masses including weathered granites and based on this increased database the system was updated. Granite formations existent in the north of Portugal including Porto city are predominantly granites. Some years ago a light rail infrastructure was built in the city of Porto and surrounding municipalities which involved considerable challenges due to the high heterogeneity levels of the granite formations and the difficulties involved in their geomechanical characterization. In this work it is intended to provide also a contribution to improve the characterization of these formations with special emphasis to the weathered horizons. A specific subsystem applicable to the weathered formations was developed. The results of the validation of these systems are presented and show acceptable performances in identifying the correct class using less information than with the RMR system.
基金supported by Nation Key R&D Program of China(2021YFE0102400).
文摘Compensating for photovoltaic(PV)power forecast errors is an important function of energy storage systems.As PV power outputs have strong random fluctuations and uncertainty,it is difficult to satisfy the grid-connection requirements using fixed energy storage capacity configuration methods.In this paper,a method of configuring energy storage capacity is proposed based on the uncertainty of PV power generation.A k-means clustering algorithm is used to classify weather types based on differences in solar irradiance.The power forecast errors in different weather types are analyzed,and an energy storage system is used to compensate for the errors.The kernel density estimation is used to fit the distributions of the daily maximum power and maximum capacity requirements of the energy storage system;the power and capacity of the energy storage unit are calculated at different confidence levels.The optimized energy storage configuration of a PV plant is presented according to the calculated degrees of power and capacity satisfaction.The proposed method was validated using actual operating data from a PV power station.The results indicated that the required energy storage can be significantly reduced while compensating for power forecast errors.
