Two great earthquakes occurred in the sea northwest of Sumatra,Indonesia,on December 26,2004 and March 29,2005.The observation of water levels in Yunnan yielded abundant information about the two earthquakes.This pape...Two great earthquakes occurred in the sea northwest of Sumatra,Indonesia,on December 26,2004 and March 29,2005.The observation of water levels in Yunnan yielded abundant information about the two earthquakes.This paper presents the water level response to the two earthquakes in Yunnan and makes a preliminary analysis.It is observed that the large earthquake-induced abnormal water level change could be better recorded by analog recording than by digital recording.The large earthquake-caused water level rise or decline may be attributed to the effect of seismic waves that change the stress in tectonic units,and is correlated with the geological structure where the well is located.The water level response mode in a well is totally the same for earthquakes occurring on the same fault and with the same fracture mode.The only difference is that the response amplitude increases with the growth of the earthquake magnitude.展开更多
With the development of advanced metering infrastructure(AMI),large amounts of electricity consumption data can be collected for electricity theft detection.However,the imbalance of electricity consumption data is vio...With the development of advanced metering infrastructure(AMI),large amounts of electricity consumption data can be collected for electricity theft detection.However,the imbalance of electricity consumption data is violent,which makes the training of detection model challenging.In this case,this paper proposes an electricity theft detection method based on ensemble learning and prototype learning,which has great performance on imbalanced dataset and abnormal data with different abnormal level.In this paper,convolutional neural network(CNN)and long short-term memory(LSTM)are employed to obtain abstract feature from electricity consumption data.After calculating the means of the abstract feature,the prototype per class is obtained,which is used to predict the labels of unknown samples.In the meanwhile,through training the network by different balanced subsets of training set,the prototype is representative.Compared with some mainstream methods including CNN,random forest(RF)and so on,the proposed method has been proved to effectively deal with the electricity theft detection when abnormal data only account for 2.5%and 1.25%of normal data.The results show that the proposed method outperforms other state-of-the-art methods.展开更多
基金sponsored by the "Personnel Training" of Yunnan Province (2006PY0139)the National Key Technology R & D Program for the 11th "Five-Year Plan"(Grant No.2006BAC01B020302),China
文摘Two great earthquakes occurred in the sea northwest of Sumatra,Indonesia,on December 26,2004 and March 29,2005.The observation of water levels in Yunnan yielded abundant information about the two earthquakes.This paper presents the water level response to the two earthquakes in Yunnan and makes a preliminary analysis.It is observed that the large earthquake-induced abnormal water level change could be better recorded by analog recording than by digital recording.The large earthquake-caused water level rise or decline may be attributed to the effect of seismic waves that change the stress in tectonic units,and is correlated with the geological structure where the well is located.The water level response mode in a well is totally the same for earthquakes occurring on the same fault and with the same fracture mode.The only difference is that the response amplitude increases with the growth of the earthquake magnitude.
基金supported by National Natural Science Foundation of China(No.52277083).
文摘With the development of advanced metering infrastructure(AMI),large amounts of electricity consumption data can be collected for electricity theft detection.However,the imbalance of electricity consumption data is violent,which makes the training of detection model challenging.In this case,this paper proposes an electricity theft detection method based on ensemble learning and prototype learning,which has great performance on imbalanced dataset and abnormal data with different abnormal level.In this paper,convolutional neural network(CNN)and long short-term memory(LSTM)are employed to obtain abstract feature from electricity consumption data.After calculating the means of the abstract feature,the prototype per class is obtained,which is used to predict the labels of unknown samples.In the meanwhile,through training the network by different balanced subsets of training set,the prototype is representative.Compared with some mainstream methods including CNN,random forest(RF)and so on,the proposed method has been proved to effectively deal with the electricity theft detection when abnormal data only account for 2.5%and 1.25%of normal data.The results show that the proposed method outperforms other state-of-the-art methods.
