In real-world many internet-based service companies need to closely monitor large amounts of data in order to ensure stable operation of their business.However,anomaly detection for these data with various patterns an...In real-world many internet-based service companies need to closely monitor large amounts of data in order to ensure stable operation of their business.However,anomaly detection for these data with various patterns and data quality has been a great challenge,especially without labels.In this paper,we adopt an anomaly detection algorithm based on Long Short-Term Memory(LSTM)Network in terms of reconstructing KPIs and predicting KPIs.They use the reconstruction error and prediction error respectively as the criteria for judging anomalies,and we test our method with real data from a company in the insurance industry and achieved good performance.展开更多
For many Internet companies,a huge amount of KPIs(e.g.,server CPU usage,network usage,business monitoring data)will be generated every day.How to closely monitor various KPIs,and then quickly and accurately detect ano...For many Internet companies,a huge amount of KPIs(e.g.,server CPU usage,network usage,business monitoring data)will be generated every day.How to closely monitor various KPIs,and then quickly and accurately detect anomalies in such huge data for troubleshooting and recovering business is a great challenge,especially for unlabeled data.The generated KPIs can be detected by supervised learning with labeled data,but the current problem is that most KPIs are unlabeled.That is a time-consuming and laborious work to label anomaly for company engineers.Build an unsupervised model to detect unlabeled data is an urgent need at present.In this paper,unsupervised learning DBSCAN combined with feature extraction of data has been used,and for some KPIs,its best F-Score can reach about 0.9,which is quite good for solving the current problem.展开更多
Most of the isolated electrical systems throughout the world suffer from similar problems of fragility and high dependence on external resources to generate energy. Smart Grid solutions and integration of renewable en...Most of the isolated electrical systems throughout the world suffer from similar problems of fragility and high dependence on external resources to generate energy. Smart Grid solutions and integration of renewable energies in order to solve their problems have increased, although it is necessary to know their specific characteristics to select the optimal solutions for each case. Therefore, as the overall objective of INSULAE Project, the development of an Investment Planning Tool, IPT, is on the way. This paper provides a view on a characterization methodology developed for the set of Reference Islands and how it will help to exploit the IPT developed. For that, characterization vectors have been defined based on a selection of Key Performance Indicators (KPIs). And Reference Islands have been obtained from the analysis of KPIs data gathered from EU islands considering the vectors formed. The linkage of new islands to reference islands helps provide the new islands with an assessment on the possibility space of their investment plans with the aim of being a decarbonization plan considering the demonstrations already evaluated.展开更多
This paper builds a self-adaptive management process in the power system dispatching area, aiming to effectively monitor the grid operation, dynamically adjust control strategy, optimize working process and ensure the...This paper builds a self-adaptive management process in the power system dispatching area, aiming to effectively monitor the grid operation, dynamically adjust control strategy, optimize working process and ensure the continuous improvement of operational performance. By building a negative feedback and dynamic balanced management mechanism, ECPRCB （East China Power Regulation Center Branch） is able to keenly sense the internal and external changes, efficiently coordinate all kinds of resources and improve the operational performance. As a result, self-adaptive management effectively boosts ECPRCB to reach the goal of being a world-class dispatching center with high operational performance, competent internal operation, adequate resources support and strong growth motion.展开更多
Haptic is the modality that complements traditional multimedia,i.e.,audiovisual,to evolve the next wave of innovation at which the Internet data stream can be exchanged to enable remote skills and control applications...Haptic is the modality that complements traditional multimedia,i.e.,audiovisual,to evolve the next wave of innovation at which the Internet data stream can be exchanged to enable remote skills and control applications.This will require ultra-low latency and ultra-high reliability to evolve the mobile experience into the era of Digital Twin and Tactile Internet.While the 5th generation of mobile networks is not yet widely deployed,Long-Term Evolution(LTE-A)latency remains much higher than the 1 ms requirement for the Tactile Internet and therefore the Digital Twin.This work investigates an interesting solution based on the incorporation of Software-defined networking(SDN)and Multi-access Mobile Edge Computing(MEC)technologies in an LTE-A network,to deliver future multimedia applications over the Tactile Internet while overcoming the QoS challenges.Several network scenarios were designed and simulated using Riverbed modeler and the performance was evaluated using several time-related Key Performance Indicators(KPIs)such as throughput,End-2-End(E2E)delay,and jitter.The best scenario possible is clearly the one integrating MEC and SDN approaches,where the overall delay,jitter,and throughput for haptics-attained 2 ms,0.01 ms,and 1000 packets per second.The results obtained give clear evidence that the integration of,both SDN and MEC,in LTE-A indicates performance improvement,and fulfills the standard requirements in terms of the above KPIs,for realizing a Digital Twin/Tactile Internet-based system.展开更多
文摘In real-world many internet-based service companies need to closely monitor large amounts of data in order to ensure stable operation of their business.However,anomaly detection for these data with various patterns and data quality has been a great challenge,especially without labels.In this paper,we adopt an anomaly detection algorithm based on Long Short-Term Memory(LSTM)Network in terms of reconstructing KPIs and predicting KPIs.They use the reconstruction error and prediction error respectively as the criteria for judging anomalies,and we test our method with real data from a company in the insurance industry and achieved good performance.
文摘For many Internet companies,a huge amount of KPIs(e.g.,server CPU usage,network usage,business monitoring data)will be generated every day.How to closely monitor various KPIs,and then quickly and accurately detect anomalies in such huge data for troubleshooting and recovering business is a great challenge,especially for unlabeled data.The generated KPIs can be detected by supervised learning with labeled data,but the current problem is that most KPIs are unlabeled.That is a time-consuming and laborious work to label anomaly for company engineers.Build an unsupervised model to detect unlabeled data is an urgent need at present.In this paper,unsupervised learning DBSCAN combined with feature extraction of data has been used,and for some KPIs,its best F-Score can reach about 0.9,which is quite good for solving the current problem.
文摘Most of the isolated electrical systems throughout the world suffer from similar problems of fragility and high dependence on external resources to generate energy. Smart Grid solutions and integration of renewable energies in order to solve their problems have increased, although it is necessary to know their specific characteristics to select the optimal solutions for each case. Therefore, as the overall objective of INSULAE Project, the development of an Investment Planning Tool, IPT, is on the way. This paper provides a view on a characterization methodology developed for the set of Reference Islands and how it will help to exploit the IPT developed. For that, characterization vectors have been defined based on a selection of Key Performance Indicators (KPIs). And Reference Islands have been obtained from the analysis of KPIs data gathered from EU islands considering the vectors formed. The linkage of new islands to reference islands helps provide the new islands with an assessment on the possibility space of their investment plans with the aim of being a decarbonization plan considering the demonstrations already evaluated.
文摘This paper builds a self-adaptive management process in the power system dispatching area, aiming to effectively monitor the grid operation, dynamically adjust control strategy, optimize working process and ensure the continuous improvement of operational performance. By building a negative feedback and dynamic balanced management mechanism, ECPRCB （East China Power Regulation Center Branch） is able to keenly sense the internal and external changes, efficiently coordinate all kinds of resources and improve the operational performance. As a result, self-adaptive management effectively boosts ECPRCB to reach the goal of being a world-class dispatching center with high operational performance, competent internal operation, adequate resources support and strong growth motion.
文摘Haptic is the modality that complements traditional multimedia,i.e.,audiovisual,to evolve the next wave of innovation at which the Internet data stream can be exchanged to enable remote skills and control applications.This will require ultra-low latency and ultra-high reliability to evolve the mobile experience into the era of Digital Twin and Tactile Internet.While the 5th generation of mobile networks is not yet widely deployed,Long-Term Evolution(LTE-A)latency remains much higher than the 1 ms requirement for the Tactile Internet and therefore the Digital Twin.This work investigates an interesting solution based on the incorporation of Software-defined networking(SDN)and Multi-access Mobile Edge Computing(MEC)technologies in an LTE-A network,to deliver future multimedia applications over the Tactile Internet while overcoming the QoS challenges.Several network scenarios were designed and simulated using Riverbed modeler and the performance was evaluated using several time-related Key Performance Indicators(KPIs)such as throughput,End-2-End(E2E)delay,and jitter.The best scenario possible is clearly the one integrating MEC and SDN approaches,where the overall delay,jitter,and throughput for haptics-attained 2 ms,0.01 ms,and 1000 packets per second.The results obtained give clear evidence that the integration of,both SDN and MEC,in LTE-A indicates performance improvement,and fulfills the standard requirements in terms of the above KPIs,for realizing a Digital Twin/Tactile Internet-based system.
