With the development of the Cyber-Physical-Social Systems(CPSS), a large number of multidimensional time series have been generated in today's world, such as: sensor data for industrial equipment operation, vehicle driving data, and cloud server operation and maintenance data and so on. The key task of Cloud-Fog-Edge Computing in managing these systems is how to detect anomalous data in a specific time series to facilitate operator action to solve potential system problems. So multidimensional time series outlier detection become an important direction of CPSS data mining and Cloud-Fog-Edge Computing research, and it has a wide range of applications in industry, finance, medicine and other fields. This paper proposes a framework called Multidimensional time series Outlier detection based on a Time Convolutional Network AutoEncoder (MOTCN-AE), which can detect outliers in time series data, such as identifying equipment failures, dangerous driving behaviors of cars, etc. Specifically, this paper first uses a feature extraction method to transform the original time series into a feature-rich time series. Second, the proposed TCN-AE is used to reconstruct the feature-rich time series data, and the reconstruction error is used to calculate outlier scores. Finally, the MOTCN-AE framework is validated by multiple time series datasets to demonstrate its effectiveness in detecting time series outliers. INDEX TERMS Time series, outlier detection, time convolution network, autoencoder.

Complex network reconfiguration has always been an important task in complex network research. Simple and effective complex network reconstruction methods can promote the understanding of the operation of complex systems in the real world. There are many complex systems, such as stock systems, social systems and thermal power systems. These systems generally produce correlated time series of data. Discovering the relationships among these multivariate time series is the focus of this research. This paper proposes a Spearman coefficient reconstruction network (SCRN) method based on the Spearman correlation coefficient. In the SCRN method, we select entities in the real world as the nodes of the network and determine connection weights of the network edges by calculating the Spearman correlation coefficients among nodes. In this paper, we selected a stock system and boiler equipment in a thermal power generation system to construct two complex network models. For the stock network model, we used the classic Girvan–Newman (GN) algorithm for community discovery to determine whether the proposed network topology is reasonable. For the boiler network model, we built a predictive model based on an support vector regression (SVR) model in machine learning, and we verified the rationality of the boiler model by predicting the amount of boiler steam.

No abstract