Real-time anomaly detection of massive data streams is an important research topic nowadays due to the fact that a lot of data is generated in continuous temporal processes. There is a broad research area, covering ma...Real-time anomaly detection of massive data streams is an important research topic nowadays due to the fact that a lot of data is generated in continuous temporal processes. There is a broad research area, covering mathematical, statistical, information theory methodologies for anomaly detection. It addresses various problems in a lot of domains such as health, education, finance, government, etc. In this paper, we analyze the state-of-the-art of data streams anomaly detection techniques and algorithms for anomaly detection in data streams (time series data). Critically surveying the techniques’ performances under the challenge of real-time anomaly detection of massive high-velocity streams, we conclude that the modeling of the normal behavior of the stream is a suitable approach. We evaluate Holt-Winters (HW), Taylor’s Double Holt-Winters (TDHW), Hierarchical temporal memory (HTM), Moving Average (MA), Autoregressive integrated moving average (ARIMA) forecasting models, etc. Holt-Winters (HW) and Taylor’s Double Holt-Winters (TDHW) forecasting models are used to predict the normal behavior of the periodic streams, and to detect anomalies when the deviations of observed and predicted values exceeded some predefined measures. In this work, we propose an enhancement of this approach and give a short description about the algorithms and then they are categorized by type of pre-diction as: predictive and non-predictive algorithms. We implement the Genetic Algorithm (GA) to periodically optimize HW and TDHW smoothing parameters in addition to the two sliding windows parameters that improve Hyndman’s MASE measure of deviation, and value of the threshold parameter that defines no anomaly confidence interval [1]. We also propose a new optimization function based on the input training datasets with the annotated anomaly intervals, in order to detect the right anomalies and minimize the number of false ones. The proposed method is evaluated on the known anomaly detection benchmarks NUMENTA and Yahoo datasets with annotated anomalies and real log data generated by the National education information system (NEIS)1 in Macedonia.展开更多
In recent years, there is a wide interest in Sarkovskii's theorem ami the related study. According to Sarkovskii's theoren if the continuous self-mapf of the closed interval has a 3-pcriodic orbit, then fmust ...In recent years, there is a wide interest in Sarkovskii's theorem ami the related study. According to Sarkovskii's theoren if the continuous self-mapf of the closed interval has a 3-pcriodic orbit, then fmust has an n-pcriodic orbit for any positive integer n. But f can not has all n-periodic orbits for some n.For example, letEvidently, f has only one kind of 3-periodic orbit in the two kinds of 3-periodic orbits. This explains that it isn't far enough to uncover the relation between periodic orbits by information which Sarkovskii's theorem has offered. In this paper, we raise the concept of type of periodic orbits, and give a feasible algorithm which decides the relation of implication between two periodic orbits.展开更多
In this paper Genetic Algorithm has been integrated with Fouquet modal analysis to optimize radiation pattern of coupled periodic antenna. Floquet analysis is used with MoM-GEC (Moment-Generalized Equivalent Circuit) ...In this paper Genetic Algorithm has been integrated with Fouquet modal analysis to optimize radiation pattern of coupled periodic antenna. Floquet analysis is used with MoM-GEC (Moment-Generalized Equivalent Circuit) method to study a finite periodic array with uniform amplitude and linear phase distribution. This method is very advantageous for studying large antenna array since it considerably reduces the computation time and the number of operations. In this way, Genetic algorithm is introduced and combined with Floquet analysis to optimize the radiation pattern distribution of this coupled periodic antenna. The goal of the optimization is to provide a better radiation characteristic for the coupled periodic antenna with maximum side lobe level reduction.展开更多
Winding and web transport systems are subjected to quasi-periodic disturbances of the web tension due to the eccentricity and the non-circularity of the reel and rolls. The disturbances induced by the non-circularity ...Winding and web transport systems are subjected to quasi-periodic disturbances of the web tension due to the eccentricity and the non-circularity of the reel and rolls. The disturbances induced by the non-circularity and eccentricity of the rolls are quasi-periodic with a frequency that varies with their rotation speed. An adaptive method of rejection of these disturbances is proposed in this paper. It is based on a phase-locked loop structure that estimates simutaneously the phase and magnitude of the perturbation and then cancels it. This algorithm can be plugged in an existing industrial controller. The stability and robustness of the algorithm are also discussed. The ability of the algorithm to reject quasi-periodic disturbances with slowly varying frequencies is shown through simulation results.展开更多
文摘Real-time anomaly detection of massive data streams is an important research topic nowadays due to the fact that a lot of data is generated in continuous temporal processes. There is a broad research area, covering mathematical, statistical, information theory methodologies for anomaly detection. It addresses various problems in a lot of domains such as health, education, finance, government, etc. In this paper, we analyze the state-of-the-art of data streams anomaly detection techniques and algorithms for anomaly detection in data streams (time series data). Critically surveying the techniques’ performances under the challenge of real-time anomaly detection of massive high-velocity streams, we conclude that the modeling of the normal behavior of the stream is a suitable approach. We evaluate Holt-Winters (HW), Taylor’s Double Holt-Winters (TDHW), Hierarchical temporal memory (HTM), Moving Average (MA), Autoregressive integrated moving average (ARIMA) forecasting models, etc. Holt-Winters (HW) and Taylor’s Double Holt-Winters (TDHW) forecasting models are used to predict the normal behavior of the periodic streams, and to detect anomalies when the deviations of observed and predicted values exceeded some predefined measures. In this work, we propose an enhancement of this approach and give a short description about the algorithms and then they are categorized by type of pre-diction as: predictive and non-predictive algorithms. We implement the Genetic Algorithm (GA) to periodically optimize HW and TDHW smoothing parameters in addition to the two sliding windows parameters that improve Hyndman’s MASE measure of deviation, and value of the threshold parameter that defines no anomaly confidence interval [1]. We also propose a new optimization function based on the input training datasets with the annotated anomaly intervals, in order to detect the right anomalies and minimize the number of false ones. The proposed method is evaluated on the known anomaly detection benchmarks NUMENTA and Yahoo datasets with annotated anomalies and real log data generated by the National education information system (NEIS)1 in Macedonia.
基金Projects Supported by the National Natural Science Foundation of China
文摘In recent years, there is a wide interest in Sarkovskii's theorem ami the related study. According to Sarkovskii's theoren if the continuous self-mapf of the closed interval has a 3-pcriodic orbit, then fmust has an n-pcriodic orbit for any positive integer n. But f can not has all n-periodic orbits for some n.For example, letEvidently, f has only one kind of 3-periodic orbit in the two kinds of 3-periodic orbits. This explains that it isn't far enough to uncover the relation between periodic orbits by information which Sarkovskii's theorem has offered. In this paper, we raise the concept of type of periodic orbits, and give a feasible algorithm which decides the relation of implication between two periodic orbits.
文摘In this paper Genetic Algorithm has been integrated with Fouquet modal analysis to optimize radiation pattern of coupled periodic antenna. Floquet analysis is used with MoM-GEC (Moment-Generalized Equivalent Circuit) method to study a finite periodic array with uniform amplitude and linear phase distribution. This method is very advantageous for studying large antenna array since it considerably reduces the computation time and the number of operations. In this way, Genetic algorithm is introduced and combined with Floquet analysis to optimize the radiation pattern distribution of this coupled periodic antenna. The goal of the optimization is to provide a better radiation characteristic for the coupled periodic antenna with maximum side lobe level reduction.
文摘Winding and web transport systems are subjected to quasi-periodic disturbances of the web tension due to the eccentricity and the non-circularity of the reel and rolls. The disturbances induced by the non-circularity and eccentricity of the rolls are quasi-periodic with a frequency that varies with their rotation speed. An adaptive method of rejection of these disturbances is proposed in this paper. It is based on a phase-locked loop structure that estimates simutaneously the phase and magnitude of the perturbation and then cancels it. This algorithm can be plugged in an existing industrial controller. The stability and robustness of the algorithm are also discussed. The ability of the algorithm to reject quasi-periodic disturbances with slowly varying frequencies is shown through simulation results.
