Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userLow Power Unsupervised Anomaly Detection by Non-Parametric Modeling of Sensor Statistics
89
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
This work presents AEGIS, a novel mixed-signal framework for real-time anomaly detection by examining sensor stream statistics. AEGIS utilizes Kernel Density Estimation (KDE)-based non-parametric density estimation to generate a real-time statistical model of the sensor data stream. The likelihood estimate of the sensor data point can be obtained based on the generated statistical model to detect outliers. We present CMOS Gilbert Gaussian cell-based design to realize Gaussian kernels for KDE. For outlier detection, the decision boundary is defined in terms of kernel standard deviation ($sigma_{Kernel}$) and likelihood threshold ($P_{Thres}$). We adopt a sliding window to update the detection model in real-time. We use time-series dataset provided from Yahoo to benchmark the performance of AEGIS. A f1-score higher than 0.87 is achieved by optimizing parameters such as length of the sliding window and decision thresholds which are programmable in AEGIS. Discussed architecture is designed using 45nm technology node and our approach on average consumes $sim$75 $mu$W power at a sampling rate of 2 MHz while using ten recent inlier samples for density estimation. textcolor{red}{Full-version of this research has been published at IEEE TVLSI}
rate research
Read More
Coherent groups of generators, i.e., machines with perfectly correlated rotor angles, play an important role in power system stability analysis. This paper introduces a real-time methodology based on hierarchical clustering techniques for discovering the degree of coherency among generators using the synchronization coefficient and the correlation coefficient of the generators rotor angle as the coherency index. Furthermore, the Power Transient Stability Indices (PTSI) were employed to examine the versatile response of the power system. The method uses power systems transients Stability indices, i.e., power Connectivity Factor (CF) index which presents coherently strong generators within the groups, the power Separation Factor (SF) index which unveils to the extent that the generators in different groups tend to swing against the other groups in the event of a disturbance, and the overall system separation index which demonstrates the overall system separation status (CF/SF). The approach is assessed on an IEEE-39 test system with a fully dynamic model. The simulation results presented in this paper demonstrate the efficiency of the proposed approach.
Non-stationary forced oscillations (FOs) have been observed in power system operations. However, most detection methods assume that the frequency of FOs is stationary. In this paper, we present a methodology for the analysis of non-stationary FOs. Firstly, Fourier synchrosqueezing transform (FSST) is used to provide a concentrated time-frequency representation of the signals that allows identification and retrieval of non-stationary signal components. To continue, the Dissipating Energy Flow (DEF) method is applied to the extracted components to locate the source of forced oscillations. The methodology is tested using simulated as well as real PMU data. The results show that the proposed FSST-based signal decomposition provides a systematic framework for the application of DEF Method to non-stationary FOs.
On-line detection of anomalies in time series is a key technique used in various event-sensitive scenarios such as robotic system monitoring, smart sensor networks and data center security. However, the increasing diversity of data sources and the variety of demands make this task more challenging than ever. Firstly, the rapid increase in unlabeled data means supervised learning is becoming less suitable in many cases. Secondly, a large portion of time series data have complex seasonality features. Thirdly, on-line anomaly detection needs to be fast and reliable. In light of this, we have developed a prediction-driven, unsupervised anomaly detection scheme, which adopts a backbone model combining the decomposition and the inference of time series data. Further, we propose a novel metric, Local Trend Inconsistency (LTI), and an efficient detection algorithm that computes LTI in a real-time manner and scores each data point robustly in terms of its probability of being anomalous. We have conducted extensive experimentation to evaluate our algorithm with several datasets from both public repositories and production environments. The experimental results show that our scheme outperforms existing representative anomaly detection algorithms in terms of the commonly used metric, Area Under Curve (AUC), while achieving the desired efficiency.
Existing coordinated cyber-attack detection methods have low detection accuracy and efficiency and poor generalization ability due to difficulties dealing with unbalanced attack data samples, high data dimensionality, and noisy data sets. This paper proposes a model for cyber and physical data fusion using a data link for detecting attacks on a Cyber-Physical Power System (CPPS). Two-step principal component analysis (PCA) is used for classifying the systems operating status. An adaptive synthetic sampling algorithm is used to reduce the imbalance in the categories samples. The loss function is improved according to the feature intensity difference of the attack event, and an integrated classifier is established using a classification algorithm based on the cost-sensitive gradient boosting decision tree (CS-GBDT). The simulation results show that the proposed method provides higher accuracy, recall, and F-Score than comparable algorithms.
Random projection is a common technique for designing algorithms in a variety of areas, including information retrieval, compressive sensing and measuring of outlyingness. In this work, the original random projection outlyingness measure is modified and associated with a neural network to obtain an unsupervised anomaly detection method able to handle multimodal normality. Theoretical and experimental arguments are presented to justify the choice of the anomaly score estimator. The performance of the proposed neural network approach is comparable to a state-of-the-art anomaly detection method. Experiments conducted on the MNIST, Fashion-MNIST and CIFAR-10 datasets show the relevance of the proposed approach.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
