Denoising and Bad Data Detection in Distribution Phasor Measurements using Filtering, Clustering and Koopman Mode Analysis

Distribution-level phasor measurement units (D-PMU) data are prone to different types of anomalies given complex data flow and processing infrastructure in an active power distribution system with enhanced digital automation. It is essential to pre-process the data before being used by critical applications for situational awareness and control. In this work, two approaches for detection of data anomalies are introduced for offline (larger data processing window) and online (shorter data processing window) applications. A margin-based maximum likelihood estimator (MB-MLE) method is developed to detect anomalies by integrating the results of different base detectors including Hampel filter, Quartile detector and DBSCAN. A smoothing wavelet denoising method is used to remove high-frequency noises. The processed data with offline analysis is used to fit a model to the underlying dynamics of synchrophasor data using Koopman Mode Analysis, which is subsequently employed for online denoising and bad data detection (BDD) using Kalman Filter (KF). The parameters of the KF are adjusted adaptively based on similarity to the training data set for model fitting purposes. Developed techniques have been validated for the modified IEEE test system with multiple D-PMUs, modeled and simulated in real-time for different case scenarios using the OPAL-RT Hardware-In-the-Loop (HIL) Simulator.

Denoising and Bad Data Detection in Distribution Phasor Measurements using Filtering, Clustering and Koopman Mode Analysis

Distribution-level phasor measurement units (D-PMU) data are prone to different types of anomalies given complex data flow and processing infrastructure in an active power distribution system with enhanced digital automation. It is essential to pre-process the data before being used by critical applications for situational awareness and control. In this work, two approaches for detection of data anomalies are introduced for offline (larger data processing window) and online (shorter data processing window) applications. A margin-based maximum likelihood estimator (MB-MLE) method is developed to detect anomalies by integrating the results of different base detectors including Hampel filter, Quartile detector and DBSCAN. A smoothing wavelet denoising method is used to remove high-frequency noises. The processed data with offline analysis is used to fit a model to the underlying dynamics of synchrophasor data using Koopman Mode Analysis, which is subsequently employed for online denoising and bad data detection (BDD) using Kalman Filter (KF). The parameters of the KF are adjusted adaptively based on similarity to the training data set for model fitting purposes. Developed techniques have been validated for the modified IEEE test system with multiple D-PMUs, modeled and simulated in real-time for different case scenarios using the OPAL-RT Hardware-In-the-Loop (HIL) Simulator.

PSO Supported Ensemble Algorithm for Bad Data Detection Against Intelligent Hacking Algorithm

Power system cybersecurity has recently become important due to cyber-attacks. Due to advanced computer science and machine learning (ML) applications being used by malicious attackers, cybersecurity is becoming crucial to creating sustainable, reliable, efficient, and well-protected cyber-systems. Power system operators are needed to develop sophisticated detection mechanisms. In this study, a novel machine-learning-based detection algorithm that combines the five most popular ML algorithms with Particle Swarm Optimizer (PSO) is developed and tested by using an intelligent hacking algorithm that is specially developed to measure the effectiveness of this study. The hacking algorithm provides three different types of injections: random, continuous random, and slow injections by adaptive manner. This would make detection harder. Results shows that recall values with the proposed algorithm for each different type of attack have been increased.

Stacked Autoencoder Framework of False Data Injection Attack Detection in Smart Grid

The advanced communication technology provides new monitoring and control strategies for smart grids. However, the application of information technology also increases the risk of malicious attacks. False data injection (FDI) is one kind of cyber attacks, which cannot be detected by bad data detection in state estimation. In this paper, a data-driven FDI attack detection framework of the smart grid with phasor measurement units (PMUs) is proposed. To enhance the detecting accuracy and efficiency, the multiple layer autoencoder algorithm is applied to abstract the hidden features of PMU measurements layer by layer in an unsupervised manner. Then, the features of the measurements and corresponding labels are taken as inputs to learn a softmax layer. Last, the autoencoder and softmax layer are stacked to form a FDI detection framework. The proposed method is applied on the IEEE 39-bus system, and the simulation results show that the FDI attacks can be detected with higher accuracy and computational efficiency compared with other artificial intelligence algorithms.

False Data Detection in Smart Grid using Artificial Intelligence

In order to increase the utilization of artificial intelligence in smart grids, it is necessary to have an accurate state estimation. This criterion is an essential aspect, along with other functionalities for successful control and monitoring. As the internet and utility network form an increasing interconnectivity, it leaves the state estimators in a state of vulnerability to various attacks like bad data detection and false data injection. Though there are many research-works done on detectors for false data detection, depending on the contingencies, the counter measure will also vary. A sudden change physically will have a high impact on the available data, resulting in incorrect classification of the future instances. As a means of addressing this issue, we have analyzed the differences between data manipulation change and physical grid change for better understanding. Focusing on distribution change, we used outage and have introduced analysis of historical data. The goal is to determine the important aspects thereby identifying the scope. We have also used statistical hypothesis and dimensionality reduction for testing purpose. We have used IEEE 14 bus system for evaluation based on the scenario of attack: under concept drift and without concept drift. The result shows a more accurate output when compared with the other previously existing methodologies using concept drift.

IoT based Smart Grid Attack Resistance in AC and DC State Estimation

Use of automation and intelligence in smart grids has led to implementation in a number of applications. When internet of things is incorporated it will result in the significant improvement a number of factors such as fault recovery, energy delivery efficiency, demand response and reliability. However, the collaboration of internet of things and smart grid gives rise to a number of security issues and threats. This is especially the case when using internet based protocols and public communication infrastructure. To address these issues we should ensure that the data stored is secure and critical information from the data is extracted in a careful manner. If any threat to its security is detective an early blackout warning should be issued immediately. In this paper we have proposed a geometric view point for big data attacks which is capable of bypassing bad data detection. We have created an environment where replay scheme is used launch blind energy big data attack. The defence mechanism of our proposed work is studied and found to be efficient. Experimental evidence supports our theory and we have found our methodology to efficiently improve error detection rate.