Early anomaly detection and localisation in distribution network: a data‐driven approach

This paper proposes a novel distribution line parameter estimation method, driven by the probabilistic data fusion of the distributed phasor measurement unit (D-PMU) and the advanced measurement infrastructure. The synchronized and high-precision D-PMU is utilized to tackle the challenge risen by the a-synchronization of smart meters. Correspondingly, a time-alignment algorithm is proposed to obtain the time-synchronous error (TSE) dataset for the up-stream smart meter. The non-parametric estimation method is performed then to evaluate the probabilistic density curve of TSE. Furthermore, TSE data of down-stream smart meters are generated by implementing the acceptance-rejection process based on the obtained probabilistic density curve. Leveraging the generated TSE dataset, a new time-shifted D-PMU curve is probabilistically aligned or fused with the down-stream advanced measurement infrastructure curves. According to the complete voltage drop model, the line parameter estimation of resistance and reactance is formulated as a quadratic programming problem and solved by Optimal Toolbox in MATLAB by conducting multi-run Monte-Carlo simulations under various scenarios. Simulation results demonstrate the effectiveness and robustness of the proposed methodology.

Section: Motivation and Incitementmentioning

confidence: 99%

Distribution line parameter estimation driven by probabilistic data fusion of D‐PMU and AMI

Xiao

Xie

Fang

et al. 2021

“…First, the character vectors in attack features are transformed into numeric vectors [23]. For continuous numerical features, the measurement methods of each feature are different, so the features are normalized in Equation (7) to avoid the impact on attack detection due to the difference of feature measurements.…”

Section: A Pso-bpnn Attack Detection Model In the Cyber Layermentioning

confidence: 99%

“…Cyber-attack has become a factor threatening the safe operation of the DCPS [6,7]. For instance, a substation in Ukrainian was attacked by the BalckEnergy3 in 2015 [8] and European industrial manufacturing systems were attacked by the Havex Trojan in 2014 [9].…”

Section: Introductionmentioning

confidence: 99%

A cyber‐attack detection method for load control system based on cyber and physical layer crosscheck mechanism

Shi

et al. 2021

Cyber‐attack has become one of the main threats to the safe operation of the load control system in the distribution network. Currently, a large number of methods are developed to detect cyber‐attacks. However, most of these methods make decisions based on information from a single cyber layer or physical layer. If the decision is wrong, there is no way to verify it. Therefore, if the separate decisions can be made in the cyber layer and physical layer at the same time, and a final judgment is received throughout the two decisions, then the accuracy and credibility of cyber‐attack detection can be improved. First, the particle swarm optimization‐based backpropagation neural network is adopted to detect attacks in the cyber layer, at the same time, the criteria importance though intercrieria correlation‐based grey relational analysis is adopted for the assessment of node risk level in the physical layer. Then, based on the incidence matrix model between the cyber layer and the physical layer, a crosscheck mechanism is proposed to detect the cyber‐attack through the verification of the two decisions. Finally, taking the IEEE 33‐node system as an example, the validity of the proposed method to detect the denial of service attack is verified.

“…In [13], an ensemblebased unsupervised approach is proposed to identify outliers based on the anomaly score detected by three different base detectors (Chebyshev-based, DBSCAN-based, and Regressionbased detectors). Other techniques such as dynamic time warping classifier [14], multi-class SVM [15,16], Bayesian Network [17], and random matrix theory [18][19][20] are also used to address the anomaly detection problem.…”

Section: Introductionmentioning

confidence: 99%

Convolutional autoencoder anomaly detection and classification based on distribution PMU measurements

Ehsani

Aminifar

Mohsenian‐Rad

2022

The huge volume of data that is streamed from distribution phasor measurement units (DPMU) toward distribution management system (DMS) can rarely be used in the raw format. Data‐driven analysis to extract information out of massive raw data has revealed promising opportunities to overcome this challenge. This paper utilizes convolutional autoencoders (Conv‐AE) for the sake of anomaly detection based on the DPMU measurements in distribution systems. The Conv‐AE is unsupervised and independent of the event type. It dispenses preprocessing and feature extraction along with obtaining the essential information directly from the measured data. The anomaly detection is followed by a supervised classifier which is developed to identify the type of anomaly. This classifier is a convolutional neural network that is designed and fine‐tuned for the problem at hand. It uses an ensemble learning method to augment the fully‐labelled dataset. Performance of the proposed methodology is evaluated on modified IEEE 34‐node and IEEE 123‐node test feeders in various operating conditions, presence of noise, and different scenarios of DPMU outage. Moreover, a real‐world DPMU dataset is utilized to evaluate performance of the Conv‐AE model in practical conditions. Results confirmed effectiveness of the proposed technique to be used in future DMS platforms.