Intrusion detection system in the Smart Distribution Network: A feature engineering based AE-LightGBM approach

Article Chemistry, Analytical

Intrusion Detection System in the Advanced Metering Infrastructure: A Cross-Layer Feature-Fusion CNN-LSTM-Based Approach

Ruizhe Yao et al.

Summary: The paper proposes a cross-layer feature-fusion AMI intrusion detection model based on CNN and LSTM networks, which comprehensively considers regional features, global features, and periodic features, to more accurately identify intrusion information in AMI.

SENSORS (2021)

Add to Collection

Article Engineering, Electrical & Electronic

Electricity theft detection in low-voltage stations based on similarity measure and DT-KSVM

Xiangyu Kong et al.

Summary: The paper proposes an electricity theft detection method based on similarity measure and decision tree combined K-Nearest Neighbor and support vector machine. Features of power consumption are obtained using feature selection strategy and KFCM algorithm, while 1D-WGAN is used to generate simulated stealing data. Numerical and morphological features are comprehensively considered for preliminary detection of NTL, with DT-KSVM used for secondary detection and identification of suspicious customers. Simulation experiments verify the effectiveness of the method.

INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS (2021)

Add to Collection

Article Engineering, Electrical & Electronic

Electricity theft detection based on stacked sparse denoising autoencoder

Yifan Huang et al.

Summary: This paper presents a method for detecting electricity theft based on autoencoders, using a stacked sparse denoising autoencoder approach. The technique trains on electricity data from honest users, reconstructs inputs, and identifies theft users by setting an appropriate error threshold. Sparsity, noise, and particle swarm optimization are used to improve feature extraction and robustness, with the proposed approach evaluated on a electricity dataset in Fujian, China.

INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS (2021)

Add to Collection

Article Energy & Fuels

A novel feature engineered-CatBoost-based supervised machine learning framework for electricity theft detection

Saddam Hussain et al.

Summary: This paper introduces a novel supervised machine learning-based electric theft detection approach using CatBoost and SMOTETomek algorithms, with k-Nearest neighbor technique for missing data imputation and bias mitigation. The Feature Extraction and Scalable Hypothesis algorithm is utilized to extract relevant features, and the model is trained with CatBoost algorithm, achieving high accuracy and detection rate.

ENERGY REPORTS (2021)

Add to Collection

Article Computer Science, Information Systems

A fast network intrusion detection system using adaptive synthetic oversampling and LightGBM

Jingmei Liu et al.

Summary: The study proposes a network intrusion detection system based on ADASYN oversampling technology and LightGBM, which addresses the issue of imbalanced data by preprocessing and increasing minority samples to enhance detection accuracy.

COMPUTERS & SECURITY (2021)

Add to Collection

Article Computer Science, Theory & Methods

An adaptive synthesis to handle imbalanced big data with deep siamese network for electricity theft detection in smart grids

Nadeem Javaid et al.

Summary: The paper proposes a robust big data analytics technique to address the challenges of electricity theft detection, utilizing adaptive synthesis (ADASYN) to handle imbalanced class problems and integrating convolutional neural network (CNN) and long-short term memory (LSTM) for feature discrimination. Through analysis on real-time smart meters data, the model's effectiveness is validated in terms of high area under the curve, F-1-Score, precision and recall.

JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING (2021)

Add to Collection

Article Energy & Fuels