A5071896704- Electricity Theft Detection Techniques
- Imbalanced Data Classification Techniques
- Electric Power System Optimization
- Power Systems and Renewable Energy
- Currency Recognition and Detection
- Optimal Power Flow Distribution
- Microgrid Control and Optimization
- Vehicle License Plate Recognition
- Power Systems and Technologies
- Aerogels and thermal insulation
- Electrical Fault Detection and Protection
- HVDC Systems and Fault Protection
- Thermal Analysis in Power Transmission
- Advanced Data and IoT Technologies
- High-Voltage Power Transmission Systems
- Frequency Control in Power Systems
- Advanced Cellulose Research Studies
- Power System Reliability and Maintenance
- Water Systems and Optimization
- Smart Grid Energy Management
- Smart Grid Security and Resilience
- Energy Load and Power Forecasting
University of Technology Malaysia
2020-2022
Pusan National University
2019
This paper presents a novel supervised machine learning-based electric theft detection approach using the feature engineered-CatBoost algorithm in conjunction with SMOTETomek algorithm. Contrary to previous literature, where missing observations data are either ignored or imputed average values, this work utilizes k-Nearest neighbor technique for imputation; thus, an accurate and realistic estimation of is achieved. To mitigate biasness majority class, proposed model algorithm, which...
Abstract This paper presents a novel, sequentially executed supervised machine learning‐based electric theft detection framework using Jaya‐optimized combined Kernel and Tree Boosting (KTBoost) classifier. It utilizes the intelligence of XGBoost algorithm to estimate missing values in acquired dataset during data pre‐processing phase. An oversampling based on Robust‐SMOTE technique is utilized avoid unbalanced class distribution issue. Afterward, with aid few very significant statistical,...
This paper presents a novel data-oriented unsupervised machine learning-based theft detection approach for efficiently identifying the fraudster consumers. It accomplishes above-mentioned objective by exploiting intelligence of robust principal component analysis (ROBPCA) algorithm in conjunction with outlier removal clustering (ORC) algorithm. To avoid irregularities acquired consumers' data from power utility, statistical features are extracted each consumer's consumption patterns using an...
This study presents a novel feature-engineered-natural gradient descent ensemble-boosting (NGBoost) machine-learning framework for detecting fraud in power consumption data. The proposed was sequentially executed three stages: data pre-processing, feature engineering, and model evaluation. It utilized the random forest algorithm-based imputation technique initially to impute missing entries acquired smart meter dataset. In second phase, majority weighted minority oversampling (MWMOTE)...
This paper proposes a two-stage planning approach for the efficient utilization of distributed generation (DG) and capacitor banks (CBs) simultaneous grid-connected islanded operations balanced distribution networks. The first stage determines optimal installation locations capacities DGs CBs using an improved variant Jaya optimization algorithm (IJaya) to minimize power losses voltage deviation during operation network. second identifies operating point DG CB combination carry maximum load...
The active and reactive power can be controlled exchanged in a High-Voltage Direct Current (HVDC) transmission system using Modular-Multi level converter (MMC). HVDC based upon MMC improves exchange capability, transient stability, oscillation damping. challenging task of is controlling its output current, additive currents, internal energy, sum capacitor voltages. A control design structure for the modular-multilevel presented this paper. current decomposed into dq-axis components park...
This paper proposes parameter-free two best-worst optimizers (BWOs) that combine the searching capabilities of Jaya and Rao-1 algorithms to intensify their exploration exploitation capabilities. For proposed first optimizer, BWO-1, parallel taxonomy has been adopted for combination obtain dual sets updated solutions a given current solution set. second BWO-2, along with taxonomy, subloop random scaling factors introduced in updating mechanism both standard generate multiple single iteration....
Available transfer capability (ATC) is an important metric used to measure the techno-economic viability of transmission networks. Several methods have been presented in literature for ATC assessment, however, only some few articles incorporate CBM calculation and those papers considered conventional power generation sources evaluation. a function reliability generating units. This paper presents inter-area presence wind energy source using graph theory technique results are incorporated...
From lightening a house to functionalizing industry energy plays significant role. Generating is not the only issue but generating and then utilizing it appropriately main issue. 2011, Pakistan has shifted its sight on renewable cut power starved national grid’s dependency from fossil fuel plants. Since number of solar parks wind plants have been initiated few them are already operational. Focusing just energy, by August 2016 total six fully functional capable producing 300 MW generation as...