A deep learning technique Alexnet to detect electricity theft in smart grids DOI Creative Commons
Nitasha Khan, Muhammad Amir Raza,

Darakhshan Ara

и другие.

Frontiers in Energy Research, Год журнала: 2023, Номер 11

Опубликована: Ноя. 8, 2023

Electricity theft (ET), which endangers public safety, creates a problem with the regular operation of grid infrastructure and increases revenue losses. Numerous machine learning, deep mathematical-based algorithms are available to find ET. Still, these models do not produce best results due problems like dimensionality curse, class imbalance, improper hyper-parameter tuning learning models, etc. We present hybrid model for effectively detecting electricity thieves in smart grids while considering abovementioned concerns. Pre-processing techniques first employed clean up data from meters. Then, feature extraction technique, AlexNet, addresses curse dimensionality. The effectiveness proposed method is evaluated through simulations using real dataset Chinese intelligent To conduct comparative analysis, various benchmark implemented as well. Our achieves accuracy, precision, recall, F1, 86%, 89%, 84%, respectively.

Язык: Английский

Modeling of intelligent controllers for solar photovoltaic system under varying irradiation conditions DOI Creative Commons

Malhar Khan,

Muhammad Amir Raza, Touqeer Ahmed Jumani

и другие.

Frontiers in Energy Research, Год журнала: 2023, Номер 11

Опубликована: Ноя. 2, 2023

The increasing demand for solar renewable energy resources, driven by the global crisis and depletion of conventional sources, has underscored importance harnessing energy. Solar photovoltaic (PV) systems, however, exhibit nonlinear output power due to their weather-dependent nature, impacting overall system efficiency. This study focuses on development comparative analysis three intelligent Maximum Power Point Tracking (MPPT) controllers using MATLAB Simulink. employ distinct methodologies, namely, Artificial Neural Networks (ANN), Adaptive Fuzzy Inference System (ANFIS), Logic Controller (FLC). results demonstrate that ANFIS achieved highest accuracy at 99.50%, surpassing ANN FLC with accuracies 97.04% 98.50%, respectively, thus establishing as superior MPPT controller. Additionally, positives negatives all MPPT-based algorithms are also compared in this work.

Язык: Английский

Процитировано

15

An Intelligent Frequency Control Scheme for Inverting Station in High Voltage Direct Current Transmission System DOI Creative Commons

Saleem Saleem,

Muhammad Amir Raza,

Syed Waqar Umer

и другие.

Engineering Reports, Год журнала: 2025, Номер 7(1)

Опубликована: Янв. 1, 2025

ABSTRACT Power system stability is crucial for the reliable and efficient operation of electrical grids. One key factors affecting power frequency alternating current (AC) while connected with High Voltage Direct Current (HVDC) transmission system. Changes in load demand can lead to deviations, which have detrimental effects on performance Frequency should therefore be controlled within predefined limits order prevent unexpected disturbances that may cause problems loads or even entire fail. A broad simulation model HVDC developed using MATLAB software evaluate effectiveness proposed controllers such as Adaptive Neuro‐Fuzzy Inference System (ANFIS), Artificial Neural Network (ANN), optimization Proportional‐Integral‐Derivative (PID) controller Particle Swarm Optimization (PSO) based control strategy addressing instability problems. To assess how well ANFIS, ANN, PID‐PSO controls system, several situations were simulated, including changes operational circumstances. The result reveals ANN performs more accurate results than other and, displaying its capacity successfully reduce deviations maintained a 50 Hz. Adopted method suggested easy integration AC grid enhances quality stability.

Язык: Английский

Процитировано

0

The impact and carbon reduction effect on biomass energy in power system DOI
Hongli Liu, Han Zhang, Yumin Chen

и другие.

Biomass Conversion and Biorefinery, Год журнала: 2025, Номер unknown

Опубликована: Фев. 8, 2025

Язык: Английский

Процитировано

0

PINN and KAN temperature prediction of carbon Fiber/Epoxy composite materials irradiated by nuclear environment simulated intense heat fluxes DOI
Xiaoxiang Han, Jun Li, Lin Yuan

и другие.

Annals of Nuclear Energy, Год журнала: 2025, Номер unknown, С. 111454 - 111454

Опубликована: Апрель 1, 2025

Язык: Английский

Процитировано

0

A bibliometric analysis of research trends in electric vehicle power electronics: global perspectives and future directions DOI Creative Commons

Gaddala Anusha,

A. V. V. Sudhakar,

Ram Deshmukh

и другие.

Deleted Journal, Год журнала: 2025, Номер 7(5)

Опубликована: Май 13, 2025

Язык: Английский

Процитировано

0

Demand side management through energy efficiency measures for the sustainable energy future of Pakistan DOI Creative Commons
Arshad Chughtai, Mohammad Aslam Uqaili, Nayyar Hussain Mirjat

и другие.

Heliyon, Год журнала: 2024, Номер 10(15), С. e34798 - e34798

Опубликована: Июль 21, 2024

Pakistan is facing energy crises due to localized shortages, market manipulation, infrastructure disruption, rising demand, governance issues, climate and geopolitical events. In this situation Demand Side Management (DSM) a promising solution overcome the problem of crises. DSM strategy helps manage consumer demand through conservation rather than addition new power capacity. study, Low Emissions Analysis Platform (LEAP) develops an model for period 2022-2050. Three scenarios has been constructed namely Baseline (BAS), Green Energy Policy (GEP), Efficiency (ENE) predict future production, carbon emissions investment cost which covers capital, operational maintenance costs. The results suggest that targets should be achieved implementation ENE scenario. Predicted production consumption under scenario are substantially less those BAS country can meet its 635.83,000 GWh with 747.15,000 production. Non-renewable sources produce 171.27,000 GWh, whilst renewable 575.88,000 GWh. According scenario, by 2050, CO

Язык: Английский

Процитировано

2

Spatial-temporal evolution characteristics and driving factors analysis of regional energy supply and demand in China DOI Creative Commons
Weijun He, Jingyi Sun, Min An

и другие.

Energy Strategy Reviews, Год журнала: 2024, Номер 55, С. 101542 - 101542

Опубликована: Сен. 1, 2024

Язык: Английский

Процитировано

2

Demand-side management scenario analysis for the energy-efficient future of Pakistan: Bridging the gap between market interests and national priorities DOI Creative Commons
Arshad Chughtai, Mohammad Aslam Uqaili, Nayyar Hussain Mirjat

и другие.

Frontiers in Energy Research, Год журнала: 2024, Номер 12

Опубликована: Сен. 17, 2024

Pakistan is facing the worst level of energy and economic crisis its history. The underlying reason basically due to crisis. Various approaches have been adopted tackle crises which country for over 2 decades. Demand-side management (DSM) most potent cost-effective option redress crisis, which, unfortunately, has neglected strategy in Pakistan. DSM potential save up 10.0%–15.0% primary ensure country’s future security. Laws codes do exist, but ironically, vital element DSM’s policy framework implementation mechanism missing. Hence, main objective this research develop a model analyze reference scenarios. Low-emission analysis platform (LEAP) used Pakistan’s study period 2021–2050. three alternative scenarios developed include efficiency (EEF), conservation (EC), load (LOM), they are all analyzed. results estimate electricity demand forecast 1009.8 TWh under scenario 2050, whereas tends result 26.38% decreased compared scenario. also outperform In EC reduces consumption by 178.0 GHG emissions 19.20 million metric tons, EEF 110.30 10.04 LOM suggests reduced 101.0 6.20 tons. This concluded that must be institutionalized building robust regulatory execution at government utility levels.

Язык: Английский

Процитировано

1

A deep learning technique Alexnet to detect electricity theft in smart grids DOI Creative Commons
Nitasha Khan, Muhammad Amir Raza,

Darakhshan Ara

и другие.

Frontiers in Energy Research, Год журнала: 2023, Номер 11

Опубликована: Ноя. 8, 2023

Electricity theft (ET), which endangers public safety, creates a problem with the regular operation of grid infrastructure and increases revenue losses. Numerous machine learning, deep mathematical-based algorithms are available to find ET. Still, these models do not produce best results due problems like dimensionality curse, class imbalance, improper hyper-parameter tuning learning models, etc. We present hybrid model for effectively detecting electricity thieves in smart grids while considering abovementioned concerns. Pre-processing techniques first employed clean up data from meters. Then, feature extraction technique, AlexNet, addresses curse dimensionality. The effectiveness proposed method is evaluated through simulations using real dataset Chinese intelligent To conduct comparative analysis, various benchmark implemented as well. Our achieves accuracy, precision, recall, F1, 86%, 89%, 84%, respectively.

Язык: Английский

Процитировано

3