iEVEM: Big Data-Empowered Framework for Intelligent Electric Vehicle Energy Management

Systems, Год журнала: 2025, Номер 13(2), С. 118 - 118

Опубликована: Фев. 13, 2025

Recent years have witnessed an unprecedented boom of Electric Vehicles (EVs). However, EVs’ further development confronts critical bottlenecks due to EV Energy (EVE) issues like battery hazards, range anxiety, and charging inefficiency. Emerging data-driven EVE Management (EVEM) is a promising solution but still faces fundamental challenges, especially in terms reliability efficiency. This article presents iEVEM, the first big data-empowered intelligent EVEM framework, providing systematic support essential driver-, enterprise-, social-level applications. Particularly, layered data architecture from heterogeneous management knowledge-enhanced design provided, edge–cloud collaborative for networked system proposed reliable efficient EVEM, respectively. We conducted proof-of-concept case study on typical task (i.e., energy consumption outlier detection) using real driving 4000+ EVs within three months. The experimental results show that iEVEM achieves significant boost efficiency up 47.48% higher detection accuracy at least 3.07× faster response speed compared with state-of-art approaches). As expected inspire more applications exploiting skyrocketing data.

Язык: Английский

---

SolarNexus: A deep learning framework for adaptive photovoltaic power generation forecasting and scalable management

Applied Energy, Год журнала: 2025, Номер 391, С. 125848 - 125848

Опубликована: Апрель 11, 2025

Язык: Английский

---

BWO–ICEEMDAN–iTransformer: A Short-Term Load Forecasting Model for Power Systems with Parameter Optimization

Algorithms, Год журнала: 2025, Номер 18(5), С. 243 - 243

Опубликована: Апрель 24, 2025

Maintaining the equilibrium between electricity supply and demand remains a central concern in power systems. A response program can adjust load from side to promote balance of demand. Load forecasting facilitate implementation this program. However, as consumption patterns become more diverse, resulting data grows increasingly irregular, making precise difficult. Therefore, paper developed specialized scheme. First, parameters improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) were optimized using beluga whale optimization (BWO). Then, nonlinear decomposed into multiple subsequences ICEEMDAN. Finally, each subsequence was independently predicted iTransformer model, overall forecast derived by integrating these individual predictions. Data Singapore selected for validation. The results showed that BWO–ICEEMDAN–iTransformer model outperformed other comparison models, an R2 0.9873, RMSE 48.0014, MAE 66.2221.

Язык: Английский

---

Revolving Gate Fourier Transform (RGFT): A novel time-frequency model for forecasting energy efficiency and CO2 neutrality

Energy Conversion and Management, Год журнала: 2025, Номер 341, С. 120020 - 120020

Опубликована: Июнь 4, 2025

Язык: Английский

---

Smart grid stability prediction model using two-way attention based hybrid deep learning and MPSO

Energy Exploration & Exploitation, Год журнала: 2024, Номер unknown

Опубликована: Авг. 19, 2024

In smart grid management, precise stability prediction is a complicated task that adds to the effective allocation of resources with stability. Specifically, demand-side management considered an essential element overall Smart Grids system. Hence, predicting future energy demands crucial regulating consumption by aligning utility offerings consumer demand. This research presents hybrid deep learning model (Convolutional Neural Network [CNN] Bi-LSTM) two-way attention method and multi-objective particle swarm optimization (MPSO) for short-term load from grid. The proposed utilizes at its encoding decoding stages, in which layer helps recognize all features input vector, resolve fixed context vector problem offering better memory capacity. A CNN Bi-LSTM are used capture dataset. We also utilize t-Nearest Neighbours algorithm pre-process initial An MPSO combines methods, resulting accuracy. As far as we know, it first work suggest dynamic considers different significant enables outcomes. performance existing well-known models such Recurrent Network, Gated Unit, Long Short-Term Memory (LSTM), Time Series Transformer, CNN-LSTM various measuring parameters MAE, MSE, MAPE RMSE calculated on online UCI dataset (Electrical Grid Stability Simulated Dataset). achieved result, proves efficiency model.

Язык: Английский

---

Metaheuristic Algorithm‐Based Optimal Energy Operation Scheduling and Energy System Sizing Scheme for PV‐ESS Integrated Systems in South Korea

International Journal of Energy Research, Год журнала: 2024, Номер 2024(1)

Опубликована: Янв. 1, 2024

To efficiently utilize the power generated by a photovoltaic (PV) system, integrating it with an energy storage system (ESS) is essential. Furthermore, maximizing economic benefits of such PV‐ESS integrated systems requires selecting optimal capacity and performing operation scheduling. Although many studies rely on rule‐based scheduling, these methods prove inadequate for complex real‐world scenarios. Moreover, they often focus solely determining ESS to integrate into existing PV systems, thereby limiting possibility achieving benefits. address this issue, we propose scheduling sizing scheme based metaheuristic algorithms. The proposed employs zero‐shot forecasting model estimate potential generation from planned system. A systematic analysis installation, operation, maintenance costs then incorporated analysis. We conducted extensive experiments comparing various capacities using real electrical load data collected private university in South Korea estimated data. According results, most effective algorithm simulated annealing (SA). Additionally, battery, conversion are 13,000 kW each, 10% capacity, 60% battery respectively. annual electricity tariff calculated used experiment $3,315,484. In contrast, SA‐based achieved $875,000, improvement approximately 7% over $817,730.

Язык: Английский

---

Designing adaptive smart buildings: an RNN and Migrating Birds Optimization approach for occupancy prediction

Asian Journal of Civil Engineering, Год журнала: 2023, Номер 25(3), С. 2653 - 2663

Опубликована: Дек. 15, 2023

Язык: Английский

---

Deep Learning Method to Analyze the Bi-LSTM Model for Energy Consumption Forecasting in Smart Cities

Опубликована: Ноя. 15, 2023

Smart cities and IoT solutions are improving urban efficiency, resource optimization, public safety by using modern technologies. Deep residual Bi-LSTM (Long Short-Term Memory) models can analyze forecast complicated time-varying data. This study examines how the deep Bi-LS TM model might improve smart city solutions. The has broad use since it captures long-term interdependence extracts meaningful representations from sequential Traffic prediction, energy consumption forecasting, environmental monitoring, predictive maintenance, safety, emergency response discussed. provides realtime insights, accurate forecasts, quick data processing to systems solutions, making more sustainable, efficient, secure.

Язык: Английский

---

Novel STAttention GraphWaveNet Model for Residential Household Appliance Prediction and Energy Structure Optimization

Energy, Год журнала: 2024, Номер 307, С. 132582 - 132582

Опубликована: Июль 26, 2024

Язык: Английский

---