Optimal Scheduling of Biomass-Hybrid Microgrids with Energy Storage: An LSTM-PMOEVO Framework for Uncertain Environments

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(5), P. 2702 - 2702

Published: March 3, 2025

The microgrid is a small-scale, independent power system that plays crucial role in the transition to carbon-neutral energy systems. Combined heat and (CHP) systems with storage reduce waste within microgrids, enhancing utilization efficiency. key challenge for integrated combined determining optimal configuration operation duration under different scenarios meet users’ electricity demands while minimizing both economic environmental costs. Thus, this paper presents bi-objective mathematical model solve scheduling problem of microgrid. Long Short-Term Memory–Parallel Multi-Objective Energy Valley Optimizer (LSTM-PMOEVO) framework incorporates load prediction using LSTM planning solved via PMOEVO. These strategies address challenges posed by unpredictable fluctuations complexity solving such Finally, public dataset was utilized experiments verify performance proposed algorithm. Comparisons discussions show optimization significantly improve PMOEVO, demonstrating marked advantages over six classical algorithms. In conclusion, PMOEVO developed performs excellently Scheduling Problem Biomass-Hybrid microgrids considering uncertainty. work presented provides new solution microgrid-scheduling future research, will be further advanced application real-world scenarios.

Language: Английский

---

A robust energy flow predictor based on CNN-LSTM for prosumer-oriented microgrids considering changes in biogas generation

Energy, Journal Year: 2025, Volume and Issue: unknown, P. 136050 - 136050

Published: April 1, 2025

Language: Английский

---

A hybrid Monte Carlo quantile EMD-LSTM method for satellite in-orbit temperature prediction and data uncertainty quantification

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: 255, P. 124875 - 124875

Published: July 26, 2024

Language: Английский

---

Performance analysis of machine learning based prediction models in assessing optimal operation of microgrid under uncertainty

Results in Control and Optimization, Journal Year: 2024, Volume and Issue: 15, P. 100407 - 100407

Published: March 7, 2024

Of late, the exponential rise in global population is driving higher energy demand. However, rapid depletion of conventional fossil fuels and growing environmental concerns have prompted evolution alternative sources. To this end, Microgrid (MG) with Renewable Energy Sources (RES) has emerged as popular means small-scale localized power grid. planning MG operation pose challenges due to inherent variability stochasticity RES output On account this, present study introduces a Stochastic Management Strategy (SEMS) for grid-connected incorporating Micro-Turbine, Fuel-Cell, RES, Battery Storage, electrical heat The forecasted through hybrid prediction model (sARIMA-GRU) uncertain demand estimated via 'Monte Carlo Simulation.' proposed problem formulated dynamic non-linear stochastic optimization problem. It seeks minimize expected value operational cost satisfying practical constraints. Addressing newly developed 'Artificial Electric Field Algorithm (AEFA)' utilized. Several case studies are performed assess under varied operating conditions. Moreover, analyses impact uncertainty on contribution from DER, grid dependency cost. Comparative analysis reveals that sARIMA-GRU outperforms other contemporary models. noteworthy superior accuracy leads lower costs. statistical convergence confirm proficiency applied AEFA over state-of-the-art Grey Wolf Optimization Firefly solving

Language: Английский

---

Research on optimization of improved short-term load composite forecasting model based on AM–CNN–Bi–LSTM

AIP Advances, Journal Year: 2024, Volume and Issue: 14(5)

Published: May 1, 2024

Accurate load prediction is a prerequisite for the design, operation, scheduling, and management of energy systems. In context development smart grids, extensive integration highly volatile distributed generation into power system has brought new challenges to accuracy, reliability, real-time performance, intelligence short-term forecasting. Therefore, this article proposes novel composite model based on AM–CNN–Bi–LSTM. First, CNN used extract relevant feature quantities coupling characteristics. Then, AM evaluate importance data, highlighting features that have greater impact results. Finally, Bi-LSTM network captures bidirectional temporal information from multiple time steps prediction. Taking one year measured data as an example, error comparison results overlay shows compared with other models, improved extraction, generalization ability, aspects. The research improve accuracy forecasting while providing effective references decision-making in optimization safe reasonable pricing.

Language: Английский

---

Improved Bacterial Foraging Optimization Algorithm with Machine Learning-Driven Short-Term Electricity Load Forecasting: A Case Study in Peninsular Malaysia

Algorithms, Journal Year: 2024, Volume and Issue: 17(11), P. 510 - 510

Published: Nov. 6, 2024

Accurate electricity demand forecasting is crucial for ensuring the sustainability and reliability of power systems. Least square support vector machines (LSSVM) are well suited to handle complex non-linear load series. However, less optimal regularization parameter Gaussian kernel function in LSSVM model have contributed flawed accuracy random generalization ability. Thus, these parameters need be chosen appropriately using intelligent optimization algorithms. This study proposes a new hybrid based on optimized by improved bacterial foraging algorithm (IBFOA) short-term daily Peninsular Malaysia. The IBFOA sine cosine equation addresses limitations fixed chemotaxis constants original (BFOA), enhancing its exploration exploitation capabilities. Finally, LSSVM-IBFOA constructed mean absolute percentage error (MAPE) as objective function. comparative analysis demonstrates model, achieving highest determination coefficient (R2) 0.9880 significantly reducing average MAPE value 28.36%, 27.72%, 5.47% compared deep neural network (DNN), LSSVM, LSSVM-BFOA, respectively. Additionally, exhibits faster convergence times BFOA, highlighting practicality forecasting.

Language: Английский

---

A hybrid forecasting model for general hospital electricity consumption based on mixed signal decomposition

Energy and Buildings, Journal Year: 2024, Volume and Issue: 325, P. 115006 - 115006

Published: Nov. 6, 2024

Language: Английский

---

Energy management in networked microgrids: Leveraging predictive analytics for renewable sources

Energy Conversion and Management X, Journal Year: 2024, Volume and Issue: unknown, P. 100828 - 100828

Published: Dec. 1, 2024

Language: Английский

---

An Overview of Short-Term Load Forecasting for Electricity Systems Operational Planning: Machine Learning Methods and the Brazilian Experience

Energies, Journal Year: 2023, Volume and Issue: 16(21), P. 7444 - 7444

Published: Nov. 4, 2023

The advent of smart grid technologies has facilitated the integration new and intermittent renewable forms electricity generation in power systems. Advancements are driving transformations context energy planning operations many countries around world, particularly impacting short-term horizons. Therefore, one primary challenges this environment is to accurately provide forecasting load demand. This a critical task for creating supply strategies, system reliability decisions, price formation markets. In context, nonlinear models, such as Neural Networks Support Vector Machines, have gained popularity over years due advancements mathematical techniques well improved computational capacity. academic literature highlights various approaches improve accuracy these machine learning including data segmentation by similar patterns, input variable selection, from hierarchical data, net forecasts. Brazil, national independent operator operation short term through DESSEM model, which uses forecast models day-ahead system. Consequently, study provides comprehensive review methods used forecasting, with particular focus on those based discusses Brazilian Experience.

Language: Английский

---

An IHPO-WNN-Based Federated Learning System for Area-Wide Power Load Forecasting Considering Data Security Protection

Energies, Journal Year: 2023, Volume and Issue: 16(19), P. 6921 - 6921

Published: Oct. 1, 2023

With the rapid growth of power demand and advancement new system intelligence, smart energy measurement data quality security are also facing influence diversified factors. To solve series problems such as low prediction efficiency, poor perception, “data islands” system, this paper proposes a federated learning based on Improved Hunter–Prey Optimizer Optimized Wavelet Neural Network (IHPO-WNN) for whole-domain load prediction. An improved HPO algorithm Sine chaotic mapping, dynamic boundaries, parallel search mechanism is first proposed to improve generalization ability wavelet neural network models. Further considering privacy in each station area potential threat cyber-attacks, localized differential privacy-based architecture designed by using above IHPO-WNN base model. In paper, actual dataset master selected, simulation experiments carried out through MATLAB software test examine performance federal respectively, results show that method has high accuracy excellent practical performance.

Language: Английский

---