Journal of Renewable Energy and Smart Grid Technology

Sliding Window-Based Deep Learning Approach for Solar Power Forecasting in Malaysia Utility-Scale PV Systems

Thu, 04 Dec 2025 00:00:00 +0700

Accurate forecasting of solar power in utility-scale photovoltaic (USPV) systems is critical for grid stability but remains challenging due to meteorological variability and the large spatial scale of these systems. However, the choice of sliding window size in time-series forecasting remains underexplored. This study introduces a deep learning-based forecasting framework that systematically evaluates
the impact of sliding window size on forecasting accuracy using multivariate time-series data. The data collected from a 25 MWac USPV system in Malaysia between August 2022 and April 2023, comprises 5-minute interval measurements of solar irradiance, module temperature and solar power output. Multiple deep learning (DL) models, namely LSTM, CNN and GRU across window sizes ranging from 12 to 288 steps and forecasting horizons of 1 to 12 hours were investigated. Results show that a 144-step window consistently improves accuracy over conventional one-step input methods, with LSTM outperforming other models by achieving up to 23.1% RMSE reduction, 30.7% MAE reduction and a 8.6% increase in R² at 60 minutes forecasting horizon. This work emphasizes the importance of window size selection in optimizing forecasting accuracy for USPV systems and supporting renewable energy grid integration. By improving forecasting capabilities, this research is expected to provide critical insights to enhance renewable energy integration into the grid system.

Optimal Sizing of Stand-Alone Photovoltaic Systems using Meerkat Optimization Algorithm: A Cost-Based Performance Assessment

Wed, 17 Dec 2025 00:00:00 +0700

Stand-Alone Photovoltaic (SAPV) systems play a vital role in providing clean and reliable electricity for remote and off-grid communities where grid expansion is economically or technically unfeasible. Their economic feasibility and technical reliability, however, depend strongly on accurate component sizing and system configuration, which require advanced optimization techniques. In this study, the Meerkat Optimization Algorithm (MOA) is applied to optimize two SAPV configurations. System 1 integrates a photovoltaic array, battery storage, and a hybrid inverter, while System 2 consists of a photovoltaic array, battery storage, a solar inverter, and a charge controller. The optimization focuses on minimizing Life Cycle Cost (LCC) and Levelized Cost of Energy (LCOE), which are widely recognized as reliable indicators of long-term cost-effectiveness and financial viability. To validate the performance of MOA, its results are benchmarked against three well-established metaheuristic algorithms: Particle Swarm Optimization (PSO), Firefly Algorithm (FA), and Slime Mould Algorithm (SMA). Simulation results show that System 1 consistently achieves lower LCC and LCOE compared to System 2, primarily due to its reduced component count and simplified integration. Moreover, MOA demonstrates enhanced optimization performance by converging more rapidly and delivering more stable solutions across multiple independent runs. In contrast, PSO, FA, and SMA exhibit slower convergence and greater variability in outcomes. Importantly, the performance differences are statistically meaningful, as MOA achieved consistently lower mean values and smaller standard deviations. These findings highlight MOA as an effective and reliable optimization tool for SAPV systems and provide practical insights to support sustainable rural electrification planning.

Optimized Power Quality in Grid Systems Using PV-Based UPFC and Advanced ANN Control Approach

Wed, 24 Dec 2025 00:00:00 +0700

Nowadays, grid power losses and Power Quality (PQ) issues are inducing various problems in power systems, which need to be addressed and rectified for attaining enhanced and smooth functioning. These PQ issues are generated as a result of differing values between generated and load power, which further produces fluctuations within the power supply. Hence, to overcome these limitations, an innovative control approach is proposed for attaining optimum power flow by using a Unified Power Flow Controller (UPFC). The proposed UPFC is combined with an Artificial Neural Network (ANN) controller for improving the performance efficiency of the UPFC. The ANN controller-aided UPFC rectifies the PQ issues, including sag and swell. Additionally, to provide a consistent and unlimited power supply to the DC link, a photovoltaic (PV) system is incorporated with a single-switch boost-Cuk (SSBC) converter for boosting the PV power generation process. For attaining maximum power extraction and tracking maximum power, a new Pelican Optimized Recurrent Neural Network (RNN)-based Maximum Power Point Tracking (MPPT) technique is utilized. Furthermore, to validate the proposed model, MATLAB/Simulink is utilized and the obtained results depict improved PQ with reduced losses. Therefore, the overall system attains improved power quality, thereby, enhancing the power system functioning.