Journal of Renewable Energy and Smart Grid Technology

A Review of Machine Learning Models and Evaluation Techniques for Single String Anomaly Detection in Malaysian Large-Scale Photovoltaic Systems

Muhammad Ashraf ฺBin Ghazali — 2026-07-03

The rapid expansion of solar photovoltaic (PV) systems has increased the need for reliable and intelligent fault detection techniques to maintain system performance and operational efficiency. This paper presents a comprehensive review of machine learning (ML) models and evaluation techniques for single-string anomaly detection in large-scale PV systems, with particular emphasis on applications in Malaysia’s tropical environment. The review examines the transition from conventional monitoring approaches, such as I–V curve tracing and threshold-based diagnostics, to advanced data-driven methods. Various ML and deep learning models, including Artificial Neural Networks (ANN), Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) and hybrid frameworks such as CNN–SVM and AE–GRU, are analysed in terms of their anomaly detection capabilities and performance. Common evaluation metrics, including RMSE, MAE, MAPE, R², and F1-score, are also reviewed to assess model effectiveness and reliability. In addition, recent Malaysian studies employing K-Means–LSTM clustering, Random Forest-based fault diagnosis, Vision Transformers (ViT) and Vertices Principal Component Analysis (VPCA) are discussed, highlighting their potential for PV anomaly detection under tropical operating conditions. The review identifies key challenges, particularly the dependence on labelled datasets, which limits scalability and early fault detection. Emerging opportunities in unsupervised and semi-supervised learning approaches, including autoencoders, Isolation Forests, and clustering-based reconstruction methods, are also explored as promising solutions for real-time and adaptive anomaly detection. The findings provide insights into current research trends, existing limitations, and future directions for enhancing the reliability and maintenance of large-scale PV systems in Malaysia.

Comparative Review of Bio-Oil and Liquid Smoke in Terms of Characteristics, Composition, and Applications

Delphy Yustisia Ayu Praja — 2026-07-09

Thermal decomposition process or commonly referred as pyrolysis is one of the methods that can be utilized to obtain several chemical products. Through the pyrolysis process, there are three products produced, each of which can be used; char, tar and bio-oil or liquid smoke. This article will specifically discuss about bio-oil and liquid smoke which are often assumed as the same, but are actually two different things. It is based on the used feedstock, composition of compounds in the product and the utilization of both which are different. To support the writing of the article, a Systematic Literature Review (SLR) method was used to obtain data to analyze and critically compare the differences between bio-oil and liquid smoke. The findings indicate that the used feedstock may affect the quality and characteristics of pyrolysis liquid products. Liquid products obtained from feedstocks contained organic compounds (e.g. biomass) tend to produce high acid and phenol content, called as liquid smoke. While liquid products produced from feedstocks with long carbon chain bonds (e.g. several types of plastics and scrap tires) tend to produce high hydrocarbon compounds called as bio-oil.

Agrivoltaics Systems in Egypt: Opportunities, Challenges, and Future Prospects

Mohammed Elsayed Samier Madboli Ibrahim — 2026-08-04

Agrivoltaic systems are defined as the integration of renewable energy generation and agricultural activities within the same landscape. This innovative approach could address some of the most pressing challenges facing both sectors in Egypt and provide a sustainable and reliable long-term solution. The agricultural sector currently faces serious issues, including water scarcity, land degradation, and the fact that nearly 95% of the country’s land is arid. In response, the Egyptian government’s Vision 2030 highlights the importance of expanding renewable energy, particularly solar energy, due to Egypt’s abundant sunlight. Because of the country’s dry conditions, agrivoltaic systems would allow land to be used for both farming and electricity generation. The installed solar panels could provide necessary shade for crops, reduce water evaporation, improve the surrounding microclimate, and make crops more resilient to harsh environmental conditions. Successful studies and applications in countries such as India, France, and Spain have shown that agrivoltaics can increase land productivity, generate clean energy, and improve water-use efficiency. These results suggest that Egypt has strong potential to make agrivoltaics a leading initiative in both the agricultural and energy sectors. However, implementation in Egypt may face several challenges, including high initial costs, technical complexity, and the need for supportive policies. This paper examines the potential of agrivoltaic systems in Egypt, focusing on sustainable agriculture, renewable energy generation, and rural development. By addressing financial, technical, and policy barriers, Egypt could use agrivoltaics to support its Vision 2030 goals and provide an example for countries facing similar challenges.

Modeling and Real-World Validation of Home Energy Management Systems: Photovoltaic Generation, Battery Energy Storage, and Bidirectional Grid Connection

Supakan Janthong — 2026-07-13

This article presents comprehensive modeling, simulation, and real-world validation of a Home Energy Management System (HEMS) using MATLAB/Simulink. The model integrates photovoltaic (PV) panels with single-diode equivalent circuits, lithium-ion battery energy storage systems (BESS) with electrochemical modeling, unidirectional boost converters for PV voltage regulation, bidirectional DC/DC converters for battery control, bidirectional DC/AC inverters with sinusoidal PWM for grid synchronization, and utility grid connection to enhance energy efficiency and reduce grid dependency. Five operational scenarios were designed and tested: (1) normal operation with PV as primary source, (2) high load exceeding PV capacity, (3) low solar irradiance requiring battery support, (4) battery charging during PV surplus, and (5) grid export of excess energy. Each scenario was analyzed under varying environmental conditions (solar irradiance 0–650 W/m²) and load profiles. Simulation results demonstrate effective energy balancing through priority-based control (PV → Battery → Grid), with response times under 200 ms, DC bus voltage stability within ±2%, and AC power quality meeting grid standards (THD < 3%, power factor > 0.98). Real-world validation used 30 days of operational data from a 5 kW PV system with 10 kWh battery storage monitored through the FusionSolar platform. Experimental results showed strong model accuracy with
an average MAPE of 4.2%. Performance metrics demonstrated 65.2% PV self-consumption under normal conditions, 37% peak demand reduction during high-load scenarios, 92.4% battery charging efficiency, and 96.1% grid export inverter efficiency. The validated model confirms the HEMS performance evaluation and demonstrates the practical feasibility and economic viability of integrated PV-battery-grid systems.

An Experimental Investigation of Organic Rankine Cycle with Scroll Expander for Low-Temperature Heat Sources

Suppachai Chumnumwat — 2026-07-21

Effective utilization of low-grade waste heat remains a significant challenge in sustainable energy recovery. This study presents an experimental investigation of an organic Rankine cycle (ORC) system utilizing R134a as the working fluid, integrated with a scroll expander modified from an electric automotive compressor. The primary objective was to evaluate the system's thermodynamic performance under varying low-temperature heat source conditions (50°C to 90°C) and heat sink temperatures (8°C to 16°C). The experimental results demonstrated that the system performance is heavily dependent on the thermal gradient, which governs the pressure ratio and enthalpy drop. Under the operating condition (evaporating temperature of 90°C and condensing temperature of 8°C), the ORC system generated a maximum electrical power output of 488.55 W. At this peak point, the ORC system achieved a thermal efficiency of 2.61%, with the scroll expander exhibiting a maximum isentropic efficiency of 21.14%. Detailed analysis reveals that the efficiency limitations are primarily attributed to under-expansion losses caused by the mismatch between the system pressure ratio and the expander's fixed built-in volume ratio, compounded by internal leakage. Despite the lower efficiency compared to specifically built expanders, this study confirms the techno-economic feasibility of using low-cost, off-the-shelf automotive components for small-scale decentralized power generation, offering a robust and cost-effective solution for low-grade heat recovery.

Development of Empirical Models for Estimating Daily Downward Longwave Radiation Using Meteorological Data in Thailand

Rungrat Wattan — 2026-07-24

This study developed models to estimate daily downward longwave radiation under general sky conditions using meteorological data. The input variables included air temperature, vapor pressure, daily minimum and maximum temperatures, relative humidity, and cloud cover. Data were collected from four meteorological stations, namely Chiang Mai, Ubon Ratchathani, Nakhon Pathom, and Songkhla, during the period from 2018 to 2023. Model development was based on data from 2018 to 2021 using graphical analysis and multiple linear regressions to identify strong relationships and build predictive equations. Seven models were formulated, each using different combinations of meteorological variables. These models were then validated using independent data from 2022–2023. The results showed that downward longwave radiation is largely correlated with meteorological parameters. The developed models provided accurate estimates, closely matching observed radiation values. The root mean square error relative to the mean measured values (RMSE) ranged from 1.7% to 4.4%, and mean bias error relative to the mean measured values (MBE) ranged from -1.4% to 3.9%. The developed model gave a more accurate value of daily longwave radiation. The findings indicate that the proposed models are effective tools for estimating longwave radiation in areas without direct measurements, supporting further research in climate and energy studies.

Simulation and Analysis of a Solar-Powered Electric Vehicle Charging Station Based on MPPT

Prashant Singh Rajpoot — 2026-08-10

As electric vehicles (EVs) are rapidly increasing, the need for clean, efficient, and grid-independent EV charging stations is growing. Existing charging stations are prone to efficiency degradation, voltage fluctuation and low flexibility to changing solar irradiance. In this research, a PV-Storage-based EV charging infrastructure combined with a boost converter besides a hybrid maximum power point tracking (MPPT) algorithm (P&O and PID) is introduced. This system is capable of providing stable charging and high efficiency of power conversion under different operating conditions. The simulation has demonstrated the peak efficiency of 92.3%, MPPT tracking efficiency of 98.5%, and the voltage gain ratio of 6.73, which allows the system to deliver a high output above 250 V even with low irradiance (720 W/m²). The hybrid MPPT approach significantly reduces the oscillations and response time, given that the hybrid approach converges more quickly than conventional approaches. The robustness and scalability of the system is validated through benchmarking with the latest state-of-the-art (2020-2025). The findings confirmed the system as viable and sustainable for next-generation EV charging infrastructure, which is promising for energy security, less reliance on the traditional grid and promoting a more efficient renewable energy utilization. Future research will take into account the hardware implementation of the system using WBG devices with hybrid renewable energy sources and vehicle-to-grid (V2) energy exchange for improving the resilience and adaptability of the system.