Related references
Note: Only part of the references are listed.
Article
Chemistry, Analytical
Hamad Alharkan et al.
Summary: This paper introduces a deep learning-based method combining dual-stream convolutional neural network (CNN) and long short-term memory (LSTM) network for accurate prediction of solar power generation. The method learns spatial and temporal features, and incorporates a self-attention mechanism to select optimal features for further processing, achieving good performance in short-term solar power prediction.
Article
Chemistry, Analytical
Reham A. Eltuhamy et al.
Summary: This article presents the development of an adaptive neuro-fuzzy inference system (ANFIS) for PV fault classification based on statistical and mathematical features extracted from outdoor infrared thermography (IRT) and I-V measurements of thin-film PV modules. The proposed methodology achieves high classification accuracy through dimension reduction using PCA and selection of effective features.
Article
Chemistry, Analytical
Radityo Fajar Pamungkas et al.
Summary: Photovoltaic (PV) systems have great potential for clean energy generation and their adoption has grown significantly in recent years. Faults in PV systems can lead to safety risks, reduced lifespan, and waste. This paper highlights the importance of accurately classifying faults in PV systems to maintain optimal efficiency and increase financial return. The proposed lightweight coupled UdenseNet model shows significant improvements in fault classification accuracy and efficiency compared to previous studies, and image augmentation techniques further enhance performance on unbalanced datasets.
Article
Energy & Fuels
Mariam Ibrahim et al.
Summary: This paper evaluates the performance of different machine learning schemes and applies them to detect anomalies on photovoltaic components, providing models to identify healthy and abnormal behaviors in photovoltaic systems and offering clear insights for solving complex anomaly detection problems.
Article
Computer Science, Information Systems
Ali Reza Sajun et al.
Summary: This paper presents an anomaly detection system that utilizes Siamese-twin neural networks for detecting anomalies on edge devices in a solar farm. The model achieves high accuracy and efficiency, as evaluated on multiple hardware platforms.
Article
Energy & Fuels
Masoud Emamian et al.
Summary: This paper proposes an Intelligent Monitoring System (IMS) for Photovoltaic (PV) systems, which utilizes affordable hardware and lightweight software to easily implement in different locations and types of PV power plants. IMS uses IoT platform for data handling and communication among devices and components. It also includes a personal cloud server for computing and storing PV system data. The system utilizes deep ensemble models for fault detection and power prediction, improving energy yield and identifying malfunctions in PV plants. The IMS is scalable, interoperable, and replicable, offering a holistic approach to PV plant monitoring.
Article
Chemistry, Analytical
Sufyan Ali Memon et al.
Summary: This paper presents an intelligent model to detect faults in PV panels using Convolutional Neural Network. Training and testing results show a higher accuracy compared to previous studies conducted on the dataset.
Article
Chemistry, Analytical
Jose Miguel Paredes-Parra et al.
Summary: With the rapid changes in power system configuration and performance, grids are now being influenced by microgeneration systems connected in low and medium voltage. However, these facilities often provide little to no information to distribution and transmission system operators, causing management problems. This paper proposes a monitoring solution using IoT technologies to address these issues, specifically for photovoltaic self-consumption installations. Field tests show that the proposed solution provides reliable and accurate results, making it a popular choice for those who cannot afford professional monitoring platforms.
Article
Chemistry, Analytical
Luqman Ahsan et al.
Summary: This article describes a low-cost SCADA system for a PV plant with local data logging. It uses open-source hardware and software solutions to process, log, and visualize PV and environmental data. With the cooperation of remote terminal units and Raspberry Pi, data collection and storage are achieved, while providing a user-friendly dashboard.
Review
Physics, Applied
Arnulf Jaeger-Waldau
Summary: Over the past decade, photovoltaic electricity generation has become the fastest-growing power source globally, with solar energy capacity reaching new heights. The Chinese market's recovery, along with ongoing growth in Europe, the USA, and emerging markets, has been a major driver. The decreasing prices in battery storage technology have fostered the development of local storage systems for solar farms and decentralized electricity generation systems. It is evident that renewable electricity will play a crucial role in powering electric vehicles and producing green hydrogen in the future.
Article
Engineering, Electrical & Electronic
Wenchao Miao et al.
Summary: The article proposes an LL fault detection technique for PV systems based on string-current behavior and current sensing, which can determine the occurrence, type, location, and percentage of faults. The technique is effective in detecting LL faults despite the effects of MPPT, partial shading, and blocking diodes.
IEEE TRANSACTIONS ON MAGNETICS
(2021)
Article
Chemistry, Analytical
Isaias Gonzalez et al.
Summary: This paper presents a novel I-V curve tracer based on IoT open-source hardware and software, using a Raspberry Pi microcomputer as the hardware level, while the applied software comprises mariaDB, Python, and Grafana. All tasks required for curve tracing are automated, and the system development and experimental results demonstrate the suitability and validity of the proposal, particularly for I-V curve tracing of a monocrystalline PV generator under real operating conditions.
Article
Automation & Control Systems
B. Pradeep Kumar et al.
Summary: Traditional protection devices in PV arrays struggle to detect faults under low irradiance levels, leading to reliability issues and fire risks. This article proposes a new fault localization method based on differential voltage measurement between PV modules for timely and accurate fault detection and localization, ensuring sustainable power generation and safety.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2021)
Article
Green & Sustainable Science & Technology
Jose Maria Portalo et al.
Summary: Smart grids and smart microgrids require proper monitoring for their operation, and this paper presents a monitoring system based on open-source hardware and software for tracking the temperature of the photovoltaic generator in such a SMG. The monitoring system consists of a network of digital temperature sensors connected to an Arduino microcontroller, which feeds the acquired data to a Raspberry Pi microcomputer, accessed by a cloud-enabled user/operator interface implemented in Grafana.
