1,721,040 research outputs found
A proof of the compositional Delta conjecture
Full text linkWe prove a compositional refinement of the Delta conjecture (rise version) of Haglund, Remmel and Wilson [16] for Δejavax.xml.bind.JAXBElement@350ecd2e′en which was stated in [8] in terms of Theta operators
Deep learning neural networks for short-term photovoltaic power forecasting
Full text linkAccurate short-term forecasting of photovoltaic (PV) power is indispensable for controlling and designing smart energy management systems for microgrids. In this paper, different kinds of deep learning neural networks (DLNN) for short-term output PV power forecasting have been developed and compared: Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), Gated Recurrent Unit (GRU), Bidirectional GRU (BiGRU), One-Dimension Convolutional Neural Network (CNN1D), as well as other hybrid configurations such as CNN1D-LSTM and CNN1D-GRU. A database of the PV power produced by the microgrid installed at the University of Trieste (Italy) is used to train and comparatively test the neural networks. The performance has been evaluated over four different time horizons (1 min, 5 min, 30 min and 60 min), for one-Step and multi-step ahead. The results show that the investigated DLNNs provide very good accuracy, particularly in the case of 1 min time horizon with one-step ahead (correlation coefficient is close to 1), while for the case of multi-step ahead (up to 8 steps ahead) the results are found to be acceptable (correlation coefficient ranges between 96.9% and 98%)
Application of artificial neural networks for the prediction of a 20-kWp grid-connected photovoltaic plant power output
No full textLeast squares support vector machine for short-term prediction of meteorological time series
No full textA study on the mismatch effect due to the use of different photovoltaic modules classes in large-scale solar parks
No full textIn this paper, a study on the mismatch effect due to the use of different photovoltaic (PV) modules classes in large-scale solar parks is presented. For this purpose, a new model for simulating current-voltage and power-voltage characteristics is introduced. The model is then applied for calculating mismatch losses in a number of case studies for a PV plant built in Bari, southern Italy. First, in order to test the effectiveness of the model, this is applied to homogeneous strings and field showing that the mismatch losses are zero. Subsequently, the use of inhomogeneous strings (i.e. made of modules belonging to different power classes) is investigated. Finally, the behaviour of 1MWp homogeneous and inhomogeneous PV fields is investigated, again with a focus on the mismatch effect. The operational conditions have been introduced starting from the definition of European efficiency. The use of standard test conditions can in fact lead to gross approximations because mismatch losses depend, as well as, on PV module characteristics, electrical connections and electrical architecture, also on the location of the PV system. The results presented in this work can be used both by PV system designers for carrying out yield calculations, and by operation and maintenance personnel for substituting modules during operation without compromising the productivity of the plant
An embedded system for remote monitoring and fault diagnosis of photovoltaic arrays using machine learning and the internet of things
Full text linkIn this paper a novel embedded system for remote monitoring and fault diagnosis of photovoltaic systems is introduced. The idea is to embed machine leaning algorithms into a low-cost edge device for real-time deployment. First, an artificial neural network is developed to detect faults. Then an effective stacking ensemble learning algorithm is developed to classify the nature of the fault. The method performance is evaluated through common error metrics such as RMSE, MAE, MAPE, r and confusion matrix. Additional algorithms are also embedded into the edge device in order to remotely control the photovoltaic array parameters. Users can be notified by email and SMS about the state of their photovoltaic array. The Blynk IoT platform is used to monitor remotely the photovoltaic array parameters. The experimental results demonstrate the ability of the proposed embedded system to diagnose and monitor the photovoltaic array with a good accuracy
A machine learning and internet of things-based online fault diagnosis method for photovoltaic arrays
Full text linkIn this paper, a novel fault detection and classification method for photovoltaic (PV) arrays is introduced. The method has been developed using a dataset of voltage and current measurements (I–V curves) which were collected from a small-scale PV system at the RELab, the University of Jijel (Algeria). Two different machine learning-based algorithms have been used in order to detect and classify the faults. An Internet of Things-based application has been used in order to send data to the cloud, while the machine learning codes have been run on a Raspberry Pi 4. A webpage which shows the results and informs the user about the state of the PV array has also been developed. The results show the ability and the feasibility of the developed method, which detects and classifies a number of faults and anomalies (e.g., the accumulation of dust on the PV module surface, permanent shading, the disconnection of a PV module, and the presence of a short-circuited bypass diode in a PV module) with a pretty good accuracy (98% for detection and 96% classification)
TinyML Model for Fault Classification of Photovoltaic Modules Based on Visible Images
No full textIn this paper, a Tiny Machine Learning (TinyML) model is developed for fault classification of photovoltaic (PV) modules. A dataset based on visible images of healthy and faulty PV modules has been collected at different locations. The examined defects are: discolored cells, cracked PV modules, bubble formation, bird droppings, dirt accumulation, sand deposit, corrosion, shading effect, and snail trails. The Edge Impulse platform has been used to develop and optimize our TinyML model, which is then integrated onto a low-power microcontroller for a real time application. The simulation results show a good overall classification accuracy of 92% whereas experimental results demonstrate the ability of the developed TinyML model to be deployed for real-world and low-cost applications
An iterative adaptive virtual impedance loop for reactive power sharing in islanded meshed microgrids
Full text linkThis paper proposes a control strategy for the optimization of the reactive power sharing based on an iterative adaptive virtual impedance (IAVI). The IAVI includes two elements: the first is proportional to the reactive power delivered by the distributed generation units at the current iteration, while the second is proportional to the sum of the reactive power variations at the previous iterations. The proposed control strategy has been verified under a Matlab/Simulink environment for an islanded meshed microgrid with three distributed generators. The simulation of different scenarios considering feeder impedance mismatches, different microgrid configurations, and variable loads has shown a good accuracy in the sharing of the reactive power in the microgrid. The control strategy proposed in this paper can be easily implemented as it does not require any communication link between the generators, any knowledge regarding the feeder impedances, and any local load measurement
- …
