Benchmark dataset: 3 years of processed sky images (64×64) and concurrent PV power generation data with 1-min interval that are ready-to-use for deep learning model development; Raw dataset: Overlapping high resolution sky video footage (2048×2048) recorded at 20 frames per second, sky ima...
The first 70 % of each dataset was used as the training set, 20 % as the validation set, and the remaining 10 % as the test set.Fig. 3 shows the time series records of the PV power data in four datasets. Table 1. Statistical analysis of the four datasets. DatasetsNumbersStatistical ...
Therefore, this study proposes a hybrid deep learning model for PV power forecast that is successfully developed using the combination of the bidirectional long short-term memory (BLSTM) and convolutional neural network (CNN) and is applied to the actual dataset collected in the DKASC PV system ...
When large-scale photovoltaic (PV) power stations are connected to the power grid, it will have a serious impact on the security and stability of the power system1,2. Therefore, it is of great significance to predict the power generation of PV power stations quickly and accurately3. At pres...
The data used in this study were obtained from the publicly available dataset of the Desert Knowledge Australia Solar Center (DKASC) [36]. The selected PV array area number is 23 and the rated output power is 5.4 kW. After data were screened and cleaned, the data measured from 2014 to 20...
The algorithm extracts key features that significantly contribute to photovoltaic power prediction, forming an optimal feature subset from the dataset. The dataset is initially processed, followed by the utilization of three feature selection algorithms: CMI feature selection algorithm, Pearson feature ...
of Descartes Labs Inc., the company that builds and maintains the cloud computation infrastructure used to conduct this research. J.F. and L.B. are employees of the World Resources Institute, a not-for-profit organization which will host and publicly visualize a copy of our dataset. ...
small datasets to assist in the fine-tuning process.\nOn the PVEL-AD dataset, we validated the effectiveness of our proposed\nELCN-YOLOv7 method. It... HS Fu,G Cheng - 《Energy Sources Part A Recovery Utilization & Environmental Effects》 被引量: 0发表: 2023年 Defect detection of photov...
In particular, GRU can be used to save computational costs and time when dealing with a large dataset (i.e., historical weather data). With the potential of superior quantum computers, quantum machine learning (QML) has emerged as a potentially faster solution than its classical counterparts [...
Thus, it is extremely important for generation, transmission and energy management [6]. Energy production for the next day has to be planned on the previous day and this PV power forecasting process for the next day is a daily routine for the PV power generating station. The forecasting error...