The volatility and uncertainty of wind power often affect the quality of electric energy, the security of the power grid, the stability of the power system, and the fluctuation of the power market. In this case, the research on wind power forecasting is of great significance for ensuring the better development of wind power grids and the higher quality of electric energy. Therefore, a lot of new forecasting methods have been put forward. In this paper, a new forecasting model based on a convolution neural network and LightGBM is constructed. The procedure is shown as follows. First, we construct new feature sets by analyzing the characteristics of the raw data on the time series from the wind field and adjacent wind field. Second, the convolutional neural network (CNN) is proposed to extract information from input data, and the network parameters are adjusted by comparing the actual results. Third, in consideration of the limitations of the single-convolution model in predicting wind power, we innovatively integrated the LightGBM classification algorithm at the model to improve the forecasting accuracy and robustness. Finally, compared with the existing support vector machines, LightGBM, and CNN, the fusion model has better performance in accuracy and efficiency. INDEX TERMS Convolutional neural network, fusion model, LightGBM, ultra-short-term wind power forecasting, wind energy.
Due to the volatility and randomness of the photovoltaic power generation, it is difficult for traditional models to predict it accurately. To solve the problem, we established a model based on the selfattention mechanism and multi-task learning to predict the ultra-short-term photovoltaic power generation. First, we selected the data with the optimal timing length and input the data into the Encoder-Decoder network based on the self-attention. The validity of features extracted by the encoder was checked by the decoder. Then, we added a restriction to the middle layer of the Encoder-Decoder network to prevent the autoencoder from copying the input to the output mechanically. This condition is used to predict the photovoltaic power generation, so a multi-task learning model was established. Finally, to take full advantage of the features that are efficiently expressed and allow our main task, the prediction task, to learn some unique features autonomously, we proposed a step-by-step training method and have validated the effectiveness of this view in experiments. Through experimental contrast, it is found that compared with the Encoder-Decoder network based on CNN and LSTM, the performance of the proposed method has been increased by 14.82% and 8.09% respectively. The RMSE and MAE of the Encoder-Decoder model based on the self-attention mechanism using step-by-step training are 0.071 and 0.040 respectively. INDEX TERMS Energy, PV power prediction, self-attention mechanism, multi-task learning, autoencoder.
The ASIR technique tends to improve the image quality of VMS imaging. Dual-energy computed tomography pulmonary angiography with ASIR can reduce contrast medium volume and produce images of comparable quality with those of standard CTPA.
Background Reducing radiation dose inevitably increases image noise, and thus, it is important in low-dose computed tomography (CT) to maintain image quality and lesion detection performance. Purpose To assess image quality and lesion conspicuity of ultra-low-dose CT with model-based iterative reconstruction (MBIR) and to determine a suitable protocol for lung screening CT. Material and Methods A total of 120 heavy smokers underwent lung screening CT and were randomly and equally assigned to one of five groups: group 1 = 120 kVp, 25 mAs, with FBP reconstruction; group 2 = 120 kVp, 10 mAs, with MBIR; group 3 = 100 kVp, 15 mAs, with MBIR; group 4 = 100 kVp, 10 mAs, with MBIR; and group 5 = 100 kVp, 5 mAs, with MBIR. Two radiologists evaluated intergroup differences with respect to radiation dose, image noise, image quality, and lesion conspicuity using the Kruskal-Wallis test and the Chi-square test. Results Effective doses were 61-87% lower in groups 2-5 than in group 1. Image noises in groups 1 and 5 were significantly higher than in the other groups ( P < 0.001). Overall image quality was best in group 1, but diagnostic acceptability of overall image qualities in groups 1-3 was not significantly different (all P values > 0.05). Lesion conspicuities were similar in groups 1-4, but were significantly poorer in group 5. Conclusion Lung screening CT with MBIR obtained at 100 kVp and 15 mAs enables a ∼60% reduction in radiation dose versus low-dose CT, while maintaining image quality and lesion conspicuity.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.