a b s t r a c tAs a type of clean and renewable energy source, wind power is widely used. However, owing to the uncertainty of wind speed, it is essential to build an accurate forecasting model for large-scale wind power penetration. Numerical weather prediction (NWP) and data-driven modeling are two typical paradigms. NWP is usually unavailable or spatially insufficient. Data-driven modeling is an effective candidate. As to some newly-built wind farms, sufficient historical data is not available for training an accurate model, while some older wind farms may have long-term wind speed records. A question arises regarding whether the prediction model trained by data coming from older farms is also effective for a newly-built farm. In this paper, we propose an interesting trial of transferring the information obtained from data-rich farms to a newly-built farm. It is well known that deep learning can extract a high-level representation of raw data. We introduce deep neural networks, trained by data from data-rich farms, to extract wind speed patterns, and then finely tune the mapping with data coming from newly-built farms. In this way, the trained network transfers information from one farm to another. The experimental results show that prediction errors are significantly reduced using the proposed technique.

Through making full use of the solar wind and interplanetary magnetic field data accumulated by ACE satellites we improve the prediction accuracy of the Kp geomagnetic index and accurately predict the occurrence of geomagnetic storms (Kp ≥ 5). Specially, we use long short‐term memory to train the Kp forecast model described in this study. Based on the large‐scale data, we build the Kp forecasting model with solar wind, interplanetary magnetic field parameters, and the historical Kp value as input. In this study, we first analyze the distribution of Kp and the effect of the data imbalance on the prediction of geomagnetic storms. Second, we analyze the correlation between the different input parameters and Kp. Thus, the input parameters of the model are selected by the results of the correlation. We consider two types of forecasting: one is the overall Kp forecasting and the other is the geomagnetic storm (Kp ≥ 5) forecasting. Hence, we design an integrated model which is then compared with other models. Some evaluation parameters are introduced: the root‐mean‐square error, the mean‐absolute error, and the correlation coefficient, as well as the measurement of geomagnetic storms (Kp ≥ 5) F1. The root‐mean‐square error and mean‐absolute error of our model are 0.4765 and 0.6382, respectively. The experimental results show that the proposed model with long short‐term memory improve the Kp forecasting.

Deep ConvNets have shown its good performance in image classification tasks. However it still remains as a problem in deep video representation for action recognition. The problem comes from two aspects: on one hand, current video ConvNets are relatively shallow compared with image ConvNets, which limits its capability of capturing the complex video action information; on the other hand, temporal information of videos is not properly utilized to pool and encode the video sequences.Towards these issues, in this paper, we utilize two stateof-the-art ConvNets, i.e., the very deep spatial net (VGGNet [29]) and the temporal net from Two-Stream ConvNets [28], for action representation. The convolutional layers and the proposed new layer, called frame-diff layer, are extracted and pooled with two temporal pooling strategy: Trajectory pooling and line pooling. The pooled local descriptors are then encoded with VLAD to form the video representations. In order to verify the effectiveness of the proposed framework, we conduct experiments on UCF101 and HMDB51 datasets. It achieves the accuracy of 93.78% on UCF101 which is the state-of-the-art and the accuracy of 65.62% on HMDB51 which is comparable to the state-of-the-art.

No abstract

No abstract