In this study, horizontal wind (U and V), air temperature (T), and relative humidity (RH) modelled by the European Centre for Medium-Range Weather Forecasts Reanalysis Interim (ERA-Interim), the National Aeronautics and Space Administration (NASA) Modern Era Retrospective Analysis for Research and Applications (MERRA), the Japanese 55-year Reanalysis (JRA-55), the National Centers for Environmental Prediction (NCEP) Climate Forecast System Version 2 (CFSv2), and the NCEP Final Operational Global Analysis data and the NCEP Final Operational Global Analysis data (NCEP-FNL) products have been compared with observations at 11 radiosonde stations over the eastern and downstream regions of the Tibetan Plateau (TP) from late June until the end of July during 2011 to 2015. The mean bias of all variables for the five gridded datasets (GDs) in the Sichuan Basin (SCB) is larger than that for the TP. The mean values of U, V, and T from each grid dataset are generally consistent with the radiosonde values, whereas considerable bias in the mean RH exists at upper levels. The diurnal variation of the mean bias and root-mean-square (RMS) error in the basin are stronger than those in the TP and the negative/positive peak usually occurs at 06:00 UTC and 18:00 UTC in the basin or at 12:00 UTC in the TP. The inter-annual variations in the basin are significantly stronger, and the maximum values of the variations usually occur at upper levels or near the surface, except for V. The weather conditions have a crucial influence on the performance of the gridded datasets. The mean bias and RMS error of T in the TP on cloudy days are obviously larger than those during sunny conditions. Considerable but unsteady differences occur in the mean bias and RMS error of U and V in different weather conditions. On average, the four variables in the TP are more sensible to the weather conditions.

Accurate wind speed prediction is significantly important for the full utilization of wind energy resources and the improvement in the economic benefits of wind farms. Because the ensemble forecast takes into account the uncertainty of information about the atmospheric motion, domestic and foreign weather service forecast centers often choose to use the ensemble numerical forecast to achieve the fine forecast of wind speed. However, due to the unavoidable systematic errors of the ensemble numerical forecast model, it is necessary to correct the deviation in the ensemble numerical forecast wind speed. Considering the typical spatio-temporal characteristics of the grid prediction data of the wind field, based on Convolutional Long–Short Term Memory (ConvLSTM) units and attention mechanism, this paper takes the complex and representative North China region as the research area, aiming to reveal the shortcomings of existing deep learning integrated prediction correction models in extracting temporal features of grid prediction data. We propose a new ensemble prediction wind field correction model integrating multi-factor and spatio-temporal characteristics. This model uses reanalyzed land data provided by the European Center for Medium-Range Weather Forecasts as the real data to correct the deviation in the near-surface 10 m wind field data predicted by the regional ensemble numerical prediction model of the China Meteorological Administration. We used the reanalyzed land data provided by the European Centre for Medium-Range Weather Forecasts (ECMWF) as the live data to correct the deviation in the near-surface 10 m wind field data predicted by the regional ensemble numerical forecast model of the China Meteorological Administration (CMA). At the same time, Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) were used as the scoring indicators, and the results of the China Meteorological Administration–Regional Ensemble Prediction System (CMA–REPS) ensemble average, multiple linear regression method correction, Long–Short Term Memory (LSTM) method correction, and U-net (UNET) method correction were compared. Compared with the UNET model method, the experimental results show that when processing the 10 m zonal wind data, 10 m meridional wind data, and 10 m average wind speed data of CMA–REPS 24 h forecasts, the correction results of our model can reduce the RMSE score index by 9.15%, 4.83%, and 7.79%. At the same time, when processing the 48 h and 72 h near-surface 10 m wind field data of the CMA–REPS forecast, our model can improve the prediction accuracy of CMA–REPS near-surface wind forecast data. Therefore, the correction effect of the proposed model in a complex terrain area is evidently better compared to other methods.

Short-term precipitation forecasting is essential for agriculture, transportation, urban management, and tourism. The radar echo extrapolation method is widely used in precipitation forecasting. To address issues like forecast degradation, insufficient capture of spatiotemporal dependencies, and low accuracy in radar echo extrapolation, we propose a new model: MS-DD3D-RSTN. This model employs spatiotemporal convolutional blocks (STCBs) as spatiotemporal feature extractors and uses the spatial-temporal loss (STLoss) function to learn intra-frame and inter-frame changes for end-to-end training, thereby capturing the spatiotemporal dependencies in radar echo signals. Experiments on the Sichuan dataset and the HKO-7 dataset show that the proposed model outperforms advanced models in terms of CSI and POD evaluation metrics. For 2 h forecasts with 20 dBZ and 30 dBZ reflectivity thresholds, the CSI metrics reached 0.538, 0.386, 0.485, and 0.198, respectively, representing the best levels among existing methods. The experiments demonstrate that the MS-DD3D-RSTN model enhances the ability to capture spatiotemporal dependencies, mitigates forecast degradation, and further improves radar echo prediction performance.

The sudden outbreak of the COVID-19 pandemic has had a serious impact on the health and daily life of people. The problem of determining where to put household goods during an epidemic is particularly important. This paper proposes a siting method for the distribution of supplies in disaster area relief missions. Firstly, a qualitative analysis of the SIR prediction model for infectious diseases is established, and it is learned that how subsistence supplies are distributed has a direct impact on the direction of the number of infected people in the disaster area. Secondly, the entropy weight is applied to comprehensively evaluate the reasonableness of the number of material drop-off points. The optimal number, location, and size of drop-off points are determined by the GA-K-means clustering algorithm. This algorithm resulted in a 22.88% reduction in the overall number of task drop-off points. The average contour coefficient is then used to find the number and location of large-scale material sorting sites. The siting method for drop-off points provides practical implications for the solution of material or personnel deployment problems during various natural disasters.

An accurate prediction of the remaining useful life (RUL) of lithium-ion batteries is of great practical significance. In this work, a RUL prediction method for lithium-ion batteries is proposed. Firstly, the battery data are considered as matrices and the health indicators (HIs) are extracted using SVD that does not require parameter settings and is adaptable for different lithium-ion battery datasets. Meanwhile, considering that the degradation information contained in the matrices has an important impact on the ability of extracted HIs to quantify the degradation status of battery, the original measurement data is reanalyzed and the feature extraction object containing more degradation information are obtained. Then, the strongly correlated HIs are selected by spearman correlation analysis and degradation trend analysis. After that, the GPR algorithm is used and the final prediction results with uncertainty expression are obtained. Finally, the feasibility and validity of the proposed prediction model are verified by four batteries provided by NASA. And an additional battery dataset from MIT is selected to verify the adaptability of feature extraction method. From the experimental results, it can be seen that the proposed RUL prediction framework has satisfactory prediction performance and the SVD feature extraction method has good adaptivity. The HIs extracted in this paper show a significant improvement in prediction accuracy compared with those extracted from measurement data.