With the advancement of technologies such as sensor networks and global positioning systems， the volume of meteorological data with both temporal and spatial characteristics has exploded， and the research on deep learning models for Spatiotemporal Sequence Forecasting （STSF） has developed rapidly. However， the traditional machine learning methods applied to weather forecasting for a long time have unsatisfactory effects in extracting the temporal correlations and spatial dependences of data， while the deep learning methods can extract features automatically through artificial neural networks to improve the accuracy of weather forecasting effectively， and have a very good effect in encoding long-term spatial information modeling. At the same time， the deep learning models driven by observational data and Numerical Weather Prediction （NWP） models based on physical theories are combined to build hybrid models with higher prediction accuracy and longer prediction time. Based on these， the application analysis and research progress of deep learning in the field of weather forecasting were reviewed. Firstly， the deep learning problems in the field of weather forecasting and the classical deep learning problems were compared and studied from three aspects： data format， problem model and evaluation metrics. Then， the development history and application status of deep learning in the field of weather forecasting were looked back， and the latest progress in combining deep learning technologies with NWP was summarized and analyzed. Finally， the future development directions and research focuses were prospected to provide a certain reference for future deep learning research in the field of weather forecasting.

Focusing on the issue that the 3-point and 4-point hill climbing algorithms have high calculation and low efficiency in enhancing the nonlinearity of a Substitution box (S-box), an algorithm named Combination of Hill Climbing (CHC), which could apply multiple swap elements at a time, was proposed. The algorithm defined the behavior of swapping 2 output data of an S-box as a swap element, and used weighting prioritizing function to select swap elements that have larger contribution to the enhancement of nonlinearity, then simultaneously applied multiple selected swap elements to enhance the nonlinearity of an S-box. In the experiments, a maximum of 12 output data were swapped at a time by using the CHC algorithm, and most of the random 8-input and 8-output S-boxes' nonlinearity surpassed 102, with a maximum of 106. The experimental results show that the proposed CHC algorithm not only reduces the amount of calculation, but also enhances the nonlinearity of random S-boxes more significantly in comparison with the 3-point and 4-point hill climbing algorithms.