神经网络算法及对未来一月的天气状况预测

Neural network algorithm and forecast of weather conditions in the coming monthAbstract With the improvement of the meteorological data observation methods and technology,the accuracy of each type of data obtained or obtained has been better surpassed,and the predictionof meteorological temperature is the most important need to be solved in the current forecast dataOne of the problems.Regarding how to choose to predict the weather conditions within a month,this time using neuralnetwork method to solve this problem,because this method has a strong adaptability in dealing withnonlinear problems,and it is perfectly suitable for the weather with great uncertainty Data research.First use MATLAB software to build a weather index model,most of which use the neural networktoolbox to achieve most operations,and then use BP and RBF algorithms to improve its predictionaccuracy,the use of these two different algorithms in neural networks It proves that this method isvery effective,and the weather forecast based on neural network can play a very good role in airpollution and local precise monitoring for the society,and give a new solution to the detection ofworld meteorology.Throughout the construction of the entire model and the design of parameters,the issue of howto monitor the meteorological temperature in the next month is continuously analyzed.Not only isthe theory of deep learning and meteorological prediction organized and optimized in this article,but also the experimental results have proved that The traditional shallow neural and vectormachine approach is more accurate.Subsequently,the traditionally used vector machine model isintegrated into the neural network method to improve it,thus greatly improving the efficiency of thealgorithm.Keywords:Deep leaming;Weather forecast;Neural Networks