Research On Precipitation Data Analysis Algorithm Based On Machine Learning

Precipitation is a very important topic in weather forecasts.Weather forecasts,especially precipitation prediction,poses complex tasks because they depend on various parameters to predict the dependent variables like temperature,humidity,wind speed and direction,which are changing from time to time and weather calculation varies with the geographical location along with its atmospheric variables.To improve the prediction accuracy of precipitation,this context proposes a prediction model for rainfall forecast which is based on Support Vector Machine with Particle Swarm Optimization(PSO-SVM)to replace the linear threshold used in traditional precipitation.This model is proposed to replace the traditional linear threshold method which is used in traditional precipitation forecast.Parameter selection has a critical impact on the predictive accuracy of SVM,to solve this problem,we use PSO to find the optimal parameters for SVM.The PSO-S VM algorithm was used for the training of a model by using the historical data for precipitation prediction,which can be useful information and used by people of all kinds of occupations to make wise and intelligent decisions.The simulations demonstrate that prediction models indicate that the performance of the proposed algorithm has much better accuracy than the direct prediction model which is based on a set of experimental data if other things are equal.On the other hand,simulation results demonstrate the effectiveness and advantages of the SVM-PSO model.Based on the above research,in order to improve the efficiency of the model,Deep belif networks are used to analyze and predict meteorological data with large data size.The analysis results show that the model is no longer rebuilt when the model is applied,due to the use of batch processing technology in data processing,DBNs reduce the time of model establishment and it is more efficient when being applied to large-scale datasets.