The Revised Research And Visualization Technology Of The Meteorological Precision Grid Point Prediction Data

On the basis of the National Meteorological Center guidance forecast after cut,using the moving average method,the first order adaptive Kalman filtering method,the optimal dynamic deviation correction method and the optimal dynamic deviation correction set SVM correction method,this paper corrects the 003 limitation grid temperature forecast in the range of 104.15 °E-107.80 °E to 35.10 °N-35.10 °N from October 1,2017 to November 30 at 08,and compare the results of the correction.The results show that these revised methods have a good effect,the temperature prediction error is decreased after correction,and the prediction accuracy is improved to a certain extent.But compared to the effect of several correction methods,one of the most obvious being the moving average method,second for optimal dynamic deviation correction method,the third for the first order adaptive kalman filtering method,relatively poor for the optimal dynamic deviation correction set SVM correction method.The correction results of moving average method and optimal dynamic deviation correction method for reduction fluctuate greatly,and appear a small part of negative effect,the results of the first-order adaptive Kalman filtering method and the optimal dynamic deviation correction collection SVM correction method are relatively stable,and less affected by individual data.It can be seen that the optimized method is a first-order adaptive Kalman filter method.Therefore,it is effective and efficient to use the historical data of adjacent forecasts to correct grid prediction.In addition,the optimal correction method has been implemented by program.The revised method was written in the Java language,and the development of the program was realized through the Eclipse tool,and the graphical display of the revised grid prediction was realized by means of WebGIS.At the same time,this paper has realized the visualization of grid prediction data,which mainly includes multiple grid prediction view,subjective correction,single point view and multi-category grid prediction automatic test.Mainly through the JavaScript scripting language for web development,a variety of grid forecast view and subjective correction was realized by using the key technology such as localStorage cache method and Ajax provided by ArcGIS API for JavaScript,HTML 5,and also,a single point of view and more categories grid forecast automatic inspection was realized with the aid of Echarts rich charting library.