How to design and implement the graduation thesis of China Weather Quality Data Display and Analysis System?

With the development of meteorological data acquisition technology and the improvement of meteorological center algorithm, the accuracy of numerical weather forecast is constantly improving. However, in the face of the demand for high-precision weather forecast in today's society, the model weather forecast from various meteorological centers can no longer meet people's requirements. Therefore, meteorological departments in various regions need to re-forecast the future weather phenomena in their own areas according to their own conditions and refer to the guidance and forecast of their superiors, so as to improve the forecast accuracy. However, due to the limitation of its own equipment, it is difficult for regional meteorological departments to obtain local meteorological data with guiding significance. Based on the above reasons, this paper designs and implements the "Zhejiang Weather Forecast Data Analysis System" by investigating the actual forecast work of Zhejiang Meteorological Observatory and combining with the quality inspection requirements issued by the National Bureau, so as to solve the practical problems encountered in the weather forecast process. Through the analysis of the system result data, the weather forecast module based on neural network method is designed and implemented. This module will add the weather forecast results as a new forecast model to the data analysis system, providing more data support for the weather forecast work. The main contents of this paper are as follows:

1. According to the data requirements of the meteorological forecast work of Zhejiang Meteorological Observatory, through the analysis of the existing meteorological data structure, a set of meteorological data warehouse with long time span and high query efficiency is designed and implemented. Through the implementation of this data warehouse, the annual data statistics time is reduced from several hundred seconds to several seconds.

2. Design and implement a meteorological data analysis system. A data analysis system with rich and intuitive data expression is developed by using C#, HTML, JavaScript and T-SQL voice, which has the ability of data mining, quality inspection and auxiliary prediction. Through the application of this system, the forecast accuracy of weather forecasters can be effectively improved.

3. A data mining model based on neural network is proposed to construct numerical reference data of weather forecast. The BP neural network method is used to build the model, and through the continuous learning of historical data, the errors of each reference prediction model are predicted, and then new reference data are given according to the prediction results as new models for forecasters' reference. Compared with the data of other models, the reference data greatly improves the accuracy and stability of prediction. The application practice of meteorological numerical forecast analysis system of Zhejiang Meteorological Observatory proves that the meteorological numerical forecast data analysis system designed and implemented in this paper can meet the demand of numerical forecast team for forecast error control. Not only the predicted reference data calculated according to the data learning model has good accuracy. Moreover, the provided quality inspection function also provides strong data support for the forecasting work of the numerical forecasting team.