Research And Application Of Rice Yield Prediction System Based On GIS And Time Series Algorithm

China has a population of more than 1.3 billion food production has an important role in the decision for subsistence and development of human beings.Rice is one of the three major food crops throughout the world, and about 60% of the world population take it as basic food.Rice production plays an important role in China’s grain production, so it is very important to accurately predict the grain security of rice., and it affects macro rice planting decisions made by the government.With the deepening of the research of statistics, the time series algorithm has become a hot research direction. The continuous rise of computer software and hardware technology, geographic information system has made great achievements in the spatial data model, spatial data analysis, decision support and spatial visualization and other theoretical issues. Geographic information system is a set of query, localization, map and data analysis functions in one integrated system. In this paper, the rice yield prediction system is geographic information system(GIS) and yield prediction model of integrated application development. Through the acquisition of Heilongjiang area from 1991 to 2010, the rice planting area, total and per mu yield information. Rice planting situation and yield were obtained by early data processing, and the yield and average annual yield of rice in 1991 to 2009 were obtained. In this paper, the SAS software is used to analyze the time series of the rice yield data, which will be converted to a stationary sequence, after white noise processing, parameter estimation and so on.On the basis of obtaining the rice yield prediction model, the rice yield prediction system based on GIS and timing algorithm was developed by using MapObject component and C# language.. The system can realize the map information import, the output data display and query, the spatial data visualization. It is of great significance to the formulation of national macro policy.In this paper, taking rice yield prediction in Heilongjiang Province as focus, the research on rice yield prediction system based on GIS and time series algorithm is carried out around the time series algorithm. In addition, this paper describes the research status of geographic information system, spatial visualization technology, data mining technology, time series algorithm and a series of theoretical basis as well as the GIS system and SAS technology support.. Based on the analysis of the demand of rice yield prediction system, the design principle of the system is designed, and the functions of the system are designed, such as spatial location, data acquisition, database operation, function module design and so on.The main work is as follows: 1.Generation of time series of rice yield. 2. Processing time series data. Mainly using the SAS software for processing, in the time series of process through multiple differential points will be non stationary time is transformed into stationary time series, the timing diagram of function and autocorrelation function judgement P, D, Q value, model fitting, and the eventual establishment of a ARIMA model. 3. The forecast model of rice yield in Heilongjiang province was established. Due to the impact of the terrain of rice in Heilongjiang area, the impact of natural disasters, the acquisition of continuous, normal data is more difficult. The forecast model of rice yield in Heilongjiang province was established. Due to the impact of the terrain of rice in Heilongjiang area, the impact of natural disasters, the acquisition of continuous, normal data is more difficult.