Analysis Of Grain Production Fluctuations And Prediction In Grain Yield Trend Of Qitai

Based on relevant information and data from Xinjiang Statistical Yearbook(Year 1985～2007), social and economic statistical information in Qitai County(1949～2007) and Land and Resources Bureau of Qitai County, this essay chooses Qitai County in Xinjiang Uygur Autonomous Region as the regional object of study. Through analysis of the background of grain production in Qitai County as well as the collecting and processing of historical data, this essay analyses the history, present situation and driving factors of fluctuations in grain yield in Qitai County. At the same time, this essay adopts the methods of wavelet analysis, artificial neural network and grey prediction model to make a qualitative and quantitative prediction of grain yield trend in Qitai County and further propose some constructive comments on how to stabilize grain production in Qitai County. The conclusions are as follows:(1)The fluctuation of the grain yield, the total grain yield and sown area in Qitai County can be divided into two stages: 1949～1983 and 1984～2006.(2)The yield per unit, total yield and sown area in Qitai County all belong to typical classical fluctuations. The yield per unit fluctuates regularly. The fluctuating length from 1949 to 2006 is from 2 to 5 years, with an average of 3.5 years. The fluctuating margins are violent, with an average of 46.39%. The fluctuating margins concerning sown area are from 15.15%～39.79%, with an average of 27.97%. The fluctuating length is irregular, with a relatively small wave trough and crest. The average fluctuating length of total yield is 3.8 years. The fluctuating margin is comparatively violent, with an average of 58.13%.(3)Cluster Analysis is adopted to study the changing rule of grain production. The changing process of the 58-year grain production can be divided into two stages: from 1949 to 1983--the first time period; from 1984 to 2006--the second time period. Principal Component Analysis is adopted to sift driving factors. In the first time period, agricultural system policy and population factors are the main driving factors for grain production; in the second time period, scientific and technological advances in agriculture and social and economic development are the main driving factors in grain production.(4)Wavelet analysis shows that the total grain yield in Qitai County has obvious annual change and stage features. From 1970 to 1990, 3-year, 6-year and 12-year characteristic scales are all reflected to some extent. The 3-year and 6-year time scale after 2006 will keep improving or remain in a relatively high level. The 12-year time scale after 2006 will decrease.(5)The predicted result of the trained artificial neural network is more accurat, in 2004 and 2005 , the errors between predictive value and the actual value is -0.01868,0.026529. The predicted total grain yields in 2010 and 2020 are 29.54×104t and30.43×104t。(6)The average relative error of "the optimal grey prediction model of embedded knowledge" is 2.34%. The 2010 total grain production from extrapolation is 31.54×104t。...