Grain Output Prediction Of Gray-Markov Model

The food problem has been an important issue related to the survival and development of our country.China is not only a large population,but also a large agricultural country since ancient times.food has been closely related to the lives of the people,which is not only related to the development of the national economy,but also related to the stability and security of the country,so food is crucial to our country.At present,due to the pressure of population,food is always in a state of tension.Although China has a lot of land,the arable area is very small,so it is very important to predict the grain yield.The research on the law of grain yield and its change can provide reliable basis and guarantee for the relevant government departments to formulate agricultural policies.In view of the total grain yield of Henan and the related factors affecting grain yield,this paper studies the grain yield forecast based on the gray-Markov model.Grey system theory is an emerging theory.It is mainly used to study small sample and poor information data,which can reveal the overall law of the development of things.It is widely used in agricultural science.The main prediction model of this system is GM model.Due to the many factors affecting grain production and in order to find out the main influencing factors,we first use the classical Deng's correlation degree in the grey system theory to analyze the influencing factors affecting grain yield.According to the correlation degree from large to small,the main four influencing factors were selected,and the gray GM(0,4)model was established by using the grain data and the four main influencing factors to predict and analyze the total grain yield data of Henan.The grey GM(0,4)model can basically reflect the overall trend of food production data,but does not reflect the volatility and instability of food data.So as to reflect these characteristics of the food data and improve the prediction accuracy,Markov chain was introduced later.it predicts the state of the data at the next moment based on the position and trend of the system data which can well reflect the volatility and instability of the data.The gray system is combined with the Markov chain to establish a gray-Markov model for prediction and correction,which improves the prediction accuracy and reflects the volatility and instability of the grain data.Finally,in order to further improve the accuracy of prediction,a genetic optimization algorithm is introduced to establish a genetic algorithm-grey Markov model for correction.The algorithm has wide coverage,excellent parallelism,self-learning and self-adaptation,which is beneficial to parameter optimization and further improve prediction accuracy.The experimental results show that the genetic algorithm-gray Markov model has better prediction accuracy than the gray GM(0,4)model and gray Markov model.