Research On Optimization Of Heilongjiang Farming Cultivation Based On Artificial Neural Network

With the largest number of population in the world, agriculture is the industry related tothe people’s livelihood and even national security in china. Heilongjiang is a big agriculturalprovince, and its ultimate grain production in2011was55.705billion kilogram, therefore, itbecame to be the first highest grain yield province. As the basis of agriculture in Heilongjiang,reclamation economy receives much concern. Heilongjiang agricultural reclamation economyhas a good momentum of development at present, meanwhile, it also faces realistic problemof structure adjustment. Reclamation economy system in Heilongjiang province is studied byusing artificial neural network method and cobb-douglass formula, and decision scheme ofinvestment optimization for further developing economy is given in this paper. Therefore, theoptimization of industrial structure of Heilongjiang agricultural reclamation economy isrealized.According to nonlinear characteristic of input-output relationship of Heilongjiangagricultural reclamation economy, the artificial neural network method is used to modeling tosimulate such nonlinear mapping in this paper. The linear mapping between Neurons oftraditional BP algorithm is changed to nonlinear, and an improved BP algorithm is proposed.Numerical results show that the improved algorithm is effective for the simulation ofinput-output relationship of Heilongjiang agricultural reclamation economy.As a basic work of modeling the Heilongjiang agricultural reclamation economy system,the investment decision status of reclamation economy system in Heilongjiang province isevaluated by using Data Envelopment Analysis(DEA) in the paper. Based on analysis andresearch of input-output situation of agricultural reclamation area in Heilongjiang province,DEA evaluation model is established for production situation by selecting time decisionmaking units and input-output indexes. By solving DEA evaluation model, the analysis ofrelative efficiency of input-output between internal industries of reclamation economy systemin Heilongjiang province is given in the paper.According to the conditions of capital, labor and output of Heilongjiang agriculturalreclamation economy, this paper uses cobb-douglass productive function to analyze and study. For the convenience of application, cobb-douglass productive function is transformedmathematically. The productive function is refined to primary industries, secondary industriesand tertiary industries for farm system in Heilongjiang province, which is simulated byartificial neural network method. The self-feedback loop RNN model is established forHeilongjiang agricultural reclamation economy. Numerical results show that this method iseasy to be realized, and has the characteristics of good simulation effect and high predictionprecision.An important work in this paper is to analyze the effect of industrial structure ofreclamation economy system in Heilongjiang province on economic benefit. The mappingrelationships between investment and total output value of Heilongjiang agriculturalreclamation economy in primary industries, secondary industries and tertiary industries areestablished by using BP network model. Nonlinear programming model for reclamationeconomy system in Heilongjiang province is established, in which the objective function istotal output value and the independent variables are three industries investments. Optimalinvestment strategy of three industries for the maximum total output value is gained finallyby optimal search. Numerical results show that the model is effective.