Environmental impact assessment model of overall land-use planning based on BP artificial neural network

In view of the problems existing in the evaluation methods of land overall planning, such as excessive evaluation error and large root mean square error, this paper proposes to design an effective environmental impact assessment model for land overall planning. The influence of social, economic and ecological environment in the overall land planning was analysed to build the environmental impact index system of the overall land planning. GRNN was used to screen the indexes with different degrees of influence, MIV was introduced to screen the index values, and rank correlation coefficient among the indexes was controlled. The indexes obtained after screening were input into BP artificial neural network to build an environmental impact assessment model of overall land planning. The experimental show that the mean absolute error is about 1.6%, and minimum root-mean-square error is about 0.01.