Intelligent integrated model for forewarning evaluation of watershed water security

In the paper,the accelerating genetic algorithm based on the fuzzy analytic hierarchy process is used to filtrate the index system and determine the weights of both the indexes and the subsystems of the water security forewarning evaluation system,the back-propagation neural networks updating model is used to predict the values of the evaluation indexes,and the set pair analysis method is used to determine the relative membership degree function values of the variable fuzzy set of the assessed samples according the criterion grades of watershed water security,and the intelligent methods integrated forewarning evaluation model(IMI-FEM)is established for the watershed water security. The results of IMI-FEM applied to evaluating the water security system of Chaohu watershed in the following future five years show that the security states of its water resource security sub-system,water environment sub-system and water disaster sub-system are 2.73～2.74,3.30～3.32,and 2.54～3.43 grade respectively,the security state of its society&economy subsystem is 3.33 grade,the whole state of the watershed water security system is 2.98～3.20 grade in the critical security state,and the restrict indexes of Chaohu watershed water security system are as follows:the water resource utilization ratio,the water resource amount of per person,the water resource amount of unit area,the water resource demand amount of per ten-thousand Yuans of agriculture production value,the fertilizer wastage of unit area,the water resource demand rate of eco-environment,the environment protection investment of unit GDP,storage capacity of water conservancy projections of unit area,the water conservancy projection investment of unit GDP, the regional exploitation exponent,the population density,and the science&technology investment of unit GDP.