Improvement Of Ant Colony Algorithm And Its Application In Water Resources System

Water resources system engineering existed a great deal of nonlinear,multi-dimensional combination optimization. Traditional solutions were analytical and exhaustive method, but these methods had different disadvantages. This subject could seek more effective and convenient solutions. Ant colony algorithm was a global optimization method, and was improved on previous research foundation. It had been established that new models of ant colony algorithm in water resources optimization and prediction, and had been implemented by MATLAB language. The model could better solve multi-element complicated problems of water resources system in practice and effect was good.This paper obtained three aspects of achievements below:(1) this paper introduced a model of ant colony algorithm of local randomness search strategy, which the model could improve solving precision and searching efficiency to common function optimization. The model had been applied successfully in critical depth calculation for section in ditch and computing water depth of section downstream overflow dam, and effect was better, precision was higher.(2) combination point had been found that ant colony algorithm and Marr wavelet neural network. Multi-dimensional ant colony algorithm in continuous spaces insteaded gradient descent method, and adjusted weight values of BP network and expansion and translational coefficients of wavelet function, which Marr of it insteaded S-type activation function in BP network, and population standard deviation in wavelet network was used as objective function of ant colony algorithm. The results of the model was satisfied in forecasting rice water requirement.(3) ant colony algorithm had very good ability of global optimization. Continuous multi-dimensional ant colony algorithm—genetic algorithm had been proposed on real coding based accelerating genetic algorithm. The new model had been applied in irrigation regime of rice at Chahayang irrigation area. Calculating and practical water were same basically, which proved that the model had very good practical value.Based on the above established model, the theory and the practice had been associated organically. The research result offered new method for the complex non-linear problem in the domain of hydrology and water resource. At the same time, the applied range of ant colony algorithm model was amplified. Moreover, the theory was improved accordingly.