The new optimization algorithm, wind-driven optimization(WDO)algorithm, is simulated and verified by using six typical test functions. The simulated results are compared with the optimization ones by the cuckoo search(CS)algorithm, differential evolution(DE)algorithm, shuffled frog leaping algorithm(SFLA), particle swarm optimization(PSO)algorithm, ant colony optimization(ACO)algorithm, imperialist competitive algorithm(ICA)and simulated annealing(SA)algorithm. The carrying capacity of water resources of Wenshan Prefecture from 2006 to 2015 is calculated by the wind-driven optimization-projection pursuit(WDO-PP)model by selecting 10 evaluation indices such as development and utilization of water resources and precipitation. Its change trends and the carrying capacities of water resources in the planning years of 2020 and 2030 are predicated and evaluated. The results show that: (1)For either unimodal or multimodal functions, the optimization results by the DWO algorithm is much better than those by the CS, DE, SFLA, PSO, ACO, ICA and SA algorithms, and it has better optimization accuracy, convergence rate, excellent capability of seeking the extreme values and convergence stability and reliability. (2)The evaluated results of carrying capacity of water resources of Wenshan Prefecture by WDO-PP model from 2006 to 2014 are basically capable(grade Ⅲ), and those in 2015 is capable(grade Ⅱ). The carrying capacity of water resources increases with time, and its trend is significant. The predicated results of carrying capacities of water resources in the planning years of 2020 and 2030 are capable(grade Ⅱ)and absolutely capable(grade Ⅰ), respectively.