The invention discloses a generalized-entropy-theory-based dynamic intelligent comprehensive analysis method for water environment risk management. According to the method, on the basis of integration of the hydro science, intelligent science, information science, and risk theory, risk factors of the water yield, water quality, and water ecology and the like in water environment risk management are explored from the perspectives of the integrated structure and components; and unified description and physical parsing are carried out on various uncertain information in the system by using a general distribution function and a generalized entropy theory and thus a generalized-entropy dynamic intelligent comprehensive analysis method for water environment risk management is constructed based on the genetic algorithm, fuzzy analytic hierarchy process, and projection pursuit. With the method, a simulation result that is obtained by analyzing the water environment risk meets an actual measurement early-warning result; and the accuracy is high. Therefore, the water environment risk can be managed by using indexes reasonably to the certain extent. The method can be applied to researches of the water environment evaluation and has the important significance in scientific early warning of the water environment.