The Reservoir Operation And Flood Disaster Evaluation Theory Method Based On Chaotic Evolution

As a important component part of water resource system, flood possesses dual attribute of hazard and profit. It is a penetrating contradiction of flood disaster and water resources shortage, and is a theoretical and practical issue how to solve this contradiction reasonably. Furthermore, the transformation mode of flood disaster and resource is one of major content of flood resource management. Hence, in view of the characteristics of flood resource, this article presents several population evolution algorithms based on chaotic searcher for dealing with four objects including profit (power generation scheduling), prevention(flood control), flood classification and flood disaster grade evaluation, and obtains some theoretical and practical results. The research work and innovation achievements can be described as follows:(1) Aiming at the problem of cascaded hydropower system optimal dispatchling, we propose a novel chaotic particle swarm optimization (CPSO) algorithm and compares the efficiency of three one-dimensional chaotic maps within symmetrical region. This approach, which process chaotic search nearby the optimal solution computed by PSO algorit hm, improves the global optimization capability of CPSO algorithm in order to solve the problem that particle swarm optimization (PSO) algorithm is easy to trap in local optima and logistic map can't search in the negative region effectively. Moreover, a piecewise linear interpolation function is employed to transform all constraints into restrict upriver water level for solving the objective function. Numerical results and comparisons demonstrate the effect and speed of different algorithms on the Three Gorges cascaded hydroelectric system(2) In order to solve the problem of reservoir flood control, a chaotic particle swarm optimization (CPSO) algorithm based on the improved logistic map is presented, which uses the discharge flow process as the decision variables combined with the death penalty function. According to the principle of maximum eliminating flood peak, the flood control operation model has been eatablished with the goal of minimum the discharge flow. The case of two typical floods on Gorges reservoir compares the effect of proposed model and algorithm with existing models and algorithms. Moreover, the risk rate that discharge flow overtakes the safety standard under the different accuracy of flood forecasting is investigated while the reservoir encounters the over-level flood.(3) Flood disaster is a kind of frequent natural hazards. The objective of flood classification is to establish hazard assessment model for managing flood and preventing disaster. Base on the chaotic optimization theory, a chaotic differential evolution algorithm (CDE) is proposed to solve a fuzzy clustering iterative model for flood disaster classification. By using improved logistic chaotic map and penalty function, the objective function can be solved more perfectly and ave taken into account two practical flood disaster cases so as to test the effect of novel hybrid method. Simulation results and comparisons show that the chaotic differential evolution algorithm is competitive and stable in performance with simple differential evolution and other optimization approaches presented in literatures.(4) In allusion to the problem of evaluating flood disaster grade, we establish the projection pursuit polynomial function model, and propose a chaotic differential evolution (CDE) algorithm using piecewise linear chaotic map for optimizing the projection index function, and adopt penalty function method of dealing with constraints to detect optimal projection direction. The coefficients of polynomial function are solved by projecting the index boundary values of flood disaster grade table so as to implement the division of continuous flood disaster grade. The calculation results of two example show this scheme is effective and easy to implement in comparison with other models and algorithms, which can be applied to many scientific field.