The Application And Improvement Of Invasive Weed Optimization

Intelligent optimization algorithm has been widely applied to many fields because it has lots of advantages, such as strong robustness, fast convergence speed and high quality of solution. Therefore, the research of intelligent optimization algorithm is very important.This article mainly researchs the application and the improvement of invasive weed optimization. Invasive weed optimization algorithm is applied to determine the parameters of the flow of aquifer and the parameters of one-dimensional hydrodynamic dispersion experiment. In order to improve the defect of invasive weed optimization and increase the performance of the invasive weed optimization, mutation strategy is used to improve the algorithm. Improved invasive weed optimization algorithm was applied to determine the parameter of the anisotropy aquifer. The main contents are as follows:1、Introduce the origin of the invasive weed optimization algorithm and invasive weed optimization algorithm is introduced in the domestic and foreign research situation. The principle of invasive weed optimization and its basic steps are introduced detailedly. Analyze the advantages and disadvantages of invasive weeds optimization algorithm.2、Invasive weed optimization algorithm was applied to determine the parameters of the flow of aquifer and the parameters of the one-dimensional hydrodynamic dispersion experiment. Comparing the results of calculation with other results and comparing the calculating experiment data with actual data, these show that the results of invasive weed optimization algorithm are reliable. Invasive weed optimization algorithm can get more precise results in a relatively short time. The rang of initial estimated parameters have slimly effect on rate of convergence and running time, but it does not affect the invasive weeds optimization calculation results and the convergence of the algorithm.3、Aiming at the shortcomings of the invasive weed optimization algorithm, the mutation strategy is used to improve the algorithm. The gaussian chaotic mutation operator and differential evolution mutation operator and their hybrid mutation operator are applied to improve the invasive weeds optimization algorithm. In testing function and determining parameters of anisotropy aquifer, numerical experiment of the three kinds of improved weed optimization algorithm show that the mutation strategy which are used to improve the invasive weed optimization algorithm are effective and feasible from three factors which are the calculation results and the running time and two parameters of the algorithm.