基于复杂度指数的GA-SVR及其在气体水合物中的应用

摘要

To solve the difficulty in determining optimum craft parameters of the gas hydrate desalination effectively ,a complexi-ty index based GA-SVR model denoted as CI-GA-SVR was proposed .The GA-SVR model’s theoretical principle and experimen-tal results were studied ,and in view of the insufficiency in the model ,the complexity index of regression problems was defined to optimize the GA-SVR model .By studying the influence rules of different complexity index on the SVR parameters ,the CI-GA-SVR model was constructed .Finally ,with examples of surfactant affecting the gas hydrate ,the simulation experimental results show that the model and algorithm are feasible and effective .%为有效解决气体水合物海水淡化工艺中最佳参数难以确定的问题，提出基于复杂度指数的遗传算法优化支持向量回归模型（CI-GA-SVR）。在分析遗传算法优化支持向量回归模型（GA-SVR）的理论原理及其实验效果的基础上，针对模型中存在的不足，利用回归问题的复杂度指数对其进行进一步优化；通过研究不同复杂度指数对SVR参数的影响规律，构建CI-GA-SVR模型。结合表面活性剂对气体水合物生成影响的实例，对该模型及算法进行仿真，仿真结果表明了该模型的可行性和有效性。