Research On Genetic Algorithm To Simplify The Mechanism Of Reaction Kinetics

With the development of CFD technology, numerical simulation of the fire made rapid development. In order to accurate simulation of fire phenomena, should be able to more accurately describe the use of combustion system of the multi-step, multi-component chemical reaction mechanism in detail. However, the combustion of chemical reactions are complex, often by hundreds of elementary reaction composed of several to dozens of related intermediate products. Although the level of development as the computer, the use of detailed reaction mechanism can be achieved has been calculated, but in reality, due to the computer's memory capacity and computing speed limits should be coupled to the combustion reaction mechanism in the numerical calculation of fire, is still unrealistic. Therefore, Reasonable to simplify the complexity of the reaction mechanism, has become the key to the fire numerical simulation.In recent years, emerging means of the use of mathematical methods to simplify the reaction mechanism, such as the computational singular perturbation theory, intrinsic low-dimensional manifolds technology and auto-adapted schedule method and so on. When the detailed reaction mechanism can be simplified, the simplification of the reaction mechanism is often difficult to determine the coefficients, for which the application of genetic algorithm in this article to simplify the reaction mechanism of the optimization and to determine the coefficients in order to simplify the reaction mechanism allows the application can also obtain a detailed reaction mechanism in combustion the process of fine structure and the flow field in the generation of pollutants and so on the same predictable results.Optimized to simplify the results of the reaction mechanism, can be applied to numerical simulation of fire and fire-fighting to the development of clean and highly efficient method has great significance.In this paper, is mainly applies the genetic algorithm optimization and the determination hydrogen/air, the methane/air combustion simplification mechanism's reaction rate. Based on methane 19-step reduced chemical reaction kinetics, using genetic algorithm to optimize the methane/air reduced reaction mechanism's coefficient and obtaining the optimized reduced mechanism. Contrasts the two mechanism's PSR model and PREMIX model result and the detailed mechanism's result, and the result showed that, the optimized mechanism computed result is more accurate, can achieve the detailed mechanism's precision in certain performance.Genetic algorithm combined with Perfectly Stirred Reactor model to determine the reaction rate and reaction order of methane two-step mechanism. Compared with methane detailed reaction mechanism's result which were obtained through PSR and PREMIX model, the new methane two-step mechanism based on GA have high accuracy in computation the species concentration of O2,CO2 and CO. Generally speaking, it improves a lot than the previous methane two-step mechanism.