Analysis Of Steady And Transient Fluid-solid Coupling Heat Transfer Simulation For Turbocharged Diesel Engine Exhaust Manifold

Affecting the gas exchange process and determining the reliability of the turbocharger as well as its efficiency directly, the exhaust manifold has taken the role of a key component of the supercharged engine’s after-treatment system. Nowadays, on account of the engine’s specified power design, the exhaust gas temperature has a steep rising, resulting in an issue of operational reliability with respect to the manifold’s material and structure. Consequently, the numerical simulation and analysis of fluid-solid coupling are essential in the preliminary process of a new manifold design. The conventional form of flow heat transfer simulating is kinda of transient or steady. How to properly model the complex exhaust flow and figure up the heat flux between gas and pipe is an intractable matter for CFD engineers, a number of who prefer the former consider that the transient simulation is identical to the actural flow and the steady one is nonobjective, while the rest of who take the opposite view and hold that the transient one is in need of more assumptions and higher computing resource, which is very prone to errors.Actually, macro-parameters like velocity and temperature are to some extent the macro-reflection of the exhaust gas’micro-parameters. The reliable simulation depends on modeling the domain in the use of discrete distribution and discrete equations that should reflect the characteristics of the flowing, and the optimized simulation is to properly weaken equations’ solving difficulties that exactly refer to equations’nonlinearity and compressibility. However, a large number of engineering calculations have put the cart before the horse, excessively simplifying the puzzles and even proposing a distorted simulation result that fails to withstand close scrutiny in either a transient model or a steady one. As to the flow heat transfer simulation, the general problem is a certain degree of confusion of combination among discrete distribution, equation of state, viscous fluid-flow model, boundary condition and solution method. The key point is to work out a precisely estimation of the exhaust flow determined by those above aspects. The root or radical measure is to constitute and optimize a reasonable model based on computational fluid mechanics that is viable as well as theoretical.This paper is based on an exhaust manifold finite element analysis program, which proposed an issue of reliability about the numerical simulation of fluid-solid coupling. A series of models including both transient and steady ones have been put up and compared, aimed at establishing proper strategies which should be calculable and businesslike. Naturally, the association of discrete distribution, equation of state, viscous fluid-flow model, boundary condition and solution method cannot be more clearly outlined. On one hand, some common models have been run and analyzed, which have an obsession with simplifying calculation and ignore the relationship of key points for the most part; On the other hand, according with the well-posedness of fluid mechanics theory and software, the impact of discrete distribution, equation of state, viscous fluid-flow model, boundary condition and solution method of a model has been illustrated step by step in an easy-to-understand way. At the same time, strategies for systematic modeling and simplifying have been carried out, with verified via temperature tests. It turns out that the usual transient model of the exhaust flow is very prone to errors aroused by viscous fluid-flow model, discrete distribution as well as time step in consideration of limited computational resource, so the usual contours are higher than the actual when key points are valued and computational resource is guaranteed. If not, the transient makes a nonsence. To sum up, it is the exhaust gas defined with piecewise-linear properties and the solution method applied with pressure-based solver that are to be trusted, compared with those above optimized approachs. Furthermore, the steady model can reach the requirements as long as the outlined association of key points are appreciated and put a high value on.