The porous medium approach is widely used to represent high-resistance devices, such as catalysts, filters or heat exchangers. Because of its computational efficiency, it is invaluable when flow losses need to be predicted on a system level. One drawback of using the porous medium approach is the loss of detailed information downstream of the device. Correct evaluation of the turbulence downstream affects the calculation of the related properties, e.g. heat and mass transfer.
The novel approach proposed in the current study is based on a modified distribution of the resistance across the porous medium, which allows to account for the single jets developing in the small channels, showing an improved prediction of the turbulence at the exit of the device, while keeping the low computational demand of the porous medium approach.
The benefits and limitations of the current approach are discussed and presented by comparing the results with different numerical approaches and experiments. The flexibility of the proposed approach in terms of describing the device geometry is demonstrated via an optimisation study where the size of the monolith channels is modified to obtain a more uniform distribution of the flow. With the rapid development of additive technologies, the proposed method offers endless possibilities for catalyst and similar device design improvement.
Although here the new approach is applied to a monolith commonly used in automotive exhaust after-treatment systems, it can be generalized to other high resistance devices with multiple flow passages.