A cfd simulation method based on gpf regenerated soot layer collapse process

Abstract

The application discloses a CFD simulation method based on GPF regeneration soot layer collapse process, and comprises a regeneration soot layer collapse process judgment step after the CFD simulation method ends the time step iteration step. The application accurately simulates the soot layer thickness change caused by the collapse process, and improves the prediction accuracy.

Description

Technical Field

[0001] This invention relates to the field of fluid mechanics, specifically to a CFD simulation method based on the collapse process of GPF-regenerated carbon soot layer. Background Technology

[0002] With increasingly stringent vehicle emission regulations, higher demands are being placed on the simulation accuracy of aftertreatment systems. For the most efficient particulate filter (GPF), high-precision CFD simulation of the combustion and regeneration process of accumulated soot is particularly important. Traditional GPF regeneration simulation methods, such as those described in Di Sarli, V. and A. Di Benedetto (2015). "Modeling and simulation of soot combustion dynamics in a catalyticdiesel particulate filter." Chemical Engineering Science 137:69-78., involve solving the governing conservation equations and the local soot concentration scalar equation within the computational domain. The flow resistance between the soot layer and the porous media of the filter wall is represented as a negative source term in the momentum equation. This flow resistance determines the flow field distribution during GPF regeneration, which in turn determines the convective heat transfer process of combustion, ultimately affecting the regeneration temperature and soot concentration distribution. Existing techniques, such as those described in reference Serrano, JR, et al. (2013). "Packed bed of spherical particles approach for pressure drop prediction in wall-flow DPFs (diesel particulate filters) under soot loading conditions." Energy 58:644-654., use the spherical packed bed assumption to calculate porous flow resistance. This assumption assumes that the soot layer and filter wall are filled with spherical particles in fixed spatial positions. As regeneration progresses, the microstructural parameters of the soot layer, such as porosity and permeability, gradually increase, characterizing the change in flow resistance. However, research shows that the change in flow resistance during GPF regeneration is more affected by the thickness of the soot layer than by the porous microstructure. During soot combustion, due to the uneven spatial distribution of soot consumption rate, the soot structure at the bottom layer is too "loose" to support the mass of the upper soot layer. The resulting "collapse" phenomenon causes the soot layer thickness to change significantly in a very short time, thereby altering the regeneration characteristics. As mentioned earlier, the existing soot layer spherical fixed bed assumption cannot simulate this collapse phenomenon, that is, it cannot simulate the soot layer thickness change caused by this phenomenon. Ultimately, this leads to inaccurate predictions of regeneration temperature and soot consumption patterns, and the simulation results cannot guide the design and selection of GPFs. Summary of the Invention

[0003] In order to overcome the shortcomings and deficiencies of the existing technology, this invention provides a CFD simulation method based on the collapse process of GPF regenerated carbon soot layer.

[0004] This invention addresses the technical problem that traditional CFD simulation methods cannot simulate the thickness reduction caused by carbon soot layer collapse during GPF regeneration, thus failing to accurately simulate the temperature and carbon soot concentration distribution. It provides a carbon soot layer collapse process simulation method based on C language and coupled to the basic CFD solver, which accurately simulates the carbon soot layer thickness change caused by the collapse process and improves prediction accuracy.

[0005] The present invention adopts the following technical solution:

[0006] A CFD simulation method based on the GPF regenerated soot layer collapse process is provided, wherein the CFD simulation method further includes a regenerated soot layer collapse process judgment step after the end of the time step iteration step.

[0007] Furthermore, the step for determining the collapse process of the regenerated soot layer includes:

[0008] Determine the collapse flag CF value. If it is 0, enter the carbon soot layer mesh loop. If it is 1, exit the mesh loop and enter the start time step iteration step in the CFD simulation method.

[0009] Enter the carbon soot layer grid cycle, set the center grid Cell. central That is, the current circulating grid, the grid cell. neighbor To determine the relative positional relationship between the central grid and adjacent grids, the following steps are taken:

[0010] If the grid surface is the interface with other computational domains, there is no carbon soot mass exchange; otherwise, the relative positional relationship between the central grid and adjacent grids is determined. The relative positional relationship includes the vertical relationship and the horizontal relationship. Only the vertical relationship may have collapse.

[0011] If the central grid and its adjacent grids are in a vertical relationship, then the process continues to determine whether the collapse condition is met. The specific collapse condition is as follows:

[0012] Scenario 1: Porosity ε of the upper grid neighbor Greater than the critical value ε limit1 This indicates that the carbon soot structure is too loose to support its own structure, and the porosity ε of the central grid is also high. central Greater than the critical value ε limit2 This indicates that the central grid has enough space to accommodate the mass of collapsed soot;

[0013] Scenario 2: Porosity ε of the upper mesh neighbor Less than the critical value ε limit3 This indicates that the upper grid has a sufficiently large tendency to collapse, while the porosity ε of the central grid is high. central Greater than the critical value ε limit4 This indicates that the strength of the central grid carbon soot structure is insufficient to support the collapse trend;

[0014] Scenario 1 and Scenario 2 are ORed;

[0015] If the collapse condition is met, adding this constraint to Scenario 1 will cause the program to be unable to exit, resulting in an infinite loop. After the collapse occurs, for Scenario 1, update the carbon soot mass m of the central mesh. central_update =m central_old +m neighbor_old Update the soot quality m of adjacent grid cells. neighbor_update =0, for scenario 2, update the central soot mass m central_update =0, update the soot mass m of adjacent grid cells. neighbor_update =m neighbor_old +m central_old After a collapse occurs, the collapse flag CF is set to 1.

[0016] Furthermore, the four critical values ​​in the judgment criteria are porosity ε. limit1 ε limit2 ε limit3 ε limit4 The values ​​are 0.95, 0.5, 0.7, and 0.92 respectively.

[0017] Furthermore, the determination of the relative positional relationship between the central grid and adjacent grids specifically involves:

[0018] Calculate the unit vectors of the center point of the central grid and the center point of the surface;

[0019] Calculate the dot product of the unit vector connecting the center of the grid and the center of the grid surface with the vertical downward unit vector (0, -1). Determine the positional relationship between the central grid and adjacent grids based on the dot product result. If the dot product result is greater than the first threshold, the adjacent grid is considered to be below the central grid. If the dot product result is less than the second threshold, the adjacent grid is considered to be above the central grid.

[0020] Furthermore, the first threshold and the second threshold are 0.5 and -0.5, respectively.

[0021] Furthermore, the above method is coupled into the CFD solver using the C language.

[0022] Furthermore, the porosity of the upper grid must be less than 1, otherwise it will fall into a dead loop.

[0023] Furthermore, during the grid cycle, if at least one collapse is detected, the collapse marker is set to 1. After the grid cycle is completed, if the marker is still set to 1, it is considered that there is still a possibility of collapse, and the grid cycle continues until the marker is 0. At this point, the carbon soot layer is considered to have reached a "dense" state, the current grid cycle ends, and the next time step iteration begins.

[0024] A computer storage medium storing a computer program thereon, which, when executed by a processor, implements the CFD simulation method.

[0025] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the CFD simulation method.

[0026] The beneficial effects of this invention are:

[0027] (1) This invention uses the porosity of adjacent grids in the carbon soot layer computational domain as the criterion, and exchanges the carbon soot mass in the grid to simulate the collapse phenomenon of the carbon soot layer during GPF regeneration, thereby achieving the purpose of simulating the change in carbon soot layer thickness. The existing spherical fixed-bed filling assumption assumes that the resistance of the porous medium in the carbon soot layer is entirely determined by its microstructure, such as porosity, pore size, and permeability, making it impossible to simulate the change in carbon soot layer thickness. However, research shows that the flow resistance of the carbon soot layer is more affected by its thickness, thus limiting the simulation accuracy of traditional methods. To improve upon the shortcomings of traditional methods, this invention considers both the microstructure and thickness of the carbon soot layer, more realistically reflecting the change in flow resistance of the porous medium, and ultimately improving the accuracy of temperature and carbon soot concentration prediction during the regeneration process.

[0028] (2) The present invention uses the inner product of the vector connecting the center of the grid and the center of the grid surface and the vertical downward unit vector as the basis to determine the relative positional relationship of the grid. There is no possibility of carbon soot mass exchange in the left and right positional relationship, but there is a possibility of carbon soot mass exchange in the up and down positional relationship. The inner product threshold is set with a margin to eliminate the positional relationship judgment error caused by numerical error.

[0029] (3) After determining the grid position relationship, use porosity as the standard to judge whether the carbon soot mass exchange, i.e., collapse condition, is met between the upper and lower grids. By setting the porosity threshold, we can ensure normal grid loop exit and avoid dead loop.

[0030] (4) The above algorithm is coupled into the CFD solver using C language to simulate the carbon soot layer collapse process, solving the problem that traditional methods cannot simulate the carbon soot layer thickness change caused by collapse, and more accurately predicting the temperature and carbon soot concentration distribution during the regeneration process. Attached Figure Description

[0031] Figure 1 This diagram illustrates the grid relationship within the computational domain of the soot layer, showing the positional relationship between the central grid and adjacent grids.

[0032] Figures 2(a) and 2(b) are schematic diagrams of the collapse satisfying scenario 1 and scenario 2.

[0033] Figure 3This is a graph showing the carbon soot concentration at various axial positions of the soot layer during regeneration after the coupling collapse step.

[0034] Figure 4 This describes the overall solution process of the collapse step algorithm in the CFD solver.

[0035] Figure 5 This is a flowchart of the collapse process. Detailed Implementation

[0036] The present invention will be further described in detail below with reference to the embodiments and accompanying drawings, but the embodiments of the present invention are not limited thereto.

[0037] Example

[0038] like Figure 4 As shown, a CFD simulation method based on the GPF regenerated soot layer collapse process is presented. The CFD main program first adjusts the custom parameters, then solves the mass conservation, momentum conservation, energy conservation, and composition conservation equations in sequence. Next, it uses the corresponding turbulence model to solve the turbulence equation and the soot concentration scalar equation. Finally, it updates the mesh characteristic parameters such as physical property parameters and porous medium characteristic parameters. After the current time step iteration ends, it enters the soot layer collapse process judgment step and begins the soot layer collapse process simulation.

[0039] like Figure 5 As shown, the steps for determining the collapse process of the soot layer include:

[0040] First, set the Collapse Flag (CF) to 0. The purpose of this is to set CF to 1 when exiting the collapse model to end the mesh cycle, and set it to 0 when entering it again to perform at least one mesh cycle.

[0041] Determine the CF value. If it is 0, enter the soot layer mesh loop. If it is 1, exit the mesh loop and enter the next time step of the CFD main program iteration. Since the CF was set to 0 in the previous step, the soot layer mesh loop will definitely be entered when entering the collapse model for the first time.

[0042] After entering the grid loop program, as shown in the attached document. Figure 1 As shown, the central grid cell central This refers to the current cyclic grid, Cell. neighbor To determine the relative positional relationship with adjacent grids, iterate through each grid face of the current grid. If a grid face is an interface with other computational domains, it indicates that the Cell... neighbor If adjacent grids are located in other computational domains, such as pure fluid channels or filter wall computational domains, there is no possibility of carbon soot mass exchange. If the grid surface is not an interface, proceed to the next step of grid position relationship determination.

[0043] Cell central The center coordinates are represented as (x central ,y central Cell central The coordinates of the center of the surface are represented as (x face ,y face If , then the unit vector connecting the two points is represented as:

[0044]

[0045] The purpose of normalizing a vector to a unit vector is that the mesh size of the computational domain is very small. Directly calculating the dot product between the unnormalized vector and the (0, -1) vector will result in an uncertain and very small value, which cannot be used to determine the positional relationship of the mesh.

[0046] The dot product of the unit vector connecting the center of the grid and the center of the grid surface with the vertically downward unit vector (0, -1) is calculated. The positional relationship between the central grid and adjacent grids is determined based on the dot product result. If the dot product result is greater than 0.5, the adjacent grid is considered to be below the central grid. If the dot product result is less than -0.5, the adjacent grid is considered to be above the central grid. The purpose of setting the critical values ​​of 0.5 and -0.5 is that the non-strict regularity of the computational domain grid division and numerical errors may cause the dot product result of parallel unit vectors to be non-strictly 1 or -1, and the dot product result of orthogonal unit vectors to be non-strictly 0. Setting a tolerance value of 0.5 is sufficient to eliminate the above influences and accurately determine the relative position of the grids. This method assumes that grids with left-right relationships will not collapse, while grids with up-down relationships may collapse, i.e., carbon soot mass exchange.

[0047] After determining that the central grid and adjacent grids are in an upper-lower relationship, the next step is to determine whether the collapse conditions are met. As shown in Figures 2(a) and 2(b), these represent two scenarios where collapse occurs. Scenario 1: The porosity ε of the upper grid... neighbor Greater than the critical value ε limit1 This indicates that the carbon soot structure is too loose to support its own structure, and the porosity ε of the central grid is also high. central Greater than the critical value ε limit2 This indicates that the central grid has enough space to accommodate the collapsed soot mass. Scenario 2: The porosity ε of the upper grid neighbor Less than the critical value ε limit3 This indicates that the upper grid has a sufficiently large tendency to collapse, while the porosity ε of the central grid is high. central Greater than the critical value ε limit4 This indicates that the strength of the central grid carbon soot structure is insufficient to support the collapse trend. The conditions for the collapse to occur are as follows:

[0048] senario1:(ε neighbor ≥ε limit1 )&&(ε neighbor<1)&&(ε central ≥ε limit2 )

[0049] senario2:(ε neighbor ≤ε limit3 )&&(ε central ≥ε limit4 )

[0050] Scenario 1 and Scenario 2 are ORed; if either condition is true, a collapse occurs. Scenario 1 requires ε. neighbor The condition <1 is valid because when the porosity is 1, i.e., the soot mass is 0, collapse does not occur. Without this constraint, the program would be unable to exit and would enter an infinite loop. After collapse occurs, for scenario 1, the soot mass m of the central mesh is updated. central_update =m central_old +m neighbor_old Update the soot quality m of adjacent grid cells. neighbor_update =0, for scenario 2, update the central soot mass m central_update =0, update the soot mass m of adjacent grid cells. neighbor_update =m neighbor_old +m central_old After a collapse occurs, the collapse flag CF is set to 1. The critical porosity ε in the above collapse judgment criteria... limit1 ε limit2 ε limit3 ε limit4 It needs to be determined by combining carbon soot microstructure analysis and experimental calibration, as shown in the attached document. Figure 3 The results show the CFD simulation of GPF regeneration when the critical porosity is 0.95, 0.5, 0.7 and 0.92, respectively. The simulation represents the change of soot concentration at each monitoring point along the filter axis during regeneration. Soot collapse occurs sequentially from back to front along the axis. The last part receives the collapse mass above and does not collapse downwards into the grid, which basically matches the regeneration characteristics under the simulation conditions.

[0051] After completing one full mesh cycle for the soot layer computational domain, check the collapse flag CF value. If it is 1, it indicates that at least one collapse has occurred. At this time, reset CF to 0 and continue meshing to look for possible collapses. If the CF value is 0, it indicates that no collapse has occurred and the soot layer as a whole has reached a "dense" state. Set CF to 1, the program returns to the mesh cycle entry, and checks the CF value again. If the loop exit condition is met, the collapse model program ends and enters the next time step iteration solution until all preset time step iterations are completed, and the regeneration process simulation ends.

[0052] This invention addresses the technical problem that traditional CFD simulation methods cannot simulate the thickness reduction caused by carbon soot layer collapse during GPF regeneration, thus failing to accurately simulate the temperature and carbon soot concentration distribution. It provides a carbon soot layer collapse process simulation method based on C language and coupled to the basic CFD solver, which accurately simulates the carbon soot layer thickness change caused by the collapse process and improves prediction accuracy.

[0053] Embodiments of the present invention also provide a computer storage medium storing a computer program thereon, which, when executed by a processor, implements the CFD simulation method described above.

[0054] Embodiments of the present invention also provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the CFD simulation method.

[0055] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the embodiments described above. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A CFD simulation method based on the collapse process of GPF-regenerated soot layer, characterized in that, The CFD simulation method also includes a step to determine the collapse process of the regenerated soot layer after the end of the time step iteration step. The steps for determining the collapse process of the regenerated soot layer include: Determine the collapse flag CF value. If it is 0, enter the carbon soot layer mesh loop. If it is 1, exit the mesh loop and enter the start time step iteration step in the CFD simulation method. Enter the carbon soot layer grid cycle, set the center grid Cell. neighbor That is, the current cyclic network, the grid cell. neighbor To determine the relative positional relationship between the central grid and adjacent grids, the following steps are taken: If the grid surface is the interface with other computational domains, there is no carbon soot mass exchange; otherwise, the relative positional relationship between the central grid and adjacent grids is determined. The relative positional relationship includes the vertical relationship and the horizontal relationship. Only the vertical relationship may have collapse. If the central grid and its adjacent grids are in a vertical relationship, then the process continues to determine whether the collapse condition is met. The specific collapse condition is as follows: Scenario 1: Porosity of the upper mesh Greater than the critical value This indicates that the carbon soot structure is too loose to support its own structure, and the porosity of the central grid is also high. Greater than the critical value This indicates that the central grid has enough space to accommodate the mass of collapsed soot; Scenario 2: Porosity of the upper mesh Less than the critical value This indicates that the upper grid has a sufficiently large tendency to collapse, while the porosity of the central grid is... Greater than the critical value This indicates that the strength of the central grid carbon soot structure is insufficient to support the collapse trend; Scenario 1 and Scenario 2 are ORed; If the collapse condition is met, adding this constraint to Scenario 1 will cause the program to be unable to exit, resulting in an infinite loop. After the collapse occurs, for Scenario 1, the carbon soot quality of the central mesh will be updated. Update the carbon soot quality of adjacent grids. For scenario 2, update the carbon soot quality at the center. Update the carbon soot quality of adjacent grids. After a collapse occurs, the collapse flag CF is set to 1.

2. The CFD simulation method according to claim 1, characterized in that, , , , The values ​​are 0.95, 0.5, 0.7, and 0.92 respectively.

3. The CFD simulation method according to claim 1 or 2, characterized in that, The determination of the relative positional relationship between the central grid and adjacent grids is specifically as follows: Calculate the unit vectors of the center point of the central grid and the center point of the surface; Calculate the dot product of the unit vector connecting the center of the grid and the center of the grid surface with the vertical downward unit vector (0, -1). Determine the positional relationship between the central grid and adjacent grids based on the dot product result. If the dot product result is greater than the first threshold, the adjacent grid is considered to be below the central grid. If the dot product result is less than the second threshold, the adjacent grid is considered to be above the central grid.

4. The CFD simulation method according to claim 3, characterized in that, The first threshold and the second threshold are 0.5 and -0.5, respectively.

5. The CFD simulation method according to claim 1, characterized in that, The above method is coupled into the CFD solver using the C language.

6. The CFD simulation method according to claim 1, characterized in that, The porosity of the upper grid must be less than 1, otherwise it will fall into an infinite loop.

7. The CFD simulation method according to claim 1, characterized in that, During the grid cycle, if at least one collapse is detected, the collapse flag is set to 1. After the grid cycle is completed, if the flag is set to 1, it is considered that there is still a possibility of collapse, and the grid cycle continues until the flag is not set and the value is 0. At this point, the soot layer is considered to have reached a "dense" state, the current grid cycle ends, and the next time step iteration begins.

8. A computer storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the CFD simulation method according to any one of claims 1 to 7.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the CFD simulation method according to any one of claims 1 to 7.

Images