Graphical analysis of results

In presenting the results it would be tempting to try to correlate penetration depth with porosity, but as discussed previously, porosity is difficult to define in a meaningful way. The meaningful variables are those used to drive the heuristic model geometry definition tool, namely the maximum pore diameter, the spacing factor and the exclusion length, which is the product of the first two.
The analysis results quantity, PEEQ penetration depth, is tabulated in the right hand columns of Table 3. The values shown from left to right correspond with the FEA analysis result images shown in Figures 7 and 8. The penetration depth was measured from the FEA results by identifying the left-most element having a PEEQ value greater than 1 × 10-7, in other words, the left-most coloured element. The left-most and right-most nodes of that element were examined, and the PEEQ penetration depth set equal to the average, and the ± error being half the difference. This is illustrated in Figure 9.
The results shown in Figure 10 compare the effect of maximum pore diameter on the PEEQ penetration depth, for constant spacing factor, equal to 3. The data plotted corresponds to the first four rows of Table 3, with three data points per maximum pore diameter result. The error bar for each individual result was determined using the above method, and for that reason, each result has a somewhat different error bar size. It should be noted that this error is attributable to the particular geometry mesh only: since the geometries are randomly created, there should be no expectation that the results for different geometries created using the same parameters should be equal to within a tolerance defined by these error bars. The dashed line shows the linear trend of the combined results.