diff --git a/README.md b/README.md index e38aa37..3e7d02e 100644 --- a/README.md +++ b/README.md @@ -132,36 +132,48 @@ The PDF report associated with the mass flow rate example can be found [here](ex #### Standard Definitions The report includes the standard definitions pictured here. -![](readme_media/mass_flow_rate_definitions.png) + + #### System Setup There is then a page dedicated for the system that the report was generated for. It begins with the System Governing equation. -![](readme_media/mass_flow_rate_governing_equation.png) + + There is then a list of the errors used for each sensor. -![](readme_media/mass_flow_rate_errors.png) + + #### System Results Getting into the Monte Carlo results, the overall system has various numerical results that are shown in text. -![](readme_media/mass_flow_rate_system_results.png) + + There are then two sets of figures. The first figures are histograms, with and without outliers. Immediately next are two scatter plots, again, one with outliers and the other without. -![](readme_media/mass_flow_rate_system_histogram.png) -![](readme_media/mass_flow_rate_system_scatter.png) + + + + Here we can see the first real results, where the sensor system over estimates the true output due to the various thermal affects that do not average out. #### Sensitivity Analysis If there is no sensitivity analysis or range analysis desired, this is the end of the report. For each sensor with sensitivity analysis, there will be a table of numerical values followed by two plots looking at how the "error gain" applied to the specific sensor under analysis affects the second standard deviation error of the system. This first table and set of plots is with the sensor "isolated", meaning every other sensor in the system has zero error. -![](readme_media/mass_flow_rate_sensitivity_iso_table.png) -![](readme_media/mass_flow_rate_sensitivity_iso_plots.png) + + + + The simulation does its best to rpedict the relationship between entire system error and the "error gain" on the specific sensor. Continuing with sensitivity analysis, the next table and sets of plots are "non-isolated", meaning the rest of the sensors will have their standard error applied while the sensor under analysis has an "error gain" applied to its base error. -![](readme_media/mass_flow_rate_sensitivity_noniso_table.png) -![](readme_media/mass_flow_rate_sensitivity_noniso_plots.png) + + + + Comparing between isolated and non-isolated sensitivity analysis shows that the flow sensor has minimal impact to the entire system error due to the relatively higher error threshold of the other sensors. #### Range Analysis Similar to sensitivity analysis, range analysis is optional. For each sensor that undergoes range analysis, there is only one plot. This plot sweeps over the possible values that the sensor could have read, giving the user an idea of the error as a function of that sensor's reading. This is primarily impactful for the DensityLookup function. -![](readme_media/mass_flow_rate_range_plot.png) + + + Through range analysis, it is possible to see how small differences in pressure measurement could result in large variations in reported vlue. In this case, the DensityLookup has the fluid changing from a gas to a liquid, where the two-phase mixture has large variations in density and therefore flow rate. \ No newline at end of file