[807] | 1 | |
---|
[1337] | 2 | Examples for event biasing: B01 and B02 |
---|
| 3 | --------------------------------------- |
---|
[807] | 4 | |
---|
[1337] | 5 | B01 and B02 applications demonstrate the usage of different variance |
---|
| 6 | reduction techniques supported in Geant4, or possible from the user |
---|
| 7 | applications. |
---|
[807] | 8 | |
---|
| 9 | General remark to variance reduction |
---|
| 10 | ------------------------------------ |
---|
| 11 | The tools provided for importance sampling (or geometrical splitting and |
---|
| 12 | Russian roulette) and for the weight window technique require the user to |
---|
| 13 | have a good understanding of the physics in the problem. This is because |
---|
| 14 | the user has to decide which particle types have to be biased, define the |
---|
| 15 | cells (physical volumes, replicas) and assign importances or weight |
---|
| 16 | windows to that cells. If this is not done properly it can not be |
---|
| 17 | expected that the results describe a real experiment. The examples given |
---|
| 18 | here only demonstrate how to use the tools technically. They don't intend |
---|
| 19 | to produce physical correct results. |
---|
| 20 | |
---|
| 21 | General remark to scoring |
---|
| 22 | ------------------------- |
---|
| 23 | A interface G4VScorer is provided for the user. The user may create his |
---|
| 24 | own class to perform the desired scoring. The user defined class |
---|
| 25 | therefore should inherit from the interface G4VScorer. |
---|
| 26 | An example of an implementation of a scorer is G4Scorer |
---|
| 27 | which may be found in source/event. |
---|
| 28 | The scoring in these examples is done with a G4Scorer. |
---|
| 29 | The variance reduction techniques and scoring do not support all options |
---|
| 30 | of the Geant4 geometry. It only supports physical volumes and simple |
---|
| 31 | replicas. |
---|
| 32 | To identify a physical volume (or replica) objects of the class |
---|
| 33 | G4GeometryCell are used. Scoring is done according to these |
---|
| 34 | cells and importance values or the weight windows may be assigned to |
---|
| 35 | them. |
---|
| 36 | When scoring is done in a parallel geometry special action has to be taken |
---|
| 37 | to prevent counting of "collisions" with boundaries of the mass geometry |
---|
| 38 | as interactions. This is differently handled when scoring is done in the |
---|
| 39 | mass geometry. |
---|
| 40 | |
---|
| 41 | --> G4GeometryCell of the parallel geometry must not share boundaries with |
---|
| 42 | the world volume! <-- |
---|
| 43 | |
---|
| 44 | Known problems |
---|
| 45 | -------------- |
---|
| 46 | In the following scenario it can happen that a particle is not |
---|
| 47 | biased and it's weight is therefore not changed even if it crosses |
---|
| 48 | a boundary where biasing should happen. |
---|
| 49 | Importance and weight window sampling create particles on boundaries |
---|
| 50 | between volumes. If the GPIL method of a physical process returns |
---|
| 51 | 0 as step length for a particle on a boundary and if the PostStepDoIt of |
---|
| 52 | that process changes the direction of the particle to go back in the |
---|
| 53 | former volume the biasing won't be invoked. |
---|
| 54 | This will produce particles with weights that do not correspondent to the |
---|
| 55 | importance of the current volumes. |
---|
| 56 | |
---|
| 57 | Further information: |
---|
| 58 | -------------------- |
---|
| 59 | Short description of importance sampling and scoring: |
---|
| 60 | http://cern.ch/geant4/working_groups/geometry/biasing/Sampling.html |
---|
| 61 | |
---|
| 62 | Example B01 |
---|
| 63 | =========== |
---|
| 64 | |
---|
| 65 | The example uses importance sampling or the weight window technique |
---|
| 66 | according to an input parameter. It uses scoring in both cases. |
---|
| 67 | Importance values or weight windows are defined according to the mass |
---|
| 68 | geometry. In this example the weight window technique is configured such |
---|
| 69 | that it behaves equivalent to importance sampling: The window is actually |
---|
| 70 | not a window but simply the inverse of the importance value and only |
---|
| 71 | one energy region is used that covers all energies in the problem. |
---|
| 72 | The user may change the weight window configuration by changing the |
---|
| 73 | initialization of the weight window algorithm in example,cc. |
---|
| 74 | Different energy bounds for the weight window technique may be specified |
---|
| 75 | in B01DetectorConstruction. |
---|
| 76 | |
---|
| 77 | The executable takes one optional argument: 0 or 1. Without argument or |
---|
| 78 | with argument: 0, the importance sampling is applied with argument: 1, |
---|
| 79 | the weight window technique is applied. |
---|
| 80 | |
---|
| 81 | |
---|
| 82 | Example B02 |
---|
| 83 | =========== |
---|
| 84 | |
---|
| 85 | This example uses a parallel geometry to define G4GeometryCell objects |
---|
| 86 | for scoring and importance sampling. In addition it customizes |
---|
| 87 | the scoring. In this example one scorer creates a histogram. |
---|
| 88 | |
---|
| 89 | Compiling and running |
---|
| 90 | --------------------- |
---|
[1337] | 91 | To compile this example you need AIDA installed. To link |
---|
| 92 | and run it you need a AIDA compliant analysis package. |
---|
| 93 | Histograms are saved in HBOOK format. |
---|
| 94 | You need to set the following variable in your environment: |
---|
[807] | 95 | "G4ANALYSIS_USE" |
---|
| 96 | The example stores the plot in the file b02.hbook. |
---|
| 97 | |
---|
[1337] | 98 | ___________________________________________________________________________ |
---|
[807] | 99 | |
---|
| 100 | |
---|
[1337] | 101 | Reverse MonteCarlo Technique example: ReverseMC01 |
---|
| 102 | ------------------------------------------------- |
---|
[807] | 103 | |
---|
[1337] | 104 | Example ReverseMC01 |
---|
| 105 | =================== |
---|
[807] | 106 | |
---|
[1337] | 107 | Example illustrating the use of the Reverse Monte Carlo (RMC) mode in a Geant4 |
---|
| 108 | application. See details in ReverseMC01/README. |
---|
[807] | 109 | |
---|