Generate model based on an origin in the fault to improve calculation accuracy (#11267)

Generate model based on an origin in the fault to improve calculation accuracy.
This commit is contained in:
jonjenssen authored and GitHub committed 2024-03-08 15:50:05 +01:00
1 parent bfda6519db
commit 2386ad8985
4 files changed
+84 -39

No files matched your search

@@ -50,7 +50,7 @@ public:
void reset();
void generateGeometry( const std::array<cvf::Vec3d, 12>& inputPoints,
const std::vector<cvf::Vec3d>& reservoirLayers,
const std::vector<double>& reservoirZ,
double maxCellHeight,
double cellSizeFactor,
const std::vector<double>& horizontalPartition,
@@ -59,6 +59,7 @@ public:
double topHeight,
int nFaultZoneCells );
void shiftNodes( const cvf::Vec3d offset );
void generateLocalNodes( const cvf::Mat4d transform );
void setUseLocalCoordinates( bool useLocalCoordinates );
@@ -89,7 +90,6 @@ protected:
static cvf::Vec3d stepVector( cvf::Vec3d start, cvf::Vec3d stop, int nSteps );
static std::vector<double> generateConstantLayers( double zFrom, double zTo, double maxSize );
static std::vector<double> generateGrowingLayers( double zFrom, double zTo, double maxSize, double growfactor );
static std::vector<double> extractZValues( const std::vector<cvf::Vec3d>& points );
void generateVerticalMeshlines( const std::vector<cvf::Vec3d>& cornerPoints, const std::vector<double>& horzPartition );