next up previous print clean
Next: Preprocessing Up: The Proposed Approach Previous: Exploding Reflector Modeling

Geometry Optimization

The most important part of my approach is using the information computed from the rays to optimize the layout of the sources and receivers. There are several ways in which this can be accomplished according to how much departure from the standard geometry we are prepared to accept.
1.
Choose an orthogonal, parallel, or slanted geometry with parallel receiver lines as well as parallel source lines. In this case the optimization consists of choosing receiver and shot sampling interval, receiver and source line separations, number of channels per active receiver lines and number of active receiver lines per shot. These parameters can be changed spatially to conform with the requirements of the image, but the basic geometry is unchanged. In other words, the optimum geometry is the composite of regular geometry patches each of which is optimum for a given part of the survey area.
2.
Choose a dense receiver patch with similar receiver sampling in the inline and the cross-line direction and optimize the location of the shots which will not form continuous source lines. The source positions will be computed to optimize uniformity of the subsurface illumination. Optimizing the source positions is particularly attractive in environments in which the layout of the receivers is easy but the drilling of shot holes is expensive. The use of vibrators may also be optimized, since the road paths can be used to constrain the source positions at the design stage.
3.
Allow shots and receivers to be placed in geometries other than continuous source and receiver lines. The constraints of offset and azimuth distribution as well as inline and cross-line offset sampling migration requirements can then be explicitly posed as optimization constraints.

For the sake of this example, I used the first approach, which is the least ambitious.


 
next up previous print clean
Next: Preprocessing Up: The Proposed Approach Previous: Exploding Reflector Modeling
Stanford Exploration Project
7/8/2003