Apparatus and method for fusing diverse data

• US 5,729,451 A
• Filed: 12/01/1995
• Issued: 03/17/1998
• Est. Priority Date: 12/01/1995
• Status: Expired due to Fees

• Current Assignee: CyTerra Corp. (L3Harris Technologies, Inc.)
• Original Assignee: Coleman Research Corp. (L3Harris Technologies, Inc.)
• Inventors: Porter, David W., Yancey, William E., Gibbs, Bruce P.
• Primary Examiner(s): McElheny, Jr., Donald E.
• Application Number:
  US08/566,353
• Time in Patent Office: 837 Days
• Field of Search: 395/420, 395/421, 395/422, 395/578
• US Class Current: 702/12
• CPC Class Codes: G06F 17/17 Function evaluation by approximation methods, e.g. inter- or extrapolation, smoothing, least mean square method G06F17/18 takes precedence ; interpolation for numerical control G05B19/18

Abstract

A data fusion workstation apparatus and method utilizes algorithms and can be used for applications such as, e.g., hydrogeological modeling, steady-state hydrological flow modeling, transport uncertainty determination, flow/transport fusion, oil reserve management, water supply development, geo-technical engineering projects, and management decision optimization. The invention uses a spatial continuity model which is extended to 3D and equations are provided for computation of the spatial AR coefficients from a desired covariance function. The data fusion workstation provides more robust optimization iterations by use of an algorithm that includes Trust Region or Step Halving methods to limit the Gauss Newton step in solving non-linear least squares problems. Measurement biases and error parameters, and parameters which define a hydrological flow model can be included in the common set of estimated states. An LSQR or other iterative algorithm is used to reduce the computation time for each step of a nonlinear optimization, thereby allowing computation of problems having a very large number of states. The invention preferably provides transport uncertainty calculations in a fusion post-processing step. The invention allows contaminant measurements to be used with hydraulic head and conductivity measurements for better calibration of flow and transport models.

Claims

1. A data fusion workstation for creating a reconstruction of a site from measurement data and computerized models, comprising:

means for storing measurement equations describing at least one physical characteristic of a site;

means for storing pseudomeasurement equations describing said at least one physical characteristic;

means for generating a site model comprising a computer reconstuction of said site;

means for optimizing said site model, comprising;

means for comparing data produced by said site model to data produced by said measurement equations and for generating a first quantification of error based upon differences therebetween;

means for comparing data produced by said site model to data produced by said pseudomeasurement equations and for generating a second quantification of error based upon differences therebetween;

means for calculating a least squares solution which reduces a cost function related to said first and second quantifications of error, said means for calculating comprising means for executing a trust region algorithm which limits Gauss-Newton steps;

means for using said least squares solution to adjust said site model;

means for displaying a graphical representation of said site model.

2. A data fusion workstation for creating a reconstruction of a site from measurement data and computerized models, comprising:

means for storing measurement equations describing at least one physical characteristic of a site;

means for storing pseudomeasurement equations describing said at least one physical characteristic;

means for generating a site model comprising a computer reconstuction of said site;

means for optimizing said site model, comprising;

means for comparing data produced by said site model to data produced by said measurement equations and for generating a first quantification of error based upon differences therebetween;

means for comparing data produced by said site model to data produced by said pseudomeasurement equations and for generating a second quantification of error based upon differences therebetween;

means for calculating a least squares solution which reduces a cost function related to said first and second quantifications of error, said means for calculating comprising means for executing a step-halving algorithm which limits Gauss-Newton steps;

means for using said least squares solution to adjust said site model;

means for displaying a graphical representation of said site model.

3. A data fusion workstation for creating a reconstruction of a site from measurement data and computerized models, comprising:

means for storing measurement equations describing at least one physical characteristic of a site;

means for storing pseudomeasurement equations describing said at least one physical characteristic;

means for generating a site model comprising a computer reconstuction of said site;

means for optimizing said site model, comprising;

means for comparing data produced by said site model to data produced by said measurement equations and for generating a first quantification of error based upon differences therebetween;

means for comparing data produced by said site model to data produced by said pseudomeasurement equations and for generating a second quantification of error based upon differences therebetween;

means for calculating a least squares solution which reduces a cost function related to said first and second quantifications of error, said means for calculating comprising an iterative linear least squares solver for calculating Gauss-Newton steps;

means for using said least squares solution to adjust said site model;

means for displaying a graphical representation of said site model.

4. A data fusion workstation for creating a reconstruction of a site from measurement data and computerized models, comprising:

means for storing a spatial continuity model;

means for receiving user input of a covariance function;

means for computing spatial continuity AR coefficients from said covariance function;

means for using said spatial continuity AR coefficients in a spatial continuity model;

means for using said spatial continuity model to create a site model;

means for displaying a graphical representation of said site model.

5. A data fusion workstation for creating a steady-state hydrological flow model, comprising:

means for storing measurement equations describing at least one physical characteristic of a site;

means for storing pseudomeasurement equations describing said at least one physical characteristic;

means for generating a hydrological flow model comprising a computer reconstuction of said site;

means for optimizing said site model, comprising;

means for comparing data produced by said site model to data produced by said measurement equations and for generating a first quantification of error based upon differences therebetween;

means for comparing data produced by said site model to data produced by said pseudomeasurement equations and for generating a second quantification of error based upon differences therebetween;

means for calculating a least squares solution which reduces a cost function related to said first and second quantifications of error;

means for using said least squares solution to adjust said hydrological flow model;

means for displaying a graphical representation of said hydrological flow model.

6. A data fusion workstation for creating a site model which includes information on groundwater flow, comprising:

means for storing measurement equations describing at least one physical characteristic of a site;

means for storing pseudomeasurement equations describing said at least one physical characteristic;

means for generating a site model comprising a computer reconstuction of said site and a quantification of uncertainty on groundwater flow;

means for optimizing said site model, comprising;

means for comparing data produced by said site model to data produced by said measurement equations and for generating a first quantification of error based upon differences therebetween;

means for comparing data produced by said site model to data produced by said pseudomeasurement equations and for generating a second quantification of error based upon differences therebetween;

means for calculating a least squares solution which reduces a cost function related to said first and second quantifications of error;

means for using said least squares solution to adjust said site model;

means for displaying a graphical representation of said site model.

7. A data fusion workstation for creating a site model which includes information on oil reservoir properties, comprising:

means for storing measurement equations describing at least one physical characteristic of a reservoir site;

means for storing pseudomeasurement equations describing said at least one physical characteristic;

means for generating a site model comprising a computer reconstuction of said site and a quantification of uncertainty associated with said reconstruction;

means for optimizing said site model, comprising;

means for comparing data produced by said site model to data produced by said measurement equations and for generating a first quantification of error based upon differences therebetween;

means for comparing data produced by said site model to data produced by said pseudomeasurement equations and for generating a second quantification of error based upon differences therebetween;

means for calculating a least squares solution which reduces a cost function related to said first and second quantifications of error;

means for using said least squares solution to adjust said reservoir site model;

means for displaying a graphical representation of said reservoir site model.