초록 ▼

A method and system may be provided to perform a process for controlling a target system. In one aspect of the invention, the process may include determining a design space (D8) including a first set of control data values associated with a set of control variables (X). A model (800) may be generate

A method and system may be provided to perform a process for controlling a target system. In one aspect of the invention, the process may include determining a design space (D8) including a first set of control data values associated with a set of control variables (X). A model (800) may be generated that reflects a relationship between the first set of control data values and a first set of response data values associated with a set of response variables (Y). Further, the process may test the model (800) to determine whether a set of predicted response data values associated with the set of response variables (Y) meets a predetermined criteria based on a set of actual response data values associated with the set of response variables (Y). The design space (D8) may then be modified to obtain a relationship between a second set of control data values associated with the set of control variables and a second set of response data values associated with the set of response variables, and the second set of control data values may then be applied to the target system.

대표청구항 ▼

What is claimed is: 1. A method for controlling a target system, comprising: determining a design space including a first set of control data values associated with a set of control variables; generating a model reflecting a relationship between the first set of control data values and a first set

What is claimed is: 1. A method for controlling a target system, comprising: determining a design space including a first set of control data values associated with a set of control variables; generating a model reflecting a relationship between the first set of control data values and a first set of response data values associated with a set of response variables; testing the model to determine whether a set of predicted response data values associated with the set of response variables meet a predetermined criteria based on a corresponding set of actual response data values associated with the set of response variables; modifying, when the predetermined criteria is not met, the design space to obtain a relationship between a second set of control data values associated with the set of control variables and a second set of response data values associated with the set of response variables; and applying the second set of control data values to the target system. 2. The method of claim 1, wherein testing the model includes: applying the first set of control data values to the model to obtain the set of predicted response data values; and applying the first set of control data values to the target system to obtain the set of actual response data values. 3. The method of claim 1, wherein generating a model includes: applying the first set of control data values to the target system to obtain the first set of response data values. 4. The method of claim 1, wherein the model includes a mathematical function and modifying the design space includes: designating an initial data point and a target data point on the function based on minimum and maximum values associated with the set of control variables, respectively; mapping, to the function, a midpoint between the initial and target data points; determining whether the mapped midpoint is converging toward a desired value associated with the at least one response variable; stepping toward the target data point when the mapped midpoint is determined to converge toward a desired value associated with the at least one response variable; mapping a new midpoint to the function based on the stepping and repeating the determining and stepping steps until the mapped new midpoint diverges from the desired value; stepping back from the midpoint to determine a stepped data point between the midpoint and the initial data point when the mapped new midpoint is determined to diverge from the desired value; mapping the stepped data point to the function; modifying the design space such that the mapped stepped data point is designated as the initial data point and the mapped new midpoint is designated as the target data point; and generating a new model based on the modified design space. 5. The method of claim 4, wherein generating a new model includes: generating a new model based on the modified design space to obtain the relationship between the second set of control data values and the second set of response data values. 6. The method of claim 4, wherein generating a new model includes: repeating the generating, testing, and modifying the design space until the set of predicted response data values meet the predetermined criteria. 7. The method of claim 1, wherein the target system is associated with an engine and each control variable is an independent variable that may control operations of the engine. 8. The method of claim 7, wherein each response variable is a dependent variable associated with a response characteristic of the engine and may change values based on the set of control variables. 9. The method of claim 1, wherein the method for controlling target system is performed for a plurality of modes of operations, wherein each mode of operation is associated with at least one parameter that remains constant. 10. The method of claim 1, wherein generating a model includes: determining a first mathematical function that accurately represents a first relationship between a group of control data values corresponding to the set of control variables and a group of response data values corresponding to the set of response variables; determining a second mathematical function that accurately represents a second relationship between a second group of control data values corresponding to the set of control variables and a second group of response data values corresponding to the set of response variables; and generating the model based on the first and second relationships. 11. The method of claim 1, wherein the predetermined criteria is a threshold value associated with a relationship between the set of predicted response data values and the set of actual response data values. 12. The method of claim 11, wherein the set of predicted response data values is determined to meet the predetermined criteria when at least one predicted response data value is within the threshold value of a corresponding at least one actual response data value. 13. The method of claim 1, wherein modifying the design space includes: modifying the data values of at least one of the control data values associated with the set of control variables. 14. A computer implemented method configured to control a target system, comprising: generating a model reflecting a relationship between at least one control variable and at least one response variable, wherein the relationship is represented as a mathematical function included in a design space bounded by a minimum and maximum value associated with the at least one control variable; designating an initial data point and a target data point on the function based on the minimum and maximum values, respectively; mapping, to the function, a midpoint between the initial and target data points; determining whether the mapped midpoint is converging toward a desired value associated with the at least one response variable; designating the mapped midpoint value as the initial data point on the function; repeating the mapping and determining until the mapped midpoint diverges from the desired value; modifying the design space such that a last midpoint that converged toward the desired value is designated as the initial data point; and generating a new model based on the modified design space. 15. A method for controlling a target system, comprising: creating a parameter map including a relationship between a set of parameters and a set of control variables associated with the target system; collecting parameter data for each parameter in the parameter set reflecting a dynamic operation of the target system, wherein the parameter data varies as a function of time; separating the collected parameter data into temporal based segments; determining an average value of each parameter included in each segment; populating the parameter map with the average values of each parameter; segmenting the parameter map into bins; and for each bin, generating a model reflecting a relationship between the set of control variables and a set of response variables associated with an operation of the target system based on the average value for each parameter corresponding to the bin; determining a set of predicted response data values associated with the set of response variables based on the model, adjusting the model until the predicted response data values meet a predetermined criteria, and determining a set of control data values corresponding to the set of control variables that are associated with the predicted response variables that meet the predetermined criteria. 16. The method of claim 15, wherein generating a model includes: applying the set of control variables to the target system to obtain the set of response variables; and determining one or more mathematical functions that accurately represent the relationship between the set of control variables and the set of response variables. 17. The method of claim 16, wherein determining one or more mathematical functions includes: determining a first mathematical function that accurately represents a first relationship between a first set of data values corresponding to the set of control variables and a first set of data values corresponding to the set of response variables; determining a second mathematical function that accurately represents a second relationship between a second set of data values corresponding to the set of control variables and a second set of data values corresponding to the set of response variables; and generating the model based on the first and second relationships. 18. A method for controlling a target system, comprising: determining a base of information including a first set of control data values associated with a set of control variables; generating a model reflecting a relationship between the first set of control data values and a first set of response data values associated with a set of response variables; testing the model to determine whether a set of predicted response data values associated with the set of control variables meet a predetermined criteria based on a set of actual response data values associated with the set of response variables; adding the set of predicted response data values and the set of actual response data values to the base of information when the set of predicted response data values does not meet the predetermined criteria; repeating the generating, testing, and adding until the set of predicted response data values meets the predetermined criteria; determining a second set of control data values associated with the set of control variables that, when applied to the target system, will produce a second set of response data values associated with the set of response variables based on the repeating; and applying the second set of control data values to the target system. 19. A system configured to control the performance of a target system, comprising: a data collection module for determining a design space including a first set of control data values corresponding to a set of control variables; a relationship mapping module for generating a model reflecting a relationship between the first set of control data values and a first set of response data values corresponding to a set of response variables; a testing module for testing the model to determine whether a set of predicted response data values associated with the set of response variables meets a predetermined criteria based on a set of actual response data values associated with the set of response variables; and a training module for modifying the design space to obtain a relationship between a second set of control data values corresponding to the set of control variables and a second set of response data values corresponding to the set of response variables, wherein the system applies the second set of control data values corresponding to the set of control variables to the target system. 20. The system of claim 19, wherein the testing module is further configured to apply the first set of control data values to the model to obtain the set of predicted response data values and apply the first set of control data values to the target system to obtain the set of actual response data values. 21. The system of claim 19, wherein the relationship mapping module is further configured to apply the first set of control data values to the target system to obtain the first set of response data values. 22. The system of claim 19, wherein the model includes a mathematical function and the training module is further configured to: designate an initial data point and a target data point on the function based on minimum and maximum values associated with the set of control variables, respectively; map, to the function, a midpoint between the initial and target data points; determine whether the mapped midpoint is converging toward a desired value associated with the at least one response variable; step toward the target data point when the mapped midpoint is determined to converge toward a desired value associated with the at least one response variable; map a new midpoint to the function based on the stepping and repeating the determining and stepping steps until the mapped new midpoint diverges from the desired value; step back from the midpoint to determine a stepped data point between the midpoint and the initial data point when the mapped new midpoint is determined to diverge from the desired value; map the stepped data point to the function; and modify the design space such that the mapped stepped data point is designated as the initial data point and the mapped new midpoint is designated as the target data point, wherein the system generates a new model based on the modified design space. 23. The system of claim 22, wherein the system generates a new model by the relationship mapping module repeating its generating functions, the testing module repeating its testing functions, and the training module repeating its modifying functions until the at least one predicted response variable meets the predetermined criteria. 24. The system of claim 19, wherein the target system is associated with an engine and each control variable is an independent variable that may control operations of the engine. 25. The system of claim 24, wherein each response variable is a dependent variable associated with a response characteristic of the engine and may change values based on the set of control variables. 26. The system of claim 19, wherein the system is an engine control unit and the second set of control data values are stored in a memory device associated with the engine control unit. 27. The system of claim 19, wherein the system is configured to control the performance of the target system operating in a plurality of modes of operations, wherein each mode of operation is associated with at least one target system parameter that remains constant during operation. 28. The system of claim 19, wherein the relationship mapping module is further configured to: determine a first mathematical function that accurately represents a first relationship between a group of control data values corresponding to the set of control variables and a group of response data values corresponding to the set of response variables; determine a second mathematical function that accurately represents a second relationship between a second group of control data values corresponding to the set of control variables and a second group of response data values corresponding to the set of response variables; and generate the model based on the first and second relationships. 29. The system of claim 19, wherein the predetermined criteria is a threshold value associated with a relationship between the set of predicted response data values and the set of actual response data values. 30. The system of claim 29, wherein the set of predicted response data values is determined to meet the predetermined criteria when at least on predicted response data value is within the threshold value of a corresponding at least one actual response data value. 31. A computer-readable medium including instructions for performing a method, when executed by a processor, for controlling the performance of a target system, the method comprising: determining a design space including a first set of control data values associated with a set of control variables; generating a model reflecting a relationship between the first set of control data values and a first set of response data values associated with a set of response variables; testing the model to determine whether a set of predicted response data values associated with set of response variables meets a predetermined criteria based on a set of actual response data values associated with the set of response variables; modifying, when the predetermined criteria is not met, the design space to obtain a relationship between a second set of control data values associated with the set of control variables and a second set of response data values associated with the set of response variables; and applying the second set of control data values to the target system. 32. The computer-readable medium of claim 31, wherein testing the model includes: applying the first set of control data values to the model to obtain the set of predicted response data values; and applying the first set of control data values to the target system to obtain the set of actual response data values. 33. The computer-readable medium of claim 31, wherein generating a model includes: applying the first set of control data values to the target system to obtain the first set of response data values. 34. The computer-readable medium of claim 31, wherein the model includes a mathematical function and modifying the design space includes: designating an initial data point and a target data point on the function based on minimum and maximum values associated with the set of control variables, respectively; mapping, to the function, a midpoint between the initial and target data points; determining whether the mapped midpoint is converging toward a desired value associated with the at least one response variable; stepping toward the target data point when the mapped midpoint is determined to converge toward a desired value associated with the at least one response variable; mapping a new midpoint to the function based on the stepping and repeating the determining and stepping steps until the mapped new midpoint diverges from the desired value; stepping back from the midpoint to determine a stepped data point between the midpoint and the initial data point when the mapped new midpoint is determined to diverge from the desired value; mapping the stepped data point to the function; modifying the design space such that the mapped stepped data point is designated as the initial data point and the mapped new midpoint is designated as the target data point; and generating a new model based on the modified design space. 35. The computer-readable medium of claim 34, wherein generating a new model includes: generating a new model based on the modified design space to obtain the relationship between the second set of data values and the second set of response data values. 36. The computer-readable medium of claim 34, wherein generating a new model includes: repeating the generating, testing, and modifying the design space until the set of predicted response data values meets the predetermined criteria. 37. The computer-readable medium of claim 31, wherein the target system is associated with an engine and each control variable is an independent variable that may control operations of the engine. 38. The computer-readable medium of claim 37, wherein each response variable is a dependent variable associated with a response characteristic of the engine and may change values based on the set of control variables. 39. The computer-readable medium of claim 31, wherein the method for controlling the performance of the target system is performed for a plurality of modes of operations, wherein each mode of operation is associated with at least one parameter that remains constant. 40. The computer-readable medium of claim 31, wherein the generating a model includes: determining a first mathematical function that accurately represents a first relationship between a group of control data values corresponding to the set of control variables and a group of response data values corresponding to the set of response variables; determining a second mathematical function that accurately represents a second relationship between a second group of control data values corresponding to the set of control variables and a second group of response data values corresponding to the set of response variables; and generating the model based on the first and second relationships. 41. The computer-readable medium of claim 31, wherein the predetermined criteria is a threshold value associated with a relationship between the set of predicted response data values and the set of actual response data values. 42. The computer-readable medium of claim 41, wherein the set of predicted response data values is determined to meet the predetermined criteria when it includes at least one predicted response data value within a threshold value of a corresponding at least one actual response data value. 43. A computer implemented method for controlling the performance of a target system, comprising: generating a model reflecting a relationship between at least one control variable and at least one response variable, wherein the relationship is represented as a mathematical function included in a design space bounded by a minimum and maximum value associated with the at least one control variable; designating an initial data point and a target data point on the function based on minimum and maximum values, respectively; mapping, to the function, a midpoint between the initial and target data points; determining whether the mapped midpoint is converging toward a desired value associated with the at least one response variable; stepping toward the target data point when the mapped midpoint is determined to converge toward a desired value associated with the at least one response variable; mapping a new midpoint to the function based on the stepping and repeating the determining and stepping steps until the mapped new midpoint diverges from the desired value; stepping back from the midpoint to determine a stepped data point between the midpoint and the initial data point when the mapped new midpoint is determined to diverge from the desired value; mapping the stepped data point to the function; modifying the design space such that the mapped stepped data point is designated as the initial data point and the mapped new midpoint is designated as the target data point; and generating a new model based on the modified design space. 44. A computer-readable medium including instructions for performing a method, when executed by a processor, for controlling the performance of a target system, the method comprising: creating a parameter map including a relationship between a set of parameters and a set of control variables associated with the target system; collecting parameter data for each parameter in the parameter set reflecting a dynamic operation of the target system, wherein the parameter data varies as a function of time; separating the collected parameter data into temporal based segments; determining an average value of each parameter included in each segment; populating the parameter map with the average values of each parameter; segmenting the parameter map into bins; and for each bin, generating a model reflecting a relationship between the set of control variables and a set of response variables associated with an operation of the target system based on the average value for each parameter corresponding to the bin; determining a set of predicted response data values associated with the set of response variables based on the model, adjusting the model until the predicted response data values meet a predetermined criteria, and determining a group of data values corresponding to the set of control variables that are associated with the predicted response data values that meet the predetermined criteria. 45. The computer-readable medium of claim 44, wherein generating a model includes: applying the set of control variables to the target system to obtain the set of response variables; and determining one or more mathematical functions that accurately represent the relationship between the set of control variables and the set of response variables. 46. The computer-readable medium of claim 45, wherein determining one or more mathematical functions includes: determining a first mathematical function that accurately represents a first relationship between a set of control data values corresponding to the set of control variables and a set of response data values corresponding to the set of response variables; determining a second mathematical function that accurately represents a second relationship between a second set of control data values corresponding to the set of control variables and a second set of response data values corresponding to the set of response variables; and generating the model based on the first and second relationships. 47. A computer-readable medium including instructions for performing a method, when executed by a processor, for controlling the performance of a target system, the method comprising: determining a design space including a set of control variables each having a range of values; generating a model reflecting a relationship between the set of control variables and a set of response variables; testing the model to determine whether a set of predicted response data values associated with the set of response variables meets a predetermined criteria based on a set of actual response data values associated with the set of response variables; adjusting the design space based on the set of predicted response data values when the set of predicted response data values does not meet the predetermined criteria; repeating the generating, testing, and adjusting until the set of predicted response data values meet the predetermined criteria; determining a set of control data values corresponding to the set of control variables that, when applied to the target system, will produce a set of response data values corresponding to the set of response variables based on the repeating; and applying the set of control data values corresponding to the set of control variables to the target system. 48. A method for controlling a target system, comprising: determining a design space including a first set of control data values associated with a set of control variables; generating a model reflecting a relationship between the control variables and a set of response variables; testing the model to determine whether a set of predicted response data value associated with the set of response variables meet a predetermined criteria based on a corresponding set of actual response data values associated with the set of response variables; modifying the design space to obtain a second relationship between said control variables and said response variables; and applying a second set of control data values associated with said control variables to the target system based on said second relationship.