Optimization of a design
First Claim
Patent Images
1. A computer readable medium containing instructions that, when executed by a processor, cause the processor to perform a method for finding at least one local optimum model, the method comprising:
- a. specifying a base model that includes a variable having a value,b. specifying a goal which identifies a characteristic of the base model that is to be optimized;
c. specifying a maximum number of optimizations to be performed;
d. specifying a minimum value and a maximum value for the variable;
e. specifying a list of different values for the variable between the minimum and maximum values for the variable;
f. creating a model for every value of the variable included in the list;
g. running each model created to determine a value of the characteristic for each model;
h. comparing each model run with its adjacent model or models;
i. identifying each model as a local model if the model has a better value of the characteristic than its adjacent model or models;
j. ranking all local models identified;
k. optimizing each local model to find a local optimum for each local model up to the maximum number of optimizations specified; and
l. identifying and storing the at least one local optimum model.
1 Assignment
0 Petitions
Accused Products
Abstract
An apparatus, system, and method for optimization. The apparatus, system, and method include a mechanism for ending an optimization when all models within one design tolerance of an optimum model have been simulated.
34 Citations
40 Claims
-
1. A computer readable medium containing instructions that, when executed by a processor, cause the processor to perform a method for finding at least one local optimum model, the method comprising:
-
a. specifying a base model that includes a variable having a value, b. specifying a goal which identifies a characteristic of the base model that is to be optimized; c. specifying a maximum number of optimizations to be performed; d. specifying a minimum value and a maximum value for the variable; e. specifying a list of different values for the variable between the minimum and maximum values for the variable; f. creating a model for every value of the variable included in the list; g. running each model created to determine a value of the characteristic for each model; h. comparing each model run with its adjacent model or models; i. identifying each model as a local model if the model has a better value of the characteristic than its adjacent model or models; j. ranking all local models identified; k. optimizing each local model to find a local optimum for each local model up to the maximum number of optimizations specified; and l. identifying and storing the at least one local optimum model.
-
-
2. A computer readable medium containing instructions that, when executed by a processor, cause the processor to perform a method for finding at least one local optimum model, the method comprising:
-
a. specifying a base model that includes two or more variables each having a value, b. specifying a goal which identifies a characteristic of the base model that is to be optimized; c. specifying a maximum number of optimizations to be performed; d. specifying for each variable, a minimum value and a maximum value; e. specifying a list of different values for each variable between their minimum and maximum values; f. creating a model for combinations of the values of the variables included in the list; g. running each model created to determine a value of the characteristic for each model; h. comparing each model run with its adjacent model or models; i. identifying each model as a local model if the model has a better value of the characteristic than its adjacent model or models; j. ranking all local models identified; k. optimizing the highest ranked local models to find the local optimum model up to the maximum number of optimizations specified; and l. storing the local optimum model for each of the highest ranked local models.
-
-
3. A computer readable medium containing instructions that, when executed by a processor, cause the processor to perform an optimization method, the method comprising:
-
a. specifying a characteristic to be optimized and a desired value for the characteristic; b. specifying a variable to be varied during the optimization; c. setting a design tolerance equal to a minimum increment by which the variable is to be varied; d. selecting a design configuration that includes a value for the variable and all other values necessary to simulate the design configuration as a base design configuration; e. simulating the design configuration to arrive at a value for the characteristic; f. selecting design configurations having variable values adjacent to the base design configuration variable value; g. simulating the selected design configurations; h. setting the base design configuration variable value to the design configuration value of an adjacent design configuration having a characteristic value nearest to the desired characteristic value if an adjacent design configuration has a characteristic value nearer to the desired characteristic than the base design configuration; i. selecting design configurations having variable values more nearly adjacent to the base design configuration variable value if no adjacent design configuration has a characteristic value nearer to the desired characteristic value than the base design configuration; j. repeating steps g, h, and i until all design configurations having variable values adjacent to the base design configuration by one design tolerance have been simulated and the characteristic value of the base design configuration is nearer the desired characteristic value than any other design configuration simulated; and k. storing the base design configuration variable value. - View Dependent Claims (4, 5, 6, 7, 8)
-
-
9. A computer readable medium containing instructions that, when executed by a processor, cause the processor to perform a method for selecting a best value for a variable in a model, the method comprising:
-
a. specifying a model that includes an input variable having a tolerance and a step having an initial magnitude greater than the tolerance, and specifying an output characteristic having an objective; b. running the model with the input variable set to a selected value to arrive at a resulting output characteristic; c. running the model with the input variable set to the selected value changed by the step to arrive at a second resulting output characteristic; d. selecting the input variable value that results in an output characteristic that more nearly conforms to the objective; e. reducing the step if the selected input variable value is unchanged at d; f. setting the step equal to the tolerance if the step would otherwise be reduced to less than the tolerance; g. repeating c through f until the step reaches the tolerance; and h. storing the selected input variable value.
-
-
10. A computer readable medium containing instructions that, when executed by a processor, cause the processor to perform a method for selecting a best value for a variable in a model, the method comprising:
-
a. specifying a model that includes an input variable having a tolerance and a step having an initial magnitude greater than the tolerance, and specifying an output characteristic having an objective; b. running the model with the input variable set to a selected value to arrive at a resulting output characteristic; c. running the model with the input variable set to the selected value changed by the step to arrive at a second resulting output characteristic; d. selecting the input variable value that results in an output characteristic that more nearly conforms to the objective; e. selecting the input variable value changed by a step if the output characteristic resulting when the selected input variable is set to the selected value and the output characteristic resulting when the selected input variable is set to the selected value changed by the step are the same; f. reducing the step if the selected input variable value is unchanged at d and e; g. repeating c through f until the step reaches the tolerance; and h. storing the selected input variable value.
-
-
11. A computer readable medium containing instructions that, when executed by a processor, cause the processor to perform a method for selecting a best value for a variable in a model, the method comprising:
-
a. specifying a model that includes an input variable having a tolerance and a step having an initial magnitude greater than the tolerance, and specifying an output characteristic having an objective; b. running the model with the input variable set to a selected value to arrive at a resulting output characteristic; c. running the model with the input variable set to the selected value changed by the step to arrive at a second resulting output characteristic; d. selecting the input variable value that results in an output characteristic that more nearly conforms to the objective, wherein the output characteristic that more nearly conforms to the objective is the output characteristic having the largest value; e. reducing the step if the selected input variable value is unchanged at d; f. repeating c through e until the step reaches the tolerance; and g. storing the selected input variable value.
-
-
12. A computer readable medium containing instructions that, when executed by a processor, cause the processor to perform a method for selecting a best value for a variable in a model, the method comprising:
-
a. specifying a model that includes an input variable having a tolerance and a step that is an integer multiple of the tolerance and has an initial magnitude greater than the tolerance, and specifying an output characteristic having an objective; b. running the model with the input variable set to a selected value to arrive at a resulting output characteristic; c. running the model with the input variable set to the selected value changed by the step to arrive at a second resulting output characteristic; d. selecting the input variable value that results in an output characteristic that more nearly conforms to the objective; e. reducing the step if the selected input variable value is unchanged at d; f. repeating c through e until the step reaches the tolerance; and g. storing the selected input variable value.
-
-
13. A computer readable medium containing instructions that, when executed by a processor, cause the processor to perform a method for selecting a best value for a variable in a model, the method comprising:
-
a. specifying a model that includes a first input variable having a first tolerance and a first step having an initial magnitude greater than the first tolerance, specifying a second input variable having a second tolerance equal to the first tolerance and a second step equal to the first step, the second step having an initial magnitude equal to the first initial magnitude and greater than the second tolerance, and specifying an output characteristic having an objective; b. running the model with the first input variable set to a first selected value and the second input variable set to a second selected value to arrive at a resulting output characteristic; c. running the model with the first input variable set to the first selected value changed by the first step and the second input variable set to the second selected value changed by the second step to arrive at a second resulting output characteristic; d. selecting the first input variable value and the second input variable value that result in an output characteristic that more nearly conforms to the objective; e. reducing the first step and the second step if the first selected input variable value and the second selected input variable value are both unchanged at d; f. repeating c through e until the first step reaches the first tolerance and the second step reaches the second tolerance; and g. storing the selected first input variable value and the selected second input variable value.
-
-
14. A computer readable medium containing instructions that, when executed by a processor, cause the processor to perform a method for selecting a best value for a variable in a model, the method comprising:
-
a. specifying a model that includes an input variable having a tolerance and a step having an initial magnitude greater than the tolerance, wherein the input variable is a dimension of a component of a machine and the model simulates the machine including that component, and specifying an output characteristic having an objective; b. running the model with the input variable set to a selected value to arrive at a resulting output characteristic; c. running the model with the input variable set to the selected value changed by the step to arrive at a second resulting output characteristic; d. selecting the input variable value that results in an output characteristic that more nearly conforms to the objective; e. reducing the step if the selected input variable value is unchanged at d; f. repeating c through e until the step reaches the tolerance; and g. storing the selected input variable value.
-
-
15. A computer readable medium containing instructions that, when executed by a processor, cause the processor to perform a method for finding a variable value that results in an optimum value for an output characteristic by varying the magnitude of the variable value, the method comprising:
-
a. setting the variable to a base value; b. setting a tolerance to a minimum step; c. determining a first output characteristic value with the variable set to the base value; d. determining a second output characteristic value with the variable set to the base value plus a whole number multiple of the tolerance; e. choosing one of the first output characteristic value and the second output characteristic value, whichever is nearest a desired output characteristic value;
dimension of a component of a machine and the model simulates the machine including that component, and specifying an output characteristic having an objective;b. running the model with the input variable set to a selected value to arrive at a resulting output characteristic; c. running the model with the input variable set to the selected value changed by the step to arrive at a second resulting output characteristic; d. selecting the input variable value that results in an output characteristic that more nearly conforms to the objective; e. reducing the step if the selected input variable value is unchanged at d; f. repeating c through e until the step reaches the tolerance; g. storing the selected input variable value. f. setting the base value to the variable value providing the chosen output characteristic value; g. reducing the whole number multiple if the base value is unchanged; h. repeating d through g until the whole number multiple is one; and i. storing the variable value. - View Dependent Claims (16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40)
-
Specification