초록 ▼

An adaptive mutation operator for a genetic algorithm that programmatically mutates individuals in a constrained optimization for a modeled system is discussed. The mutation operator takes into account linear and bound constraints in generating new mutated individuals. The mutation operator generate

An adaptive mutation operator for a genetic algorithm that programmatically mutates individuals in a constrained optimization for a modeled system is discussed. The mutation operator takes into account linear and bound constraints in generating new mutated individuals. The mutation operator generates random mutation direction vectors and random initial step sizes. A mutated individual is generated and moved along a randomly chosen mutation direction vector a distance equal to the initial step size. The generated mutated individual is compared to the linear and bound constraints. In the event the generated mutated individual is located in an infeasible region, the illustrative embodiment of the present invention automatically adjusts the step size to a smaller value and generates another mutated individual along the chosen mutation direction vector. The process iterates until the generated individual is within the feasible region. The number of available valid mutation directions increases as the step size decreases.

대표청구항 ▼

I claim: 1. In a computing device, a computer-implemented method of solving an optimization problem for a modeled system using an adaptive mutation operator in a genetic algorithm, comprising: providing, to the computing device, a first population of a plurality of individuals for an optimization p

I claim: 1. In a computing device, a computer-implemented method of solving an optimization problem for a modeled system using an adaptive mutation operator in a genetic algorithm, comprising: providing, to the computing device, a first population of a plurality of individuals for an optimization problem for a modeled system, the individuals being individual solutions to the optimization problem, the optimization problem including linear and bound constraints; selecting an individual for mutation using the computing device; generating programmatically, using the computing device, an initial step size; generating programmatically, using the computing device, a plurality of random mutation direction vectors extending from a plotted location representing the individual selected for mutation, the plurality of random mutation direction vectors based on the linear and bound constraints; generating programmatically, using the computing device, a mutated individual and moving a distance equal to the initial step size from the plotted location representing the individual selected for mutation along a randomly chosen one of the plurality of mutation direction vectors; and assessing programmatically, using the computing device, whether the mutated individual is in a feasible region, the feasible region representing a region encompassing acceptable solutions to the optimization problem considering the linear and bound constraints, the assessing further comprising: storing the mutated individual as part of second population in a computer-readable medium based on an assessment that the mutated individual is in the feasible region, or based on an assessment that the mutated individual is not in the feasible region, iteratively mutating a different mutated individual along the chosen mutation direction vector using a step size programmatically reduced from the initial step size and assessing whether the different mutated individual is in the feasible region until the different mutated individual is assessed to be in the feasible region, the reduced step size based on the linear and bound constraints, and storing the different mutated individual as part of a second population in a computer-readable medium based on an assessment that the different mutated individual is in the feasible region; and using the second population to solve the optimization problem. 2. The method of claim 1, further comprising: generating a plurality of mutation direction vectors after selecting the individual for mutation; and generating a different mutated individual along each of the plurality of mutation direction vectors using the initial step size. 3. The method of claim 2; further comprising: assessing programmatically each mutated individual generated along the plurality of mutation direction vectors to determine whether each mutated individual is in the feasible region, the feasible region representing an acceptable solution to the optimization problem considering the linear and bound constraints; identifying at least one mutated individual that is not in the feasible region on its selected mutation direction vector; and for each of the identified individuals, mutating another mutated individual along the selected mutation direction vector using a second step size programmatically reduced from the initial step size. 4. The method of claim 1, further comprising: based on a programmatic assessment that the individual mutated using the programmatically reduced step size is not in the feasible region, mutating another individual along the chosen mutation direction vector using a different step size programmatically reduced from the step size used in the previous mutation. 5. The method of claim 1, further comprising: generating programmatically, in response to the generation of a reduced step size, at least one new mutation direction vector and at least one new mutated individual for the new mutation direction. 6. The method of claim 5 wherein the number of mutation directions is increased for each reduction in step size. 7. The method of claim 1 wherein the mutated individual is located in the feasible region and becomes part of a new population. 8. The method of claim 7 wherein the new population also includes individuals created by crossover. 9. The method of claim 7 wherein the new population also includes individuals selected directly from the first population. 10. The method of claim 7 wherein the new population includes pools of individuals created separately in a parallel computing environment. 11. The method of claim 1 wherein the linear and bound constraints are incorporated into the calculation of the generation of the mutation direction vector. 12. The method of claim 1, further comprising: providing a plurality of computing devices communicating over a network with the computing device, the plurality of computing devices each providing a pool of individuals for the optimization problem, the individuals being individual solutions to the optimization problem, the pools collectively representing the first population. 13. The method of claim 12 wherein the plurality of computing devices perform crossover between individuals in the pools before providing the first population to the computing device. 14. One or more computer-readable media holding computer-executable instructions that when executed on a computing device solve an optimization problem for a modeled system using an adaptive mutation operator in a genetic algorithm, the media holding one or more instructions for: providing a first population of a plurality of individuals for an optimization problem for a modeled system using the computing device, the individuals being individual solutions to the optimization problem, the optimization problem including linear and bound constraints; selecting an individual for mutation, using the computing device; generating programmatically, using the computing device; an initial step size; generating programmatically, using the computing device, a plurality of random mutation direction vectors extending from a plotted location representing the individual selected for mutation, the plurality of random mutation direction vectors based on the linear and bound constraints; generating programmatically, using the computing device, a mutated individual and moving a distance equal to the initial step size from the plotted location representing the individual selected for mutation along a randomly chosen one of the plurality of mutation direction vectors; assessing programmatically, using the computing device, whether the mutated individual is in the feasible region, the feasible region representing a region encompassing acceptable solutions to the optimization problem considering the linear and bound constraints, the assessing further comprising: storing the mutated individual as part of second population in a computer-readable medium based on an assessment that the mutated individual is in the feasible region, or based on an assessment that the mutated individual is not in the feasible region, iteratively mutating a different mutated individual along the mutation direction vector using a step size programmatically reduced from the initial step size and assessing whether the different mutated individual is in the feasible region until the different mutated individual is assessed to be in the feasible region, the reduced step size based on the linear and bound constraints, and storing the different mutated individual as part of a second population in a computer-readable medium based on an assessment that the different mutated individual is in the feasible region; and using the second population to solve the optimization problem. 15. The media of claim 14, wherein the media comprises one or more instructions for: generating a plurality of mutation direction vectors after selecting the individual for mutation; and generating a different mutated individual along each of the plurality of mutation direction vectors using the initial step size. 16. The media of claim 15 wherein the media comprises one or more instructions for: assessing programmatically each mutated individual generated along the plurality of mutation direction vectors to determine whether each mutated individual is in the feasible region representing an acceptable solution to the optimization problem considering the linear and bound constraints; identifying at least one mutated individual that is not in feasible region on its selected mutation direction vector; and for each of the identified individuals, mutating another mutated individual along the selected mutation direction vector using a second step size programmatically reduced from the initial step size. 17. The media of claim 14, wherein the media comprises one or more instructions for: mutating another mutated individual along the chosen mutation direction vector using a different step size programmatically reduced from the step size used in the previous mutation based on a programmatic assessment that the mutated individual mutated using the second step size is not in the feasible region. 18. The media of claim 14, wherein the media comprises one or more instructions for: generating programmatically, in response to the generation of a reduced step size, at least one new mutation direction vector and at least one new mutated individual for the new mutation direction. 19. The media of claim 14 wherein the number of mutation directions is increased for each reduction in step size. 20. The media of claim 14 wherein the mutated individual is located in the feasible region and becomes part of a new population. 21. The media of claim 14 wherein the linear and bound constraints are incorporated into the calculation of the generation of the mutation direction vector.