Multi-objective design optimization using adaptive classification
Definition of a design space and an objective space for conducting multi-objective design optimization of a product is received in a computer system having a design optimization application module installed thereon. Design space is defined by design variables while objective space is defined by design objectives. First set of designs in the design space is selected. Each of the first set is evaluated in the objective space for non-dominance. Design space is partitioned into first and second regions using a multi-dimensional space division scheme (e.g., SVM). The first region is part of the design space containing all of the non-dominated design alternatives while the second region contains remaining of the design space. Second set of designs is selected within the first region. Each of the second set and existing non-dominated design alternatives are evaluated for non-dominance. Multi-objective optimization repeats the partition and evaluation until an end condition is reached.
1. A method of conducting multi-objective design optimization of a product using adaptive classification for selecting design alternatives comprising: (a) receiving a definition of a design space and an objective space for conducting multi-objective design optimization of a product in a computer system having a design optimization application module installed thereon, the design space being defined by a plurality of design variables while the objective space being defined by a plurality of design objectives; (b) initially selecting, by the design optimization application module, a first set of design alternatives within the design space; (c) determining, by the design optimization application module, whether each of the first set is a non-dominated design alternative by evaluating said each of the first set in the objective space, whereby the non-dominated design alternative does not improve all of the design objectives; (d) partitioning, by the design optimization application module, the design space into first and second regions using a multi-dimensional space division scheme, the first region being part of the design space containing all of the non-dominated design alternatives while the second region containing remaining of the design space; (e) selecting, by the design optimization application module, a second set of design alternatives within the first region according to a predetermined selection scheme; (f) determining, by the design optimization application module, whether each of the second sets and the previously-determined non-dominated design alternatives is non-dominated in the objective space; (g) repartitioning, by the design optimization application module, the design space into the first region and the second region using the multi-dimensional space division scheme with respect to the newly-determined non-dominated design alternatives; and (h) repeating (e)-(g) until an end condition of the multi-objective design optimization is reached;
wherein the first region is increased by a buffer region along the first and second region's boundary.
2. The method of claim 1 , wherein said first set of design alternatives are uniformly distributed in the design space.
3. The method of claim 1 , wherein each of said first and said second regions contains one or more contiguous portions of the design space.
4. The method of claim 1 , wherein said predetermined selection scheme comprises placing each of the second set at a location having a largest minimum distance to the existing non-dominated design alternatives.
5. The method of claim 1 , wherein the multi-dimensional space division scheme comprises support vector machine.
6. The method of claim 1 , wherein said predetermined selection scheme comprises placing some of the second set in the buffer region and the remaining of the second set in the first region.
7. A system for conducting multi-objective design optimization of a product using adaptive classification for selecting design alternatives comprising: a main memory for storing computer readable code for a design optimization application module; at least one processor coupled to the main memory, said at least one processor executing the computer readable code in the main memory to cause the design optimization application module to perform operations by a method of: (a) receiving a definition of a design space and an objective space for conducting multi-objective design optimization of a product, the design space being defined by a plurality of design variables while the objective space being defined by a plurality of design objectives; (b) initially selecting a first set of design alternatives in the design space; (c) determining whether each of the first set is a non-dominated design alternative by evaluating said each of the first set in the objective space, whereby the non-dominated design does not improve all of the design objectives; (d) partitioning the design space into first and second regions using a multi-dimensional space division scheme, the first region being part of the design space containing all of the non-dominated design alternatives while the second region containing remaining of the design space; (e) selecting a second set of design alternatives within the first region according to a predetermined selection scheme; (f) determining whether each of the second sets and the previously-determined non-dominated design alternatives is non-dominated in the objective space; (g) repartitioning the design space into the first region and the second region using the multi-dimensional space division scheme with respect to the newly-determined non-dominated design alternatives; and (h) repeating (e)-(g) until an end condition of the multi-objective design optimization is reached;
wherein the first region is increased by a buffer region along the first and second region's boundary.
8. The system of claim 7 , wherein said first set of design alternatives are uniformly distributed in the design space.
9. The system of claim 7 , wherein each of said first and said second regions contains one or more contiguous portions of the design space.
10. The system of claim 7 , wherein said predetermined selection scheme comprises placing each of the second set at a location having a largest minimum distance to the existing non-dominated design alternatives.
11. The system of claim 7 , wherein the multi-dimensional space division scheme comprises support vector machine.
12. The system of claim 7 , wherein said predetermined selection scheme comprises placing some of the second set in the buffer region and the remaining of the second set in the first region.
13. A non-transitory computer readable storage medium containing instructions for conducting multi-objective design optimization of a product using adaptive classification for selecting design alternatives by a method comprising: (a) receiving a definition of a design space and an objective space for conducting multi-objective design optimization of a product in a computer system having a design optimization application module installed thereon, the design space being defined by a plurality of design variables while the objective space being defined by a plurality of design objectives; (b) initially selecting, by the design optimization application module, a first set of design alternatives within the design space; (c) determining, by the design optimization application module, whether each of the first set is a non-dominated design alternative by evaluating said each of the first set in the objective space, whereby the non-dominated design alternative does not improve all of the design objectives; (d) partitioning, by the design optimization application module, the design space into first and second regions using a multi-dimensional space division scheme, the first region being part of the design space containing all of the non-dominated design alternatives while the second region containing remaining of the design space; (e) selecting, by the design optimization application module, a second set of design alternatives within the first region according to a predetermined selection scheme; (f) determining, by the design optimization application module, whether each of the second sets and the previously-determined non-dominated design alternatives is non-dominated in the objective space; (g) repartitioning, by the design optimization application module, the design space into the first region and the second region using the multi-dimensional space division scheme with respect to the newly-determined non-dominated design alternatives; and (h) repeating (e)-(g) until an end condition of the multi-objective design optimization is reached.
14. The non-transitory computer readable storage medium of claim 13 , wherein each of said first and said second regions contains one or more contiguous portions of the design space;
wherein the first region is increased by a buffer region along the first and second region's boundary.
15. The non-transitory computer readable storage medium of claim 13 , wherein said predetermined selection scheme comprises placing each of the second set at a location having a largest minimum distance to the existing non-dominated design alternatives.
16. The non-transitory computer readable storage medium of claim 13 , wherein the multi-dimensional space division scheme comprises support vector machine.
17. The non-transitory computer readable storage medium of claim 13 , wherein said predetermined selection scheme comprises placing some of the second set in the buffer region and the remaining of the second set in the first region.