Multi-objective optimization

• US 20020099929A1
• Filed: 11/09/2001
• Published: 07/25/2002
• Est. Priority Date: 11/14/2000
• Status: Active Grant

• Current Assignee: Honda Research Institute Europe GmbH (Honda Motor Company)
• Original Assignee: Honda Research Institute Europe GmbH (Honda Motor Company)
• Inventors: Sendhoff, Bernhard, Jin, Yaochu
• Granted Patent: US 7,363,280 B2
• Time in Patent Office: Days
• US Class Current: 712/220
• CPC Class Codes: G06N 3/126 Evolutionary algorithms, e.g. genetic algorithms or genetic programming
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202

Abstract

In the field of multi-objective optimization using evolutionary algorithms conventionally different objectives are aggregated and combined into one objective function using a fixed weight when more than one objective needs to be optimized. With such a weighted aggregation, only one solution can be obtained in one run. Therefore, according to the present invention two methods to change the weights systematically and dynamically during the evolutionary optimization are proposed. One method is to assign uniformly distributed weight to each individual in the population of the evolutionary algorithm. The other method is to change the weight periodically when the evolution proceeds. In this way a full set of Pareto solutions can be obtained in one single run.

Claims

1. A method for optimizing multi-objective problems using evolutionary algorithms, the method comprising the steps of:

setting up an initial population as parents;

reproducing the parents to create a plurality of offspring individuals;

evaluating a quality of the offspring individuals by means of a fitness function, wherein the fitness function includes a sum of weighted sub-functions that represent an objective, said weights for the sub-functions are changed dynamically during the optimization; and

selecting one or more offspring having a highest evaluated quality value as parents for a next evolution cycle, wherein the weights for the sub-functions are changed periodically with the process of optimization.

2. The method of claim 1, wherein each offspring has a same weight in a same generation.

3. The method of claim 1, wherein the weights for the sub-functions are changed gradually between 0 and 1 with the process of optimization.

4. The method of claim 3, wherein the periodic change has the shape of a sine function applied on the generation number.

5. The method of claim 1 further comprising the step of:

recording, in an archive, Pareto solutions found as optimal solutions for a multi-objective problem.

6. The method of claim 1, wherein a pressure loss and an outlet angle calculated by a Navier-Stokes-solver and geometric constraints are used to optimize an aerodynamic body.

7. A computer program stored in a computer readable medium for performing the method of claim 1.