Evolutionary Algorithm API for Pareto-Optimal Design Exploration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Classical multiple-objective optimization techniques are inadequate for real-world problems like satellite constellation design and airline network scheduling, as they fail to find all Pareto-optimal solutions, require a priori knowledge of objective importance, and are limited by discontinuous and nonconvex objective spaces, leading to incomplete exploration of the feasible solution space.
Innovation Solution
A system and method utilizing evolutionary algorithms with a processor and memory to execute evolutionary algorithms, allowing specification of parameters, definition of chromosome data structures, identification of objective functions, and generation of optimal chromosome data structures as designs, enabling parallel processing and infeasible space management to overcome these limitations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If classical multiple-objective optimization techniques are used, then convergence proofs exist and decision maker knowledge is utilized, but all Pareto-optimal solutions cannot be found in discontinuous and nonconvex objective spaces
Solution Approach 1:
The patent transforms the multiple-objective optimization problem into a single-objective optimization problem by changing the parameter representation. It uses a utility function that combines multiple objectives with weights, allowing the algorithm to search for solutions that optimize the combined objective while maintaining the ability to explore diverse solution spaces including discontinuous and nonconvex regions.
Solution Approach 2:
The patent segments the solution space exploration by using multiple random restarts and maintaining a diverse population of solutions. This allows the algorithm to explore different regions of the objective space independently, increasing the likelihood of finding all Pareto-optimal solutions even in complex, discontinuous landscapes.
2Ease of operation
If classical optimization methods require a priori knowledge of objective importance, then the optimization can be guided, but discovery in the feasible solution space is limited
Solution Approach 1:
The patent implements dynamic weighting of objectives during the optimization process. The weights assigned to different objectives change over time based on the progress of the search and the diversity of solutions found, allowing the algorithm to adaptively explore different regions of the objective space without requiring fixed a priori knowledge of objective importance.
Solution Approach 2:
The algorithm performs self-adjustment by automatically modifying its search behavior based on the solutions it discovers. Through mechanisms like adaptive weighting and diversity maintenance, the system serves itself by learning from its own search progress, enabling discovery without external guidance about objective priorities.
3Use of energy by moving object
If only one Pareto-optimal solution is found with one run of a classical algorithm, then computational resources are conserved, but the diversity of optimal solutions is limited
Solution Approach 1:
The patent maintains continuous exploration of the solution space by running multiple optimization processes in parallel or sequentially with different initial conditions. This continuous action ensures that diverse Pareto-optimal solutions are found without wasting computational resources, as each run contributes unique information about the solution landscape.
Solution Approach 2:
The patent merges results from multiple optimization runs by maintaining a combined population of solutions and applying selection pressure to preserve diversity. This combining approach allows the system to leverage computational effort from multiple runs while ensuring that the final result includes a diverse set of Pareto-optimal solutions.
Data Source
AI summary
Systems and methods may include an application program interface that enables a user to: specify parameters associated with an evolutionary algorithm, where an execution of the evolutionary algorithm is in accordance with the specified parameters; define a chromosome data structure that includes a plurality of variables that are permitted to evolve in value in accordance with the execution of the evolutionary algorithm in order to generate one or more child chromosome data structures; identify one or more objective functions for evaluating chromosome data structures, including the generated one or more child chromosome data structures; and define an output format for providing one or more optimal chromosome data structures of the evaluated generated child chromosome data structures as designs to the identified objective functions.