A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Hybrid single objective genetic algorithm coupled with the simulated annealing optimization method for building optimization
https://orcid.org/0000-0002-2309-1545
Highlights Introduction of a hybrid optimization algorithm combining the GA and SA. Goal: Improvement of reliability of genetic algorithms used for building optimization. Genetic algorithm does not always provide results close to the global optimum. Reduction of computation time compared to repetitive use of the GA. Result: Reliable and robust results of the hybrid algorithm.
Abstract Evolutionary genetic optimization algorithms (GA) have been used for thermal building optimization in the past. However, the results of these algorithms can differ significantly from each other because of random search and it is not guaranteed that the optimal solution is close to the global optimum. Furthermore, the use of these algorithms for non-expert users is limited. In this study, a hybrid single objective building optimization algorithm is introduced, which combines an evolutionary genetic algorithm with a modified simulated annealing algorithm. The goal of this paper is (1) to illustrate that the GA does not always provide solutions close to the global optimum and (2) to provide a building optimization method, which provides a higher reliability than what the GA alone can provide by using a relatively short computation time. Results illustrate that the hybrid GA coupled with the modified SA provides solutions close to the global optimum in all of the test runs in this study. The proposed algorithm therefore provides more reliable results than the GA without the addition of the modified SA.
Hybrid single objective genetic algorithm coupled with the simulated annealing optimization method for building optimization
https://orcid.org/0000-0002-2309-1545
Highlights Introduction of a hybrid optimization algorithm combining the GA and SA. Goal: Improvement of reliability of genetic algorithms used for building optimization. Genetic algorithm does not always provide results close to the global optimum. Reduction of computation time compared to repetitive use of the GA. Result: Reliable and robust results of the hybrid algorithm.
Abstract Evolutionary genetic optimization algorithms (GA) have been used for thermal building optimization in the past. However, the results of these algorithms can differ significantly from each other because of random search and it is not guaranteed that the optimal solution is close to the global optimum. Furthermore, the use of these algorithms for non-expert users is limited. In this study, a hybrid single objective building optimization algorithm is introduced, which combines an evolutionary genetic algorithm with a modified simulated annealing algorithm. The goal of this paper is (1) to illustrate that the GA does not always provide solutions close to the global optimum and (2) to provide a building optimization method, which provides a higher reliability than what the GA alone can provide by using a relatively short computation time. Results illustrate that the hybrid GA coupled with the modified SA provides solutions close to the global optimum in all of the test runs in this study. The proposed algorithm therefore provides more reliable results than the GA without the addition of the modified SA.
Hybrid single objective genetic algorithm coupled with the simulated annealing optimization method for building optimization
https://orcid.org/0000-0002-2309-1545
Highlights Introduction of a hybrid optimization algorithm combining the GA and SA. Goal: Improvement of reliability of genetic algorithms used for building optimization. Genetic algorithm does not always provide results close to the global optimum. Reduction of computation time compared to repetitive use of the GA. Result: Reliable and robust results of the hybrid algorithm.
Abstract Evolutionary genetic optimization algorithms (GA) have been used for thermal building optimization in the past. However, the results of these algorithms can differ significantly from each other because of random search and it is not guaranteed that the optimal solution is close to the global optimum. Furthermore, the use of these algorithms for non-expert users is limited. In this study, a hybrid single objective building optimization algorithm is introduced, which combines an evolutionary genetic algorithm with a modified simulated annealing algorithm. The goal of this paper is (1) to illustrate that the GA does not always provide solutions close to the global optimum and (2) to provide a building optimization method, which provides a higher reliability than what the GA alone can provide by using a relatively short computation time. Results illustrate that the hybrid GA coupled with the modified SA provides solutions close to the global optimum in all of the test runs in this study. The proposed algorithm therefore provides more reliable results than the GA without the addition of the modified SA.
Hybrid single objective genetic algorithm coupled with the simulated annealing optimization method for building optimization
Junghans, Lars (author) / Darde, Nicholas (author)
Energy and Buildings ; 86 ; 651-662
2014-10-19
12 pages
Article (Journal)
Electronic Resource
English
Simulated Annealing-Genetic Algorithm for Transit Network Optimization
| British Library Online Contents | 2006
Multiobjective simulated annealing optimization of concrete building frames
| British Library Conference Proceedings | 2006