Papers by Keyword: Cultural Algorithm

Abstract: This paper presents an alternative algorithm for solving the laser cutting path problem which was modeled as Generalized Traveling Salesman Problem (GTSP). The objective is to minimize the traveling distance of laser cutting of all profiles in a given layout, where a laser beam makes a single visit and then does the complete cut of individual profile in an optimum sequence. This study proposed a hybrid method combining population-based simulated annealing (SA) with an adaptive large neighborhood search (ALNS) algorithm to solve the cutting path problem. Recombination procedures were executed alternately using swap, reversion, insertion and removal-insertion through a fitness proportionate selection mechanism. In order to reduce the computing time and maintain the solution quality, the 35% proportion of population were executed in each iteration using the cultural algorithm selection method. The results revealed that the algorithm can solve several ranges of problem size with an acceptable percentage of error compared to the best known solution.

Abstract: The substance of the deployment of radar network is a multi-parameter optimization problem. This paper presents an objective function to deploy the radar network and a shuffled frog leaping algorithm (SFLA) is proposed to implement the radar network deployment. The proposed cultural shuffled frog leaping algorithm (CSFLA) makes use of mechanism of cultural evolution to update the locations of cultural frogs. Simulation results show that the proposed CSFLA has stronger abilities of exploitation and exploration by designing new leaping equations based on knowledge strategy and information communication, which may obviously improve the performance of SFLA. The radar network deployment based on the CSFLA is superior to previous deployment based on particle swarm optimization (PSO) and the SFLA in the convergence speed and optimization results. It provides a new idea to the radar network deployment.

Abstract: In the speech emotion recognition system, voice signal recognition is the most critical step, the simple signal recognition can lead to errors. In this paper the cultural genetic method applied in speech recognition optimizes the voice features combination to find the optimal solution, and it provides effective method to improve the efficiency of the speech recognition.

Abstract: For substation locating and sizing in the distribution network, the cost of land and geographic information should be considered. A new algorithm based on cultural differential evolution algorithm (CDEA) considering geographic information factor is presented. Based on the improved technique for order preference by similarity to ideal solution (ITOPSIS), geographic information factor model is set up. CDEA is proposed by designing three kinds of knowledge in belief space to guide evolution. The new algorithm overcomes the premature phenomenon of differential evolution algorithm and improves significantly ability of global optimization. Finally the case proves that the proposed model and method is correct, having certain practical value.

Abstract: For problem of limited resources in the RFID (Radio frequency identification) system, a power resource allocation scheme is proposed. The method aims to maximize the system throughput, using cultural algorithm (CA) to search for the optimal power allocation scheme. By dynamically adjusting the signal transmission power of the reader, the overlap area between the reader can be reduced so that the maximum reading range can be obtained. Simulation results show that the algorithm has better performance in the system throughput and energy consumption, reducing the impact of interference between the readers and efficiently using the resources in RFID system.

Abstract: As for the typeIassembly line balancing problem in complex products, the mathematical model of optimization goal is established with its optimization goal of minimizing the number of workstations as well as minimizing the differences in assembly complex relationship, and with the introduction of the framework of cultural evolution, a multi-cultural particle swarm algorithm is presented. The algorithm used arranged code so that the particle can still be able to meet the job constraints after its been decoded; using crowded distance to sort operator and remove the extra particles, in order to ensure a uniform distribution of the Pareto front; in order to improve the efficiency of the convergence of the algorithm, we adjust the flight parameters of particle basing on the dynamic changes in crowding distance. Through the comparison of standard test problems with other algorithms, we indicate the effectiveness of the proposed algorithm.

Abstract: The course of socio-cultural transition is not a phenomenon of aimless, arbitrary development, but towards a clear goal. Such process of moving towards higher-level soul experience and mental state is a common goal of social species' evolution. In this paper, the model of cultural development is imitated to be the system thinking frame for developing an evolution algorithm, namely cultural evolution algorithm. It consists of several search methods with similar thinking and then proposes four strategies of the cultural evolution algorithm. Seven benchmark functions are utilized to validate the search performance of the proposed algorithm. The results show that all of the four strategies of cultural evolution algorithm have better performance when compared with relevant literatures.

Abstract: In allusion to the deficiencies existing in current structural optimization algorithm of excavator boom such as the inefficiency in expressing and handling the constraints, the insufficiency in adopting the task knowledge to direct constraint handling, and the difficulty in obtaining and adopting the optimal process knowledge, a new method of the constraints expression and handling based on cultural algorithm for excavator boom structural optimization is put forward. The mechanism of hierarchical constraints expression and handing is established to improve the efficiency of the constraint handling in the optimal process with the cultural algorithm as frame and the task knowledge as guide. And the hierarchical topographical knowledge is formed to express tacit knowledge in the population space. Then the process of constraints expressing and handling as well as its interaction with the knowledge base is discussed under the direction of dual evolution in cultural algorithm while the aim of acquisition, expression and handling for tacit constraints knowledge in optimal process is realized. Finally, structural optimization of an excavator boom is taken as an example to demonstrate the effectiveness of this method.

Abstract: This study investigated the characteristics of various water resources in Tianjin, explored the needs for the study of water resources allocation, and identified the current issues of water resources allocation in Tianjin. The water resources allocation framework was established by adopting the cultural algorithm approach. A case study was done for Tianjin by utilizing this framework to evaluate the water resources allocation in 2008. It shows that the integrated approach of water resources allocation can effectively relieve the water shortage pressure of Tianjin, maximize the benefits for the limited water resources. This study provides a theoretical guidance for the water resources allocation in this area.

Abstract: In existing multi-population multi-objective cultural algorithms, information are exchanged among sub-populations by individuals. However, migrated individuals can not reflect the evolution information enough, which limits the evolution performance.In order to enhance the migration efficiency, a novel multi-population multi-objective cultural algorithm adopting knowledge migration is proposed. Implicit knowledge extracted from the evolution process of each sub-population directly reflects the information about dominant search space. By migrating the knowledge among sub-populations at the constant interval, the algorithm realizes more effective interaction with less communication cost. Taken benchmark functions as the examples, simulation results indicate that the algorithm can effectively obtain the Pareto-optimal sets of multi-objective optimization problems. The distribution performance is also improved.