Self-adaptive, multipopulation differential evolution in dynamic environments

Novoa-Hernández, Pavel; Corona, Carlos Cruz; Pelta, David A.

doi:10.1007/s00500-013-1022-x

Cited by 36 publications

(29 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…(12)) found in the first experiment, was compared to the DOP DE, DOP jDE (i.e., equivalent to single agent MAS-jDE) and the following stateof-the-art algorithms: jDE* [2], CDDE Ar [5], MLSDO [6] and mSQDE-i [9]. Algorithms DOP DE and DOP jDE are original DE [11] and jDE [1] algorithms that were adopted to solve DOP, by randomly reinitializing population when change in environment is detected.…”

Section: Resultsmentioning

confidence: 99%

“…Lepagnot et al [6] presented MLSDO algorithm, which is based on several coordinated local searches and on the archiving of the found local optima, in order to track them after a change in the objective function. Novoa et al [9] proposed an algorithm mSQDEi, a multi-population algorithm with self-adaptive strategy for controlling the population diversity and an interaction mechanism between individuals. This paper proposes multi-agent system based on self-ada-ptive differential evolution (MAS-jDE) for solving DOP.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multi-agent System Based on Self-adaptive Differential Evolution for Solving Dynamic Optimization Problems

Čep¹,

Fister²

2017

StuCoSReC. Proceedings of the 2017 4th Student Computer Science Research Conference.

View full text Add to dashboard Cite

This article presents the multi-agent system based on a selfadaptive differential evolution algorithm for solving dynamic problems. Characteristics of dynamic problems are that objective function, with which the quality of solutions is evaluated, changes over time. These changes can occur either after some predefined number of generations or, more commonly, in each generation, where the online responses are desired by the evolutionary algorithms. In this study, the former kind of problems were taken into consideration and were solved using multi-agent systems. Each agent in the systems is implemented as a self-adapted differential evolution with constant population size. In our comparative study, multiagent systems consisting of various population sizes in combination with various number of agents were compared, by solving the benchmark functions provided for CEC'09 Competition on Dynamic Optimization with respect to the same number of the fitness function evaluation. The main obstacle when using smaller populations is to prevent the stagnation of the population. That was partially overcome by using additional mechanisms such as: diversity measurement and information sharing between agents. As a result, the most appropriate multi-agent system was searched for, and the results of the best found multi-agent system were compared with the state-of-the-art algorithms.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Multi-agent System Based on Self-adaptive Differential Evolution for Solving Dynamic Optimization Problems

Čep¹,

Fister²

2017

StuCoSReC. Proceedings of the 2017 4th Student Computer Science Research Conference.

View full text Add to dashboard Cite

show abstract

“…More recently [21] pointed out that another alternative is to implement self-adaptive strategies [20] to cope with changes. This latter approach provides the algorithm, the ability to intelligently react to environment variations, as was shown by [22,24].…”

Section: Dynamic Optimization Problemsmentioning

confidence: 98%

“…On the other hand, in the context of dynamic environments, using multi-population and self-adaptive approaches have shown to be very effective [9,22,24], specially, when combined with the differential evolution metaheuristic [28]. While the use of several populations enables a proper exploration of the search space, self-adaptation contributes to enhance the algorithm diversity and the optimum tracking over time.…”

Section: Proposed Approachmentioning

confidence: 99%

A Hybrid Approach for Solving Dynamic Bi-level Optimization Problems

Samaniego¹,

Novoa-Hernández²

2018

CyS

Self Cite

View full text Add to dashboard Cite

Several real-life decision scenarios are hierarchical, which are commonly modeled as bi-level optimization problems (BOPs).As other decision scenarios, these problems can be dynamic, that is, some elements of their mathematical model can change over time. This kind of uncertainty imposes an extra level of complexity on the model, since the algorithm needs to find the best bi-level solution over time. Despite the importance of studying these problems, the literature reflects just a few works on dynamic bi-level optimization problems (DBOPs). In this context, this work addresses the solution of DBOPs from the viewpoint of metaheuristic methods. Our hypothesis is that, by hybridizing successful solving approaches from both bi-level and dynamic optimization fields, an effective method for DBOPs can be obtained. In this regard, we propose a hybrid method that combines a coevolutionary approach and a self-adaptive, multipopulation algorithm. Experimental results assert our hypothesis, specially for certain information exchange mechanisms.

show abstract

“…Moreover, a dexterous method called "ensemble learning" is used for further enhancing the solution diversity to reach global optimum point within a reasonable computational time. This strategy was previously applied on the variants of Particle Swarm Optimization [29,30] and Differential Evolution [31,32] algorithms. With the application of ensemble learning strategy, Global Best algorithm aims to benefit variety of mutation schemes and control parameters of the Differential evolution method as well as maintaining a favorable balance between global and local search mechanisms.…”

Section: Fundamentals Of Global Best Algorithmmentioning

confidence: 99%

Global Best Algorithm Based Parameter Identification of Solar Cell Models

Turgut¹

2017

ijisae

View full text Add to dashboard Cite

Effectivity of the solar energy systems is thoroughly dependent of successful modeling of the I-V characteristic curves. However, due to the lack of information about the precise model parameters those are profoundly involved in characterizing governing equations; an efficient design has not been accurately accomplished by researchers yet. This article proposes Global Best Algorithm (GBEST) in order to extract unknown parameters of solar cell models accurately. In order to test the performance of the proposed optimizer, nine different unconstrained optimization test functions are evaluated and their statistical results are compared with the recently developed metaheuristic algorithms. GBEST is applied on PV module, single and double diode models which are mathematically formulated as multi-dimensional and highly nonlinear in their nature. Results reveal that GBEST is superior to the other methods in terms of solution accuracy and efficiency.

show abstract

Self-adaptive, multipopulation differential evolution in dynamic environments

Cited by 36 publications

References 38 publications

Multi-agent System Based on Self-adaptive Differential Evolution for Solving Dynamic Optimization Problems

Multi-agent System Based on Self-adaptive Differential Evolution for Solving Dynamic Optimization Problems

A Hybrid Approach for Solving Dynamic Bi-level Optimization Problems

Global Best Algorithm Based Parameter Identification of Solar Cell Models

Contact Info

Product

Resources

About