Cooperative Co-evolution with delta grouping for large scale non-separable function optimization

Omidvar, Mohammad Nabi; Li, Xiaodong; Yao, Xin

doi:10.1109/cec.2010.5585979

Cited by 208 publications

(113 citation statements)

References 8 publications

Supporting

Mentioning

113

Contrasting

Order By: Relevance

“…The number of decision variables to be optimized for the LSGO problem can reach up to 1,000 or more, whose general formulation can be expressed as follows: For solving such an LSGO problem, two typical methods have been widely discussed: the groupingbased method and non-grouping method [27]. The former method is applied for high-dimensional LSGO problems and classifies all the decision variables into several supgroups by some grouping methods, such as the random grouping method [28][29], delta grouping method [30], and differential grouping methods [31][32]. Then the decision variables located in different sup-groups are optimized cooperatively and their results are combined into the ultimate solution.…”

Section: Optimization Methodologymentioning

confidence: 99%

Optimizing Hierarchical Power Distribution of Multiple Local Energy Network Systems in Grid-Connected Mode

Zhang¹,

Zhang²,

Wang³

2017

Pol. J. Environ. Stud.

View full text Add to dashboard Cite

IntroductionWith increasing environmental pollution and the reduction of fossil energy storage, decentralized renewable energy sources (DRES) have been considered to be effective alternatives for future energy requirements [1][2]. For adapting to the characteristics of DRES (e.g., geographical dispersion), local energy network systems (LENs) are employed to strengthen the collaborative relationship between DRES and the smart grid. LENs are composed of many small-scale but self-governing LENs whose basic members include renewable energy production units (e.g., wind turbines and photovoltaic generators), controllable storage units, and intelligent consumption units [3][4]. Different LENs are coupled Pol. J. Environ. Stud. Vol. 26, No. 5 (2017), 1981-2000 Original AbstractThis paper focuses on the hierarchical power distribution optimization of multiple local energy network (LEN) systems that are formed in three levels and can be operated in six typical modes. The decentralized optimal model for each LEN (the first level) and LENs (the second level) as well as the concentrated optimal model for the top level of the system are built, respectively. For each LEN, all the basic unities such as power generated by wind turbines and photovoltaic, and their upper nodes are considered. For LENs, the aggregated results (e.g., supply-demand requirements) from each LEN are dispatched. Furthermore, in the concentrated optimal control model (the third level), the ultimate supply-demand requirements of networked LENs together with other resources such as electric vehicles are considered. Due to the large number of control resources, the whole system is formulated as a large-scale global optimization (LSGO) problem. The self-adaptive differential evolution with neighborhood search method (SaNSDE) modified with the Lagrange multiplier method is used to solve the problem. The algorithm is firstly examined on 10 constrained benchmark functions, then it is applied to our problem. Experimental results show that the proposed model and algorithm are effective and efficient.Keywords: local energy network (LEN), energy internet (EI), decentralized renewable energy sources (DRES), adaptive neighborhood search technique, electric vehicles (EVs)

show abstract

Section: Optimization Methodologymentioning

confidence: 99%

Optimizing Hierarchical Power Distribution of Multiple Local Energy Network Systems in Grid-Connected Mode

Zhang¹,

Zhang²,

Wang³

2017

Pol. J. Environ. Stud.

View full text Add to dashboard Cite

show abstract

“…Omidvar et al [15] proposed DECC-D (Differential Evolution with Cooperative Coevolution using Delta-Grouping) which uses delta value to identify interacting variables. Delta value corresponding to a decision variable is calculated by measuring the amount of change in it, in successive iterations.…”

Section: Based Variant Of Cma-es Called Cc-cma-es For Largementioning

confidence: 99%

A Survey on Metaheuristics for Solving Large Scale Optimization Problems

Singh¹,

Jana²

2017

IJCA

View full text Add to dashboard Cite

In recent years, there has been a remarkable improvement in the computing power of computers. As a result, numerous realworld optimization problems in science and engineering, possessing very high dimensions, have appeared. In the research community, they are generally labeled as Large Scale Global Optimization (LSGO) problems. Several Metaheuristics has been proposed to tackle these problems. Broadly these algorithms can be categorized in 3 groups: Standard Evolutionary Algorithms, Cooperative Co-evolution (CC) based Evolutionary Algorithms and Memetic Algorithms. This paper gives a brief introduction of some state-of-the-art Metaheuristics used in the field of LSGO, discusses their performance in CEC Competition on LSGO and finally, future scope in this field is presented.

show abstract

“…Other decomposition methods such as delta grouping [14], variable interaction learning [9], and differential grouping [6] are more sophisticated techniques that have a higher accuracy in detecting interacting variables. Since the focus of this research is not on investigating the effect of various grouping techniques, we confined our experiments to random grouping only.…”

Section: A Cooperative Co-evolutionmentioning

confidence: 99%

“…The loop on line 4 forms the main co-evolutionary cycle. At the beginning of each cycle the subcomponents are reformed if a dynamic grouping algorithm such as random grouping [5] or delta grouping [14] is used. Static grouping algorithms such as differential grouping [6] are only called once at the beginning of the algorithm.…”

Section: A Cooperative Co-evolutionmentioning

confidence: 99%

See 1 more Smart Citation

A novel hybridization of opposition-based learning and cooperative co-evolutionary for large-scale optimization

Kazimipour

Omidvar

et al. 2014

2014 IEEE Congress on Evolutionary Computation (CEC)

Self Cite

View full text Add to dashboard Cite

Abstract-Opposition-based learning (OBL) and cooperative co-evolution (CC) have demonstrated promising performance when dealing with large-scale global optimization (LSGO) problems. In this work, we propose a novel framework for hybridizing these two techniques, and investigate the performance of simple implementations of this new framework using the most recent LSGO benchmarking test suite. The obtained results verify the effectiveness of our proposed OBL-CC framework. Moreover, some advanced statistical analyses reveal that the proposed hybridization significantly outperforms its component methods in terms of the quality of finally obtained solutions.

show abstract

Cooperative Co-evolution with delta grouping for large scale non-separable function optimization

Cited by 208 publications

References 8 publications

Optimizing Hierarchical Power Distribution of Multiple Local Energy Network Systems in Grid-Connected Mode

Optimizing Hierarchical Power Distribution of Multiple Local Energy Network Systems in Grid-Connected Mode

A Survey on Metaheuristics for Solving Large Scale Optimization Problems

A novel hybridization of opposition-based learning and cooperative co-evolutionary for large-scale optimization

Contact Info

Product

Resources

About