Browsing by Author "Ruan, Gan"

Now showing 1 - 10 of 10
  • Thumbnail Image
    ItemOpen Access
    An adaptation reference-point-based multiobjective evolutionary algorithm
    (Elsevier, 2019-03-11) Zou, Juan; Fu, Liuwei; Yang, Shengxiang; Zheng, Jinhua; Ruan, Gan; Pei, Tingrui; Wang, Lei
    It is well known that maintaining a good balance between convergence and diversity is crucial to the performance of multiobjective optimization algorithms (MOEAs). However, the Pareto front (PF) of multiobjective optimization problems (MOPs) affects the performance of MOEAs, especially reference point-based ones. This paper proposes a reference-point-based adaptive method to study the PF of MOPs according to the candidate solutions of the population. In addition, the proportion and angle function presented selects elites during environmental selection. Compared with five state-of-the-art MOEAs, the proposed algorithm shows highly competitive effectiveness on MOPs with six complex characteristics.
  • Thumbnail Image
    ItemOpen Access
    The effect of diversity maintenance on prediction in dynamic multi-objective optimization
    (Elsevier, 2017-05-10) Ruan, Gan; Yu, Guo; Zheng, Jinhua; Zou, Juan; Yang, Shengxiang
    There are many dynamic multi-objective optimization problems (DMOPs) in real-life engineering applications whose objectives change over time. After an environmental change occurs, prediction strategies are commonly used in dynamic multi-objective optimization algorithms to find the new Pareto optimal set (POS). Being able to make more accurate prediction means the algorithm requires fewer computational resources to make the population approximate to the Pareto optimal front (POF). This paper proposes a hybrid diversity maintenance method to improve prediction accuracy. The method consists of three steps, which are implemented after an environmental change. The first step, based on the moving direction of the center points, uses the prediction to relocate a number of solutions close to the new Pareto front. On the basis of self-defined minimum and maximum points of the POS in this paper, the second step applies the gradual search to produce some well-distributed solutions in the decision space so as to compensate for the inaccuracy of the first step, simultaneously and further enhancing the convergence and diversity of the population. In the third step, some diverse individuals are randomly generated within the region of next probable POS, which prompts the diversity of the population. Eventually the prediction becomes more accurate as the solutions with good convergence and diversity are selected after the non-dominated sort on the combined solutions generated by the three steps. Compared with three other prediction methods on a series of test instances, our method is very competitive in convergence and diversity as well as the speed at which it responds to environmental changes.
  • Thumbnail Image
    ItemOpen Access
    An evolutionary dynamic multi-objective optimization algorithm based on center-point prediction and sub-population autonomous guidance
    (2018-11) Zhou, Jianwei; Zou, Juan; Yang, Shengxiang; Ruan, Gan; Ou, Junwei; Zheng, Jinhua
    Dynamic multi-objective optimization problems (DMOPs) provide a challenge in that objectives conflict each other and change over time. In this paper, a hybrid approach based on prediction and autonomous guidance is proposed, which responds the environmental changes by generating a new population. According to the position of historical population, a part of the population is generated by predicting roughly and quickly. In addition, another part of the population is generated by autonomous guidance. A sub-population from current population evolves several generations independently, which guides the current population into the promising area. Compared with other three algorithms on a series of benchmark problems, the proposed algorithm is competitive in convergence and diversity. Empirical results indicate its superiority in dealing with dynamic environments.
  • Thumbnail Image
    ItemOpen Access
    High-dimensional multi-objective optimization strategy based on decision space oriented search
    (2019-09) Zheng, Jinhua; Dong, Jiangnan; Ruan, Gan; Zou, Juan; Yang, Shengxiang
    Traditional multi-objective evolutionary algorithm (MOEA) have sound performance when solving low dimensional continuous multi-objective optimization problems. However, as the optimization problems’ dimensions increase, the difficulty of optimization will also increase dramatically. The main reasons are the lack of algorithms’ search ability, and the smaller selection pressure when the dimension increases as well as the difficulty to balance convergence and distribution conflicts. In this study, after analyzing the characteristics of the continuous multi-objective optimization problem, a directional search strategy based on decision space (DS) is proposed to solve high dimensional multi-objective optimization problems. This strategy can be combined with the MOEAs based on the dominating relationship. DS first samples solutions from the population and analyzes them, and obtains the controlling vectors of convergence subspace and distribution subspace by analyzing the problem characteristics. The algorithm is divided into convergence search stage and distribution search stage, which correspond to convergent subspace and distributive subspace respectively. In different stages of search, sampling analysis are used results to macroscopically control the region of offspring generation. The convergence and distribution are divided and emphasized in different stages to avoid the difficulty of balancing them. Additionally, it can also relatively focuses the search resources on certain aspect in certain stages, which facilitates the searching ability of the algorithm. In the experiment, NSGA-II and SPEA2 algorithms are compared combining DS strategy with original NSGA-II and SPEA2 algorithms, and DS-NSGA-II is used as an example to compare it with other state-of-the-art high-dimensional algorithms, such as MOEAD-PBI, NSGA-III, Hype, MSOPS, and LMEA. The experimental results show that the introduction of the DS strategy greatly improves the performance of NSGA-II and SPEA2 when addressing high dimensional multi-objective optimization problems. It is also shown that DS-NSGA-II is more competitive when compared the existing classical high dimensional multi-objective algorithms.
  • Thumbnail Image
    ItemOpen Access
    A Pareto-based evolutionary algorithm using decomposition and truncation for dynamic multi-objective optimization
    (Elsevier, 2019-08-21) Ou, Junwei; Zheng, Jinhua; Ruan, Gan; Hu, Yaru; Zou, Juan; Li, Miqing; Yang, Shengxiang; Tan, Xu
    Maintaining a balance between convergence and diversity of the population in the objective space has been widely recognized as the main challenge when solving problems with two or more conflicting objectives. This is added by another difficulty of tracking the Pareto optimal solutions set (POS) and/or the Pareto optimal front (POF) in dynamic scenarios. Confronting these two issues, this paper proposes a Pareto-based evolutionary algorithm using decomposition and truncation to address such dynamic multi-objective optimization problems (DMOPs). The proposed algorithm includes three contributions: a novel mating selection strategy, an efficient environmental selection technique and an effective dynamic response mechanism. The mating selection considers the decomposition-based method to select two promising mating parents with good diversity and convergence. The environmental selection presents a modified truncation method to preserve good diversity. The dynamic response mechanism is evoked to produce some solutions with good diversity and convergence whenever an environmental change is detected. In the experimental studies, a range of dynamic multi-objective benchmark problems with different characteristics were carried out to evaluate the performance of the proposed method. The experimental results demonstrate that the method is very competitive in terms of convergence and diversity, as well as in response speed to the changes, when compared with six other state-of-the-art methods.
  • Thumbnail Image
    ItemOpen Access
    A performance indicator for reference-point-based multiobjective evolutionary optimization
    (2018-11) Hou, Zhanglu; Yang, Shengxiang; Zou, Juan; Zheng, Jinhua; Yu, Guo; Ruan, Gan
    Aiming at the difficulty in evaluating preference-based evolutionary multiobjective optimization, this paper proposes a new performance indicator. The main idea is to project the preferred solutions onto a constructed hyperplane which is perpendicular to the vector from the reference (aspiration) point to the origin. And then the distance from preferred solutions to the origin and the standard deviation of distance from each mapping point to the nearest point will be calculated. The former is used to measure the convergence of the obtained solutions. The latter is utilized to assess the diversity of preferred solutions in the region of interest. The indicator is conducted to assess different algorithms on a series of benchmark problems with various features. The results show that the proposed indicator is able to properly evaluate the performance of preference-based multiobjective evolutionary algorithms.
  • Thumbnail Image
    ItemOpen Access
    A predictive strategy based on special points for evolutionary dynamic multi-objective optimization
    (Springer, 2018-01-27) Li, Qingya; Zou, Juan; Yang, Shengxiang; Zheng, Jinhua; Ruan, Gan
    There are some real-world problems in which multiple objectives conflict with each other and the objectives change with time. These problems require an optimization algorithm to track the moving Pareto front or Pareto set over time. In this paper, we propose a predictive strategy based on special points (SPPS) which consists of three mechanisms. The first one is that the non-dominated set is predicted directly by feed-forward center points, which can eliminate many useless individuals predicted by traditional prediction using feed-forward center points. The second one is that a special point set(such as boundary point, knee point, etc.) is introduced into the predicted population which can track Pareto front or Pareto set more accurately. The third one is the adaptive diversity maintenance mechanism based on boundary points and center points. The mechanism can introduce diverse individuals of the corresponding number according to the degree of difficulty of the problem to keep the diversity of the population. The number of these diverse individuals is strongly related to the center points. Then, they are generated evenly throughout the decision space between the boundary points. The proposed strategy is compared with the four other state-of-the-art strategies. The experimental results show that SPPS can do well for dynamic multi-objective optimization.
  • Thumbnail Image
    ItemOpen Access
    A proportion-based selection scheme for multi-objective optimization
    (IEEE Press, 2018-02-08) Fu, Liuwei; Zou, Juan; Yang, Shengxiang; Ruan, Gan; Zheng, Jinhua; Ma, Zhongwei
    Classical multi-objective evolutionary algorithms (MOEAs) have been proven to be inefficient for solving multiobjective optimizations problems when the number of objectives increases due to the lack of sufficient selection pressure towards the Pareto front (PF). This poses a great challenge to the design of MOEAs. To cope with this problem, researchers have developed reference-point based methods, where some well-distributed points are produced to assist in maintaining good diversity in the optimization process. However, the convergence speed of the population may be severely affected during the searching procedure. This paper proposes a proportion-based selection scheme (denoted as PSS) to strengthen the convergence to the PF as well as maintain a good diversity of the population. Computational experiments have demonstrated that PSS is significantly better than three peer MOEAs on most test problems in terms of diversity and convergence.
  • Thumbnail Image
    ItemOpen Access
    A random benchmark suite and a new reaction strategy in dynamic multiobjective optimization
    (Elsevier, 2021-03-27) Ruan, Gan; Zheng, Jinhua; Juan, Zou; Ma, Zhongwei; Yang, Shengxiang
    In the domain of evolutionary computation, more and more attention has been paid to dynamic multiobjective optimization. Generally, artificial benchmarks are effective tools for the performance evaluation of dynamic multiobjective evolutionary algorithms (DMOEAs). After reviewing existing benchmarks and highlighting their weaknesses, this paper proposes a new benchmark suite to promote the comprehensive testing of algorithms. This proposed benchmark suite has eight random instances in which the randomness is produced by designed random time sequences. Also, this suite introduces challenging but rarely considered characteristics, including diverse features in fitness landscape (e.g. deception, multimodality, and bias) and complex trade-off geometries (e.g. convexity-concavity mixed geometry and disconnected geometry). Empirical studies have shown that the proposed benchmark poses reasonable challenges to DMOEAs in terms of convergence and diversity. Besides, a center matching strategy (CMS) is suggested to track random changes in these problems, which applies the history individual information in a global scope for population prediction. Compared with other reaction strategies, CMS has been demonstrated to be very competitive in dealing with random problems.
  • Thumbnail Image
    ItemOpen Access
    Solving dynamic multi-objective problems with an evolutionary multi-directional search approach
    (Elsevier, 2019-11-08) Hu, Yaru; Ou, Junwei; Zheng, Jinhua; Zou, Juan; Yang, Shengxiang; Ruan, Gan
    The challenge of solving dynamic multi-objective optimization problems is to effectively and efficiently trace the varying Pareto optimal front and/or Pareto optimal set. To this end, this paper proposes a multi-direction search strategy, aimed at finding the dynamic Pareto optimal front and/or Pareto optimal set as quickly and accurately as possible before the next environmental change occurs. The proposed method adopts a multi-directional search approach which mainly includes two parts: an improved local search and a global search. The first part uses individuals from the current population to produce solutions along each decision variable’s direction within a certain range and updates the population using the generated solutions. As a result, the first strategy enhances the convergence of the population. In part two, individuals are generated in a specific random method along every dimension’s orientation in the decision variable space, so as to achieve good diversity as well as guarantee the avoidance of local optimal solutions. The proposed algorithm is measured on several benchmark test suites with various dynamic characteristics and different difficulties. Experimental results show that this algorithm is very competitive in dealing with dynamic multi-objective optimization problems when compared with four state-of-the-art approaches.