The CFD-based design optimization of turbomachinery cascades is a typical high-dimensional expensive blackbox (HEB) problem. Specifically, to fully consider the interactions between vanes and stages and thus achieving...
详细信息
The CFD-based design optimization of turbomachinery cascades is a typical high-dimensional expensive blackbox (HEB) problem. Specifically, to fully consider the interactions between vanes and stages and thus achieving desirable solutions, the cascades of two or more stages have to be optimized simultaneously. The number of variables can reach to more than 100 while the affordable CFD simulations within a design cycle can be only a few hundred or thousand. Hence, how to achieve the optimal design of turbomachinery cascades on budget is very challenging. To solve this problem, a generalized surrogate-assisted differential evolution (DE) algorithm is proposed, which is labeled as GSDE. In GSDE, radial basis function is incorporated into the population regeneration process, which helps to enhance the local exploitation in each iteration and thus achieving a good balance in between global and local search. By validating on benchmark functions ranging from 50 to 100 dimensions, the proposed GSDE algorithm is observed to obtain the true optimal solution with less than 1000 function calls and have far better convergence rate than other state-of-the-art algorithms. Furthermore, GSDE is then tested on the cascades design optimization of a 3.5-stage compressor with 126 variables. Within 1500 CFD evaluations, the total-to-total efficiency is increased by 1.104%. In contrast, classic DE algorithm achieves a similar efficiency gain with nearly 20,000 CFD evaluations. In other words, the computational cost of GSDE amount to only 1/13 of frequently-used evolutionaryalgorithms. Therefore, the efficiency of the proposed GSDE algorithm for solving high-dimensional turbomachinery optimization problems is well demonstrated.
surrogate-assisted evolutionary algorithms (SAEAs) have recently received increasing attention in solving computationally expensive engineering optimization problems. Existing studies have shown that surrogate modelin...
详细信息
surrogate-assisted evolutionary algorithms (SAEAs) have recently received increasing attention in solving computationally expensive engineering optimization problems. Existing studies have shown that surrogate modeling techniques based on different radial basis functions (RBF) can highly affect the search capability of an optimizer. However, without any prior knowledge about the optimization problem to be solved, it is very hard for a designer to decide which modeling techniques should be used. To defeat this issue, we suggested a brand-new model management strategy based on multi-RBF parallel modeling technology in this paper. The proposed strategy aims to adaptively select a high-fidelity surrogate from a pre-specified set of RBF modeling techniques during the optimization process. At each evolutionary interaction, the most promising RBF surrogate was employed to help neighborhood field optimizer (NFO) perform fitness evaluation, and the proposed algorithm is named aRBF-NFO. Moreover, a detailed experimental analysis was given to show the effectiveness of the proposed method, and an overall comparison was made between the aRBF-NFO and two state-of-the-art SAEAs on a commonly-used test set as well as an antenna optimization problem. Experimental results demonstrate the proposed algorithm is robust and efficient.
surrogate-assisted evolutionary algorithms (SAEAs) have been proven to be powerful optimization tools for tackling Expensive Optimization Problems (EOPs) where a limited number of function evaluations are available. H...
详细信息
surrogate-assisted evolutionary algorithms (SAEAs) have been proven to be powerful optimization tools for tackling Expensive Optimization Problems (EOPs) where a limited number of function evaluations are available. However, many SAEAs are only designed for low- or medium-dimensional EOPs. Existing SAEAs are challenging to address High-dimensional EOPs (HEOPs) owing to the curse of dimensionality and lack of powerful exploitation capacity. To tackle HEOPs efficiently, a surrogate-assisted Fully-informed Particle Swarm Optimization (SA-FPSO) algorithm is proposed in this paper. Firstly, a generation-based Social Learning based PSO (SLPSO) is adopted to explore the whole decision space with the help of the global surrogate model. Secondly, the fully-informed search scheme is incorporated into the framework of SLPSO to improve its exploitation capacity in the surrogate-assisted search environment. Thirdly, a local space identification strategy is proposed to determine the search range for the local surrogate-assisted search. Seven commonly used expensive benchmark functions with dimensions ranging from 30D to 300D are used to verify the performance of SA-FPSO for HEOPs. Experiment results indicate that SA-FPSO obtains superior performance over several state-of-the-art SAEAs both in terms of convergence speed and solution accuracy.
Radial basis function (RBF) models have attracted a lot of attention in assisting evolutionaryalgorithms for solving computationally expensive optimization problems. However, most RBFs cannot directly provide the unc...
详细信息
Radial basis function (RBF) models have attracted a lot of attention in assisting evolutionaryalgorithms for solving computationally expensive optimization problems. However, most RBFs cannot directly provide the uncertainty information of their predictions, making it difficult to adopt principled infill sampling criteria for model management. To overcome this limitation, an inverse distance weighting (IDW) and RBF based surrogateassistedevolutionaryalgorithm, named IR-SAEA, is proposed to address high -dimensional expensive multi -objective optimization problems. First, an RBF-IDW model is developed, which can provide both the predicted objective values and the uncertainty of the predictions. Moreover, a modified lower confidence bound infill criterion is proposed based on the RBF-IDW for the balance of exploration and exploitation. Extensive experiments have been conducted on widely used benchmark problems with up to 100 dimensions. The empirical results have validated that the proposed algorithm is able to achieve a competitive performance compared with state-of-the-art SAEAs.
Architectures generation optimization has been received a lot of attention in neural architecture search (NAS) since its efficiency in generating architecture. By learning the architecture representation through unsup...
详细信息
Architectures generation optimization has been received a lot of attention in neural architecture search (NAS) since its efficiency in generating architecture. By learning the architecture representation through unsupervised learning and constructing a latent space, the prediction process of predictors is simplified, leading to improved efficiency in architecture search. However, searching for architectures with top performance in complex and large NAS search spaces remains challenging. In this paper, an approach that combined a ranker and generative model is proposed to address this challenge through regularizing the latent space and identifying architectures with top rankings. We introduce the ranking error to gradually regulate the training of the generative model, making it easier to identify architecture representations in the latent space. Additionally, a surrogate-assistedevolutionary search method that utilized neural network Bayesian optimization is proposed to efficiently explore promising architectures in the latent space. We demonstrate the benefits of our approach in optimizing architectures with top rankings, and our method outperforms state-of-the-art techniques on various NAS benchmarks. The code is available at .
Multi-objective optimization problems (MOPs) involve optimizing multiple conflicting objectives simultaneously, resulting in a set of Pareto optimal solutions. Due to the high computational or financial cost associate...
详细信息
Multi-objective optimization problems (MOPs) involve optimizing multiple conflicting objectives simultaneously, resulting in a set of Pareto optimal solutions. Due to the high computational or financial cost associated with evaluating fitness, expensive multi-objective optimization problems (EMOPs) further complicate the optimization process. surrogate-assisted evolutionary algorithms (SAEAs) have emerged as a promising approach to address EMOPs by substituting costly evaluations with computationally efficient surrogate models. This paper introduces the self-organizing surrogate-assisted non-dominated sorting differential evolution (SSDE), which uses surrogate model based on a self-organizing map (SOM) to approximate the fitness function. SSDE offers advantages such as reduced computational cost, improved accuracy, and the speed of enhanced convergence. The SOM-based surrogate models effectively capture the underlying structure of the Pareto optimal set and Pareto optimal front, leading to superior approximations of the fitness function. Experimental results on benchmark functions and real-world problems, including Model-Free Adaptive Control (MFAC) and the Yagi-Uda Antenna design, demonstrate the competitiveness and efficiency of SSDE compared to other algorithms.
Risers with high-grade steel are widely used in offshore oil and gas industry at present. The extreme weight, lower fatigue and corrosion resistance of steel risers significantly limited the exploitation depths and th...
详细信息
Risers with high-grade steel are widely used in offshore oil and gas industry at present. The extreme weight, lower fatigue and corrosion resistance of steel risers significantly limited the exploitation depths and the production capacity. Nowadays, it is acknowledged that using fibre-reinforced polymer composites to manufacture risers can be a better option. The prototypes of composite risers fabricated and tested confirm that fibre-reinforced polymer composites have an obvious advantage over steel risers on weight saving. Three different approaches are developed here to minimise composite risers' weights: (1) enhancing the riser with only axial-direction and hoop-direction fibre;(2) off-axis reinforcements are included using an iterative approach of manual inspection and selection and (3) employing the optimisation technique of surrogate-assisted evolutionary algorithm. These design approaches have been applied to eight different material combinations to achieve the minimum structural weight by optimising their laminate configurations. The designs are conducted in accordance with the Standards, considering both local load cases and global - functional as well as environmental loads using ANSYS 15.0. The results show that comparing with steel risers, weight savings achieved by different design approaches and material combinations are different.
暂无评论