Zhongmei Gao

An active learning variable-fidelity metamodelling approach based on ensemble of metamodels and objective-oriented sequential sampling

ABSTRACT Computational simulation models with different fidelity have been widely used in complex systems design. However, running the high-fidelity (HF) simulation models tends to be very time-consuming, while incorporating low-fidelity (LF), inexpensive models into the design process may result in inaccurate design alternatives. To make a trade-off between high accuracy and low expense, an active learning variable-fidelity (VF) metamodelling approach aiming to integrate information from both LF and HF models is proposed. In the proposed VF metamodelling approach, a model fusion technology based on ensemble of metamodels is employed to map the difference between the HF and LF models. Furthermore, an active learning strategy based on a generalised objective-oriented sequential sampling strategy is introduced to make full use of the already-acquired information of difference characteristics between the HF and LF models. Several numerical and engineering cases verify the applicability of the proposed VF metamodelling approach. Different types of test cases, sample sizes, and metamodel performance evaluation measures including accuracy and robustness are considered.

An active learning metamodeling approach by sequentially exploiting difference information from variable-fidelity models

We propose an active learning variable-fidelity metamodeling approach (AL-VFM).Information from high-fidelity and low-fidelity models is integrated in AL-VFM.An active learning strategy is introduced to use the already-acquired information.Numerical and engineering cases verify the applicability of the proposed approach. Complex system engineering design optimization based on simulation is a very time-consuming, even computationally prohibitive process. To relieve the computational burden, metamodels are commonly used to replace the computation-intensive simulations. In this paper, an active learning variable fidelity (VF) metamodeling approach (AL-VFM) is proposed for the purpose of integrating information from both low-fidelity (LF) and high-fidelity (HF) models. In AL-VFM, Kriging metamodel is adopted to map the difference between the HF and LF models aiming to approach the HF model on the entire domain. Besides, a general active learning strategy is introduced in AL-VFM to make full use of the already-acquired information to guide the VF metamodeling. The already-acquired information represents the location of regions where the differences between the HF and LF models are multi-model, non-smooth and have abrupt changes. Several numerical and engineering cases with different degrees of difficulty verify the applicability of the proposed VF metamodeling approach.

Optimization of laser brazing onto galvanized steel based on ensemble of metamodels

Laser brazing (LB) provides a promising way to join the galvanized steel in automotive industry for its significant advantages including high speed, small heat-affected zone, and high welding seam quality. The process parameters of LB have significant effects on the bead profile and hence the quality of joint. Since the relationships between the process parameters and bead profile cannot be expressed explicitly, it is impractical to determine the optimal process parameters intuitively. This paper proposes an optimization methodology by combining genetic algorithm (GA) and ensemble of metamodels (EMs) to address the process parameters optimization of the bead profile in LB with crimping butt. Firstly, Taguchi experimental design is adopted to generate the experimental points. Secondly, the relationships between process parameters (i.e., welding speed, wire feed rate, gap) and the bead geometries are fitted using EMs based on the experimental data. The comparative results show that the EMs can take advantage of the prediction ability of each stand-alone metamodel and thus decrease the risk of adopting inappropriate metamodels. Then, the GA is used to facilitate design space exploration and global optimum search. Besides, the main effects and contribution rates of multiple process parameters on bead profile are analyzed. Eventually, the verification experiments are carried out to demonstrate the effectiveness and reliability of the obtained optimal parameters. Overall, the proposed hybrid approach, GA–EMs, exhibits great capability of guiding the actual LB processing and improving welding quality.

Optimization of Process Parameters of Hybrid Laser–Arc Welding onto 316L Using Ensemble of Metamodels

Hybrid laser–arc welding (LAW) provides an effective way to overcome problems commonly encountered during either laser or arc welding such as brittle phase formation, cracking, and porosity. The process parameters of LAW have significant effects on the bead profile and hence the quality of joint. This paper proposes an optimization methodology by combining non-dominated sorting genetic algorithm (NSGA-II) and ensemble of metamodels (EMs) to address multi-objective process parameter optimization in LAW onto 316L. Firstly, Taguchi experimental design is adopted to generate the experimental samples. Secondly, the relationships between process parameters (i.e., laser power (P), welding current (A), distance between laser and arc (D), and welding speed (V)) and the bead geometries are fitted using EMs. The comparative results show that the EMs can take advantage of the prediction ability of each stand-alone metamodel and thus decrease the risk of adopting inappropriate metamodels. Then, the NSGA-II is used to facilitate design space exploration. Besides, the main effects and contribution rates of process parameters on bead profile are analyzed. Eventually, the verification experiments of the obtained optima are carried out and compared with the un-optimized weld seam for bead geometries, weld appearances, and welding defects. Results illustrate that the proposed hybrid approach exhibits great capability of improving welding quality in LAW.

A multi-fidelity information fusion metamodeling assisted laser beam welding process parameter optimization approach

A multi-fidelity (MF) metamodel assisted welding parameter optimization is proposed.A 3D thermal finite element model is developed as a low-fidelity model.A laser welding physical experiment is taken as the high-fidelity model.The MF metamodel is built based on two different levels fidelity information fusion.Verification experiments illustrated the reliability of the final optima. Selecting reasonable laser beam welding (LBW) process parameters is very helpful for obtaining a good welding bead profile and hence a high quality of the welding joint. Existing process parameter optimization approaches for LBW either based on cost-expensive physical experiments or low-fidelity (LF) computer simulations. This paper proposes a multi-fidelity (MF) metamodel based LBW process parameter optimization approach, in which different levels fidelity information, both from LF computer simulations and high-fidelity (HF) physical experiments can be integrated and fully exploited. In the proposed approach, a three-dimensional thermal finite element model is developed as the LF model, which is fitted with a LF metamodel firstly. Then, by taking the LF metamodel as a base model and scaling it using the HF physical experiments, a MF metamodel is constructed to approximate the relationships between the LBW process parameters and the bead profile. Two metrics are adopted to compare the prediction accuracy of the MF metamodel with the single-fidelity metamodels solely constructed with physical experiments or computer simulations. Results illustrate that the MF metamodel outperforms the single-fidelity metamodels both in global and local accuracy. Finally, the fast elitist non-dominated sorting genetic algorithm (NSGA-II) is used to facilitate LBW process parameter space exploration and multi-objective Pareto optima search. LBW verification experiments verify the effectiveness and reliability of the obtained optimal process parameters.

A multi-objective robust optimization approach for engineering design under interval uncertainty

Purpose Engineering system design and optimization problems are usually multi-objective, constrained and have uncertainties in the inputs. These uncertainties might significantly degrade the overall performance of engineering systems and change the feasibility of the obtained solutions. In this paper, a multi-objective robust optimization approach based on Kriging metamodel (K-MORO) is proposed to obtain the robust Pareto set under the interval uncertainty. Design/methodology/approach In K-MORO, the nested optimization structure is reduced into a single loop optimization structure to ease the computational burden. Considering the interpolation uncertainty from the Kriging metamodel may affect the robustness of the Pareto optima, an objective switching and sequential updating strategy is introduced in K-MORO to determine (1) whether the robust analysis or the Kriging metamodel should be used to evaluate the robustness of design alternatives, and (2) which design alternatives are selected to improve the pre...