Figure - available from: Structural and Multidisciplinary Optimization
This content is subject to copyright. Terms and conditions apply.
Source publication
The front longitudinal beam (FLB) is the most important energy-absorbing and crashing force–transmitting structure of a vehicle under front-impact collision. For better weight reduction and crashworthiness of the FLB, a new structure, variable rolled blank–variable cross-sectional shape FLB (VRB-VCS FLB), is proposed. It has both the continuous var...
Citations
... The design of the L7e's S-rail for crashworthiness is challenging because of the limited design space in the crumple zone. Some researchers have proposed S-rail designs for passenger cars with various cross-sectional configurations [4][5][6][7]. Esmaeili-Marzdashti et al. [8] also conducted a parametric study and optimization of the multi-corner S-rail design for passenger cars using the response surface method (RSM) to construct a surrogate model. The results show that the multicell thin-walled member, particularly the octagonal double-cell, outperforms other geometric designs. ...
Heavy electric quadricycles (L7e) typically have the problem of crash incompatibility because their size and weight are smaller and lighter than other vehicles. The S-rails, located in front of the automotive underbody, are the main components for absorbing the impact energy of the L7e. The design of the L7e's S-rail for crashworthiness is challenging because the design space is relatively limited. High computational resources are commonly required for crash simulation due to a vast amount of data from nonlinear explicit dynamic analyses. Machine learning techniques can more quicky provide and extract accurate crash predictions from a combination of feasible solutions. This work employs a finite element analysis via LS-DYNA to obtain discrete results of the peak crushing force and the specific energy absorption for S-rail impact. The artificial neural network (ANN) module in MATLAB and response surface methodology (RSM) were then applied to predict the crushing-performance surrogate models based on the S-rail's thickness and geometry. To conduct an accurate learning process of the ANN, 81 data points for the crashworthiness simulation were adopted as training inputs with ten test data points. Furthermore, the number of neurons and activation function are examined to acquire the most accurate results for predicting the dynamic crushing of an S-rail. The investigation shows that ANN is more accurate than RSM in predicting the S-rail's peak crash force and specific energy absorption, especially when applying ReLU activation functions. With its high accuracy and capability to predict multiple targets, the ANN-ReLU is recommended to perform surrogate-based model optimization.
... Yu et al. (2021) applied the VRB process to the frontend structure of PEV and performed the crashworthiness and lightweight based on the Kriging model and the multiobjective particle swarm optimization algorithm, achieving a weight reduction of 12.8%. Duan et al. (2019b) proposed a VRB front longitudinal beam (FLB) that met the VRB rolling process and carried out the crashworthiness optimization design of the VRB FLB, which achieved 15.21% Responsible Editor: Makoto Ohsaki * Zhanpeng Du dzp1014@163.com weight reduction. ...
Variable-thickness rolled blank (VRB) structures can offer excellent crashworthiness and weight reduction potential with its large-scale applications with satisfying manufacturing constraints, whose crashworthiness optimization is classified into the high-dimensional expensive problem including explicit and implicit constraints. Therefore, an efficient parallel constrained Bayesian optimization (PCBO) algorithm is proposed to improve the global searching accuracy and efficiency from three aspects: (1) the bilog transformation for implicit constraints is introduced to reduce the difficulty of identifying the feasibility of "expensive" sample points near constraint boundaries; (2) the trust region updating strategy is introduced to balance the exploration and exploitation of the searching process by dynamically updating the searching space; (3) the parallel high-quality points addition strategy based on multiple acquisition functions (PPA-MAF) is proposed, which not only increases the diversity of the optimal solutions but also achieves the multi-task parallel computation. Seven classical cases are adopted to validate the convergence and robustness of PCBO algorithm by comparing with several popular algorithms. Finally, the crashworthiness optimization of a VRB bumper system is performed by the proposed algorithm which can get better lightweight case under satisfying the manufacturing and performance constraints.
... Given the complexity of the vehicle body structure and the diversity of linear/non-linear working conditions, the design of its components usually relies on parametric modelling and optimization method to ensure the cycle and efficiency of product development (Duan et al., 2019;Wang et al., 2018). However, the introduction of carbon fiber reinforcement plastic (CFRP) in body materials brings a great challenge to address these issues for multi-material bodies. ...
Combined application of steel, aluminum and CFRP is the main direction of future lightweight body development. However, the anisotropy and additional lamination design variables of CFRP parts poses significant challenges for the development of multi-material bodies. This study establishes a parametric design method for the variable-thickness lamination scheme based on non-uniform rational B-splines (NURBS), it can be coupled with existing parametric design methods for structural shapes to formulate a complete parametric design and modelling of CFRP components. On this basis, a homogenized intermediate material property is derived from classic laminate theory by introducing lamination assumptions, it enables a stepwise multi-material body optimization method to solve the challenge that components’ material design variables switching between CFRP and alloy will introduce/eliminate lamination design variables iteratively, posing a great optimization convergence difficulty. The proposed parametric modeling method for CFRP components was validated by experimental tests of a fabricated roof beam, and the proposed optimization method was applied to a vehicle body, achieving 15.9%, 23.9%, 18.6%, 12.2% increase in bending and tortional stiffness and modal frequencies; 20.2%, 9.3%, 12.7% reduction of weight and peak acceleration in frontal and side collisions. This study enables the forward design of multi-material bodies compatible with CFRP parts.
... It is well known that the reliability and efficiency of a manufactured structure depends on geometrical aspects. Therefore, it is not surprising that optimal shape design problems have attracted the interest of many engineers and mathematicians (Haslinger and M€ akinen 2003;Duan et al. 2019). In shape optimization, the design variables are the parameters that describe the structure geometry. ...
... The manufacturing method of VRB is illustrated in Figure 1A as the flexible rolling process. 28 The linear transition 29,30 was adopted in the thickness transition zones (TTZ) between the constant thickness zones (CTZ) in this VRB as shown in Figure 1B. In consideration of different sequences, metals, and combinations, in total 11 kinds of hybrid metal/GFRP laminates were used, as listed in Table 1. ...
The hybrid metal/GFRP laminate is a kind of typical hybrid structure consisting of thin metal sheets and glass fiber reinforced plastic laminates, which combines the benefits of low cost, high strength, excellent impact resistance, and good corrosion resistance. In this work, a semi‐experimental method based on a nondestructive test and the analytic solution of the thin plate's deflection for the flexural and sinking stiffnesses identification of hybrid metal/GFRP laminates was proposed and validated by corresponding numerical simulations. Then, in total 17 different specimens (including metal, GFRP and hybrid metal/GFRP) were prepared and tested by the proposed semi‐experimental method to investigate their flexural and sinking stiffness performances in multiple perspectives (including stiffness per unit mass/volume/price/surface density). The analytic solutions show that the flexural and sinking stiffnesses are proportional to macro equivalent moduli of laminate specimens and proportional to the third power of thicknesses of specimens. The test results show that the stiffnesses per unit mass/density of the 7‐ply GFRP laminate can be comparable to those of the variable‐thickness rolled blank (VRB) made of steels, but the stiffnesses per volume/price of the 7‐ply GFRP laminate are only one fifth to one quarter of those of VRB. It is also found that though the hybrid metal/GFRP laminate with the combination sequence of aluminum plate/5‐ply GFRP/aluminum plate has the lowest GFRP layers' volume fraction, but has the best performances of stiffnesses per unit mass/density. Besides, the VRB part can significantly improve the stiffnesses per unit volume of hybrid metal/GFRP laminates.
... Thin-walled frame structure has been commonly implemented in car body frame, bus frame, battery pack frame et al. Especially, car body is a typical space frame structure assembled by the thin-walled structures of multiple materials (such as aluminum alloy, high-strength steel, etc.) [1,2], whose crashworthiness performance has great effects on the safety of millions of occupants and pedestrians. Therefore, this paper focuses on the material optimization of car body frame. ...
... 3 Therefore, researchers have carried out lots of cross-sectional optimization based on UT body structures, and designed a variable cross-sectional shape (VCS) body structure. As shown in Figures 1 and 2, with maturing of variablethickness rolled blanks (VRB) in recent years, 4 customized VRB sheet with continuously changing thickness can be rolled along the rolling direction by adjusting the rolling space during the flexible rolling process, which can be employed to manufacture VRB car body structures. 5 The thickness distribution of VRB car body structures can be optimized based on the loading conditions, resulting in better material utilization. ...
... Multiple performance evaluation indicators (R 2 ) and Root Mean Square Error (RMSE) are chosen to verify the accuracy of metamodel. 4,40 The closer R 2 is set to be 1.0 and the smaller RMSE values, the higher the accuracy of metamodel is. 41 The RSM constructed according to Table 5 is accurate enough. ...
... Flexible rolling process of VRB.4 ...
After fully understanding the limitations of traditional variable rolled blank (VRB) and variable cross-sectional (VCS) double-hat shaped beam structures on the energy absorption ability and collision safety optimal design, we propose a variable rolled blank and variable cross-sectional (VRB-VCS) double-hat shaped beam structure with piecewise exponential function. Firstly, the thickness distribution function and the mathematical model of constraint conditions are derived under the constraints of manufacturing rollability. The corresponding finite element (FE) model of the VRB-VCS structure is validated by the dropping hammer impact axial crush experiments, which can predict the crushing process with high fidelity. Secondly, the parametric researches of the energy absorption characteristic of VRB-VCS structure are carried out based on the FE model. The reasonable range of each parameter is obtained by analyzing the influence of the thickness distribution and other key cross-sectional parameters on the energy absorption characteristic and deformation mode of VRB-VCS structure. Finally, the Pareto frontier solutions of the linear and piecewise exponential function VRB-VCS structures are obtained by the multi-objective optimization of the corresponding structures. In comparison to VCS, VRB and linear VRB-VCS structures, the proposed VRB-VCS structure has a higher energy absorption capacity and greater crashworthiness.
... Various multi-cell configurations were introduced into the thin-walled tubes to improve the crashworthiness performance under axial and oblique impact loading (Murat Altin 2019; Murat Altin et al. 2019;Acar et al. 2019). Numerous optimization methods were employed to determine the geometric and material parameters of the energy absorbers (Mohammadiha and Ghariblu 2017;Xie et al. 2018;Xiong et al. 2018;Duan et al. 2019;Abdullahi and Gao 2020). ...
A novel structure with enhanced energy absorption is proposed by introducing thin-walled tube as the inner liner tube of the cylindrical structures with negative Poisson’s ratio (C-NPR). The energy absorption performances of C-NPR structure with inner tube (C-NPR-IT) are compared to other configurations like the single NPR structure, single thin-walled tube, and C-NPR structure with outer tube (C-NPR-OT) to show its superiority. It is found that the interaction between the NPR structure and inner tube in C-NPR-IT can be enhanced. Then, the parametric analysis of the geometric parameters on the crashworthiness performance of C-NPR-IT structures are performed with finite element method. To achieve the best configuration, the surrogate modeling technique and the multi-objective particle swarm optimization (MOPSO) algorithm are employed to optimize the C-NPR-IT structures. The results show that the optimized structure improves the specific energy absorption (SEA) from 3.97 to 10.26 kJ/kg by almost 2.5× by controlling peak crushing force (PCF) less than 80 kN. Therefore, the C-NPR-IT structure has an application prospective in the energy absorber.
... A 3 will be used as the boundary and loading condition of the crash simulation at the component level. This could overcome the limitations of the traditional method, which uses roughly estimated or assumed initial velocity [24,25] Based on this formulation and a component model parameterized with geometric design variables, a set of design alternatives is created, and their response is simulated through FEA to form the training set. Learned from this dataset, the rules (B 1− 2 ) can be generated. ...
Keyword: Data-driven mechanical design Decision tree Uncertainty Reliability Vehicle crashworthiness A B S T R A C T In this research, a new data mining-based design approach has been developed for designing complex mechanical systems such as a crashworthy passenger car with uncertainty modeling. The method allows exploring the big crash simulation dataset to design the vehicle at multi-levels in a top-down manner (main energy absorbing system-components-geometric features) and derive design rules based on the whole vehicle body safety requirements to make decisions towards the component and sub-component level design. Full vehicle and component simulation datasets are mined to build decision trees, where the interrelationship among parameters can be revealed and the design rules are derived to produce designs with good performance. This method has been extended by accounting for the uncertainty in the design variables. A new decision tree algorithm for uncertain data (DTUD) is developed to produce the desired designs and evaluate the design performance variations due to the uncertainty in design variables. The framework of this method is implemented by combining the design of experiments (DOEs) and crash finite element analysis (FEA), and then demonstrated by designing a passenger car subject to front impact. The results show that the new methodology could achieve the design objectives efficiently and effectively. By applying the new method, the reliability of the final designs is also increased greatly. This approach has the potential to be applied as a general design methodology for a wide range of complex structures and mechanical systems.
... On the other hand, it is well known that the efficiency and reliability of manufactured structures depend on geometrical aspects. Therefore, it is not surprising that optimal shape design problems have attracted the interest of many engineers and mathematicians, Haslinger and Makinen [19] and Duan et al. [9]. In shape optimization, the design variables are the parameters that describe the structure geometry. ...
Nonuniform microbeams made of functionally graded materials (FGMs) have been studied extensively in literature to predict their mechanical and thermal behavior, those demonstrated that each of material variation, non-uniformity and micro-scale effects have significant influences on the static stability sand dynamic behavior. Therefore, this research exploited the multi-objective shape optimization method to optimize the beam shape and its volume fraction distribution in order to maximize the critical buckling loads and fundamental frequencies while minimizing the mass and cost of the FG microbeam, for the first time. Modified continuum model based on both Euler-Bernoulli beam theory as kinematic assumptions and constitutive equation of modified couple stress theory, is developed to derive equilibrium equations (in static analysis) and equations of motion (in dynamic analysis) of axially FGMs nonuniform microbeam. To control the variation of height and width along the beam length, three different shape functions are proposed in the analysis. The multiobjective particle swarm optimization (MOPSO) is adopted to get the Pareto optimal solutions. In addition to the FGM power index, the shape functions types and parameters are considered as the design variables. Several optimization problems are studied to demonstrate the multi-objective optimal shape design of axially functionally graded microbeams.
