Search results

Meta-model method has been widely used in structural reliability optimization design. The main limitation of this method is that it is difficult to quantify the error caused by the meta-model approximation, which leads to the inaccuracy of the optimization results of the reliability evaluation. Taking the local high efficiency of the proxy model, this paper aims to propose a local effective constrained response surface method (LEC-RSM) based on a meta-model.

The operating mechanisms of LEC-RSM is to calculate the index of the local relative importance based on numerical theory and capture the most effective area in the entire design space, as well as selecting important analysis domains for sample changes. To improve the efficiency of the algorithm, the constrained efficient set algorithm (ESA) is introduced, in which the sample point validity is identified based on the reliability information obtained in the previous cycle and then the boundary sampling points that violate the constraint conditions are ignored or eliminated.

The computational power of the proposed method is demonstrated by solving two mathematical problems and the actual engineering optimization problem of a car collision. LEC-RSM makes it easier to achieve the optimal performance, less feature evaluation and fewer algorithm iterations.

This paper proposes a new RSM technology based on proxy model to complete the reliability design. The originality of this paper is to increase the sampling points by identifying the local importance of the analysis domain and introduce the constrained ESA to improve the efficiency of the algorithm.

Because of the high computational efficiency, response surface method (RSM) has been widely used in structural reliability analysis. However, for a highly nonlinear limit state function (LSF), the approximate accuracy of the failure probability mainly depends on the design point, and the result is that the response surface function composed of initial experimental points rarely fits the LSF exactly. The inaccurate design points usually cause some errors in the traditional RSM. The purpose of this paper is to present a hybrid method combining adaptive moving experimental points strategy and RSM, describing a new response surface using downhill simplex algorithm (DSA-RSM).

In DSA-RSM, the operation mechanism principle of the basic DSA, in which local descending vectors are automatically generated, was studied. Then, the search strategy of the basic DSA was changed and the RSM approximate model was reconstructed by combining the direct search advantage of DSA with the reliability mechanism of response surface analysis.

The computational power of the proposed method is demonstrated by solving four structural reliability problems, including the actual engineering problem of a car collision. Compared to specific structural reliability analysis methods, the approach of modified DSA interpolation response surface for structural reliability has a good convergent capability and computational accuracy.

This paper proposes a new RSM technology based on proxy model to complete the reliability analysis. The originality of this paper is to present an improved RSM that adjusts the position of the experimental points judiciously by using the DSA principle to make the fitted response surface closer to the actual limit state surface.

The purpose of this paper is to propose physically based varying fidelity surrogates to be used in structural design optimization of space trusses. The main aim is to demonstrate its efficiency in reducing the number of high fidelity (HF) runs in the optimization process.

In this work, surrogate models are built for space truss structures. This study uses functional as well as physical surrogates. In the latter, a grid analogy of the space truss is used thereby reducing drastically the analysis cost. Global and local approaches are considered. The latter will require a globalization scheme (sequential approximate optimization (SAO)) to ensure convergence.

Physically based surrogates were proposed. Classical techniques, namely Taylor series and kriging, are also implemented for comparison purposes. A parameter study in kriging is necessary to select the best kriging model to be used as surrogate. A test case was considered for optimization and several surrogates were built. The CPU time is reduced when compared with the HF solution, for all surrogate‐based optimization performed. The best result was achieved combining the proposed physical model with additive corrections in a SAO strategy in which C1 continuity was imposed at each trust region center. Some guidance for other engineering applications was given.

This is the first time that physical‐based surrogates for optimum design of space truss systems are used in the SAO framework. Physical surrogates typically exhibit better generalization properties than other surrogates forms, produce faster solutions, and do not suffer from dimensionality curse when used in approximate optimization strategies.

Optimization of structural systems under reliability‐based performance constraints is an important problem at present receiving too little attention. This problem is investigated in this paper. In developing the reliability‐based optimization approach to the design of framed structures, we review first the general formulation of the deterministic optimization problem and present some of the main features of two general‐purpose deterministic optimization programs. A computer‐automated reliability‐based optimum design procedure is then presented by which the concept of reliability analysis with regard to both serviceability and ultimate performance constraints is combined with that of the minimum weight design to achieve an optimum trade‐off between the global reliability and the total cost. The procedure is feasible for application in system optimization of both steel and reinforced concrete structures.

The aim of this paper is to present a novel multifactor structural optimisation method incorporating reliability performance.

This research addresses structural optimisation problems in which the design is required to satisfy multiple performance criteria, such as strength, stiffness, mass and reliability under multiple loading cases simultaneously. A MOST technique is extended to accommodate the reliability‐related optimisation. Structural responses and geometrical sensitivities are analysed by a FE method, and reliability performance is calculated by a reliability loading‐case index (RLI). The evaluation indices of performances and loading cases are formulated, and an overall performance index is presented to quantitatively evaluate a design.

The proposed method is applicable to multi‐objective, multi‐loading‐case, multi‐disciplinary and reliability‐related optimisation problems. The applications to a star‐like truss structure and a raised‐access floor panel structure confirmed that the method is highly effective and efficient in terms of structural optimisation.

A systematic method is proposed. The optimisation method combines the MOST technique with a RLI (a new alternative route to calculate the reliability index at multiple loading cases) using a parametric FE model.

Considering the randomness of physical parameters of structural material, dynamic characteristic topology optimization mathematical model based on reliability of planar continuum structures is built in this paper. In which topology information variables of the structure are taken as design variables, minimizing the mean value of total structural weight as objective function and satisfying the reliability requirement of structural dynamic characteristic as constraints. In the process of optimization, the ESO method based on probability is adopted as solution strategy. At the same time, distribution function method is utilized to convert the reliability constraints into conventional constraints formally. A square thin plate with four sides fixed is used as an example to demonstrate the rationality and validity of the presented model.

The paper attempts to establish the connection between structural reliability and structural optimization for the particular case of plastic structures. Along this line, the paper outlines a reliability‐based optimization approach to design plastic structures with uncertain interdependent strengths and acted on by random interdependent loads. The importance of such interdependencies, and of some of the other statistical parameters used as input data in probabilistic computations, is demonstrated by several examples of sensitivity studies on both the probability of collapse failure as well as the reliability‐based optimum solution.

Structural optimisation and reliability theory are considered, and described. A general reliability‐based structural optimisation problem is formulated, and consideration given to procedures for solving it. Two different examples suggest the efficacy of these procedures. The amount of calculations depends to a great degree on the definition of failure of the structure. In order to reduce this by improving optimisation procedures, more research is needed, and the convergence of the optimisation is very dependent on accurate evaluation of the gradients of the reliability constraints.

The purpose of this paper is to propose a novel non-probabilistic reliability-based topology optimization (NRBTO) method for continuum structural design under interval uncertainties of load and material parameters based on the technology of 3D printing or additive manufacturing.

First, the uncertainty quantification analysis is accomplished by interval Taylor extension to determine boundary rules of concerned displacement responses. Based on the interval interference theory, a novel reliability index, named as the optimization feature distance, is then introduced to construct non-probabilistic reliability constraints. To circumvent convergence difficulties in solving large-scale variable optimization problems, the gradient-based method of moving asymptotes is also used, in which the sensitivity expressions of the present reliability measurements with respect to design variables are deduced by combination of the adjoint vector scheme and interval mathematics.

The main findings of this paper should lie in that new non-probabilistic reliability index, i.e. the optimization feature distance which is defined and further incorporated in continuum topology optimization issues. Besides, a novel concurrent design strategy under consideration of macro-micro integration is presented by using the developed RBTO methodology.

Uncertainty propagation analysis based on the interval Taylor extension method is conducted. Novel reliability index of the optimization feature distance is defined. Expressions of the adjoint vectors between interval bounds of displacement responses and the relative density are deduced. New NRBTO method subjected to continuum structures is developed and further solved by MMA algorithms.

Mechanical products usually require deterministic finite element analysis in the design phase to determine whether their structures meet the requirements. However, deterministic design ignores the influence of uncertainties in the design and manufacturing process of mechanical products, leading to the problem of a lack of design safety or excessive redundancy in the design. In order to improve the accuracy and rationality of the design results, a robust design method for structural reliability based on an active-learning marine predator algorithm (MPA)–backpropagation (BP) neural network is proposed.

The MPA was used to obtain the optimal weights and thresholds of a BP neural network, and an active-learning function applicable to neural networks was proposed to efficiently improve the prediction performance of the BP neural network. On this basis, a robust optimization design method for mechanical product reliability based on the active-learning MPA-BP model was proposed. Random moving quadrilateral sampling was used to obtain the sample points required for training and testing of the neural network, and the reliability sensitivity corresponding to each sample point was calculated by subset simulated significant sampling (SSIS). The total mass of the mechanical product and the structural reliability sensitivity of the trained active-learning MPA-BP model output were taken as the optimization objectives, and a multi-objective reliability-robust optimization design model was constructed, which was solved by the second-generation non-dominated ranking genetic algorithm (NSGA-II). Then, the dominance function was used in the obtained Pareto solution set to make a dominance-seeking decision to obtain the final reliability-robust optimization design solution. The feasibility of the proposed method was verified by a reliability-robust optimization design example of the bogie frame.

The prediction error of the active-learning MPA-BP neural network was smaller than those of the particle swarm optimization (PSO)-BP, marine predator algorithm (MPA)-BP and genetic algorithm (GA)-BP neural networks under the same basic parameter settings of the algorithm, which indicated that the improvement strategy proposed in this paper improved the prediction accuracy of the BP neural network. To ensure the reliability of the bogie frame, the reliability sensitivity and total mass of the bogie frame were reduced, which not only realized the lightweight design of the bogie frame, but also improved the reliability and robustness of the bogie.

The MPA algorithm with a higher optimization efficiency was introduced to find the weights and thresholds of the BP neural network. A new active-learning function was proposed to improve the prediction accuracy of the MPA-BP neural network.