Papers by Keyword: Response Surface Method (RSM)

Abstract: The response surface method is adopted to analyze the structural reliability. This paper presents a new response surface method with the uniform design method to predict the failure probability of structures. It is the response surface method based on Fourier orthogonal basis function (RSM-Fourier). To reduce computational costs in structural reliability analysis, approximate Fourier response surface functions for reliability assessment have been suggested. The method involves the selection of training datasets for establishing a model by the uniform design points, the approximation of the limit state function by the trained model and the estimation of the failure probability using first-order reliability method (FORM). The proposed method is applied to examples, compared with other methods to demonstrate its effectiveness.

Abstract: Heterogeneous metal sandwich plate is composed of high strength material face-sheets and low-density material core. In this paper, the crashworthiness property of the heterogeneous metal sandwich plate was investigated with combining the response surface method and finite element simulation. The specific absorption energy of sandwich plate is made as the crashworthiness performance index, and approximation written as a third-order function of the section parameters by means of the response surface method. According to the BB experiment design method, 29 groups of sandwich plate with different section structure parameters were selected and finite element numerical simulations were conducted to study on deformation behavior of sandwich plates during the collision process. Then, the coefficients of approximate specific absorption energy function were solved with regressive analysis. Consequently, the relationship between the crashworthiness performances of sandwich plate and the section parameters α, β, D, d was obtained.

Abstract: Optimization of culture conditions for lipopeptide production of lipopeptide by Bacillus subtilis NEL-01 was carried out in shaker flask batch fermentations using composite central design of response surface methodology (RSM). A five-level three-factor central composite design was employed to determine the maximum lipopeptide production at optimum levels for culture temperature, initial pH and culture cycle. Culture temperature and culture cycle showed the significant linear main effects, while pH had no significant linear effect. The production was also significantly affected by quadratic effect of culture temperature and initial pH. Optimum fermentation parameters were predicted at temperature, initial pH, and culture cycle of 34.81 °C, 7.33g/l, 49.26 h, respectively. The prediction lipopeptide yield was 1879.56 mg/l. The subsequent experiments confirmed the prediction.

Abstract: Based on Mindlin first order shear effect plate theory, the structural-acoustic optimization of laminated composite structures under external loading was investigated. For improving the optimization efficiency, the response surfaces method (RSM) is introduced, and the uniform Latin square method was used to select the most appropriate sample points. In the end, taking the laminated composite plate acoustic radiation as an example, the mathematical model of structural structural-acoustic optimization is established. The results show that the acoustic radiation of laminated composite structures can be reduced by optimizing the stacking sequence parameters such as layers thickness and layers angle. The results prove the validity of the calculating ways.

Abstract: Using the static and dynamic test data simultaneously to update the finite element model can increase the available information for updating. It can overcome the disadvantages of updating based on static or dynamic test data only. In this paper, the response surface method is adopted to update the finite element model of the structure based on the static and dynamic test. Using the reasonable experiment design and regression techniques, a response surface model is formulated to approximate the relationships between the parameters and response values instead of the initial finite element model for further updating. First, a numerical example of a reinforced concrete simply supported beam is used to demonstrate the feasibility of this approach. Then, this approach is applied to update the finite element model of a prestressed reinforced concrete rigid frame-continuous girders bridge based on in-situ static and dynamic test data. Results show that this approach works well and achieve reasonable physical explanations for the updated parameters. The results from the updated model are in good agreement with the results from the in-situ measurement. The updated finite element model can accurately represent mechanical properties of the bridge and it can serve as a benchmark model for further damage detection and condition assessment of the bridge.

Abstract: Thin-walled structure absorbs most impact kinetic energy during collision accident,and they are widely used as energy-absorbing element. In order to improve crashworthiness of them, regular pyramidal ripple is added on the thin-walled square tube’s surface. Explicit finite element technology is applied to simulate the behavior of the tube under axial impact load. Simulation data was delt with by Response Surface Method to form a function of variables and response,and the new structure was optimized. Research results show that, the thin-walled square tube with pyramidal ripples can improve controllable of structure deformation obviously and Optimized structure can absorb and dissipate much more impact kinetic energy.

Abstract: A method is established to determine eight kinds of residual solvents in paper package by HS-GC/MS, such as benzene, toluene etc. Response surface method is used to optimize pretreatment temperature and time, the terminal pre-temperature is 100 °C, with time 40min. The result reveals that the pretreatment-temperature has a significant impact on peak area and pretreatment-time has no significant influence. The residual solvents in sample was separated using HP-5MS chromatographic column; Flow velocity of He:1ml/min; Inlet temperature:250°C; programmed temperature: keep 2 min at initial temperature of 35 °C, then rise to 100 °C at a rate of 10 °C/min and keep 3min; the total time is 14.5min. The eight kinds of residual solvents can be effectively separated by this method. The linear range was 0.330~7.956mg/m2 and the correlation coefficient was between 0.995~0.999.The Recovery rate was 99%~114.8% and the relative standard deviation (n=5) was 0.916%~1.333%.

Abstract: HHT is widely used to analyze nonlinear and non-stationary signals. But how to extend boundaries of signals in decomposition processes is a key problem of HHT. A new technique based on response surface method (RSM), which establishes the recursive relations between sample points of signals, is presented to deal with this difficult problem. Besides, the boundary extension problem arising from HHT can be described by mathematical least squares problem but traditional gradient algorithms may diverge when the Hessian matrix of the object function of the least squares problem is non-positive. It has been proved that the generalized inverse of the linear equations (derived from the linear least squares problem) by singular value decomposition is the solution of original linear least squares problems. Thereby the divergence problem is also solved. Analysis results with respect to simulation signals and measured signals show that the method with new boundary extension technique performs successfully for HHT.

Abstract: The establishment of an effective finite element model for bridge structure is essential in the health monitoring system for Bridge. A new updating method for static model using response surface method is proposed in this paper, and the main procedures are given with an example of a special-shaped bridge. Firstly, the bridge deflection and strain data in designed load case are obtained. Several groups of combined parameters which are chosen based on the principle of uniform design method are selected to conduct calculation through finite element software. Finally through response surface fitting and optimization, the updated bridge finite element model is obtained. The results show that the updated model is approximate to the real bridge and this updating method is rational and practical.

Abstract: Due to the massive volume of body and complex structure, the dynamic reliability analysis on arch dam under earthquake action draws more attention. In this paper calculated seismic dynamic action on arch dam by applying mode superposition response spectrum method. By taking the elastic modulus of the dam concrete and the rock foundation beneath the dam, and concrete tensile and compressive strength as random variables, adapting response surface method to establish limit state equation. The reliability index of each element in the dam was calculated via gradient optimization method, as a result to obtain the variation law of reliability index of dam body. The calculation of real case shows that the tensile index is comparatively low at some region of dam heel and abutment upstream, which requires the measure for aseismic measures, while the reliability index for tensile and compressive stress on other part of dam generally meets requirement.