Determination of residual stresses and material properties by an energy-based method using artificial neural networks

Jin, Haotian; Yang, Wenchao; Yan, Lin

doi:10.3176/proc.2012.4.04

Cited by 8 publications

(5 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The determination of these parameters are often used to assess the mechanical reliability of materials (e.g. fatigue, fracture, corrosion and wear) [5], [11], [12]. The nanoindentation test is an extremely small-scale test carried out with nanometer order displacements, so the indenter tip is difficult to position exactly over the material surface at the beginning of the experiment.…”

Section: Case Studymentioning

confidence: 99%

“…A classic procedure is to adopt a regression using the power law as suggested by Oliver and Pharr [5], [11]. The equation of the experimental response, including the calibration, is restated as…”

Section: Case Studymentioning

confidence: 99%

“…On the other hand, robust methodologies have been developed for calculating metal-mechanical properties [2]. However, very few attempts, such in [5], focused on computationally efficient methods. Furthermore, most of the existing literature concentrates in the physical interpretation of the results, and little has been done on approaches based on DACE.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Method for the Identification of Mechanical Properties Using Surrogate Models

Gutierrez¹,

Li²,

Toda³

et al. 2014

IJCTE

View full text Add to dashboard Cite

Abstract-Identification of material properties involves physical experimentation followed by modeling, simulation and manual optimization. However, the last step tends to be computational expensive. This paper investigates an artificial neural network (ANN) surrogate model for identifying material parameters. The proposed approach is illustrated with a case study based on a nano-indentation test. IndexTerms-Surrogate models, optimization, metal-mechanic properties, infill sampling, inverse analysis.

show abstract

Section: Case Studymentioning

confidence: 99%

Section: Case Studymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Method for the Identification of Mechanical Properties Using Surrogate Models

Gutierrez¹,

Li²,

Toda³

et al. 2014

IJCTE

View full text Add to dashboard Cite

show abstract

“…-остаточных напряжений в металле [20]; -начального значения предела текучести [21]. Механические характеристики достоверно описываются трехпараметрическим законом распределения Вейбулла 1 : -сжато-изгибаемом (верхние пояса ферм покрытия); -растянуто-изгибаемом (нижние пояса ферм покрытия); -растянутом (растяжки).…”

unclassified

Variation Coefficient of Metal Yield Strength in New and Long-Used Building Structures

Vernezi

2023

Safety of Technogenic and Natural Systems

View full text Add to dashboard Cite

Introduction. Non-destructive methods are most often used to assess the condition of the metal structure. Dangerous stress is determined by the value of the yield strength. This approach has weaknesses. This is, firstly, the probabilistic nature of the methodology (the minimum value of the indicator obtained during laboratory tests is entered into the regulatory and technical documentation). Secondly, the limitations on the number of samples should be overcome. Thirdly, the different duration of operation causes a significant difference in the mechanical characteristics of the metal, which to a certain extent complicates the long-term prediction of the condition of the structure. The presented work is designed to solve these problems within the framework of the study of new and long-operated facilities in the Rostov region. The scientific research objective is to analyze fatigue changes and determine possible degradation of the metal.Materials and Methods. The mechanical characteristics of the material under study were reliably described by the Weibull distribution law through the shear parameter (the minimum possible value of the characteristic) and the shape parameter (magnitude dispersion). For scientific research, the indentation method based on a modified Rockwell hardness estimation method was used as part of the work. A conical indenter was embedded in the surface, then the reaction of the metal was analyzed. To implement the method, an analog-to-digital converter and a laptop were used. For correlation analysis, intermediate characteristics were taken: depth, maximum and minimum velocities, maximum and minimum acceleration of cone insertion. A correlation was established with the mechanical characteristics determined by standard tensile and hardness tests of the metal.Results. Objects with zero and long-term operation were studied. The measurements were carried out in a warehouse, production site, stadium, bridge, Palace of Sports and on a power line support. From the group of new and used structures, one was selected for a detailed fixation of the values of yield strength. So, before the start of operation, the condition of three metal trusses of the warehouse was analyzed. It was established that the lowest value of the yield strength here was 240 MPa, the maximum was 345 MPa. On the power transmission line poles, which have been in operation for 43 years, the lowest recorded value of the yield strength was 235 MPa, the highest was 384 MPa. For each of the six structures, the minimum and average distribution of the metal yield strength values was given, and the coefficients of variation of this indicator were given. The recorded values were summarized in the form of a table. The average values for all new and used designs were calculated. Graphically presented data illustrate the growth of the coefficients of variation of the yield strength with increasing service life.Discussion and Conclusion. A comparative analysis of the obtained values of the yield strength of building structures of approximately the same strength class suggests that the influence of operating time can both increase and decrease the studied indicator. At the same time, long-term operation is a factor that increases the average value of the coefficient of variation. To monitor the strength capabilities of the structure, it is advisable to use a non-destructive method, selectively monitoring the mechanical characteristics of the elements before and during operation.

show abstract

“…In the field of material science and engineering, Jin et al [16] used surrogate models combined with sharp indentation, dimensional analysis, and an energy method for calculating residual stress. The surrogate model was constructed as an Artificial Neural Network (ANN) that was trained with 240 finite element simulations, which were validated with other 40 simulations chosen randomly.…”

Section: Introductionmentioning

confidence: 99%

Identification of material properties using nanoindentation and surrogate modeling

Gutierrez²,

Toda

et al. 2016

International Journal of Solids and Structures

View full text Add to dashboard Cite

In theory, identification of material properties of microscopic materials, such as thin film or single crystal, could be carried out with physical experimentation followed by simulation and optimization to fit the simulation result to the experimental data. However, the optimization with a number of finite element simulations tends to be computationally expensive. This paper proposes an identification methodology based on nanoindentation that aims at achieving a small number of finite element simulations. The methodology is based on the construction of a surrogate model using artificial neural-networks. A sampling scheme is proposed to improve the quality of the surrogate model. In addition, the differential evolution algorithm is applied to identify the material parameters that match the surrogate model with the experimental data.The proposed methodology is demonstrated with the nanoindentation of an aluminum matrix in a die cast aluminum alloy. The result indicates that the methodology has good computational efficiency and accuracy.

show abstract

Determination of residual stresses and material properties by an energy-based method using artificial neural networks

Cited by 8 publications

References 26 publications

A Method for the Identification of Mechanical Properties Using Surrogate Models

A Method for the Identification of Mechanical Properties Using Surrogate Models

Variation Coefficient of Metal Yield Strength in New and Long-Used Building Structures

Identification of material properties using nanoindentation and surrogate modeling

Contact Info

Product

Resources

About