Neuro-fuzzy modeling of rotation capacity of wide flange beams

https://doi.org/10.1016/j.eswa.2010.10.070Get rights and content

Abstract

This study is a pioneer work that investigates the feasibility of neuro-fuzzy (NF) approach for the modeling of rotation capacity of wide flange beams. The database for the NF modeling is based on experimental studies from literature. The results of the NF model are compared with numerical results obtained by a specialized computer programme and existing analytical and genetic programming based equations. The results indicate that the proposed NF model performs better. By using the proposed NF model, a wide range of parametric studies are also performed to evaluate the main effects of each variable on rotation capacity.

Research highlights

► Neuro-fuzzy (NF) approach for the modelling of rotation capacity of wide flange beams. ► Based on experimental studies from literature. ► Comparison is made with existing analytical and genetic programming based equations. ► Proposed NF model performs better. ► Parametric studies are also performed to evaluate the main effects of each variable.

Introduction

The evaluation of rotation capacity of steel beams has been the subject of numerous numerical and experimental studies as it is an extremely important phenomenon for plastic and seismic analysis and design of steel structures. Rotation capacity is used to express whether plastically designed sections possess the required ductility or not. In the same way, the moment redistribution in a steel structure also depends on the rotation capacity of the section (Dinno, 2002).

Although many studies have been conducted on this topic, there is still lack of reliable analytical models to describe the concept of rotation capacity of steel beams in all respects. Theoretical, empirical and approximate methods have been proposed so far for the determination of available rotation capacity of wide flange steel beams in literature which have been reported by Gioncu and Petcu, 1997a, Gioncu and Petcu, 1997b. Besides, a relatively new approach is also growing rapidly as an alternative tool to handle the rotation capacity of steel beams: soft computing techniques. Guzelbey, Cevik, and Gogus (2006) have proposed neural networks (NN) as an alternative approach for the prediction of rotation capacity of wide flange beams based on experimental results collected from literature. The proposed NN model showed perfect agreement with experimental results (R = 0.997) where its accuracy was also quite high. Moreover, Guzelbey et al. (2006) have also presented the proposed NN model in explicit form as a mathematical function. Furthermore Cevik proposed an alternative genetic programming (GP) based formulation which performed quite well compared to existing analytical equations (Cevik, 2007).

This study investigates the feasibility of another soft computing technique namely as neuro-fuzzy (NF) approach for the modeling of available rotation capacity of wide flange steel beams for the first time in this field. Similar to previous soft computing models, the proposed NF model is also based on experimental results collected from literature. The results of the proposed NF model are compared with numerical results and existing analytical equations and found to be more accurate. Main effects of each variable on rotation capacity are also obtained by using the proposed NF model.

Section snippets

Definition of rotation capacity

There are various definitions of rotation capacity in literature as a non-dimensional parameter.

According to Lay and Galambos (1965) rotation capacity is, R = θh/θp, in which θp is the elastic rotation at the initial attainment of the plastic moment Mp and θh is the plastic rotation at the point when moment drops below Mp.

A widely used definition for rotation capacity is proposed by ASCE (Fig. 1):

R = θ2/θ1 where θ1 refers to the theoretical rotation at which the full plastic capacity is achieved

Fuzzy logic

Over the last decade, fuzzy logic invented by Zadeh (1965) has been applied to a wide range of applications covering engineering, process control, image processing, pattern recognition and classification, management, economics and decision making (Rutkowski, 2004).

Fuzzy systems can be defined as rule-based systems that are constructed from a collection of linguistic rules which can represent any system with accuracy, i.e., they work as universal approximators. The rule-based system of fuzzy

Numerical application

The main aim of this article is the NF modeling of rotation capacity of wide flange steel based on experimental results from literature. Therefore an extensive literature survey has been performed for experimental results shown in Table 2. The experimental results in this field are dispersed. Standard beams are used in experimental studies (SB1, SB2) shown in Fig. 8, Fig. 9. SB1 is used in the experimental studies given in Table 2. The geometry of cross-section variables of tested beams is shown

Main effects of variables on rotation capacity

The Main Effect plot is an important graphical tool to visualize the independent impact of each variable on rotation capacity. This graphical tool enables a better and simple picture of the overall importance of variable effects on the output which is the rotation capacity for the case and will provide a general snapshot. In main effects plot, the mean output (rotation capacity) is plotted at each factor level which is later connected by a straight line. The slope of the line for each variable

Conclusion

This paper presents a pioneer work for the modeling of available rotation capacity of wide flange beams using Neuro-fuzzy approach for the first time in literature. The proposed NF model is a rule-based model based on experimental results collected from literature. To show its effectiveness, the simplest possible NF model is constructed. The results of the proposed NF model show very good agreement with experimental results (COV = 0.15). Numerical results of the same experimental database are

Acknowledgement

This research was supported by Gaziantep University Project Research Unit.

References (27)

  • P. Boreave et al.

    Elastoplastic behaviour of steel frameworks

    Journal of Constructional Steel Research

    (1993)
  • Dinno, D. K. (2002). Strength and ductility of steel members based on post-yield stability and collapse behaviour....
  • Eurocode 3: Design of steel structures – Part 1-1: General rules and rules for buildings. prEN 1993-1-1, Brussels,...
  • Cited by (8)

    • Rotational capacity and optimum bracing point of high strength steel I-girders

      2013, Journal of Constructional Steel Research
      Citation Excerpt :

      Further, for efficient estimation of the rational capacity of an I-girder, Guzelbey et al. [11] proposed the prediction model of the available rotation capacity of wide flange beams by using neural networks theory. Similarly, genetic programming and neuro-fuzzy modeling can be used for accurate prediction of the rotational capacity of an I-girder [12,13]. However, these researches should be carefully examined for application to I-girders with high strength steel since the rotational capacity mainly results from the material properties of the steel, and the yield stress of the steel has recently become dramatically increased.

    • Soft Computing Applied to Rotation Capacity of Wide Flange Beams

      2018, Iranian Journal of Science and Technology - Transactions of Civil Engineering
    • Support vector machines in structural engineering: A review

      2015, Journal of Civil Engineering and Management
    View all citing articles on Scopus
    View full text