Fig 8 - uploaded by Yasser Sharifi
Content may be subject to copyright.
The Error Given by the Validation Set Increases from Some Point Outwards. At that Point The Network is Optimally Fitted; Beyond is Over-fitted, Before is Under-fitted (Hristev, 1998)
Source publication
Half-through girders are not affected by conventional lateral-torsional buckling. I-section beams of simply supported half-through girders experience compression in their top flanges and tension in their bottom flanges. In this condition, the compression flange is restrained only by the stiffness of the web, and the buckling mode is generally restr...
Similar publications
![Figure 1: Muwagama Steel Bridge [1]](profile/Hasitha-Hidallana-Gamage/publication/305045171/figure/fig1/AS:381645367595008@1468002754823/Muwagama-Steel-Bridge-1_Q320.jpg)
This paper investigates the failure of the Meeliyadda Bridge, which occurred, when two trucks carrying metal were crossing the bridge. It was reported that some of the lateral bracings of the top chord had removed with the failure of the bridge. Finite element (FE) analysis was carried out using the SAP2000 FE software, where two different FE model...
Citations
... The authors considered a total of 396 samples. Tohidi and Sharifi [56] used a neural network model to predict the restrained distortional buckling strength of half-through I-section bridge girders. They developed a FEM model to build a dataset of 396 samples used to train and develop their model. ...
Most design standards require the calculation of the elastic critical bending moment for the design and verification of steel beams. Formulae exists for uniform beams with double or mono-symmetric cross-sections but, for tapered beams, simple design formulae are yet to be developed because of the complexity associated with the non-uniform geometry of these members.
This work proposes a neural network model to calculate the critical moment. The model is developed using the Backpropagation and Levenberg–Marquardt algorithms and considering 60549 data samples for training and 8526 samples for validation. The samples are generated by a numerical model using the Finite Element Method (FEM). An innovative methodology to reduce the number of samples necessary to train the model is implemented based on the concept of random sample generation with constrained geometric proportions.
The accuracy of the developed model is further illustrated on some particular cases and against the FEM results of other authors. For the uniform beams, the results of the proposed model are compared against those from existing formulae for uniform members showing its improved accuracy.
Finally, it is expected that this investigation demonstrates the benefits of the use of neural networks based solutions as fast assessment tools during the search of optimal structural solutions in the early design stages.
... This algorithm is based on the Newton method and is very suitable for training the networks with up to a few hundred weights (Marquardt 1963;Hagan and Menhaj 1994). In previous studies (Tohidi and Sharifi 2016; Sharifi et al. 2019a;Sharifi et al. 2019b;Sharifi et al. 2020), its effectiveness in solving various problems has been proven. The LM algorithm is very efficient compared to other methods and in many cases when other back-propagation algorithms diverge, this algorithm converges. ...
... Three-layer ANN model with one hidden layer was considered for the prediction of ultimate strength of plates with a circular hole in the centre. Based on the studies that have been done so far, this type of network has a good predictive ability to solve common engineering problems (Wong and Kim 2018;Tohidi and Sharifi 2016;Hosseinpour et al. 2018;Koçer et al. 2019;Hosseinpour et al. 2020). ...
In the current paper, using finite element models (FEM), an extensive numerical study is performed on the behaviour of the steel plates with a circular hole in their centre subjected to compressive axial loading. For this purpose, 270 perforated steel plates were modelled and analysed using ABAQUS software. The effects of four main variables including plate length, hole diameter, plate thickness, and yield stress were discussed. Then, using the database provided by FEM, the artificial neural network (ANN) method was used to develop a predictive model to estimate the ultimate strength of steel plates with a circular hole in the centre. Finally, an ANN-based formula was proposed to predict the ultimate strength of perforated steel plates and its accuracy was compared with the formulations presented in previous studies. Based on the results, the proposed formula provided high accuracy and can be used as a reliable formula in practice.
... In addition, there are four possible ANN solutions for different problems, namely auto-association, pattern recognition, classification, and unique detection. There are many ANN applications worldwide in various fields of knowledge that were found in the literature [1,4,[7][8][9][10]. Therefore, in this paper, the ANN model was developed to predict the damage level of the cable-stayed bridge. ...
The current condition of a bridge should be given the highest priority to ensure its safety and its serviceability to the bridge users. Therefore, this paper explained the development of an intelligent decision-support tool in Bridge Health Monitoring system using a Neural Network model as a prediction of seismic damage-level for cable-stayed bridge. A total of eight earthquake loads scaled to various Peak Ground Acceleration (PGA) values. The input and output data which were fed into Feed Forward Artificial Neural Network (ANN) for damage level prediction model were developed based on acceleration responses from the Non-Linear Time History (NLTH) analysis of the cable-stayed bridge. A total of 16620 data were used as the input data. The damage-level categorization is based on FEMA 356. Data used for the ANN training are 70% for training, 15% for validation, and 15% for testing. The damage level prediction can greatly help bridge authority in order to maintain their bridge structure integrity by identifying and predicting the probability of damage occurring under earthquake loads.
... This mode is described with simultaneous occurrence of lateral translation, twisting, and web distortion; and usually occurs in the beams with intermediate length, slender web, and thick flanges [12]. It should be mentioned that generally, distortional buckling has been recognized in two modes comprising: "lateraldistortional buckling" (LDB) and "restrained distortional buckling" (RDB) (see Fig. 1); furthermore, RDB may occur in any situation where the tension flange is restrained against lateral translation and twisting, e.g., in the negative moment regions of composite beams or the half-through girder bridges [13][14][15]. ...
... Therefore, the geometric imperfections were applied to the models by considering the lateral imperfection as a half-sine wave with amplitude L b /1000, and L b denotes the effective unbraced length of the beam. The value of L b /1000 has been recognized as an average value for test measurements in previous researches [12][13][14][15][30][31][32]. Furthermore, it is worth mentioning that the residual stresses arising from the manufacturing process have little effect on the buckling behavior of various structural elements and have often been ignored by researchers [12][13][14][15][30][31][32]. ...
... The value of L b /1000 has been recognized as an average value for test measurements in previous researches [12][13][14][15][30][31][32]. Furthermore, it is worth mentioning that the residual stresses arising from the manufacturing process have little effect on the buckling behavior of various structural elements and have often been ignored by researchers [12][13][14][15][30][31][32]. ...
The effects of web distortion on the buckling behavior of castellated steel beams have rarely been studied; and so far, no exact formula has been proposed to predict the lateral-distortional buckling (LDB) capacity of such beams. Therefore, in the following paper, an attempt has been made to propose new formulas for predicting the LDB capacity of castellated beams using two regression models. To do this, at the first, an extensive parametric study including 480 finite element (FE) models was performed. Both material nonlinearities and initial geometric imperfections were carefully applied to the models and they were also well validated against experimental results. The next, based on the parametric study, nine non-dimensional parameters that affect the LDB behavior of castellated beams were considered, and using them as input parameters, two predictive models including stepwise regression (SR) and multiple regression (MR) were developed. The models were then used to provide new formulations for estimating the LDB strength of castellated steel beams. In the end, the proposed formulas were compared with the existing design guides available for estimating the buckling strength in AS4100, EC3, and AISC codes. The results showed that the proposed formulas for use by practical engineers are accurate enough.
... This mode is described with simultaneous occurrence of lateral translation, twisting and web distortion; and usually occurs in the intermediate-length beams with slender web and thick flanges [12]. It should be mentioned that generally distortional buckling has been recognized in two modes including: "lateral-distortional buckling" (LDB) and "restrained distortional buckling" (RDB) (See Fig. 1); furthermore, RDB, may occur in any situation where the tension flange is restrained against lateral translation and twisting, e.g. in the negative moment regions of composite beams or in the half-through girder bridges [13][14][15]. ...
... In this paper, the geometric imperfection were applied to the models by considering a half-since wave lateral imperfection with amplitude L b /1000, where L b is the length between the points of effective bracing. In general, the magnitude of L b /1000 has been recognized as an average value for test measurements in previous researches [12][13][14][15][30][31][32], this specific value for initial imperfection is also stated in AS4100 as a fabrication tolerance for flexural elements. ...
... Furthermore, it is worth mentioning that based on the previous literatures, the residual stresses due to the manufacturing process had small influences on the buckling behavior of different structural elements and researchers had often ignored this in their researches [12][13][14][15][32][33][34][35][36]; as a result, residual stresses are not comprised in this study, however, it is relevant in this type of analysis. ...
This paper examined the effects of web distortion on the buckling behavior of castellated steel beams. To this purpose, a series of nonlinear finite element (FE) models was constructed and well verified against an experimental work on the distortional buckling of castellated beams; both material nonlinearities and initial geometric imperfections were carefully applied to the models. Next, an extensive parametric study was performed using the finite element models to investigate the effects of beam length, steel grade and cross-section dimensions on the ultimate buckling load and buckling modes of castellated steel beams. The results showed that the use of low grade steel and thick flanges makes an economical design in the castellated steel beams. Moreover, it is concluded that lateral-distortional buckling (LDB) mode is more common in the castellated beams with intermediated overall slenderness, thick flanges and slender web. Finally, the ultimate loads obtained from finite element analysis (FEA) were compared with the results predicted by AS4100, EC3 and AISC codes. It was concluded that all three Specifications provide unsafe estimates for most specimens in this study.
... Time history adalah Percepatan,pemindahan serta Kecepatan riwayat waktu terhadap jembatan yang mengalami getaran gempa (BMS, 1993), (Lydon , 2016). Tingkat kinerja untuk analisis dinamik nonlinear (Tohidi, 2016) (Suryanita, 2015), (Darmawan, 2017) . beban gempa yang digunakan berasal dari gempa Loma Prieta yang dikonversikan terhadap beban gempa respons spektrum Pekanbaru. ...
The behavior of the arch bridge structure has its own peculiarities in distributing loading. The curved shape of the structure allows the structure's own weight to be transferred to the foundation as the normal compressive force without bending. the load is transferred through the semicircular structure to the abutments on either side. The curved (semicircular) design will naturally shift the load received by the floor of the bridge vehicle towards the abutment which keeps both sides of the bridge from moving sideways. Earthquake is a dominant horizontal cyclic loading of a structure. The performance of bridge structures due to the effects of earthquake loading is an important aspect that needs to be taken into account in designing and evaluating the bridge structure. The method used in this study is time history analysis. This article aims to evaluate the performance of arch bridge structures due to earthquake loads by using time-history analysis method. The results of the structure performance show that. The performance of the bridge due to the loading of the Loma prieta earthquake is generated from the period scale that has been analyzed with the 10 second period of the first earthquake for each result. Based on the results of Midas analysis, the strongest location where the damage occurs is in the middle of the bridge road. At the coordinates of 19 the largest displacement is at joint 268 of 1.177169 mm and for U2 of 0.0194 mm and U3 of 31.763771 mm.
... The achievements of these studies using ANN include a rapid estimation of the seismic demand, probability of failure and the reliability analysis of large concrete dams, computational efficient stability assessment, accurate predictions of the structural response at a fraction of computational time, seismic damage prediction of reinforced concrete respectively. The accuracy and efficiency of the ANN has been already proved in the previous studies (Pang et al., 2014;Tohidi and Sharifi, 2016) which motivates researchers to use this method for assessment of prediction of bridge component responses and seismic fragility of HCSIG bridges. Among various types of soft computing methods, ANN are chosen due to the capture of adequate nonlinearity behavior and to reduce the calculations needed for the prediction of responses. ...
This study presents a new hybrid method to develop seismic fragility curves for horizontally-curved steel I-girder bridges using Artificial neural network and logistic regression methods. The approach for developing fragility curves based on the assumption that engineering demand parameters follow the lognormal distribution for calculating the probability of damage occurrence. A sufficient number of input data including a set of earthquake ground motion records and macro-structural parameters together with the output data resulting from nonlinear structural analyses was assigned to neural network structure to achieve satisfactory approximations of responses. Logistic regression statistical method was used to determine the probability of occurrence or non-occurrence of limit states for earthquake ground motion parameters and structural characteristics. In this study, based on the estimation of engineering demand parameters, the proposed method is compared with the neural network method, simplified mathematical model and analytical method. The nonlinear time history analysis of three dimensional horizontally curve bridges were performed using the OpenSEES software. The statistical results indicate the accuracy and efficiency of the predicted limit state occurrence of the proposed method at a low computational cost. Comparison of fragility curves using the mentioned methods represent a proper estimation for slight, moderate, extensive and collapse limit states at different levels of seismic intensity.
... The achievements of these studies using ANN include a rapid estimation of the seismic demand, probability of failure and the reliability analysis of large concrete dams, computational efficient stability assessment, accurate predictions of the structural response at a fraction of computational time, seismic damage prediction of reinforced concrete respectively. The accuracy and efficiency of the ANN has been already proved in the previous studies (Pang et al., 2014;Tohidi and Sharifi, 2016) which motivates researchers to use this method for assessment of prediction of bridge component responses and seismic fragility of HCSIG bridges. Among various types of soft computing methods, ANN are chosen due to the capture of adequate nonlinearity behavior and to reduce the calculations needed for the prediction of responses. ...
This study presents a new hybrid method to develop seismic fragility curves for horizontally-curved steel I-girder bridges using Artificial neural network and logistic regression methods. The approach for developing fragility curves based on the assumption that engineering demand parameters follow the lognormal distribution for calculating the probability of damage occurrence. A sufficient number of input data including a set of earthquake ground motion records and macro-structural parameters together with the output data resulting from nonlinear structural analyses was assigned to neural network structure to achieve satisfactory approximations of responses. Logistic regression statistical method was used to determine the probability of occurrence or non-occurrence of limit states for earthquake ground motion parameters and structural characteristics. In this study, based on the estimation of engineering demand parameters, the proposed method is compared with the neural network method, simplified mathematical model and analytical method. The nonlinear time history analysis of three dimensional horizontally curve bridges were performed using the OpenSEES software. The statistical results indicate the accuracy and efficiency of the predicted limit state occurrence of the proposed method at a low computational cost. Comparison of fragility curves using the mentioned methods represent a proper estimation for slight, moderate, extensive and collapse limit states at different levels of seismic intensity.
... Girder= 600x600 cm 10 Tahapan selanjutnya yaitu membuat komponen gelagar, diafragma, dan perletakan jembatan. Setelah itu dilakukan penginputan beban jembatan seperti beban mati, beban lalu lintas dan beban gempa berupa respons spectrum untuk masing masing lokasi [10]. ...
... Girder= 600x600 cm 10 Tahapan selanjutnya yaitu membuat komponen gelagar, diafragma, dan perletakan jembatan. Setelah itu dilakukan penginputan beban jembatan seperti beban mati, beban lalu lintas dan beban gempa berupa respons spectrum untuk masing masing lokasi [10]. Perhitungan pembebanan digunakan untuk menentukan beban yang bekerja di jembatan. ...
[ID] Perencanaan struktur jembatan baja pelengkung harus memperhatikan kemampuan respon strukturnya yang rentan terhadap deteriorasi akibat fatik, ancaman gempa bumi kuat atau angin topan, khususnya diwilayah sumatera yang mempunyai resiko gempa yang tinggi. Penelitian ini fokus memprediksi struktur jembatan pelengkung baja dengan analisis repons spectra dengan bantuan software analisis struktur gempa berdasarkan SNI 1726-2012. Percepatan gempa yang diambil berasal dari beberapa kota seperti Kota Aceh, Padang, Tanjung Pinang, dan Pekanbaru yang memliki karakteristik. Hasil analisis menunjukkan respon struktur jembatan terbesar terjadi di Padang dengan nilai perpindahan sebesar 0,016267 m dan percepatan sebesar 0,0235 m. Sementara itu, respons struktur terkecil terjadi di kota tanjung pinang dengan nilai perpindahan sebesar 0,01552 m dan nilai percepatan sebesar 0,0208 m. Diharapkan dengan diketahuinya hasil prediksi kesehatan struktur jembatan dapat digunakan sebagai referensi/masukan bagi pemerintah dan pihak yang terkait dalam usaha memperbaiki jembatan dengan tepat, sehingga diharapkan dapat mencegah terjadinya keruntuhan struktur jembatan. [EN] Curved steel bridge structure planning must pay attention to the responsiveness of the structure that is vulnerable to deterioration due to fatigue, the threat of strong earthquakes or hurricanes, especially in the region of Sumatra which has a high earthquake risk. This study focuses on predicting the structure of steel curved bridges with spectral response analysis with the help of earthquake structure analysis software based on SNI 1726-2012. The earthquake acceleration taken came from several cities such as Aceh City, Padang, Tanjung Pinang, and Pekanbaru which have characteristics. The analysis shows the largest bridge structure response occurred in Padang with a displacement value of 0.016267 and acceleration of 0.0235. Meanwhile, the smallest structural response occurred in Tanjung Pinang city with a displacement value of 0.01552 and an acceleration value of 0.0208. It is expected that by knowing the results of the bridge structure health predictions can be used as a reference / input for the government and related parties in an effort to repair the bridge appropriately, so that it is expected to prevent the collapse of the bridge structure.
... Girder= 600x600 cm 10 Tahapan selanjutnya yaitu membuat komponen gelagar, diafragma, dan perletakan jembatan. Setelah itu dilakukan penginputan beban jembatan seperti beban mati, beban lalu lintas dan beban gempa berupa respons spectrum untuk masing masing lokasi [10]. ...
... Girder= 600x600 cm 10 Tahapan selanjutnya yaitu membuat komponen gelagar, diafragma, dan perletakan jembatan. Setelah itu dilakukan penginputan beban jembatan seperti beban mati, beban lalu lintas dan beban gempa berupa respons spectrum untuk masing masing lokasi [10]. Perhitungan pembebanan digunakan untuk menentukan beban yang bekerja di jembatan. ...
[ID] Perencanaan struktur jembatan baja pelengkung harus memperhatikan kemampuan respon strukturnya yang rentan terhadap deteriorasi akibat fatik, ancaman gempa bumi kuat atau angin topan, khususnya diwilayah sumatera yang mempunyai resiko gempa yang tinggi. Penelitian ini fokus memprediksi struktur jembatan pelengkung baja dengan analisis repons spectra dengan bantuan software analisis struktur gempa berdasarkan SNI 1726-2012. Percepatan gempa yang diambil berasal dari beberapa kota seperti Kota Aceh, Padang, Tanjung Pinang, dan Pekanbaru yang memliki karakteristik. Hasil analisis menunjukkan respon struktur jembatan terbesar terjadi di Padang dengan nilai perpindahan sebesar 0,016267 m dan percepatan sebesar 0,0235 m. Sementara itu, respons struktur terkecil terjadi di kota tanjung pinang dengan nilai perpindahan sebesar 0,01552 m dan nilai percepatan sebesar 0,0208 m. Diharapkan dengan diketahuinya hasil prediksi kesehatan struktur jembatan dapat digunakan sebagai referensi/masukan bagi pemerintah dan pihak yang terkait dalam usaha memperbaiki jembatan dengan tepat, sehingga diharapkan dapat mencegah terjadinya keruntuhan struktur jembatan. [EN] Curved steel bridge structure planning must pay attention to the responsiveness of the structure that is vulnerable to deterioration due to fatigue, the threat of strong earthquakes or hurricanes, especially in the region of Sumatra which has a high earthquake risk. This study focuses on predicting the structure of steel curved bridges with spectral response analysis with the help of earthquake structure analysis software based on SNI 1726-2012. The earthquake acceleration taken came from several cities such as Aceh City, Padang, Tanjung Pinang, and Pekanbaru which have characteristics. The analysis shows the largest bridge structure response occurred in Padang with a displacement value of 0.016267 and acceleration of 0.0235. Meanwhile, the smallest structural response occurred in Tanjung Pinang city with a displacement value of 0.01552 and an acceleration value of 0.0208. It is expected that by knowing the results of the bridge structure health predictions can be used as a reference / input for the government and related parties in an effort to repair the bridge appropriately, so that it is expected to prevent the collapse of the bridge structure.
