Dr Yidu Bu PhD

The conventional surface treatments on stainless steel plates yield slip factors less than the required value for a friction grip, which hinders the application of stainless steel high-strength bolts. The slip factor and the treatment process of stainless steel surfaces are not clearly specified in most of the existing design codes. Existing studies also show different slip factors from similarly treated surfaces. In this paper, a new surface treatment is proposed: the two touching surfaces of the bolted connection are grit-blasted on one side and coated with high-velocity oxygen fuel (HVOF) on the other side. The slip behaviour of the new method was compared with four conventional methods. The roughness and hardness of different surfaces were measured prior to a total of 30 slip tests where the slip factors of multiple stainless steel surfaces were tested. It was shown that the new proposed surface treatment method can yield a slip factor of 0.61, much higher than the conventional surface treatments, meeting the requirements as a slip-resistant connection. The preload relaxation of stainless steel high-strength bolts was also monitored and compared to standard grade 10.9 high-strength bolts. We found that the preload relaxation in stainless steel high-strength bolts is negligible.

Earthquake-induced fractures in steel structures are characterised by high-strain low-cycle conditions. In order to investigate the ultra-low cyclic fatigue fracture of steel welded joints under earthquakes, two most commonly used structural steels (Q235B and Q345B) and the corresponding welds were studied by experiments and numerical analysis in this paper. Specimens were extracted from the base material, the weld metal and the heat affected zone to investigate the behaviour in different parts of the welded joint. Eighteen smooth round bars were tested under large strain amplitudes, the hysteretic properties, damage degradation characteristics and failure process were analyzed. Constitutive model named Chaboche model was calibrated to describe the cyclic hardening behaviour of these materials. Seventy-two notched round bars with three different notch sizes and two loading protocols were tested to study the fracture behaviour of different materials at different stress triaxialities and different strain amplitudes. Two micromechanical fracture models: cyclic void growth model and degraded significant plastic strain model were calibrated based on the test results. The micromechanical models and Chaboche model were incorporated into numerical simulations by software ABAQUS with subroutine VUMAT to predict the materials fracture. The results show that the failure process under cyclic loads is opposite to that of monotone loads. The dissipation capacity of Q345B is superior to that of Q235B. The fracture resistance deteriorate more in the weld zone under the same loading conditions. The validated models can be used to effectively and accurately evaluate the fracture in steel welded connections under ULCF conditions.

As one of the means to inhibit offshore pipeline corrosion, lining a thin layer of stainless steel within the carbon steel pipeline provides an economical and durable design. Current design codes EN1993-1-1 and DNV–OS–F101 ignore the contribution of the liner pipes towards the cross-section capacity, which might lead to inefficient design. The focus of this study is to investigate the interaction of the two pipe layers arises during manufacture and the development of local buckling of the lined pipe under compression. Experiments to study the interaction behaviour of the liner pipe and the outer pipe, including saw tests and ring-split tests were performed. Then a series of short lined pipes were tested under axial compression, with a built-in camera to capture the development of liner wrinkling. After the tests, numerical models were developed and validated, and a series of parametric studies were conducted. The obtained test and FE results were used to assess the applicability of the codified provisions for the design of lined pipes under compression. Improved design method was developed considering the liner pipe contribution and material strain hardening. The proposed approach provides better prediction accuracies on the structural behaviours of lined pipes under axial compression.

Steel structures have been widely used in constructions due to their advantages of lightweight, high strength, short construction time and high recycling and reuse potential. Fracture failure in steel structures should be prevented to avoid collapse of the whole structures. Micromechanical fracture models can capture the fracture initiation mechanisms and therefore can be used to predict ductile fractures in steel. Twelve smooth round bars were carried out to obtain the material properties and 36 notched round bars were tested to calibrate the parameters of stress modified critical strain (SMCS) model and the void growth model (VGM) for structural steels (Q235B and Q345B) and the corresponding welds. Specimens were extracted from the base metal, the weld metal and the heat affected zone (HAZ) to investigate fracture behaviour in different parts of the welded joint. Scanning electron microscope (SEM) measurements were taken and finite element models were developed to calibrate the models. The test results and calibrated parameters are reported. Moreover, the calibrated models are applied to analyses the fracture behaviour of welded joints and their accuracy are validated. The calibrated and validated models can be used for further numerical fracture analysis in welded steel structures.

Unequal-leg angles can be used to provide more efficient design in specific cases, but the behaviour of unequal-leg angles has received much less attention than equal-leg angles. This paper studies the flexural and torsional buckling modal participation and investigates the structural behaviour of aluminium alloy unequal-leg angle columns. The experimental programme, which was carried out on two unequal-leg angle cross-sections, included material coupon tests, initial geometric imperfection measurements and a total of eight column tests. The test setup, procedure and results are fully reported. Following the experimental investigation, numerical models were developed and validated against the test results. Upon validation, a series of parametric studies were carried out to analyse the influence of width-to-thickness ratio and aspect ratio on the flexural and torsional buckling modal participation. The ultimate resistances obtained from simulations were used to assess the accuracy of the current design provisions given in the European standard for aluminium alloy structures. The result comparison shows that the existing design provisions yield inaccurate predictions for unequal-leg angle columns. Considering mode interaction, a more accurate and convenient design method was proposed for unequal-leg angle columns.

Longitudinally profiled (LP) steel plates are plates rolled to variable thickness along the rolling direction in order to suit practical loading distributions in construction applications. The elastic buckling load of LP steel plates is different to normal flat plates due to the change in thickness and it is hard to find analytical solutions to the governing equation. In this paper, the elastic buckling coefficients of rectangular LP plates were calculated using the Galerkin and Rayleigh-Ritz method (GRM). The plates were under uniform compression and four typical boundary conditions were considered. The element expression of the feature matrix was obtained by theoretical derivation and the elastic buckling coefficients were obtained by calculating the eigenvalues of the feature matrix numerically with MATLAB. Numerical models of LP plates were developed to validate the derivation. Upon validation it was further found that the elastic buckling coefficients of LP plates can be approximately expressed by those of constant thickness steel plates together with variable thickness coefficients. A series of approximation formulae of elastic buckling coefficients of LP rectangular steel plates were proposed. Simplified design formulae of elastic buckling coefficients of LP plates were proposed for the first time to promote the application of LP plates in engineering practice.

Lining internally a thin corrosion resistance alloy (CRA) layer within an ordinary carbon steel pipe provides one economical approach, when transmitting highly corrosive offshore hydrocarbons. The lined pipe is expected to make optimize usage of the two types of materials, providing significant corrosion and structural resistance. Existing studies mostly covered the lined pipe under bending, corresponding to the load case during Reel-lay procedure. The behaviour of the lined pipe under axial compression, which can be induced by thermal field when transporting hot hydrocarbons, however, has received much less attention. The present paper describes a series of laboratory tests performed on lined pipes, including tensile coupon tests of the outer and the liner materials, initial geometric imperfection measurements and lined pipes tested under axial compression. Results of the column tests were used to validate numerical models and employed for subsequent parametric investigations. Based on the results, design provisions DNV–OS–F101 and EN 1993-1-1 were assessed for lined pipes under axial compression and a buckling curve was proposed for lined pipes for the first time.

The structural performance of aluminium alloy unequal angle columns was investigated in this study. Fourteen columns made of two extruded aluminium alloy unequal angle cross-sections were tested. Complementary material property tests and geometric imperfection measurements were also carried out. The test setup, procedure and results were fully reported. After the tests, numerical models were developed to replicate the tested pin-ended extruded aluminium alloy unequal angle column behaviour. The validated finite element models were then adopted to analyse the effect of the configuration of initial geometric imperfections, which proved to be pivotal in mode interaction of the column buckling performance. Parametric studies were carried out with specific initial imperfection configuration to generate further numerical data. The ultimate resistances derived from the experiments and finite element analyses were employed to evaluate the applicability of the current codified standards. The result comparison shows that current design provisions yield unsafe predictions for pin-ended unequal angle columns since the buckling mode interaction is not considered. The development of applicable design proposals and further investigation into buckling modal participation for pin-ended extruded aluminium alloy unequal angle columns are required.

The flexural-torsional buckling response and design of stainless steel I-section beam-columns are investigated in this paper. First, a series of laboratory tests on laser-welded stainless steel I-section beam-columns susceptible to flexural-torsional buckling is presented. The results obtained are supplemented by further data generated by means of numerical parametric studies on both conventionally arc-welded and laser-welded stainless steel members covering a wide range of member slenderness and combinations of loading. Existing provisions for the design of welded stainless steel I-section elements against flexural-torsional buckling are then assessed and found to require improvement. Finally, new formulae for the design of stainless steel I-section beam-columns susceptible to flexural-torsional buckling are proposed. The new proposals yield improved accuracy and consistency over existing provisions and their suitability for inclusion in the upcoming version of the European structural stainless steel design code EN 1993-1-4 is confirmed by reliability analysis in accordance with EN 1990.

Bolted stiffened extended end-plate (BSEP) joint is a commonly used beam-to-column joint in steel frames. In this paper, five full-scale stainless steel BSEP joints were tested under cyclic loading. The tested joints were made of Grade S31608 (EN 1.4401) stainless steel welded I-sections, assembled by four types of bolts, including galvanized high strength bolts of Grades 10.9 and 8.8, and austenitic stainless steel bolts of Grade A4-70 and A4-80. Variations of end-plates of different thicknesses (16 mm and 20 mm) were also considered in the design of the test specimens. Material properties of the plates were tested and the slip-resistant behaviour of the bolts with surface treatments were obtained prior to the joint tests. The test setup, loading protocol and instrumentation employed in the tests are described herein. The obtained test results are reported, including the full moment-rotation curves and the failure modes of the joints. It is observed from the seismic tests that hysteresis curves of all specimens show some sliding and pinching, especially for connections with stainless steel bolts or thin end-plates. The test results were used to validate finite element (FE) models, which can be employed for further numerical investigations. The test results were also employed to assess the stiffness and moment resistance capacity predictions according to the current European design provisions.

The stability and design of laser-welded stainless steel I-section beam-columns are explored in this study. Owing to the high precision and low heat input of laser-welding, structural cross-sections produced using this fabrication method have smaller heat affected zones, lower thermal distortions and lower residual stresses than would typically arise from traditional welding processes. Eighteen laser-welded stainless steel beam-columns were tested to investigate the member buckling behaviour under combined compression and bending. Two I-section sizes were considered in the tests: I-50 × 50 × 4 × 4 in grade EN 1.4301 and I-102 × 68 × 5 × 5 in grade EN 1.4571 austenitic stainless steel. The two cases of minor axis bending plus compression and major axis bending plus compression with lateral restraints were investigated. The initial loading eccentricities in the beam-column tests were varied to provide a wide range of bending moment-to-axial load ratios. The test results obtained herein and from a previous experimental study were used to validate finite element (FE) models, which were subsequently employed for parametric investigations to generate further structural performance data over a wider range of cross-section sizes, member lengths and loading combinations. The obtained test and FE results were utilized to evaluate the accuracy of the beam-column capacity predictions according to the current European and North American design provisions and a recent proposal by Greiner and Kettler. Finally, an improved approach for the design of stainless steel I-section beam-columns is proposed.

Design guidance for stainless steel structures has become more comprehensive and widely available in recent years. This, coupled with a growing range of structural products and increasing emphasis being placed on sustainable and durable infrastructure, has resulted in greater use of stainless steel in construction. A recent addition to the range of structural stainless steel products is that of laser-welded sections. Owing to the high precision and low heat input of the fabrication process, the resulting sections have smaller heat affected zones, lower thermal distortions and lower residual stresses than would typically arise from traditional welding processes. There currently exists very limited experimental data on laser-welded stainless steel members and their design is not covered by current design standards. The focus of this study is therefore to investigate the cross-sectional behaviour of laser-welded stainless steel I-sections in bending. The present paper describes a series of laboratory tests performed on laser-welded stainless steel I-sections, including tensile coupon tests, initial geometric imperfection measurements and in-plane bending tests. Results of the bending tests are used to validate finite element (FE) models, which are subsequently employed for parametric investigations. The obtained experimental and FE results are used to assess the applicability of the existing design provisions of EN 1993-1-4, AISC Design Guide 27 and the continuous strength method (CSM) to laser-welded stainless steel sections. It was found that the scope of application of these existing design provisions may be safely extended to laser-welded sections.

Stainless steel is widely used in construction due to its combination of excellent mechanical properties, durability and aesthetics. Towards more sustainable infrastructure, stainless steel is expected be more commonly specified and to feature in more substantial structural applications in the future; this will require larger and typically welded cross-sections. While the structural response of cold-formed stainless steel sections has been extensively studied in the literature, welded sections have received less attention to date. The stability and design of conventionally welded and laser-welded austenitic stainless steel compression members are therefore the focus of the present research. Finite element (FE) models were developed and validated against a total of 59 experiments, covering both conventionally welded and laser-welded columns, for which different residual stress patterns were applied. A subsequent parametric study was carried out, considering a range of cross-section and member geometries. The existing experimental results, together with the numerical data generated herein, were then used to assess the buckling curves given in European, North American and Chinese design standards. Following examination of the data and reliability analysis, new buckling curves were proposed, providing, for the first time, design guidance for laser-welded stainless steel members.

Doubler-plate reinforcement does not impinge on the appearance of tubular joints, and thus is widely used to improve their strength. To understand the influence of doubler-plates on the axial hysteretic behavior of square hollow section (SHS) tubular joints, two groups of unreinforced and doubler-plate reinforced (DPR) joints were experimental tested and numerically simulated under axial cyclic loading. Parametric studies were conducted to identify key parameters that influence the hysteretic behavior of DPR joints. The design of the specimen and set up of the tests are presented, and failure sequences from the experimental tests are described. The hysteretic curves, skeleton curves, energy-dissipation capacity, ductility, strength, as well as the hysteretic mechanisms of the joints, are discussed. The doubler-plate is found to effectively prevent a crack from developing towards the chord flange and to protect the integrity of the chord, which maintains the compressive capacity of the DPR joints after cracking. The doubler-plate can also increase the strength of the joints, but with a decrease in the energy dissipation capacity and ductility. The brace-chord width ratio and the doubler-plate to chord width ratio are the key parameters that influence the load transfer mechanisms and failure modes of the DPR joints, and the configuration suggestions for DPR joints are proposed.

Laser-welding is a high precision fabrication process suitable for joining a wide range of steels and stainless steels. Laser-welded structural stainless steel members, for which there are currently little experimental data owing to their recent introduction to the construction industry, are the focus of the present study. To address the lack of test data and to investigate their structural response, a total of 9 stub column tests and 22 flexural buckling tests (14 buckling about the minor axis and 8 about the major axis) have been performed on laser-welded austenitic stainless steel I-section members. Complementary tensile coupon tests, initial geometric imperfection measurements, and residual stress measurements have also been carried out and are reported herein. Based on the results obtained, a representative residual stress pattern is proposed, the design provisions of Eurocode 3 Part 1.4 and the continuous strength method are assessed, and column buckling curves for laser-welded stainless steel I-section members are recommended.

Corrosion of carbon steel reinforcing bar can lead to deterioration of concrete structures, especially in regions where road salt is heavily used or in areas close to sea water. Although stainless steel reinforcing bar costs more than carbon steel, its selective use for high risk elements is cost-effective when the whole life costs of the structure are taken into account. Considerations for specifying stainless steel reinforcing bars and a review of applications are presented herein. Attention is then given to the elevated temperature properties of stainless steel reinforcing bars, which are needed for structural fire design, but have been unexplored to date. A programme of isothermal and anisothermal tensile tests on four types of stainless steel reinforcing bar is described: 1.4307 (304L), 1.4311 (304LN), 1.4162 (LDX 2101®) and 1.4362 (2304). Bars of diameter 12 mm and 16 mm were studied, plain round and ribbed. Reduction factors were calculated for the key strength, stiffness and ductility properties and compared to equivalent factors for stainless steel plate and strip, as well as those for carbon steel reinforcement. The test results demonstrate that the reduction factors for 0.2% proof strength, strength at 2% strain and ultimate strength derived for stainless steel plate and strip can also be applied to stainless steel reinforcing bar. Revised reduction factors for ultimate strain and fracture strain at elevated temperatures have been proposed. The ability of two-stage Ramberg-Osgood expressions to capture accurately the stress-strain response of stainless steel reinforcement at both room temperature and elevated temperatures is also demonstrated.