Engineered cementitious composites (ECC) with superior tensile properties are highly promising materials for 3D concrete printing (3DCP). Due to the inherent interlayer interfaces, 3DCP structures are more susceptible to durability issues, while the understanding of which is still lacking. This study systematically investigates the durability performance of 3D printed ECC (3DP-ECC) exposed to dry-wet cycling sulfate attack. The appearance damage, mass change, compressive strength, and interfacial bond strength of 3DP-ECC during 160 cycles of erosion were evaluated and discussed. The deterioration mechanism of 3DP-ECC was explored by analyzing the microstructure, phase composition and pore structure characteristics. Results suggest that the 3DP-ECC specimens met the requirements of sulfate resistance but had inferior resistance compared to the mold-cast ones, primarily owing to the interfacial defects. Microscopic analysis revealed that the sulfate chemical erosion in 3DP-ECC mainly produced gypsum, accompanied by the physical erosion of sodium sulfate crystals. The unique ellipsoidal pores were identified as key factors in the degradation of sulfate resistance and macroscale anisotropy in the 3DP-ECC specimens. Additionally, a prediction model of compressive strength of 3DP-ECC based on pore structure damage evolution was proposed, and a service life prediction model of 3DP-ECC under sulfate attack was also developed, showing favourable accuracy and reliability. This research provides valuable insights and theoretical references for the application of 3DP-ECC in salt lake or saline-alkaline land areas.

While 3D concrete printing is attracting attention due to its intelligent and automatic advantages, introducing reinforcement in the layer-by-layer process remains a major challenge. To address this issue, high-performance fiber-reinforced cementitious composites (HPFRCC) with self-reinforcing characteristic have been developed. This paper presents a comprehensive review of the state-of-the-art on 3D printed HPFRCC (3DP-HPFRCC) with respect to fresh, hardened and durability properties, as well as microstructural features. The relationships between key fresh properties and printability are discussed in detail, including flowability, setting time, green strength and rheology. The anisotropy in the mechanical properties of 3DP-HPFRCC is evaluated and analyzed. The pore structure and fiber orientation are further examined to determine the mechanisms governing the mechanical properties of 3DP-HPFRCC. This review is expected to provide a systematic understanding on the performance of 3DP-HPFRCC and identify future research priorities for the widespread engineering applications.

Recycled aggregate concrete-filled steel tubular (RACFST) columns can effectively improve the inferior mechanical properties of recycled aggregate concrete (RAC), which is a promising and efficient structural member. To ensure structural safety, it is crucial to examine the reliability of RACFST columns. In this study, a reliability-based design analysis for square RACFST stub columns under axial compression was performed. First, the accuracy and applicability of existing design specification models were evaluated using a comprehensive experimental database with 107 square RACFST stub columns. Then, the uncertainties in calculation models, material properties, geometric parameters, and loads were identified, and a design space of 1,872 square RACFST stub columns covering a wide range of design cases was constructed by considering various stochastic factors. Subsequently, the reliability was analyzed based on the Monte Carlo simulation method. It was revealed that the reliability indices of the square RACFST stub columns designed using the existing design factors were not consistent with the target reliability indices, indicating an unbalanced design between structural safety and economic efficiency. To this end, a combined partial factor (gamma cs) was further proposed and calibrated for the design of axially loaded square RACFST stub columns. This work offers a theoretical basis for the safe and reasonable design of RACFST structures.

A three-dimensional meso-scale finite element model, explicitly considering the ribbed geometry of FRP bars, was established in this study to investigate the refined bond behavior between engineered cementitious composites (ECC) and fiber-reinforced polymer (FRP) bars. Mesh sensitivity analysis, together with experimental validation of failure modes, bond-slip behavior, and stress distribution of FRP bars, confirmed the accuracy and reliability of the proposed model. The meso-scale numerical simulations could further reveal the interaction between the FRP bar and ECC, the stress distributions of ECC at FRP ribs and grooves and the ECC crack propagation during loading. Parametric analyses indicated that the FRP bar diameter had limited influence on bond performance with bar diameter smaller than 22 mm. Increasing the anchorage-to-diameter ratio could apparently reduce the bond strength, whereas increasing the cover-thickness-to-diameter ratio could significantly enhance the bond strength by about 550%. Further, increasing the fiber content or tensile performance of ECC could improve the bond strength by 68% and introduce a pronounced strain-hardening stage in the ascending branch of the bond-slip curve. A novel predictive model incorporating ECC tensile strength and the FRP geometric parameters is developed using the genetic algorithm based on 342 datasets. The proposed model is further validated against experimental results and compared with existing models, demonstrating its potential for practical design applications of FRP-reinforced ECC structures.

Recycled fine aggregates (RFA), though widely demonstrating various weaknesses in properties, could be purposely designed in high-strength engineered cementitious composites (HS-ECC) to enhance the mechanical performance and balance the sustainability. This study aims to provide a comprehensive understanding of the roles of RFA as internal curing and artificial flaw agents in HS-ECC. RFA with different moisture contents and three particle size ranges (0.15-1.18 mm, 0.15-2.36 mm, and 0.15-4.75 mm) were used to fully replace quartz sands (QS). Internal curing and artificial flaw mechanism of RFA were elucidated by multiple microcharacterization techniques. Results show that the use of water-saturated RFA effectively mitigated matrix selfdesiccation and significantly reduced early autogenous shrinkage compared to QS. In-situ 1H NMR revealed the internal curing water transport process within the HS-ECC matrix, demonstrating enhanced cement hydration with water-saturated RFA. Despite a lower compressive strength, the tensile performance of HS-ECC with watersaturated RFA was significantly enhanced compared to that with QS. The artificial flaw effect of RFA with different particle sizes enhanced crack propagation, as confirmed by micromechanical modeling, promoting more saturated multiple cracking and higher tensile strain capacity. These findings provide valuable guidance for optimizing RFA use in HS-ECC to mitigate autogenous shrinkage and achieve ultra-high ductility.

Engineered cementitious composites (ECC) are emerging as crucial materials for 3D concrete printing, yet the design process remains largely dependent on a non-quantitative/inefficient trial-and-error approach. The controllable rheological performance plays an important role in developing 3D-printable ECC (3DP-ECC). This study proposes a machine learning (ML)-powered inverse design strategy for predicting and tailoring the rheological properties of ECC. The effects of chemical admixtures and resting time on the rheological properties of ECC were quantitatively evaluated using a factorial design approach, revealing the combined effects. Further, interpretable ML prediction models integrated with multi-objective optimization technique were established to inversely design ECC for specified rheological properties targets. The results, validated based on the database and experimental testing, demonstrate the accuracy and effectiveness of ML strategy. This work enhances understanding of rheology control and offers a highly efficient tool for the design of ECC, which will advance the development and optimization of 3DP-ECC.

Using fiber reinforced polymer (FRP) to confine recycled aggregate concrete (RAC) has been considered as a value-added and promising solution to improve the inferior mechanical properties of RAC. However, accurate quantification of the axial compressive performance of FRP-confined RAC columns and further inverse design remain a significant challenge, due to the complex failure mechanisms involved and RAC's inherent defects. This study aims to develop machine learning (ML) models in conjunction with multi-objective optimization method for performance prediction and inverse design of FRP-confined RAC columns. A comprehensive database consisting of 213 sets of experimental results on the ultimate conditions of FRP-confined RAC columns (i.e., ultimate compressive strength (fcc) and strain (epsilon cu)) was first established. Grey relational analysis (GRA) was then performed to assess the parametric sensitivity of the axial compressive behavior of FRP-confined RAC columns. Subsequently, three different ML models were developed to evaluate the fccand epsilon cu of FRP-confined RAC columns using Bayesian hyperparameter optimization, which showed reasonable accuracy and were superior to existing empirical equations in the literature. The least-squares boosting (LSBoost) models outperformed the other ML techniques with the best prediction accuracy and generalization ability. Lam and Teng's constitutive model was further calibrated with LSBoost, well reproducing the axial stress-strain curves of FRP-confined RAC columns. Finally, combined with the developed LSBoost prediction models, a multi-objective non-dominated sorting genetic algorithm of type II (NSGA-II) was successfully used to obtain the design parameters of FRP-confined RAC columns targeting specific performance requirements. In addition, four data-driven, probabilistic, explicit prediction equations were proposed for practical design of FRP-confined RAC columns using a Bayesian model updating approach. This study offers a ML-powered performance-based design strategy for FRP-confined RAC columns, which will promote efficient development and applications of this innovative composite member.

Reliable bonding performance between steel bar and high-strength engineered cementitious composites (HS-ECC) is crucial for efficient structural members, the in-depth understanding of which, however, is still lacking. This study develops a 3D meso-scale numerical model to investigate the interfacial bonding characteristics between ribbed steel bars and HS-ECC, and the effectiveness of the proposed numerical model is verified. The failure modes, the matrix cracking pattern, the refined bond stress distribution and proportion of bond strength of HS-ECC specimens are vividly demonstrated. Parametric studies are further conducted to investigate the influence of matrix strength, fiber content, steel bar diameter, embedment length, and thickness of matrix cover on the bond behavior. Furthermore, utilizing a Bayesian updating approach, three probabilistic models are further proposed based on the numerical modelling results to predict the bond strength between ribbed steel bars and HS-ECC. The proposed probabilistic models achieve good prediction significantly reducing randomness by fully accounting for the uncertainty of geometric parameters of steel bar, geometry and strength parameters of ECC, offering a robust approach to evaluate the bond strength. This work provides a comprehensive understanding of the bond behavior between ribbed steel bars and HS-ECC based on meso-scale simulation, and establishes a design method based on probabilistic models for engineering applications.

Engineered cementitious composite (ECC), known for its saturated multiple cracking and extraordinary tensile ductility, is considered as one of the most promising cement-based materials in construction industry. To ensure structural safety, it is crucial to evaluate the reliability of ECC and to develop a reliability-based design approach. In this study, mechanical tests on 222 compression and 120 direct tensile specimens across five ECC strength grades (C30-C50) were conducted. A statistically consistent yield strength was defined as 0.81 times the tensile peak strength of ECC. The statistical characteristics of key mechanical properties for reliability analysis were determined. Partial safety factors (PSF) for ECC were calibrated using First Order Reliability Method (FORM) to meet varying target reliability indices. The results show that the material PSF values for tensile failure and compressive failure of ECC were suggested as 1.33 and 1.29, respectively, which were validated to achieve a balanced design between structural safety and economic efficiency. This work offers a practical and codecompatible reliability-based design approach for ECC in structural applications.

The superior tensile ductility of engineered cementitious composites (ECC) offers a promising solution to the challenge of integrating conventional steel reinforcement in three-dimensional (3D) concrete printing (3DCP). However, the widespread adoption of 3D printed ECC (3DP-ECC) is hindered by the reliance on trial-and-error design process. The complex material component and inherent anisotropy of 3DP-ECC pose challenges for accurate property prediction and inverse design. This paper introduces a performance-based design strategy for 3DP-ECC, leveraging machine learning (ML) and multi-objective optimization. The anisotropic-mechanical properties including compressive strength and flexural strength were experimentally and statistically investigated; further, ML prediction models conbined with multi-objective optimization algorithm were developed to inversely design 3DP-ECC for specific mechanical performance requirements, while reducing carbon footprint and material cost. Specifically, an extensive database was assembled, followed by grey relational analysis (GRA) to identify the parametric sensitivity of the mechanical properties of 3DP-ECC. Three representative ML techniques were employed, with the back-propagation artificial neural network (BPANN) demonstrating superior predictive accuracy. Model interpretability analyses uncovered the importance of input parameters and their influence on predicted outcomes. Lastly, non-dominated Sorting Genetic Algorithm II (NSGA-II) integrated with the BPANN models was applied to perform the inverse design of 3DP-ECC, showing good effectiveness and accuracy. This work offers an efficient and viable avenue for performance-based design for 3DP-ECC, along with the potential to develop low-carbon cost-effective 3DP-ECC.

Engineered cementitious composites (ECC) has emerged as a promising self-reinforced material for 3D printed concrete structures, which could potentially remove the dependence on steel reinforcement. The interfacial crack resistance of 3D printed ECC (3DP-ECC) should be emphasized due to the inherent layered stacking process. Tensile strength and fracture toughness are two critical fracture parameters in describing the crack resistance. Determining realistic fracture parameters is crucial for guiding structural safety design. This study aims to develop a fracture mechanics model for determining the size-independent interfacial tensile strength and fracture toughness of 3DP-ECC based on boundary effect model (BEM). Firstly, the interfacial fracture behavior of 3DP-ECC was experimentally investigated by three-point bending tests. A fracture mechanics model was subsequently proposed to predict the size-independent tensile strength and fracture toughness by incorporating the material heterogeneity and discontinuity. The results show that the interfacial tensile strength and fracture toughness of 3DP-ECC could be extrapolated analytically once the peak load was obtained by the three-point bending fracture test, and the predicted values of tensile strength and fracture toughness were proved to follow normal distribution. Additionally, the peak load prediction lines and fracture failure curves with 95 % confidence interval for 3DP-ECC were further constructed using the determined fracture parameters, demonstrating good accuracy and reliability. This work offers a theoretical basis for the safe and reasonable design of 3DP-ECC structural members.

The integration of reinforcement in 3D concrete printing (3DCP) presents a major challenge in current 3DCP industry. Engineered cementitious composites (ECC) with self-reinforcing characteristic provides a promising solution. However, the conflicting requirements between extrudability and buildability are amplified for 3Dprintable ECC (3DP-ECC) due to the existence of a large-amount of fibers. This study investigates the printability region for 3DP-ECC based on standardized field-friendly workability test. The effects of superplasticizer (SP) dosage, hydroxypropyl methylcellulose (HPMC) dosage and fiber content on the workability and printability of 3DP-ECC at varying resting times were explored. The results of grey relational analysis (GRA) show that the workability can be regulated by SP dosage, HPMC dosage, fiber content and resting time. The workability test results in conjunction with the printability evaluation results were used to define the printable region. The suitable ranges of workability parameters including spread diameter, slump, and penetration depth for 3DP-ECC are 130-142.5 mm, 1-31 mm, and 7.5-22 mm, respectively. In addition, the hardened mechanical properties of 3DP-ECC with satisfactory printability were examined by tensile and compressive performance. This study proposes a simple but effective method for predicting the printability of 3DP-ECC and the findings provide guidance for researchers and engineers to easily develop printable ECC.

This study presents the implementation of machine learning (ML) techniques for mechanical properties prediction and analysis of polyethylene fiber-reinforced ECC (PE-ECC). A comprehensive database including different mechanical properties of PE-ECC was first constructed, with total 50 compressive strengths, 123 tensile strengths and 123 tensile strain capacities being assembled. Grey relational analysis was used to investigate the sensitivity of the critical parameters of PE-ECC’s mechanical properties. The evaluation results showed that the supplementary cementitious materials-to-binder ratio, water-to-binder ratio, sand-to-binder ratio, and fiber reinforcing index have significant effects on the mechanical properties of PE-ECC. Three representative ML techniques were utilized and demonstrated good predictive performance. A parametric study was further undertaken to quantify the effects of the selected parameters on the mechanical properties of PE-ECC based on the ML models. This study aims to help researchers and engineers estimate material properties of PE-ECC more effectively and provide supports for ECC design.

In this study, four machine learning (ML) algorithms, namely Support Vector Machine (SVM), Back-propagation Artificial Neural Network (BP-ANN), Adaptive Boosting (AdaBoost), and Gradient Boosted Regression Tree (GBRT), were employed to conduct an in-depth analysis of the global estimation model of the compressive strength and splitting tensile strength of steel fiber recycled aggregate concrete (SFR-RAC). A database containing 465 compressive strength sets and 339 splitting tensile strength sets with different mix proportions was established, and the ML model was trained and tested in combination with Bayesian optimization. The effects of multiple components on the strength of SFR-RAC were studied using partial correlation analysis and SHapley Additive exPlanations (SHAP) analysis. The results showed that AdaBoost and GBRT performed well according to the evaluation indicators, and the deviation between the predicted data and the actual data remained within 20%. The developed models were quantitatively analyzed to study the relationship between characteristics and compressive/splitting tensile strengths, and recommendations were given for key parameters in the SFR-RAC mix proportion. This study suggests improving the modeling accuracy by further incorporating out-of-range data and features for future research.

It has been well-reported that using steel tubes to confine recycled aggregate concrete (RAC) is an effective solution to strengthen the imperfection of RAC, namely RAC-filled steel tubes (RAC-FST). Yet to date available design codes applicable to the axial compressive behavior of RAC-FST stub columns are still limited, and very scarce studies have concentrated on investigating the structural safety. This paper presents a full-scope reliability-based design analysis for the load-carrying capacity of circular RAC-FST stub columns under axial compression. First, the performances of several code-based models were evaluated in terms of model uncertainty based on a reliable experimental database that comprises 94 circular RAC-FST stub columns, respectively. Then, the statistical characteristics for characteristic compressive strength of RAC were specially identified, and other uncertainties associated with steel strength, geometrical configurations, and loads were taken from available publications. Subsequently, the reliability of circular RAC-FST stub columns designed using existing design factors in current normal concrete filled steel tubes (CFST) codes was investigated in conjunction with Monte Carlo simulation (MCS) technique. Results indicated that in order to achieve the reasonable design of circular RAC-FST stub columns, existing design factors need to be calibrated. Accordingly, a combined partial factor directly reducing the whole resistance of composite cross-section was proposed, and the corresponding recommendations were provided based on the target reliability index-oriented calibration process.

Engineered cementitious composite (ECC) has been intensively studied due to its excellent tensile performance. However, classical micro-mechanical design theory of ECC is qualitative and fails to give detailed ECC mixtures at specific tensile parameters. This study aims to develop a performance-based mixture design model to generate ECC mixtures using generative AI method. An experimental database consisting of 129 polyethylene fiber reinforced ECC (PE-ECC) records has been built. The database was used to train one invertible neural network model and two artificial neural network models. A series of PE-ECC mixtures were generated by the proposed model based on desired mechanical performance and sustainable requirements. Based on the experimental results, the developed model was proven to compose PE-ECC mixtures that satisfy the target requirements with a maximum deviation of less than 16%. The neural network-based model can be used in various application scenarios (e.g., low-cost ECC and low-carbon ECC), thus promoting the development of ECC materials in the area of research and engineering application.

Engineered cementitious composites (ECC) is famous for its excellent tensile deformability and crack width control ability. Steel reinforced ECC beams under shear loading exhibit brittle behaviors, which leads to a lack of structural safety. The reliability-based shear capacity model of steel reinforced ECC beams needs to be established for ensuring the structural shear safety. In this research, a test database of steel reinforced ECC beams under shear loading was constructed based on the given selecting principles. Through considering the shear resistance provided by fibers, the shear capacity model of steel reinforced ECC beams was preliminary established. The statistical characteristics of random variables were described and determined, and reliability index was calculated through Monte Carlo simulation. It is found that reliability index increases with increasing ECC strength, while decreases with increasing stirrup strength or stirrup ratio, especially at the high partial factor of ECC. For different target reliability indexes, the recommended partial factors of ECC are obtained through calibration. This research supplies a useful reference for the shear design of steel reinforced ECC beams in the actual engineering.

It is widely accepted that the inferior characteristics of recycled concrete aggregates (RCAs) reduce the me-chanical properties of structural concrete and the available prediction models receive undesirable accuracies in estimating the compressive behavior of recycled aggregate concrete (RAC). To this end, this paper presents a series of data-driven analysis on unified modeling of the compressive behavior of unconfined and confined RACs. First, four reliable experimental databases containing the results of natural aggregate concrete (NAC) and RAC under uniaxial compression, RAC under triaxial compression, and fiber reinforced polymer (FRP)-confined RAC under axial compression were assembled through an extensive literature review, respectively. Based on the collected data, Mohr-Coulomb failure criterion was then employed to construct a model for estimation of the failure envelope of RAC under axial compression and different lateral active confinement stresses. Finally, unified models for predicting the compressive properties (i.e., compressive strength/peak stress and strain at peak stress) of unconfined and confined concretes incorporating natural aggregates (NAs) and/or RCAs were developed so as to provide a design-oriented procedure for material and structural assessments of both NAC and RAC. The results demonstrate that the proposed models exhibit acceptable predictions, thereby making various stakeholders gain sufficient confidence in use of RAC products.

Dispersing steel fibers into plain concrete has been verified as an effective way to enhance the bearing capacity of structural components. To date, however, the effects of fibers on the basic mechanical properties of Steel Fiber Reinforced Concrete (SFRC) and their uncertainties are not well documented and there are no accurate models that can predict their mechanical properties. This greatly limits the application in real constructions. Considering such a situation, three aspects of works were conducted in this study to alleviate that difficulty, those are, building test database and checking the statistical distribution, reviewing and assessing existing models, and establishing highly-accurate models applied to estimate the compressive, tension and flexural properties of SFRC. To achieve the first two goals, a large database containing 1038 experimental records was compiled from existing publications and 145 predictive expressions were collected from the available literature. The last object was obtained by conducting Bayesian model updating analysis. It was finally found from the present study that: (a) Steel fibers can effectively enhance plain concrete's compressive, tensile and flexural strengths, but induce greater variability in those indexes. That variability is well described with a lognormal probability distribution. (b) Expressions proposed by Abbass and by Padmarajaiah generally give the most accurate predictions for the compressive strength and elastic modulus, while the formulas of Chinese code JG/T472–2015 and Padmarajaiah give the most accurate tensile and flexural strengths. (c) Because of combining advantages of the experience of the prior model and the accuracy of the test records, Bayesian updating method based analytical model usually possesses an obviously high precision.