Component-based Model Research Articles

The development of a component-based model for extended endplate joints in fire-induced progressive collapse scenarios

This paper develops a generalized component-based model (CBM) for extended endplate (EP) joints designed to be applicable in progressive collapse scenarios induced by fire, which is capable of capturing the nonlinear full-range behavior of EP joints from initial yielding to ultimate failure. Based on refined FE modeling established in preceding studies, a simplified CBM method is established to simulate the collapse resistance of EP joints subjected to fire-induced column removal scenarios. For panel zones of fire-exposed EP joints, a trilinear model is applied to characterize the three-phase force-deformation behaviors of “linear-hardening-unloading” for axial springs, while a relatively large shear stiffness is used to characterize the minor shear deformation at the panel zones. For connecting zones of fire-exposed EP joints, the respective constitutive models and corresponding spring parameters for diverse spring components are established based on the strain distributions and loading transmission mechanisms of equivalent T-stub components. The proposed simplified CBMs with detailed spring parameters as well as their constitutive laws were incorporated into the ABAQUS program to validate against both experimental results and refined FE modeling. A comparative study with test data manifests that the proposed CBM method can precisely capture the collapse response of EP joints at elevated temperatures with high computing efficiency.

Optimised rotational-spring component-based modelling strategy for seismic resistant steel-concrete composite joints and frames with continuous or isolated slab

Joints and frames in steel-concrete composite systems represent complex mechanical assemblies that require specific calculation procedures to optimise their detailing and structural capacity, particularly under seismic loads. To this aim, component-based modelling approaches should be able to account for the most relevant mechanisms and resistance/stiffness behaviours of individual members, and their mutual interaction. In this paper, two different simplified non-linear approaches are considered for steel-concrete composite beam-to-column joints, and are specifically applied to a seismic resistant case-study frame with X-concentric bracings. Both beam-to-column joints with or continuous (“JA” joint) or fully isolated (“JB” joint) slab are examined. First, non-linear axial springs are assembled and calibrated on the base of a previous study (“Type 1″ model (“T1″)), according to force-displacement relationships proposed in the DPC-ReLUIS Italian guidelines. Successively, a novel modelling approach based on non-linear rotational springs is presented (“Type 2″ model (“T2″)), to further simplify the computational cost of T1 strategy, and allow to efficiently account for the moment-rotation behaviour of the examined joints. The preliminary numerical validation is carried out based on past literature experiments. Moreover, the optimized T2 approach is used to explore the in-plane lateral, seismic performance of a 2D steel-concrete composite frame, which is specifically designed with X-concentric bracings. The seismic capacity of the frame (and the associated interaction of components, especially the joint zone with the bracing system) is addressed on the base of pushover analyses.

Component-based modeling of cascading failure propagation in directed dual-weight software networks

Software vulnerabilities often lead to cascading failures, resulting in service unavailability and potential breaches of user data. However, existing models for cascading failure propagation typically focus solely on the static design’s calling relationships, disregarding dynamic runtime propagation paths. Moreover, current network topology models primarily consider function calling frequency while overlooking critical factors like internal failure probability and component failure tolerance rates. Yet, these factors significantly influence the actual propagation of software cascading failures. In this study, we address these limitations by incorporating internal failure probabilities and calling frequencies as node and edge weights, respectively. This forms the basis of our component-based directed dual-weight software network cascading failure propagation model. This model encompasses the evaluation of cascading failure propagation through intra-component and inter-component propagation probabilities, alongside the constraint of component failure tolerance rates. Through extensive experiments conducted on six real-world software applications, our model has demonstrated its effectiveness in predicting software cascading failure propagation processes. This method deepens our understanding of software failures and structures, equipping software testers with the knowledge to make well-informed judgments regarding software quality concerns.

Seismic behavior of high-rise modular buildings with simplified models of inter-module connections

Inter-module connections serve as crucial components, horizontally and vertically connecting individual modules to form modular buildings. This paper presents a component-based model that divides inter-module connections into vertical and horizontal connection components, facilitating the simplification of shear and axial behaviors for input into the global numerical model. Initially, experimental studies and numerical simulations were conducted to investigate the shear and axial behavior of these connection components, following which simplified connection component models were derived. Subsequently, a component-based model of inter-module connections was established, in which the shear and axial properties of the vertical and horizontal components were represented by nonlinear connector elements. Finally, the proposed component-based model was incorporated directly into the global frame models of high-rise steel modular buildings, and the seismic behavior was evaluated. The accuracy of the proposed component-based model was investigated by comparing the natural vibration characteristics and global building responses of high-rise steel modular buildings with traditional empirical models established based on assumptions of rigid or pinned fixities. The results reveal that traditional empirical models fail to accurately represent the discontinuous properties of inter-module connections, resulting in an overestimation of structural lateral stiffness. In contrast, the component-based model accounts for the actual load-transfer mechanisms of vertical and horizontal components, providing a more accurate understanding of the natural vibration characteristics and seismic response of high-rise steel modular buildings. The proposed component-based model establishes a critical link between isolated connection behavior and structural response assessment for high-rise steel modular buildings, making it well-suited for accurately predicting global building responses.

Component-based modelling of single-leg bolted steel angle connections at elevated temperatures

As a vital part of contemporary society, steel lattice transmission towers and communication towers play an important role in the power transmission grid and telecommunication, respectively. Being often located in rural areas, they are prone to bushfire attacks due to heatwaves and droughts, which are becoming increasingly commonplace. However, unlike the steel members found in a lattice tower, the behaviour and material properties of which at elevated temperatures have been comprehensively studied and codified in design standards, the semi-rigid behaviour of the bolted connection between such members at elevated temperatures has not been investigated thoroughly. Therefore, this paper aims to investigate both the axial and rotational behaviour of single-leg bolted steel angle connections at both ambient and elevated temperatures. A theoretical model using the component method is firstly proposed to model the axial load-slip behaviour incorporating the temperature increase, with the effects of the bolt shear, plate bearing, friction between clamped members, bolt slip, and member elongation being considered. Once the axial load-slip model is proposed, the component-based model for describing the moment-rotation behaviour of the bolted connection is derived based on the load-deformation relations of individual bolts in the bolt group. The reliability of the theoretical model is then validated by modelling the connection behaviour obtained from previous connection test results. A three-dimensional finite element model is also constructed using ABAQUS software to further validate the accuracy of the proposed theoretical model for predicting the axial and rotational behaviour of single-leg bolted angle connections at elevated temperatures. It is demonstrated that the proposed component-based theoretical model for the bolted angle connection accounting for axial and rotational semi-rigidity as well as temperature elevation is capable of effectively capturing the load-deformation relationship for the single-leg bolted angle steel connections, and would be applicable to the study of steel lattice towers under bushfire attack.

A component method approach for single-sided beam-to-column joints with CHS column and welded double-tee beam

This paper focuses on predicting the flexural stiffness and strength of beam-to-column joints with Circular-Hollow-Section (CHS) columns and externally welded double-tee beams using the Component Method approach. Currently, Eurocode 3 Part 1.8 lacks guidelines for assessing the stiffness of these joints, resulting in their practical modelling using extreme assumptions like pin or full restraint. Nevertheless, the literature demonstrates that these joints can exhibit semi-rigid behaviour due to the relatively high local deformability of the tube in the joint area. Another concern relates to the strength of these joints. In fact, recent studies have shown that Eurocode 3 Part 1.8 formula for predicting the yield resistance of these joints is over-conservative.Within this framework, the purpose of this study is to fill the current knowledge gaps (over-conservative strength prediction and lack of rules for determining the stiffness) by undertaking a thorough investigation involving numerical and analytical activities. Specifically, the research presents a component-based modelling approach for predicting stiffness and strength of CHS to double-tee beam connections.In this regard, the work presented in this paper has involved the selection of experimental tests from literature to validate a Finite Element (FE) model, specifically focusing on X- and T-joints and internal and external beam-to-column joints. The validated numerical models have been then employed to simulate the response under monotonic loading conditions of additional sets comprising forty T-joints and thirty external beam-to-column joints, respectively. The currently available formulations for predicting the resistance of T-joints were applied to the forty cases simulated, evaluating their accuracy. This allowed the individuation of the most accurate literature formula for the strength prediction of T-joints, showing that the recent proposal of Voth and Packer (2012) yields sufficiently accurate results. Instead, analytical formulations for predicting the stiffness of T-joints were properly derived in a closed-form solution and were validated against the results of the parametric study. Finally, drawing upon knowledge regarding the stiffness and strength of T-joints, the flexural response of external beam-to-columns connections was evaluated proposing a component model, whose accuracy was verified against the cases simulated in the parametric analysis.

A non-overlapping optimization-based domain decomposition approach to component-based model reduction of incompressible flows

We present a component-based model order reduction procedure to efficiently and accurately solve parameterized incompressible flows governed by the Navier-Stokes equations. Our approach leverages a non-overlapping optimization-based domain decomposition technique to determine the control variable that minimizes jumps across the interfaces between sub-domains. To solve the resulting constrained optimization problem, we propose both Gauss-Newton and sequential quadratic programming methods, which effectively transform the constrained problem into an unconstrained formulation. Furthermore, we integrate model order reduction techniques into the optimization framework, to speed up computations. In particular, we incorporate localized training and adaptive enrichment to reduce the burden associated with the training of the local reduced-order models. Numerical results are presented to demonstrate the validity and effectiveness of the overall methodology.

Energetic and exergetic metrics of a cargo aircraft turboprop propulsion system by using regression method for dynamic flight

Nowadays, the development of aviation in line with sustainability goals is one of the current challenges. Being a source of environmental impact, aero-engines have been the main research and design object for achieving these aims. In this study, thermodynamic analysis involving exergy and sustainability computations is performed at different flight conditions where Mach varies between 0 and 0.7 and altitude changes between 0 and 7.7 km. Based on this analysis, modeling exergy parameters for each component are carried out with multiple regression approach. The exergy efficiency of turboprop engine at dynamic flight conditions varies between 15% and 25.9% whereas it is measured between 76 and 77% for the combustor and between 93 and 99% for gas turbine. Moreover, exergy destruction of the turboprop engine changes between 2.15 MW and 7.55 MW. Exergetic improvement potential rate (IPR) of the combustor resides between 0.4 MW and 1.21 MW throughout flight conditions whereas it is found at orders of 0.1 MW or less for other components. On the other hand, linear and quadratic modelings are performed for several parameters of turboprop engine components including exergy efficiency, exergy destruction and IPR under dynamic flight conditions. The determination coefficient (R2) of the models changes between 0.69 and 0.98 in linear modeling, whereas R2 in quadratic modeling improves to 0.97 and 0.99 ranges. It is thought that component-based modeling by considering different flight conditions could contribute to determining points where component efficiency is the highest.

Bond-slip model of headed bar and its application to component-based model for precast concrete joints under accidental loads

This study focused on investigating bond-slip behaviour of headed bars embedded in well-confined concrete. Test results based on eight pull-out specimens under displacement-controlled incremental loading showed that recommended development lengths from ACI318–19 are conservative in developing ultimate strength of headed bars associated with smaller slips than straight bars. Moreover, it was found that different (but sufficiently long) embedment lengths had a significant effect on the free-end slip but negligible effect on the loaded-end slip of headed bars. Additionally, the study highlighted the significance of head bearing in anchorage formation, especially during inelastic stage when bond strength had deteriorated. To address inaccuracy in predicting structural behaviour caused by inadequate anchorage strength of headed bars, a macro bond-slip model was proposed based on correlation in strain distribution obtained from the pull-out tests. Subsequently, the proposed model was incorporated into a component-based model (CBM) and validated against test results of PC joints under seismic loading. A numerical investigation using the validated CBM was conducted to study behaviour of exterior PC joints with headed bars and special detailing. These included Type X joint with X-bent bars and plastic hinge relocation (PHR), and Type A joint with an additional bar layer (ABL) in the beam-joint region, subjected to extreme loads such as seismic action or progressive collapse. The results revealed that seismic behaviour of exterior joints was governed by ductile failure of the beam at the PHR and vertical joint interface in Type X and Type A joints, respectively. Progressive collapse was primarily caused by column failure, with flexural and catenary action effectively preventing collapse. Notably, under both loading conditions, presence of ABL in Type A joint and X-bent bars with PHR in Type X joint improves flexural strength of beams at critical sections and joint shear strength, ultimately enhancing structural performance of joints. Design recommendations include replacing hooked bars with headed bars as well as implementing X-bent bars with PHR and ABL in PC joints. The authors advocate the strong-column-weak-beam design approach as an effective measure to prevent progressive collapse.

Component-based models of beam-column joints exposed to fire in steel framed structures against progressive collapse

This paper is supplementary research following the previous experimental and numerical studies on the progressive collapse behaviors of beam-column substructures with different joint configurations under middle-column removal scenarios induced by fire. The component-based models (CBMs) of beam-column joints exposed to fire, which include the welded web-welded flange (WW) joint, as well as the bolted web-welded flange (BW) joint, were proposed based on refined FE models established in previous studies. Subsequently, the proposed CBMs with detailed spring arrangements and constitutive parameters were incorporated into the ABAQUS software to validate the collapse responses of beam-column joints obtained from both test results and refined solid FE models. A comparison investigation with experimental data indicates that the proposed CBMs can accurately reflect the nonlinear collapse behaviors of different fire-exposed beam-column joints with relatively high computational efficiency. Moreover, the validated CBMs are further incorporated into the collapse analysis of a multi-story planar steel framed structure exposed to fire, by compiling the VUSDFLD subroutine, to realize the component deletion process. The parameter effects of fire location, joint configurations, and loading ratios on collapse behaviors and force redistributions of framed structures exposed to fire were investigated. Based on analysis results, it is shown that the component failure temperature is quite sensitive to whether the specific joint configurations are considered in the analysis of fire-resistant progressive collapse in planar framed structures, which should be paid sufficient attention to in the performance evaluation of collapse resistance. For those planar framed structures with larger loading ratios, it is recommended to increase the section of the frame column or carry out fire protection to prevent the progressive collapse of the structure.

Multi-step ahead ozone level forecasting using a component-based technique: A case study in Lima, Peru

<abstract><p>The rise in global ozone levels over the last few decades has harmed human health. This problem exists in several cities throughout South America due to dangerous levels of particulate matter in the air, particularly during the winter season, making it a public health issue. Lima, Peru, is one of the ten cities in South America with the worst levels of air pollution. Thus, efficient and precise modeling and forecasting are critical for ozone concentrations in Lima. The focus is on developing precise forecasting models to anticipate ozone concentrations, providing timely information for adequate public health protection and environmental management. This work used hourly O$ _{3} $ data in metropolitan areas for multi-step-ahead (one-, two-, three-, and seven-day-ahead) O$ _{3} $ forecasts. A multiple linear regression model was used to represent the deterministic portion, and four-time series models, autoregressive, nonparametric autoregressive, autoregressive moving average, and nonlinear neural network autoregressive, were used to describe the stochastic component. The various horizon out-of-sample forecast results for the considered data suggest that the proposed component-based forecasting technique gives a highly consistent, accurate, and efficient gain. This may be expanded to other districts of Lima, different regions of Peru, and even the global level to assess the efficacy of the proposed component-based modeling and forecasting approach. Finally, no analysis has been undertaken using a component-based estimation to forecast ozone concentrations in Lima in a multi-step-ahead manner.</p></abstract>

Robustness assessment of precast concrete connections using component-based modelling

Employing highly optimised precast concrete product-based building solutions increases on-site productivity through elimination of formwork and reduction in propping as well as reducing waste, accidents and embodied carbon. The construction related benefits of precast concrete product-based building solutions are maximised by eliminating structural topping and designing connections between members for ease of assembly. A key challenge in the design of precast concrete buildings is the achievement of robustness under accidental loading. In this paper, sudden column removal is used to assess the robustness of a precast-concrete building system without structural topping. In this case, the development of an alternative load path under sudden column removal relies on the joint response. Joint behaviour is replicated using a component-based design procedure which captures localised failure modes. Robustness is evaluated using a ductility-centred approach and quantified in terms of the pseudo-static resistance. Two types of connection are considered for the provision of continuity at a critical half-lapped joint. The first is a plated connection which was designed initially to meet the tying requirements outlined in Eurocode 2. Under sudden column removal the plated connection’s deformational capacity is limited, which in turn reduces the pseudo-static resistance. An alternative bracketed coupler connection is proposed in which the ductility supply is controlled through debonding of reinforcement. The design concept for the bracketed connection is validated with test results from two full scale sub-assemblies. The experimental results are used to validate a component-based numerical model which is subsequently used to investigate the influence of boundary conditions, and debonding length on the pseudo-static resistance following sudden column loss. The paper shows that the pseudo-static resistance can be significantly enhanced by flexure and compressive membrane action. Consequently, the authors suggest that connection design in precast concrete structures without topping should be based on a realistic assessment of the ability of the structure to develop alternative load paths following instantaneous column removal rather than simplified tying rules.

Non-linear component-based modelling strategy for beam-to-column steel-concrete composite joints under seismic loads

This paper proposes and validates a computationally efficient, non-linear numerical macro-modelling strategy to predict and assess the global and local response of steel-concrete composite joints, when subjected to seismic loads. The reference numerical model takes the form of a component-based strategy in which each component can be efficiently characterized in resistance and stiffness terms, and each constituent material has a typical non-linear behaviour with hysteresis (Takeda model for concrete, Pivot model the T-stub components and kinematic model for all the other steel members). The major advantage is the possibility to adapt the same unified modelling strategy to different configurations of steel-concrete composite joints, which may differ for geometrical and mechanical configurations of constituent members, position of joint (i.e., interior or exterior), connection type (welded or bolted), etc. As shown, due to the presented modelling approach, even under simplified assumptions, a rather close agreement is generally found between the present numerical predictions and the cyclic response of a selection of three different steel-concrete composite joints of literature, which have been investigated on full-scale experimental configurations. Most importantly, the contribution of steel members and the concrete slab can be efficiently taken into account, and allow to develop the typical mechanisms in which the slab itself is involved for the complex performance of composite frames under seismic events.

Multi-Modal Side Channel Data Driven Golden-Free Detection of Software and Firmware Trojans

This study explores data-driven detection of firmware/software Trojans in embedded systems <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">without</i> golden models. We consider embedded systems such as single board computers and industrial controllers. While prior literature considers side channel based anomaly detection, this study addresses the following central question: is anomaly detection feasible when using low-fidelity simulated data without using data from a known-good (golden) system? To study this question, we use data from a simulator-based proxy as a stand-in for unavailable golden data from a known-good system. Using data generated from the simulator, one-class classifier machine learning models are applied to detect discrepancies against expected side channel signal patterns and their inter-relationships. Side channels fused for Trojan detection include multi-modal <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">side channel</i> measurement data (such as Hardware Performance Counters, processor load, temperature, and power consumption). Additionally, fuzzing is introduced to increase detectability of Trojans. To experimentally evaluate the approach, we generate low-fidelity data using a simulator implemented with a component-based model and an information bottleneck based on Gaussian stochastic models. We consider example Trojans and show that fuzzing-aided golden-free Trojan detection is feasible using simulated data as a baseline.

Component-Based Modeling for Steel and Composite Beam-Column Joints Subjected to Quasi-Static and Impact Loads under Column Removal Scenarios

Component-Based Modeling for Steel and Composite Beam-Column Joints Subjected to Quasi-Static and Impact Loads under Column Removal Scenarios

Engagement, communication and context: the success of the human-map nexus

ABSTRACT The communication paradigm was adopted by cartographic researchers in the 1960s as a means of describing and explaining the nature of mapping; the interaction amongst cartographers, their maps, and map users; and the way in which maps ‘work'. This paper considers some of the shortcomings of the cartographic communication paradigm, but proposes that enhancing it by incorporating modifications to its rigid linear and component-based model is useful, in trying to explain the success of maps in human society. The enhancements discussed all contribute to a deeper investigation of the ‘context' of cartographic activity, the model space which the traditional graphical representations of cartographic communication occupy. Thus, we present an exploration of several topics which it is felt could be incorporated into a deeper analysis of context: the concept of affordances; the role of cognition in human engagement with geospatial data and with maps; the nature of communication through the map medium; and the adoption of carto-pragmatics, a human-centred approach to map use. It is concluded that there are inherent relationships among these topics and each suffuses the total cartographic communication system, and each of its previously identified elements, therefore affecting the definition of ‘context’.

Experimental Study of the Bending Performance of Cold-Formed Steel Channel Beams Considering the Corner Hardening Effect

Cold-formed steel channel beam components are increasingly used in lightweight steel buildings owing to the high strength–weight ratio. However, the influence of cold working processes, in relation to corner regions, and how this impacts the bending behavior of channel beams lacks thorough evaluation. In the present study, a series of coupon and bending tests were conducted and the numerical simulation and analytical derivation were supplemented, aiming to investigate the bending performance of cold-formed steel channel beams considering the reinforcement effect in corner regions. The results show that the engineering stress–strain relationships of the flat and corner coupons conformed to the trilinear models with different characteristic parameters. The yielding and ultimate strengths of the corner specimens was increased by 50% and 7%, respectively, compared to the flat coupons due to the cold-bending technique. The strain distribution of the cold-formed channel beams accord with the plane section while the stresses at the corners were 35% higher than those at the flanges, indicting the different mechanical response of the flat and corner regions. The component-based model for cold-formed steel channel beams was established to exactly describe the influence of the cold-bending action, which was greatly validated by the experimental and numerical data with the errors of typical parameters less than 9%.

On the feasibility of a component-based approach to predict aerodynamic noise from high-speed train bogies

At speeds above 300 km/h, aerodynamic noise becomes a significant source of railway noise. In a high-speed train, the bogie area is one of the most important aerodynamic noise sources. To predict aerodynamic noise, semi-empirical component-based models are attractive as they allow fast and cheap calculations compared with numerical methods. Such component-based models have been previously applied successfully to predict the aerodynamic noise from train pantographs which consist of various cylinders. In this work, the feasibility is evaluated of applying them to the aerodynamic noise from bogies, which consist of various bluff bodies. To this end, noise measurements were carried out in an anechoic wind tunnel for different flow speeds using simple shapes representing various idealised components of a bogie. These results have been used to adjust the empirical constants of the component-based prediction model to enable its application to the bogie case. In addition, the noise from a 1:10 scale simplified bogie mock-up was measured and used for comparison with the prediction model. The results show good agreement between measurements and predictions for the bogie alone. When the bogie is located between two ramps approximating a simplified bogie cavity the noise is underpredicted at low frequencies. This is attributed to the noise from the cavity which is not included in the model.

Assessing the effect of using different APSIM model configurations on model outputs

The three major sources of uncertainty in crop models are model inputs, structure and parameters. Model structure is one of the major contributors to this uncertainty, however, its quantification is difficult due to limitations in controlling confounding effects from parameter and input uncertainty. The objective of this study was to quantify the contribution of structural uncertainty to the variance in model outputs produced by the Agricultural Production Systems sIMulator (APSIM). Outputs investigated were yield, irrigation requirements, partial gross margin, drainage and nitrogen (N) leaching. Eight model structures differing in choice of soil water model, crop model and irrigation model were developed within a single APSIM version (v.7.10) and tested under three contrasting environments (climate × soil) across 120 years. We quantified: (i) the model structure uncertainty (from soil water, crop and irrigation models) using analysis of variance (ANOVA) and deviation analysis; and (ii) the variability of outputs due to model structure and climate using the coefficient of variation. Confounding effects from inputs, parameters and model users were controlled. Most structural uncertainty resulted from first order effects of the choice of model components (crop model: 12.2–98.9%, irrigation model: 0–78.4%, soil water model:1–33.7%) rather than second order interactions between components (0.1–18.9%). Furthermore, uncertainty from choice of sub-model/model used was not necessarily related to the structural complexity of these components. The effects of structural uncertainty on predictions commonly used to inform agronomic, ecological or policy decision making were strongly impacted by site and climate conditions i.e., high rainfall site (∼1330 mm year−1) had less uncertainty and variability as compared to low rainfall site (∼610 mm year−1), highlighting the need for any uncertainty assessment to cover the entire range of conditions for model application. Here we show the value of a component-based modelling framework for quantifying uncertainty in crop modelling studies.

Machine learning-driven QSAR models for predicting the mixture toxicity of nanoparticles

Research on theoretical prediction methods for the mixture toxicity of engineered nanoparticles (ENPs) faces significant challenges. The application of in silico methods based on machine learning is emerging as an effective strategy to address the toxicity prediction of chemical mixtures. Herein, we combined toxicity data generated in our lab with experimental data reported in the literature to predict the combined toxicity of seven metallic ENPs for Escherichia coli at different mixing ratios (22 binary combinations). We thereafter applied two machine learning (ML) techniques, support vector machine (SVM) and neural network (NN), and compared the differences in the ability to predict the combined toxicity by means of the ML-based methods and two component-based mixture models: independent action and concentration addition. Among 72 developed quantitative structure–activity relationship (QSAR) models by the ML methods, two SVM-QSAR models and two NN-QSAR models showed good performance. Moreover, an NN-based QSAR model combined with two molecular descriptors, namely enthalpy of formation of a gaseous cation and metal oxide standard molar enthalpy of formation, showed the best predictive power for the internal dataset (R2test = 0.911, adjusted R2test = 0.733, RMSEtest = 0.091, and MAEtest = 0.067) and for the combination of internal and external datasets (R2test = 0.908, adjusted R2test = 0.871, RMSEtest = 0.255, and MAEtest = 0.181). In addition, the developed QSAR models performed better than the component-based models. The estimation of the applicability domain of the selected QSAR models showed that all the binary mixtures in training and test sets were in the applicability domain. This study approach could provide a methodological and theoretical basis for the ecological risk assessment of mixtures of ENPs.