skip to main content

Explore Research Products in the NSF-PAR

Title: Machine learning can predict setting behavior and strength evolution of hydrating cement systems
Abstract

Setting and strength development of ordinary Portland cement (OPC) binders involves multiple interacting chemical reactions, resulting in the formation of a solid microstructure. A long‐standing yet elusive goal has been to establish a basis for the prediction of the properties and performance of concrete using knowledge of the chemical and physical attributes of its components—PC, sand, stone, water, and chemical admixtures—together with the environmental conditions under which they react. Machine learning (ML) provides adata‐drivenbasis for the estimation of properties, and has recently been applied to estimate the 28 days (compressive) strength of concrete from knowledge of its mixture proportions (Young et al,Cem Concr Res, 2019,115:379). Building on this success, the current work uses a diverse dataset of ASTM C150 cements, the chemical composition and other attributes of which have been measured. ML estimators were trained with this dataset to estimate both paste setting time and mortar strength development. The ML estimation errors are typically similar to the measurement repeatability of the relevant ASTM test methods, and are thus able to account for the influence of binder composition and fineness. This creates new opportunities to apply data intensive methods to optimize concrete formulations under multiple constraints of cost, CO2impact, and performance attributes.

  more » « less
NSF-PAR ID:
10447418
Author(s) / Creator(s):
 ;  ;  ;  
Publisher / Repository:
Wiley-Blackwell
Date Published:
Journal Name:
Journal of the American Ceramic Society
Volume:
103
Issue:
1
ISSN:
0002-7820
Page Range / eLocation ID:
p. 480-490
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; ; ; ; ( , ACI Materials Journal)
    null (Ed.)
    The properties of concretes are controlled by the rate of reaction of their precursors, the chemical composition of the binding phase(s), and their structure at different scales. However, the complex and multiscale structure of the cementitious hydrates and the dissimilar rates of numerous chemical reactions make it challenging to eluci- date such linkages. In particular, reliable predictions of strength development in concretes remain unavailable. As an alternative route to physics- or chemistry-based models, machine learning (ML) offers a means to develop powerful predictive models for materials using existing data. Here, it is shown that ML models can be used to accurately predict concrete’s compressive strength at 28 days. This approach relies on the analysis of a large data set (>10,000 observations) of measured compressive strengths for industrially produced concretes, based on knowledge of their mixture proportions. It is demonstrated that these models can readily predict the 28-day compressive strength of any concrete based merely on the knowledge of the mixture proportions with an accuracy of approximately ±4.4 MPa (as captured by the root- mean-square error). By comparing the performance of select ML algorithms, the balance between accuracy, simplicity, and inter- pretability in ML approaches is discussed. 
    more » « less
  2. ; ; ; ; ; ; ; ( , Journal of the American Ceramic Society)
    Abstract

    Fly ash, an aluminosilicate composite consisting of disordered (major) and crystalline (minor) compounds, is a low‐carbon alternative that can partially replace ordinary portland cement (OPC) in the binder fraction of concrete. Therefore, understanding the reactivity of fly ash in the hyperalkaline conditions prevalent in concrete is critical to predicting concrete's performance; including setting and strength gain. Herein, temporal measurements of the solution composition (using inductively coupled plasma‐optical emission spectrometry: ICP‐OES) are used to assess the aqueous dissolution rate of monophasic synthetic aluminosilicate glasses analogous to those present in technical fly ashes, under hyperalkaline conditions (10 ≤ pH ≤ 13) across a range of temperatures (25°C ≤ T≤45°C). The dissolution rate is shown to depend on the average number of topological constraints per atom within the glass network (nc, unitless), but this dependence weakens with increasing pH (>10). This is postulated to be on account of: (a) time‐dependent changes in the glass’ surface structure, that is, the number of topological constraints; and/or (b) a change in the dissolution mechanism (eg from network hydrolysis to transport control). The results indicate that the topological description of glass dissolution is most rigorously valid only at very short reaction times (ie at high undersaturations), especially under conditions of hyperalkalinity. These findings provide an improved basis to understand the underlying factors that affect the initial and ongoing reactivity of aluminosilicate glasses such as fly ash in changing chemical environments, for example, when such materials are utilized in cementitious composites.

     
    more » « less
  3. ; ; ; ; ; ; ( , Scientific Reports)
    Abstract

    Carbonaceous (e.g., limestone) and aluminosilicate (e.g., calcined clay) mineral additives are routinely used to partially replace ordinary portland cement in concrete to alleviate its energy impact and carbon footprint. These mineral additives—depending on their physicochemical characteristics—alter the hydration behavior of cement; which, in turn, affects the evolution of microstructure of concrete, as well as the development of its properties (e.g., compressive strength). Numerical, reaction-kinetics models—e.g., phase boundary nucleation-and-growth models; which are based partly on theoretically-derived kinetic mechanisms, and partly on assumptions—are unable to produce a priori prediction of hydration kinetics of cement; especially in multicomponent systems, wherein chemical interactions among cement, water, and mineral additives occur concurrently. This paper introduces a machine learning-based methodology to enable prompt and high-fidelity prediction of time-dependent hydration kinetics of cement, both in plain and multicomponent (e.g., binary; and ternary) systems, using the system’s physicochemical characteristics as inputs. Based on a database comprising hydration kinetics profiles of 235 unique systems—encompassing 7 synthetic cements and three mineral additives with disparate physicochemical attributes—a random forests (RF) model was rigorously trained to establish the underlying composition-reactivity correlations. This training was subsequently leveraged by the RF model: to predict time-dependent hydration kinetics of cement in new, multicomponent systems; and to formulate optimal mixture designs that satisfy user-imposed kinetics criteria.

     
    more » « less
  4. ; ; ; ( , Sustainable Structures)

    Sheet piles are interlocked segments used for temporary or permanent soil and water retaining structures such as below-grade parking structures and sea walls. Although steel is commonly used due to its strength and ease of manufacturing, it rusts in saltwater. Fiber reinforced polymer (FRP) composite sheet piles are resistant to chlorides and have higher corrosion resistance, but their mechanical properties vary in length and width. Stress risers at corrugation corners make soil-structure interaction a challenging design issue. This research aims to develop a standardized test procedure to determine the resisting moment capacity of FRP composite sheet piles. Cantilevered FRP sheet piles fixed with a sand-concrete mixture of ~70 psi (0.48 MPa) compressive strength were tested under static loads. Strain gages and LVDTs were used to collect data on deformation response up to and beyond peak induced stress. Results suggest that the refined test procedure can assist engineers in designing efficient sheet pile structures and become a basis to develop ASTM standard.

     
    more » « less
  5. ; ; ; ; ; ; ( , Cryoletters)

    BACKGROUND:Pectins have unique properties and great potential to become an indispensable component of cryoprotective environment for platelet freezing.OBJECTIVE:To investigate the possibility of including pectins (apple pectin AU-701, tanacetan) into the composition of a cryoprotective solution for platelets during low-temperature storage.MATERIALS AND METHODS:Samples of platelet concentrates (PC) were frozen under the protection of complex solutions and stored in an electric freezer at -80 °C for 1 and 6 months.RESULT:The study showed that of the basic cryoprotectants, the best effect in the preservation of PC was with dimethylacetamide (DMAC). The use of pectins as an additive to the base solution of DMAC statistically improves the preservation of PC after exposure to low temperatures (-80 ° C) for 30 and 180 days.CONCLUSIONS:We conclude that DMAC is more promising as a basis for the development of a new combined cryoprotectant for PC freezing. Moreover, the chemical structure of pectin determines the level of its cryoprotective action in relation to the preservation of PC.

     
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email