Machine Learning Enables Rapid Screening of Reactive Fly Ashes Based on Their Network Topology

Abstract

Fly ash, a byproduct of coal combustion, can be used as supplementary cementitious material (SCM) to replace ordinary portland cement (OPC) in concrete. This generates revenue for coal power plant operators and also reduces the CO 2 intensity of the binder fraction of a concrete (each ton of OPC replaced by fly ash results in 0.9 ton of avoided CO 2 emissions, if the fly ash is considered to have no carbon footprint). However, the use of fly ash in concrete has thus far been limited to replacement levels less than 20 mass % due to uncertainties in their performance as SCM. Although the ability of a fly ash to replace cement in concrete is largely determined by the reactivity of its amorphous phase, characterizing fly ashes' amorphous phase is complex and cost prohibitive, which has thus far prevented any high-throughput screening of fly ashes to assess their suitability as SCMs. Here, we introduce a machine-learning-based methodology that enables robust screening of reactive fly ashes based solely on fast, inexpensive bulk characterization (X-ray fluorescence: XRF), by using the network topology of fly ashes' amorphous phase as a structural proxy for their reactivity. On the basis of a data set of more than 100 fly ashes, we train an artificial neural network (ANN) model that offers accurate predictions of the mass fraction of fly ashes' amorphous phase and the network topology thereof. This new method seeks to maximize the beneficial use of fly ashes obtained from routine production, as well as to identify opportunities for the reclamation of ashes that are presently stored in impoundments.

A Comparative Study on Supervised Machine Learning Algorithms for Copper Recovery Quality Prediction in a Leaching Process

“…The results of this study are somehow comparable to those of previous studies [12,14]. This is because they also find relations between elements of interest for their research in the mining context, using data-driven techniques such as ANN or SVM.…”

“…For example, the authors of [14] presented a study based on an ANN algorithm and focused on the use of fly ash as supplementary cementitious material, the authors of [15] described an interesting work using ANN, SVM and RF to generate predictive models for the behavior of organic solvent nanofiltration membranes used in the chemical industry (particularly in the pharmaceutical sector), and the authors of [16] compared the use of ANN, SVM, and RF in the formation of geological reservoirs, while the authors of [17] compared ANN and RF in rock drilling and blasting in a mining company. In this study, the authors concluded that the ANN-based model showed the best performance.…”

Machine learning enabled closed‐form models to predict strength of alkali‐activated systems

“…These chemostructural descriptors unify the various chemical descriptors of FAs into a singular, composite chemostructural descriptor; which, essentially, reduces the number of input variables, alleviates the complexities associated with data processing, introduces chemostructural information pertaining to the FAs (as opposed to only chemical composition) within the database; and serves as a proxy for FA's reactivity, which plays a significant role in development of strength in AAMs. Previous studies 13,19,27,47 only accounted for the amorphous phase of FAs in chemostructural descriptors. However, in AAMs, both amorphous and crystalline phases of FAs participate geopolymerization reaction and influence compressive strength.…”

Harnessing the power of machine learning for carbon capture, utilisation, and storage (CCUS) – a state-of-the-art review

“…Taffese et al 274 applied ANN, DT and ensemble methods to predict the carbonation depth with rationally low error, and the the ML models indicated that the CaPrM model can help designers to optimise the concrete mix or structural design as well as to define proactive maintenance plan. Song et al 275 developed a machine-learning-aided platform (ANNs) to enable the rapid, accurate, and high-throughput screening of fly ashes by predicting a structure-based proxy for their reactivity solely on the basis of bulk chemical composition, which can be potential to maximise the beneficial utilisation of fly ashes such as CO 2 adsorbents and construction materials.…”