Machine Learning Approach for Carbon Capture and Utilization - A Preliminary Investigation

Aug 3, 2026

R. R. Palkar, S. R. Palkar, and R. B. Jadeja present a combined framework that applies machine learning to Carbon Capture and Utilization (CCU) to address limitations of time-consuming and potentially inaccurate real-time laboratory experimentation. The work outlines CCU laboratory processes and details a study pipeline from system identification and carbon data generation to model development, training-validation, and optimization. The goal is faster, safer, and more efficient prediction of carbon capture behavior to mitigate environmental hazards and support climate-related industry decisions.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages3

Reading Time8 min

AccessFree to download

Introduction

Greenhouse gases and carbon emissions

Carbon capture and utilization overview

Machine learning concepts and relevance

Methodology

Combined experimental-data and ML workflow

Remarks

Prediction benefits for process industries

Data requirements and advantages

MACHINE LEARNING APPROACH FOR CARBON CAPTURE AND UTILIZATION – A PRELIMENARY INVESTIGATION R. R. Palkar, S. R. Palkar, R. B. Jadeja Marwadi university, Rajkot-360003, Gujarat, India Due to the increase in the industrialization the environment is deteriorating. The major concern is to identify the sources those are contributing to the environment change. One of the major sources of interest is carbon in this domain. The carbon capture has been carried out with different methods and data is analyzed. The process of performing real time experiments is time consuming and sometimes the accurate results may not be obtained. In order to overcome the issues mentioned, a combined approach with machine learning is presented by the authors in this article. The present work provides a detailed overview of the laboratory processes for Carbon Capture and Utilization (CCU). In addition to this a detailed investigation of machine learning along with its probable implementation is presented. The combined approach will be beneficial as it efficient, quick and safe. The proposed approach will be beneficial to the industries as well as environment. Keywords: carbon capture and utilization (CCU), machine learning, environmental hazards, climate change Introduction. The greenhouse gases (GHG) contributes to the global warming, this results from different human activities like industrialization. The major components of the GHG’s are carbon dioxide (CO2), methane (CH4), chlorofluorocarbons (CFCs) and nitrous oxide (N2O) . These GHG’s continuously contributing to the climate changes all across the globe. One of the concerns is with emission of carbon. There are several sources which contributes to the emission of CO2 viz., burning of fossil fuel, thermal power plants etc. [1] . The recent value of the carbon emission shows that it has surpassed 420 PPM, that may causes more damage to the environment [2] . It has been predicted that future global CO2 level will increase drastically if the measure have not be taken in the present. The carbon capture and utilization is promising method [3] . There are several methods used for the identification of the sources and its capture. The first step is to separate the CO2 from these sources, which pollutes the environment. The separation mechanism is preliminary operation and one of the energy intensive phases. Furthermore the techniques need advancements [4] . The process of carbon capture is studied by different research groups, the applications includes energy generation systems [5–7] . The different technologies for capturing carbon are listed in Fig. 1. Fig. 1. Carbon capture techniques The machine learning is one of the superpower the researchers have in today’s era. The importance and working of the machine learning is illustrated in the Fig. 2. The machine learning is based on the three major parts of model learning i.e. supervised, unsupervised and reinforced learning. The major application includes image classification; identify fraud detection, population growth detection, structure discovery, customer segmentation, targeted marketing etc [8] . Fig. 2. Overview of machine learning The manual working at lab scale is a tedious task. In order to achieve the better efficiency of the removal, advanced equipment’s as well as techniques needs to take care. The present work emphasize on the combined approach with machine learning for the prediction of carbon capture to mitigate the environmental hazards. Methodology. The present work highlights the use of the machine learning approach to this well-known problem. The artificial neural network approach has been studied by many researchers earlier in the domain of chemical engineering [9–13] . The steps involved in the combined studies are listed below: 1) identification of the system; 2) development of the physical experimental setup, depending upon the method applicable; 3) data generation (Carbon Capture); 4) analysis of the dat

FAQs

Why is a machine learning approach proposed for carbon capture and utilization?

Real-time laboratory experiments are time-consuming and may not always deliver accurate results. Machine learning is introduced to improve prediction efficiency and support mitigation of environmental hazards.

What are the main steps in the combined study workflow?

The workflow includes system identification, development of an experimental setup, carbon data generation, data analysis and extraction for ML, model development, training-testing-validation, optimization, and using optimized parameters for further modeling.

How does the proposed approach help industries and the environment?

By reducing risk and resource usage while enabling quicker prediction of CCU behavior, the approach supports safer and more efficient processes and helps minimize ecosystem deterioration caused by unwanted gas emissions.

Aug 3, 2026

Upload by Ava Thompson