Modeling based on machine learning to investigate flue gas desulfurization performance by calcium silicate absorbent in a sand bed reactor

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Янв. 10, 2024

Abstract Flue gas desulfurization (FGD) is a critical process for reducing sulfur dioxide (SO 2 ) emissions from industrial sources, particularly power plants. This research uses calcium silicate absorbent in combination with machine learning (ML) to predict SO concentration within an FGD process. The collected dataset encompasses four input parameters, specifically relative humidity, weight, temperature, and time, incorporates one output parameter, which pertains the of . Six ML models were developed estimate parameters. Statistical metrics such as coefficient determination (R mean squared error (MSE) employed identify most suitable model assess its fitting effectiveness. random forest (RF) emerged top-performing model, boasting R 0.9902 MSE 0.0008. model's predictions aligned closely experimental results, confirming high accuracy. hyperparameter values RF found be 74 n_estimators, 41 max_depth, false bootstrap, sqrt max_features, 1 min_samples_leaf, absolute_error criterion, 3 min_samples_split. Three-dimensional surface plots generated explore impact variables on concentration. Global sensitivity analysis (GSA) revealed weight time significantly influence integration into modeling offers novel approach optimizing efficiency effectiveness this environmentally crucial

Язык: Английский

---

Enhanced carbon dioxide adsorption using lignin-derived and nitrogen-doped porous carbons: A machine learning approaches, RSM and isotherm modeling

Case Studies in Chemical and Environmental Engineering, Год журнала: 2024, Номер 9, С. 100668 - 100668

Опубликована: Фев. 24, 2024

Язык: Английский

---

Comprehensive investigation of isotherm, RSM, and ANN modeling of CO2 capture by multi-walled carbon nanotube

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Март 1, 2024

Abstract Chemical vapor deposition was used to produce multi-walled carbon nanotubes (MWCNTs), which were modified by Fe–Ni/AC catalysts enhance CO 2 adsorption. In this study, a new realm of possibilities and potential advancements in capture technology is unveiled through the unique combination cutting-edge modeling techniques utilization recently synthesized catalyst adsorbent. SEM, BET, FTIR analyze their structure morphology. The surface area MWCNT found be 240 m /g, but after modification, it reduced 11 /g. showed increased adsorption capacity with higher pressure lower temperature, due introduction sites favorable interactions at temperatures. At 25 °C 10 bar, reached maximum 424.08 mg/g. optimal values pressure, time, temperature parameters achieved 7 2646 S 313 K. Freundlich Hill models had highest correlation experimental data. Second-Order Fractional Order kinetic fit results well. process exothermic spontaneous. has for efficient gas fields like storage or separation. regenerated M-MWCNT adsorbent demonstrated ability reused multiple times process, as evidenced study. feed-forward MLP artificial neural network model created using back-propagation training approach predict most suitable structure, selected optimization, consisted two hidden layers neurons, respectively. This trained Levenberg–Marquardt backpropagation algorithm. An created, minimum MSE performance 0.0004247 an R value 0.99904, indicating its accuracy. experiment also utilized blank spreadsheet design within framework response methodology proximity between Predicted 0.8899 Adjusted 0.9016, difference less than 0.2, indicates high level similarity. suggests that exceptionally reliable future observations, highlighting robustness.

Язык: Английский

---

Modeling of CO2 solubility and partial pressure in blended diisopropanolamine and 2-amino-2-methylpropanol solutions via response surface methodology and artificial neural network

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Янв. 13, 2025

In this study, Response Surface Methodology (RSM) and Artificial Neural Networks (ANN) were developed to estimate the equilibrium solubility partial pressure of CO2 in blended aqueous solutions diisopropanolamine (DIPA) 2-amino-2-methylpropanol (AMP). several key parameters analyzed understand behavior DIPA/AMP system for capture. Including DIPA (9–21 wt%), AMP temperature (323.15–358.15 K), (2.140–332 kPa) (0.0531–0.8796 mol/mole). The results RSM analysis indicate that model demonstrates a strong fit, as evidenced by Pred-R² 0.9601, an adjusted R² 0.9481, highly significant F-value 80.22. high predicted 0.9601 0.9292 values predictor variables can explain substantial amount variability response variable. multilayer perceptron (MLP) architecture demonstrated correlation capabilities, featuring one hidden layer with 10 5 neurons, respectively. Its topology was structured 4-10-1 predicting 4-5-1 pressure. accuracy predictions notably high, coefficients determination 0.99581 0.99839 pressure, achieved using Levenberg-Marquardt algorithm. Upon further analysis, it concluded MLP exhibited lowest error rates, mean square errors 0.00009085 0.00316632 findings emphasized not only outperformed but also greater adaptability handling intricate associated capture technologies.

Язык: Английский

---

Prediction of methane hydrate equilibrium in saline water solutions based on support vector machine and decision tree techniques

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Апрель 5, 2025

The formation of clathrate hydrates offers a powerful approach for separating gaseous substances, desalinating seawater, and energy storage at low temperatures. On the other hand, this phenomenon may lead to practical challenges, including blockage pipelines, in some industries. Consequently, accurately predicting equilibrium conditions hydrate is crucial. This study was undertaken design reliable models capable state methane saline water solutions. A comprehensive collection measured data, consisting 1051 samples, assembled from published sources. prepared databank encompassed temperature (HFTM) presence 26 different machine learning modeling through implementation Decision Tree (DT) Support Vector Machine (SVM) approaches. While both had excellent performance, latter achieved higher accuracy estimating HFTM with mean absolute percentage error (MAPE) 0.26%, standard deviation (SD) 0.78% validation process. Furthermore, more than 90% values predicted by novel fell within [Formula: see text]1% bound. It found that intelligent also favorably describe physical variations operational factors. An examination using William's plot acknowledged truthfulness gathered data suggested estimation techniques. Ultimately, order significance factors governing clarified sensitivity analysis.

Язык: Английский

---

Modeling of carbon dioxide absorption into aqueous alkanolamines using machine learning and response surface methodology

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Окт. 14, 2024

This research focuses on modeling CO

Язык: Английский

---

Mixed MDEA-PZ amine solutions for CO2 capture: Modeling and optimization using RSM and ANN approaches

Case Studies in Chemical and Environmental Engineering, Год журнала: 2023, Номер 8, С. 100509 - 100509

Опубликована: Окт. 5, 2023

The process of capturing and reducing carbon dioxide (CO2) emissions through chemical absorption is widely acknowledged as the most effective technique, especially in dealing with natural gas streams or flue gases produced by fossil fuel power plants. In this research, we delve into modeling optimization CO2 mass transfer flux (NCO₂). To accomplish this, employed a combination Piperazine (PZ) Methyldiethanolamine (MDEA) amines for absorption. approach utilized artificial neural networks (ANN) response surface methodology (RSM). We used Pi-Buckingham theory to derive dimensionless numbers input variables both ANNs RSM. resulting models offer satisfactory outcomes effectively influence independent their interactions on objective function, thereby optimizing capture process. RSM employs quadratic model. Through optimization, were fine-tuned achieve lowest error closest fit experimental data. Both demonstrated acceptable performance predicting data, maximum R2 values 0.99924 0.9663, respectively. Considering mean squared 5.2 × 10−4 obtained from simulations, ANN recommended preferred method developing simulation models.

Язык: Английский

---

Analysis of effective area and mass transfer in a structure packing column using machine learning and response surface methodology

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Авг. 24, 2024

The study examined mass transfer coefficients in a structured CO2 absorption column using machine learning (ML) and response surface methodology (RSM). Three correlations for the fractional effective area (af), gas phase coefficient (kG), liquid (kL) were derived with of determination (R2) values 0.9717, 0.9907 0.9323, respectively. To develop these correlations, four characteristics packings, including packing (ap), corrugation angle (θ), channel base (B), crimp height (h), used. ML used five models, represented as random forest (RF), radial basis function neural network (RBF), multilayer perceptron (MLP), XGB Regressor, Extra Trees Regressor (ETR), best models being (RBF) af (R2 = 0.9813, MSE 0.00088), RBF kG 0.9933, 0.00056), (MLP) kL 0.9871, 0.00089). had most impact on kL, while affected most. Although RSM method produced adequate equations each output variable good predictability, provides superior modeling capabilities.

Язык: Английский

---

Analysis of CO2 solubility in ionic liquids as promising absorbents using response surface methodology and machine learning

Journal of CO2 Utilization, Год журнала: 2025, Номер 93, С. 103043 - 103043

Опубликована: Фев. 28, 2025

Язык: Английский

---

Effect of Diethanolamine after Carbon Dioxide Absorption and Desorption on Mechanical Strength of Cement Mortar and Mechanism

Langmuir, Год журнала: 2025, Номер unknown

Опубликована: Март 12, 2025

Diethanolamine (DEA) can be used not only as a cement admixture but also to capture carbon dioxide (CO2). However, the waste liquid treatment still faces problems of high energy consumption and increasing environmental burden. The effects DEA (WL-DEA) with multiple cycles CO2 absorption desorption on setting time, hydration temperature, mechanical strength, microstructure cement-based materials were explored. It was found that adding WL-DEA could significantly reduce time enhance strength. This improvement mainly attributed two aspects: one hand, alcoholamine itself boost hydration, which accelerate generation products such AFt/AFm, CH, C–A–S–H, C–S–H, thereby improving early strength refining products; other contained little CO32– HCO3–, reacted Ca2+ produce CaCO3. above reactions cooperated complexation effect further promote optimize densification products. application in effectively their recycling liquid, provided new method for promoting development circular economy.

Язык: Английский

---