Energetic, exergetic analysis and machine learning of methane chlorination process for methyl chloride production

Abstract

Nowadays, with the growing demand for energy and effective utilization of various available sources with the exorable techniques and approaches to maximize the efficiency of energy systems. This work has developed the synthesis of Methyl chloride (MC) from the methane chlorination process using the ASPEN HYSYS simulation tool. A Searchable analysis has been done on thermodynamic derivatives (likely Energy, Exergy) to probation on the entire process. This analysis calculates all process components’ energy loss, destruction and energy, and exergy efficiencies. A heavier energy loss has been found at Reactor (ERV) with 1785.5 kW and exergy destruction of 18.8% share. Heat Exchanger Network (HEN) has energy loss (960.32kW) & exergy destruction (791.29kW). The proposed new retrofit sustainable model recovered the waste heat from the HEN and achieved energy efficiency of 87.6% and exergy efficiency of 87.3% of the total MC process. Four Machine learning models were developed for the reactor (ERV) process to predict exergy destruction. The artificial Neural network (ANN) gave good testing predictions, followed by the Random Forest (RF) with a determination coefficient (R2) of 0.999957 and 0.999981.

Assessing Exergy Efficiency in Computer-Aided Modeled Large-Scale Production of Chitosan Microbeads Modified with Thiourea and Magnetite Nanoparticles

“…Notably, the top-stage condenser emerged as the component with the most significant irreversibility (accounting for 45% of the total), closely followed by the methanol/DME synthesis reactor (24%) [47]. In the synthesis of Methyl chlorid an exergy efficiency of 87.3% was reached, in which the reactor had the highest contribution to exergy destruction, followed by the heat exchange network [48]. The recovery of low-emission alcohol from wastewater showed losses in the conventional extractive distillation process (CEDP), heat pump-assisted extractive distillation process (HPEDP), thermally coupled extractive distillation processes (TCEDP), and heat pump-assisted thermally coupled extractive distillation process (HPTCEDP) of 64.54%, 51.69%, 56.77%, and 46.04%, respectively.…”

Influence of quality of work life on psychological capital of organizational leaders using artificial neural networks: a study on learning in hybrid workplace

“…The data set was split into a training set (70%) and a testing set (30%). The scikit-learn library was used to construct the ANN model, and the hyperparameters were tuned (Gollangi & Nagamalleswara Rao, 2022). The performance of the ANN was evaluated with the help of a test dataset using the The topology of the neural network for this study included input parameters such as working conditions, social integration, opportunities, work and space for recreation and remuneration and compensation.…”

“…The data set was split into a training set (70%) and a testing set (30%). The scikit-learn library was used to construct the ANN model, and the hyperparameters were tuned (Gollangi & Nagamalleswara Rao, 2022). The performance of the ANN was evaluated with the help of a test dataset using the coefficient of determination ( R 2 ), mean square error (MSE) and mean absolute error (MAE).…”

Threat and Risk Analysis-Based Neural Network for a Chemical Explosion (TRANCE) Model to Predict Hazards in Petroleum Refinery

“…The performance of the TRANCE model can also be evaluated from another essential parameter called the coefficient of determination (R 2 ). The mathematical expressions for MSE and R 2 are represented as follows [33]: A TRANCE-based model is developed by utilizing preprocessed industrial dataset of refinery plant failure scenarios. The dataset preprocessing is performed using Matlab 2022a with a computational time of 30 min on a 5 Core computer with 8 GB RAM.…”