Digital Chemical Engineering最新文献

This paper introduces BibMon, a Python package that provides predictive models for data-driven fault detection and diagnosis, soft sensing, and process condition monitoring. Key features include regression and reconstruction models, preprocessing pipelines, alarms, and visualization through control charts and diagnostic maps. BibMon also includes real and simulated datasets for benchmarking, comparative performance analysis of different models, and hyperparameter tuning. The package is designed to be highly extensible, allowing for easy integration of new models and methodologies through its object-oriented implementation. Currently, BibMon is in production at Petrobras, a major player in the energy industry, monitoring numerous industrial assets and enabling real-time detection and diagnosis of equipment and process faults. The software is open source and available at: https://github.com/petrobras/bibmon.

The incomplete combustion of fossil fuels results in the emission of soot, a carbonaceous, solid fine powder that causes harm to human health and the environment. This study compares multiple linear regression (MLR) with three different machine learning (ML) models for predicting the threshold sooting index (TSI), a commonly employed index for measuring the sooting propensity of fuels. The dataset used for model development consists of experimental TSI data for 342 fuels, including various chemical classes, including oxygenated components like ethers and alcohols. Ten input features were employed, comprising eight functionalities, molecular weight, and the branching index (BI). These parameters used as input features have been demonstrated to affect fuels' physical and thermochemical properties. The ML models employed in this study are support vector regression with Nu parameter (NuSVR), extra trees regression (ETR), and extreme gradient boosting regression (XGBR). The models were trained, validated, and tested using randomly split datasets, with 56 % for training, 14 % for validation, and 30 % for testing. The accuracy of the MLR, NuSVR, ETR, and XGBR models for the entire dataset was 91 %, 96 %, 98 %, and 96 %, respectively. The mean absolute errors (MAE) of prediction were 3.4, 0.022, 0.011, and 0.028 for MLR, NuSVR, ETR, and XGBR respectively. These results highlight the effectiveness of the ML models in making predictions, with error levels similar to the uncertainties observed in experimental measurements. The developed ML models have been validated to ensure generalizability and can be used to predict petroleum fuels' TSI.

Whey is a very perishable food and a potent source of protein. It might be more commercialized through the process of spray drying, which would enhance its shelf life. No CFD computational models applied to the drying of fresh sweet whey have been reported in the literature to investigate transport phenomena in spray dryers and to predict crucial design parameters such as deposition, powder recovery, and drying efficiency. Using an Eulerian-Langragean technique, the behavior of multiphase flow as well as heat and mass transfer were investigated. The influence of the inlet air temperature was evaluated in relation to the velocity profiles of the continuous and discrete phases, the residence time distribution, the particle diameter formed, the air temperature near the injection zone, the particle temperature, and the mass fraction of evaporated water. Furthermore, performance parameters such as powder recovery, particle deposition, and drying efficiency were projected for varied air inlet temperatures. The velocity, residence time, and particle size distribution patterns were similar for the different air inlet temperatures. Greater variations might be noticed in the injection zone, especially in the temperature and mass fraction profiles. The lowest drying temperature resulted in the lowest particle deposition and the best thermal efficiency, making it the optimal process condition. This investigation can serve as a benchmark for the design of optimized spray dryers with greater thermal efficiency and yield to produce powdered whey.

In previous studies, the comprehensive scaling-up of nickel electroforming on a lab-scale rotating disk electrode (RDE) suggested that secondary current distribution could adequately simulate such a forming process. In this work, the use of a 3-D, time-dependent, secondary current distribution model, developed in COMSOL Multiphysics®, was examined to validate the nickel electroforming of an industrial mechanical vane, a low-tolerance part with a demanding thickness profile of great interest to the aerospace industry. A set of experiments were carried out in an industrial pilot tank with computations showing that the model can satisfactorily predict the experimental findings. In addition, these experiments revealed that the local applied current density was related to the surface appearance (shiny vs matt) of the electroform.

Simulations of the process at applied current densities $\le 5A/d{m}^2$ satisfactorily predicted the experimentally observed thickness distribution while, simulations of the process at applied current densities $\ge 5A/d{m}^2$ underpredicted the experimentally achieved thicknesses. Nevertheless, it is proposed that the model can be used for either quantitative or qualitative studies, respectively, depending on the required operating current density on a case-by-case basis. Scanning electron microscopy was used to determine the microstructure of the electroforms and determine the purity of nickel (i.e., if nickel oxide is formed), with imaging suggesting that pyramid-shaped nickel particles evolve during deposition. Another interesting observation revealed a periodicity in the deposit's growth mechanism which leads to “necklace”-like deposit layers at the areas where the electroforms presented the highest thickness.

The calculation of chemical equilibria in detailed reactor simulations frequently requires elaborate numerical solution of the governing equations in an iterative way, which is often computationally expensive and can significantly increase the overall computation time. In order to reduce these computational costs, we introduce a ready-to-use tool, ^ANNH₃, for calculation of equilibrium composition for synthesis and cracking of ammonia based on a neural network. This tool provides excellent agreement with the conventional approach in the range of 135–1000 °C and 1–100 bar and is ca. 100 times faster than conventional stoichiometry-based concepts by replacing the iterative solution process with neural network inference. While speed-up is significant even for the relatively simple case of ammonia synthesis and decomposition, we expect an even higher performance gain for the equilibrium calculation in reaction systems where more components and multiple reactions are involved.

The simultaneous flow of gas, oil & water is frequently encountered in pipelines during upstream petroleum operations. The multiphase flow results in different types of flow patterns based on the flow rates of fluids, physical properties and geometry of the flow domain. The flow behavior is characterized based on the governing flow patterns. Hence, the information about the flow patterns, regime maps and resulting pressure loss are important for multiphase flow system design and optimization. The current work is focused on construction of gas, oil and water, three-phase flow regime maps and developing pressure loss prediction correlations for the flow through vertical riser downstream 90° bend. The pipe internal diameter (ID) is 6 inch and the bending radius to pipe diameter ratio is 1. The observed gas-liquid flow patterns are slug, churn, and semi-annular churn flow at the given range of superficial velocities of fluids. The flow pattern data has been used to construct flow regime maps to analyze the variation in flow patterns with flow rates of fluids and compared with the available works in the literature. In addition, the change in pressure loss with respect to flow patterns has been analyzed. Previous models are used for the prediction of pressure loss. However, according to the assessment, the models underpredicted the pressure loss. Based on three-phase pressure loss data, multiple linear regression analysis has been carried out to propose new correlations for pressure loss prediction. Comparison of the calculated and experimental data showed good agreement between the results. The knowledge of flow regime variation and pressure loss correlations can help flow assurance engineers in designing and optimization of multiphase flow systems.

Particle mixing is a crucial operation in various industrial production processes. However, phenomena like segregation or local accumulation can arise, especially when particles differ in properties like radius and density. Numerical simulation of particles using Discrete Element Method (DEM) allows for the manipulation of control variables in batches, generating a large amount of data and facilitating quantitative research. In this study, the mixing behaviors of binary particles in rotating drums are systematically investigated. The DEM model is first validated with experimental data and then rotating drums with varying obstacles, rotation speeds, particle radii, and densities are simulated. Moreover, a Gaussian process-based optimization is conducted by correlating Lacey mixing index (MI) and parameterized shape of obstacle to find the optimized mixing condition. Experimental validations are further performed on the optimized condition to verify the design. It is shown that this integrated approach holds significant potential for enhancing the efficiency, effectiveness of industrial mixing processes and the consideration of energy consumption when balancing the mixing efficiency and optimal rotating speed.