Search

Wire coating is a commercial method to insulate wires for mechanical intensity and environmental protection. In this experimental study, the technique of computational intelligence is used for nonlinear wire coating analysis by soaking the wires in Oldroyd 8-constant fluid under a constant pressure gradient with the help of feed forward artificial neural networks (ANNs). The system of partial differential equations generated for the process of wire coating is transformed into a nonlinear dimensionless ordinary differential equation. One of the salient features of ANNs is the mathematical modeling of transformed equations by exploring the unsupervised error. A new scheme based on inverse multiquadric neural networks (IMQNNs) is used with a hybridization process of well-known genetic algorithms (GAs) and sequential quadratic programming (SQP) to obtain expeditious convergence, i.e., IMQNNs-GA-SQP. The applicability of the problem is investigated by altering the values of the dilatant constant, pressure gradient, shear stress, and pseudo-plastic constant, the outcome of which is in the form of varying polymer coating thickness. Comparison of highly accurate results in the shape of error analysis based on absolute errors of obtained results with those of the reference solution, calculated using the Adams numerical technique using MATHEMATICA software, statistical analysis such as root-mean-square error, Theil's coefficient for inequality, E-R2 (error function based on the coefficient of determination), E-VAF (variance accounted for), E-NSE (Nash–Sutcliffe efficiency), mean absolute error, box plot analyses, and the cumulative distributive function through histogram analyses, is also carried out in this research, which guarantees the effectiveness of the used scheme. [ABSTRACT FROM AUTHOR]

This study is based on gray-box modeling for the prediction of optimum mass flow rates of inlet streams of a Plate and Fin Heat Exchanger under uncertainty. A first principle model of the Plate and Fin Heat Exchanger was developed in Aspen Exchanger Design and Rating. Genetic algorithm was integrated with the first principle model to achieve the highest possible exit temperature of the cold process stream under uncertainty. A dataset of uncertain process conditions and their corresponding optimum inlet flow rates derived through the first principle-genetic algorithm integration was used to develop an artificial neural networks model. The artificial neural networks model was then integrated with the first principle model by replacing the genetic algorithm to form a novel gray-box framework. The proposed gray box model, i.e., artificial neural networks and first principle integration, achieved a higher effectiveness and higher outlet temperature than those derived through the straight run first principle model. The performance of the gray box framework was also comparable to the first principle integrated genetic algorithm approach and significantly minimized the computation time needed for estimating the optimum process conditions. First principle integrated genetic algorithm approach enhanced the effectiveness of the straight run model by 3.05%. Performance of the proposed gray-box model was comparable to the integrated framework of the genetic algorithm with the first principle model but was significantly faster. The developed artificial neural networks model was employed as a surrogate in Sobol and FAST sensitivity analysis framework to identify the impact of input variables on output variables. The proposed gray box based method enhanced the capability of the plate and fin heat exchanger to recover energy from the process stream and its robustness to cope with uncertainty. The proposed approach is suitable for real-time application and would contribute to laying a foundation for petroleum refinery 4.0. • A Gray-box (GB) modeling framework was developed for the optimization of process conditions of a plate and fin heat exchanger of a tile industry under uncertainty. • The GB framework was based on the integration of first principle (FP) model and Artificial Neural Networks (ANN) model. • The proposed GB approach attained the optimum conditions in a significantly less period of time than the conventional FP-GA framework. [ABSTRACT FROM AUTHOR]

The current study explores a new direction of research in which the magnetohydrodynamic (MHD) effects on two-dimensional nanofluid boundary layer flow under the influence of radiation are investigated in a porous medium using inverse multiquadric (IMQ) radial basis neural networks (RBNNs) i.e. IMQ-RBNNs. A quite new computational approach is used here in which artificial neural networks (ANNs) acquired IMQ radial basis function as an activation function in an amalgam of globally searching solver named genetic algorithms (GAs) together with an efficient local searching sequential quadratic programming (SQP) solver. The basic equations that represent the fluid flow are first obtained in the form of partial differential equations (PDEs) and then converted into dimensionless ordinary differential equations (ODEs) by applying similarity transformations. The dynamical properties of the nanofluid including velocity, temperature along mass concentration are analyzed by varying physical parameters including radiation parameter, magnetic parameter, constant moving parameter, heat source parameter, Brownian motion parameter and Eckert number. Moreover, the escalation in the value of the constant moving parameter diminishes the fluid velocity but this effect is reversed in the case of magnetic parameter. The increasing values of heat source parameter, Brownian motion parameter and Eckert number inflate the nanofluid temperature but a reversal effect is observed in the case of radiation parameter. It is also observed that larger values of heat Brownian motion parameter and radiation parameter amplify the nanofluid concentration but an opposite effect is obtained in the case of heat source parameter. The numerical results of the proposed model obtained through the IMQ-RBNNs based solver optimized with GASQP algorithm. The efficacy and tendency of rapid convergence of the proposed solver are endorsed through graphical illustrations and tables using well-known statistical operators with various fitness analyses on MATLAB are also part of this research. [ABSTRACT FROM AUTHOR]

Conventional fuels for power systems are now plaguing the environment more than ever before. Efforts to curb the environmental impacts of fossil fuels are now driving modern microgrids (MGs) to utilize green energy sources. In this regard, DC microgrids are well known for their versatility. Considering the rising significance of DC-MGs in future, this paper presents an aggregated DC microgrid model comprising of wind resource, photovoltaics, fuel cell, battery, and a supercapacitor. The model also includes DC–DC power converters and a two-level control structure. Firstly, the high-level of the proposed control structure constitutes an artificial neural network for maximum power point tracking and an intelligent energy management powered by a fuzzy logic controller. At this control level, the unified operation of the fuel cell and storage components under various dynamic conditions is ensured. Furthermore, the low-level control structure is based on an adaptive terminal sliding mode non-linear controller that derives the control signal for the converters. Finally, results showed that the energy management system along with the adaptive terminal sliding mode control successfully regulated the DC bus voltage by keeping the sources within their operational constraints. Moreover, the proposed non-linear controller demonstrated model's robustness, successfully converged the sliding surface to zero in finite time, ensured global asymptotic stability and effectively dealt with the model uncertainties. All the simulation leading to these results were performed in MATLAB/Simulink and further evaluation was carried out by hardware in loop experiments using C2000 Delfino microcontroller F28397D launchpad. • An optimum design of hybrid DC microgrid integrated with two renewable energy resources, an auxiliary source and hybrid energy storage system. • Multi-level control based on ANN, fuzzy Logic and adaptive terminal sliding mode non-linear controller for power balance and control. • HIL verification of controller performance using Delfino Microcontroller F28397D launchpad. [ABSTRACT FROM AUTHOR]