Estimation of the thermal conductivity λ(T,P) of ionic liquids using a neural network optimized with genetic algorithms

Abstract

In this study, an artificial neural network was optimized using a genetic algorithm in order to estimate the thermal conductivity of ionic liquids at different temperatures and pressures. Experimental thermal conductivity data of 41 ionic liquids (400 experimental data points) in the range from 0.10 to 0.22 W m À1 K À1 were used to obtain the proposed method for the temperature range of 273e390 K and the pressure range of 100 e20,000 kPa. In addition, the molecular mass M and structure of molecules, represented by the number of well-defined groups forming the molecule, were provided as input parameters in order to characterize the different molecules of ionic liquids. A heterogeneous set of ionic liquids includes cations such as imidazolium, ammonium, phosphonium, pyrrolidinium, and pyridinium. It also includes anions such as halides, sulfonates, tosylates, imides, borates, phosphates, acetates, and amino acids. The whole dataset was divided into a training set with 300 experimental data points and a prediction set with 100 experimental data points. Several architectures were studied, and the optimum weights for the network were determined. The results showed that the proposed method to estimate the thermal conductivity of ionic liquids at different temperatures and pressures presented a good accuracy with lower deviations such as AARD less than 0.91% and R 2 of 0.9969 for the training set, and AARD less than 0.84% with R 2 of 0.9963 for the prediction set.

Thermal Conductivity of Ionic Liquids

“…Therefore, the prediction of the thermal conductivity of an IL based on its structure and other physical properties would be very useful for designing novel ILs. So far, an empirical-prediction method [13,20,21], group-contribution method [22][23][24], quantitative structure-property relationship method [25], prediction Thermal Conductivity of Ionic Liquids http://dx.doi.org/10.5772/intechopen.76559 method using a neural network [26,27], and many other methods [28][29][30][31][32][33][34][35] have been proposed for the prediction of the thermal conductivities of ILs. In this section, the empirical-prediction method based on other physical properties and prediction method using group-contribution method are introduced.…”

GA-optimized neural network for forecasting the geomagnetic storm index

“…This ANN consists of one input layer with N inputs, one hidden layer with q units and one output layer with n outputs. The output of this model can be expressed as (Lazzús, 2016):…”

“…The algorithm repeatedly modifies a population of individual solutions into a search space by relying on bioinspired operators such as mutation, crossover, and selection (Davis, 1991). Due to facts, GA may offer significant benefits over the more typical search of optimization algorithms, and it can be used to optimize the update weights process of an ANN with better results than the traditional back-propagation algorithm (Lazzús, 2016). With this, we propose an improved method to forecast the Dst variation based on measurements at ground level.…”

Thermal Conductivity Enhancement of Aqueous Ionic Liquid and Nanoparticle Suspension

“…However, in the case of heat transfer applications, thermal conductivity is an essential property. Its understanding is a pre-requisite for evaluating the heat transfer coefficient, which is required for the design of heat transfer equipment and for selection of heat transfer fluid (Lazzús, 2015a). The experimental data of this property is available for only a limited number of ionic liquids.…”

“…The experimental data of this property is available for only a limited number of ionic liquids. Various authors have proposed correlations and predictive models for estimation of thermal conductivity (Albert and Muller, 2014;Carrete et al, 2012;Chen et al, 2014;Hezave et al, 2012;Hosseini et al, 2016;Lazzús, 2015aLazzús, , 2015bLazzús and Pulgar-Villarroel, 2015;Shojaee et al, 2013).…”