Artificial neural network modeling of p-cresol photodegradation

Abstract

BackgroundThe complexity of reactions and kinetic is the current problem of photodegradation processes. Recently, artificial neural networks have been widely used to solve the problems because of their reliable, robust, and salient characteristics in capturing the non-linear relationships between variables in complex systems. In this study, an artificial neural network was applied for modeling p-cresol photodegradation. To optimize the network, the independent variables including irradiation time, pH, photocatalyst amount and concentration of p-cresol were used as the input parameters, while the photodegradation% was selected as output. The photodegradation% was obtained from the performance of the experimental design of the variables under UV irradiation. The network was trained by Quick propagation (QP) and the other three algorithms as a model. To determine the number of hidden layer nodes in the model, the root mean squared error of testing set was minimized. After minimizing the error, the topologies of the algorithms were compared by coefficient of determination and absolute average deviation.ResultsThe comparison indicated that the Quick propagation algorithm had minimum root mean squared error, 1.3995, absolute average deviation, 3.0478, and maximum coefficient of determination, 0.9752, for the testing data set. The validation test results of the artificial neural network based on QP indicated that the root mean squared error was 4.11, absolute average deviation was 8.071 and the maximum coefficient of determination was 0.97.ConclusionArtificial neural network based on Quick propagation algorithm with topology 4-10-1 gave the best performance in this study.

Low pressure solubilities of CO2 in guanidinium trifluoromethanesulfonate–MDEA systems

“…18 shows, the ANN model contains three variables ([gua][OTf], MDEA and CO 2 ) in input, 13 nodes in hidden and one response (CO 2 solubility) in output layers (QP-3-13-1). In the productive process, the importance of the fabrication effective variables is determined to control the excess amount and determinate the effectivity of the input variables [27]. The QP-3-13-1) model was used to determine the importance of the three effective input variables of the [gua]…”

“…[OTf]-MDEA systems for low pressure CO 2 solubility was carried out by NeuralPower software version 2.525 [27,30,50] . The concentration of [gua][OTf], MDEA and the concentration of injected CO 2 in mole were selected as independent variables (inputs); while mole of CO 2 absorbed was selected as dependent variable (output).…”

“…The complexities cause major challenges for conventional methods such as one-variable-one-time [27]. CO 2 solubility does not depend on any single parameter alone and a combination of parameters affects the absorption of CO 2.…”

Enhancement of electronic protection to reduce e-waste

“…On the other side, the multivariate methods such as response surface methodology (RSM) and artificial neural networks (ANNs) contemplate the simultaneous effect of the input variables on the output free of the mentioned complexity [40][41][42][43][44]. However, RSM involves the complicated statistical calculation of fitting process as well as the regression analysis while ANNs are free of the mathematic functionalization [16,42,45,46].…”

Scheduling the blended solution as industrial CO2 absorber in separation process by back-propagation artificial neural networks

“…In addition, the number of the experiment is increased by increasing the input variables and number of the input is limited to 32 variables [30]. On the contrary, ANNs simultaneously consider the effect of all input variables on the output response(s) free of the mention difficulties by using the universal mathematical learning algorithms which is able to industrial scale up [20,31,32]. To the best of our knowledge, the method has not been investigated yet in this field of studies.…”