Intercriteria analysis (ICA) is a new method, which is based on the concepts of index matrices and intuitionistic fuzzy sets, aiming at detection of possible correlations between pairs of criteria, expressed as coefficients of the positive and negative consonance between each pair of criteria. Here, the proposed method is applied to study the behavior of one type of neural networks, the modular neural networks (MNN), that combine several simple neural models for simplifying a solution to a complex problem. They are a tool that can be used for object recognition and identification. Usually the inputs of the MNN can be fed with independent data. However, there are certain limits when we may use MNN, and the number of the neurons is one of the major parameters during the implementation of the MNN. On the other hand, a high number of neurons can slow down the learning process, which is not desired. In this paper, we propose a method for removing part of the inputs and, hence, the neurons, which in addition leads to a decrease of the error between the desired goal value and the real value obtained on the output of the MNN. In the research work reported here the authors have applied the ICA method to the data from real datasets with measurements of crude oil probes, glass, and iris plant. The method can also be used to assess the independence of data with good results.

The approach of InterCriteria Analysis (ICA) was applied for the aim of reducing the set of variables on the input of a neural network, taking into account the fact that their large number increases the number of neurons in the network, thus making them unusable for hardware implementation. Here, for the first time, with the help of the ICA method, correlations between triples of the input parameters for training of the neural networks were obtained. In this case, we use the approach of ICA for data preprocessing, which may yield reduction of the total time for training the neural networks, hence, the time for the network’s processing of data and images.

The concept of an intuitionistic fuzzy deep neural network (IFDNN) is introduced here as a demonstration of a combined use of artificial neural networks and intuitionistic fuzzy sets, aiming to benefit from the advantages of both methods. The investigation presents in a methodological way the whole process of IFDNN development, starting with the simplest form—an intuitionistic fuzzy neural network (IFNN) with one layer with single-input neuron, passing through IFNN with one layer with one multi-input neuron, further subsequent complication—an IFNN with one layer with many multi-input neurons, and finally—the true IFDNN with many layers with many multi-input neurons. The formulas for strongly optimistic, optimistic, average, pessimistic and strongly pessimistic formulas for NN parameters estimation, represented in the form of intuitionistic fuzzy pairs, are given here for the first time for each one of the presented IFNNs. To demonstrate its workability, an example of an IFDNN application to biomedical data is here presented.

The exactitude of petroleum fluid molecular weight correlations affects significantly the precision of petroleum engineering calculations and can make process design and trouble-shooting inaccurate. Some of the methods in the literature to predict petroleum fluid molecular weight are used in commercial software process simulators. According to statements made in the literature, the correlations of Lee–Kesler and Twu are the most used in petroleum engineering, and the other methods do not exhibit any significant advantages over the Lee–Kesler and Twu correlations. In order to verify which of the proposed in the literature correlations are the most appropriate for petroleum fluids with molecular weight variation between 70 and 1685 g/mol, 430 data points for boiling point, specific gravity, and molecular weight of petroleum fluids and individual hydrocarbons were extracted from 17 literature sources. Besides the existing correlations in the literature, two different techniques, nonlinear regression and artificial neural network (ANN), were employed to model the molecular weight of the 430 petroleum fluid samples. It was found that the ANN model demonstrated the best accuracy of prediction with a relative standard error (RSE) of 7.2%, followed by the newly developed nonlinear regression correlation with an RSE of 10.9%. The best available molecular weight correlations in the literature were those of API (RSE = 12.4%), Goosens (RSE = 13.9%); and Riazi and Daubert (RSE = 15.2%). The well known molecular weight correlations of Lee–Kesler, and Twu, for the data set of 430 data points, exhibited RSEs of 26.5, and 30.3% respectively.

Petroleum refining has been, is still, and is expected to remain in the next decades the main source of energy required to drive transport for mankind. The demand for automotive and aviation fuels has urged refiners to search for ways to extract more light oil products per barrel of crude oil. The heavy oil conversion processes of ebullated bed vacuum residue hydrocracking (EBVRHC) and fluid catalytic cracking (FCC) can assist refiners in their aim to produce more transportation fuels and feeds for petrochemistry from a ton of petroleum. However, a good understanding of the roles of feed quality and catalyst characteristics is needed to optimize the performance of both heavy oil conversion processes. Three knowledge discovery database techniques—intercriteria and regression analyses, and artificial neural networks—were used to evaluate the performance of commercial FCC and EBVRHC in processing 19 different heavy oils. Seven diverse FCC catalysts were assessed using a cascade and parallel fresh catalyst addition system in an EBVRHC unit. It was found that the vacuum residue conversion in the EBVRHC depended on feed reactivity, which, calculated on the basis of pilot plant tests, varied by 16.4%; the content of vacuum residue (VR) in the mixed EBVRHC unit feed (each 10% fluctuation in VR content leads to an alteration in VR conversion of 1.6%); the reaction temperature (a 1 °C deviation in reaction temperature is associated with a 0.8% shift in VR conversion); and the liquid hourly space velocity (0.01 h-1 change of LHSV leads to 0.85% conversion alteration). The vacuum gas oil conversion in the FCC unit was determined to correlate with feed crackability, which, calculated on the basis of pilot plant tests, varied by 8.2%, and the catalyst ΔCoke (each 0.03% ΔCoke increase reduces FCC conversion by 1%), which was unveiled to depend on FCC feed density and equilibrium FCC micro-activity. The developed correlations can be used to optimize the performance of FCC and EBVRHC units by selecting the appropriate feed slate and catalyst.