decision tree algorithm

Pore pressure is one of the most important reservoir-drilling parameters and knowledge of this pressure is essential for drilling costs, well safety and prevention of potential hazards. Research has shown that experimental equations have good performance accuracy only for certain regions. Most of these experimental equations have been compiled and developed based on a limited data set. Therefore, these correlations are valid in the range of changes in the parameters of those fields and are not valid for other areas. Therefore, artificial intelligent methods have given way to empirical equations. In this study, 2827 data related to three wells from one of the oil fields located in the southwest of Iran have been used. The input variables used in this paper to predict the pore pressure include 9 variables that have been selected using the feature selection method. In this study, 4 artificial intelligence algorithms include; random forest algorithm, support vector regression algorithm, artificial neural network algorithm and decision tree algorithm have been used to predict the pore pressure. After reviewing the results, it was found that the performance accuracy of the decision tree algorithm is higher than the other three algorithms (performance accuracy for the entire data set including R2 = 0.9985 and RMSE = 14.460 psi). Among the advantages of this algorithm compared to other algorithms are the best results without the need for statistical knowledge, separation of unnecessary data, short time to prepare data and reduction of relative error by finding the main node of the decision maker and analyzing it. Therefore, it can be concluded that with the development of this technique, it is possible to have high performance accuracy for a small amount of data from each field.

In recent decades, construction of massive structures has increased around the world. These structures which are built for different purposes, should have a useful life-time. Therefore, composite structures can increase the construction costs and useful life of such structures. Therefore, CFDST members can practically increase the efficiency of composite structures. Since the diameter and thickness of the outer tube have the greatest impact on determining the bearing capacity of CFDST columns, this study investigated four different values of D/t for the outer tube including 86, 85.8, 45.6 and 44. Two types of concrete were used to fill CFDST samples including C10 and C20. Besides, a comprehensive formula for estimating the bearing capacity of CFDST columns using artificial intelligence methods is proposed. The results showed that CFDST columns with lower Do / to value have higher load capacity and with increasing Do / to, it is observed that the ductility decreases. As a comparison between the filled concretes in CFDST samples of C10 and C20 grades, the stress-strain curves have shown that with increasing compressive strength, the bearing capacity for CFDST columns increased by about 6%. In the modeling section, the R2 for artificial neural network (ANN), support vector machines (SVM) and model tree (M5-MT) in the testing stage were determined to be 0.95, 0.96 and 0.97, respectively. Accordingly, the M5-MT method in the testing stage, had a better performance than other proposed methods for estimating the bearing capacity of CFDST columns. Comparison of traditional equations and AI models in estimating the bearing capacity of CFDST columns show that the formula presented by M5-MT with an average value of 1.01 and a standard deviation of 0.19 It has performed better in modeling the bearing capacity of CFDST columns than other intelligent models and traditional relationships.