A Prediction and Optimization Method of Oil and Gas Drilling ROP Based on Cart Algorithm

Abstract

The present invention relates to a method for predicting and optimizing ROP of oil and gas drilling machinery based on the CART algorithm, comprising: step 1, collecting data; step 2, performing data preprocessing separately, using 8 drilling parameters as different characteristic attributes and including The drilling data of the well is used as the input variable X, the ROP is used as the output variable Y, and the initial data set D is obtained 1 ; Step 3, data correlation analysis, to obtain different training data sets D 2 ; Step 4, use the CART algorithm to establish different training data sets D 2 The regression tree model between the input variable and the ROP; step 5, analyze the information of each leaf node of the generated binary tree, and the mean value of the leaf node is used as the predicted value of the ROP; step 6, traverse each leaf node from top to bottom According to the node division results of layers, different recommended values ​​of drilling parameters are obtained; step 7, optimal judgment of ROP. The invention can shorten the drilling period and reduce the drilling cost, thereby greatly improving the development efficiency of oil and gas resources.

Description

technical field [0001] The invention relates to a method for predicting and optimizing ROP in the field of petroleum exploration and development, in particular to a method for predicting and optimizing ROP based on a CART algorithm. Background technique [0002] In recent years, with the increase in the scale of oil and gas resource exploration and the strengthening of development efforts, the oil and gas exploration targets of various oilfields have gradually shifted from shallow formations to deep formations. The best way to develop. In the process of cobalt mining in deep and ultra-deep wells, due to complex geological conditions and harsh downhole conditions, engineering construction operations face great challenges. In order to better develop deep oil and gas resources, it is the general trend to shorten the drilling cycle, reduce drilling costs and improve drilling efficiency, and the most direct and effective way to solve these problems is to increase the ROP. There...

AI Technical Summary

Problems solved by technology

In recent years, with the rapid development of big data technology and the rapid growth of drilling data, there have been many cases where machine learning methods have been used to mine data and applied to the drilling industry. In drill bit optimization (Bi Xueliang, Yan Tie, Tao Lijie. Research on drill bit selection by neural network method in Qingshen Oilfield[J]. Journal of Harbin Engineering University, 2006, 27(z1):111-114), lithology identification (Shan Jingfu, Chen Xinxin, Zhao Zhongjun, etc. Using BP neural network The identification of complex lithology of tight sandstone gas reservoirs using the method[J]. Geophysics Advances, 2015(3):1257-1263) has been effectively applied in fields such as However, there are few studies and reports on the optimization of ROP

Images