CN-121997689-A - Method and system for determining anti-deformation perforation parameters based on SVM model and numerical simulation
Abstract
The disclosure belongs to the technical field of casing damage, and particularly relates to a method and a system for determining anti-casing deformation perforation parameters based on SVM (support vector machine) model and numerical simulation. The method comprises the steps of establishing a double-perforating bullet numerical model, obtaining sleeve stress through the double-perforating bullet numerical model, taking the sleeve stress as a data set, constructing a pre-trained sleeve perforation stress calculation model based on the data set and an SVM learning algorithm, and determining anti-sleeve-deformation perforation parameters by using the pre-trained sleeve perforation stress calculation model. The method overcomes the defects of multiple input parameters, complex parameter adjustment, low calculation speed, need of post-treatment and the like when the traditional numerical calculation software is used for calculating the casing stress, has the characteristics of being friendly to users, high in calculation efficiency, convenient to use and the like, and can rapidly calculate the effective stress of the casing according to perforation parameter data selected by users, judge whether the casing is invalid or not and provide the optimal perforation parameter combination.
Inventors
- ZHANG PEIJUN
- Leng Youheng
- WANG SIQI
- WEI ZHANJUN
- CHENG MUWEI
- LIU RONGHE
- CHEN PENGYU
- ZHANG WENQI
Assignees
- 中国石油天然气集团有限公司
- 中国石油国际勘探开发有限公司
- 中石油阿姆河天然气勘探开发(北京)有限公司
- 中国石油集团科学技术研究院有限公司
Dates
- Publication Date
- 20260508
- Application Date
- 20241106
Claims (10)
- 1. A method for determining anti-casing deformation perforation parameters based on an SVM model and numerical simulation, the method comprising: establishing a double perforating bullet numerical model; Obtaining the casing stress through a double perforating bullet numerical model, and taking the casing stress as a data set; constructing a pre-trained casing perforation stress calculation model based on the data set and an SVM learning algorithm; and determining the sleeve deformation preventing perforation parameters by using a pre-trained sleeve perforation stress calculation model.
- 2. The method for determining anti-casing perforation parameters based on SVM model and numerical simulation according to claim 1, wherein, Obtaining the casing stress through the double perforating bullet numerical model comprises the following steps: determining a geometric parameter combination, a main control factor parameter change range value and a main control factor parameter change step length in the dual-perforating bullet numerical model, and executing the following steps for each group of geometric parameters: Inputting an initial value of a main control factor parameter into a double perforation numerical model to obtain a casing stress corresponding to a current input value; And adjusting the parameter value of the main control factor according to the change step length of the parameter of the main control factor, and inputting the parameter value into the double perforation numerical model again to obtain the casing stress corresponding to the current input value.
- 3. The method for determining anti-casing perforation parameters based on SVM model and numerical simulation according to claim 1, wherein, Based on the data set and SVM learning algorithm, constructing a pre-trained casing perforation stress calculation model, comprising: Constructing an initial casing perforation stress calculation model; And (3) fine tuning model parameters of the initial casing perforation stress calculation model by taking the model loss of the initial casing perforation stress calculation model as a target through a data set and an SVM learning algorithm to obtain a pre-trained casing perforation stress calculation model.
- 4. The method for determining anti-set perforation parameters based on SVM model and numerical simulation according to claim 3, By means of the data set and SVM learning algorithm, the model loss of the initial casing perforation stress calculation model is minimized, fine tuning model parameters of the initial casing perforation stress calculation model to obtain a pre-trained casing perforation stress calculation model, wherein the method comprises the following steps of: Obtaining an input vector and an expected output, wherein the input vector is a geometric parameter combination, and the expected output is a sleeve stress; inputting an input vector to an initial casing perforation stress calculation model to obtain an actual output value; Comparing the expected output with the actual output to obtain an output deviation; judging whether the output deviation meets a preset condition or not; if not, judging whether the current training reaches a training frequency threshold value; When the training frequency threshold is reached, ending the current training to obtain a pre-trained casing perforation stress calculation model; If the training frequency threshold is not reached, calculating the unit error of the hidden layer, calculating the error gradient and adjusting the weight of each function of the hidden layer.
- 5. The method for determining anti-set perforation parameters based on SVM model and numerical simulation according to claim 2, wherein, Determining anti-casing deformation perforation parameters using a pre-trained casing perforation stress calculation model, comprising: inputting the geometric parameter combination and the main control parameter into a pre-trained casing perforation stress calculation model to obtain actual casing stress; judging whether the casing strength fails according to the Mises yield strength criterion and the actual casing stress, and determining the casing deformation prevention perforation parameters.
- 6. A system for determining anti-set-up perforation parameters based on an SVM model and numerical modeling, the system comprising: the building module is used for building a double-perforating bullet numerical model; the data set module is used for obtaining the sleeve stress through the double perforating bullet numerical model and taking the sleeve stress as a data set; the construction module is used for constructing a pre-trained casing perforation stress calculation model based on the data set and the SVM learning algorithm; and the determining module is used for determining the anti-casing deformation perforation parameters by using the pre-trained casing perforation stress calculation model.
- 7. The system for determining anti-set perforation parameters based on SVM models and numerical modeling of claim 6, A data set module for deriving casing stress from a dual charge numerical model, comprising: The data set module is used for determining the geometric parameter combination, the main control factor parameter change range value and the main control factor parameter change step length in the double-perforating bullet numerical model, and executing the following steps for each group of geometric parameters: Inputting an initial value of a main control factor parameter into a double perforation numerical model to obtain a casing stress corresponding to a current input value; And adjusting the parameter value of the main control factor according to the change step length of the parameter of the main control factor, and inputting the parameter value into the double perforation numerical model again to obtain the casing stress corresponding to the current input value.
- 8. The system for determining anti-set perforation parameters based on SVM models and numerical modeling of claim 6, The construction module is used for constructing a pre-trained casing perforation stress calculation model based on the data set and an SVM learning algorithm, and comprises the following steps: the construction module is used for constructing an initial casing perforation stress calculation model; And (3) fine tuning model parameters of the initial casing perforation stress calculation model by taking the model loss of the initial casing perforation stress calculation model as a target through a data set and an SVM learning algorithm to obtain a pre-trained casing perforation stress calculation model.
- 9. The system for determining anti-set perforation parameters based on SVM models and numerical modeling of claim 6, By means of the data set and SVM learning algorithm, the model loss of the initial casing perforation stress calculation model is minimized, fine tuning model parameters of the initial casing perforation stress calculation model to obtain a pre-trained casing perforation stress calculation model, wherein the method comprises the following steps of: Obtaining an input vector and an expected output, wherein the input vector is a geometric parameter combination, and the expected output is a sleeve stress; inputting an input vector to an initial casing perforation stress calculation model to obtain an actual output value; Comparing the expected output with the actual output to obtain an output deviation; judging whether the output deviation meets a preset condition or not; if not, judging whether the current training reaches a training frequency threshold value; When the training frequency threshold is reached, ending the current training to obtain a pre-trained casing perforation stress calculation model; If the training frequency threshold is not reached, calculating the unit error of the hidden layer, calculating the error gradient and adjusting the weight of each function of the hidden layer.
- 10. The system for determining anti-set perforation parameters based on SVM models and numerical modeling of claim 7, The determining module is used for determining the anti-casing deformation perforation parameters by using a pre-trained casing perforation stress calculation model and comprises the following steps: the determining module is used for inputting the geometric parameter combination and the main control parameter value into a pre-trained casing perforation stress calculation model to obtain actual casing stress; judging whether the casing strength fails according to the Mises yield strength criterion and the actual casing stress, and determining the casing deformation prevention perforation parameters.
Description
Method and system for determining anti-deformation perforation parameters based on SVM model and numerical simulation Technical Field The disclosure belongs to the technical field of casing damage, and particularly relates to a method and a system for determining anti-casing deformation perforation parameters based on SVM (support vector machine) model and numerical simulation. Background The perforating technology is a well completion technology for connecting a bottom stratum with a shaft in a perforating mode, so that a connecting channel of the bottom stratum and the shaft is effectively improved, the working efficiency of hydraulic fracturing operation is improved, and further the productivity of an oil well is remarkably improved. However, the perforation technology has some disadvantages and shortcomings, in the perforation engineering, the perforating bullet melts the liner through the huge heat generated instantaneously by the explosion of the explosive, so that the liner forms metal jet flow, and the metal jet flow penetrates through the rock stratum and the shaft structure under the action of detonation force generated by the explosion, and the jet flow inevitably causes damage deformation of the sleeve when passing through the metal sleeve in the perforation penetration stage, thereby affecting the structural integrity of the sleeve. After the casing is deformed, the uniformity of the whole stress of the casing is damaged, the deformed part becomes a weak point, and under the action of complex ground stress and local high stress, the problem of casing deformation can further aggravate the damage of the integral structure of the shaft, and the productivity and service life of an oil-gas well are reduced. In particular, in recent years, the problems of casing deformation of Fuling shale gas wells, jiyang shale oil horizontal wells, tarim oil fields and the like in China are remarkable, and the adverse effects of increasing exploitation cost, greatly reducing yield and the like are brought, so that the development process of shale gas in China is severely restricted. In this regard, it is important to ensure the stability of the casing structure during perforating operations. The perforation parameters are key factors of perforation operation, the design of the perforation parameters directly influences the damage deformation degree of the perforation sleeve, aiming at the problem of influence of the perforation parameters on the structural integrity of the sleeve, some scholars put forward a perforation parameter optimization method, such as Wang Haiyang, a perforation well stratum fracture pressure prediction method and a construction parameter optimization design method, the construction parameters are adjusted and optimized by utilizing the size of perforation well stratum fracture pressure, the perforation parameters and the sleeve types are optimally designed by predicting the stratum fracture pressure of a vertical well and a horizontal well under any perforation angle, hu Zheyu and the like utilize elastic mechanics and a bisection method, the influence rule of the perforation phase angle on the fracture pressure is put forward when corresponding to the perforation of different perforation phase angles is carried out, li Zhongdeng determines the maximum productivity of oil fields corresponding to different perforation process parameters, and a perforation parameter based on productivity prediction under the perforation parameters is put forward. The perforation parameter design optimization method is provided based on ensuring the productivity efficiency of the oil and gas well, ignores the damage of perforation parameter design to the integrity of the shaft during perforation operation, has certain potential safety hazard to the stability of the oil and gas well, and is not beneficial to the long-term development of the oil and gas well. In addition, related scholars develop a series of researches, G.E.King performs a test research on the outward extrusion load of the perforated casing, the influence of high perforation density on the outward extrusion load of the casing is researched to obtain a perforation casing extrusion resistance intensity coefficient Ktest, dou Yihua and the like research the mechanical properties of the perforated casing in a test mode to obtain a calculation formula of a casing hole stress intensity factor and a correction coefficient thereof and a calculation formula of the residual extrusion resistance intensity of the perforated casing, guo Yonggui and the like perform a high-density perforation test based on a deep water reservoir of the gulf of Mexico, the perforation and non-perforated casing are analyzed by adopting finite element simulation, the failure mode and the instability damage mechanism of the perforated casing are explained to a certain extent, liu Xianbo and the like utilize a double-perforation-bullet finite element model, and the influence of p