Perforation monitoring intelligent judgment method for oil and gas well

Abstract

The invention is suitable for the technical field of oil and gas well perforation vibration monitoring, and provides an oil and gas well perforation monitoring intelligent judgment method which comprises the following steps: step 1, acquiring and establishing a perforation vibration original signal database through a high-precision wellhead vibration acquisition system; 2, variational mode decomposition is carried out on the original signals in the wellhead perforation vibration database; 3, jointly constructing a perforation feature vector by using the perforation parameters and the frequency and the amplitude obtained by variational mode decomposition; 4, obtaining a neural network model of the optimal neural network hidden tree through parameter tuning, and training the model; and step 5, performing perforation prediction by using the BP neural network model obtained by training. According to the method, whether oil pipe transmission perforation succeeds or not is determined by monitoring wellhead vibration data, dependence of a traditional wellhead perforation monitoring method on technicians is eliminated, and perforation monitoring accuracy can be effectively improved.

Description

Technical Field

[0001] The invention belongs to the technical field of oil and gas well perforation vibration monitoring, and in particular relates to an intelligent judgment method for perforation monitoring of oil and gas wells. Background Art

[0002] With the long-term exploitation of conventional oil and gas resources on land in my country, the reserves are declining. Unconventional oil and gas resources represented by deep-sea oil and gas fields and shale gas have become new growth points for my country's oil and gas production. However, the geological depth of most shale gas in my country exceeds 3,500 meters, and the average operating water depth of deepwater gas fields in the western South China Sea is 1,500 meters. Most of these wells are deep wells, and casing perforation is generally used for completion. Oilfield technicians need to know whether the perforation is completed successfully in order to take the next construction measures. Therefore, perforation monitoring of these wells is particularly important.

[0003] As an important part of well completion engineering and a major part of oil testing technology, perforation technology is a technology that uses high-energy explosives to form jets to penetrate the casing wall, cement ring, contaminated zone and part of the formation of oil and gas wells, and establish oil and gas channels between the oil and gas layer and the wellbore. Perforation completion has the advantages of good connectivity, unobstructed channels for formation fluids to flow to the wellbore, effective support provided by the casing to prevent well wall collapse, effective isolation of oil and water layers, simple process steps and low cost, and is widely used around the world.

[0004] Abnormal detonation or refusal of perforation is directly related to the safety of the wellbore and the personal safety of workers, so accurate monitoring of perforation is required. In the past, operators often used the method of listening to the sound at the wellhead and touching the oil pipe to feel the vibration, but this method has many unsafe factors. Since perforation is completed in an instant, the time is short, the energy is released quickly, and the impact duration is short, so analyzing the vibration signals generated during the perforation process and extracting useful signals from them has become an important means to detect the perforation effect. At present, operators generally use blasting vibration recorders for monitoring, that is, operators identify the detonation characteristic signals from the various signals collected by the recorder as the standard for judging the detonation of the perforating gun. Due to the different experiences of operators and the uncertainty of various well conditions, this passive operation method has a high possibility of misjudgment. Therefore, how to accurately extract the detonation characteristic signals and actively identify and judge them by the computer has become a major difficulty in current monitoring technology. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide an intelligent judgment method for perforation monitoring of oil and gas wells, aiming to solve the problems raised in the above-mentioned background technology.

[0006] The embodiment of the present invention is implemented as follows: an intelligent judgment method for perforation monitoring of oil and gas wells, comprising the following steps:

[0007] Step 1: Establish a database of perforating vibration raw signals through the high-precision wellhead vibration acquisition system;

[0008] Step 2: Perform variational mode decomposition on the original signal in the wellhead perforation vibration database;

[0009] Step 3: Use the perforation parameters and the frequency and amplitude obtained by variational mode decomposition to jointly construct the perforation feature vector;

[0010] Step 4: Obtain the neural network model of the best neural network hidden tree through parameter tuning and train the model;

[0011] Step 5: Use the trained BP neural network model to perform perforation prediction.

[0012] According to a further technical solution, the high-precision wellhead vibration acquisition system includes a computer, a high-precision acquisition card and a high-precision vibration sensor;

[0013] The high-precision vibration sensor is installed on the four-way flange of the Christmas tree to obtain the original signal containing the perforation vibration characteristics;

[0014] The high-precision vibration sensor is connected to the high-precision acquisition card via a communication cable, and the high-precision acquisition card is connected to the computer via a USB cable;

[0015] The computer is installed with data acquisition and data processing programs.

[0016] According to a further technical solution, the high-precision vibration sensor is a piezoelectric sensor with a high response frequency, which is installed on the four-way flange by magnetic adsorption or threaded connection.

[0017] According to a further technical solution, in step 1, the data collection process includes the following specific steps:

[0018] Step 1.1: Install a high-precision vibration sensor at the wellhead;

[0019] Step 1.2: Complete the line connection, open the data acquisition and data processing program, and ensure that the signal transmission is normal;

[0020] Step 1.3: The sampling frequency is set to above 5000 Hz to ensure that the data acquisition system can collect the perforation vibration characteristic signal;

[0021] Step 1.4: If the detonation is by throwing the stick, click Collect Data after throwing the stick to start data collection; if the detonation is by pressurization, click Collect Data after pressurization to start data collection.

[0022] Step 1.5: After perforation is completed, save the data and complete the collection of raw data.

[0023] A further technical solution is that in step 2, the collected original signal is subjected to variational mode decomposition to obtain a number of IMF eigenfunctions, and a frequency spectrum corresponding to the eigenfunction is obtained by fast Fourier transform.

[0024] A further technical solution is that in step 3, the main parameters that affect the transmission of perforating vibration: the depth of the perforating section h, the total charge of the perforating bullet W, the number of perforating bullets n, the length of the perforating section l1, the interlayer length l2, and the presence or absence of a splitter, the maximum amplitude and frequency of the eigenfunction components obtained by variational mode decomposition, together constitute the perforating characteristic vector.

[0025] According to a further technical solution, in step 4, the establishment and training of the neural network model includes the following steps:

[0026] Step 4.1: Divide the collected perforation vibration sample data into a training set and a test set, with the training set accounting for 80% and the test set accounting for 20%;

[0027] Step 4.2: Normalize the training set to be in the range [0,1];

[0028] Step 4.3: Normalize the test set to be in the range of [-1,1];

[0029] Step 4.4: Establish a neural network model, with the perforation feature vector as the input layer and the output layer as the perforation result judgment, 1 indicates successful perforation, and 0 indicates no perforation or perforation failure; by setting the number of training times, learning rate and training target minimum error, and through training iterations, the optimal number of hidden layers is obtained;

[0030] Step 4.5: Establish the BP neural network with the best hidden layer and train it with the training set;

[0031] Step 4.6: Perform a test set test on the trained BP neural network model, denormalize the test results, and calculate the error between the test value and the true value; until the error is within an acceptable range, a more ideal BP neural network training model is obtained.

[0032] According to a further technical solution, step 5 comprises the following specific steps:

[0033] Step 5.1: Collection of wellhead perforation vibration signals;

[0034] Step 5.2: Construct perforation feature vector;

[0035] Step 5.3: Put the perforation vibration feature vector into the trained BP neural network for prediction.

[0036] An embodiment of the present invention provides an intelligent judgment method for perforation monitoring of oil and gas wells. The method determines whether the oil pipe transmission perforation is successful by monitoring the wellhead vibration data, gets rid of the dependence of traditional wellhead perforation monitoring methods on technicians, and can effectively improve the accuracy of perforation monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A flow chart of an intelligent judgment method for perforation monitoring of oil and gas wells provided by an embodiment of the present invention;

[0038] Figure 2 This is a structural diagram of a high-precision wellhead vibration acquisition system;

[0039] Figure 3 is the original signal diagram containing perforation characteristics;

[0040] Figure 4 The time domain and frequency domain diagrams obtained by variational mode decomposition;

[0041] Figure 5 This is the diagram of the constructed BP neural network model.

[0042] In the attached drawings: computer 1; high-precision acquisition card 2; high-precision vibration sensor 3; Christmas tree 4; four-way flange 5. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0044] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0045] like Figure 1 As shown, an intelligent judgment method for perforation monitoring of oil and gas wells provided by an embodiment of the present invention includes the following steps:

[0046] Step 1: Establish a database of perforating vibration raw signals through the high-precision wellhead vibration acquisition system;

[0047] Step 2: Perform variational mode decomposition on the original signal in the wellhead perforation vibration database;

[0048] Step 3: Use the perforation parameters and the frequency and amplitude obtained by variational mode decomposition to jointly construct the perforation feature vector;

[0049] Step 4: Obtain the neural network model of the best neural network hidden tree through parameter tuning and train the model;

[0050] Step 5: Use the trained BP neural network model to perform perforation prediction.

[0051] like Figure 2 As shown, as a preferred embodiment of the present invention, the high-precision wellhead vibration acquisition system includes a computer 1, a high-precision acquisition card 2 and a high-precision vibration sensor 3;

[0052] The high-precision vibration sensor 3 is installed on the four-way flange 5 of the Christmas tree 4 to obtain the original signal containing the perforation vibration characteristics;

[0053] The high-precision vibration sensor 3 is connected to the high-precision acquisition card 2 via a communication cable, and the high-precision acquisition card 2 is connected to the computer 1 via a USB cable;

[0054] The computer 1 is installed with data acquisition and data processing programs.

[0055] In the embodiment of the present invention, the high-precision vibration sensor 3 is a piezoelectric sensor with a high response frequency, which is installed on the cross-way flange 5 by magnetic adsorption or threaded connection, and can be installed on the casing by magnetic adsorption. The high-precision acquisition card has a 24-bit analog-to-digital conversion accuracy, and a single-channel sampling rate of more than 100ks / s.

[0056] As a preferred embodiment of the present invention, in step 1, the data collection process includes the following specific steps:

[0057] Step 1.1: Install a high-precision vibration sensor 3 at the wellhead;

[0058] Step 1.2: Complete the line connection, open the data acquisition and data processing program, and ensure that the signal transmission is normal;

[0059] Step 1.3: The sampling frequency is set to above 5000 Hz to ensure that the data acquisition system can collect the perforation vibration characteristic signal;

[0060] Step 1.4: If the detonation is by throwing the stick, click Collect Data after throwing the stick to start data collection; if the detonation is by pressurization, click Collect Data after pressurization to start data collection.

[0061] Step 1.5: After perforation is completed, save the data and complete the collection of raw data.

[0062] In the embodiment of the present invention, based on the above steps, the original signal containing the perforation characteristics collected by the high-precision wellhead vibration acquisition system under the condition of pressurized detonation is as follows: Figure 3 shown.

[0063] As a preferred embodiment of the present invention, in step 2, the collected original signal is subjected to variational mode decomposition to obtain a number of IMF eigenfunctions, and a fast Fourier transform is used to obtain a frequency spectrum corresponding to the eigenfunction.

[0064] In the embodiment of the present invention, the time domain diagram and frequency domain diagram obtained after the original signal is subjected to variational mode decomposition are as follows: Figure 4 shown.

[0065] As a preferred embodiment of the present invention, in step 3, the main parameters affecting the perforating vibration transmission: the well depth h of the perforating section, the total charge W of the perforating bullet, the number n of the perforating bullets, the length l1 of the perforating section, the length l2 of the interlayer, and the presence or absence of a divider (1 if there is a divider and 0 if there is no divider), and the maximum amplitude and frequency of the eigenfunction components obtained by variational mode decomposition, together constitute the perforating characteristic vector.

[0066] In the embodiment of the present invention, taking the perforating section depth h, the total charge of the perforating bullet W, the number of perforating bullets n, the perforating section length l1, the interlayer length l2, and the perforating construction without a packer as an example, the original signal is decomposed into 7 variational modal components to obtain the maximum amplitude and frequency A1, A2, A3, A4, A5, A6, A7 and f1, f2, f3, f4, f5, f6, f7 respectively, and the eigenvector is (h, W, n, l1, l2, 0, A1, f1, A2, f2, A3, f3, A4, f4, A55, f5, A6, f6, A7, f7).

[0067] As a preferred embodiment of the present invention, in step 4, the establishment and training of the neural network model includes the following steps:

[0068] Step 4.1: Divide the collected perforation vibration sample data into a training set and a test set, with the training set accounting for 80% and the test set accounting for 20%;

[0069] Step 4.2: Normalize the training set to be in the range [0,1];

[0070] Step 4.3: Normalize the test set to be in the range of [-1,1];

[0071] Step 4.4: Create Figure 5 In the neural network model shown, the perforation feature vector is the input layer, and the output layer is the perforation result judgment. 1 indicates successful perforation, and 0 indicates no perforation or perforation failure. By setting the number of training times, learning rate and minimum error of training target, the optimal number of hidden layers is obtained through training iterations.

[0072] Step 4.5: Establish the BP neural network with the best hidden layer and train it with the training set;

[0073] Step 4.6: Perform a test set test on the trained BP neural network model, denormalize the test results, and calculate the error between the test value and the true value; until the error is within an acceptable range, a more ideal BP neural network training model is obtained.

[0074] As a preferred embodiment of the present invention, step 5 includes the following specific steps:

[0075] Step 5.1: Collection of wellhead perforation vibration signals;

[0076] Step 5.2: Construct perforation feature vector;

[0077] Step 5.3: Put the perforation vibration feature vector into the trained BP neural network for prediction.

[0078] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent judgment method for perforation monitoring of oil and gas wells, characterized in that: The following steps are involved: Step 1: Establish a database of perforating vibration raw signals through the high-precision wellhead vibration acquisition system; Step 2: Perform variational mode decomposition on the original signal in the wellhead perforation vibration database; Step 3: Use the perforation parameters and the frequency and amplitude obtained by variational mode decomposition to jointly construct the perforation feature vector; Step 4: Obtain the neural network model of the best neural network hidden tree through parameter tuning and train the model; Step 5: Use the trained BP neural network model to perform perforation prediction.

2. The intelligent judgment method for perforation monitoring of oil and gas wells according to claim 1 is characterized in that: The high-precision wellhead vibration acquisition system includes a computer, a high-precision acquisition card and a high-precision vibration sensor; The high-precision vibration sensor is installed on the four-way flange of the Christmas tree to obtain the original signal containing the perforation vibration characteristics; The high-precision vibration sensor is connected to the high-precision acquisition card via a communication cable, and the high-precision acquisition card is connected to the computer via a USB cable; The computer is installed with data acquisition and data processing programs.

3. The intelligent judgment method for perforation monitoring of oil and gas wells according to claim 2 is characterized in that: The high-precision vibration sensor is a piezoelectric sensor with a high response frequency, which is installed on the four-way flange by means of magnetic adsorption or threaded connection.

4. The intelligent judgment method for perforation monitoring of oil and gas wells according to any one of claims 2 and 3, characterized in that: In step 1, the data collection process includes the following specific steps: Step 1.1: Install a high-precision vibration sensor at the wellhead; Step 1.2: Complete the line connection, open the data acquisition and data processing program, and ensure that the signal transmission is normal; Step 1.3: The sampling frequency is set to above 5000 Hz to ensure that the data acquisition system can collect the perforation vibration characteristic signal; Step 1.4: If it is a rod-throwing detonation, click Collect Data after throwing the rod to start data collection; if it is a pressurized detonation, click Data Collection after pressurization to start data collection; Step 1.5: After perforation is completed, save the data and complete the collection of raw data.

5. The intelligent judgment method for perforation monitoring of oil and gas wells according to claim 4 is characterized in that: In step 2, the collected original signal is subjected to variational mode decomposition to obtain a number of IMF eigenfunctions, and a frequency spectrum corresponding to the eigenfunction is obtained by fast Fourier transform.

6. The intelligent judgment method for perforation monitoring of oil and gas wells according to claim 5 is characterized in that: In step 3, the main parameters that affect the transmission of perforating vibration: the depth of the perforating section h, the total charge of the perforating bullet W, the number of perforating bullets n, the length of the perforating section l1, the length of the interlayer l2, and the presence or absence of a splitter, the maximum amplitude and frequency of the eigenfunction components obtained by variational mode decomposition, together constitute the perforating characteristic vector.

7. The intelligent judgment method for perforation monitoring of oil and gas wells according to claim 6 is characterized in that: In step 4, the establishment and training of the neural network model includes the following steps: Step 4.1: Divide the collected perforation vibration sample data into a training set and a test set, with the training set accounting for 80% and the test set accounting for 20%; Step 4.2: Normalize the training set to be in the range [0,1]; Step 4.3: Normalize the test set to be in the range of [-1,1]; Step 4.4: Establish a neural network model, with the perforation feature vector as the input layer and the output layer as the perforation result judgment, 1 indicates successful perforation, and 0 indicates no perforation or perforation failure; by setting the number of training times, learning rate and training target minimum error, and through training iterations, the optimal number of hidden layers is obtained; Step 4.5: Establish the BP neural network with the best hidden layer and train it with the training set; Step 4.6: Perform test set test on the trained BP neural network model, denormalize the test results, and calculate the error between the test value and the true value; until the error meets the requirements, the ideal BP neural network training model is obtained.

8. The intelligent judgment method for perforation monitoring of oil and gas wells according to claim 7 is characterized in that: The step 5 comprises the following specific steps: Step 5.1: Collection of wellhead perforation vibration signals; Step 5.2: Construct perforation feature vector; Step 5.3: Put the perforation vibration feature vector into the trained BP neural network for prediction.

Images