Oil pipe screwing-on quality evaluation model construction method, oil pipe screwing-on quality evaluation method and intelligent screwing-on torquemeter

Abstract

The embodiment of the invention relates to the technical field of petroleum and natural gas development, and provides an oil pipe make-up quality evaluation model construction and evaluation method and an intelligent make-up torquemeter, and the method comprises the steps: obtaining an original oil pipe make-up torque curve data set; carrying out expansion processing on the data set by utilizing a pre-trained data expansion model to obtain an expanded oil pipe make-up torque curve data set; constructing a make-up torque curve classification model based on a convolutional neural network; and training the make-up torque curve classification model by using the expanded oil pipe make-up torque curve data set to obtain a trained make-up torque curve classification model. According to the embodiment of the invention, the accuracy and efficiency of oil pipe screwing-on quality evaluation can be improved.

Description

Technical Field

[0001] The embodiments in this specification relate to the field of oil and gas development technology, and in particular to the construction of a pipeline fastening quality evaluation model, an evaluation method, and an intelligent fastening torque meter. Background Technology

[0002] Tubing coupling is a crucial step in connecting tubing to the pipeline system within the well, playing a key role in preventing the leakage of high-temperature, high-pressure, and hydrogen sulfide gases from the tubing into the atmosphere. However, even minor damage to the tubing during transportation or errors during connection can hinder the formation of an effective seal between the tubing and the coupling, leading to gas leakage and significant environmental damage and economic losses. Therefore, diagnosing the quality of tubing connections is essential for leak prevention. Traditional tubing connection quality diagnosis primarily involves manually inspecting the torque curve generated during coupling to determine if it conforms to a standard form, thus diagnosing tubing damage or incorrect connections. However, manual inspection heavily relies on the expertise and experience of field engineers, resulting in a lack of objectivity and quantifiability in the assessment.

[0003] With the development of artificial intelligence, researchers have used machine learning algorithms such as logistic regression to classify torque curves. By dividing the curves into qualified and unqualified categories, they have achieved preliminary intelligent recognition capabilities. Although this method reveals the distribution characteristics of the statistical and spatial features of the curves under different forms, as a traditional machine learning model, it largely relies on manual feature extraction. However, mining high-quality features requires highly specialized domain knowledge and numerous trials, making the feature extraction process very time-consuming and labor-intensive, thus affecting the efficiency and accuracy of curve classification. Therefore, there is an urgent need for a method for evaluating the quality of torque curves on pipelines that can improve the efficiency and accuracy of torque curve evaluation. Summary of the Invention

[0004] To address the aforementioned problems in the prior art, the purpose of the embodiments in this specification is to provide a tubing buckle quality evaluation model construction, evaluation method, and intelligent buckle torque meter, so as to solve the problems of low efficiency and accuracy in tubing buckle quality evaluation in the prior art.

[0005] To solve the above-mentioned technical problems, the specific technical solutions of the embodiments in this specification are as follows:

[0006] On the one hand, some embodiments of this specification provide a method for constructing a quality evaluation model for tubing couplings, the method comprising:

[0007] Obtain the raw torque curve dataset on the tubing;

[0008] The dataset is augmented using a pre-trained data augmentation model to obtain an augmented tubing torque curve dataset.

[0009] A classification model for upper torque curves is constructed based on a convolutional neural network;

[0010] The expanded tubing upper torque curve dataset is used to train the upper torque curve classification model to obtain a trained upper torque curve classification model.

[0011] Furthermore, the data augmentation model is a deep convolutional generative adversarial network model, which includes a generator and a discriminator;

[0012] The training process of the data augmentation model includes:

[0013] S1: Initialize the weight parameters of the generator and discriminator using a Gaussian distribution;

[0014] S2: Input the random noise vector into the generator and output the generated torque curve;

[0015] S3: Input the generated torque curve and the real torque curve as training samples into the discriminator to obtain the probability value of each sample;

[0016] S4: Construct a first loss function based on the generated torque curve and the probability value, and construct a second loss function based on the real torque curve and the generated torque curve;

[0017] S5: Update the weight parameters of the generator and discriminator according to the first loss function and the second loss function using the backpropagation algorithm;

[0018] S6: Repeat steps S2-S5 until the predetermined number of training iterations is reached.

[0019] Furthermore, updating the weight parameters of the generator and discriminator using the backpropagation algorithm based on the first loss function and the second loss function includes:

[0020] After each training session, the generator's weight parameters are updated using the backpropagation algorithm based on the first loss function.

[0021] After the number of training iterations reaches a preset value, the weight parameters of the discriminator are updated using the backpropagation algorithm based on the second loss function.

[0022] Furthermore, the upper torque curve classification model includes an input layer, a feature extraction layer, and an output layer;

[0023] The input layer is used to receive torque curve data from the tubing.

[0024] The feature extraction layer consists of multiple stacked convolutional layers, pooling layers, and batch normalization layers, used to capture features in the torque curve data on the tubing to obtain feature representations;

[0025] The output layer consists of a fully connected layer and a Softmax layer, which is used to map the feature representation to the corresponding probability distribution and output the final curve classification result.

[0026] On the other hand, some embodiments of this specification also provide a device for constructing a quality evaluation model for tubing couplings, the device comprising:

[0027] The acquisition module is used to acquire the raw torque curve dataset on the tubing.

[0028] An expansion module is used to expand the dataset using a pre-trained data expansion model to obtain an expanded oil pipe torque curve dataset.

[0029] The building block is used to construct an upper torque curve classification model based on a convolutional neural network;

[0030] The training module is used to train the upper torque curve classification model using the expanded upper torque curve dataset of the tubing, so as to obtain the trained upper torque curve classification model.

[0031] Based on the same inventive concept, some embodiments of this specification also provide a method for evaluating the quality of tubing couplings, the method comprising:

[0032] Receive the torque and number of turns data of the oil pipe under test, and obtain the upper torque curve based on the torque and number of turns data;

[0033] The upper torque curve is input into the upper torque curve classification model trained using the method described in any of the foregoing embodiments for classification and identification, so as to obtain the upper torque curve classification result of the oil pipe under test;

[0034] The quality of the upper clamping of the oil pipe under test is evaluated based on the classification results of the upper clamping torque curve.

[0035] Furthermore, the classification results of the upper torque curve include: standard upper torque curve, abnormal fluctuation upper torque curve, and upper torque curve without inflection point;

[0036] The evaluation of the connection quality of the oil pipe under test based on the classification results of the connection torque curve includes:

[0037] If the classification result of the upper buckle torque curve is a standard upper buckle torque curve, then the upper buckle quality evaluation result of the oil pipe under test is normal.

[0038] If the classification result of the upper buckling torque curve is either an abnormal fluctuation upper buckling torque curve or an upper buckling torque curve without an inflection point, then the upper buckling quality evaluation result of the oil pipe under test is abnormal.

[0039] On the other hand, some embodiments of this specification also provide a tubing thread quality evaluation device, the device comprising:

[0040] The receiving module is used to receive the torque and number of turns data of the oil pipe under test, and to obtain the upper torque curve based on the torque and number of turns data;

[0041] The classification and recognition module is used to input the upper torque curve into the upper torque curve classification model trained by the method described in any of the foregoing embodiments for classification and recognition, so as to obtain the classification result of the upper torque curve of the oil pipe under test;

[0042] The evaluation module is used to evaluate the up-coil quality of the oil pipe under test based on the classification results of the up-coil torque curve.

[0043] On another note, some embodiments of this specification also provide an intelligent torque meter, including a memory, a processor, and a computer program stored in the memory, which, when run by the processor, executes instructions for the above-described method.

[0044] On the other hand, some embodiments of this specification also provide a tubing coupling quality evaluation system, which includes: tubing, tubing wrenches and intelligent coupling torque meter;

[0045] The tubing clamp is used to fasten the tubing and generate torque and turn data;

[0046] The intelligent torque meter is connected to the tubing wrench and is used to receive the torque and number of turns data, evaluate the quality of tubing connection based on the torque and number of turns data, and output the evaluation result.

[0047] Furthermore, the intelligent upper buckle torque meter also includes: a communication module, a classification and recognition module, a drawing and display module, and a power supply module;

[0048] The communication module is used to interact with the tubing wrench, receive torque and rotation data sent by the tubing wrench and send it to the drawing and display module;

[0049] The drawing and display module is used to receive the torque and revolution count data, draw the upper torque curve image based on the torque and revolution count data, and display it;

[0050] The classification and recognition module is used to classify and recognize the upper torque curve to obtain the classification result of the upper torque curve;

[0051] The power supply module is used to provide power to the communication module, the classification and recognition module, and the drawing and display module.

[0052] Furthermore, the intelligent buckle torque meter also includes: indicator lights and an alarm module;

[0053] The indicator light is used to indicate whether the power supply of the intelligent upper clamp torque meter is on;

[0054] The alarm module is used to issue an alarm and sound when the quality evaluation result of the top tap of the oil pipe under test is detected to be abnormal.

[0055] Furthermore, the intelligent buckle torque meter also includes: a chassis;

[0056] The chassis is used to house and fix the communication module, classification and identification module, drawing and display module and power supply module, and to shield external electromagnetic radiation signals.

[0057] In another aspect, some embodiments of this specification also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor of a computer device, performs instructions for any of the methods described above.

[0058] In another aspect, embodiments of this specification also provide a computer program product, which, when run by the processor of a computer device, executes instructions for any of the methods described above.

[0059] Some embodiments of this specification provide one or more technical solutions, which have at least the following technical effects:

[0060] The embodiments in this specification automatically acquire the original tubing torque curve dataset and augment it using a pre-trained data augmentation model to obtain an augmented dataset. This augmented dataset provides a large number of training samples for the subsequent curve classification model, thereby ensuring the classification accuracy of the curve classification model. Then, a curve classification model is constructed based on a convolutional neural network (CNN), and the model is trained using the augmented dataset. The CNN can automatically extract features from the torque curves, avoiding the subjectivity and instability of manual feature extraction, thus improving the accuracy of tubing torque quality assessment. Furthermore, compared to existing methods using logistic regression for curve classification, using a CNN does not require a significant amount of time and numerous attempts to mine features, thereby improving the efficiency of tubing torque quality assessment.

[0061] The above description is merely an overview of some embodiments of the technical solutions in this specification. In order to better understand the technical means of some embodiments of this specification and to implement them in accordance with the content of the specification, and to make the above and other objects, features and advantages of the embodiments of this specification more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0062] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0063] Figure 1 A flowchart illustrating a method for constructing a quality evaluation model for tubing couplings, as shown in some embodiments of this specification, is presented.

[0064] Figure 2 Schematic diagrams of three upper clamping torque curves are shown in some embodiments of this specification;

[0065] Figure 3 This specification shows schematic diagrams of the data augmentation model structure in some embodiments;

[0066] Figure 4 A flowchart illustrating the data augmentation model training process in some embodiments of this specification is shown.

[0067] Figure 5 This specification shows a schematic diagram of the structure of the upper torque curve classification model in some embodiments;

[0068] Figure 6 A flowchart illustrating a method for evaluating the quality of tubing connections is shown in some embodiments of this specification.

[0069] Figure 7 This specification shows a schematic diagram of a tubing fastener quality evaluation model construction device in some embodiments;

[0070] Figure 8 This specification shows a schematic diagram of the structure of a tubing fastener quality evaluation device in some embodiments;

[0071] Figure 9 This specification shows a schematic diagram of the structure of a tubing fastener quality evaluation system;

[0072] Figure 10 A schematic diagram of the structure of the intelligent buckle torque meter in some embodiments of this specification is shown.

[0073] Explanation of symbols in the attached drawings:

[0074] 1. Oil pipe;

[0075] 2. Oil pipe wrench;

[0076] 3. Intelligent buckle torque meter;

[0077] 4. Sealing buckle;

[0078] 311. Communication module;

[0079] 312. Classification and Recognition Module;

[0080] 313. Draw the display module;

[0081] 314. Power supply module;

[0082] 315. Indicator lights;

[0083] 316. Alarm module;

[0084] 317. Chassis;

[0085] 701. Acquisition Module;

[0086] 702. Expansion Module;

[0087] 703. Building Modules;

[0088] 704, Training Module;

[0089] 801. Receiver module;

[0090] 802. Classification and Recognition Module;

[0091] 803. Evaluation Module. Detailed Implementation

[0092] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this specification.

[0093] To address the aforementioned issues, this specification provides a method for constructing a quality evaluation model for oil pipe couplings. Figure 1This is a flowchart illustrating a method for constructing a quality evaluation model for oil pipe couplings, as provided in the embodiments of this specification. This specification provides the operational steps described in the embodiments or flowcharts, but based on conventional or non-inventive labor, more or fewer operational steps may be included. The order of steps listed in the embodiments is merely one possible execution order among many and does not represent the only possible execution order. In actual system or device products, the methods shown in the embodiments or accompanying drawings can be executed sequentially or in parallel.

[0094] It should be noted that the terms "first," "second," etc., used in this specification, claims, and the foregoing drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0095] Reference Figure 1 As shown in the embodiments of this specification, a method for constructing a quality evaluation model for tubing couplings is provided. The method includes:

[0096] S101: Obtain the raw tubing torque curve dataset;

[0097] S102: The dataset is augmented using a pre-trained data augmentation model to obtain an augmented tubing torque curve dataset.

[0098] S103: Constructing a classification model for upper torque curves based on convolutional neural networks;

[0099] S104: The upper torque curve classification model is trained using the expanded tubing upper torque curve dataset to obtain a trained upper torque curve classification model.

[0100] The embodiments in this specification automatically acquire the original tubing torque curve dataset and augment it using a pre-trained data augmentation model to obtain an augmented dataset. This augmented dataset provides a large number of training samples for the subsequent curve classification model, thereby ensuring the classification accuracy of the curve classification model. Then, a curve classification model is constructed based on a convolutional neural network (CNN), and the model is trained using the augmented dataset. The CNN can automatically extract features from the torque curves, avoiding the subjectivity and instability of manual feature extraction, thus improving the accuracy of tubing torque quality assessment. Furthermore, compared to existing methods using logistic regression for curve classification, using a CNN does not require a significant amount of time and numerous attempts to mine features, thereby improving the efficiency of tubing torque quality assessment.

[0101] Tubing coupling is a crucial operational step in the oil and gas industry, referring to the connection of tubing to pipelines or other related equipment within the well using a sealing coupling. The tubing coupling torque curve describes the variation of torque value with the number of rotations during the coupling process. As the threads tighten, the torque exhibits a certain trend. Initially, the torque value is relatively low due to the threads just making contact, gradually increasing as the threads engage further, eventually reaching a peak. Analyzing the tubing coupling torque curve allows for assessment of the connection quality between the tubing and the well pipeline, helping to identify potential problems promptly and ensuring the safety of tubing transportation. In the embodiments of this specification, to enable timely evaluation of the tubing connection quality during coupling, a coupling torque curve classification model is constructed to automatically classify the torque curves generated during the coupling process, and the tubing connection quality is evaluated based on the classification results. Based on extensive experimental results, coupling torque curves can be divided into three types: standard coupling torque curves, abnormal fluctuation coupling torque curves, and coupling torque curves without inflection points. Figure 2 As shown, the standard torque curve indicates a normal torque curve, signifying that the tubing connection quality meets the standard. Abnormally fluctuating torque curves and torque curves without an inflection point indicate abnormal torque curves, signifying that the tubing connection quality does not meet the standard. In these cases, the tubing needs to be re-tightened until it meets the standard. Since the probability of abnormal torque curves in reality is relatively small, the abnormal torque curve dataset has too few samples. However, training convolutional neural network models typically requires a large number of training samples. To ensure the performance of the torque curve classification model built based on a convolutional neural network, the original tubing torque curve dataset needs to be expanded to generate a large number of abnormal curve samples for training the torque curve classification model.

[0102] In some embodiments, the original tubing torque curve dataset includes the three types of tubing torque curves mentioned above and their corresponding labels. To ensure the data quality of the training dataset, data preprocessing is required before expanding the original curve dataset, including unifying the horizontal and vertical coordinates of the curves, denoising, and normalization. Unifying the horizontal and vertical coordinates involves unifying the horizontal coordinate (number of rotations) and vertical coordinate (torque value) of the upper torque curve to ensure consistency in the range of values ​​for both coordinates. In this embodiment, the horizontal coordinate is unified to 7 rotations, and the vertical coordinate is unified to 5000 N·m. If the number of rotations of the torque curve is less than 7, the range of the horizontal coordinate is directly changed. If the number of rotations of the input torque curve is greater than 7, the value is taken upwards from the maximum number of rotations, ultimately retaining the torque value of the last 7 rotations of the original data. After unifying the horizontal and vertical coordinates of the torque curve, denoising is required for each data point in the torque curve. To retain as much detail as possible in the original data and maximize the noise reduction function, this embodiment uses Gaussian filtering to smooth the data points in the torque curve to remove high-frequency noise. The specific steps are as follows:

[0103] Step a: Construct a Gaussian filter with a filter size of 3 and a standard deviation of 1. The Gaussian filter formula is shown below:

[0104]

[0105] Where G(x) is a Gaussian filter, x is the torque value, and σ is the standard deviation of the torque value.

[0106] Step b: For each data point in the upper torque curve, align the center of the Gaussian filter with that point and calculate the convolution result between the filter and the curve. The convolution calculation can use linear convolution operation, that is, weighted summation of the elements of the filter and the data at the corresponding positions of the curve.

[0107] Step c: Repeat the above steps to perform filtering operations on each data point on the curve in turn.

[0108] In some embodiments, to eliminate differences between torque curve data, the denoised data is normalized. After normalization, the original data values ​​are all on the same order of magnitude. The specific steps are as follows: First, analyze the torque curve dataset to determine the range of torque values, obtain the maximum and minimum values, and then for each torque value T, apply the following formula to normalize each torque value, converting it to a range between 0 and 1:

[0109]

[0110] Among them, T noris the normalized torque value, T is the denoised torque value, min is the minimum torque value, and max is the maximum torque value.

[0111] To ensure the quality of the augmented dataset, this specification employs a Deep Convolutional Generative Adversarial Network (DCGAN) model as the data augmentation model in its embodiments. DCGAN uses convolutional neural networks to construct a generator and a discriminator to generate high-quality data. The generator network learns the distribution characteristics of real torque curve data and maps random noise vectors to the torque curve data space through deconvolution operations, generating realistic and diverse torque curve data. The discriminator network distinguishes between real and generated torque curves, and adversarial training enables the generator network to generate even more realistic torque curve data. A schematic diagram of the data augmentation model structure in this specification embodiment is shown below. Figure 3 As shown, the generator of this model contains six two-dimensional deconvolutional layers, while the discriminator consists of five two-dimensional convolutional layers. The generator takes a 100-dimensional random noise vector as input and scales it through a series of deconvolutional layers to generate data of the same size as the real samples. The discriminator's recognition process is similar to a convolutional neural network performing a classification task. It downsamples and extracts features from both real and generated samples, and generates a scalar of dimension 1 through the output layer, representing the probability that the input sample is a real sample. The closer the probability value is to 1, the greater the probability that the sample is a real sample.

[0112] In some embodiments, reference is made to Figure 4 The training process of the data augmentation model includes the following steps:

[0113] S1: Initialize the weight parameters of the generator and discriminator using a Gaussian distribution;

[0114] S2: Input the random noise vector into the generator and output the generated torque curve;

[0115] S3: Input the generated torque curve and the real torque curve as training samples into the discriminator to obtain the probability value of each sample;

[0116] S4: Construct a first loss function based on the generated torque curve and the probability value, and construct a second loss function based on the real torque curve and the generated torque curve.

[0117] The first loss function represents the generator loss function. By minimizing this loss function, the samples generated by the generator are made closer to the real samples. The calculation formula is as follows:

[0118] L gen = -log(D(G(z)));

[0119] Among them, L genLet G(z) represent the generator loss function, z represent the random noise vector, G(z) represent the generated torque curve, and D(G(z)) represent the probability that the discriminator judges the generated torque curve as real.

[0120] The second loss function is the discriminator loss function. Minimizing this loss function improves the discriminator's ability to distinguish between generated samples and real samples. Its calculation formula is as follows:

[0121] L dis = -logD(x) - log(1 - D(G(z)));

[0122] Among them, L dis Let represent the discriminator loss function, x represent the true torque curve, D(x) represent the probability that the discriminator judges the true torque curve as true, z represent the random noise vector, G(z) represent the generated torque curve, and D(G(z)) represent the probability that the discriminator judges the generated torque curve as true.

[0123] S5: Update the weight parameters of the generator and discriminator using the backpropagation algorithm based on the first loss function and the second loss function.

[0124] To balance the training progress of the generator and discriminator during training and avoid the problem of the generator's loss function being unable to be optimized, this embodiment of the specification uses an asynchronous weight optimization method to update the weight parameters of the generator and discriminator. Specifically, after each training session, the generator's weight parameters are updated using the backpropagation algorithm according to the first loss function; after the number of training sessions reaches a preset value, the discriminator's weight parameters are updated using the backpropagation algorithm according to the second loss function. In this embodiment of the specification, the preset number of training sessions is 4, meaning that the generator's weight parameters are updated with each training session, while the discriminator's weight parameters are updated every four training sessions. Furthermore, the weights used to update the discriminator are obtained by averaging the weight changes over four training sessions, and the discriminator's weight update formula is as follows:

[0125]

[0126] Among them, W dis L represents the weight parameters of the discriminator. dis (i) represents the loss function of the discriminator in the i-th training iteration, w(i) represents the weight parameters in the i-th training iteration, and n represents the preset number of training iterations.

[0127] S6: Repeat steps S2-S5 until the predetermined number of training iterations is reached.

[0128] Therefore, through the above training process, a trained data augmentation model can be obtained. Using this data augmentation model to augment the original tubing upper torque curve dataset, a large number of data samples containing normal torque curves and abnormal torque curves can be obtained. In order to meet the input data format of the convolutional neural network, the curve data needs to be converted into an image format to obtain the upper torque curve image, thereby obtaining the final tubing upper torque curve dataset.

[0129] To automatically extract features from torque curve images and thus achieve automatic classification of torque curves, this embodiment of the specification constructs an upper torque curve classification model based on a convolutional neural network, referring to... Figure 5 As shown, the torque curve classification model includes an input layer, a feature extraction layer, and an output layer. The input layer receives the torque curve dataset from the pipeline. The feature extraction layer consists of three stacked convolutional layers. Each convolutional layer uses a set of learnable filters or kernels to extract features and details from the torque curves, capturing features from the pipeline torque curve data to obtain a feature map. To accelerate the training process and improve model stability, a max-pooling layer and a batch normalization layer are applied after each convolutional layer. The max-pooling layer downsamples the feature map, reducing its size and thus accelerating training. The batch normalization layer normalizes the images in each training batch, thereby improving the stability of model training. Furthermore, to reduce overfitting, a Dropout layer is introduced in the last convolutional layer. After feature extraction in the last convolutional layer, Dropout randomly discards some of the neuron outputs, thereby reducing the complex dependencies between neurons in the convolutional neural network. The output layer consists of a fully connected layer and a softmax layer, which are used to map the feature representation to the corresponding probability distribution and output the final curve classification result.

[0130] In some embodiments, the training data augmented by DCGAN is first input into the CNN model for training. Then, the cross-entropy loss function is used to measure the difference between the model output and the true label. Finally, the Adam optimization algorithm is used to minimize the loss function and update the model's weights and biases until a preset number of training rounds are reached. Thus, a well-trained upper torque curve classification model can be obtained.

[0131] Based on the same inventive concept, referring to Figure 6 In some embodiments, a method for evaluating the quality of tubing connections may include:

[0132] S601: Receives torque and number of turns data of the oil pipe under test, and obtains the upper torque curve based on the torque and number of turns data;

[0133] S602: Input the upper torque curve into the upper torque curve classification model trained by the method described in any of the foregoing embodiments for classification and identification, and obtain the upper torque curve classification result of the oil pipe under test;

[0134] S603: Evaluate the up-clamping quality of the oil pipe under test based on the classification results of the up-clamping torque curve.

[0135] In some embodiments, the torque curve of the tubing under test is input into a torque curve classification model trained using the method described in any of the aforementioned embodiments. This yields a classification result for the torque curve of the tubing under test. The classification result includes the type of torque curve and its corresponding probability. If the probability of "standard torque curve" is the highest in the output probability, it indicates that the curve type of the tubing under test is a standard torque curve, and so on. The tubing under test is then evaluated for its torque curve quality based on the classification result. If the classification result is a standard torque curve, the evaluation result is considered normal, indicating that the connection quality of the tubing meets the standard requirements. If the classification result is either an abnormally fluctuating torque curve or a torque curve without an inflection point, the evaluation result is considered abnormal. For torque curves with abnormal evaluation results, the curve can be analyzed to identify the cause of the abnormality, and corresponding countermeasures can be provided.

[0136] Based on the above-described method for constructing a quality evaluation model for tubing top fasteners, this specification also provides a corresponding apparatus for constructing such a model. The apparatus may include a system (including a distributed system), software (application), module, component, server, client, etc., using the method described in this specification, combined with necessary hardware implementation. Based on the same innovative concept, the apparatuses in one or more embodiments provided in this specification are as described in the following embodiments. Since the implementation schemes and methods for solving the problem are similar, the implementation of specific apparatuses in this specification can refer to the implementation of the aforementioned method, and repeated details will not be elaborated further. As used below, the terms "unit" or "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatuses described in the following embodiments are preferably implemented in software, hardware implementations, or a combination of software and hardware, are also possible and contemplated.

[0137] Specifically, Figure 7 This is a schematic diagram of the module structure of an embodiment of a tubing fastening quality evaluation model construction device provided in this specification. (Refer to...) Figure 7 As shown in the embodiments of this specification, a tubing fastening quality evaluation model construction device includes:

[0138] Module 701 is used to acquire the raw torque curve dataset on the tubing.

[0139] The expansion module 702 is used to expand the dataset using a pre-trained data expansion model to obtain an expanded oil pipe torque curve dataset.

[0140] Module 703 is used to build an upper torque curve classification model based on a convolutional neural network;

[0141] Training module 704 is used to train the upper torque curve classification model using the expanded upper torque curve dataset of the tubing to obtain a trained upper torque curve classification model.

[0142] Based on the same inventive concept, and corresponding to the above-described tubing splice quality evaluation method, some embodiments of this specification also provide a tubing splice quality evaluation device, see reference. Figure 8 As shown, in some embodiments, the apparatus may include:

[0143] The receiving module 801 is used to receive the torque and number of turns data of the oil pipe under test, and to obtain the upper torque curve based on the torque and number of turns data;

[0144] The classification and recognition module 802 is used to input the upper torque curve into the upper torque curve classification model trained by the method described in any of the foregoing embodiments for classification and recognition, so as to obtain the upper torque curve classification result of the oil pipe under test;

[0145] Evaluation module 803 is used to evaluate the up-coil quality of the oil pipe under test based on the classification results of the up-coil torque curve.

[0146] The beneficial effects obtained by the apparatus provided in the embodiments of this specification are consistent with the beneficial effects obtained by the methods described above, and will not be repeated here.

[0147] Some embodiments of this specification also provide an intelligent tubing buckle torque meter, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes instructions for a tubing buckle quality evaluation method when executing the computer program.

[0148] Some embodiments of this specification also provide a tubing splice quality evaluation system, see reference Figure 9 As shown, the tubing connection quality evaluation system includes: tubing 1, tubing wrench 2, intelligent connection torque meter 3, and sealing buckle 4;

[0149] The sealing buckle 4 is used to connect the oil pipe 1 to the pipeline inside the oil well to prevent leakage during oil and gas transmission.

[0150] The tubing clamp 2 is used to fasten the tubing 1 and generate torque and number of turns data;

[0151] The intelligent torque meter 3 is connected to the tubing clamp 2 and is used to receive the torque and number of turns data, evaluate the quality of tubing connection based on the torque and number of turns data, and output the evaluation result.

[0152] In some embodiments, the tubing wrench contains a built-in torque sensor. During the tightening process, the torque sensor converts the sensed number of turns and torque data into electrical signals, which are then transmitted to the intelligent tightening torque meter. The intelligent tightening torque meter can then output the torque curve type and evaluation result of the corresponding tubing based on the torque and number of turns data. The classification speed is fast and the classification result is highly accurate.

[0153] In some embodiments, reference is made to Figure 10 As shown, the intelligent upper-clamp torque meter 3 further includes: a communication module 311, a classification and recognition module 312, a drawing and display module 313, and a power supply module 314. The communication module 311 is used for data interaction with the tubing clamp 2, receiving torque and rotation data sent by the tubing clamp 2 and sending it to the drawing and display module. Furthermore, the communication module 311 is also used for data transmission between modules of the intelligent upper-clamp torque meter. By setting parameters such as serial port number, baud rate, parity bit, data bits, and stop bits, the parameters of the data receiving end and the data sending end can be kept consistent, thus enabling data transmission between modules. Specifically, the working principle of the communication module is as follows: It opens the serial port according to communication requirements, listens to the serial port status, and displays the current serial port connection status. If it displays "not connected," it means that the intelligent upper-clamp torque meter has not established communication with the on-site tubing clamp; if it displays "connected," it means that the intelligent upper-clamp torque meter has established communication with the on-site tubing clamp, receives the torque value, rotation number, and time information sent by the tubing clamp, and transmits the above information to the drawing and display module and the classification and recognition module.

[0154] The drawing and display module 313 can be a common electronic output device, used to receive the torque and revolution count data, draw and display the torque curve image based on the torque and revolution count data. Furthermore, the drawing and display module is also used to display serial port information, well information, classification result information, verification result information, and file directory information. Serial port information includes serial port number, baud rate, parity bit, data bits, and stop bits; well information includes well depth, maximum torque value, optimal torque value, minimum torque value, maximum shoulder torque, minimum shoulder torque, maximum abscissa, save path, and construction time; classification result information includes the tubing quality evaluation result, curve type, and classification probability; verification result information includes classification results and verification results, where verification results refer to the results of manual verification of the model classification results. If the model classification results are inconsistent with the perceived results or if the model classification results contain errors, manual verification can be used to recalibrate the torque curve category and tubing quality evaluation results, and the model classification results and manual verification results are saved to the corresponding documents; file directory information refers to the storage information corresponding to the documents containing classification results and verification results, including file name, size, type, file path, and time.

[0155] The classification and recognition module 312 is used to classify and recognize the upper torque curve to obtain the classification result of the upper torque curve. This can be understood as encapsulating the above-mentioned upper torque curve classification model and embedding it into the classification and recognition module, thereby realizing automatic classification of the tubing torque curve and evaluation of the upper torque curve quality.

[0156] The power supply module 314 is used to provide power to the communication module, the classification and recognition module and the drawing and display module.

[0157] In some embodiments, the intelligent torque meter further includes an indicator light 315 and an alarm module 316; wherein the indicator light 315 is used to indicate whether the power supply of the intelligent torque meter is on; the alarm module 316 can be a buzzer alarm, used to issue an alarm sound when the torque meter's torque evaluation result is detected to be abnormal.

[0158] In some embodiments, the intelligent torque meter further includes a chassis 317, which is used to house and fix the communication module, classification and identification module, drawing and display module, and power supply module, and to shield against external electromagnetic radiation signals. In some embodiments, a power interface can be provided on the outside of the chassis to power the internal modules via an external power supply, and a power converter can be placed inside the chassis to distribute the external power to each module. In some embodiments, to ensure the stability of data transmission and network communication, multiple USB ports and network ports can be provided on the outside of the chassis, while a wireless network card can be provided inside the chassis.

[0159] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the acquisition, storage, use, and processing of data in the technical solutions described in the embodiments of this application all comply with relevant regulations.

[0160] Corresponding to, for example Figures 1 to 6 In addition to the method shown, embodiments of this specification also provide a computer-readable storage medium storing a computer program that, when executed by a processor, performs the steps of the above-described method.

[0161] This specification also provides computer-readable instructions, wherein when a processor executes the instructions, the program therein causes the processor to perform the following... Figures 1 to 6 The method shown.

[0162] This specification also provides a computer program product, including at least one instruction or at least one program segment, wherein the at least one instruction or the at least one program segment is loaded and executed by a processor to achieve the following: Figures 1 to 6 The method shown.

[0163] It should be understood that in the various embodiments of this specification, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this specification.

[0164] It should also be understood that, in the embodiments of this specification, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this specification generally indicates that the preceding and following related objects have an "or" relationship.

[0165] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this specification can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this specification.

[0166] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0167] In the several embodiments provided in this specification, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces, devices, or units, or they may be electrical, mechanical, or other forms of connection.

[0168] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments described in this specification, depending on actual needs.

[0169] Furthermore, the functional units in the various embodiments of this specification can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0170] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this specification, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this specification. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0171] This specification uses specific embodiments to illustrate the principles and implementation methods of this specification. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this specification. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this specification. Therefore, the content of this specification should not be construed as a limitation of this specification.

Claims

1. A method for constructing a quality evaluation model for oil pipe couplings, characterized in that, The method includes: Obtain the raw torque curve dataset on the tubing; The dataset is augmented using a pre-trained data augmentation model to obtain an augmented tubing torque curve dataset. A classification model for upper torque curves is constructed based on a convolutional neural network; The expanded tubing upper torque curve dataset is used to train the upper torque curve classification model to obtain a trained upper torque curve classification model.

2. The method according to claim 1, characterized in that, The data augmentation model is a deep convolutional generative adversarial network model, which includes a generator and a discriminator. The training process of the data augmentation model includes: S1: Initialize the weight parameters of the generator and discriminator using a Gaussian distribution; S2: Input the random noise vector into the generator and output the generated torque curve; S3: Input the generated torque curve and the real torque curve as training samples into the discriminator to obtain the probability value of each sample; S4: Construct a first loss function based on the generated torque curve and the probability value, and construct a second loss function based on the real torque curve and the generated torque curve; S5: Update the weight parameters of the generator and discriminator according to the first loss function and the second loss function using the backpropagation algorithm; S6: Repeat steps S2-S5 until the predetermined number of training iterations is reached.

3. The method according to claim 2, characterized in that, The step of updating the weight parameters of the generator and discriminator using the backpropagation algorithm based on the first loss function and the second loss function includes: After each training session, the generator's weight parameters are updated using the backpropagation algorithm based on the first loss function. After the number of training iterations reaches a preset value, the weight parameters of the discriminator are updated using the backpropagation algorithm based on the second loss function.

4. The method according to claim 1, characterized in that, The upper torque curve classification model includes an input layer, a feature extraction layer, and an output layer; wherein... The input layer is used to receive torque curve data from the tubing. The feature extraction layer consists of multiple stacked convolutional layers, pooling layers, and batch normalization layers, used to capture features in the torque curve data on the tubing to obtain feature representations; The output layer consists of a fully connected layer and a Softmax layer, which is used to map the feature representation to the corresponding probability distribution and output the final curve classification result.

5. A method for evaluating the quality of oil pipe couplings, characterized in that, The method includes: Receive the torque and number of turns data of the oil pipe under test, and obtain the upper torque curve based on the torque and number of turns data; The upper torque curve is input into the upper torque curve classification model trained by the method of any one of claims 1-4 for classification and identification, so as to obtain the upper torque curve classification result of the oil pipe under test; The quality of the upper clamping of the oil pipe under test is evaluated based on the classification results of the upper clamping torque curve.

6. The method according to claim 5, characterized in that, The classification results of the upper torque curve include: standard upper torque curve, abnormal fluctuation upper torque curve, and upper torque curve without inflection point; The evaluation of the connection quality of the oil pipe under test based on the classification results of the connection torque curve includes: If the classification result of the upper buckle torque curve is a standard upper buckle torque curve, then the upper buckle quality evaluation result of the oil pipe under test is normal. If the classification result of the upper buckling torque curve is either an abnormal fluctuation upper buckling torque curve or an upper buckling torque curve without an inflection point, then the upper buckling quality evaluation result of the oil pipe under test is abnormal.

7. A device for constructing a quality evaluation model for oil pipe couplings, characterized in that, The device includes: The acquisition module is used to acquire the raw torque curve dataset on the tubing. An expansion module is used to expand the dataset using a pre-trained data expansion model to obtain an expanded oil pipe torque curve dataset. The building block is used to construct an upper torque curve classification model based on a convolutional neural network; The training module is used to train the upper torque curve classification model using the expanded upper torque curve dataset of the tubing, so as to obtain the trained upper torque curve classification model.

8. A quality evaluation device for oil pipe couplings, characterized in that, The device includes: The receiving module is used to receive the torque and number of turns data of the oil pipe under test, and to obtain the upper torque curve based on the torque and number of turns data; The classification and recognition module is used to input the upper torque curve into the upper torque curve classification model trained by the method of any one of claims 1-4 for classification and recognition, so as to obtain the classification result of the upper torque curve of the oil pipe under test; The evaluation module is used to evaluate the up-coil quality of the oil pipe under test based on the classification results of the up-coil torque curve.

9. An intelligent torque meter, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 5 to 6.

10. A quality evaluation system for oil pipe couplings, characterized in that, The tubing connection quality evaluation system includes: tubing, tubing wrenches, and an intelligent connection torque meter; wherein... The tubing clamp is used to fasten the tubing and generate torque and turn data; The intelligent torque meter is connected to the tubing wrench and is used to receive the torque and number of turns data, evaluate the quality of tubing connection based on the torque and number of turns data, and output the evaluation result.

11. The tubing connection quality evaluation system according to claim 10, characterized in that, The intelligent torque meter also includes: a communication module, a classification and recognition module, a drawing and display module, and a power supply module; wherein The communication module is used to interact with the tubing wrench, receive torque and rotation data sent by the tubing wrench and send them to the drawing and display module; The drawing and display module is used to receive the torque and revolution count data, draw the upper torque curve image based on the torque and revolution count data, and display it; The classification and recognition module is used to classify and recognize the upper torque curve to obtain the classification result of the upper torque curve; The power supply module is used to provide power to the communication module, the classification and recognition module, and the drawing and display module.

12. The tubing connection quality evaluation system according to claim 10, characterized in that, The intelligent upper buckle torque meter also includes: indicator lights and an alarm module; wherein The indicator light is used to indicate whether the power supply of the intelligent upper clamp torque meter is on; The alarm module is used to issue an alarm sound when the quality evaluation result of the top tap of the oil pipe under test is detected to be abnormal.

13. The tubing connection quality evaluation system according to claim 11, characterized in that, The intelligent upper buckle torque meter also includes: a chassis; The chassis is used to house and fix the communication module, classification and identification module, drawing and display module and power supply module, and to shield external electromagnetic radiation signals.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 6.

15. A computer program product, characterized in that, It includes at least one instruction or at least one program segment, said at least one instruction or said at least one program segment being loaded and executed by a processor to implement the method as claimed in any one of claims 1 to 6.

Images