Residual life prediction method for corrosion fatigue monitoring of high-strength drill rod steel

Abstract

The invention relates to the technical field of drill rod monitoring, in particular to a residual life prediction method for high-strength drill rod steel corrosion fatigue monitoring, which comprises the following steps: acquiring working data of a drill rod, including stress and corrosion environment data; quantifying the influence of the stress and the corrosion environment on the fatigue life, and establishing a stress-corrosion environment-fatigue life quantitative relationship; constructing an improved damage accumulation model considering corrosion rate and crack propagation to obtain a preliminary drill rod residual life prediction result; fusing real-time monitoring data and historical operation data by using a deep learning algorithm, introducing an attention mechanism to dynamically adjust the weight of the model, and correcting and updating the prediction result of the remaining life of the drill rod; and setting multi-stage safety thresholds, and carrying out graded early warning on the failure time of the drill rod. By means of the method, intelligent management of the whole life cycle of the drill rod can be achieved, and the safety, reliability and economical efficiency of drilling operation are improved.

Description

Technical Field

[0001] The invention relates to the technical field of drill pipe monitoring, in particular to a remaining life prediction method for high-strength drill pipe steel corrosion fatigue monitoring. Background Art

[0002] In oil drilling operations, drill pipe is a key drilling tool, and its performance directly affects the safety and efficiency of drilling operations. During service, drill pipe is very prone to failure due to the complex and variable loads (including static loads and dynamic loads) and the influence of various factors such as corrosion and wear. The main forms of drill pipe failure include excessive deformation, fracture and surface damage, among which corrosion fatigue fracture is one of the main causes of drill pipe failure. Drill pipe failure will not only lead to the suspension or interruption of drilling operations, but also cause the scrapping and replacement of drill pipe, thereby increasing drilling costs. Therefore, in order to reduce the economic losses and safety risks caused by drill pipe failure, it is particularly important to predict the remaining life of drill pipe.

[0003] During the service process, drill pipes will be affected by various environmental factors, which may easily cause the performance of drill pipe materials to change, thus affecting their fatigue life. Therefore, how to integrate the interference of various environmental factors and conduct fatigue monitoring on drill pipes to predict the remaining life of drill pipes and evaluate their safety in use is an urgent problem to be solved.

[0004] Therefore, a remaining life prediction method for corrosion fatigue monitoring of high-strength drill pipe steel is proposed. Summary of the invention

[0005] The present invention provides the following technical solutions:

[0006] A method for predicting the remaining life of high-strength drill pipe steel corrosion fatigue monitoring, comprising:

[0007] Collect the working data of the drill pipe, including stress and corrosion environment data;

[0008] Simulate real working conditions to conduct corrosion fatigue experiments, quantify the effects of stress and corrosion environment on fatigue life, and establish a quantitative relationship between stress, corrosion environment and fatigue life;

[0009] Based on the established quantitative relationship, an improved damage accumulation model considering corrosion rate and crack propagation was constructed. Combined with multi-source monitoring data, fatigue damage was quantitatively evaluated and preliminary prediction results of the remaining life of the drill pipe were obtained.

[0010] Use deep learning algorithms to fuse real-time monitoring data with historical operation data, introduce attention mechanisms to dynamically adjust model weights, and calibrate and update drill pipe remaining life prediction results;

[0011] According to the remaining life prediction results, multi-level safety thresholds are set to provide graded warnings for the failure time of the drill pipe.

[0012] Preferably, the stress σ, vibration frequency f and corrosion environment parameters of the drill pipe are collected, wherein the corrosion environment parameters include pH value and chloride ion concentration.

[0013] The collected data are preprocessed, including outlier removal, denoising and smoothing, and numerical data normalization.

[0014] Preferably, the actual working conditions of the drill pipe are simulated, and a multi-factor orthogonal test method is used to obtain different stress levels σ, pH values ​​and chloride ion concentrations under different temperature gradients. The influence of factors on the fatigue life N of drill pipe;

[0015] The drill pipe samples are subjected to different combinations of stress and corrosive environmental conditions through a fatigue testing machine, and the fatigue life data of the drill pipe are recorded.

[0016] Preferably, the corrosion fatigue influence law modeling is based on the test data to construct a quantitative relationship model of stress-corrosion environment-fatigue life:

[0017]

[0018] Where N is the life of the drill pipe under corrosive environment and stress conditions, C is the material constant, m, a, b are the influence coefficients of stress, pH value and chloride ion concentration respectively;

[0019] The nonlinear least square method is used to fit the test data and determine the parameters of the quantitative relationship model of stress-corrosion environment-fatigue life.

[0020] Preferably, the corrosion rate v is established for the drill pipe service environment. c Computational model:

[0021]

[0022] Where k is the rate constant, E a is the apparent activation energy, R is the gas constant, T is the absolute temperature, n and p are the reaction orders; the corrosion rate v is determined by corrosion weight loss test and electrochemical test method c Calculate the parameters of the model;

[0023] Using fracture mechanics methods, the stress intensity factor range ΔK and crack growth rate are constructed Relationship:

[0024]

[0025] Among them, C and η are material constants.

[0026] Preferably, based on the constructed linear damage accumulation model, the corrosion rate v is introduced c and crack growth rate The correction factor λ is used to construct an improved damage accumulation model:

[0027]

[0028] Where D is the damage variable, n l is the number of cycles under the lth stress level, N l is the drill pipe life under the lth stress level; λ is the comprehensive corrosion rate v c and crack growth rate Correction factor of

[0029] Combined with multi-source data from real-time monitoring, the improved damage accumulation model is used to calculate the fatigue damage value D of the current drill pipe, and the preliminary remaining life prediction results are obtained:

[0030]

[0031] Where RUL is the remaining useful life, v D is the damage accumulation rate, calculated from the monitoring data.

[0032] Preferably, a deep learning prediction model integrating real-time monitoring data and historical operation data is constructed. The deep learning prediction model adopts a structure combining a long short-term memory network LSTM model with an attention mechanism:

[0033] h t =LSTM(x t ,h t-1 ),

[0034] α ti =Attention(h t ,h i )

[0035]

[0036] Among them, x t is the input feature vector at time t, h t is the hidden state vector output by the LSTM model at time t, α ti is the attention weight of time t to time i, is the remaining life of the drill pipe predicted at time t, f NN It is a fully connected neural network.

[0037] Preferably, the deep learning prediction model is trained offline using historical operation data, and the model performance is evaluated and hyperparameters are tuned using a cross-validation method;

[0038] During the operation of the drill pipe, the model is updated online using real-time monitoring data. The sliding window method is used to dynamically adjust the model input to continuously predict the remaining life of the drill pipe. At the same time, new monitoring data is collected to perform incremental training on the model.

[0039] Preferably, according to the prediction result of the remaining life of the drill pipe Set the third-level warning threshold τ 1 , τ 2 , τ 3 , and give early warning of the failure time of the drill pipe:

[0040] when When the drill pipe enters the warning period, a first-level warning is issued, indicating that manual attention is required;

[0041] when When the drill pipe is in a sub-healthy state, a secondary warning is issued, indicating that a maintenance plan needs to be formulated;

[0042] when When the drill pipe fails, a third-level warning is issued, indicating that the drill pipe has failed and needs to be replaced.

[0043] A system for predicting a remaining life of high-strength drill pipe steel corrosion fatigue monitoring method, comprising: a data acquisition module, a drill pipe life quantification module, a drill pipe remaining life prediction module, a drill pipe remaining life update module, and a safety warning module;

[0044] The data acquisition module is used to collect the working data of the drill pipe, including stress and corrosion environment data;

[0045] The drill pipe life quantitative module performs corrosion fatigue experiments based on simulated real working conditions, quantifies the effects of stress and corrosion environment on fatigue life, and establishes a quantitative relationship between stress, corrosion environment, and fatigue life;

[0046] The drill pipe remaining life prediction module constructs an improved damage accumulation model that takes into account corrosion rate and crack propagation based on the established quantitative relationship, and quantitatively evaluates fatigue damage in combination with multi-source monitoring data to obtain a preliminary drill pipe remaining life prediction result;

[0047] The drill pipe remaining life update module uses a deep learning algorithm to fuse real-time monitoring data with historical operation data, introduces an attention mechanism to dynamically adjust the model weights, and corrects and updates the drill pipe remaining life prediction results;

[0048] The safety warning module sets a multi-level safety threshold according to the remaining life prediction result, and provides a graded warning for the failure time of the drill pipe.

[0049] Beneficial effects of the present invention: By collecting the working data of the drill pipe, including stress and corrosion environment data, the present invention can fully understand the actual stress state and corrosion environment conditions of the drill pipe, and provide necessary data support for subsequent corrosion fatigue life prediction.

[0050] The present invention simulates real working conditions to conduct corrosion fatigue experiments, quantifies the effects of stress and corrosion environment on fatigue life, and establishes a quantitative relationship between stress-corrosion environment-fatigue life, which can reveal the inherent mechanism of corrosion fatigue failure of drill pipes and provide a theoretical basis for the design and material selection of drill pipes.

[0051] Based on the established quantitative relationship, the present invention constructs an improved damage accumulation model that takes into account corrosion rate and crack propagation, and combines multi-source monitoring data to quantitatively evaluate fatigue damage, which can more comprehensively and accurately describe the damage process of the drill pipe and improve the reliability of the remaining life prediction.

[0052] The present invention utilizes a deep learning algorithm to fuse real-time monitoring data with historical operation data, and introduces an attention mechanism to dynamically adjust the model weights, which can fully mine the degradation laws and fault precursors contained in the data and improve the intelligence level of remaining life prediction.

[0053] The present invention sets a multi-level safety threshold according to the remaining life prediction result, and provides a graded early warning for the failure time of the drill pipe, so that the potential risks of the drill pipe can be identified in advance and targeted prevention and maintenance measures can be taken.

[0054] The present invention collects the working data of the drill pipe, combines corrosion fatigue tests and theoretical modeling, constructs a damage accumulation model that takes into account the influence of multiple factors, uses a deep learning algorithm to predict the remaining life, and sets up a graded early warning mechanism, so as to realize intelligent management of the entire life cycle of the drill pipe and improve the safety, reliability and economy of drilling operations.

[0055] The method of the present invention can effectively reduce the risk of drill pipe failure, extend the service life of the drill pipe, reduce unplanned downtime, improve drilling efficiency, and provide important technical support for oil and gas field development. At the same time, the method of the present invention can be promoted and applied to the health monitoring and life prediction of other key drilling equipment, and has broad application prospects and economic value. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0057] in:

[0058] Figure 1 This is a flow chart of a method for predicting the remaining life of high-strength drill pipe steel corrosion fatigue monitoring according to the present invention;

[0059] Figure 2 This is a structural diagram of a remaining life prediction system for corrosion fatigue monitoring of high-strength drill pipe steel according to the present invention. DETAILED DESCRIPTION

[0060] In order to make the above-mentioned purposes, features and advantages of the present invention more understandable, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0061] Example 1

[0062] Reference Figure 1 The first embodiment of the present invention provides a method for predicting the remaining life of high-strength drill pipe steel corrosion fatigue monitoring.

[0063] Step 1: Collect the working data of the drill pipe, including stress and corrosion environment data.

[0064] Collect the stress σ of the drill pipe and the corrosion environment parameters, including pH value and chloride ion concentration.

[0065] The collected data are preprocessed, including outlier removal, denoising and smoothing, and numerical data normalization.

[0066] Considering the structural characteristics and stress conditions of the drill pipe, the density of measuring points is increased in stress concentration areas and fatigue sensitive areas.

[0067] When preprocessing data, wavelet transform method is used to filter the noise and improve the signal-to-noise ratio.

[0068] Step 2: Conduct corrosion fatigue experiments simulating real working conditions, quantify the effects of stress and corrosion environment on fatigue life, and establish a quantitative relationship between stress-corrosion environment-fatigue life.

[0069] Simulating the actual working conditions of the drill pipe, a multi-factor orthogonal test method was used to study the effects of different stress levels σ, pH values ​​and chloride ion concentrations under different temperature gradients. The influence of factors on the fatigue life N of drill pipe.

[0070] The drill pipe samples are subjected to different combinations of stress and corrosive environmental conditions through a fatigue testing machine, and the fatigue life data of the drill pipe are recorded.

[0071] When conducting orthogonal factor experiments, stress level, pH value, chloride ion concentration and temperature were selected as the main influencing factors, and their interactions were considered; during the experiment, the fracture morphology was observed using a scanning electron microscope (SEM) to analyze the corrosion fatigue failure mechanism; the elemental composition of the fracture was analyzed using an energy dispersive spectrometer (EDS) to study the influence of micro-region composition changes on fatigue fracture; through a comprehensive analysis of fatigue life data and microscopic analysis results, the relationship between macroscopic fatigue performance and microscopic corrosion behavior was established.

[0072] Modeling of the influence of corrosion fatigue Based on test data, a quantitative relationship model of stress-corrosion environment-fatigue life is constructed:

[0073]

[0074] Wherein, N is the drill pipe life under corrosive environment and stress conditions, C is the material constant, and m, a, and b are the influence coefficients of stress, pH value, and chloride ion concentration, respectively.

[0075] The nonlinear least square method is used to fit the test data and determine the parameters of the quantitative relationship model of stress-corrosion environment-fatigue life.

[0076] Step 3: Based on the established quantitative relationship, an improved damage accumulation model considering corrosion rate and crack propagation is constructed. The fatigue damage is quantitatively evaluated by combining multi-source monitoring data to obtain preliminary prediction results of the remaining life of the drill pipe.

[0077] According to the corrosion kinetics research, the corrosion rate v is established for the drill pipe service environment. c Computational model:

[0078]

[0079] Where k is the rate constant, E a is the apparent activation energy, R is the gas constant, T is the absolute temperature, n and p are the reaction orders; the corrosion rate v is determined by corrosion weight loss test and electrochemical test method c Calculate the parameters of the model.

[0080] Using fracture mechanics methods, the stress intensity factor range ΔK and crack growth rate are constructed Relationship:

[0081]

[0082] Where C and η are material constants, which can be obtained from fatigue crack growth tests on compact tension (CT) specimens.

[0083] Based on the constructed linear damage accumulation model, the corrosion rate v is introduced cand crack growth rate The correction factor λ is used to construct an improved damage accumulation model:

[0084]

[0085] Where D is the damage variable, n l is the number of cycles under the lth stress level, N l is the drill pipe life under the lth stress level, which is calculated by the established quantitative relationship model; λ is the corrosion rate v c and crack growth rate The correction factor is determined by numerical fitting method.

[0086] Combined with multi-source data from real-time monitoring, the improved damage accumulation model is used to calculate the fatigue damage value D of the current drill pipe, and the preliminary remaining life prediction results are obtained:

[0087]

[0088] Where RUL is the remaining useful life, v D is the damage accumulation rate, calculated from the monitoring data.

[0089] Step 4: Use deep learning algorithms to fuse real-time monitoring data with historical operation data, introduce attention mechanisms to dynamically adjust model weights, and correct and update drill pipe remaining life prediction results.

[0090] Construct a deep learning prediction model that integrates real-time monitoring data and historical operation data. The deep learning prediction model uses a structure that combines the long short-term memory network LSTM model with the attention mechanism (Attention):

[0091] h t =LSTM(x t ,h t-1 ),

[0092] α ti =Attention(h t ,h i )

[0093]

[0094] Among them, x t is the input feature vector at time t, h t is the hidden state vector output by the LSTM model at time t, α ti is the attention weight of time t to time i, is the remaining life of the drill pipe predicted at time t, f NN It is a fully connected neural network.

[0095] The importance of different historical moments to the current prediction is dynamically adjusted through the attention mechanism to improve the model's ability to capture the drill pipe degradation trend; the appropriate loss function (such as mean absolute error MAE) and optimization algorithm (such as Adam) are selected to train the model.

[0096] During the model training process, the early stopping strategy is adopted to monitor the changes in the loss function of the validation set to prevent model overfitting. At the same time, the learning rate scheduler is used to dynamically adjust the learning rate to accelerate model convergence. In view of the large amount of drill pipe monitoring data and high annotation cost, the transfer learning method is introduced to initialize the network parameters using the pre-trained model under similar working conditions to reduce training time and annotation requirements.

[0097] The deep learning prediction model is trained offline using historical operation data, and the model performance is evaluated and hyperparameters are tuned using cross-validation methods.

[0098] During the operation of the drill pipe, the model is updated online using real-time monitoring data. The sliding window method is used to dynamically adjust the model input to continuously predict the remaining life of the drill pipe. At the same time, new monitoring data is collected to perform incremental training on the model.

[0099] During the online update process of the model, the adaptive window method is used to dynamically adjust the length of the input sequence to balance the computational efficiency and prediction accuracy. When the monitoring data fluctuates abnormally, the model retraining mechanism is triggered, and the model is fine-tuned using the newly collected data to improve the adaptability and robustness of the model.

[0100] Step 5: According to the remaining life prediction results, set multi-level safety thresholds to provide graded warnings for the failure time of the drill pipe.

[0101] According to the prediction results of the remaining life of the drill pipe Set the third-level warning threshold τ 1 , τ 2 , τ 3 , and provide early warning of the failure time of the drill pipe.

[0102] when When the drill pipe is in the early warning period, a first-level warning is issued, indicating that the drill pipe has entered the early warning period and requires manual attention.

[0103] when When the drill pipe is in a sub-healthy state, a secondary warning is issued, indicating that the drill pipe has entered a sub-healthy state and a maintenance plan needs to be formulated.

[0104] when When the drill pipe fails, a level 3 warning is issued, indicating that the drill pipe is close to failure and needs to be replaced as soon as possible.

[0105] Through the graded early warning mechanism, the health status of the drill pipe can be timely understood, and the maintenance and replacement of the drill pipe can be reasonably arranged to ensure that the drilling operation is carried out safely and efficiently.

[0106] When determining the early warning threshold, factors such as the material properties, operating conditions and safety margin of the drill pipe are comprehensively considered, and a reasonable threshold range is determined by combining expert experience and data analysis. When the predicted remaining life is lower than the early warning threshold, the corresponding early warning measures are triggered, including increasing the monitoring frequency, adjusting drilling parameters, preparing spare drill pipes, etc., to minimize the impact of drill pipe failure on drilling operations.

[0107] Example 2

[0108] Reference Figure 2 The second embodiment of the present invention provides a remaining life prediction system for high-strength drill pipe steel corrosion fatigue monitoring.

[0109] The system comprises: a data acquisition module, a drill rod life quantification module, a drill rod remaining life prediction module, a drill rod remaining life update module and a safety early warning module.

[0110] The data acquisition module is used to collect the working data of the drill pipe, including stress and corrosion environment data.

[0111] The drill pipe life quantitative module performs corrosion fatigue experiments based on simulated real working conditions, quantifies the effects of stress and corrosion environment on fatigue life, and establishes a quantitative relationship between stress-corrosion environment-fatigue life.

[0112] The drill pipe remaining life prediction module constructs an improved damage accumulation model that takes into account corrosion rate and crack propagation based on the established quantitative relationship, and quantitatively evaluates fatigue damage in combination with multi-source monitoring data to obtain a preliminary drill pipe remaining life prediction result.

[0113] The drill pipe remaining life update module uses a deep learning algorithm to fuse real-time monitoring data with historical operation data, introduces an attention mechanism to dynamically adjust model weights, and corrects and updates the drill pipe remaining life prediction results.

[0114] The safety warning module sets a multi-level safety threshold according to the remaining life prediction result, and provides a graded warning for the failure time of the drill pipe.

[0115] Example 3

[0116] The third embodiment of the present invention is different from the previous embodiment in that:

[0117] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0118] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0119] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0120] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0121] Additionally, in order to provide a concise description of exemplary embodiments, all features of an actual embodiment (ie, those features that are not relevant to the best mode presently contemplated for carrying out the invention or those that are not relevant to implementing the invention) may not be described.

[0122] It should be understood that in the development of any actual implementation, as in any engineering or design project, numerous implementation-specific decisions may be made. Such a development effort may be complex and time-consuming, but for those of ordinary skill having the benefit of this disclosure, without undue experimentation, the development effort will be a routine task of design, fabrication, and production.

[0123] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for predicting the remaining life of high-strength drill pipe steel corrosion fatigue monitoring, characterized in that: include: Collect the working data of the drill pipe, including stress and corrosion environment data; Simulate real working conditions to conduct corrosion fatigue experiments, quantify the effects of stress and corrosion environment on fatigue life, and establish a quantitative relationship between stress, corrosion environment and fatigue life; Based on the established quantitative relationship, an improved damage accumulation model considering corrosion rate and crack propagation was constructed. Combined with multi-source monitoring data, fatigue damage was quantitatively evaluated and preliminary prediction results of the remaining life of the drill pipe were obtained. Use deep learning algorithms to fuse real-time monitoring data with historical operation data, introduce attention mechanisms to dynamically adjust model weights, and calibrate and update drill pipe remaining life prediction results; According to the remaining life prediction results, multi-level safety thresholds are set to provide graded warnings for the failure time of the drill pipe.

2. The method for predicting the remaining life of high-strength drill pipe steel by corrosion fatigue monitoring according to claim 1 is characterized in that: Collect the stress σ, vibration frequency f and corrosion environment parameters of the drill pipe, the corrosion environment parameters include pH value and chloride ion concentration The collected data are preprocessed, including outlier removal, denoising and smoothing, and numerical data normalization.

3. The method for predicting the remaining life of high-strength drill pipe steel by corrosion fatigue monitoring according to claim 2 is characterized in that: Simulating the actual working conditions of the drill pipe, a multi-factor orthogonal test method was used to obtain different stress levels σ, pH values ​​and chloride ion concentrations under different temperature gradients. The influence of factors on the fatigue life N of drill pipe; The drill pipe samples are subjected to different combinations of stress and corrosive environmental conditions through a fatigue testing machine, and the fatigue life data of the drill pipe are recorded.

4. The method for predicting the remaining life of high-strength drill pipe steel corrosion fatigue monitoring according to claim 3 is characterized in that: Modeling of the influence of corrosion fatigue Based on test data, a quantitative relationship model of stress-corrosion environment-fatigue life is constructed: Where N is the life of the drill pipe under corrosive environment and stress conditions, C is the material constant, m, a, b are the influence coefficients of stress, pH value and chloride ion concentration respectively; The nonlinear least square method is used to fit the test data and determine the parameters of the quantitative relationship model of stress-corrosion environment-fatigue life.

5. The method for predicting the remaining life of high-strength drill pipe steel corrosion fatigue monitoring according to claim 4 is characterized in that: According to the service environment of drill pipe, the corrosion rate v is established c Computational model: Where k is the rate constant, E a is the apparent activation energy, R is the gas constant, T is the absolute temperature, n and p are the reaction orders; the corrosion rate v is determined by corrosion weight loss test and electrochemical test method c Calculate the parameters of the model; Using fracture mechanics methods, the stress intensity factor range ΔK and crack growth rate are constructed Relationship: Among them, C and η are material constants.

6. The method for predicting the remaining life of high-strength drill pipe steel corrosion fatigue monitoring according to claim 5, characterized in that: Based on the constructed linear damage accumulation model, the corrosion rate v is introduced c and crack growth rate The correction factor λ is used to construct an improved damage accumulation model: Where D is the damage variable, n l is the number of cycles under the lth stress level, N l is the drill pipe life under the lth stress level; λ is the comprehensive corrosion rate v c and crack growth rate Correction factor of Combined with multi-source data from real-time monitoring, the improved damage accumulation model is used to calculate the fatigue damage value D of the current drill pipe, and the preliminary remaining life prediction results are obtained: Where RUL is the remaining useful life, v D is the damage accumulation rate, calculated from the monitoring data.

7. The method for predicting the remaining life of high-strength drill pipe steel corrosion fatigue monitoring according to claim 6 is characterized in that: Construct a deep learning prediction model that integrates real-time monitoring data and historical operation data. The deep learning prediction model uses a structure that combines the long short-term memory network LSTM model with the attention mechanism: h t =LSTM(x t ,h t-1 ), a ti =Attention(h t ,h i ) Among them, x t is the input feature vector at time t, h t is the hidden state vector output by the LSTM model at time t, α ti is the attention weight of time t to time i, is the remaining life of the drill pipe predicted at time t, f NN It is a fully connected neural network.

8. The method for predicting the remaining life of high-strength drill pipe steel corrosion fatigue monitoring according to claim 7 is characterized in that: Use historical operation data to train the deep learning prediction model offline, and use cross-validation methods to evaluate model performance and tune hyperparameters; During the operation of the drill pipe, the model is updated online using real-time monitoring data. The sliding window method is used to dynamically adjust the model input to continuously predict the remaining life of the drill pipe. At the same time, new monitoring data is collected to perform incremental training on the model.

9. The method for predicting the remaining life of high-strength drill pipe steel corrosion fatigue monitoring according to claim 8, characterized in that: According to the prediction results of the remaining life of the drill pipe Set three-level warning thresholds τ1, τ2, and τ3 to warn of the failure time of the drill pipe: when When the drill pipe enters the warning period, a first-level warning is issued, indicating that manual attention is required; when When the drill pipe is in a sub-healthy state, a secondary warning is issued, indicating that a maintenance plan needs to be formulated; when When the drill pipe fails, a third-level warning is issued, indicating that the drill pipe has failed and needs to be replaced.

10. A system based on the method for predicting the remaining life of high-strength drill pipe steel corrosion fatigue monitoring according to any one of claims 1 to 9, characterized in that: include: Data acquisition module, drill pipe life quantification module, drill pipe remaining life prediction module, drill pipe remaining life update module and safety warning module; The data acquisition module is used to collect the working data of the drill pipe, including stress and corrosion environment data; The drill pipe life quantitative module performs corrosion fatigue experiments based on simulated real working conditions, quantifies the effects of stress and corrosion environment on fatigue life, and establishes a quantitative relationship between stress, corrosion environment, and fatigue life; The drill pipe remaining life prediction module constructs an improved damage accumulation model that takes into account corrosion rate and crack propagation based on the established quantitative relationship, and quantitatively evaluates fatigue damage in combination with multi-source monitoring data to obtain a preliminary drill pipe remaining life prediction result; The drill pipe remaining life update module uses a deep learning algorithm to fuse real-time monitoring data with historical operation data, introduces an attention mechanism to dynamically adjust the model weights, and corrects and updates the drill pipe remaining life prediction results; The safety warning module sets a multi-level safety threshold according to the remaining life prediction result, and provides a graded warning for the failure time of the drill pipe.

Images