Soluble bridge plug fault early warning method and system

Abstract

The invention discloses a soluble bridge plug fault early warning method and system, and relates to the technical field of soluble bridge plug fault early warning. According to the soluble bridge plug fault early warning method, based on a set time period, dissolving state time sequence data and dissolving environment time sequence data of a soluble bridge plug are obtained; based on a pre-trained time sequence feature perception model, respectively extracting a state evolution feature set and an environment evolution feature set; generating a state evolution characteristic index and an environment evolution characteristic index of the soluble bridge plug; based on the state evolution characteristic index and the environment evolution characteristic index, whether the soluble bridge plug has a fault risk or not is judged, and a fault alarm is sent when the fault risk exists. Abnormal behaviors in the dissolving process can be found in time, potential faults can be warned in advance, lagged diagnosis after traditional deblocking failure is avoided, and therefore operation efficiency is improved, and safety risks are reduced.

Description

Technical Field

[0001] This invention relates to the field of soluble bridge plug failure early warning technology, specifically to a soluble bridge plug failure early warning method and system. Background Technology

[0002] Dissolvable bridge plugs are a common sealing device used in oil and gas wells or other underground engineering projects. They are typically used to control fluid flow within the wellbore or to isolate pressure in different areas. During drilling, completion, or construction of oil and gas wells, the bridge plug forms a sealed space through contact with the wellbore wall, ensuring effective isolation of fluids or gases within the well. However, as downhole operations continue, the dissolution process of the bridge plug can be affected by environmental factors (such as downhole temperature, pressure, and fluid composition), potentially leading to bridge plug failure or malfunction.

[0003] During the dissolution process of bridge plugs, various environmental factors can cause malfunctions such as localized cracking, loosening, or even complete failure. Without timely detection and warning, bridge plug failures can severely impact subsequent operations, potentially leading to damage to downhole equipment, operational delays, or safety risks for downhole personnel. Therefore, timely identification of the bridge plug's dissolution state and potential failure risks is crucial for improving operational safety, reducing production costs, and optimizing operational efficiency.

[0004] The limitations of existing technologies include at least the following problems: In monitoring the operational status of soluble bridge plugs, existing technologies generally lack structured processing and intelligent early warning mechanisms for continuous dynamic data on the dissolution process. Most solutions rely solely on fixed-point parameter readings or downhole intervention feedback during the dissolution process, making it difficult to capture and analyze early abnormal evolution trends in the dissolution behavior. This leads to delayed fault identification and makes it easy to miss the optimal window for fault intervention. Especially in complex well conditions, the dissolution state of the bridge plug is highly coupled with environmental disturbances. If its dynamic behavior characteristics are not detected in a timely manner, problems such as bridge plug retention, incomplete unsealing, or misjudgment as normal can easily occur, increasing on-site risks and the difficulty of intervention. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for early warning of soluble bridge plug failures. This solves the problem that existing technologies are unable to perform multi-dimensional behavioral perception based on continuous time-series data during the dissolution process of soluble bridge plugs, resulting in delayed identification of bridge plug failure trends, delayed early warning mechanisms, and delayed intervention opportunities.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for early warning of soluble bridge plug failure, comprising the following steps: acquiring time-series data of the dissolution state and the dissolution environment of the soluble bridge plug based on a set time period; performing dynamic evolution analysis on the time-series data of the dissolution state and the dissolution environment of the soluble bridge plug based on a pre-trained time-series feature perception model, and extracting state evolution feature sets and environment evolution feature sets respectively; generating state evolution feature indices and environment evolution feature indices of the soluble bridge plug based on the state evolution feature sets and environment evolution feature sets; determining whether there is a failure risk in the soluble bridge plug based on the state evolution feature indices and environment evolution feature indices, and sending a failure alarm when there is a failure risk.

[0007] Furthermore, the time-series feature perception model includes a dissolution state evolution perception model and a dissolution environment evolution perception model. The dissolution state evolution perception model includes a state data processing layer, a behavioral feature extraction layer, a local change recognition layer, a feature result conversion layer, and a state feature output layer. The dissolution environment evolution perception model includes an environmental data processing layer, a trend feature extraction layer, an interval fluctuation recognition layer, a trend feature conversion layer, and an environmental feature output layer.

[0008] Furthermore, the time-series data of the dissolution state includes surface resistance values, dissolution rate values, and acoustic emission amplitudes at several time points. The state evolution feature set includes the frequency value of resistance drop behavior, the maximum resistance drop amplitude, the maximum dissolution rate value, the dissolution rate fluctuation amplitude value, and the proportion of high-energy acoustic emission points.

[0009] Further, the specific steps for extracting the state evolution feature set are as follows: In the state data processing layer of the dissolution state evolution perception model, the dissolution state time series data of the soluble bridge plug is standardized to obtain a standardized dissolution state time series input sequence; in the behavior feature extraction layer of the dissolution state evolution perception model, the standardized dissolution state time series input sequence is parsed using a sliding window structure and segmented into behavior patterns to extract local behavior segments such as sudden drop segments, rapid rise segments, and high-energy excitation segments, resulting in a set of behavior segments; in the local change identification layer of the dissolution state evolution perception model, the set of behavior segments is subjected to segment resistance reduction calculation, dissolution rate extreme value search, and high-amplitude acoustic emission segment identification to obtain the original state change index set; in the feature result conversion layer of the dissolution state evolution perception model, the original state change index set is subjected to unit time frequency statistics, maximum amplitude extraction, fluctuation amplitude calculation, and proportional mapping to generate structured state evolution features; in the state feature output layer of the dissolution state evolution perception model, the structured state evolution features are arranged in an orderly manner and encapsulated into fields to finally output the state evolution feature set.

[0010] Further, the specific steps for generating the state evolution characteristic index of the soluble bridge plug are as follows: read the frequency value of the resistance drop behavior, the maximum resistance drop amplitude, the maximum dissolution rate value, the dissolution rate fluctuation amplitude value, and the proportion of high-energy acoustic emission points of the soluble bridge plug; perform weighted analysis on the frequency value of the resistance drop behavior, the maximum resistance drop amplitude, the maximum dissolution rate value, the dissolution rate fluctuation amplitude value, and the proportion of high-energy acoustic emission points of the soluble bridge plug to obtain the state evolution characteristic index of the soluble bridge plug.

[0011] Furthermore, the time series data of the dissolution environment includes local well temperature values, local well pressure values, viscosity values ​​of dissolution products, and fluid disturbance intensity at several time points. The environmental evolution feature set includes the lowest well temperature value, well pressure fluctuation amplitude value, average viscosity rise slope, and disturbance fluctuation ratio.

[0012] Further, the specific steps for extracting the environmental evolution feature set are as follows: In the environmental data processing layer of the dissolution environment trend perception model, the dissolution environment time series data of soluble bridge plugs are standardized to obtain a standardized dissolution environment time series input sequence; in the trend feature extraction layer, the standardized dissolution environment time series input sequence is fitted with trend curves and analyzed in multiple segments to extract the temperature dip segment, pressure disturbance peak segment, viscosity rise segment, and disturbance flicker segment, thus obtaining a set of trend structure segments; in the interval fluctuation identification layer, the set of trend structure segments is processed by extracting the lowest temperature point, calculating the pressure fluctuation amplitude, regressing the viscosity rise rate, and statistically analyzing the proportion of disturbance fluctuations, thus obtaining a set of original trend change indicators; in the trend feature transformation layer, the original trend change indicator set is normalized and fused to construct a standardized environmental trend feature vector; in the environmental feature output layer, the standardized environmental trend feature vector is formatted and output to finally generate the environmental evolution feature set.

[0013] Further, the specific steps for generating the environmental evolution characteristic index of soluble bridge plugs are as follows: read the lowest well temperature value, well pressure fluctuation amplitude value, average viscosity rise slope, and disturbance fluctuation ratio of the soluble bridge plug; perform weighted analysis on the lowest well temperature value, well pressure fluctuation amplitude value, average viscosity rise slope, and disturbance fluctuation ratio of the soluble bridge plug to obtain the environmental evolution characteristic index of the soluble bridge plug.

[0014] Furthermore, the specific steps for determining whether a soluble bridge plug has a failure risk are as follows: The state evolution characteristic index and environmental evolution characteristic index of the soluble bridge plug are compared with a preset failure interval set for analysis, and the failure interval set includes both state evolution characteristic failure intervals and environmental evolution characteristic failure intervals; when the state evolution characteristic index and environmental evolution characteristic index are within the preset failure interval set, the soluble bridge plug is considered to have a failure risk.

[0015] A soluble bridge plug fault early warning system includes: a dissolution data acquisition unit, used to acquire time-series data of the dissolution state and dissolution environment of the soluble bridge plug based on a set time period; a data evolution analysis unit, used to perform dynamic evolution analysis on the time-series data of the dissolution state and dissolution environment of the soluble bridge plug based on a pre-trained time-series feature perception model, and extract state evolution feature sets and environmental evolution feature sets respectively; an evolution feature analysis unit, used to generate state evolution feature indices and environmental evolution feature indices of the soluble bridge plug based on the state evolution feature sets and environmental evolution feature sets; and a fault judgment and early warning unit, used to judge whether there is a fault risk in the soluble bridge plug based on the state evolution feature indices and environmental evolution feature indices, and send a fault alarm when there is a fault risk.

[0016] The present invention has the following beneficial effects:

[0017] (1) The soluble bridge plug failure early warning method constructs a complete process of continuous monitoring, feature extraction and dynamic analysis based on the state data and environmental data of the bridge plug during the dissolution process. By acquiring data such as resistance, dissolution rate and acoustic emission at several time points, combined with environmental parameters such as well temperature, well pressure, liquid viscosity and disturbance intensity, it can promptly detect whether the dissolution behavior deviates from the normal trajectory. When there are characteristics such as sudden drop in resistance frequency, violent fluctuation in rate or concentrated acoustic emission energy, potential dissolution anomalies can be identified by the model. Compared with the traditional method of relying on unsealing failure to make a diagnosis, this method can issue an early warning earlier and avoid operation delays or even safety risks caused by soluble bridge plug retention, partial dissolution or unsealing failure. It has significant application effectiveness and reliability guarantee in high-density downhole operation scenarios.

[0018] (2) The soluble bridge plug fault early warning method divides the bridge plug's own state and surrounding environment data into domains, extracts key evolution features, and performs dynamic analysis on the dissolution state time series data and environmental time series data through the model. It extracts feature indicators that can represent the current operating trend, such as the frequency of resistance drop behavior, the amplitude of dissolution rate fluctuation, and the amplitude of well pressure disturbance. Based on these features, a weighted analysis method is used to construct the state evolution feature index and the environmental evolution feature index, thereby more clearly distinguishing the main body fault and the environmental impact. Compared with the single-dimensional judgment method, this method not only improves the accuracy of fault identification, but also enhances the adaptability to different well conditions.

[0019] (3) The soluble bridge plug fault early warning system adopts a modular design, which independently divides the functional units such as dissolution data acquisition, data evolution analysis, evolution feature analysis and fault judgment and early warning. This allows each unit to work independently while also working together to ensure the efficient operation of the system. Each functional unit can be customized and adjusted according to the actual needs of the site, and has strong flexibility and scalability. For example, the dissolution data acquisition unit supports different types of dissolution state and environmental data input and can adapt to the access of various sensors and detection devices. The data evolution analysis unit and the evolution feature analysis unit can continuously improve the early warning effect through model optimization and training, and adapt to complex well site environmental changes. The fault judgment and early warning unit can judge whether there is a fault risk in the first time by comprehensively analyzing the two major indices and issue an alarm in a timely manner, providing operators with clear and reliable decision-making basis. The modular architecture of the system makes it easy to deploy in different well sites and different application scenarios, and provides rich operating space and efficient support platform for possible future technology upgrades or functional expansion, thereby enhancing the sustainability and market competitiveness of the system.

[0020] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0021] Figure 1 This is a flowchart of a soluble bridge plug fault early warning method according to the present invention.

[0022] Figure 2 This is a flowchart illustrating the specific steps involved in extracting environmental evolution feature sets in a soluble bridge plug fault early warning method according to the present invention.

[0023] Figure 3 This is a block diagram of a soluble bridge plug fault early warning system according to the present invention. Detailed Implementation

[0024] Please see Figure 1 This invention provides a technical solution: a method for early warning of soluble bridge plug failures, comprising the following steps: acquiring time-series data of the dissolution state and dissolution environment of the soluble bridge plug based on a set time period; performing dynamic evolution analysis on the time-series data of the dissolution state and dissolution environment of the soluble bridge plug based on a pre-trained time-series feature perception model, and extracting state evolution feature sets and environment evolution feature sets respectively; generating state evolution feature indices and environment evolution feature indices of the soluble bridge plug based on the state evolution feature sets and environment evolution feature sets; determining whether there is a failure risk in the soluble bridge plug based on the state evolution feature indices and environment evolution feature indices, and sending a failure alarm to relevant personnel when there is a failure risk.

[0025] The time-series feature perception model includes a dissolution state evolution perception model and a dissolution environment evolution perception model. The dissolution state evolution perception model includes a state data processing layer, a behavioral feature extraction layer, a local change recognition layer, a feature result conversion layer, and a state feature output layer. The dissolution environment evolution perception model includes an environmental data processing layer, a trend feature extraction layer, an interval fluctuation recognition layer, a trend feature conversion layer, and an environmental feature output layer.

[0026] The pre-training steps for the dissolution state evolution perception model are as follows:

[0027] At the start of pre-training, a large amount of time-series data on the dissolution state is collected, including time-series data on physical quantities such as surface resistance, dissolution rate, and acoustic emission amplitude. This data should be timestamped and consistent within the same time window. Next, discrete smoothing and noise removal techniques are used to process the raw data. Specifically, a three-point moving average method is used to smooth each data sequence, effectively removing high-frequency noise. Then, interpolation techniques are used to fill in any missing data to ensure data continuity. Finally, to eliminate the dimensional differences between different physical quantities, z-score standardization is applied to each class of data, making its mean 0 and standard deviation 1, thus ensuring that all time-series data are within the same scale range, facilitating subsequent model training.

[0028] After data preprocessing, the next step is feature extraction. In this stage, the time-series data is first divided into different time periods using the sliding window method, and each segment is analyzed in detail. Specifically, we need to identify and extract behavioral patterns such as sudden drops, rapid rises, and high-energy excitation segments in the dissolution state. These behavioral segments correspond to different dissolution processes, such as sudden drops in resistance or rapid changes in dissolution rate, reflecting key dynamic characteristics of the dissolution process. Next, by calculating local features such as the magnitude of resistance drops, changes in dissolution rate, and fluctuations in acoustic emission energy, these behavioral segments are transformed into specific physical quantities, providing important input data for subsequent model training.

[0029] After feature extraction, the next step is to input the processed data into a pre-trained neural network model for training. Temporal processing models such as LSTM (Long Short-Term Memory) or GRU (Gated Recurrent Unit) are employed to capture long-term dependencies and change patterns in the time-series data. The training performance of the model is evaluated by setting an appropriate loss function (such as mean squared error (MSE)) to optimize the weight parameters in the network. During training, a large amount of labeled time-series data is required. The model aims to automatically extract the corresponding state evolution feature set (such as the frequency of resistance drops and the amplitude of dissolution rate fluctuations) from the input dissolution state data. The model is evaluated using cross-validation (K-fold validation), and adjustments and optimizations are made based on the validation results to ensure that the trained model performs stably on different datasets and has good generalization ability.

[0030] The pre-training steps for the dissolution environment evolution perception model are as follows:

[0031] First, sufficient environmental time-series data is collected, including environmental parameters such as local well temperature, local well pressure, viscosity of dissolved products, and fluid disturbance intensity. This data must be temporally consistent and continuous, and undergo rigorous data preprocessing. First, smoothing algorithms (such as Gaussian filtering or low-pass filtering) are used to remove noise from the raw data. Then, interpolation techniques are used to fill in any missing values ​​to avoid affecting subsequent analysis. Finally, to ensure that all environmental features are trained on the same scale, z-score normalization is applied to all environmental data, ensuring that all input data can be learned within the same data distribution range.

[0032] After data preprocessing, the feature extraction stage begins. The main task of this stage is to extract trend structures and fluctuation characteristics from the time-series data of the dissolution environment. First, the environmental data is modeled using trend fitting (such as multinomial regression or local regression) to extract trend change segments such as temperature decline segments, pressure disturbance peak segments, and viscosity increase segments. Next, fluctuation analysis is used to calculate the fluctuation amplitude, rate of change, and percentage of fluctuation for each segment. These fluctuation indicators effectively reflect the impact of various disturbance factors in the dissolution environment on the dissolution rate, serving as important features for model learning. Particularly in the pressure disturbance peak segments and viscosity change segments, the extracted fluctuation information provides crucial information for subsequent fault early warning.

[0033] After feature extraction, the processed environmental data is input into advanced temporal learning models such as Temporal Convolutional Neural Networks (TCNs) or Transformers for training. The training objective of the network is to automatically learn environmental evolution features related to the dissolution process (such as minimum well temperature and well pressure fluctuations) from the input environmental time-series data. By selecting an appropriate loss function (such as mean squared error (MSE)) for model optimization, the network can accurately capture long-term trends and local fluctuations in the environmental data. During training, a large amount of labeled time-series data is used, and cross-validation is combined to ensure the model's generalization ability. The model hyperparameters (such as learning rate, batch size, and number of network layers) are adjusted based on the evaluation results, ultimately obtaining a trained environmental evolution awareness model with high predictive accuracy.

[0034] Specifically, the time series data of the dissolution state includes surface resistance values, dissolution rate values, and acoustic emission amplitudes at several time points. The state evolution feature set includes the frequency value of resistance drop behavior, the maximum resistance drop amplitude, the maximum dissolution rate value, the dissolution rate fluctuation amplitude value, and the proportion of high-energy acoustic emission points.

[0035] Among them, the surface resistance value refers to the change in the conductivity of the surface layer of the bridge plug material during the dissolution process, which can be obtained by measuring the micro resistance measurement probe set on the surface of the bridge plug.

[0036] The dissolution rate value refers to the rate of change of mass or volume of the bridge plug per unit time, which is used to characterize the overall dissolution progress. It can be calculated by combining real-time mass estimation and volume change analysis. The required raw data can be obtained by the downhole ultrasonic ranging device and differential pressure sensor.

[0037] Acoustic emission amplitude refers to the intensity of the acoustic signal generated during the dissolution of the bridge plug due to material fracture, microcrack propagation, or stress concentration release. It can be measured by a piezoelectric acoustic emission sensor placed near the well section.

[0038] The specific steps for extracting the state evolution feature set are as follows:

[0039] In the state data processing layer of the dissolution state evolution sensing model, the dissolution state time series data of soluble bridge plugs are standardized (missing point completion, outlier removal and normalization) to obtain a standardized dissolution state time series input sequence. Specifically, the three types of raw data, namely bridge plug surface resistance value, dissolution rate value and acoustic emission amplitude, are preprocessed respectively. First, the linear interpolation method is used to continuously complete the time discontinuity data in the resistance and rate curves. Then, the 3σ statistical criterion is used to remove abrupt outliers in the acoustic emission data. Finally, each sequence is uniformly mapped to the [0,1] interval to form a normalized value format, thereby constructing a standardized dissolution state time series input sequence.

[0040] In the behavioral feature extraction layer of the dissolution state evolution perception model, the standardized dissolution state time-series input sequence is parsed using a sliding window structure and segmented into behavioral patterns. Local behavioral segments such as sudden drop segments, rapid rise segments, and high-energy excitation segments are extracted to obtain a set of behavioral segments. Specifically, a sliding window of fixed length 5 seconds is used to sequentially traverse the state input sequence. The resistance change curve is analyzed within each window. When the resistance drop exceeds a set threshold (e.g., the normalized value drop > 0.3), it is marked as a resistance sudden drop segment. When the derivative of the dissolution rate remains positive within the window and the average rise rate is greater than twice the global mean, it is marked as a rate rapid rise segment. If the acoustic emission amplitude shows three consecutive amplitude points greater than 0.8 within any window, the segment is identified as a high-energy excitation segment. These three types of segments constitute the set of local behavioral segments.

[0041] In the local change recognition layer of the dissolution state evolution perception model, the set of behavioral segments is subjected to segment resistance reduction calculation, dissolution rate extreme value search, and high-amplitude acoustic emission segment identification to obtain the original state change index set. Specifically, the maximum initial and final resistance difference is selected as the "maximum resistance drop amplitude" in all sudden drop segments, and the maximum single-point rate value is extracted as the "maximum dissolution rate value" in the rate rise segment. Then, the number of times the acoustic emission amplitude is greater than 0.8 in all time points is calculated, and the ratio of this number to the total number of sampling points is used as the "proportion of high-energy acoustic emission points". In addition, the occurrence frequency of all sudden drop segments is counted to provide a basis for the derivation of subsequent frequency indicators, and finally, the original state change index set is formed.

[0042] In the feature result transformation layer of the dissolution state evolution perception model, the original set of state change indicators is statistically analyzed per unit time, the maximum amplitude is extracted, the fluctuation amplitude is calculated and the proportion is mapped to generate structured state evolution features. Specifically, the number of occurrences of sudden drop behavior is divided by the total data acquisition time to obtain the "resistance sudden drop behavior frequency value", which measures the stability of the dissolution process by the number of behaviors per unit time; the difference between the maximum and minimum values ​​in the dissolution rate sequence is defined as the "dissolution rate fluctuation amplitude value", which reflects the non-uniformity of rate change; other indicators such as the maximum sudden drop amplitude and the proportion of high-energy acoustic emission points are all constructed into vector fields after retaining three significant digits, forming a set of state evolution features with a regular structure and consistent units.

[0043] In the state feature output layer of the dissolution state evolution perception model, the structured state evolution features are arranged in an orderly manner and encapsulated in fields, and finally the state evolution feature set is output. Specifically, the "resistance drop behavior frequency value, maximum resistance drop amplitude, maximum dissolution rate value, dissolution rate fluctuation amplitude value" and "proportion of high-energy acoustic emission points" are arranged in a predetermined field order, and each item is encapsulated in the feature output structure in the form of floating point numbers, which is used as the formal output of the state evolution feature set.

[0044] In this implementation scheme, standardization eliminates data discrepancies and ensures consistency of time-series data across different dissolution states. This process includes not only data completion and outlier removal but also normalization to ensure data uniformity under the same dimension, effectively improving the reliability of subsequent analysis. In the behavioral feature extraction stage, a sliding window structure is used to accurately analyze resistance changes, dissolution rates, and acoustic emission amplitudes, enabling timely capture of key sudden drops, rapid rises, and high-energy excitation behaviors during the dissolution process. By extracting these behavioral fragments, the model can identify potential fault signals and provide early warnings of anomalies during the dissolution process. Furthermore, in the feature conversion and output stage, the analysis and structured processing of key indicators such as the frequency of sudden drops and the amplitude of dissolution rate fluctuations provide a practically meaningful set of state evolution features. These features effectively help the system determine whether the bridge plug's dissolution state has entered the fault risk range, providing a scientific basis for subsequent fault warnings and emergency response. This method improves the accuracy and response speed of the early warning system, enabling timely and accurate prediction of potential faults in soluble bridge plugs, ensuring the safety and efficiency of downhole operations.

[0045] Specifically, the steps for generating the state evolution characteristic index of the soluble bridge plug are as follows: read the frequency value of the resistance drop behavior, the maximum resistance drop amplitude, the maximum dissolution rate value, the dissolution rate fluctuation amplitude value, and the proportion of high-energy acoustic emission points of the soluble bridge plug; perform weighted analysis on the frequency value of the resistance drop behavior, the maximum resistance drop amplitude, the maximum dissolution rate value, the dissolution rate fluctuation amplitude value, and the proportion of high-energy acoustic emission points of the soluble bridge plug to obtain the state evolution characteristic index of the soluble bridge plug.

[0046] In this implementation plan, by weighting key characteristic values ​​such as the frequency of resistance drop behavior, the maximum resistance drop amplitude, the maximum dissolution rate, the dissolution rate fluctuation amplitude, and the proportion of high-energy acoustic emission points, various dynamic behaviors during the bridge plug dissolution process can be comprehensively reflected, thereby accurately assessing its state. Integrating these characteristics through weighting not only enhances the representativeness of the characteristics but also improves the sensitivity and accuracy of fault early warning. Ultimately, the obtained state evolution characteristic index provides a reliable basis for subsequent fault risk assessment, helping operators to identify potential fault risks in a timely manner and ensuring the safety and efficiency of downhole operations.

[0047] Specifically, the time series data of the dissolution environment includes local well temperature, local well pressure, viscosity of dissolution products, and fluid disturbance intensity at several time points. The environmental evolution feature set includes the lowest well temperature, well pressure fluctuation amplitude, average viscosity rise slope, and disturbance fluctuation percentage.

[0048] The local well temperature value refers to the liquid temperature in the well section where the soluble bridge plug is located. It is used to analyze the influence of the current thermal environment on the dissolution reaction. It can be measured by thermocouple temperature sensors installed near the bridge plug.

[0049] Local well pressure refers to the static or dynamic pressure within the well section where the bridge plug is actually in a dissolved state. It is used to determine the effect of environmental pressure on the bridge plug structure and can be measured by a piezoelectric pressure sensor installed downhole.

[0050] The viscosity value of the dissolved product refers to the viscosity of the released material in the well fluid after the bridge plug dissolves. It is used to assess the product flowability and the risk of dissolution residue. It can be measured by an online rotary viscometer installed in the flow path.

[0051] The fluid disturbance intensity value refers to the degree of local disturbance of the liquid around the bridge plug, including shear fluctuations, micro-vortices, or velocity variations. It can be obtained by combining multi-point micro velocity probes and shear stress sensors deployed in the well section.

[0052] like Figure 2 As shown, the specific steps for extracting the environmental evolution feature set are as follows:

[0053] In the environmental data processing layer of the dissolution environment trend perception model, the dissolution environment time series data of soluble bridge plugs are standardized to obtain a standardized dissolution environment time series input sequence. Specifically, the original well temperature, well pressure, viscosity and disturbance intensity are preprocessed sequentially. First, the three-point moving average method is used to smooth each time series to suppress high-frequency noise. Then, the upper and lower quantile determination rules are used to correct the maximum and minimum values ​​to avoid trend deviation caused by short-term interference. Finally, each physical quantity is z-score standardized to make its mean 0 and variance 1, thus obtaining a standardized environmental input sequence with a uniform format and clear data form.

[0054] In the trend feature extraction layer, trend curve fitting and multi-segment structure analysis are performed on the normalized dissolution environment time series input sequence to extract the temperature dip segment, pressure disturbance peak segment, viscosity rise segment, and disturbance stroboscopic segment, resulting in a set of trend structure segments. Specifically, local linear fitting and sliding window derivative analysis methods are used on the well temperature, well pressure, viscosity, and disturbance sequences respectively. In the well temperature sequence, regions with negative slopes and continuous drops exceeding three points are identified as temperature dip segments. In the well pressure sequence, segments where the amplitude peak change point differs from the mean by more than two standard deviations are identified as pressure disturbance peak segments. First-derivative regression analysis is performed on the viscosity data, and segments with a slope continuously greater than zero and an uninterrupted rise exceeding 10 seconds are defined as viscosity rise segments. For the fluid disturbance sequence, segments where the number of disturbance peaks per unit time exceeds a set threshold (e.g., 3 times / 10 seconds) are identified as disturbance stroboscopic segments. All the above-mentioned marked regions together constitute the set of trend structure segments.

[0055] In the interval fluctuation identification layer, the set of trend structure segments is analyzed by extracting the lowest temperature point, calculating the pressure fluctuation amplitude, regressing the viscosity rise rate, and statistically analyzing the proportion of disturbance fluctuations to obtain the original set of trend change indicators. Specifically, the minimum point of the temperature curve is extracted from all temperature-progressing segments and recorded as the "lowest well temperature value"; the amplitude (difference between the maximum and minimum values) of each segment in the pressure disturbance peak segment is statistically analyzed, and the maximum amplitude is selected as the "well pressure fluctuation amplitude value"; linear regression curves are fitted to all viscosity rise segments, and their average slope is calculated as the "average viscosity rise slope"; the proportion of the time length of high-frequency disturbance behavior in the disturbance flicker segment is statistically analyzed to the total time and recorded as the "disturbance fluctuation proportion";

[0056] In the trend feature transformation layer, the original trend change index set is normalized and feature fused to construct a standardized environmental trend feature vector. Specifically, the above four original trend indices are linearly normalized to map their values ​​to the [0,1] interval, ensuring that each index is comparable at the same scale. The normalization results are then encapsulated into a unified format environmental trend feature vector, which includes the minimum temperature point corresponding value, the normalized value of the maximum pressure amplitude, the standardized value of the average viscosity rise slope, and the ratio of the proportion of disturbance fluctuations.

[0057] In the environmental feature output layer, the standardized environmental trend feature vector is formatted and output as fields, ultimately generating an environmental evolution feature set. Specifically, the four standardized environmental trend features are encapsulated into field structures in a preset order and named "Minimum Well Temperature Value", "Well Pressure Fluctuation Amplitude Value", "Average Viscosity Rise Slope", and "Percentage of Disturbance Fluctuation". All fields are presented in floating-point format and stored in the structured dataset as the formal output of the environmental evolution feature set.

[0058] In this implementation scheme, by standardizing data such as local well temperature, well pressure, viscosity of dissolved products, and fluid disturbance intensity, noise and outliers in the original data are eliminated, ensuring data quality and consistency. Based on this, a trend feature extraction layer performs trend analysis on various environmental data, identifying typical patterns of key environmental changes such as temperature, pressure, viscosity, and disturbance, providing valuable structured information for in-depth analysis. Furthermore, an interval fluctuation identification layer further refines the capture of dynamic environmental changes by extracting key fluctuation indicators such as the fluctuation amplitude of minimum well temperature, maximum well pressure, and viscosity rise rate. These features are normalized by a feature transformation layer, ensuring a uniform scale for all indicators, facilitating subsequent weighted analysis and fusion. Finally, the generated standardized environmental evolution feature set accurately reflects the changing trends of the downhole environment, providing precise data support for fault risk assessment. This process not only improves the comparability of environmental data but also effectively enhances the accuracy and reliability of soluble bridge plug fault prediction, thereby ensuring the safety and efficiency of downhole operations.

[0059] Specifically, the steps for generating the environmental evolution characteristic index of soluble bridge plugs are as follows: read the lowest well temperature value, well pressure fluctuation amplitude value, average viscosity rise slope, and disturbance fluctuation ratio of the soluble bridge plug; perform weighted analysis on the lowest well temperature value, well pressure fluctuation amplitude value, average viscosity rise slope, and disturbance fluctuation ratio of the soluble bridge plug to obtain the environmental evolution characteristic index of the soluble bridge plug.

[0060] In this implementation plan, by reading key environmental parameters such as minimum well temperature, well pressure fluctuation amplitude, viscosity rise slope, and disturbance fluctuation ratio, and performing weighted analysis, the impact of environmental changes on the dissolution process of soluble bridge plugs can be more comprehensively reflected. The weighted analysis of each indicator takes into account its importance in the overall dissolution process, ensuring that the final environmental evolution characteristic index can accurately capture the comprehensive effect of environmental changes. Such a characteristic index provides a more accurate and reliable basis for subsequent fault judgment and early warning, helps to improve the monitoring accuracy and response speed of the dissolution process, and ensures the safety and efficiency of downhole operations.

[0061] Specifically, the steps for determining whether a soluble bridge plug has a failure risk are as follows: The state evolution characteristic index and environmental evolution characteristic index of the soluble bridge plug are compared with a preset failure interval set for analysis, and the failure interval set includes a state evolution characteristic failure interval and an environmental evolution characteristic failure interval; when the state evolution characteristic index and environmental evolution characteristic index are within the preset failure interval set, the soluble bridge plug is considered to have a failure risk.

[0062] In this implementation scheme, by comparing the state evolution characteristic index and the environmental evolution characteristic index with the preset fault range, the system can accurately predict the fault risk of soluble bridge plugs. By setting a clear fault range, the system can accurately identify key risk points in the dissolution process. When the state and environmental characteristic values ​​are within the fault range, the system will issue an alarm in a timely manner. This data-driven judgment method not only improves the accuracy of fault risk identification, but also responds in real time to various environmental changes and operating conditions, ensuring safety in the dissolution process. Compared with the limitations of traditional methods that rely on a single parameter or experience, this multi-dimensional and multi-feature analysis method is more comprehensive, can provide early warning of potential faults, avoid equipment damage and operation delays, and improve the stability and efficiency of operations.

[0063] Please see Figure 3 This invention provides a technical solution: a soluble bridge plug fault early warning system, comprising: a dissolution data acquisition unit, used to acquire time-series data of the dissolution state and dissolution environment of the soluble bridge plug based on a set time period; a data evolution analysis unit, used to perform dynamic evolution analysis on the time-series data of the dissolution state and dissolution environment of the soluble bridge plug based on a pre-trained time-series feature perception model, and extract state evolution feature sets and environmental evolution feature sets respectively; an evolution feature analysis unit, used to generate a state evolution feature index and an environmental evolution feature index of the soluble bridge plug based on the state evolution feature sets and the environmental evolution feature sets; and a fault judgment and early warning unit, used to judge whether there is a fault risk in the soluble bridge plug based on the state evolution feature index and the environmental evolution feature index, and send a fault alarm to relevant personnel when there is a fault risk.

[0064] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0065] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for failure warning of dissolvable bridge plug, characterized in that, The method comprises the following steps: obtaining the dissolving state time series data and the dissolving environment time series data of the dissolvable bridge plug based on a set time period; based on the pre-trained time series feature perception model, the dissolving state time series data and the dissolving environment time series data of the dissolvable bridge plug are dynamically evolved and analyzed to extract the state evolution feature set and the environment evolution feature set respectively; based on the state evolution feature set and the environment evolution feature set, the state evolution feature index and the environment evolution feature index of the dissolvable bridge plug are generated; based on the state evolution feature index and the environment evolution feature index, it is judged whether the dissolvable bridge plug has a fault risk, and a fault alarm is sent when there is a fault risk.

2. The method of claim 1, wherein, The time series feature perception model comprises a dissolving state evolution perception model and a dissolving environment evolution perception model, the dissolving state evolution perception model comprises a state data processing layer, a behavior feature extraction layer, a local change identification layer, a feature result conversion layer and a state feature output layer, and the dissolving environment evolution perception model comprises an environment data processing layer, a trend feature extraction layer, an interval fluctuation identification layer, a trend feature conversion layer and an environment feature output layer.

3. The method of claim 2, wherein, The dissolving state time series data comprises surface resistance values, dissolving rate values and acoustic emission amplitude values at a plurality of time points, and the state evolution feature set comprises resistance drop behavior frequency values, maximum resistance drop amplitudes, maximum dissolving rate values, dissolving rate fluctuation amplitude values and high-energy acoustic emission point proportions.

4. The method of claim 3, wherein, The specific steps for extracting the state evolution feature set are as follows: in the state data processing layer of the dissolving state evolution perception model, the dissolving state time series data of the dissolvable bridge plug is standardized to obtain a standardized dissolving state time series input sequence; in the behavior feature extraction layer of the dissolving state evolution perception model, the standardized dissolving state time series input sequence is analyzed by sliding window structure and behavior mode segmentation to extract local behavior segments of drop segments, rapid rise segments and high-energy excitation segments, and a behavior segment set is obtained; in the local change identification layer of the dissolving state evolution perception model, the behavior segment set is subjected to segment resistance drop amplitude calculation, dissolving rate extreme value search and high-amplitude acoustic emission segment identification to obtain an original state change index set; in the feature result conversion layer of the dissolving state evolution perception model, the original state change index set is subjected to unit time frequency statistics, maximum amplitude extraction, fluctuation amplitude calculation and proportional mapping to generate a structured state evolution feature; in the state feature output layer of the dissolving state evolution perception model, the structured state evolution feature is sequentially arranged and field encapsulated, and finally the state evolution feature set is output.

5. The method of claim 3, wherein, The specific steps for generating the state evolution feature index of the dissolvable bridge plug are as follows: read the resistance drop behavior frequency value, the maximum resistance drop amplitude, the maximum dissolving rate value, the dissolving rate fluctuation amplitude value and the high-energy acoustic emission point proportion of the dissolvable bridge plug; the resistance drop behavior frequency value, the maximum resistance drop amplitude, the maximum dissolving rate value, the dissolving rate fluctuation amplitude value and the high-energy acoustic emission point proportion of the dissolvable bridge plug are subjected to weighted analysis to obtain the state evolution feature index of the dissolvable bridge plug.

6. The method of claim 2, wherein, The time series data of the dissolution environment includes local well temperature, local well pressure, viscosity of dissolution products, and fluid disturbance intensity at several time points. The environmental evolution feature set includes the lowest well temperature, well pressure fluctuation amplitude, average viscosity rise slope, and disturbance fluctuation percentage.

7. The method of claim 6, wherein, The specific steps for extracting environmental evolution feature sets are as follows: In the environmental data processing layer of the dissolution environment trend perception model, the dissolution environment time series data of soluble bridge plugs are standardized to obtain a standardized dissolution environment time series input sequence. In the trend feature extraction layer, trend curve fitting and multi-segment structure analysis are performed on the normalized dissolution environment time series input sequence to extract the temperature probing segment, pressure disturbance peak segment, viscosity rise segment and disturbance stroboscopic segment, and obtain a set of trend structure segments; In the interval fluctuation identification layer, the set of trend structure segments is subjected to the extraction of the lowest temperature point, the calculation of pressure fluctuation amplitude, the regression of viscosity rise rate and the statistics of the proportion of disturbance fluctuation, to obtain the original set of trend change indicators. In the trend feature transformation layer, the original set of trend change indicators is normalized and feature fused to construct a standardized environmental trend feature vector; In the environmental feature output layer, the standardized environmental trend feature vectors are formatted and output as fields, ultimately generating an environmental evolution feature set.

8. The method of claim 6, wherein, The specific steps for generating the environmental evolution characteristic index of soluble bridge plugs are as follows: Read the lowest well temperature value, well pressure fluctuation amplitude value, average viscosity rise slope and disturbance fluctuation ratio of soluble bridge plug; The environmental evolution characteristic index of soluble bridge plugs was obtained by weighting the minimum well temperature, well pressure fluctuation amplitude, average viscosity rise slope, and disturbance fluctuation ratio.

9. The method of claim 1, wherein, The specific steps for determining whether a soluble bridge plug has a risk of failure are as follows: The state evolution characteristic index and environmental evolution characteristic index of the soluble bridge plug are respectively compared with the preset fault interval set for judgment and analysis, and the fault interval set includes the state evolution characteristic fault interval and the environmental evolution characteristic fault interval. When the state evolution characteristic index and the environmental evolution characteristic index are within the preset fault interval set, the soluble bridge plug is considered to have a fault risk.

10. A soluble bridge plug failure warning system applying the soluble bridge plug failure warning method of any one of claims 1-9, characterized in that, include: The dissolution data acquisition unit is used to acquire time-series data of the dissolution state and dissolution environment of the soluble bridge plug based on a set time period. The data evolution analysis unit is used to perform dynamic evolution analysis on the dissolution state time-series data and dissolution environment time-series data of soluble bridge plugs based on a pre-trained time-series feature perception model, and extract the state evolution feature set and environment evolution feature set respectively. The evolutionary feature analysis unit is used to generate the state evolution feature index and environmental evolution feature index of soluble bridge plugs based on the state evolution feature set and the environmental evolution feature set. The fault judgment and early warning unit is used to determine whether there is a fault risk in the soluble bridge plug based on the state evolution characteristic index and the environmental evolution characteristic index, and to send a fault alarm when there is a fault risk.

Images