CN117275209A

CN117275209A - Monitoring and early warning method based on distributed optical fiber acoustic wave sensing and related device

Info

Publication number: CN117275209A
Application number: CN202311561608.7A
Authority: CN
Inventors: 蔡毅; 游东东; 丁楠; 张军; 向海民; 蔡立辉
Original assignee: Guangdong Lichuang Information Technology Co ltd
Current assignee: Guangdong Lichuang Information Technology Co ltd
Priority date: 2023-11-22
Filing date: 2023-11-22
Publication date: 2023-12-22
Anticipated expiration: 2043-11-22
Also published as: CN117275209B

Abstract

The application is applicable to the technical field of data processing, and provides a monitoring and early warning method based on distributed optical fiber acoustic wave sensing and a related device, wherein the method comprises the following steps: acquiring sound wave transmission data in an oil and gas pipeline; carrying out neighborhood construction and correlation extraction from sound wave transmission data through a multi-source data characteristic analysis model to obtain acoustic transmission scene characteristics corresponding to each oil and gas pipeline segment; predicting the acoustic transmission scene characteristics through a dynamic self-adaptive positioning model to obtain detection event predicted values corresponding to each oil and gas pipeline segment in the oil and gas pipeline; and for the target oil gas pipeline segment with the predicted value of the detection event meeting the early warning condition, displaying dynamic early warning information corresponding to the target oil gas pipeline segment in a dynamic early warning monitoring distribution diagram according to the geographic position of the target oil gas pipeline segment. Therefore, the monitoring accuracy of the oil gas pipeline can be effectively improved, and the safety of the oil gas pipeline is guaranteed.

Description

Monitoring and early warning method based on distributed optical fiber acoustic wave sensing and related device

Technical Field

The application belongs to the field of data processing, and particularly relates to a monitoring and early warning method based on distributed optical fiber acoustic wave sensing and a related device.

Background

An oil-gas pipeline is a pipeline system for conveying petroleum, natural gas and other energy products. It is usually composed of high strength steel pipes, connectors, valves, pumping stations, etc. and by applying different techniques and equipment, reliable transportation and supply of energy is ensured. The oil and gas pipelines can be divided into long-distance pipelines and short-distance pipelines, and the long-distance pipelines generally cross country or region boundaries and are composed of a few large-scale pipelines; short-circuit pipelines are distributed in urban or industrial areas, which are generally small and serve local economic development and energy consumption needs.

At present, the geographical position distribution of oil gas pipeline is extensive, and the coverage is great, and the inspection degree of difficulty is big. Taking cities as an example, the arrangement of oil and gas pipelines is complex, and generally, the complex underground environments such as buildings, underground pipelines, public facilities and the like are involved, and the traditional manual inspection is difficult to cover the oil and gas pipelines in all areas, so that the monitoring data is difficult to guarantee accuracy, and certain potential safety hazards exist. In summary, a technical solution is needed to overcome the above technical problems.

Disclosure of Invention

The embodiment of the application provides a monitoring and early warning method based on distributed optical fiber acoustic wave sensing and a related device, which can solve the problems that the traditional manual inspection is difficult to cover oil and gas pipelines in all areas, so that the monitoring data is difficult to guarantee accuracy and certain potential safety hazards exist.

In a first aspect, an embodiment of the present application provides a monitoring and early warning method based on distributed optical fiber acoustic wave sensing, including:

acquiring sound wave transmission data in an oil and gas pipeline; the sound wave transmission data comprise multi-source sound wave transmission data in different environments; the data collected by each distributed optical fiber sensor in different environments is used as a data source;

carrying out neighborhood construction and correlation extraction from the acoustic transmission data through a multi-source data characteristic analysis model to obtain acoustic transmission scene characteristics corresponding to each oil and gas pipeline segment;

predicting the acoustic transmission scene characteristics through a dynamic self-adaptive positioning model to obtain detection event predicted values corresponding to each oil and gas pipeline segment in the oil and gas pipeline; the higher the predicted value of the detection event is, the higher the probability of the detection event in the target oil and gas pipeline segment is;

For a target oil gas pipeline segment with a predicted value of a detection event meeting an early warning condition, displaying dynamic early warning information corresponding to the target oil gas pipeline segment in a dynamic early warning monitoring distribution diagram according to the geographic position of the target oil gas pipeline segment; the display form of the dynamic early warning information changes along with the change of the predicted value of the detection event of the target oil and gas pipeline section.

In one possible implementation manner of the first aspect, acquiring acoustic transmission data in an oil and gas pipeline includes:

acquiring optical fiber vibration data acquired by each distributed optical fiber sensor under different environments;

preprocessing the optical fiber vibration data for each distributed optical fiber sensor in different environments;

extracting multi-scale sound wave transmission characteristics from the preprocessed optical fiber vibration data to obtain sound wave time lag relation characteristics, sound wave self-adaptive filtering characteristics and sound rate universality characteristics corresponding to each data source;

and carrying out feature fusion processing on the sound wave time lag relation feature, the sound wave self-adaptive filtering feature and the acoustic rate universality feature corresponding to each data source to obtain sound wave transmission fusion feature data corresponding to each data source.

In a possible implementation manner of the first aspect, the multi-source data feature analysis model includes a fine granularity analysis layer, a neighborhood construction layer, a correlation analysis layer, a dimension reduction processing layer and a feature fusion layer;

carrying out neighborhood construction and correlation extraction from the acoustic transmission data through a multi-source data characteristic analysis model to obtain acoustic transmission scene characteristics corresponding to each oil and gas pipeline segment, wherein the method comprises the following steps:

carrying out multi-source data fine-granularity analysis on the sound wave transmission data through the fine-granularity analysis layer so as to obtain fine-granularity sound wave transmission characteristics; the fine-grained acoustic wave transmission characteristics are distributed in different frequency domain sub-bands;

carrying out neighborhood construction processing on the fine-granularity sound wave transmission characteristics in different frequency domain sub-bands through the neighborhood construction layer so as to obtain sound wave transmission neighborhood characteristics; the sound wave transmission neighborhood characteristics are used for representing the interactive relation between different positions of each oil and gas pipeline section and acoustic transmission transformation factors in different neighborhood;

acquiring correlation measurement information between each characteristic point in the acoustic wave transmission neighborhood characteristic and surrounding neighborhood characteristic points through the correlation analysis layer; the relevance measurement information is used for representing the association relation between acoustic transmission transformation factors corresponding to the oil and gas pipeline segments;

Performing dimension reduction on the sound wave transmission neighborhood characteristics based on the correlation measurement information through the dimension reduction processing layer to obtain sound wave transmission dimension reduction characteristics;

and carrying out feature fusion on the sound wave transmission dimension reduction features through the feature fusion layer so as to obtain the acoustic transmission scene features.

In a possible implementation manner of the first aspect, the dynamic adaptive positioning model includes an adaptive reinforcement learning layer, a dynamic adaptive value learning layer, and an experience playback layer;

predicting the acoustic transmission scene characteristics through a dynamic self-adaptive positioning model to obtain detection event predicted values corresponding to each oil and gas pipeline segment in the oil and gas pipeline, wherein the detection event predicted values comprise:

extracting multi-scale abstract features from the acoustic transmission scene features through the self-adaptive reinforcement learning layer, and performing detection event prediction processing based on the extracted multi-scale abstract features to obtain initial prediction values of detection events corresponding to all the oil and gas pipeline segments;

and calculating the initial predicted value and the historical adaptive value stored in the experience playback layer through the dynamic adaptive value learning layer to obtain a detection event target predicted value corresponding to each oil and gas pipeline segment.

In a possible implementation manner of the first aspect, after the calculating the initial predicted value and the historical adaptive value stored in the experience playback layer to obtain the detected event target predicted value corresponding to each oil and gas pipeline segment, the method further includes:

circularly storing the detection event target predicted value into the experience playback layer;

updating a historical adaptive value stored in the experience playback layer based on the detection event target predicted value by adopting a preset dynamic adaptive value updating rule;

in the dynamic adaptive value updating rule, the historical adaptive value to be updated is obtained through the following formulaThe method comprises the following steps:

；

wherein,is to detect event target predictive value in history +.>The adaptive value +.>Instant rewards obtained later, < >>Is a newly generated detection event target predictive value, < >>Is learning rate (I/O)>Is a discount factor, < >>Is the maximum of all the selectable adaptation values under the newly generated detection event target prediction value.

In a possible implementation manner of the first aspect, the dynamic adaptive positioning model further includes a target network layer;

the target network layer has the same network structure and the same network parameters as the self-adaptive reinforcement learning layer; the network parameter updating frequency of the target network layer is lower than the network parameter updating frequency of the self-adaptive reinforcement learning layer;

After the initial predicted value and the historical adaptive value stored in the experience playback layer are calculated to obtain the detection event target predicted value corresponding to each oil and gas pipeline segment, the method further comprises the following steps:

extracting multi-scale abstract features from the acoustic transmission scene features through the target network layer, and performing detection event prediction processing based on the extracted multi-scale abstract features to obtain detection event reference predicted values corresponding to the oil and gas pipeline segments;

calculating a quantile regression loss value between the reference predicted value and the initial predicted value;

and carrying out gradient descent optimization processing on the self-adaptive reinforcement learning layer based on the fractional regression loss value so as to update the self-adaptive network parameters in the self-adaptive reinforcement learning layer.

In a possible implementation manner of the first aspect, the experience playback layer further includes an experience playback memory bank;

randomly sampling from the experience playback memory library to obtain a history detection event target predicted value with discrete distribution, and extracting corresponding acoustic transmission scene characteristics;

Extracting multi-scale abstract features from the extracted acoustic transmission scene features through the target network layer, and performing detection event prediction processing based on the extracted multi-scale abstract features to obtain detection event sampling predicted values corresponding to each oil and gas pipeline segment;

and carrying out gradient descent optimization processing on the self-adaptive reinforcement learning layer based on the sampled historical detection event target predicted value and the corresponding detection event sampling predicted value so as to update the self-adaptive network parameters in the self-adaptive reinforcement learning layer.

In a possible implementation manner of the first aspect, the dynamic early warning monitoring distribution map is a GIS geographic location grid coordinate map; the GIS geographic position grid coordinate graph is provided with a GIS geographic position mapping table, and the GIS geographic position mapping table stores GIS geographic position grid coordinates corresponding to the oil gas pipeline segments;

for a target oil gas pipeline segment with a predicted value of a detection event meeting an early warning condition, displaying dynamic early warning information corresponding to the target oil gas pipeline segment in a dynamic early warning monitoring distribution diagram according to the geographic position of the target oil gas pipeline segment, wherein the method comprises the following steps:

Judging whether the predicted value of the detection event corresponding to each oil and gas pipeline segment accords with the early warning condition;

for a target oil gas pipeline segment with a predicted value of a detection event meeting an early warning condition, acquiring GIS (geographic information system) geographic position grid coordinates where the target oil gas pipeline segment is positioned from a GIS geographic position mapping table to obtain a GIS geographic position grid coordinate graph;

generating dynamic early warning information matched with the predicted value of the detection event, and displaying the dynamic early warning information in the GIS geographic position grid coordinate graph;

or the dynamic early warning monitoring distribution map is a three-dimensional local scene map of the oil and gas pipeline;

generating dynamic early warning information matched with the predicted value of the detection event, and displaying the dynamic early warning information in the GIS geographic position grid coordinate graph, and then further comprising:

constructing a three-dimensional local scene map of the oil and gas pipeline based on the GIS geographic position grid coordinate graph and a preset GIS geographic two-dimensional panorama; the display main body in the three-dimensional local scene map of the oil and gas pipeline is a target oil and gas pipeline segment indicated by the dynamic early warning information and an early warning event state of the target oil and gas pipeline segment.

In a second aspect, an embodiment of the present application provides a monitoring and early warning device based on distributed optical fiber acoustic wave sensing, including:

the acquisition module is used for acquiring sound wave transmission data in the oil and gas pipeline; the sound wave transmission data comprise multi-source sound wave transmission data in different environments; the data collected by each distributed optical fiber sensor in different environments is used as a data source;

the multi-source data characteristic analysis module is used for carrying out neighborhood construction and correlation extraction from the sound wave transmission data through a multi-source data characteristic analysis model so as to obtain acoustic transmission scene characteristics corresponding to each oil and gas pipeline segment;

the dynamic self-adaptive positioning module is used for predicting the acoustic transmission scene characteristics through a dynamic self-adaptive positioning model so as to obtain detection event predicted values corresponding to each oil and gas pipeline segment in the oil and gas pipeline; the higher the predicted value of the detection event is, the higher the probability of the detection event in the target oil and gas pipeline segment is;

the dynamic early warning information display module is used for displaying dynamic early warning information corresponding to the target oil gas pipeline segments in a dynamic early warning monitoring distribution diagram according to the geographic positions of the target oil gas pipeline segments for the target oil gas pipeline segments with the detection event predicted values meeting the early warning conditions; the display form of the dynamic early warning information changes along with the change of the predicted value of the detection event of the target oil and gas pipeline section.

In a third aspect, embodiments of the present application provide a computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the method according to the first aspect described above when executing the computer program.

Compared with the prior art, the embodiment of the application has the beneficial effects that:

in the technical scheme provided by the embodiment of the application, firstly, acoustic wave transmission data in an oil and gas pipeline are acquired. The sound wave transmission data comprise multi-source sound wave transmission data in different environments; the data collected by each distributed optical fiber sensor in different environments is used as a data source. Therefore, through multi-source acoustic wave transmission data under different environments, the method can provide original data with more abundant information for subsequent early warning analysis for early warning monitoring of a distributed optical fiber acoustic wave sensing system matched with a complex geographic environment, and further improves the accuracy of early warning monitoring. And carrying out neighborhood construction and correlation extraction from the acoustic transmission data through a multi-source data characteristic analysis model to obtain acoustic transmission scene characteristics corresponding to each oil and gas pipeline segment. Therefore, the acoustic wave transmission data from different sources can be fully utilized, and the characteristic information contained in the acoustic wave transmission data can be extracted more comprehensively, so that the characteristic diversity and the accuracy of model characteristic extraction are improved. And then, predicting the characteristics of the acoustic transmission scene through a dynamic self-adaptive positioning model to obtain detection event predicted values corresponding to each oil and gas pipeline segment in the oil and gas pipeline. The higher the predicted value of the detection event is, the higher the probability of the detection event in the target oil and gas pipeline segment is. The predicted values of the detection events can reflect the probability of the detection events of each segment in the oil and gas pipeline, so that the quick positioning of potential problems in the pipeline is assisted, and corresponding measures can be taken in time. Finally, for the target oil gas pipeline segments with the predicted values of the detection events meeting the early warning conditions, displaying dynamic early warning information corresponding to the target oil gas pipeline segments in a dynamic early warning monitoring distribution diagram according to the geographic positions of the target oil gas pipeline segments; the display form of the dynamic early warning information changes along with the change of the predicted value of the detection event of the target oil and gas pipeline section. Through the display of the dynamic early warning information, the early warning state and the risk degree of the target oil gas pipeline section can be intuitively known, and the safe operation of the pipeline is ensured.

In the embodiment of the application, the multisource acoustic transmission data is processed through multisource data characteristic analysis and the dynamic self-adaptive positioning model, so that the monitoring accuracy of the oil and gas pipeline can be effectively improved, and particularly the monitoring accuracy in a complex underground environment is improved. Meanwhile, the early warning state and the risk degree of the target oil gas pipeline section can be displayed more intuitively in the dynamic early warning monitoring distribution diagram through the dynamic early warning information, so that abnormal treatment measures can be triggered, and the safety of the oil gas pipeline is ensured.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following description will briefly introduce the drawings that are needed in the embodiments or the description of the prior art, it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of a monitoring and early warning method based on distributed optical fiber acoustic wave sensing provided by an embodiment of the application;

fig. 2 is a schematic structural diagram of a monitoring and early warning device based on distributed optical fiber acoustic wave sensing according to an embodiment of the present application;

Fig. 3 is a schematic structural diagram of a computer device according to an embodiment of the present application.

Detailed Description

In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system configurations, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.

It should be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.

It should also be understood that the term "and/or" as used in this specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.

As used in this specification and the appended claims, the term "if" may be interpreted as "when..once" or "in response to a determination" or "in response to detection" depending on the context. Similarly, the phrase "if a determination" or "if a [ described condition or event ] is detected" may be interpreted in the context of meaning "upon determination" or "in response to determination" or "upon detection of a [ described condition or event ]" or "in response to detection of a [ described condition or event ]".

In addition, in the description of the present application and the appended claims, the terms "first," "second," "third," and the like are used merely to distinguish between descriptions and are not to be construed as indicating or implying relative importance.

Reference in the specification to "one embodiment" or "some embodiments" or the like means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," and the like in the specification are not necessarily all referring to the same embodiment, but mean "one or more but not all embodiments" unless expressly specified otherwise. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless expressly specified otherwise.

The following describes the technical solutions of the embodiments of the present application.

Referring to fig. 1, a schematic flow chart of a monitoring and early warning method based on distributed optical fiber acoustic wave sensing according to an embodiment of the present application is provided, and the method may be applied to a server, and includes the following steps:

s101, acquiring sound wave transmission data in an oil and gas pipeline.

In this embodiment of the present application, the acoustic transmission data includes multi-source acoustic transmission data under different environments. Further, the data collected by each distributed optical fiber sensor in different environments is used as a data source.

In practical applications, the multi-source acoustic transmission data includes acoustic transmission data caused by temperature variation, for example: the temperature change causes expansion and contraction of the tubing material, which in turn causes a change in acoustic transmission. The change of the surface temperature of the pipeline is monitored through the distributed optical fiber sensor, and corresponding sound wave signals are recorded, so that sound wave transmission data can be obtained at different environment temperatures.

In another example, the multi-source acoustic transmission data includes acoustic transmission data caused by water flow: the speed and flow pattern of the water flow can affect the transmission characteristics of sound waves in the pipe. The change of water flow near the oil gas pipeline is monitored through the distributed optical fiber sensor, and corresponding sound wave signals are recorded, so that sound wave transmission data caused by different environmental water flows are obtained.

In yet another example, the multi-source acoustic transmission data includes acoustic transmission data caused by geological changes: changes in the geologic structure can have an effect on the transmission of sound waves. And monitoring the change of geology around the oil and gas pipeline through a distributed optical fiber sensor, and recording corresponding sound wave signals so as to acquire sound wave transmission data caused by different environmental geology changes.

In yet another example, the multi-source acoustic transmission data includes acoustic transmission data caused by pipe vibration: vibration of the pipe may generate sound waves and affect the transmission of the sound waves in the pipe. And monitoring the vibration condition of the pipeline through a distributed optical fiber sensor, and recording corresponding sound wave signals so as to acquire sound wave transmission data caused by vibration of pipelines in different environments.

The multi-source acoustic wave transmission data are not limited to the four examples, and the acoustic transmission scene of the oil and gas pipeline can be more comprehensively known by collecting and analyzing the multi-source acoustic wave transmission data under different environments, so that richer original data are provided for subsequent early warning analysis.

Specifically, in an alternative embodiment, the acquiring the acoustic transmission data in the oil and gas pipeline in S101 may be implemented as:

s201, acquiring optical fiber vibration data acquired by each distributed optical fiber sensor in different environments;

S202, preprocessing the optical fiber vibration data of each distributed optical fiber sensor in different environments;

s203, extracting multi-scale sound wave transmission characteristics from the preprocessed optical fiber vibration data to obtain sound wave time lag relation characteristics, sound wave self-adaptive filtering characteristics and sound rate universality characteristics corresponding to each data source;

s204, carrying out feature fusion processing on the sound wave time lag relation feature, the sound wave self-adaptive filtering feature and the acoustic rate universality feature corresponding to each data source to obtain sound wave transmission fusion feature data corresponding to each data source.

Further optionally, in S203, the multi-scale acoustic wave transmission feature is extracted from the preprocessed optical fiber vibration data, so as to obtain an acoustic wave time lag relation feature, an acoustic wave adaptive filtering feature and an acoustic frequency universality feature corresponding to each data source, which may be obtained by:

for acoustic wave time lag relationship feature extraction, the acoustic wave time lag relationship feature reflects propagation delays of acoustic wave signals between different sensor nodes. The time lag difference between the sensor nodes is calculated by cross-correlation analysis. The method comprises the following specific steps: carrying out cross-correlation analysis on the sound wave signal on each sensor node and a reference signal; finding out the peak value of the cross-correlation function, wherein the position of the peak value corresponds to propagation delay; and extracting time lag differences among different sensor nodes as sound wave time lag relation characteristics.

For acoustic wave adaptive filtering feature extraction, the acoustic wave adaptive filtering feature reflects the adaptive filtering process in the acoustic wave signal. The acoustic wave adaptive filtering may be processed in the time domain or the frequency domain. The method comprises the following specific steps: performing a filtering operation on the acoustic wave signal at each sensor node using a suitable adaptive filtering algorithm, such as least mean square filtering (LMS) or an adaptive filter; features of the filtered acoustic wave signal, such as signal energy, spectral features (e.g., dominant frequency or band energy), are extracted to represent acoustic wave adaptive filtering features.

For acoustic frequency commonality feature extraction, acoustic frequency commonality features may reveal frequency components that are prevalent in acoustic wave signals. For example, the extraction method is to obtain features of the frequency domain using spectral analysis. The method comprises the following specific steps: performing a spectral analysis on the acoustic signal at each sensor node, for example using a Fast Fourier Transform (FFT); obtaining a spectrogram, and selecting a relevant frequency range according to requirements; spectral features such as frequency energy distribution, spectral morphology features (e.g., spectral peaks, frequency bandwidths, etc.) are extracted as acoustic frequency generalization features.

The relevant information in the acoustic wave transmission data is better described and the subsequent analysis and application are carried out by extracting the acoustic wave time lag relation characteristic, the acoustic wave self-adaptive filtering characteristic and the acoustic frequency universality characteristic corresponding to each data source.

S102, carrying out neighborhood construction and correlation extraction from sound wave transmission data through a multi-source data characteristic analysis model to obtain acoustic transmission scene characteristics corresponding to each oil and gas pipeline segment.

The multi-source data characteristic analysis model can be a polynomial linear regression model, which is a generalized linear model used for modeling and predicting discrete variables.

As an alternative embodiment, the multi-source data feature analysis model includes: the system comprises a fine granularity analysis layer, a neighborhood construction layer, a correlation analysis layer, a dimension reduction processing layer and a feature fusion layer.

S102, carrying out neighborhood construction and correlation extraction from acoustic transmission data through a multi-source data characteristic analysis model to obtain acoustic transmission scene characteristics corresponding to each oil and gas pipeline segment, wherein the method can be realized as follows:

s301, carrying out multi-source data fine-granularity analysis on the sound wave transmission data through a fine-granularity analysis layer so as to obtain fine-granularity sound wave transmission characteristics; wherein fine-grained acoustic transmission characteristics are distributed among different frequency domain sub-bands.

In this step, by performing fine-granularity analysis on the acoustic transmission data, the acoustic signal can be decomposed into different frequency-domain subbands, thereby obtaining more detailed acoustic transmission characteristics. The fine granularity analysis can provide more information, reveal the change rule of the acoustic wave transmission in different frequency ranges, and help to understand the characteristics of the acoustic wave transmission in different frequency domains.

S302, carrying out neighborhood construction processing on fine-grained sound wave transmission characteristics in different frequency domain sub-bands through a neighborhood construction layer so as to obtain sound wave transmission neighborhood characteristics; the sound wave transmission neighborhood characteristics are used for representing the interactive relation between different positions of each oil and gas pipeline segment and acoustic transmission transformation factors in different neighborhood.

In the step, through neighborhood construction processing, the interactive relation between different positions of each oil and gas pipeline segment and transformation factors of acoustic transmission under different neighborhoods can be captured. This helps to distinguish between acoustic transmission characteristic differences between different locations and neighborhoods, thereby better describing the acoustic transmission scenario of the oil and gas pipeline.

S303, obtaining correlation measurement information between each characteristic point in the sound wave transmission neighborhood characteristic and surrounding neighborhood characteristic points through a correlation analysis layer; the relevance measurement information is used for representing the association relation between acoustic transmission transformation factors corresponding to the oil and gas pipeline segments.

In the embodiment of the application, the spearman rank correlation coefficient may be used as the correlation measurement information, the kendel rank correlation coefficient may be used as the correlation measurement information, and the spearman rank correlation coefficient and the kendel rank correlation coefficient may be used as the correlation measurement information.

In the step, the association relation between acoustic transmission transformation factors of different oil and gas pipeline segments can be measured by analyzing the correlation measurement information between each characteristic point in the acoustic transmission neighborhood characteristic and the surrounding neighborhood characteristic points. This helps to determine the importance and dependency between different features, providing clues for further understanding of the acoustic transmission scenario.

S304, dimension reduction is carried out on the sound wave transmission neighborhood characteristics based on the correlation measurement information through a dimension reduction processing layer so as to obtain sound wave transmission dimension reduction characteristics.

In the step, the dimension reduction processing based on the correlation measurement information can reduce the dimension of transmission data, remove irrelevant or redundant features, and improve the characterization capability and calculation efficiency of the features. The acoustic wave transmission characteristics after dimension reduction are more representative, and the subsequent characteristic analysis and processing are convenient.

And S305, carrying out feature fusion on the sound wave transmission dimension reduction features through a feature fusion layer so as to obtain acoustic transmission scene features.

In the step, the acoustic wave transmission characteristics after dimension reduction are integrated through characteristic fusion, so that more comprehensive acoustic transmission scene characteristics are obtained. The feature fusion can comprehensively consider the relation among different features, enhance the description capability of the acoustic transmission scene and provide more accurate prediction or judgment.

Through the multi-source data feature analysis of steps S301 through S305, acoustic transmission scene features of individual hydrocarbon pipe segments may be more fully extracted, characterized, and described. These features help to understand in depth the relationship between the law of variation, location and neighborhood of acoustic transmission, improving the ability to monitor and predict oil and gas pipelines, thereby optimizing safe operation and risk management of the pipelines.

S103, predicting the characteristics of the acoustic transmission scene through a dynamic self-adaptive positioning model to obtain detection event predicted values corresponding to each oil and gas pipeline segment in the oil and gas pipeline.

In the embodiment of the application, the higher the predicted value of the detection event is, the higher the probability of the detection event in the target oil and gas pipeline segment is.

The dynamic self-adaptive positioning model can be constructed based on a deep neural network model.

As an alternative embodiment, the dynamic adaptive positioning model includes, but is not limited to: an adaptive reinforcement learning layer, a dynamic adaptive value learning layer and an experience playback layer. The self-adaptive reinforcement learning layer is mainly constructed by a plurality of convolutional neural network layers, and features with different sizes are extracted through convolutional kernels with different sizes, so that more comprehensive feature extraction is realized; the dynamic self-adaptive value learning layer is mainly formed by constructing a feedforward connecting layer and a plurality of hidden layers, and calculates a target predicted value of a detection event corresponding to each segment by combining an initial predicted value and a historical self-adaptive value; the experience playback layer mainly contains an experience buffer for holding history adaptation values.

S103, predicting the acoustic transmission scene characteristics through a dynamic self-adaptive positioning model to obtain detection event predicted values corresponding to each oil and gas pipeline segment in the oil and gas pipeline, wherein the detection event predicted values can be realized as follows:

s401, extracting multi-scale abstract features of the acoustic transmission scene features through the self-adaptive reinforcement learning layer, and performing detection event prediction processing based on the extracted multi-scale abstract features to obtain initial prediction values of detection events corresponding to the oil and gas pipeline segments.

S402, calculating the initial predicted value and the historical adaptive value stored in the experience playback layer through the dynamic adaptive value learning layer to obtain the detection event target predicted value corresponding to each oil and gas pipeline segment.

In this alternative embodiment, the dynamic adaptive positioning model includes an adaptive reinforcement learning layer, a dynamic adaptive value learning layer, and an empirical playback layer. Through the model, the prediction of the acoustic transmission scene characteristics can be realized, and the predicted value of the detection event of each segment in the oil and gas pipeline can be obtained.

It can be understood that the multi-scale abstract feature extraction is performed by the self-adaptive reinforcement learning layer, and the model can flexibly extract relevant features according to specific acoustic transmission scene features, so that the accuracy and the robustness of prediction are improved. And secondly, calculating a target predicted value according to the initial predicted value and the historical adaptive value through a dynamic adaptive value learning layer, and adjusting the adaptive value according to the current situation by the model in the process of updating the historical adaptive value, thereby improving the accuracy and the adaptability of prediction. In addition, the use of the experience playback layer can save and recycle the predicted value of the history detection event, and the newly generated target predicted value can be stored in the experience playback layer through the updating rule of the dynamic self-adaptive value learning layer so as to supplement the history information and improve the prediction performance. In a comprehensive view, the dynamic self-adaptive positioning model in the embodiment combines the advantages of self-adaptive reinforcement learning, dynamic self-adaptive value learning and experience playback, and can improve the prediction capability of detection events of an acoustic transmission scene and gradually optimize the performance of the model through flexible feature extraction and self-adaptive value updating.

In an alternative embodiment, in S402, after calculating the initial predicted value and the historical adaptive value stored in the experience playback layer to obtain the detected event target predicted value corresponding to each oil and gas pipeline segment, the detected event target predicted value is circularly stored in the experience playback layer; and updating the historical adaptive value stored in the experience playback layer based on the detection event target predicted value by adopting a preset dynamic adaptive value updating rule. In the dynamic adaptive value updating rule, the historical adaptive value Q (s, a) required to be updated is obtained through the following formula, namely:

；

The application of such a dynamic adaptive value update rule may bring about several beneficial effects: first, by introducing rewards and discount factors, the importance of instant rewards as well as future rewards for current actions can be taken into account, helping to better guide the updating of historical adaptive values. Second, by circularly storing the detected event target predicted value in the experience playback layer, and based on the maximum adaptive value of the newly generated predicted value, the historical adaptive value can be gradually learned and adjusted to better adapt to the current detected event predicted requirement. Thirdly, the updating rule of the dynamic self-adaptive value can flexibly adjust the self-adaptive value according to the current detection event target predicted value and the historical self-adaptive value, so that the prediction performance and the adaptability of the model are improved. In summary, by circularly storing the target predicted value of the detected event in the experience playback layer and updating the history adaptive value by using the dynamic adaptive value updating rule, learning and adaptive capabilities of the model can be promoted, and prediction accuracy and stability of the detected event can be improved.

In an alternative embodiment, the dynamic adaptive positioning model further comprises a target network layer. The target network layer has the same network structure and the same network parameters as the self-adaptive reinforcement learning layer. The network parameter update frequency of the target network layer is lower than the network parameter update frequency of the adaptive reinforcement learning layer. After S402, the multi-scale abstract feature extraction is performed on the acoustic transmission scene feature through the target network layer, and the detection event prediction processing is performed based on the extracted multi-scale abstract feature, so as to obtain the detection event reference predicted value corresponding to each oil and gas pipeline segment. Further, a quantile regression loss value between the reference predicted value and the initial predicted value is calculated. And finally, carrying out gradient descent optimization processing on the self-adaptive reinforcement learning layer based on the fractional regression loss value so as to update the self-adaptive network parameters in the self-adaptive reinforcement learning layer.

Wherein, alternatively, the mathematical expression for calculating the quantile regression loss value between the reference predicted value and the initial predicted value may be expressed as:

；

wherein,representing fractional regression loss values, +.>Representing the quantile level (0</><1），/>Representing a reference predictor->Representing the initial predicted value- >Is a weight coefficient.

By calculating the loss function L, the difference between the reference predicted value and the initial predicted value can be measured and the model is guided to adjust parameters in the adaptive reinforcement learning layer to reduce this difference. The quantile level delta can be selected according to task requirements and specific scenes.

It should be noted that, when the gradient descent optimization process is performed, the loss function L may be derived, and then the adaptive network parameters in the adaptive reinforcement learning layer are updated according to an optimization algorithm (such as a random gradient descent method) so as to optimize the adaptive network parameters in a direction of reducing the loss function. The selection of the specific weight coefficient ρ may be adjusted according to the specific situation to balance the importance between the reference predictor and the initial predictor. The appropriate weighting coefficients may be set according to the data distribution and task requirements.

The model can be helped to adapt to different acoustic transmission scenes step by calculating quantile regression loss values and optimizing the self-adaptive network parameters in the self-adaptive reinforcement learning layer based on the loss values, so that the prediction capability and accuracy of each sectional detection event in the oil and gas pipeline are improved.

Specifically, the dynamic adaptive positioning model includes a target network layer and an adaptive reinforcement learning layer, which have the same network structure and network parameters. The parameter update frequency of the target network layer is lower, and the parameter update frequency of the adaptive reinforcement learning layer is higher. Based on the above, after S402, multi-scale abstract feature extraction is performed on the acoustic transmission scene features through the target network layer, and detection event prediction processing is performed by using the extracted multi-scale abstract features, so as to obtain reference prediction values of each oil and gas pipeline segment. The reference predictor may be regarded as a predictor obtained by the target network layer in the current transmission scenario. Then, a quantile regression loss value between the reference predicted value and the initial predicted value is calculated. The quantile regression loss value is used to measure the deviation of the reference predicted value on the quantile. By calculating the loss value, the prediction performance of the target network layer in the current transmission scene can be known, and the model is optimized by taking the prediction performance as an evaluation index. And finally, carrying out gradient descent optimization processing on the self-adaptive reinforcement learning layer based on the fractional regression loss value so as to update the self-adaptive network parameters in the self-adaptive reinforcement learning layer. Therefore, the parameters of the self-adaptive reinforcement learning layer can be continuously adjusted by using the loss value as an optimization basis, so that the model is better adapted to the current acoustic transmission scene, and the accuracy and the robustness of prediction are improved.

In a specific example, the target network layer predicts the oil and gas pipeline segments according to the characteristics of the acoustic transmission scene, and obtains a reference predicted value. Then, a quantile regression loss value between the reference predicted value and the initial predicted value is calculated. If the loss value is larger, a larger deviation may exist in the prediction of the target network layer in the current acoustic transmission scene. And carrying out gradient descent optimization treatment on the self-adaptive reinforcement learning layer according to the magnitude of the loss value, and adjusting self-adaptive network parameters. Through repeated iterative optimization process, the parameters of the self-adaptive reinforcement learning layer are gradually updated, so that the model is more accurate and stable for prediction under different acoustic transmission scenes.

According to the embodiment, the method and the device can adapt to the continuously-changing acoustic transmission scene in the dynamic environment, update the model parameters by optimizing the loss value, and improve the adaptability and the prediction performance of the model. By combining training and optimizing the target network layer and the self-adaptive reinforcement learning layer, the characteristics of the acoustic transmission scene can be better processed, and the detection event prediction capability of the oil and gas pipeline segment is improved.

In an alternative embodiment, the experience playback layer further includes an experience playback memory library. After S402, historical detection event target predicted values of discrete distribution are randomly sampled from the experience playback memory bank, and corresponding acoustic transmission scene features are extracted. And extracting the multi-scale abstract features of the extracted acoustic transmission scene features through the target network layer, and performing detection event prediction processing based on the extracted multi-scale abstract features to obtain detection event sampling predicted values corresponding to the oil and gas pipeline segments. And then, performing gradient descent optimization processing on the self-adaptive reinforcement learning layer based on the sampled historical detection event target predicted value and the corresponding detection event sampling predicted value so as to update the self-adaptive network parameters in the self-adaptive reinforcement learning layer.

In the above alternative embodiment, an experience playback memory bank is also added in the experience playback layer. The experience playback memory is a data structure for storing historical detected event target predictors.

Based on this structure, specifically, after S402, historical detection event target predicted values are randomly sampled from the experience playback memory bank, and corresponding acoustic transmission scene features are extracted. The purpose of this is to train the model using historical data and enable it to take full account of past experience in predicting current events. The extracted acoustic transmission scene features are extracted through the target network layer in a multi-scale abstract mode, features on different scales can be extracted, and the characteristics of acoustic transmission data are more comprehensively described. And then, based on the extracted multi-scale abstract features, carrying out detection event prediction processing to obtain detection event sampling predicted values corresponding to the oil and gas pipeline segments. Thus, the historical experience can be combined with the current acoustic transmission data, and the accuracy and reliability of the prediction of the detection event can be improved. And then, based on the sampled historical detection event target predicted value and the corresponding detection event sampling predicted value, performing gradient descent optimization processing through the self-adaptive reinforcement learning layer so as to update the self-adaptive network parameters in the self-adaptive reinforcement learning layer. Therefore, the weight and the parameters of the model can be adjusted by comparing the learning historical data and the current data, and the performance and the adaptability of the prediction model are further improved.

In this embodiment, the historical data in the experience playback memory is utilized, and through random sampling of discrete distributions, the prediction model can learn the historical experiences more comprehensively, and the experience can be utilized to optimize the prediction accuracy. By extracting multi-scale abstract features, characteristics of acoustic transmission scenes on various scales can be captured better, so that a prediction model can more comprehensively understand and predict detection events. By using the self-adaptive reinforcement learning layer, the self-adaptive network parameters of the model can be optimized by comparing the historical data with the current data, so that the adaptability and performance of the prediction model can be further improved. In summary, on the basis of the experience playback memory library, the model can be optimized by comprehensively utilizing the historical experience and the current data, and the accuracy and the stability of the prediction of the detection event can be improved.

And S104, for the target oil gas pipeline segment with the predicted value of the detection event meeting the early warning condition, displaying dynamic early warning information corresponding to the target oil gas pipeline segment in a dynamic early warning monitoring distribution diagram according to the geographic position of the target oil gas pipeline segment.

The display form of the dynamic early warning information changes along with the change of the predicted value of the detection event of the target oil and gas pipeline section. The dynamic early warning monitoring distribution map is a GIS geographic position grid graph or a three-dimensional local scene map of the oil and gas pipeline.

In the embodiment of the present application, the geographic location is map data, and the map data refers to a data set including geospatial information. It records the geographic features, the distribution of features, the geographic coordinates, the network of roads, the distribution of water areas, the elevation of terrain, etc. of a particular region on earth. Map data exists in a digitized form and can be used in a plurality of fields such as map making, geographic Information System (GIS) analysis, navigation application and the like. In practice, the map data includes geographic coordinates, such as longitude and latitude, of various locations and areas within the target area, as well as boundary lines of various administrative areas, to aid in determining geographic locations and spatial ranges. The map data also includes the location and connection of traffic networks such as roads, highways, railways, etc. The data comprises attribute information such as road types, traffic flows, speed limits and the like, and provides support for navigation, traffic planning and traffic management. Map data is usually obtained from data sources such as geographical investigation, remote sensing images, aerial photography, satellite observation and the like, and is obtained through technical processing such as processing, integration, modeling and the like. These map data provide accurate, visual spatial information about various locations and areas on the earth, facilitating annotation analysis of the deployment of the oil and gas pipelines in different geographical areas in the present application.

Specifically, as an alternative embodiment, taking a GIS system as an example, in S104, it is determined whether the predicted value of the detection event corresponding to each oil and gas pipeline segment meets the pre-warning condition. And further, for the target oil gas pipeline segments with the predicted values of the detection events meeting the early warning conditions, the GIS geographic position grid coordinates where the target oil gas pipeline segments are located are obtained from the GIS geographic position mapping table, and a GIS geographic position grid coordinate graph is obtained. And finally, generating dynamic early warning information matched with the predicted value of the detection event, and displaying the dynamic early warning information in the GIS geographic position grid coordinate graph.

The following describes the beneficial effects of the above steps by way of a specific example:

firstly, a detection event classification model is constructed for carrying out prediction classification on detection events of each oil and gas pipeline segment. For example, the classification model may predict whether there are conditions in each hydrocarbon pipe section where pipe vibration exceeds an early warning threshold. By comparing with the early warning conditions, whether the predicted value of the detection event of each segment meets the early warning conditions can be judged. Furthermore, in the GIS geographic location mapping table, the geographic location grid coordinates where each oil and gas pipeline segment is located may be stored. And obtaining geographic position grid information of the corresponding target oil gas pipeline segment from the GIS mapping table according to the oil gas pipeline segment meeting the early warning condition, and obtaining a GIS geographic position grid coordinate graph. And then, generating corresponding dynamic early warning information according to the target oil gas pipeline segments meeting the early warning conditions. The dynamic warning information may include segment number, type of warning, countermeasures taken, etc. The early warning information is superimposed on the GIS geographic position grid graph, so that the position where the early warning condition exists in the oil gas pipeline system can be intuitively displayed, related personnel can be helped to quickly know the early warning information, and corresponding measures are taken. For example, the predicted value of the detected event in the segment A exceeds the early warning threshold value of the pipeline vibration, and meets the early warning condition. And obtaining the geographic position grid coordinates corresponding to the segment A through the GIS geographic position mapping table. Then, generating dynamic early warning information according to the early warning information, such as section A: the pipeline vibration exceeds the early warning threshold value and needs to be overhauled immediately, and the dynamic early warning information is displayed in the GIS geographic position grid coordinate graph, so that the position of the segment A and the corresponding early warning information can be clearly seen on the graph.

By the aid of the example, whether the detection event of the oil and gas pipeline section meets the early warning condition or not can be judged rapidly through the multi-source data characteristic analysis model, early warning information is dynamically displayed on the geographic position grid chart, relevant personnel can know and take corresponding measures rapidly, and safety and operation efficiency of a pipeline system are improved.

Further alternatively, dynamic early warning information matched with the predicted value of the detection event is generated, and after the dynamic early warning information is displayed in the GIS geographic position grid coordinate graph, a three-dimensional local scene map of the oil and gas pipeline is constructed based on the GIS geographic position grid coordinate graph and a preset GIS geographic two-dimensional panorama. The display main body in the three-dimensional local scene map of the oil and gas pipeline is a target oil and gas pipeline segment indicated by dynamic early warning information and an early warning event state of the target oil and gas pipeline segment. And pushing the three-dimensional local scene map of the oil and gas pipeline to a monitoring and early warning interface of the user terminal.

Specifically, the three-dimensional local scene map of the oil and gas pipeline is constructed based on the GIS geographic position and the predicted event state of the target oil and gas pipeline segment and a preset GIS geographic two-dimensional panorama. The visual display of the three-dimensional local scene of the oil and gas pipeline can be realized by combining the segmented position and state information with global geographic information data. On the three-dimensional local scene map of the oil and gas pipeline, the information such as the predicted value of the detection event, the real-time running state of the pipeline, the type of the early warning event and the like can be displayed in the form of images. For example, segment a is determined to be in the presence of pipe vibration exceeding an early warning threshold and is illustrated in a GIS geographic location grid graph. By constructing the three-dimensional local scene map of the oil and gas pipeline, the position of the segment A can be displayed in the three-dimensional scene, and the early warning state of the segment is represented by different colors or marks, so that a user can intuitively know the running condition of the oil and gas pipeline system. In a monitoring and early warning interface of a user terminal, pushing the three-dimensional local scene map of the oil and gas pipeline to a user, and improving the monitoring and early warning capability of the user on the oil and gas pipeline system through a multidimensional display mode. Meanwhile, when an early warning event occurs, a user can quickly know the real-time state and the position of the oil and gas pipeline section related to the event, take measures in time and reduce the pipeline operation risk.

Therefore, through the steps, multidimensional monitoring and early warning of the oil and gas pipeline can be realized, and the safety and efficiency of the operation of the pipeline are improved. Meanwhile, through the construction of the three-dimensional local scene map of the oil and gas pipeline, a user can more intuitively know the running state of the pipeline system and better understand monitoring and early warning information of the oil and gas pipeline segments reflected by the predicted value of the detection event.

It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic of each process, and should not limit the implementation process of the embodiment of the present application in any way.

Corresponding to the monitoring and early warning method based on distributed optical fiber acoustic wave sensing described in the above embodiments, fig. 2 shows a block diagram of the monitoring and early warning device based on distributed optical fiber acoustic wave sensing provided in the embodiments of the present application, and for convenience of explanation, only the portions related to the embodiments of the present application are shown.

Referring to fig. 2, the apparatus includes:

an acquisition module 21, configured to acquire acoustic transmission data in an oil and gas pipeline; the sound wave transmission data comprise multi-source sound wave transmission data in different environments; the data collected by each distributed optical fiber sensor in different environments is used as a data source;

the multi-source data feature analysis module 22 is configured to perform neighborhood construction and correlation extraction from the acoustic transmission data through a multi-source data feature analysis model, so as to obtain acoustic transmission scene features corresponding to each oil and gas pipeline segment;

the dynamic self-adaptive positioning module 23 is configured to predict the acoustic transmission scene feature through a dynamic self-adaptive positioning model, so as to obtain a predicted value of a detection event corresponding to each oil and gas pipeline segment in the oil and gas pipeline; the higher the predicted value of the detection event is, the higher the probability of the detection event in the target oil and gas pipeline segment is;

the dynamic early warning information display module 24 is configured to display, for a target oil gas pipeline segment whose predicted value of a detected event meets an early warning condition, dynamic early warning information corresponding to the target oil gas pipeline segment in a dynamic early warning monitoring distribution diagram according to a geographic position where the target oil gas pipeline segment is located; the display form of the dynamic early warning information changes along with the change of the predicted value of the detection event of the target oil and gas pipeline section.

It should be noted that, because the content of information interaction and execution process between the above devices/units is based on the same concept as the method embodiment of the present application, specific functions and technical effects thereof may be referred to in the method embodiment section, and will not be described herein again.

Fig. 3 is a schematic structural diagram of a computer device according to an embodiment of the present application. As shown in fig. 3, the computer device 3 of this embodiment includes: at least one processor 30, a memory 31 and a computer program 32 stored in the memory 31 and executable on the at least one processor 30, the processor 30 implementing the steps of any of the various method embodiments described above when executing the computer program 32.

The computer equipment can be a desktop computer, a notebook computer, a palm computer, a cloud server and other computing equipment. The computer device may include, but is not limited to, a processor 30, a memory 31. It will be appreciated by those skilled in the art that fig. 3 is merely an example of the computer device 3 and is not meant to be limiting as the computer device 3, and may include more or fewer components than shown, or may combine certain components, or different components, such as may also include input-output devices, network access devices, etc.

The processor 30 may be a central processing unit (Central Processing Unit, CPU), the processor 30 may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory 31 may in some embodiments be an internal storage unit of the computer device 3, such as a hard disk or a memory of the computer device 3. The memory 31 may in other embodiments also be an external storage device of the computer device 3, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card) or the like, which are provided on the computer device 3. Further, the memory 31 may also include both an internal storage unit and an external storage device of the computer device 3. The memory 31 is used for storing an operating system, application programs, boot loader (BootLoader), data, other programs etc., such as program codes of the computer program etc. The memory 31 may also be used for temporarily storing data that has been output or is to be output.

It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.

Embodiments of the present application also provide a computer readable storage medium storing a computer program which, when executed by a processor, implements steps that may implement the various method embodiments described above.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present application implements all or part of the flow of the method of the above embodiments, and may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, where the computer program, when executed by a processor, may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include at least: any entity or device capable of carrying computer program code to a computer device, a recording medium, computer Memory, read-Only Memory (ROM), random access Memory (RAM, random Access Memory), electrical carrier signals, telecommunications signals, and software distribution media. Such as a U-disk, removable hard disk, magnetic or optical disk, etc. In some jurisdictions, computer readable media may not be electrical carrier signals and telecommunications signals in accordance with legislation and patent practice.

In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus/network device and method may be implemented in other manners. For example, the apparatus/network device embodiments described above are merely illustrative, e.g., the division of the modules or units is merely a logical functional division, and there may be additional divisions in actual implementation, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection via interfaces, devices or units, which may be in electrical, mechanical or other forms.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application.

Claims

1. A monitoring and early warning method based on distributed optical fiber acoustic wave sensing is characterized by comprising the following steps:

2. The monitoring and early warning method based on distributed optical fiber acoustic wave sensing according to claim 1, wherein the acquiring of acoustic wave transmission data in an oil and gas pipeline comprises:

3. The monitoring and early warning method based on distributed optical fiber acoustic wave sensing according to claim 1, wherein the multi-source data characteristic analysis model comprises a fine granularity analysis layer, a neighborhood construction layer, a correlation analysis layer, a dimension reduction processing layer and a characteristic fusion layer;

4. The monitoring and early warning method based on distributed optical fiber acoustic wave sensing according to claim 1, wherein the dynamic self-adaptive positioning model comprises a self-adaptive reinforcement learning layer, a dynamic self-adaptive value learning layer and an experience playback layer;

5. The method for monitoring and early warning based on distributed optical fiber acoustic wave sensing according to claim 4, wherein after calculating the initial predicted value and the historical adaptive value stored in the experience playback layer to obtain the detected event target predicted value corresponding to each oil and gas pipeline segment, further comprising:

；

6. The monitoring and early warning method based on distributed optical fiber acoustic wave sensing according to claim 4, wherein the dynamic self-adaptive positioning model further comprises a target network layer;

7. The monitoring and early warning method based on distributed optical fiber acoustic wave sensing according to claim 4, wherein the experience playback layer further comprises an experience playback memory bank;

8. The monitoring and early warning method based on distributed optical fiber acoustic wave sensing according to claim 1, wherein the dynamic early warning and monitoring distribution diagram is a GIS geographic position grid coordinate diagram, the GIS geographic position grid coordinate diagram is provided with a GIS geographic position mapping table, and the GIS geographic position mapping table stores GIS geographic position grid coordinates corresponding to oil and gas pipeline segments;

9. Monitoring and early warning device based on distributed optical fiber acoustic wave sensing, characterized by comprising:

10. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the method of any of claims 1 to 8 when executing the computer program.