Method and device for generating pipeline integrity management plan

Abstract

The present invention discloses a method and device for generating a pipeline integrity management solution, which relates to the field of oilfield survey technology, and is mainly used to solve the problem of high manpower consumption in the current generation process of pipeline integrity management solutions. The method includes: training a pipeline knowledge extraction model through a preset corpus sample and a preset algorithm, wherein the preset corpus sample is obtained based on manual annotation of part of the pipeline data, and the preset algorithm is a machine learning algorithm selected based on the pipeline data type; performing knowledge extraction on the pipeline data according to the pipeline knowledge extraction model to obtain a target knowledge graph; adding data to a preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management solution, wherein the preset template is determined based on the type of the pipeline integrity management solution. The present invention is used in the generation process of a pipeline integrity management solution.

Description

Technical Field

[0001] The invention relates to the technical field of oilfield survey, and in particular to a method and device for generating a pipeline integrity management plan. Background Art

[0002] In the process of oil and gas exploration, development and production, pipeline integrity management has always been an important link because oil and gas transmission cannot be separated from pipelines. Due to the complexity of pipeline data, in the process of writing pipeline integrity management plans based on pipeline data, it is generally necessary to manually obtain the required knowledge information from the pipeline data and add the knowledge information to the corresponding template to generate a pipeline integrity management plan.

[0003] At present, in the process of generating pipeline integrity management plans, the plans are usually written manually. However, since the required knowledge information needs to be extracted from the pipeline data during the writing process, and this information often requires a lot of manpower in the search process, especially when the number of pipeline integrity management plans to be generated is large, the existing manual generation method has the problem of high manpower consumption. Summary of the invention

[0004] In view of the above problems, the present invention provides a method and device for generating a pipeline integrity management plan, the main purpose of which is to solve the problem of manpower consumption in the current process of generating a pipeline integrity management plan.

[0005] In order to solve the above technical problems, in a first aspect, the present invention provides a method for generating a pipeline integrity management solution, the method comprising:

[0006] The pipeline knowledge extraction model is trained by using a preset corpus sample and a preset algorithm, wherein the preset corpus sample is obtained based on manual annotation of part of the pipeline data, and the preset algorithm is a machine learning algorithm selected based on the pipeline data type;

[0007] Performing knowledge extraction on pipeline data according to the pipeline knowledge extraction model to obtain a target knowledge graph, wherein the pipeline data includes pipeline parameter information, work target information, and implementation detail information;

[0008] Data is added to a preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management solution, wherein the preset template is determined based on the type of the pipeline integrity management solution.

[0009] Optionally, before training the pipeline knowledge extraction model using a preset corpus sample and a preset algorithm, the method further includes:

[0010] Obtain manually annotated historical pipeline data as the preset corpus sample, wherein the manually annotated historical pipeline data is obtained by annotating each knowledge point in the pipeline data in the form of a triple, and the triple contains the knowledge points and the association relationship between the knowledge points.

[0011] Optionally, the preset algorithms include: Naive Bayes, Support Vector Machine SVM, Word2vec, Recurrent Neural Network RNN, and Long Short-Term Memory Network LSTM;

[0012] The pipeline knowledge extraction model is trained by using a preset corpus sample and a preset algorithm, including:

[0013] The pipeline knowledge extraction model is trained by using a preset corpus sample and a preset algorithm to obtain multiple candidate models, wherein each candidate model corresponds to a set of model parameters;

[0014] Evaluate the plurality of candidate models using manually annotated historical pipeline data to obtain an evaluation result corresponding to each candidate model;

[0015] A candidate model with the highest recall rate and accuracy is selected from the multiple evaluation results as the pipeline knowledge extraction model.

[0016] Optionally, after extracting knowledge from pipeline data according to the pipeline knowledge extraction model to obtain a target knowledge graph, the method further includes:

[0017] Output the target knowledge graph, and update the knowledge graph according to the user modification instruction, wherein the user modification instruction is instruction information obtained based on the user operation after the knowledge graph is output to the user, and is used to modify the knowledge points and / or knowledge point relationships in the target knowledge graph.

[0018] Optionally, adding data to a preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management solution includes:

[0019] Data is added to the preset template according to the updated target knowledge graph to obtain the corresponding pipeline integrity management solution.

[0020] Optionally, after adding data to a preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management solution, the method further includes:

[0021] Outputting the pipeline integrity management plan, and receiving a modification command for modifying the pipeline integrity management plan, wherein the modification command is a command issued by a user for modifying the pipeline integrity management plan;

[0022] The pipeline integrity management plan is modified according to the modification command.

[0023] Optionally, the pipeline parameter information includes pipeline name, pipeline type, pipeline attribute parameters and pipeline transmission medium type;

[0024] The work target information includes the length of the data collection pipeline, the risk assessment completion rate, the integrity assessment completion rate, the maintenance completion rate, and the number of times the performance assessment and audit are completed;

[0025] The implementation details include pipeline data collection data, pipeline hazard identification data, pipeline risk assessment data, inspection and evaluation data, maintenance data, performance evaluation data, failure event management data, and subsequent work plan data;

[0026] The step of adding data to a preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management solution includes:

[0027] Acquire first graph data corresponding to the pipeline parameter information from the target knowledge graph, and add data to the pipeline overview section in the preset template according to the first graph data to obtain the pipeline overview and applicable scope;

[0028] Acquire second graph data corresponding to the work target information from the target knowledge graph, and add data to the work target section in the preset template according to the second graph data to obtain the work target;

[0029] Acquire third graph data corresponding to the implementation details information from the target knowledge graph, and add data to the implementation plan section in the preset template according to the third graph data to obtain an implementation plan;

[0030] The pipeline integrity management plan is generated according to the pipeline overview and scope of application, work objectives and implementation plan.

[0031] In a second aspect, an embodiment of the present invention further provides a device for generating a pipeline integrity management solution, including:

[0032] A training unit, used to train the pipeline knowledge extraction model through a preset corpus sample and a preset algorithm, wherein the preset corpus sample is obtained based on manual annotation of part of the pipeline data, and the preset algorithm is a machine learning algorithm selected based on the pipeline data type;

[0033] An extraction unit, configured to extract knowledge from pipeline data according to the pipeline knowledge extraction model to obtain a target knowledge graph, wherein the pipeline data includes pipeline parameter information, work target information, and implementation detail information;

[0034] An adding unit is used to add data to a preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management solution, wherein the preset template is determined based on the type of the pipeline integrity management solution.

[0035] Optionally, the device further comprises:

[0036] An acquisition unit is used to acquire manually annotated historical pipeline data as the preset corpus sample, wherein the manually annotated historical pipeline data is obtained by annotating each knowledge point in the pipeline data in the form of a triple, and the triple contains the knowledge points and the association relationship between the knowledge points.

[0037] Optionally, the preset algorithms include: Naive Bayes, Support Vector Machine SVM, Word2vec, Recurrent Neural Network RNN, and Long Short-Term Memory Network LSTM;

[0038] The training unit comprises:

[0039] A training module is used to train the pipeline knowledge extraction model through a preset corpus sample and a preset algorithm to obtain multiple candidate models, wherein each candidate model corresponds to a set of model parameters;

[0040] An evaluation module, used to evaluate the plurality of candidate models using the manually annotated historical pipeline data, and obtain an evaluation result corresponding to each candidate model;

[0041] A selection module is used to select a candidate model with the highest recall rate and accuracy from the multiple evaluation results as the pipeline knowledge extraction model.

[0042] Optionally, the device further comprises:

[0043] The first interaction unit is used to output the target knowledge graph and update the knowledge graph according to the user modification instruction, wherein the user modification instruction is the instruction information obtained based on the user operation after the knowledge graph is output to the user and is used to modify the knowledge points and / or knowledge point relationships in the target knowledge graph.

[0044] Optionally, the adding unit is specifically used to add data to the preset template according to the updated target knowledge graph to obtain a corresponding pipeline integrity management solution.

[0045] Optionally, the device further comprises:

[0046] A second interaction unit, configured to output the pipeline integrity management plan and receive a modification command for modifying the pipeline integrity management plan, wherein the modification command is a command issued by a user for modifying the pipeline integrity management plan;

[0047] A modification unit is used to modify the pipeline integrity management plan according to the modification command.

[0048] In order to achieve the above-mentioned purpose, according to a third aspect of the present invention, a storage medium is provided, wherein the storage medium includes a stored program, wherein when the program is run, the device where the storage medium is located is controlled to execute the above-mentioned method for generating a pipeline integrity management solution.

[0049] In order to achieve the above objective, according to a fourth aspect of the present invention, a processor is provided, wherein the processor is used to run a program, wherein the program executes the above-mentioned method for generating a pipeline integrity management solution when running.

[0050] Optionally, the pipeline parameter information includes pipeline name, pipeline type, pipeline attribute parameters and pipeline transmission medium type;

[0051] The work target information includes the length of the data collection pipeline, the risk assessment completion rate, the integrity assessment completion rate, the maintenance completion rate, and the number of times the performance assessment and audit are completed;

[0052] The implementation details include pipeline data collection data, pipeline hazard identification data, pipeline risk assessment data, inspection and evaluation data, maintenance data, performance evaluation data, failure event management data, and subsequent work plan data;

[0053] The adding unit comprises:

[0054] A first adding module is used to obtain first graph data corresponding to the pipeline parameter information from the target knowledge graph, and add data to the pipeline overview section in the preset template according to the first graph data to obtain the pipeline overview and applicable scope;

[0055] A second adding module is used to obtain second graph data corresponding to the work target information from the target knowledge graph, and add data to the work target section in the preset template according to the second graph data to obtain the work target;

[0056] A third adding module is used to obtain third graph data corresponding to the implementation details information from the target knowledge graph, and add data to the implementation plan section in the preset template according to the third graph data to obtain an implementation plan;

[0057] A generation module is used to generate the pipeline integrity management plan according to the pipeline overview and scope of application, work objectives and implementation plan.

[0058] By means of the above technical scheme, the method and device for generating a pipeline integrity management scheme provided by the present invention solves the problem that a large amount of manpower is consumed in the generation process of the existing pipeline integrity management scheme. The present invention trains a pipeline knowledge extraction model through a preset corpus sample and a preset algorithm, and then extracts knowledge from the pipeline data according to the pipeline knowledge extraction model to obtain a target knowledge graph, and finally adds data to the preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management scheme, wherein the preset template is determined based on the type of the pipeline integrity management scheme, thereby realizing an automated generation function of the pipeline integrity management scheme. In the above scheme, the preset corpus sample is obtained based on manual annotation of part of the pipeline data, and the preset algorithm is a machine learning algorithm selected based on the pipeline data type, so as to ensure that the knowledge extraction model obtained by training can simulate the function of manual knowledge extraction, so that the obtained target knowledge graph has a higher accuracy, and then ensures that the generation result of the pipeline integrity management scheme generated subsequently is more accurate. At the same time, the present invention saves pipeline data based on the knowledge graph, and adds data to the preset template according to the target knowledge graph, thereby realizing an automated solution generation effect without manual data search and filling, thereby solving the problem of large manpower consumption in the process of generating pipeline integrity management solutions in the prior art.

[0059] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0061] Figure 1 A flow chart of a method for generating a pipeline integrity management solution provided by an embodiment of the present invention is shown;

[0062] Figure 2 A flow chart of a method for generating another pipeline integrity management solution provided by an embodiment of the present invention is shown;

[0063] Figure 3 A block diagram showing a composition of a device for generating a pipeline integrity management solution provided by an embodiment of the present invention;

[0064] Figure 4 A block diagram showing the composition of another device for generating a pipeline integrity management solution provided by an embodiment of the present invention is shown;

[0065] Figure 5 A block diagram showing a composition of a device for generating a pipeline integrity management solution provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to enable the scope of the present invention to be fully communicated to those skilled in the art.

[0067] In order to solve the problem of high manpower consumption in the current process of generating pipeline integrity management solutions, an embodiment of the present invention provides a method for generating a pipeline integrity management solution, such as Figure 1 As shown, the method includes:

[0068] 101. Train the pipeline knowledge extraction model through preset corpus samples and preset algorithms.

[0069] Among them, the preset corpus sample is obtained based on manual annotation of part of the pipeline data, and the preset algorithm is a machine learning algorithm selected based on the pipeline data type.

[0070] In this embodiment, the preset corpus sample can be obtained by manually annotating knowledge documents, wherein the knowledge document can be understood as a document containing various knowledge generated in the process of pipeline integrity management, which can include a variety of different documents in Word, PPT, PDF, Excel, jpg, PNG and other formats. Since the knowledge in these documents is unstructured knowledge, the main purpose of manual marking is to determine the relationship between each knowledge point and other knowledge points. The specific marking method can be to use a marking tool to mark the knowledge in the above document according to the needs. For example, when marking the knowledge related to the pipeline, the name, length, material, commissioning date and other knowledge points of the pipeline can be marked in the document, and the relationship between these knowledge points can be marked at the same time. There is no specific limitation on the marking method here, and it can be selected according to the knowledge extraction model that needs to be trained later.

[0071] Meanwhile, the preset algorithm described in this embodiment can be selected from common classification algorithms in the field of artificial intelligence, such as the Naive Bayes algorithm. Of course, in specific use, it can be selected based on actual needs.

[0072] 102. Perform knowledge extraction on pipeline data according to the pipeline knowledge extraction model to obtain the target knowledge graph.

[0073] Among them, the pipeline data includes pipeline parameter information, work target information and implementation detail information. Based on the description of the aforementioned steps, it can be known that the pipeline knowledge extraction model trained in the aforementioned steps of this embodiment is actually the pipeline data after the aforementioned partial annotation. Therefore, in this step, the pipeline data as a whole can be subjected to knowledge extraction through the trained pipeline knowledge extraction surface model, thereby obtaining the corresponding target knowledge graph. Among them, the form of the target knowledge graph in this embodiment is consistent with the manual annotation method in the aforementioned steps. In this way, by performing knowledge extraction through the pipeline knowledge extraction model, a target knowledge graph in the form of a knowledge graph can be extracted from the originally complex and unstructured pipeline data. Since the form of the knowledge graph is direct and clear, by taking nodes as each knowledge and using the lines between nodes to represent the relationship between knowledge points, it can be ensured that in the subsequent use process, knowledge acquisition is clearer and simpler, and it is also convenient for subsequent modification, updating and maintenance.

[0074] 103. Add data to the preset template according to the target knowledge graph to obtain the corresponding pipeline integrity management plan.

[0075] Wherein, the preset template is determined based on the type of the pipeline integrity management solution.

[0076] Since the pipeline integrity management scheme actually includes a variety of different types of schemes, in this step, in the process of generating the pipeline integrity management scheme based on the target knowledge graph, it is also necessary to select the corresponding preset template for execution. Here, the preset template can be selected based on the instructions issued by the user. In this way, when generating the pipeline integrity management scheme, it can be ensured that the generated pipeline integrity management scheme can meet the needs of the user.

[0077] The above embodiment provides a method for generating a pipeline integrity management solution. In order to solve the problem that a large amount of manpower is consumed in the generation process of the existing pipeline integrity management solution, the present invention trains a pipeline knowledge extraction model through a preset corpus sample and a preset algorithm, and then extracts knowledge from the pipeline data according to the pipeline knowledge extraction model to obtain a target knowledge graph, and finally adds data to the preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management solution, wherein the preset template is determined based on the type of the pipeline integrity management solution, thereby realizing an automated generation function of the pipeline integrity management solution. In the above solution, the preset corpus sample is obtained based on manual annotation of part of the pipeline data, and the preset algorithm is a machine learning algorithm selected based on the pipeline data type, so as to ensure that the knowledge extraction model obtained by training can simulate the function of manual knowledge extraction, so that the obtained target knowledge graph has a higher accuracy, and then ensures that the generation result of the pipeline integrity management solution generated subsequently is more accurate. At the same time, the present invention saves pipeline data based on the knowledge graph, and adds data to the preset template according to the target knowledge graph, thereby realizing an automated solution generation effect without manual data search and filling, thereby solving the problem of large manpower consumption in the process of generating pipeline integrity management solutions in the prior art.

[0078] Furthermore, as a Figure 1 The embodiment shown in the figure is refined and expanded. The embodiment of the present invention also provides another method for generating a pipeline integrity management solution, such as Figure 2 As shown, the specific steps include:

[0079] 201. Obtain manually annotated historical pipeline data as a preset corpus sample.

[0080] The manually annotated historical pipeline data is obtained by annotating each knowledge point in the pipeline data in a triplet manner, and the triplet includes the knowledge points and the association relationship between the knowledge points.

[0081] In this embodiment, in order for the model obtained by subsequent training to directly extract knowledge from pipeline data in the form of knowledge graph, the manually annotated historical pipeline data obtained in this step can be annotated in the form of triples. The triples contain two knowledge points and a line representing the association between the knowledge points. In this way, after the annotated historical pipeline data is obtained in this step, since the data is standardized in the form required by the knowledge graph, it can be directly used as a sample for the subsequent training model, that is, the preset corpus sample.

[0082] 202. Train the pipeline knowledge extraction model through preset corpus samples and preset algorithms.

[0083] The preset corpus sample is obtained based on manual annotation of some pipeline data, and the preset algorithm is a machine learning algorithm selected based on the pipeline data type. The preset algorithm may include: Naive Bayes, Support Vector Machine SVM, Word2vec, Recurrent Neural Network RNN, and Long Short-Term Memory Network LSTM.

[0084] The support vector machine algorithm (SVM) is a generalized linear classifier that performs binary classification on data in a supervised learning manner. Its decision boundary is the maximum margin hyperplane solved for the learning samples.

[0085] The Naive Bayes Classifier (NBC) algorithm originated from classical mathematical theory, has a solid mathematical foundation, and stable classification efficiency. At the same time, the algorithm requires few parameters to be estimated, is not very sensitive to missing data, and the algorithm is relatively simple.

[0086] Recurrent Neural Network (RNN) is a type of recursive neural network algorithm that takes sequence data as input, recursively in the direction of sequence evolution, and all nodes are connected in a chain.

[0087] The Long Short-Term Memory (LSTM) network algorithm is a time recurrent neural network that is specially designed to solve the long-term dependency problem of general recurrent neural networks.

[0088] Among the above algorithms, users can select any one of them to train the model based on actual needs, and there is no limitation here.

[0089] In addition, in order to further optimize the model, during the training process, the pipeline knowledge extraction model is trained by using a preset corpus sample and a preset algorithm in this step, which can be performed in the following manner, including:

[0090] Firstly, the pipeline knowledge extraction model is trained by using a preset corpus sample and a preset algorithm to obtain multiple candidate models, where each candidate model corresponds to a set of model parameters;

[0091] Then, the multiple candidate models are evaluated using the manually annotated historical pipeline data to obtain an evaluation result corresponding to each candidate model;

[0092] Finally, a candidate model with the highest recall rate and accuracy is selected from the multiple evaluation results as the pipeline knowledge extraction model.

[0093] Since multiple candidate models are evaluated according to actual manually annotated historical pipeline data, that is, it is equivalent to comparing the prediction results of the candidate models with the actual results of the manual standards. Therefore, it is possible to ensure that when there are multiple model parameters, the candidate model corresponding to the model parameters that best match the actual results is determined as the knowledge extraction model, thereby ensuring the accuracy of the model.

[0094] 203. Perform knowledge extraction on the pipeline data according to the pipeline knowledge extraction model to obtain a target knowledge graph.

[0095] Wherein, the pipeline data includes pipeline parameter information, work target information and implementation detail information. Since the target knowledge graph is automatically extracted based on the machine model, that is, extracted using the pipeline knowledge extraction model, in order to further improve the accuracy of the target knowledge graph, after the target knowledge graph is obtained in this step, since each knowledge point and the relationship between knowledge points have been marked by the model, this embodiment can also manually review the target knowledge graph to ensure the data quality of the target knowledge graph. Therefore, if there is a need to output and correct the knowledge graph, step 204 can be executed. Conversely, if there is no need to modify the knowledge graph, step 205 can be executed directly.

[0096] 204. Output the target knowledge graph, and update the knowledge graph according to the user's modification instructions.

[0097] Among them, the user correction instruction is instruction information obtained based on user operations after the knowledge graph is output to the user, which is used to modify the knowledge points and / or knowledge point relationships in the target knowledge graph.

[0098] In this step, the user can determine whether there are errors and missing items based on the output target knowledge graph, so as to make manual corrections and supplements. Specifically, the contradictions and ambiguities in the knowledge graph can be corrected based on the user's correction instructions. At the same time, in order to further improve the accuracy of the model, the knowledge points to be modified and the relationship between the knowledge points can also be used as sample corpus for training the model in the aforementioned step 202, so as to optimize the model training in the subsequent process and continuously improve the knowledge point extraction accuracy and generalization ability of the model.

[0099] In addition, in this embodiment, the target knowledge graph and the updated knowledge graph can also be saved so that relevant personnel can subsequently troubleshoot problems with the model.

[0100] 205. Add data to the preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management solution.

[0101] The preset template is determined based on the type of the pipeline integrity management solution, and includes a pipeline overview section, a work target section, and an implementation plan section.

[0102] In practical applications, the pipeline parameter information may include information such as pipeline name, pipeline type, pipeline attribute parameters, and pipeline transmission medium type; at the same time, the work target information may also specifically include: information such as the length of the pipeline for data collection, risk assessment completion rate, integrity assessment completion rate, maintenance completion rate, and performance evaluation and audit completion times; the implementation detail information may also include: such as pipeline data collection data, pipeline hazard identification data, pipeline risk assessment data, detection and evaluation data, maintenance data, performance evaluation data, failure event management data, and subsequent work plan data.

[0103] The above data actually corresponds to each section in the subsequently generated pipeline integrity management plan. That is to say, when adding data to the preset template, the corresponding pipeline data can be obtained from the knowledge graph and added to each section in the preset template. Therefore, in this step, in the process of adding data to the preset template according to the target knowledge graph to obtain the corresponding pipeline integrity management plan, its implementation method can be carried out in the following manner:

[0104] In the first aspect, when adding data to the pipeline overview section: obtain the first graph data corresponding to the pipeline parameter information from the target knowledge graph, and add data to the pipeline overview section in the preset template according to the first graph data to obtain the pipeline overview and applicable scope.

[0105] On the second aspect, when adding data to the work target section: obtain the second graph data corresponding to the work target information from the target knowledge graph, and add data to the work target section in the preset template according to the second graph data to obtain the work target.

[0106] In the third aspect, when adding data to the implementation plan section: obtaining the third graph data corresponding to the implementation details information from the target knowledge graph, and adding data to the implementation plan section in the preset template according to the third graph data to obtain the implementation plan. ;

[0107] Finally, after executing the above three methods, a pipeline integrity management plan can be generated based on the pipeline overview and scope of application obtained in the first aspect, the work objectives obtained in the second aspect, and the implementation plan obtained in the third aspect.

[0108] In this embodiment, when the target knowledge graph is updated in the above steps, this step may be specifically as follows: adding data to the preset template according to the updated target knowledge graph to obtain the corresponding pipeline integrity management solution. Since the target knowledge graph is updated, the accuracy of the obtained pipeline integrity management solution can be ensured by adding data through the updated target knowledge graph.

[0109] 206. Output the pipeline integrity management solution, and receive a modification command for modifying the pipeline integrity management solution.

[0110] The modification command is a command issued by a user to modify the pipeline integrity management plan.

[0111] After the pipeline integrity management plan is generated, in order to further improve the accuracy of the obtained plan, the pipeline integrity management plan can also be output to the user based on the method of this step, and after the user reviews it, it is detected whether there is a modification command. When the modification command is detected, it is modified based on the modification command, so that the pipeline integrity management plan automatically generated by the machine can be manually corrected when the user needs it.

[0112] 207. Modify the pipeline integrity management plan according to the modification command.

[0113] Since the modification command includes the position and content that need to be modified, the corresponding position and content of the pipeline integrity management plan can be modified based on this, thereby forming a final pipeline integrity management plan.

[0114] Furthermore, as a response to the above Figure 1 In order to realize the method shown in the figure, the embodiment of the present invention also provides a device for generating a pipeline integrity management solution for the above Figure 1 The device embodiment corresponds to the aforementioned method embodiment. For ease of reading, this device embodiment will not repeat the details of the aforementioned method embodiment one by one, but it should be clear that the device in this embodiment can correspond to all the contents of the aforementioned method embodiment. Figure 3 As shown, the device includes: a training unit 31, an extraction unit 32 and an adding unit 33, wherein

[0115] The training unit 31 can be used to train the pipeline knowledge extraction model through a preset corpus sample and a preset algorithm, wherein the preset corpus sample is obtained based on manual annotation of part of the pipeline data, and the preset algorithm is a machine learning algorithm selected based on the pipeline data type;

[0116] The extraction unit 32 may be used to extract knowledge from the pipeline data according to the pipeline knowledge extraction model trained by the training unit 31 to obtain a target knowledge graph, wherein the pipeline data includes pipeline parameter information, work target information, and implementation detail information;

[0117] The adding unit 33 can be used to add data to the preset template according to the target knowledge graph extracted by the extracting unit 32 to obtain the corresponding pipeline integrity management plan, wherein the preset template is determined based on the type of the pipeline integrity management plan, and the preset template includes a pipeline overview section, a work target section, and an implementation plan section.

[0118] Furthermore, as a response to the above Figure 2 In order to realize the method shown in the figure, the embodiment of the present invention also provides a device for generating a pipeline integrity management solution for the above Figure 2 The device embodiment corresponds to the aforementioned method embodiment. For ease of reading, this device embodiment will not repeat the details of the aforementioned method embodiment one by one, but it should be clear that the device in this embodiment can correspond to all the contents of the aforementioned method embodiment. Figure 4 As shown, the device includes: a training unit 41, an extraction unit 42 and an adding unit 43, wherein

[0119] The training unit 41 may be used to train the pipeline knowledge extraction model through a preset corpus sample and a preset algorithm, wherein the preset corpus sample is obtained based on manual annotation of part of the pipeline data, and the preset algorithm is a machine learning algorithm selected based on the pipeline data type;

[0120] The extraction unit 42 may be used to extract knowledge from the pipeline data according to the pipeline knowledge extraction model trained by the training unit 41 to obtain a target knowledge graph, wherein the pipeline data includes pipeline parameter information, work target information, and implementation detail information;

[0121] The adding unit 43 can be used to add data to the preset template according to the target knowledge graph extracted by the extracting unit 42 to obtain the corresponding pipeline integrity management plan, wherein the preset template is determined based on the type of the pipeline integrity management plan, and the preset template includes a pipeline overview section, a work target section, and an implementation plan section.

[0122] Furthermore, the device also includes:

[0123] The acquisition unit 44 can be used to acquire historical pipeline data after manual annotation as the preset corpus sample and send it to the training unit 41, wherein the historical pipeline data after manual annotation is obtained by annotating each knowledge point in the pipeline data in the form of a triple, and the triple contains the knowledge points and the association relationship between the knowledge points.

[0124] Furthermore, the preset algorithms include: Naive Bayes, Support Vector Machine (SVM), Word2vec, Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM);

[0125] The training unit 41 comprises:

[0126] The training module 411 can be used to train the pipeline knowledge extraction model through a preset corpus sample and a preset algorithm to obtain multiple candidate models, wherein each candidate model corresponds to a set of model parameters;

[0127] The evaluation module 412 can be used to evaluate the candidate models trained by the multiple training modules 411 using the manually annotated historical pipeline data to obtain an evaluation result corresponding to each candidate model;

[0128] The selection module 413 may be used to select a candidate model with the highest recall rate and accuracy rate from the evaluation results obtained by the multiple evaluation modules 412 as the pipeline knowledge extraction model.

[0129] Furthermore, the device also includes:

[0130] The first interaction unit 45 can be used to output the target knowledge graph extracted by the extraction unit 42, and update the knowledge graph according to the user modification instruction, wherein the user modification instruction is instruction information obtained based on the user operation after the knowledge graph is output to the user for modifying the knowledge points and / or knowledge point relationships in the target knowledge graph.

[0131] Furthermore, the adding unit 43 can be specifically used to add data to the preset template according to the target knowledge graph updated according to the correction instruction received by the first interaction unit 45 to obtain a corresponding pipeline integrity management solution.

[0132] Furthermore, the device also includes:

[0133] The second interaction unit 46 may be used to output the pipeline integrity management plan obtained by the adding unit 43, and receive a modification command that may be used to modify the pipeline integrity management plan, wherein the modification command is a command issued by a user to modify the pipeline integrity management plan;

[0134] The modification unit 47 may be configured to modify the pipeline integrity management solution according to the modification command received by the second interaction unit 46 .

[0135] Furthermore, the pipeline parameter information includes pipeline name, pipeline type, pipeline attribute parameters and pipeline transmission medium type;

[0136] The work target information includes the length of the data collection pipeline, the risk assessment completion rate, the integrity assessment completion rate, the maintenance completion rate, and the number of times the performance assessment and audit are completed;

[0137] The implementation details include pipeline data collection data, pipeline hazard identification data, pipeline risk assessment data, inspection and evaluation data, maintenance data, performance evaluation data, failure event management data, and subsequent work plan data;

[0138] The adding unit 43 comprises:

[0139] The first adding module 431 can be used to obtain the first graph data corresponding to the pipeline parameter information from the target knowledge graph, and add data to the pipeline overview section in the preset template according to the first graph data to obtain the pipeline overview and applicable scope;

[0140] The second adding module 432 may be used to obtain second graph data corresponding to the work target information from the target knowledge graph, and add data to the work target section in the preset template according to the second graph data to obtain the work target;

[0141] The third adding module 433 may be used to obtain third graph data corresponding to the implementation details information from the target knowledge graph, and add data to the implementation plan section in the preset template according to the third graph data to obtain an implementation plan;

[0142] The generation module 434 can be used to generate the pipeline integrity management plan according to the pipeline overview and applicable scope obtained by the first adding module 431, the work objectives obtained by the second adding module 432, and the implementation plan obtained by the third adding module 433.

[0143] By means of the above technical scheme, an embodiment of the present invention provides a method and device for generating a pipeline integrity management scheme. In order to solve the problem that a large amount of manpower is consumed in the generation process of the existing pipeline integrity management scheme, the present invention trains a pipeline knowledge extraction model through a preset corpus sample and a preset algorithm, and then extracts knowledge from the pipeline data according to the pipeline knowledge extraction model to obtain a target knowledge graph, and finally adds data to the preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management scheme, wherein the preset template is determined based on the type of the pipeline integrity management scheme, thereby realizing an automated generation function of the pipeline integrity management scheme. In the above scheme, the preset corpus sample is obtained based on manual annotation of part of the pipeline data, and the preset algorithm is a machine learning algorithm selected based on the pipeline data type, so as to ensure that the knowledge extraction model obtained by training can simulate the function of manual knowledge extraction, so that the obtained target knowledge graph has a higher accuracy, and then ensures that the generation result of the pipeline integrity management scheme generated subsequently is more accurate. At the same time, the present invention saves pipeline data based on the knowledge graph, and adds data to the preset template according to the target knowledge graph, thereby realizing an automated solution generation effect without manual data search and filling, thereby solving the problem of large manpower consumption in the process of generating pipeline integrity management solutions in the prior art.

[0144] The processor includes a kernel, and the kernel retrieves the corresponding program unit from the memory. One or more kernels can be set, and a method for automatically generating a pipeline integrity management solution is implemented by adjusting kernel parameters to solve the problem of manpower consumption caused by manual generation.

[0145] An embodiment of the present invention provides a storage medium on which a program is stored. When the program is executed by a processor, the method for generating the pipeline integrity management solution is implemented.

[0146] An embodiment of the present invention provides a processor, which is used to run a program, wherein the method for generating the pipeline integrity management solution is executed when the program is running.

[0147] The embodiment of the present invention provides a device 50, such as Figure 5 As shown, the device includes at least one processor 501, and at least one memory 502 and a bus 503 connected to the processor; wherein the processor 501 and the memory 502 communicate with each other through the bus 503; the processor 501 is used to call the program instructions in the memory to execute the above-mentioned method for generating the pipeline integrity management solution.

[0148] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0149] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program that initializes the following method steps: training a pipeline knowledge extraction model through a preset corpus sample and a preset algorithm, wherein the preset corpus sample is obtained based on manual annotation of part of the pipeline data, and the preset algorithm is a machine learning algorithm selected based on the pipeline data type; performing knowledge extraction on the pipeline data according to the pipeline knowledge extraction model to obtain a target knowledge graph; adding data to a preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management plan, wherein the preset template is determined based on the type of the pipeline integrity management plan.

[0150] Furthermore, before training the pipeline knowledge extraction model using a preset corpus sample and a preset algorithm, the method further includes:

[0151] Obtain manually annotated historical pipeline data as the preset corpus sample, wherein the manually annotated historical pipeline data is obtained by annotating each knowledge point in the pipeline data in the form of a triple, and the triple contains the knowledge points and the association relationship between the knowledge points.

[0152] Furthermore, the preset algorithms include: Naive Bayes, Support Vector Machine (SVM), Word2vec, Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM);

[0153] The pipeline knowledge extraction model is trained by using a preset corpus sample and a preset algorithm, including:

[0154] The pipeline knowledge extraction model is trained by using a preset corpus sample and a preset algorithm to obtain multiple candidate models, wherein each candidate model corresponds to a set of model parameters;

[0155] Evaluate the plurality of candidate models using manually annotated historical pipeline data to obtain an evaluation result corresponding to each candidate model;

[0156] A candidate model with the highest recall rate and accuracy is selected from the multiple evaluation results as the pipeline knowledge extraction model.

[0157] Furthermore, after extracting knowledge from pipeline data according to the pipeline knowledge extraction model to obtain a target knowledge graph, the method further includes:

[0158] Output the target knowledge graph, and update the knowledge graph according to the user modification instruction, wherein the user modification instruction is instruction information obtained based on the user operation after the knowledge graph is output to the user, and is used to modify the knowledge points and / or knowledge point relationships in the target knowledge graph.

[0159] Furthermore, the adding of data to the preset template according to the target knowledge graph to obtain the corresponding pipeline integrity management solution includes:

[0160] Data is added to the preset template according to the updated target knowledge graph to obtain the corresponding pipeline integrity management solution.

[0161] Furthermore, after adding data to the preset template according to the target knowledge graph to obtain the corresponding pipeline integrity management solution, the method further includes:

[0162] Outputting the pipeline integrity management plan, and receiving a modification command for modifying the pipeline integrity management plan, wherein the modification command is a command issued by a user for modifying the pipeline integrity management plan;

[0163] The pipeline integrity management plan is modified according to the modification command.

[0164] Furthermore, the pipeline parameter information includes pipeline name, pipeline type, pipeline attribute parameters and pipeline transmission medium type;

[0165] The work target information includes the length of the data collection pipeline, the risk assessment completion rate, the integrity assessment completion rate, the maintenance completion rate, and the number of times the performance assessment and audit are completed;

[0166] The implementation details include pipeline data collection data, pipeline hazard identification data, pipeline risk assessment data, inspection and evaluation data, maintenance data, performance evaluation data, failure event management data, and subsequent work plan data;

[0167] The step of adding data to a preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management solution includes:

[0168] Acquire first graph data corresponding to the pipeline parameter information from the target knowledge graph, and add data to the pipeline overview section in the preset template according to the first graph data to obtain the pipeline overview and applicable scope;

[0169] Acquire second graph data corresponding to the work target information from the target knowledge graph, and add data to the work target section in the preset template according to the second graph data to obtain the work target;

[0170] Acquire third graph data corresponding to the implementation details information from the target knowledge graph, and add data to the implementation plan section in the preset template according to the third graph data to obtain an implementation plan;

[0171] The pipeline integrity management plan is generated according to the pipeline overview and scope of application, work objectives and implementation plan.

[0172] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0173] In a typical configuration, the device includes one or more processors (CPU), memory and bus. The device may also include input / output interface, network interface and the like.

[0174] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip. The memory is an example of a computer-readable medium.

[0175] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0176] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0177] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0178] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for generating a pipeline integrity management plan, characterized in that: include: The pipeline knowledge extraction model is trained by using a preset corpus sample and a preset algorithm, wherein the preset corpus sample is obtained based on manual annotation of part of the pipeline data, and the preset algorithm is a machine learning algorithm selected based on the pipeline data type; Performing knowledge extraction on pipeline data according to the pipeline knowledge extraction model to obtain a target knowledge graph, wherein the pipeline data includes pipeline parameter information, work target information, and implementation detail information; Add data to a preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management solution, wherein the preset template is determined based on the type of the pipeline integrity management solution, and the preset template includes a pipeline overview section, a work target section, and an implementation plan section; Before training the pipeline knowledge extraction model using a preset corpus sample and a preset algorithm, the method further includes: Obtaining manually annotated historical pipeline data as the preset corpus sample, wherein the manually annotated historical pipeline data is obtained by annotating each knowledge point in the pipeline data in a triplet manner, wherein the triplet includes the knowledge points and the association relationship between the knowledge points; The pipeline knowledge extraction model is trained by using a preset corpus sample and a preset algorithm, including: The pipeline knowledge extraction model is trained by using a preset corpus sample and a preset algorithm to obtain multiple candidate models, wherein each candidate model corresponds to a set of model parameters; Evaluate the plurality of candidate models using manually annotated historical pipeline data to obtain an evaluation result corresponding to each candidate model; Selecting a candidate model with the highest recall rate and accuracy from the multiple evaluation results as the pipeline knowledge extraction model; After extracting knowledge from pipeline data according to the pipeline knowledge extraction model to obtain a target knowledge graph, the method further includes: Output the target knowledge graph, and update the knowledge graph according to the user modification instruction, wherein the user modification instruction is instruction information obtained based on the user operation after the knowledge graph is output to the user, and is used to modify the knowledge points and / or knowledge point relationships in the target knowledge graph.

2. The method according to claim 1, characterized in that The adding of data to the preset template according to the target knowledge graph to obtain the corresponding pipeline integrity management solution includes: Data is added to the preset template according to the updated target knowledge graph to obtain the corresponding pipeline integrity management solution.

3. The method according to claim 2, characterized in that After adding data to the preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management solution, the method further includes: Outputting the pipeline integrity management plan, and receiving a modification command for modifying the pipeline integrity management plan, wherein the modification command is a command issued by a user for modifying the pipeline integrity management plan; The pipeline integrity management plan is modified according to the modification command.

4. The method according to any one of claims 1 to 3, characterized in that The pipeline parameter information includes pipeline name, pipeline type, pipeline attribute parameters and pipeline transmission medium type; The work target information includes the length of the data collection pipeline, the risk assessment completion rate, the integrity assessment completion rate, the maintenance completion rate, and the number of times the performance assessment and audit are completed; The implementation details information includes pipeline data collection data, pipeline hazard identification data, pipeline risk assessment data, detection and evaluation data, maintenance data, performance evaluation data, failure event management data and subsequent work plan data.

5. The method according to claim 4, characterized in that The step of adding data to a preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management solution includes: Acquire first graph data corresponding to the pipeline parameter information from the target knowledge graph, and add data to the pipeline overview section in the preset template according to the first graph data to obtain the pipeline overview and applicable scope; Acquire second graph data corresponding to the work target information from the target knowledge graph, and add data to the work target section in the preset template according to the second graph data to obtain the work target; Acquire third graph data corresponding to the implementation details information from the target knowledge graph, and add data to the implementation plan section in the preset template according to the third graph data to obtain an implementation plan; The pipeline integrity management plan is generated according to the pipeline overview and scope of application, work objectives and implementation plan.

6. A device for generating a pipeline integrity management solution, characterized in that: A method for generating a pipeline integrity management solution as claimed in any one of claims 1 to 5, comprising: A training unit, used to train the pipeline knowledge extraction model through a preset corpus sample and a preset algorithm, wherein the preset corpus sample is obtained based on manual annotation of part of the pipeline data, and the preset algorithm is a machine learning algorithm selected based on the pipeline data type; An extraction unit, configured to extract knowledge from pipeline data according to the pipeline knowledge extraction model to obtain a target knowledge graph, wherein the pipeline data includes pipeline parameter information, work target information, and implementation detail information; An adding unit is used to add data to a preset template according to the target knowledge graph to obtain a corresponding pipeline integrity management plan, wherein the preset template is determined based on the type of the pipeline integrity management plan, and the preset template includes a pipeline overview section, a work target section, and an implementation plan section.

7. A storage medium, characterized in that: The storage medium includes a stored program, wherein when the program is executed, the device where the storage medium is located is controlled to execute the method for generating a pipeline integrity management solution according to any one of claims 1 to 5.

8. A device, characterized in that The device includes at least one processor, and at least one memory and bus connected to the processor; wherein the processor and memory communicate with each other through the bus; the processor is used to call program instructions in the memory to execute the method for generating a pipeline integrity management solution as described in any one of claims 1 to 5.

Images