Pipeline network risk early warning method, device, computer equipment and storage medium

Abstract

The present application relates to the field of artificial intelligence technology, and discloses a pipeline risk warning method, device, computer equipment and storage medium, the method comprising obtaining detection data; extracting text from the detection data to obtain corresponding pipeline information and defect information; constructing a knowledge graph based on the pipeline information and defect information; inputting the pipeline information and defect information corresponding to any two first nodes connected in the knowledge graph into a matching model to obtain corresponding similarities, and synchronizing the similarities into the knowledge graph; calculating the first similarity between any two second nodes in the knowledge graph based on the knowledge graph with similarities, classifying the pipelines corresponding to each second node in the knowledge graph based on the first similarity to obtain multiple pipeline categories; determining the pipeline category of the first pipeline, and issuing a risk warning for the second pipeline in the pipeline category to which the first pipeline belongs. The present application provides early warning for pipelines to avoid undetected actual defects in the pipelines.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular to a pipeline network risk early warning method, device, computer equipment and storage medium. Background Art

[0002] In the process of urban construction, the construction of drainage pipe networks is indispensable, and drainage pipe networks have always been the top priority of the drainage system. In real life, drainage pipe networks often have abnormal conditions, such as overflow, seepage and other defects in the pipe network, which will have adverse effects on the environment and the lives and even health of surrounding citizens; defects such as groundwater infiltration may increase the treatment load of sewage treatment plants and reduce treatment efficiency; defects such as pipe network siltation and foreign body penetration may affect the operation efficiency of the pipe network, thereby affecting the normal operation of the urban sewage system. The existing treatment of defects is often discovered through regular inspections, but the inspection cost is high, or often the handling personnel carry out emergency repairs after the defects have occurred. Therefore, how to provide early warning of the defects of the pipe network has become an urgent problem to be solved. Summary of the invention

[0003] The present application provides a pipeline network risk warning method, device, computer equipment and storage medium to solve the problem in the prior art of how to provide early warning of defects in the pipeline network.

[0004] To solve the above problems, this application provides a pipeline network risk early warning method, including:

[0005] Obtain test data;

[0006] Performing text extraction on the detection data to obtain pipe network information and defect information corresponding to the pipeline in the detection data;

[0007] Constructing a knowledge graph based on the pipe network information and defect information;

[0008] Inputting the pipe network information and defect information corresponding to any two first nodes connected in the knowledge graph into the matching model for matching, obtaining the corresponding similarity, and synchronizing the similarity into the knowledge graph;

[0009] According to the knowledge graph with similarity, a first similarity between any two second nodes in the knowledge graph is calculated, and according to the first similarity, the pipelines corresponding to the second nodes in the knowledge graph are classified to obtain a plurality of pipeline categories;

[0010] When it is received that a first pipeline among the pipelines has a first defect, the pipeline category to which the first pipeline belongs is determined, and a risk warning is issued for a second pipeline in the pipeline category to which the first pipeline belongs.

[0011] Further, the acquiring of detection data includes:

[0012] receiving processing instructions;

[0013] According to the processing instruction, a camera is called to shoot paper text to obtain a text image;

[0014] According to the text image, optical character recognition is used to perform scanning to obtain corresponding text data.

[0015] Furthermore, the text extraction of the detection data includes:

[0016] Using regular expressions, extracting the pipe network information in the detection data;

[0017] The defect information in the detection data is extracted using a defect extraction model, and the defect extraction model is trained based on an RCNN model.

[0018] Furthermore, the pipe network information and defect information corresponding to any two first nodes connected in the knowledge graph are input into the matching model for matching, and the corresponding similarity is obtained, including:

[0019] Extracting the pipe network information and defect information corresponding to the two first nodes connected in the knowledge graph;

[0020] According to the pipe network information and defect information corresponding to the two first nodes connected, a matching model is used to perform corresponding matching to obtain a first matching degree and a second matching degree, wherein the matching model is obtained by training based on a Bimpm model;

[0021] The first matching degree is multiplied by the second matching degree to obtain a corresponding similarity.

[0022] Furthermore, the calculating, based on the knowledge graph with similarity, a first similarity between any two second nodes in the knowledge graph, and classifying the pipelines corresponding to each second node in the knowledge graph according to the first similarity includes:

[0023] Calculate the first similarity between two second nodes in the knowledge graph according to the knowledge graph with similarity, and determine whether the first similarity is greater than or equal to a preset value;

[0024] If the first similarity between the two second nodes is greater than or equal to a preset value, the two second nodes whose first similarity is greater than or equal to the preset value are stored in the same pipeline category;

[0025] If the first similarity between two second nodes is less than a preset value, the two second nodes whose first similarity is less than the preset value are stored in different pipeline categories until the first similarity between any two nodes in the knowledge graph is judged with the preset value, thereby obtaining multiple pipeline categories, and the first similarity corresponding to any two second nodes in the same pipeline category is greater than or equal to the preset value.

[0026] Further, the calculating the first similarity between two second nodes in the knowledge graph according to the knowledge graph with similarity includes:

[0027] According to the knowledge graph, determining whether two of the second nodes are directly connected;

[0028] If the two second nodes are directly connected, obtaining the first similarity corresponding to the two second nodes;

[0029] If the two second nodes are not directly connected, the shortest connection path between the two second nodes is obtained, and based on multiple second similarities on the connection path, multiple second similarities are multiplied to obtain corresponding first similarities.

[0030] Further, when receiving a signal that a first pipeline among the pipelines has a first defect, determining a pipeline category to which the first pipeline belongs, and issuing a risk warning for a second pipeline in the pipeline category to which the first pipeline belongs includes:

[0031] determining a defect level of the first defect based on the first defect corresponding to the first pipeline;

[0032] Based on the first pipeline, determining the pipeline category to which the first pipeline belongs, and obtaining the similarity between the first pipeline and a second pipeline in the pipeline category to which the first pipeline belongs;

[0033] If the similarity between the first pipeline and the second pipeline is greater than or equal to a preset threshold, it is determined that the second pipeline and the first defect are of the same defect level and an early warning is issued;

[0034] If the similarity between the first pipeline and the second pipeline is less than a preset threshold, a preset integer is subtracted from the defect level corresponding to the first defect to obtain the defect level corresponding to the second pipeline and issue an early warning.

[0035] In order to solve the above problems, the present application also provides a pipeline network risk early warning device, including:

[0036] An acquisition module, used to acquire detection data;

[0037] An extraction module, used to perform text extraction on the detection data to obtain pipe network information and defect information corresponding to the pipeline in the detection data;

[0038] A construction module, used to construct a knowledge graph based on the pipe network information and defect information;

[0039] A similarity calculation module, used for inputting the pipe network information and defect information corresponding to any two first nodes connected in the knowledge graph into the matching model for matching, obtaining the corresponding similarity, and synchronizing the similarity into the knowledge graph;

[0040] a classification module, configured to calculate a first similarity between any two second nodes in the knowledge graph according to the knowledge graph with similarity, and classify the pipelines corresponding to the second nodes in the knowledge graph according to the first similarity to obtain a plurality of pipeline categories;

[0041] The early warning module is used to determine the pipeline category to which the first pipeline belongs when receiving a signal that the first pipeline among the pipelines has a first defect, and issue a risk early warning to a second pipeline in the pipeline category to which the first pipeline belongs.

[0042] In order to solve the above problem, the present application also provides a computer device, including:

[0043] at least one processor; and,

[0044] a memory communicatively connected to the at least one processor; wherein,

[0045] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the pipeline network risk early warning method as described above.

[0046] In order to solve the above problems, the present application also provides a non-volatile computer-readable storage medium, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor, the pipeline network risk warning method as described above is implemented.

[0047] A pipeline network risk warning method, device, computer equipment and storage medium provided in the embodiments of the present application have at least the following beneficial effects compared with the prior art:

[0048] By receiving the detection data and performing text extraction on the detection data, the data required for subsequent use, such as the pipe network information and defect information in the detection data, is extracted; the structured pipe network information and defect information are used to construct a knowledge graph; the pipe network information and defect information corresponding to any two first nodes connected in the knowledge graph are matched using a matching model to obtain corresponding similarities, and the similarities are synchronized to the knowledge graph; according to the size of the similarities, the first similarity between any two second nodes in the knowledge graph is calculated; according to the first similarity, the pipelines corresponding to the nodes are classified to obtain multiple pipeline categories; the matching model is used to classify the pipelines, so as to facilitate the subsequent acquisition of warning objects; when it is received that the first pipeline has a first defect, the pipeline category of the first pipeline is determined, and a risk warning is issued to the second pipeline in the pipeline category to which the first pipeline belongs, that is, all pipelines in the pipeline category except the first pipeline, and the pipeline is issued in advance by using the similarity between pipelines, so as to avoid the actual defects of the second pipeline being undiscovered. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the scheme in the present application, a brief introduction will be given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0050] Figure 1 A flow chart of a pipeline network risk warning method provided in one embodiment of the present application;

[0051] Figure 2 A schematic diagram of a module of a pipeline network risk warning device provided in an embodiment of the present application;

[0052] Figure 3 A schematic diagram of the structure of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0053] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0054] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly or implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0055] This application provides a pipeline network risk early warning method. Figure 1 As shown, Figure 1 A flow chart of a pipeline network risk warning method provided in one embodiment of the present application.

[0056] In this embodiment, the pipeline network risk early warning method includes:

[0057] S1. Obtaining test data;

[0058] Specifically, the detection data includes description information, historical circulation data, and historical defects of each pipeline in the pipeline network. The description information of the pipeline includes pipeline number, pipeline material, pipeline installation status, pipe length, and pipe diameter. The pipeline installation status is tilted installation, horizontal installation, or vertical installation, and also includes the angle of tilted installation. The historical circulation data includes liquid level, flow rate, and flow rate. The historical defect situation is the situation that the pipeline has been blocked or foreign objects have penetrated the pipeline in the past.

[0059] The above-mentioned pipe network detection data can be obtained from a database or input manually.

[0060] Further, the acquiring of detection data includes:

[0061] Send a call request to the preset knowledge base, the call request carries a signature verification token;

[0062] Receive the signature verification result returned by the knowledge base, and when the signature verification result is passed, call the detection data in the preset knowledge base, wherein the signature verification result is obtained by the knowledge base through RSA asymmetric encryption verification according to the signature verification token.

[0063] Since the inspection data of each pipeline under the pipeline network may involve urban safety issues, the inspection data will be saved in a preset database. Therefore, when obtaining the inspection data, the database will perform a signature verification step to ensure the security of the data and avoid data leakage.

[0064] By verifying the signature, we can ensure data security and avoid data leakage.

[0065] Further, the acquiring of detection data includes:

[0066] receiving processing instructions;

[0067] According to the processing instruction, calling the camera to shoot the paper text to obtain a text image;

[0068] According to the text image, optical character recognition is used to perform scanning to obtain corresponding text data.

[0069] Specifically, since most of the existing pipeline data are recorded in paper texts, optical character recognition is used to process offline data online. First, the paper text is photographed with a camera to obtain the corresponding text image. Based on the text image, optical character recognition is used to scan and obtain the corresponding text data.

[0070] Optical Character Recognition (OCR) refers to the process in which an electronic device (such as a scanner or a digital camera) examines characters printed on paper, determines their shapes by detecting dark and light patterns, and then translates the shapes into computer text using character recognition methods.

[0071] By using optical character recognition to convert paper inspection text into editable inspection data, the data is made online and the efficiency of subsequent processing is improved.

[0072] S2. Perform text extraction on the detection data to obtain pipe network information and defect information corresponding to the pipeline in the detection data;

[0073] Specifically, the detection data is processed by using regular expressions and defect extraction models, the pipeline network information and defect information in the detection data are extracted and stored in corresponding tables to obtain formatted data, which is convenient for subsequent construction of a knowledge graph.

[0074] Furthermore, the text extraction of the detection data includes:

[0075] Using regular expressions, extracting the pipe network information in the detection data;

[0076] The defect information in the detection data is extracted using a defect extraction model, and the defect extraction model is trained based on an RCNN model.

[0077] Specifically, since the pipe network information usually has a fixed format, for example, the specific pipe number in the pipe network information is often in the format of "pipe number: XX-XXX", regular expressions are used to extract the pipe network information, which includes the pipe number, pipe material, pipe installation status, pipe length, and pipe diameter.

[0078] In the paper reports corresponding to the inspection data, a series of descriptions of the corresponding defects of the pipeline are usually made manually. Therefore, the defect extraction model is used to summarize and extract the defects in the inspection data, that is, to classify the defects in the inspection data and determine the category to which they belong. The category to which they belong is the defect information. The same pipeline may correspond to one or more defect information. The defect information includes siltation, foreign matter inserted into the pipeline, seepage and groundwater infiltration, etc.

[0079] The RCNN model is constructed as follows: Bi-LSTM is used to obtain context information, similar to the language model; the hidden layer output obtained by Bi-LSTM is concatenated with the word vector; the concatenated vector is nonlinearly mapped to a low dimension; the value of each position in the vector is the maximum value of all time series to obtain the final feature vector, which is similar to max-pool; and finally, softmax classification is performed.

[0080] By using regular expressions and defect extraction models simultaneously to complete text extraction, the accuracy of text extraction is improved.

[0081] S3. Construct a knowledge graph based on the pipe network information and defect information;

[0082] Specifically, a knowledge graph is constructed based on the formatted pipe network information and defect information. The pipe numbers in the pipe network information are used as nodes and the defect information is used as edges. The pipe numbers corresponding to pipes with the same or similar defect information are connected to obtain the edges of the knowledge graph. The pipe network information and defect information corresponding to all pipes are matched and constructed to obtain a knowledge graph.

[0083] And each node will carry all the information of the pipeline corresponding to the pipeline number, including pipeline network information and defect information. When displayed on the front end, the node will only display the pipeline code.

[0084] S4, inputting the pipe network information and defect information corresponding to any two first nodes connected in the knowledge graph into the matching model for matching, obtaining the corresponding similarity, and synchronizing the similarity into the knowledge graph;

[0085] Specifically, according to the pipe network information and defect information corresponding to the two directly connected first nodes, the matching model is used to perform corresponding matching, and the first matching degree of the pipe network information corresponding to the two first nodes and the second matching degree of the defect information are obtained. The first matching degree is multiplied by the second matching degree to obtain the similarity between the two directly connected nodes. The similarity is synchronized to the knowledge graph, and specifically, the similarity can be displayed on the line connecting the two connected nodes.

[0086] Furthermore, the pipe network information and defect information corresponding to any two first nodes connected in the knowledge graph are input into the matching model for matching, and the corresponding similarity is obtained, including:

[0087] Extracting the pipe network information and defect information corresponding to the two first nodes connected in the knowledge graph;

[0088] According to the pipe network information and defect information corresponding to the two first nodes connected, a matching model is used to perform corresponding matching to obtain a first matching degree and a second matching degree, wherein the matching model is obtained by training based on a Bimpm model;

[0089] The first matching degree is multiplied by the second matching degree to obtain a corresponding similarity.

[0090] Specifically, by continuously extracting the pipe network information and defect information corresponding to the two first nodes connected in the knowledge graph, the pipe material, pipe installation status, pipe length and pipe diameter in the pipe network information corresponding to the two first nodes connected are used for matching, and the first matching degree of the pipe network information is obtained; and the blockage or foreign matter entering the pipe in the defect information corresponding to the two first nodes connected are used for matching, and the second matching degree of the defect information is obtained. By multiplying the first matching degree with the second matching degree, the similarity corresponding to the two connected nodes can be obtained. Thus, the similarity corresponding to all the two first nodes connected in the knowledge graph is calculated.

[0091] BiMPM (Bilateral Multi-perspective Matching) is a semantic matching model based on the "matching-aggregation" framework.

[0092] By calculating the matching degree between the corresponding pipe network information and defect information of the two first connected nodes, the first matching degree and the second matching degree are obtained respectively. By multiplying the first matching degree and the second matching degree, the similarity of the two corresponding connections can be obtained, which is convenient for the subsequent classification of each node.

[0093] S5. Calculate the first similarity between any two second nodes in the knowledge graph according to the knowledge graph with similarity, and classify the pipelines corresponding to the second nodes in the knowledge graph according to the first similarity to obtain multiple pipeline categories;

[0094] Specifically, the pipelines corresponding to the nodes are classified according to the similarity between the two first nodes. The two nodes may be directly connected, or there may be other nodes in between, that is, they are not directly connected. The similarity between the two nodes is judged by the size of the preset value to obtain multiple pipeline categories.

[0095] Further, the calculating a first similarity between any two second nodes in the knowledge graph according to the knowledge graph with similarity, and classifying the pipelines corresponding to each second node in the knowledge graph according to the first similarity includes:

[0096] Calculate the first similarity between two second nodes in the knowledge graph according to the knowledge graph with similarity, and determine whether the first similarity is greater than or equal to a preset value;

[0097] If the first similarity between the two second nodes is greater than or equal to a preset value, the two second nodes whose first similarity is greater than or equal to the preset value are stored in the same pipeline category;

[0098] If the first similarity between two second nodes is less than a preset value, the two second nodes whose first similarity is less than the preset value are stored in different pipeline categories until the first similarity between any two nodes in the knowledge graph is judged with the preset value, thereby obtaining multiple pipeline categories, and the first similarity corresponding to any two second nodes in the same pipeline category is greater than or equal to the preset value.

[0099] Specifically, since the knowledge graph contains the similarity of two connected first nodes, it is determined whether the first similarity between the two second nodes is greater than or equal to a preset value based on the similarity between the two connected first nodes. If the first similarity is greater than or equal to the preset value, the two second nodes are classified into the same pipeline category. If the first similarity is less than the preset value, the two second nodes are not classified into the same pipeline category, and the next group of two second nodes are continued to be calculated until any two nodes in the knowledge graph have been judged, and the first similarity corresponding to any two second nodes in the same pipeline category finally obtained is greater than or equal to the preset value. The two second nodes refer to two nodes randomly extracted from the knowledge graph, and the extracted two nodes may be directly connected or not directly connected in the knowledge graph.

[0100] For example, there are existing nodes 1, 2, 3, and 4, and the knowledge graph structure is that nodes 1, 2, and 3 are connected in sequence in a ring shape, and node 4 is connected to node 2. The similarity between node 1 and node 2 is calculated to be 0.85, the similarity between node 1 and node 3 is 0.85, the similarity between node 2 and node 3 is 0.85, and the similarity between node 2 and node 4 is 0.6. The preset value is 0.8. By judging the similarity between the above nodes, two pipeline categories are obtained, and the first pipeline category includes nodes 1, 2, and 3, and the second pipeline category includes node 4. In other embodiments of the present application, when the similarity between the node 2 and the node 4 is also 0.85, the corresponding classification results are that the first pipeline category includes nodes 1, 2, and 3, and the second pipeline category includes nodes 2 and node 4. The same node can appear in multiple pipeline categories.

[0101] By calculating the similarity between two second nodes, it is determined whether to store the two second nodes correspondingly or separately, thereby classifying the second nodes, so as to facilitate subsequent rapid warning of the pipelines corresponding to each second node.

[0102] Furthermore, the calculating the first similarity between two second nodes in the knowledge graph according to the knowledge graph with similarity includes:

[0103] According to the knowledge graph, determining whether two of the second nodes are directly connected;

[0104] If the two second nodes are directly connected, obtaining the first similarity corresponding to the two second nodes;

[0105] If the two second nodes are not directly connected, the shortest connection path between the two second nodes is obtained, and based on multiple second similarities on the connection path, multiple second similarities are multiplied to obtain corresponding first similarities.

[0106] Specifically, since the two second nodes to be judged may be directly connected in the knowledge graph, there may also be a situation where they are not directly connected in the knowledge graph, that is, there may be one or more intermediate nodes between the two second nodes. For example, there are existing nodes 5, 6 and 7, and the nodes 5, 6 and 7 are connected in sequence. When the second nodes are nodes 5 and 7, the two are not directly connected, but connected through intermediate nodes;

[0107] Therefore, firstly, according to the knowledge graph, determine whether the two second nodes are directly connected in the knowledge graph. When the two second nodes are directly connected, obtain the corresponding first similarity between the two second nodes from the knowledge graph; if the two second nodes are not directly connected, obtain the connection path with the least intermediate nodes between the two second nodes, that is, the shortest connection path, based on multiple second similarities on the connection path; for example, between the above-mentioned nodes 5 and 7, the connection path with the least intermediate nodes is 5-6-7, and the corresponding second similarity is the similarity between node 5 and node 6, and the similarity between node 6 and node 7. By multiplying the above-mentioned two second similarities, the corresponding first similarity can be obtained, and finally, it is determined whether the first similarity between the two second nodes is greater than or equal to the preset value.

[0108] By calculating the similarity between two second nodes, we first determine whether the two second nodes are connected. If they are connected, we directly obtain the corresponding similarity. If they are not connected, we need to obtain the shortest path between the two second nodes and then perform the similarity calculation, thereby realizing the calculation of the similarity between the two second nodes and improving the processing efficiency.

[0109] S6. When receiving a signal that a first pipeline among the pipelines has a first defect, determining the pipeline category to which the first pipeline belongs, and issuing a risk warning for a second pipeline in the pipeline category to which the first pipeline belongs.

[0110] Specifically, when receiving the first pipeline with the first defect uploaded by the offline personnel, or when the first pipeline is found to have the first defect through other systems, the pipeline category to which the first pipeline belongs is determined according to the pipeline category, and a risk warning is issued for the second pipeline in the pipeline category to which the first pipeline belongs, and the specific warning information is generated accordingly according to the first defect. When receiving the first pipeline with the first defect, the relevant information can be synchronized to the knowledge graph, so that the complete information of a pipeline can be displayed using the knowledge graph for easy traceability. The first defect includes the specific defect, the level corresponding to the defect, and the specific evaluation of the defect by the relevant personnel.

[0111] Furthermore, the first defect may carry a level corresponding to the defect, and according to the level, a graded warning may be issued for the second pipeline in the pipeline category described by the first pipeline.

[0112] Relevant personnel can conduct offline inspections on the second pipeline based on the early warning to avoid major losses due to failure to discover problems in time.

[0113] Further, when receiving a signal that a first pipeline among the pipelines has a first defect, determining a pipeline category to which the first pipeline belongs, and issuing a risk warning for a second pipeline in the pipeline category to which the first pipeline belongs includes:

[0114] determining a defect level of the first defect based on the first defect corresponding to the first pipeline;

[0115] Based on the first pipeline, determining the pipeline category to which the first pipeline belongs, and obtaining the similarity between the first pipeline and a second pipeline in the pipeline category to which the first pipeline belongs;

[0116] If the similarity between the first pipeline and the second pipeline is greater than or equal to a preset threshold, it is determined that the second pipeline and the first defect are of the same defect level and an early warning is issued;

[0117] If the similarity between the first pipeline and the second pipeline is less than a preset threshold, a preset integer is subtracted from the defect level corresponding to the first defect to obtain the defect level corresponding to the second pipeline and issue an early warning.

[0118] Specifically, the pipeline category to which the first pipeline belongs is determined through the first pipeline and the first defect corresponding to the first pipeline, and the similarity between the second pipeline and the first pipeline in the pipeline category to which the first pipeline belongs is obtained; since the similarity corresponding to any two pipelines in the same pipeline category is greater than a preset value, and the similarity corresponding to any two pipelines in the same pipeline category is known, the defect level of the first defect is determined, and the defect level is the existing defect level in the first defect.

[0119] A graded alarm is realized by setting a preset threshold value, and the preset threshold value is greater than a preset value. When the similarity between the first pipeline and the second pipeline is greater than or equal to the preset threshold value, a defect level the same as the first defect is issued to the second pipeline. When the similarity between the first pipeline and the second pipeline is less than the preset threshold value, a defect level smaller than the defect level corresponding to the first defect is issued to the second pipeline. Specifically, the defect level corresponding to the first defect is subtracted from the preset integer to obtain the defect level corresponding to the second pipeline.

[0120] When a first defect occurs in a first pipeline, a second pipeline belonging to the same category as the first defect is determined, and a graded alarm is issued to the second pipeline according to the similarity between the first pipeline and the second pipeline, so as to provide an early warning to the pipeline and avoid the actual defect of the second pipeline from being discovered.

[0121] It should be emphasized that in order to further ensure the privacy and security of the data, all the detection data can also be stored in a blockchain node.

[0122] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0123] By receiving the detection data and performing text extraction on the detection data, the data required for subsequent use, such as the pipe network information and defect information in the detection data, is extracted; the structured pipe network information and defect information are used to construct a knowledge graph; the pipe network information and defect information corresponding to any two first nodes connected in the knowledge graph are matched using a matching model to obtain corresponding similarities, and the similarities are synchronized to the knowledge graph; according to the size of the similarities, the first similarity between any two second nodes in the knowledge graph is calculated; according to the first similarity, the pipelines corresponding to the nodes are classified to obtain multiple pipeline categories; the matching model is used to classify the pipelines, so as to facilitate the subsequent acquisition of warning objects; when it is received that the first pipeline has a first defect, the pipeline category of the first pipeline is determined, and a risk warning is issued to the second pipeline in the pipeline category to which the first pipeline belongs, that is, all pipelines in the pipeline category except the first pipeline, and the pipeline is issued in advance by using the similarity between pipelines, so as to avoid the actual defects of the second pipeline being undiscovered.

[0124] This embodiment also provides a pipeline network risk warning device, such as Figure 2 The figure shows the functional module diagram of the pipeline risk warning device of the present application.

[0125] The pipe network risk warning device 100 described in the present application can be installed in an electronic device. According to the functions implemented, the pipe network risk warning device 100 may include an acquisition module 101, an extraction module 102, a construction module 103, a similarity calculation module 104, a classification module 105 and an early warning module 106. The module described in the present application can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0126] In this embodiment, the functions of each module / unit are as follows:

[0127] An acquisition module 101 is used to acquire detection data;

[0128] Further, the acquisition module includes a request sending submodule and a data calling submodule;

[0129] A request sending submodule is used to send a call request to a preset knowledge base, wherein the call request carries a signature verification token;

[0130] The data calling submodule is used to receive the verification result returned by the knowledge base, and when the verification result is passed, call the detection data in the preset knowledge base, wherein the verification result is obtained by the knowledge base through RSA asymmetric encryption verification according to the verification token.

[0131] Through the cooperation of the request sending submodule and the data calling submodule, the security of data is guaranteed and leakage is avoided through the signature verification call method.

[0132] Furthermore, the acquisition module 101 includes an instruction receiving submodule, a shooting submodule and an identification submodule;

[0133] The instruction receiving submodule is used to receive processing instructions;

[0134] The shooting submodule is used to call the camera to shoot the paper text according to the processing instruction to obtain a text image;

[0135] The recognition submodule is used to scan the text image using optical character recognition to obtain corresponding text data.

[0136] Through the cooperation of the instruction receiving submodule, the shooting submodule and the recognition submodule, optical character recognition is used to convert paper detection text into editable detection data, which realizes the onlineization of data and improves the efficiency of subsequent processing.

[0137] An extraction module 102 is used to perform text extraction on the detection data to obtain pipe network information and defect information corresponding to the pipe in the detection data;

[0138] Furthermore, the extraction module 102 includes a regular extraction submodule and a model extraction submodule;

[0139] The regular extraction submodule is used to extract the pipe network information in the detection data using a regular expression;

[0140] The model extraction submodule is used to extract the defect information in the detection data using a defect extraction model, and the defect extraction model is trained based on the RCNN model.

[0141] By cooperating with the regular extraction submodule and the model extraction submodule, regular expressions and defect extraction models are used to complete text extraction, thereby improving the accuracy of text extraction.

[0142] A construction module 103 is used to construct a knowledge graph according to the pipe network information and defect information;

[0143] The similarity calculation module 104 is used to input the pipe network information and defect information corresponding to any two first nodes connected in the knowledge graph into the matching model for matching, obtain the corresponding similarity, and synchronize the similarity into the knowledge graph.

[0144] Furthermore, the similarity calculation module 104 includes an information extraction submodule, a matching degree acquisition submodule and a corresponding multiplication submodule;

[0145] The information extraction submodule is used to extract the pipe network information and defect information corresponding to the two first nodes connected in the knowledge graph;

[0146] The matching degree acquisition submodule is used to perform corresponding matching based on the pipe network information and defect information corresponding to the two connected first nodes using a matching model to obtain a first matching degree and a second matching degree, wherein the matching model is obtained by training based on a Bimpm model;

[0147] The corresponding multiplication submodule is used to multiply the first matching degree by the second matching degree to obtain a corresponding similarity.

[0148] Through the cooperation of the information extraction submodule, the matching degree acquisition submodule and the corresponding multiplication submodule, the matching degree between the corresponding pipeline information and the defect information of the two first nodes connected is calculated, and the first matching degree and the second matching degree are obtained accordingly. By multiplying the first matching degree and the second matching degree, the two corresponding similarities can be obtained, which is convenient for the subsequent classification of each node.

[0149] A classification module 105 is used to calculate a first similarity between any two second nodes in the knowledge graph according to the knowledge graph with similarity, and classify the pipelines corresponding to each second node in the knowledge graph according to the first similarity to obtain multiple pipeline categories;

[0150] Further, the classification module includes a corresponding calculation submodule, a first storage submodule and a second storage submodule;

[0151] The corresponding calculation submodule is used to calculate the first similarity between two second nodes in the knowledge graph according to the knowledge graph with similarity, and determine whether the first similarity is greater than or equal to a preset value;

[0152] The first storage submodule is configured to store the two second nodes whose first similarity is greater than or equal to the preset value into the same pipeline category if the first similarity between the two second nodes is greater than or equal to the preset value;

[0153] The second storage submodule is used to store the two second nodes whose first similarity is less than the preset value in different pipeline categories if the first similarity between the two second nodes is less than the preset value, until the first similarity between any two nodes in the knowledge graph is judged with the preset value, thereby obtaining multiple pipeline categories, and the first similarity corresponding to any two second nodes in the same pipeline category is greater than or equal to the preset value.

[0154] By cooperating with the corresponding calculation submodule, the first storage submodule and the second storage submodule, the similarity between the two second nodes is calculated to determine whether to store the two second nodes correspondingly or separately, thereby classifying the second nodes, so as to facilitate the subsequent rapid warning of the pipelines corresponding to each second node.

[0155] Further, the corresponding calculation submodule includes a position determination unit, a first calculation unit and a second calculation unit;

[0156] The position determination unit is used to determine whether two second nodes are directly connected according to the knowledge graph;

[0157] The first calculation unit is configured to obtain a first similarity corresponding to two second nodes if the two second nodes are directly connected;

[0158] The second calculation unit is used to obtain the shortest connection path between the two second nodes if the two second nodes are not directly connected, and multiply the multiple second similarities on the connection path to obtain the corresponding first similarity.

[0159] Through the cooperation of the position judgment unit, the first calculation unit and the second calculation unit, the similarity between the two second nodes is calculated by first determining whether the two second nodes are connected. If they are connected, the corresponding similarity is directly obtained. If they are not connected, it is necessary to obtain the nearest path between the two second nodes and then perform the similarity calculation, thereby realizing the calculation of the similarity between the two second nodes and improving the processing efficiency.

[0160] The early warning module 106 is used for, when receiving a signal that a first pipeline among the pipelines has a first defect, determining the pipeline category to which the first pipeline belongs, and issuing a risk early warning for a second pipeline in the pipeline category to which the first pipeline belongs.

[0161] Further, the early warning module 106 includes a level determination submodule, a pipeline determination submodule, a first early warning submodule, and a second early warning submodule;

[0162] The level determination submodule is used to determine the defect level of the first defect based on the first defect corresponding to the first pipeline;

[0163] The pipeline determination submodule is used to determine, based on the first pipeline, the pipeline category to which the first pipeline belongs, and obtain the similarity between the second pipeline in the pipeline category to which the first pipeline belongs and the first pipeline;

[0164] The first warning submodule is configured to determine that the second pipeline and the first defect have the same defect level and issue a warning if the similarity between the first pipeline and the second pipeline is greater than or equal to a preset threshold;

[0165] The second warning submodule is used to subtract a preset integer from the defect level corresponding to the first defect to obtain the defect level corresponding to the second pipeline and issue a warning if the similarity between the first pipeline and the second pipeline is less than a preset threshold.

[0166] Through the cooperation of the level determination submodule, the pipeline determination submodule, the first warning submodule and the second warning submodule, when a first defect occurs in the first pipeline, a second pipeline belonging to the same category as the first defect is determined, and a graded alarm is issued to the second pipeline according to the similarity between the first pipeline and the second pipeline, so as to implement early warning of the pipeline and avoid the actual defect of the second pipeline from being discovered.

[0167] By adopting the above-mentioned device, the pipeline risk warning device 100 uses the acquisition module 101, the extraction module 102, the construction module 103, the similarity calculation module 104, the classification module 105 and the warning module 106 in coordination, receives the detection data, and performs text extraction on the detection data to extract the data required for subsequent use, such as the pipeline information and defect information in the detection data; constructs a knowledge graph with the structured pipeline information and defect information; uses the pipeline information and defect information corresponding to any two first nodes connected in the knowledge graph, uses the matching model to match to obtain the corresponding similarity, and synchronizes the similarity to the knowledge graph , according to the size of the similarity, the first similarity between any two second nodes in the knowledge graph is calculated, the pipelines corresponding to the nodes are classified according to the first similarity to obtain multiple pipeline categories, and the pipelines are classified by using a matching model to facilitate subsequent acquisition of warning objects. When it is received that the first pipeline has a first defect, the pipeline category of the first pipeline is determined, and a risk warning is issued for the second pipeline in the pipeline category to which the first pipeline belongs, that is, all pipelines in the pipeline category except the first pipeline. By using the similarity between pipelines, the pipeline is warned in advance, thereby avoiding the actual defect of the second pipeline from being undiscovered.

[0168] The present application also provides a computer device. Figure 3 , Figure 3 This is a basic structural block diagram of the computer device in this embodiment.

[0169] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 4 with components 41-43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0170] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0171] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the computer device 4. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions of the pipeline network risk early warning method, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0172] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run computer-readable instructions or process data stored in the memory 41, such as computer-readable instructions for running the pipeline network risk early warning method.

[0173] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0174] In this embodiment, when a processor executes computer-readable instructions stored in a memory, the steps of the pipeline network risk warning method in the above embodiment are implemented, by receiving detection data, and performing text extraction on the detection data to extract data required for subsequent use, such as pipeline network information and defect information in the detection data; constructing a knowledge graph with the structured pipeline network information and defect information; using the pipeline network information and defect information corresponding to any two first nodes connected in the knowledge graph, matching using a matching model to obtain corresponding similarities, and synchronizing the similarities to the knowledge graph; calculating the first similarity between any two second nodes in the knowledge graph according to the size of the similarity; classifying the pipelines corresponding to the nodes according to the first similarity to obtain multiple pipeline categories; classifying the pipelines by using a matching model to facilitate subsequent acquisition of warning objects; when receiving that the first pipeline has a first defect, determining the pipeline category of the first pipeline, issuing a risk warning to the second pipeline in the pipeline category to which the first pipeline belongs, that is, all pipelines in the pipeline category except the first pipeline, and issuing an early warning to the pipeline by using the similarity between pipelines, thereby avoiding the actual defects of the second pipeline from being undiscovered.

[0175] The embodiment of the present application also provides a computer-readable storage medium, which stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the above-mentioned pipeline risk warning method, by receiving detection data, and performing text extraction on the detection data to extract subsequent required data, such as pipeline information and defect information in the detection data; constructing a knowledge graph with the structured pipeline information and defect information; using the pipeline information and defect information corresponding to any two first nodes connected in the knowledge graph, using the matching model to match to obtain the corresponding similarity, and synchronizing the similarity to the In the knowledge graph, the first similarity between any two second nodes in the knowledge graph is calculated according to the size of the similarity, and the pipelines corresponding to the nodes are classified according to the first similarity to obtain multiple pipeline categories. The pipelines are classified by using a matching model to facilitate subsequent acquisition of warning objects. When it is received that the first pipeline has a first defect, the pipeline category of the first pipeline is determined, and a risk warning is issued for the second pipeline in the pipeline category to which the first pipeline belongs, that is, all pipelines except the first pipeline in the pipeline category to which the first pipeline belongs. By using the similarity between pipelines, the pipeline is warned in advance, thereby avoiding the actual defect of the second pipeline from being undiscovered.

[0176] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0177] The pipeline risk warning device, computer equipment, and computer-readable storage medium of the above-mentioned embodiments of the present application have the same technical effects as the pipeline risk warning method of the above-mentioned embodiments, and will not be expanded here.

[0178] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A pipeline network risk early warning method, characterized in that: The method comprises: Obtain test data; Performing text extraction on the detection data to obtain pipe network information and defect information corresponding to the pipeline in the detection data; By taking the pipeline numbers in the pipeline network information as nodes and the defect information as edges, the pipeline numbers with the same or similar defect information are connected to obtain the edges of the knowledge graph, and the pipeline network information and the defect information corresponding to all pipelines are matched and constructed to obtain the knowledge graph; Inputting the pipe network information and defect information corresponding to any two first nodes connected in the knowledge graph into the matching model for matching, obtaining the corresponding similarity, and synchronizing the similarity into the knowledge graph; According to the knowledge graph, determine whether two second nodes are directly connected. If the two second nodes are directly connected, obtain the first similarity corresponding to the two second nodes. If the two second nodes are not directly connected, obtain the shortest connection path between the two second nodes. Based on multiple second similarities on the connection path, multiply the multiple second similarities to obtain the corresponding first similarity. Determine whether the first similarity is greater than or equal to a preset value. If the first similarity between the two second nodes is greater than or equal to the preset value, store the two second nodes whose first similarity is greater than or equal to the preset value in the same pipeline category. If the first similarity between the two second nodes is less than the preset value, store the two second nodes whose first similarity is less than the preset value in different pipeline categories, until the first similarity between any two nodes in the knowledge graph is judged with the preset value, thereby obtaining multiple pipeline categories, wherein the first similarity corresponding to any two second nodes in the same pipeline category is greater than or equal to the preset value. When it is received that a first pipeline among the pipelines has a first defect, the pipeline category to which the first pipeline belongs is determined, and a risk warning is issued for a second pipeline in the pipeline category to which the first pipeline belongs.

2. The pipeline network risk early warning method according to claim 1 is characterized in that: The acquisition of detection data comprises: receiving processing instructions; According to the processing instruction, a camera is called to shoot paper text to obtain a text image; According to the text image, optical character recognition is used to perform scanning to obtain corresponding text data.

3. The pipeline network risk early warning method according to claim 1 is characterized in that: The text extraction of the detection data comprises: Using regular expressions, extracting the pipe network information in the detection data; The defect information in the detection data is extracted using a defect extraction model, and the defect extraction model is trained based on an RCNN model.

4. The pipeline network risk early warning method according to claim 1 is characterized in that: The pipe network information and defect information corresponding to any two first nodes connected in the knowledge graph are input into the matching model for matching, and the corresponding similarity is obtained, which includes: Extracting the pipe network information and defect information corresponding to the two first nodes connected in the knowledge graph; According to the pipe network information and defect information corresponding to the two first nodes connected, a matching model is used to perform corresponding matching to obtain a first matching degree and a second matching degree, wherein the matching model is obtained by training based on a Bimpm model; The first matching degree is multiplied by the second matching degree to obtain a corresponding similarity.

5. The pipeline network risk early warning method according to claim 1, characterized in that: When receiving that a first pipeline among the pipelines has a first defect, determining a pipeline category to which the first pipeline belongs, and issuing a risk warning for a second pipeline in the pipeline category to which the first pipeline belongs includes: determining a defect level of the first defect based on the first defect corresponding to the first pipeline; Based on the first pipeline, determining the pipeline category to which the first pipeline belongs, and obtaining the similarity between the first pipeline and a second pipeline in the pipeline category to which the first pipeline belongs; If the similarity between the first pipeline and the second pipeline is greater than or equal to a preset threshold, it is determined that the second pipeline and the first defect are of the same defect level and an early warning is issued; If the similarity between the first pipeline and the second pipeline is less than a preset threshold, a preset integer is subtracted from the defect level corresponding to the first defect to obtain the defect level corresponding to the second pipeline and issue an early warning.

6. A pipeline network risk early warning device, characterized in that: The device comprises: An acquisition module, used to acquire detection data; An extraction module, used to perform text extraction on the detection data to obtain pipe network information and defect information corresponding to the pipeline in the detection data; A construction module, used to connect the pipeline numbers with the same or similar defect information by taking the pipeline numbers in the pipeline network information as nodes and the defect information as edges to obtain the edges of the knowledge graph, and to match and construct the pipeline network information and defect information corresponding to all pipelines to obtain the knowledge graph; A similarity calculation module, used for inputting the pipe network information and defect information corresponding to any two first nodes connected in the knowledge graph into the matching model for matching, obtaining the corresponding similarity, and synchronizing the similarity into the knowledge graph; a classification module, configured to determine whether two second nodes are directly connected according to the knowledge graph, if the two second nodes are directly connected, obtain the first similarity corresponding to the two second nodes, if the two second nodes are not directly connected, obtain the shortest connection path between the two second nodes, multiply the multiple second similarities based on the multiple second similarities on the connection path to obtain the corresponding first similarity, determine whether the first similarity is greater than or equal to a preset value, if the first similarity between the two second nodes is greater than or equal to the preset value, store the two second nodes whose first similarity is greater than or equal to the preset value in the same pipeline category, if the first similarity between the two second nodes is less than the preset value, store the two second nodes whose first similarity is less than the preset value in different pipeline categories, until the first similarity between any two nodes in the knowledge graph is judged with the preset value, thereby obtaining multiple pipeline categories, wherein the first similarity corresponding to any two second nodes in the same pipeline category is greater than or equal to the preset value; The early warning module is used to determine the pipeline category to which the first pipeline belongs when receiving a signal that the first pipeline among the pipelines has a first defect, and issue a risk early warning to a second pipeline in the pipeline category to which the first pipeline belongs.

7. A computer device, characterized in that: The computer device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the pipeline network risk early warning method as described in any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the pipeline network risk early warning method as described in any one of claims 1 to 5 is implemented.

Images