Risk analysis method and device for pipe section

Abstract

The invention discloses a risk analysis method and device for a pipe section, and relates to the technical field of computers. A specific embodiment of the method comprises the following steps: acquiring pipe section feature information of a target pipe section; according to the pipe section feature information of the target pipe section, generating a representation vector of the target pipe section; acquiring a target monitoring value of the target pipe section in the target time period; and inputting the representation vector of the target pipe section and the target detection value into a risk analysis model to obtain a risk analysis result of the target pipe section. According to the implementation mode, risk analysis can be accurately carried out on the pipe section, early warning can be timely carried out on the abnormal event of the pipe section, and adverse effects on production and life of residents due to pipe section abnormity are reduced.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a risk analysis method and device for a pipe section. Background Art

[0002] The water pipe network is an important part of urban infrastructure, including water supply, drainage and other pipe networks. If the pipes in the pipe network are blocked, burst, valves fail and other incidents occur, the production and life of residents will be greatly affected. Therefore, it is necessary to strengthen the monitoring and information management of the pipe network and improve the prevention and emergency response capabilities of pipe network accidents. In the prior art, the flow rate of the pipeline is usually monitored and compared with the threshold to determine whether the pipe section is abnormal. This risk analysis method for pipe sections is not accurate enough. Summary of the invention

[0003] In view of this, an embodiment of the present invention provides a risk analysis method and device for a pipe section, which can accurately perform risk analysis on the pipe section and reduce the adverse effects on residents' production and life caused by pipe section abnormalities.

[0004] In a first aspect, an embodiment of the present invention provides a risk analysis method for a pipe segment, comprising:

[0005] Obtain pipe segment characteristic information of the target pipe segment;

[0006] Generate a representation vector of the target pipe section according to the pipe section feature information of the target pipe section;

[0007] Obtain the target monitoring value of the target pipe section in the target period;

[0008] The characterization vector and target detection value of the target pipe section are input into the risk analysis model to obtain the risk analysis result of the target pipe section.

[0009] Optionally, the pipe section characteristic information includes: pipe section time characteristics;

[0010] Obtain the pipe segment characteristic information of the target pipe segment, including:

[0011] determining at least one first time period;

[0012] Obtaining first monitoring statistics of the target pipe section in each first time period;

[0013] According to each first monitoring statistic corresponding to the target pipe section, a pipe section time characteristic of the target pipe section is generated.

[0014] Optionally, the pipe segment feature information includes: upstream and downstream correlation features;

[0015] Obtain the pipe segment characteristic information of the target pipe segment, including:

[0016] Determine the upstream and downstream equipment corresponding to the target pipe section;

[0017] determining at least one second time period and a third time period;

[0018] Obtaining second monitoring statistics of the upstream device in each second time period;

[0019] Obtaining third monitoring statistics of the downstream device in each third time period;

[0020] According to each second monitoring statistic corresponding to the upstream device and each third monitoring statistic corresponding to the downstream device, an upstream and downstream correlation feature of the target pipe section is generated.

[0021] Optionally, obtaining a target monitoring value of a target pipe section in a target period includes:

[0022] Determine the monitoring time point and the target period corresponding to the monitoring time point;

[0023] Divide the target period into multiple time slices;

[0024] Obtain the monitoring value of the target pipe section at each time slice.

[0025] Optionally, before inputting the characterization vector and the target detection value of the target pipe section into the risk analysis model to obtain the risk analysis result of the target pipe section, the following steps are further included:

[0026] Identify multiple pipe segments;

[0027] For each abnormal pipe section among the multiple pipe sections: determining at least one abnormal time period corresponding to the abnormal pipe section; generating at least one risk sample according to the monitoring value and risk classification category of the abnormal pipe section in each of the abnormal time periods;

[0028] determining a first sample size of the generated risk sample;

[0029] Determine a second sample size based on the first sample size;

[0030] Generate a plurality of normal samples of the pipe segment; wherein the number of samples of the plurality of normal samples is the second number of samples;

[0031] The risk analysis model is trained using multiple risk samples and multiple normal samples.

[0032] Optionally, the risk analysis model includes: an encoder and a classification model;

[0033] The characterization vector and target detection value of the target pipe section are input into the risk analysis model to obtain the risk analysis results of the target pipe section, including:

[0034] Inputting the characterization vector of the target pipe section and the target detection value into the encoder to obtain the target code;

[0035] The target code is input into the classification model to obtain the risk classification category and probability of the target pipe section.

[0036] Optionally, after inputting the characterization vector and the target detection value of the target pipe section into the risk analysis model to obtain the risk analysis result of the target pipe section, the following steps are further included:

[0037] Determine at least one toughness assessment indicator based on the risk analysis results of the target pipe section;

[0038] Determine the index value of each toughness assessment index for the target pipe section;

[0039] Generate a risk score for the target pipe section based on the index value of each toughness assessment index of the target pipe section;

[0040] In response to the risk score being less than a score threshold, generating alarm information for the target pipe section.

[0041] In a second aspect, an embodiment of the present invention provides a risk analysis device for a pipe section, comprising:

[0042] An information acquisition module, used to acquire pipe segment characteristic information of a target pipe segment;

[0043] A vector generation module, used for generating a representation vector of the target pipe section according to the pipe section characteristic information of the target pipe section;

[0044] A monitoring value acquisition module is used to obtain the target monitoring value of the target pipe section in the target time period;

[0045] The risk analysis module is used to input the characterization vector and target detection value of the target pipe section into the risk analysis model to obtain the risk analysis result of the target pipe section.

[0046] In a third aspect, an embodiment of the present invention provides an electronic device, including:

[0047] one or more processors;

[0048] The storage device is used to store one or more programs. When the one or more programs are executed by one or more processors, the one or more processors implement the method of any of the above embodiments.

[0049] In a fourth aspect, an embodiment of the present invention provides a computer-readable medium having a computer program stored thereon, and when the program is executed by a processor, the method of any of the above embodiments is implemented.

[0050] In a fifth aspect, an embodiment of the present invention provides a computer program product, including a computer program, which implements the method of any of the above embodiments when the computer program is executed by a processor.

[0051] An embodiment of the above invention has the following advantages or beneficial effects: a characterization vector of the target pipe segment is generated according to the pipe segment characteristic information of the target pipe segment. The characterization vector of the target pipe segment and the target detection value of the target pipe segment in the target time period are input into the risk analysis model to obtain the risk analysis result of the target pipe segment. The pipe segment characteristic information may include: basic characteristics of the pipe segment, spatial characteristics of the pipe segment, time characteristics of the pipe segment and upstream and downstream correlation characteristics. By considering the rich characteristic information of the pipe segment and using the risk analysis model, the risk analysis of the pipe segment can be accurately performed to reduce the adverse effects on residents' production and life caused by pipe segment abnormalities.

[0052] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0054] Figure 1 is a schematic diagram of a process of a risk analysis method for a pipe section provided by an embodiment of the present invention;

[0055] Figure 2 is a schematic diagram of a process of a risk analysis method for a pipe segment provided by another embodiment of the present invention;

[0056] Figure 3 is a schematic diagram of the structure of a risk analysis model provided by an embodiment of the present invention;

[0057] Figure 4 is a schematic diagram of a process of a risk analysis method for a pipe segment provided by another embodiment of the present invention;

[0058] Figure 5 It is a schematic structural diagram of a risk analysis device for a pipe section provided by an embodiment of the present invention;

[0059] Figure 6 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0061] It should be noted that the acquisition, storage, use, and processing of data in the technical solutions of the embodiments of the present invention are in compliance with the relevant provisions of national laws and regulations.

[0062] Figure 1 FIG. 1 is a schematic diagram of a process of a risk analysis method for a pipe segment provided by an embodiment of the present invention. Figure 1 As shown, the method includes:

[0063] Step 101: Obtain pipe segment characteristic information of the target pipe segment.

[0064] The pipe segment characteristic information may include: basic pipe segment characteristics, pipe segment spatial characteristics, pipe segment time characteristics, and upstream and downstream correlation characteristics. Among them, the basic pipe segment characteristics may include: pipe segment type, pipe segment model, pipe segment diameter, water pressure upper limit, flow upper limit, installation date, service life, and use category. Use categories may include: drainage and water supply.

[0065] The spatial features of the pipe segment may include: the absolute position and relative position of the pipe segment. The absolute position may include: the longitude and latitude of the pipe segment, the POI (Point of Interest) category near the pipe segment, etc. The relative position may include: the distance between the starting point of the pipe segment and the reference point, the distance between the end point of the pipe segment and the reference point, etc.

[0066] In one embodiment of the present invention, the pipe segment characteristic information includes: pipe segment time characteristics; obtaining the pipe segment characteristic information of the target pipe segment includes: determining at least one first time period; obtaining the first monitoring statistical value of the target pipe segment in each first time period; generating the pipe segment time characteristics of the target pipe segment according to each first monitoring statistical value corresponding to the target pipe segment.

[0067] The first period is a historical period. The duration of the first period can be: 1 week, 1 month, 1 year, etc. The first period can also correspond to the middle of the week, weekend, etc. A sensing device is installed on the target pipe section, which can obtain the pressure, flow, liquid level, etc. of the target pipe section. The first monitoring value may include at least one of the following: pressure, flow, liquid level, etc. The first monitoring statistic value may be a statistical value of the first monitoring value. The statistical value may be: mean value, mode, maximum value, minimum value, standard deviation, etc.

[0068] In one embodiment of the present invention, the pipe segment characteristic information includes: upstream and downstream correlation characteristics; obtaining the pipe segment characteristic information of the target pipe segment, including: determining the upstream equipment and downstream equipment corresponding to the target pipe segment; determining at least one second time period and a third time period; obtaining the second monitoring statistical value of the upstream equipment in each second time period; obtaining the third monitoring statistical value of the downstream equipment in each third time period; generating the upstream and downstream correlation characteristics of the target pipe segment according to each second monitoring statistical value corresponding to the upstream equipment and each third monitoring statistical value corresponding to the downstream equipment.

[0069] First, obtain the pipe segment connection information, which includes: the upstream and downstream connection information of the pipe network. The connection information of the pipe segment with the purpose category of water supply includes: the pump station at the starting position, the water supply pipe segment sequence, the user end at the ending position, etc. The connection information of the pipe segment with the purpose category of drainage includes: the drainage well at the starting position, the drainage pipe end sequence, the sewage treatment center, the rainwater discharge point, etc.

[0070] Upstream equipment may include: pumping stations, user terminals, drainage wells, etc. Downstream equipment may include: user terminals, discharge outlets, sewage treatment centers, rainwater discharge points, etc. Upstream equipment and downstream equipment are equipped with sensing equipment to obtain the pressure, flow, liquid level, etc. of upstream equipment and downstream equipment.

[0071] The monitoring statistics of the upstream equipment in each second period can be spliced ​​together in the order of time to obtain the upstream correlation feature of the target pipe section. The monitoring statistics of the downstream equipment in each third period can be spliced ​​together in the order of time to obtain the downstream correlation feature of the target pipe section. The upstream correlation feature and the downstream correlation feature are combined into the upstream and downstream correlation features.

[0072] The second period is a historical period. The duration of the second period may be: 1 week, 1 month, 1 year, etc. The second period may correspond to a mid-week, a weekend, etc. The second monitoring value may include at least one of the following: pressure, flow, liquid level, etc. The second monitoring statistic may be a statistic of the second monitoring value.

[0073] The third period is a historical period. The duration of the third period can be: 1 week, 1 month, 1 year, etc. The third period can correspond to the middle of the week, the weekend, etc. The third monitoring value may include at least one of the following: pressure, flow, liquid level, etc. The third monitoring statistic value may be a statistical value of the third monitoring value. The statistical value may be: average value, mode, maximum value, minimum value, standard deviation, etc.

[0074] Step 102: Generate a characterization vector of the target pipe segment according to the pipe segment feature information of the target pipe segment.

[0075] In order to reflect the correlation features between the upstream and downstream of the pipe section, and to deeply extract the basic features and spatiotemporal features of the target pipe section, a vector model is used for representation learning. The pipe section feature information of the target pipe section is input into the vector model, and the representation vector of the target pipe section is generated according to the output of the vector model. The vector model can be built based on the word2vec model.

[0076] According to the connection mode of pipe sections, the feature information of pipe sections is obtained, the basic features and spatiotemporal features of each pipe section are extracted, and the monitoring values ​​of equipment such as pump stations and water wells at the starting position, the upstream and downstream relationships of pipe section connections, etc. are considered. The vector model is used to learn the representation vector of the pipe section.

[0077] Step 103: Obtain the target monitoring value of the target pipe section in the target time period.

[0078] The target monitoring value may include at least one of the following: pressure, flow, liquid level, etc. The statistical value of the pressure, flow, liquid level, etc. of the target pipe section in the target period may be used as the target monitoring value of the target pipe section in the target period. The statistical value may be: mean value, mode, maximum value, minimum value, standard deviation, etc.

[0079] In one embodiment of the present invention, obtaining a target monitoring value of a target pipe section in a target time period includes: determining a monitoring time point and determining a target time period corresponding to the monitoring time point; dividing the target time period into multiple time slices; and obtaining the monitoring value of the target pipe section in each time slice.

[0080] The target period is determined by the monitoring time point. For example, if the monitoring time point is the current time point, the target period can be the period from the time point one week back to the current time point. If the monitoring time point is the time point when the risk event occurs, the target period can be the period from the time point 10 hours back to the time point when the risk event occurs, etc.

[0081] The length of the time slice can be set according to the needs, and the length of the time slice can be 10 minutes, 1 hour, etc. The target period can be divided into multiple time slices. Obtain the monitoring value of the target pipe section in each time slice. The monitoring value may include: pressure, flow, liquid level, etc. The monitoring value of each time slice can be spliced ​​together before and after the time slice to obtain the target monitoring value of the target pipe section in the target period. The target monitoring value obtained in this way has more reference monitoring results, and the final risk assessment result is more accurate.

[0082] Step 104: input the characterization vector and target detection value of the target pipe section into the risk analysis model to obtain the risk analysis result of the target pipe section.

[0083] The risk analysis results may include: risk classification category and probability of the target pipe section. Risk classification categories may include: pipe section blockage, pipe end burst, etc. Probability is used to characterize the probability of a risk event corresponding to the risk classification category occurring in the target pipe section.

[0084] In the solution of the embodiment of the present invention, a characterization vector of the target pipe segment is generated according to the pipe segment characteristic information of the target pipe segment. The characterization vector of the target pipe segment and the target detection value of the target pipe segment in the target period are input into the risk analysis model to obtain the risk analysis result of the target pipe segment. By considering the rich characteristic information of the pipe segment and using the risk analysis model, the risk analysis of the pipe segment can be accurately performed to reduce the adverse effects on the production and life of residents caused by pipe segment abnormalities.

[0085] The characterization vector and target detection value of the target pipe segment are input into the risk analysis model. Before obtaining the risk analysis result of the target pipe segment, the risk analysis model needs to be trained. The embodiment of the present invention also provides a method for generating training samples, including: determining multiple pipe segments; for each abnormal pipe segment in the multiple pipe segments: determining at least one abnormal time period corresponding to the abnormal pipe segment; generating at least one risk sample according to the monitoring value and risk classification category of the abnormal pipe segment in each abnormal time period; determining the first sample number of the generated risk samples; determining the second sample number according to the first sample number; generating multiple normal samples of the pipe segment; wherein the sample number of the multiple normal samples is the second sample number; and using the multiple risk samples and the multiple normal samples to train the risk analysis model.

[0086] The abnormal pipe section is a pipe section where risk events have occurred in history. The abnormal period is the historical period when risk events occurred in the abnormal pipe section. Generate risk samples corresponding to the abnormal pipe section, and the label of the risk sample is the category of the risk event that occurred. The first sample quantity is the total sample quantity of the generated risk samples. The first sample quantity can be multiplied by a coefficient to obtain the second sample quantity. The coefficient can be: 0.8, 1, 1.1, etc.

[0087] Select the pipe sections that have not experienced risk events in history, randomly select historical time periods, such as 1 week, 1 month, etc., obtain the monitoring values ​​of the pipe sections that have not experienced risk events in the time period, and generate normal samples of the second sample quantity. The label of the normal sample indicates that no risk event has occurred.

[0088] Since there are fewer abnormal situations in the pipe section, there are fewer and unevenly distributed risk events in the pipe section, and the risk samples are unbalanced. The method of the embodiment of the present invention is used to generate multiple risk samples and multiple normal samples, so that the trained model can accurately analyze and predict the risks of the pipe section, reducing the risk of model overfitting.

[0089] In one embodiment of the present invention, the risk analysis model includes: an encoder and a classification model. Figure 2 FIG. 1 is a schematic diagram of a process of a risk analysis method for a pipe segment provided by another embodiment of the present invention. Figure 2 As shown, the method includes:

[0090] Step 201: Obtain pipe segment characteristic information of the target pipe segment.

[0091] Step 202: Generate a characterization vector of the target pipe segment according to the pipe segment feature information of the target pipe segment.

[0092] Step 203: Obtain the target monitoring value of the target pipe section in the target time period.

[0093] Step 204: Input the characterization vector of the target pipe section and the target detection value into the encoder to obtain the target code.

[0094] Step 205: The target code is input into the classification model to obtain the risk classification category and probability of the target pipe section.

[0095] First, the characterization vector of the target pipe section and the target detection value are input into the encoder to obtain the target code; then the target code is input into the classification model to obtain the risk classification category and probability of the target pipe section. The target code obtained by the encoder can provide the classification model with more detailed pipe section features, making the risk analysis result of the classification model more accurate. The encoder can be constructed using a language model. The classification model can be constructed using a logistic regression model, a decision tree, a neural network, etc.

[0096] Figure 3 FIG. 1 is a schematic diagram of a risk analysis model provided by an embodiment of the present invention. Figure 3 As shown in the figure, the risk analysis model consists of an autoencoder and a deep multi-classification model. The autoencoder includes an encoder and a decoder. The encoder uses a fully connected layer of a neural network. The encoder usually gradually reduces the dimension for feature extraction and data compression. The decoder has the opposite structure to the encoder, gradually increasing the dimension and trying to reconstruct the original data.

[0097] Autoencoder is a self-supervised model that learns how to represent data by reconstructing itself during training. This data representation can achieve similar or better performance than models without self-supervised learning with a smaller amount of labeled data. The risk analysis model uses autoencoder data reconstruction for feature enhancement.

[0098] The encoder in the Autoencoder is the encoder in the embodiment of the present invention. The deep multi-classification model is the classification model in the embodiment of the present invention, which shares input and encoder with the autoencoder, and the deep multi-classification model outputs risk categories and probabilities. The risk analysis model trains the autoencoder and the multi-classification model at the same time, and the loss function of the risk analysis model is as follows:

[0099] total loss =γloss autoencoder +(1-γ)loss classify

[0100] Among them, total loss is the loss value of the risk analysis model, γ is the weight parameter, and loss autoencoder is the reconstruction error of the autoencoder, loss classifyis the classification cross entropy loss. Train the model until the loss value of the risk analysis model is less than the loss threshold, and save the model file.

[0101] For each pipe section of the water pipe network at the current time point, input the pipe section characterization vector and the target monitoring value in the most recent period into the trained risk analysis model. Obtain the risk classification category with the highest probability, and also use the probability threshold to determine the risk classification category of the target pipe section. According to the risk classification category, predict the risk analysis result of the target pipe section at the current time point, and report the hidden dangers of the risk situation. If the risk classification category is other, it means that the target pipe section has no risk.

[0102] The task of the self-supervised model Autoencoder is to extract key features through the encoder and then reconstruct the input through the decoder. The classification model constructs risk samples and normal samples for labeled training. Since the risk samples of the classification model are very small, it is difficult to achieve good results by directly training the deep multi-classification model. Therefore, training the self-supervised model and the deep risk classification at the same time and sharing the target encoding of the learned encoder output can give the deep multi-classification model more important features and improve the accuracy of classification recognition. By simultaneously training the self-supervised learning tasks and sharing the output of the encoder, the classification model's ability to understand the data is improved, making the classification more accurate.

[0103] Figure 4 FIG. 1 is a schematic diagram of a process of a risk analysis method for a pipe segment provided by another embodiment of the present invention. Figure 4 As shown, the method includes:

[0104] Step 401: Obtain pipe segment characteristic information of the target pipe segment.

[0105] Step 402: Generate a characterization vector of the target pipe segment according to the pipe segment feature information of the target pipe segment.

[0106] Step 403: Obtain the target monitoring value of the target pipe section in the target time period.

[0107] Step 404: input the characterization vector and target detection value of the target pipe section into the risk analysis model to obtain the risk analysis result of the target pipe section.

[0108] Step 405: Determine at least one toughness assessment index according to the risk analysis result of the target pipe section.

[0109] If the risk analysis results of the target pipeline section indicate that the probability of a risk event occurring in the target pipeline section exceeds the probability threshold, the resilience assessment indicators may include: redundancy of emergency facilities, reliability of backup water supply, effectiveness of activation of emergency facilities, completeness of emergency plans, whether there are measures to resist earthquakes, frost and high temperatures, average response and repair time after historical emergencies, average impact range, etc.

[0110] If the risk analysis result of the target pipeline section indicates that the probability of a risk event occurring in the target pipeline section does not exceed the probability threshold, the resilience assessment indicators may include: service life of network equipment, inspection and maintenance cycle, last inspection and maintenance time, number of historical fault repairs, water pressure stability of the water supply system, flow stability, drainage system blockage risk, drainage pressure, overall risk score of the pipeline network, etc.

[0111] Step 406: Determine the index value of each toughness assessment index of the target pipe segment; and generate a risk score of the target pipe segment according to the index value of each toughness assessment index of the target pipe segment.

[0112] The calculation formulas for each toughness assessment index and risk score can be preset in the system. According to the risk analysis results of the target pipe section, the index value of at least one toughness assessment index is calculated, and then the risk score of the target pipe section is generated.

[0113] Step 407: In response to the risk score being less than the score threshold, generating alarm information for the target pipe section.

[0114] In the scheme of the embodiment of the present invention, at least one resilience assessment index is determined based on the risk analysis result of the target pipe segment. For example, if the risk analysis result of the target pipe segment indicates that the probability of a risk event occurring in the target pipe segment exceeds the probability threshold, corresponding to the state of an emergency, the resilience assessment index is mainly used to evaluate the relevant information of fault recovery. If the risk analysis result of the target pipe segment indicates that the probability of a risk event occurring in the target pipe segment does not exceed the probability threshold, corresponding to the normal state, the resilience assessment index is mainly used to characterize the relevant information of the daily operation and maintenance of the pipe segment.

[0115] Fully consider the robustness of the pipeline network under normal conditions and the response and recovery capabilities under emergency conditions. When the risk score is less than the score threshold, an alarm message is generated for the target pipeline segment. The alarm message may include: risk score and index values ​​of each resilience assessment index. Relevant personnel check the target pipeline segment based on the alarm message to ensure that the target pipeline segment is in a normal state.

[0116] Figure 5 FIG. 1 is a schematic diagram of a risk analysis device for a pipe segment provided by an embodiment of the present invention. Figure 5 As shown, the device comprises:

[0117] The information acquisition module 501 is used to acquire the pipe segment characteristic information of the target pipe segment;

[0118] A vector generation module 502 is used to generate a characterization vector of the target pipe segment according to the pipe segment characteristic information of the target pipe segment;

[0119] A monitoring value acquisition module 503, used to acquire a target monitoring value of a target pipe section in a target period of time;

[0120] The risk analysis module 504 is used to input the characterization vector and target detection value of the target pipe section into the risk analysis model to obtain the risk analysis result of the target pipe section.

[0121] Optionally, the pipe section characteristic information includes: pipe section time characteristics;

[0122] The information acquisition module 501 is specifically used for:

[0123] determining at least one first time period;

[0124] Obtaining first monitoring statistics of the target pipe section in each first time period;

[0125] According to each first monitoring statistic corresponding to the target pipe section, a pipe section time characteristic of the target pipe section is generated.

[0126] Optionally, the pipe segment feature information includes: upstream and downstream correlation features;

[0127] The information acquisition module 501 is specifically used for:

[0128] Determine the upstream and downstream equipment corresponding to the target pipe section;

[0129] determining at least one second time period and a third time period;

[0130] Obtaining second monitoring statistics of the upstream device in each second time period;

[0131] Obtaining third monitoring statistics of the downstream device in each third time period;

[0132] According to each second monitoring statistic corresponding to the upstream device and each third monitoring statistic corresponding to the downstream device, an upstream and downstream correlation feature of the target pipe section is generated.

[0133] Optionally, the monitoring value acquisition module 503 is specifically used for:

[0134] Determine the monitoring time point and the target period corresponding to the monitoring time point;

[0135] Divide the target period into multiple time slices;

[0136] Obtain the monitoring value of the target pipe section at each time slice.

[0137] Optionally, it also includes:

[0138] A sample generation module for determining a plurality of pipe segments;

[0139] For each abnormal pipe section among the multiple pipe sections: determining at least one abnormal time period corresponding to the abnormal pipe section; generating at least one risk sample according to the monitoring value and risk classification category of the abnormal pipe section in each of the abnormal time periods;

[0140] determining a first sample size of the generated risk sample;

[0141] Determine a second sample size based on the first sample size;

[0142] Generate a plurality of normal samples of the pipe segment; wherein the number of samples of the plurality of normal samples is the second number of samples;

[0143] The risk analysis model is trained using multiple risk samples and multiple normal samples.

[0144] Optionally, the risk analysis model includes: an encoder and a classification model;

[0145] The risk analysis module 504 is specifically used for:

[0146] Inputting the characterization vector of the target pipe section and the target detection value into the encoder to obtain the target code;

[0147] The target code is input into the classification model to obtain the risk classification category and probability of the target pipe section.

[0148] Optionally, the risk analysis module 504 is further configured to:

[0149] Determine at least one toughness assessment indicator based on the risk analysis results of the target pipe section;

[0150] Determine the index value of each toughness assessment index for the target pipe section;

[0151] Generate a risk score for the target pipe section based on the index value of each toughness assessment index of the target pipe section;

[0152] In response to the risk score being less than a score threshold, generating alarm information for the target pipe section.

[0153] An embodiment of the present invention provides an electronic device, including:

[0154] one or more processors;

[0155] The storage device is used to store one or more programs. When the one or more programs are executed by one or more processors, the one or more processors implement the method of any of the above embodiments.

[0156] An embodiment of the present invention provides a computer program product, including a computer program. When the computer program is executed by a processor, the method of any of the above embodiments is implemented.

[0157] Reference below Figure 6 , which shows a schematic diagram of the structure of a computer system 600 of a terminal device suitable for implementing an embodiment of the present invention. Figure 6 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0158] like Figure 6 As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the system 600 are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0159] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage section 608 as needed.

[0160] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, the above-mentioned functions defined in the system of the present invention are executed.

[0161] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0162] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0163] The modules involved in the embodiments of the present invention may be implemented by software or hardware. The modules described may also be set in a processor, for example, they may be described as: an information acquisition module, a vector generation module, a monitoring value acquisition module, and a risk analysis module. The names of these modules do not, in some cases, constitute limitations on the modules themselves. For example, the information acquisition module may also be described as a "module for acquiring pipe segment characteristic information of a target pipe segment."

[0164] As another aspect, the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes:

[0165] Obtain pipe segment characteristic information of the target pipe segment;

[0166] Generate a representation vector of the target pipe section according to the pipe section feature information of the target pipe section;

[0167] Obtain the target monitoring value of the target pipe section in the target period;

[0168] The characterization vector and target detection value of the target pipe section are input into the risk analysis model to obtain the risk analysis result of the target pipe section.

[0169] According to the technical solution of the embodiment of the present invention, a characterization vector of the target pipe segment is generated according to the pipe segment characteristic information of the target pipe segment. The characterization vector of the target pipe segment and the target detection value of the target pipe segment in the target time period are input into the risk analysis model to obtain the risk analysis result of the target pipe segment. The pipe segment characteristic information may include: basic characteristics of the pipe segment, spatial characteristics of the pipe segment, time characteristics of the pipe segment and upstream and downstream correlation characteristics. By considering the rich characteristic information of the pipe segment and using the risk analysis model, the risk analysis of the pipe segment can be accurately performed to reduce the adverse effects on the production and life of residents caused by pipe segment abnormalities.

[0170] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A risk analysis method for a pipe section, characterized in that: include: Obtaining pipe segment characteristic information of the target pipe segment; Generating a characterization vector of the target pipe segment according to the pipe segment feature information of the target pipe segment; Obtaining a target monitoring value of the target pipe section in a target time period; The characterization vector and target detection value of the target pipe section are input into the risk analysis model to obtain the risk analysis result of the target pipe section.

2. The method according to claim 1, characterized in that: The pipe section characteristic information includes: pipe section time characteristics; The step of obtaining the pipe segment characteristic information of the target pipe segment includes: determining at least one first time period; Acquire a first monitoring statistic value of the target pipe section in each of the first time periods; A pipe section time characteristic of the target pipe section is generated according to each of the first monitoring statistical values ​​corresponding to the target pipe section.

3. The method according to claim 1, characterized in that The pipe segment characteristic information includes: upstream and downstream correlation characteristics; The step of obtaining the pipe segment characteristic information of the target pipe segment includes: Determine the upstream equipment and downstream equipment corresponding to the target pipe section; determining at least one second time period and a third time period; Obtaining a second monitoring statistic of the upstream device in each of the second time periods; Obtaining a third monitoring statistic of the downstream device in each of the third time periods; According to each second monitoring statistic corresponding to the upstream device and each third monitoring statistic corresponding to the downstream device, an upstream and downstream correlation feature of the target pipe section is generated.

4. The method according to claim 1, characterized in that The obtaining of the target monitoring value of the target pipe section in the target time period includes: Determine a monitoring time point, and determine a target time period corresponding to the monitoring time point; Dividing the target time period into a plurality of time slices; The monitoring value of the target pipe section in each of the time slices is obtained.

5. The method according to claim 1, characterized in that: Before inputting the characterization vector and the target detection value of the target pipe section into the risk analysis model to obtain the risk analysis result of the target pipe section, the method further includes: Identify multiple pipe segments; For each abnormal pipe section among the multiple pipe sections: determining at least one abnormal time period corresponding to the abnormal pipe section; generating at least one risk sample according to the monitoring value and risk classification category of the abnormal pipe section in each abnormal time period; determining a first sample size of the generated risk sample; Determining a second sample size based on the first sample size; Generate a plurality of normal samples of the pipe segment; wherein the number of samples of the plurality of normal samples is the second number of samples; The risk analysis model is trained using a plurality of the risk samples and a plurality of the normal samples.

6. The method according to claim 1, characterized in that: The risk analysis model includes: an encoder and a classification model; The characterization vector and the target detection value of the target pipe section are input into the risk analysis model to obtain the risk analysis result of the target pipe section, including: Inputting the characterization vector of the target pipe section and the target detection value into the encoder to obtain the target code; The target code is input into the classification model to obtain the risk classification category and probability of the target pipe segment.

7. The method according to claim 1, characterized in that After inputting the characterization vector and the target detection value of the target pipe section into the risk analysis model to obtain the risk analysis result of the target pipe section, the method further includes: Determining at least one toughness assessment index according to a risk analysis result of the target pipe segment; Determining an index value of the target pipe section for each of the toughness evaluation indicators; Generating a risk score of the target pipe segment according to the index value of each toughness assessment index of the target pipe segment; In response to the risk score being less than a score threshold, generating warning information for the target pipe segment.

8. A risk analysis device for a pipe section, characterized in that: include: An information acquisition module, used to acquire pipe segment characteristic information of a target pipe segment; A vector generation module, used to generate a characterization vector of the target pipe segment according to the pipe segment feature information of the target pipe segment; A monitoring value acquisition module, used to acquire a target monitoring value of the target pipe section in a target period of time; The risk analysis module is used to input the characterization vector and target detection value of the target pipe section into the risk analysis model to obtain the risk analysis result of the target pipe section.

9. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

11. A computer program product, comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images