Gas leakage detection method of gas extraction system based on laser radar

Abstract

The invention discloses a gas leakage detection method of a gas extraction system based on a laser radar, relates to the technical field of gas extraction, and solves the technical problem of delayed processing of high-risk key data caused by disordered uploading of data by an existing gas extraction system. Comprising the following steps: 1, obtaining regional pipeline data of a monitoring region corresponding to each laser radar, and generating a risk coefficient corresponding to the monitoring region according to the regional pipeline data; 2, grouping the monitoring areas according to the risk coefficients of the monitoring areas to obtain a plurality of monitoring area groups; 3, obtaining the maximum risk coefficient in each monitoring area in each monitoring area group, and obtaining the real-time monitoring data of each monitoring area; uploading the monitoring data according to a descending order of the risk coefficients; 4, analyzing the monitoring data to obtain the detection state of the corresponding pipeline; and the condition of data delay processing in a high-risk area is reduced.

Description

Technical Field

[0001] The present application belongs to the field of gas extraction technology and relates to gas extraction technology, specifically a gas leakage detection method for a gas extraction system based on laser radar. Background Art

[0002] The gas extraction system is an important part of coal mine safety production. Its main function is to extract gas from the coal seam to prevent gas accumulation from causing explosions and other safety accidents. However, in actual applications, the gas extraction system often faces the problem of pipeline leakage, which not only reduces the efficiency of gas extraction, but also may cause safety hazards.

[0003] During the monitoring process of the gas extraction system, a large amount of monitoring data needs to be uploaded to the background for analysis. The existing data uploading method of the gas extraction system often adopts a fixed order or the order of data collection. Due to the huge amount of data to be transmitted, this disordered data transmission method may cause some high-risk emergency or critical data to fail to be uploaded in time, which in turn leads to delayed processing of critical data. Gas is a flammable and explosive gas, and once leaked, it may cause very serious harm. Therefore, a gas leakage detection method based on laser radar is needed for gas extraction system. Summary of the invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art; to this end, the present application proposes a gas leakage detection method for a gas extraction system based on lidar, which is used to solve the technical problem that the existing gas extraction system uploads data in an unordered manner, resulting in delayed processing of high-risk critical data.

[0005] To achieve the above objectives, the first aspect of the present application provides a gas leakage detection method for a gas extraction system based on a laser radar, comprising the following steps:

[0006] Step 1: Obtain regional pipeline data of the corresponding monitoring area of ​​each laser radar, and generate the risk coefficient of the corresponding monitoring area according to the regional pipeline data;

[0007] Step 2: Group each monitoring area according to its risk factor to obtain several monitoring area groups;

[0008] Step 3: Obtain the maximum risk coefficient of each monitoring area in each monitoring area group, and obtain real-time monitoring data of each monitoring area; upload the monitoring data in order of the risk coefficient from large to small;

[0009] Step 4: Analyze the monitoring data to obtain the detection status of the corresponding pipeline.

[0010] The present application obtains regional pipeline data of the monitoring area corresponding to each laser radar, generates a risk coefficient of the corresponding monitoring area according to the regional pipeline data; groups each monitoring area according to each risk coefficient, and uploads the monitoring data in descending order according to the largest risk coefficient in each monitoring area within each monitoring area group; analyzes the monitoring data to obtain the detection status of the corresponding pipeline; obtains the risk coefficient of the corresponding area by comprehensively evaluating the pipelines in the monitoring area corresponding to each laser radar, groups the monitoring data according to the corresponding risk coefficient, and uploads them in an orderly manner; enables data in high-risk areas to be uploaded first, thereby reducing the occurrence of delayed data processing in high-risk areas.

[0011] Preferably, generating a risk coefficient of a corresponding monitoring area according to regional pipeline data includes:

[0012] Extract pipeline basic data, pipeline record data and pipeline handover data from regional pipeline data. When there is no pipeline handover inside the pipeline, the corresponding pipeline handover data is empty.

[0013] Extracting several basic parameters from the basic data of the pipeline, and inputting the several basic parameters into the pipeline characteristic risk scoring model to obtain the pipeline characteristic risk coefficient; the higher the pipeline characteristic risk coefficient, the stronger the pressure resistance of the pipeline, and the lower the risk of leakage; the pipeline characteristic risk scoring model is obtained by training an artificial intelligence model;

[0014] Extracting a number of maintenance data and usage data from pipeline record data, inputting pipeline maintenance data into a maintenance risk scoring model to obtain a corresponding maintenance risk coefficient; generating a corresponding usage risk coefficient according to the usage data; the maintenance risk scoring model is obtained by training an artificial intelligence model;

[0015] Generate a handover risk coefficient based on pipeline handover data; generate a risk coefficient for the corresponding monitoring area based on pipeline characteristic risk coefficient, maintenance risk coefficient, usage risk coefficient and handover risk coefficient.

[0016] Preferably, generating the risk coefficient of the corresponding monitoring area according to the pipeline characteristic risk coefficient, the maintenance risk coefficient, the use risk coefficient and the handover risk coefficient includes:

[0017] Obtain the pipeline characteristic risk coefficient, maintenance risk coefficient, use risk coefficient and handover risk coefficient in the monitoring area; and mark them as GTXi, WCXj, SFXi and JFXg respectively; GTXi is the pipeline characteristic risk coefficient corresponding to the pipeline numbered i in the monitoring area, i=1, 2,…, I; I is the total number of pipelines in the monitoring area; WCXj is the maintenance risk coefficient corresponding to the maintenance data numbered j, j=1, 2,…, J; J is the total number of maintenance data; SFXi is the use risk coefficient corresponding to the pipeline numbered i in the monitoring area; JFXg is the handover risk coefficient corresponding to the handover data numbered g, g=1, 2,…, G; G is the total number of monitoring data

[0018] By formula:

[0019]

[0020] The risk coefficient FX of the corresponding monitoring area is calculated; among them, α1, α2, α3 and α4 are proportional coefficients.

[0021] Preferably, the pipeline characteristic risk scoring model is obtained by training an artificial intelligence model, including:

[0022] Obtaining basic parameters of several pipelines and their corresponding pipeline characteristic risk scores, wherein the basic parameters include pipeline thickness, pipeline diameter, pipeline structure and pipeline material; the pipeline characteristic risk score is a comprehensive score given by experts to the pipeline based on the pipeline thickness, pipeline diameter, pipeline structure and pipeline material; the higher the pipeline characteristic risk score, the higher the possibility of abnormality; integrating the basic parameters of each pipeline and its corresponding pipeline characteristic risk score into several training data and test data;

[0023] The artificial intelligence model is trained using training data, and the trained artificial intelligence model is tested using test data; an artificial intelligence model is obtained whose input is a number of basic parameters and whose output is a pipeline characteristic risk score; the pipeline characteristic risk score is output as a pipeline characteristic risk coefficient, and finally a pipeline characteristic risk scoring model is obtained whose input is a number of basic parameters and whose output is a pipeline characteristic risk coefficient; the artificial intelligence model includes a BP neural network model and an RBF neural network model.

[0024] Preferably, the maintenance risk scoring model is obtained by training an artificial intelligence model, including:

[0025] A number of pipeline maintenance data and their corresponding maintenance risk scores are obtained, wherein the pipeline maintenance data includes maintenance area, maintenance depth, and maintenance materials, etc. The maintenance risk score is a score given by experts to the maintenance situation based on the pipeline maintenance data. The higher the score, the greater the risk of subsequent pipeline abnormalities. The pipeline maintenance data and their corresponding maintenance risk scores are integrated into a number of training data and test data.

[0026] The artificial intelligence model is trained using training data, and the trained artificial intelligence model is tested using test data; an artificial intelligence model is obtained whose input is pipeline maintenance data and whose output is maintenance risk score; the maintenance risk score is output as maintenance risk coefficient, and finally a maintenance risk score model is obtained whose input is pipeline maintenance data and whose output is maintenance risk coefficient; the artificial intelligence model includes a BP neural network model and an RBF neural network model.

[0027] Preferably, generating a corresponding usage risk coefficient according to the usage data includes:

[0028] Extract the pipeline pressure and transmission time from the usage data and mark them as GYk and CTk respectively, where k is the number of the usage data; it can be understood that the pipeline pressure and the corresponding transmission time are recorded once each time the pipeline is used for transmission; the pipeline pressure is the maximum pressure measured in the pipeline during the corresponding use process;

[0029] By formula:

[0030]

[0031] The use risk score SFX is calculated; where ZY is the maximum pressure of the corresponding pipeline; ST is the maximum service life of the corresponding pipeline; β1 is the adjustment coefficient, which can be used to remove the dimension; the specific value is set according to experience; k = 1, 2, ..., K; K is the total number of use data. It can be understood that the pressure in the pipeline is less than the maximum pressure of the pipeline; the longer the use time, the greater the risk, and the greater the pressure during use, the greater the risk.

[0032] Preferably, generating a handover risk coefficient according to pipeline handover data includes:

[0033] Extract the pipe diameter 1 and pipe diameter 2 of the two pipes to be connected in the pipe connection data, mark them as GZ1 and GZ2 respectively; and pipe thickness 1 and pipe thickness 2, mark them as GH1 and GH2 respectively;

[0034] By formula:

[0035]

[0036] The handover risk coefficient JFX between pipelines is calculated; where β2 and β3 are adjustment coefficients, and the specific values ​​are set based on experience. It can be understood that when there is no pipeline handover in the monitoring area, the handover risk coefficient is set to 0.

[0037] Preferably, the monitoring areas are grouped according to their risk coefficients to obtain a number of monitoring area groups, including:

[0038] Get the maximum amount of data CD for one data transmission and the maximum amount of data ZD for monitoring data; by formula Calculate the transmission quantity ZD of the monitoring data transmitted each time;

[0039] The monitoring areas are divided according to the number of transmissions in the order of risk coefficient from large to small, and several monitoring area groups are obtained.

[0040] The present application uploads the monitoring data in groups through the above method, and uploads as much monitoring data as possible at the same time each time, so that the efficiency of uploading the monitoring data is greatly improved, and thus the efficiency of subsequent processing of the monitoring data is also greatly improved.

[0041] Preferably, analyzing the monitoring data to obtain the pipeline detection status includes:

[0042] The monitoring data is analyzed to obtain the gas concentration, and it is determined whether the gas concentration is greater than the set leakage threshold; if so, the pipeline detection state is set to gas leakage, and the point with the highest gas concentration is taken as the leakage point; if not, the pipeline detection state is set to normal.

[0043] Preferably, the step 4 further includes obtaining a number of historical monitoring data of each laser radar, and generating a radar state of each laser radar according to the historical monitoring data; including: it can be understood that the current monitoring data also belongs to the historical monitoring data after uploading;

[0044] Extracting various performance parameters from the historical monitoring data, the performance parameters include the frequency, signal strength and monitoring range of the radar; fitting various performance parameters into performance parameter curves according to the time sequence of the historical monitoring data, and obtaining the predicted performance parameters for the next stage through various performance parameter curves;

[0045] It is determined in turn whether each predicted performance parameter is less than the corresponding performance parameter threshold; if yes, the radar state is set to an abnormal state, indicating that the radar needs to be repaired or replaced; if no, the radar state is set to a normal state.

[0046] Compared with the prior art, the beneficial effects of this application are:

[0047] 1. The present application obtains the regional pipeline data of the monitoring area corresponding to each laser radar, generates the risk coefficient of the corresponding monitoring area according to the regional pipeline data; groups the monitoring areas according to the risk coefficients, and uploads the monitoring data in descending order according to the largest risk coefficient in each monitoring area within each monitoring area group; analyzes the monitoring data to obtain the detection status of the corresponding pipeline; obtains the risk coefficient of the corresponding area by comprehensively evaluating the pipelines in the monitoring area corresponding to each laser radar, groups the monitoring data according to the corresponding risk coefficient, and uploads them in order; enables the data of high-risk areas to be uploaded first, thereby reducing the occurrence of delayed data processing in high-risk areas.

[0048] 2. This application uploads the monitoring data in groups through the above method, and uploads as much monitoring data as possible at the same time each time, which greatly improves the efficiency of uploading the monitoring data, and further greatly improves the efficiency of subsequent processing of the monitoring data. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 A schematic diagram of the steps of the gas leakage detection method of the gas extraction system in this application;

[0051] Figure 2 This is the radar status monitoring flow chart for this application. DETAILED DESCRIPTION

[0052] The technical solution of the present application will be described clearly and completely below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0053] See also Figure 1 The first embodiment of the present application provides a gas leakage detection method for a gas extraction system based on a laser radar, comprising the following steps:

[0054] Step 1: Obtain the regional pipeline data of the corresponding monitoring area of ​​each laser radar, and generate the risk coefficient of the corresponding monitoring area based on the regional pipeline data; the risk coefficient is a coefficient used to express the pipeline leakage risk in the monitoring area by comprehensively analyzing each pipeline in the monitoring area based on the regional pipeline data, and the larger the value of the risk coefficient, the greater the probability of leakage in the pipeline in the corresponding monitoring area; the corresponding monitoring data needs to be uploaded as soon as possible;

[0055] The regional pipeline data in step 1 is the data corresponding to the pipeline in the corresponding radar monitoring area, including basic pipeline data, pipeline record data and pipeline handover data; the basic pipeline data is the basic parameters of the corresponding pipeline, including pipeline thickness, pipeline diameter, pipeline structure and pipeline material, etc.; the pipeline record data is the data generated during the use of the corresponding pipeline, including maintenance data and usage data. The maintenance data is various relevant data recorded each time the pipeline is repaired; the usage data is various relevant data recorded each time the pipeline is used; the pipeline handover data is data related to the pipeline handover situation in the corresponding monitoring area;

[0056] Step 2: Group each monitoring area according to its risk coefficient to obtain a number of monitoring area groups; the monitoring area group is obtained by grouping the monitoring areas according to the risk coefficient, and is used to realize step-by-step uploading of the corresponding monitoring area data of each monitoring area;

[0057] Step 3: Obtain the maximum risk coefficient of each monitoring area in each monitoring area group, and obtain real-time monitoring data of each monitoring area; upload the monitoring data in order of the risk coefficient from large to small; when the value of the risk coefficient is larger, it means that the probability of leakage in the pipeline in the corresponding monitoring area is greater; then the corresponding monitoring data needs to be uploaded as soon as possible;

[0058] Step 4: Analyze the monitoring data to obtain the detection status of the corresponding pipeline; the detection status is the status of the pipeline in the corresponding monitoring area, including gas leakage and normal status.

[0059] This embodiment obtains regional pipeline data of the monitoring area corresponding to each laser radar, generates a risk coefficient of the corresponding monitoring area according to the regional pipeline data; groups each monitoring area according to each risk coefficient, and uploads the monitoring data in descending order according to the largest risk coefficient in each monitoring area within each monitoring area group; analyzes the monitoring data to obtain the detection status of the corresponding pipeline; obtains the risk coefficient of the corresponding area by comprehensively evaluating the pipelines in the monitoring area corresponding to each laser radar, groups the monitoring data according to the corresponding risk coefficient, and uploads them in order; enables data in high-risk areas to be uploaded first, thereby reducing the occurrence of delayed data processing in high-risk areas.

[0060] Generate the risk coefficient of the corresponding monitoring area according to the regional pipeline data, including: extracting the basic pipeline data, pipeline record data and pipeline handover data in the regional pipeline data. It can be understood that when there is no pipeline handover inside the pipeline, the corresponding pipeline handover data is empty;

[0061] Extract several basic parameters from the basic data of the pipeline, and input the basic parameters into the pipeline characteristic risk scoring model to obtain the pipeline characteristic risk coefficient; the lower the pipeline characteristic risk coefficient, the higher the comprehensive quality of the pipeline itself, the stronger the pressure resistance, and the lower the risk of pipeline leakage; the pipeline characteristic risk scoring model is obtained through artificial intelligence model training;

[0062] Extract several maintenance data and usage data from pipeline record data, input pipeline maintenance data into maintenance risk scoring model to obtain corresponding maintenance risk coefficient; the higher the maintenance risk coefficient, the greater the difficulty of maintenance when repairing the pipeline, the worse the maintenance effect, and the greater the probability of secondary leakage; generate corresponding usage risk coefficient according to usage data; the usage risk coefficient is the impact of the historical usage of the pipeline on the pipeline, and the larger the usage risk coefficient, the greater the risk of leakage in the corresponding pipeline; the maintenance risk scoring model is obtained by training the artificial intelligence model;

[0063] The handover risk coefficient is generated based on the pipeline handover data; the handover risk coefficient is the leakage risk at the pipeline joint in the corresponding monitoring area. The larger the handover risk coefficient, the greater the probability of leakage at the corresponding handover; the risk coefficient of the corresponding monitoring area is generated based on the pipeline characteristic risk coefficient, maintenance risk coefficient, use risk coefficient and handover risk coefficient.

[0064] Generate the risk coefficient of the corresponding monitoring area according to the pipeline characteristic risk coefficient, maintenance risk coefficient, use risk coefficient and handover risk coefficient, including: obtain the pipeline characteristic risk coefficient, maintenance risk coefficient, use risk coefficient and handover risk coefficient of the monitoring area; and mark them as GTXi, WCXj, SFXi and JFXg respectively; GTXi is the pipeline characteristic risk coefficient corresponding to the pipeline numbered i in the monitoring area, i=1, 2,..., I; I is the total number of pipelines in the monitoring area; WCXj is the maintenance risk coefficient corresponding to the maintenance data numbered j, j=1, 2,..., J; J is the total number of maintenance data; SFXi is the use risk coefficient corresponding to the pipeline numbered i in the monitoring area; JFXg is the handover risk coefficient corresponding to the handover data numbered g, g=1, 2,..., G; G is the total number of monitoring data

[0065] By formula:

[0066]

[0067] The risk coefficient FX of the corresponding monitoring area is calculated; among them, α1, α2, α3 and α4 are proportional coefficients.

[0068] This embodiment uses the above formula to comprehensively evaluate the pipelines in the monitoring area. When the pipeline situation in the monitoring area is more complex, the values ​​of the pipeline characteristic risk coefficient, maintenance risk coefficient, use risk coefficient and handover risk coefficient of each pipeline are larger, and the risk coefficient of the corresponding monitoring area is larger, indicating that the probability of leakage of the pipeline in the monitoring area is greater, and the area needs to be monitored in a timely manner, and the corresponding monitoring data needs to be processed in a timely manner.

[0069] The pipeline characteristic risk scoring model is obtained through artificial intelligence model training, including: obtaining the basic parameters of several pipelines and their corresponding pipeline characteristic risk scores, the basic parameters include pipeline thickness, pipeline diameter, pipeline structure and pipeline material, etc.; the pipeline characteristic risk score is a comprehensive score given by experts to the pipeline based on basic parameters such as pipeline thickness, pipeline diameter, pipeline structure and pipeline material, etc.; the higher the pipeline characteristic risk score, the worse the overall quality of the pipeline and the higher the possibility of abnormality in the pipeline; the basic parameters of each pipeline and its corresponding pipeline characteristic risk score are integrated into several training data and inspection data;

[0070] The artificial intelligence model is trained using training data, and the trained artificial intelligence model is tested using test data; specifically, the basic parameters of the pipeline in the test data are used to input the trained artificial intelligence model to obtain the corresponding pipeline characteristic risk score, and it is determined whether the difference between the output pipeline characteristic risk score and the pipeline characteristic risk score recorded in the corresponding test data is within an acceptable range. If yes, the next set of test data is tested; if not, the relevant parameters of the artificial intelligence are adjusted, and the test is performed again using the current set of test data until a set proportion of the test data passes the test, thereby obtaining an artificial intelligence model whose input is a number of basic parameters and whose output is a pipeline characteristic risk score; the pipeline characteristic risk score is output as a pipeline characteristic risk coefficient, and finally a pipeline characteristic risk score model whose input is a number of basic parameters and whose output is a pipeline characteristic risk coefficient is obtained; the artificial intelligence model includes a BP neural network model and an RBF neural network model.

[0071] The maintenance risk scoring model is obtained through artificial intelligence model training, including: obtaining a number of pipeline maintenance data and their corresponding maintenance risk scores, the pipeline maintenance data includes maintenance area, maintenance depth and maintenance materials, etc.; the maintenance risk score is the score given by experts to the maintenance situation based on the pipeline maintenance data. The higher the score, the greater the risk of subsequent pipeline abnormalities. The pipeline maintenance data and their corresponding maintenance risk scores are integrated into a number of training data and inspection data.

[0072] The artificial intelligence model is trained using the training data, and the trained artificial intelligence model is tested using the test data; specifically, the pipeline maintenance data of the pipeline in the test data is input into the trained artificial intelligence model to obtain the corresponding maintenance risk score, and it is determined whether the difference between the output maintenance risk score and the maintenance risk score recorded in the corresponding test data is within an acceptable range. If yes, the next set of test data is tested; if not, the relevant parameters of the artificial intelligence are adjusted, and the test is performed again using the current set of test data until a set proportion of the test data passes the test, thereby obtaining an artificial intelligence model whose input is the pipeline maintenance data and whose output is the maintenance risk score; the maintenance risk score is output as the maintenance risk coefficient, and finally a maintenance risk score model whose input is the pipeline maintenance data and whose output is the maintenance risk coefficient is obtained; the artificial intelligence model includes a BP neural network model and an RBF neural network model.

[0073] Generate the corresponding usage risk coefficient according to the usage data, including: extract the pipeline pressure and transmission time in the usage data, and mark them as GYk and CTk respectively, where k is the number of the usage data; it can be understood that the pipeline pressure and the corresponding transmission time will be recorded once each time the pipeline is used for transmission; the pipeline pressure is the maximum pressure measured in the pipeline during the corresponding use process;

[0074] By formula:

[0075]

[0076] The use risk score SFX is calculated; where ZY is the maximum bearing pressure of the corresponding pipeline; ST is the maximum service life of the corresponding pipeline; β1 is the adjustment coefficient, which can be used to remove the dimension; the specific value is set according to experience; k = 1, 2, ..., K; K is the total number of use data. It can be understood that the pressure in the pipeline is less than the maximum bearing pressure of the pipeline; this embodiment calculates the influence of the gas pressure in the pipeline on the pipeline during the historical use of the pipeline through the above formula. When the gas pressure in the pipeline is greater during use, the greater the influence of the gas pressure on the pipeline; the longer the use time, the greater the superimposed influence on the pipeline; the greater the risk of leakage in the pipeline.

[0077] Generating a handover risk coefficient according to the pipeline handover data includes: extracting the pipeline diameter 1 and the pipeline diameter 2 of the two pipelines handed over in the pipeline handover data, marking them as GZ1 and GZ2 respectively; and the pipeline thickness 1 and the pipeline thickness 2, marking them as GH1 and GH2 respectively;

[0078] By formula:

[0079]

[0080] The handover risk coefficient JFX between pipelines is calculated; wherein β2 and β3 are adjustment coefficients, and the specific values ​​are set according to experience. It is understandable that when there is no pipeline handover in the monitoring area, the handover risk coefficient is set to 0. This embodiment evaluates the risk of pipeline junctions through the above formula to obtain the corresponding handover risk coefficient; specifically, the greater the difference between the diameters of the two pipes being handed over, the greater the difficulty of pipeline handover, and the greater the probability of problems occurring at the pipeline junction; similarly, the greater the thickness of the pipe at the junction, the greater the difficulty of pipeline handover, and the greater the probability of problems occurring at the pipeline junction.

[0081] According to the risk coefficient of each monitoring area, each monitoring area is grouped to obtain several monitoring area groups, including: obtaining the maximum transmission data volume CD of one data transmission and the maximum data volume ZD of the monitoring data; through the formula The number of transmissions ZD of the monitoring data transmitted each time is calculated; the monitoring area is divided according to the number of transmissions in descending order of risk coefficient to obtain a number of monitoring area groups. This embodiment uploads the monitoring data in groups through the above method, and uploads as much monitoring data as possible at the same time each time, which greatly improves the efficiency of uploading the monitoring data, and further greatly improves the efficiency of subsequent processing of the monitoring data.

[0082] In another embodiment, the monitoring data of each monitoring area is acquired, and the monitoring data is arranged in order of risk coefficient from large to small to form a message queue, and the data is uploaded in order.

[0083] The monitoring data is analyzed to obtain the pipeline detection status, including: analyzing the monitoring data to obtain the gas concentration, and judging whether the gas concentration is greater than the set leakage threshold; if yes, the pipeline detection status is set to gas leakage, and the point with the highest gas concentration is taken as the leakage point; if no, the pipeline detection status is set to normal. Other methods can also be used to monitor gas leakage through the monitoring data of the laser radar.

[0084] See also Figure 2, step four also includes, obtaining several historical monitoring data of each laser radar, and generating the radar status of each laser radar according to the historical monitoring data; including: it can be understood that the current monitoring data also belongs to the historical monitoring data after uploading; extracting various performance parameters in the historical monitoring data, the performance parameters include the frequency, signal strength and monitoring range of the radar, etc.; fitting each performance parameter into a performance parameter curve according to the time sequence of the historical monitoring data, which can be fitted by interpolation or other methods; obtaining the predicted performance parameters of the next stage through each performance parameter curve; judging in turn whether each predicted performance parameter is less than the corresponding performance parameter threshold; if yes, setting the radar status to an abnormal state; if not, setting the radar status to a normal state. This embodiment also estimates the performance of the laser radar through the monitoring data of the laser radar. When the estimated result does not meet the status standard of the corresponding laser radar, it means that the relevant radar needs to be repaired or replaced; to ensure the accuracy of the corresponding monitoring data of the laser radar, and then to ensure the accuracy of the final monitoring result.

[0085] This embodiment also provides a gas leakage detection system for a gas extraction system based on a laser radar; comprising: a data acquisition module, a supervision area division module, an information transmission module, a monitoring module and a database;

[0086] The data acquisition module is used to acquire corresponding monitoring data in real time through the laser radar connected thereto;

[0087] The supervision level division module: obtains regional pipeline data of the corresponding monitoring area of ​​each laser radar through the database, generates a risk coefficient of the corresponding monitoring area according to the regional pipeline data, and generates a number of monitoring area groups according to the risk coefficient;

[0088] The information transmission module is used to obtain the maximum risk coefficient of each monitoring area in each monitoring area group; and upload the monitoring data to the leakage point monitoring module in the order of the risk coefficient from large to small;

[0089] The monitoring module analyzes the monitoring data to obtain the pipeline detection status, obtains a number of historical monitoring data of each laser radar, and generates the radar status of each laser radar according to the historical monitoring data.

[0090] It is worth noting that this system conducts risk assessment on the monitoring areas of each lidar. This can be done at fixed intervals or after each gas extraction from the pipeline, to ensure the real-time nature of the risk factor of each monitoring area.

[0091] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0092] How this application works:

[0093] The present application obtains regional pipeline data of the monitoring area corresponding to each laser radar, generates a risk coefficient of the corresponding monitoring area according to the regional pipeline data; groups each monitoring area according to each risk coefficient, and uploads the monitoring data in descending order according to the largest risk coefficient in each monitoring area within each monitoring area group; analyzes the monitoring data to obtain the detection status of the corresponding pipeline; obtains the risk coefficient of the corresponding area by comprehensively evaluating the pipelines in the monitoring area corresponding to each laser radar, groups the monitoring data according to the corresponding risk coefficient, and uploads them in an orderly manner; enables data in high-risk areas to be uploaded first, thereby reducing the occurrence of delayed data processing in high-risk areas.

[0094] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person of ordinary skill in the art should understand that the technical method of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. A gas leakage detection method for a gas extraction system based on laser radar, characterized in that: The following steps are involved: Step 1: Obtain regional pipeline data of the corresponding monitoring area of ​​each laser radar, and generate the risk coefficient of the corresponding monitoring area according to the regional pipeline data; Step 2: Group each monitoring area according to its risk factor to obtain several monitoring area groups; Step 3: Obtain the maximum risk coefficient of each monitoring area in each monitoring area group, and obtain real-time monitoring data of each monitoring area; upload the monitoring data in order of the risk coefficient from large to small; Step 4: Analyze the monitoring data to obtain the detection status of the corresponding pipeline.

2. The gas leakage detection method of the gas extraction system based on laser radar according to claim 1 is characterized in that: The generating of the risk coefficient of the corresponding monitoring area according to the regional pipeline data includes: Extract pipeline basic data, pipeline record data and pipeline handover data from regional pipeline data; Extracting several basic parameters from the basic data of the pipeline, and inputting the several basic parameters into the pipeline characteristic risk scoring model to obtain the pipeline characteristic risk coefficient; the pipeline characteristic risk scoring model is obtained by training an artificial intelligence model; Extracting a number of maintenance data and usage data from pipeline record data, inputting pipeline maintenance data into a maintenance risk scoring model to obtain a corresponding maintenance risk coefficient; generating a corresponding usage risk coefficient according to the usage data; the maintenance risk scoring model is obtained by training an artificial intelligence model; Generate a handover risk coefficient based on pipeline handover data; generate a risk coefficient for the corresponding monitoring area based on pipeline characteristic risk coefficient, maintenance risk coefficient, usage risk coefficient and handover risk coefficient.

3. The gas leakage detection method of the gas extraction system based on laser radar according to claim 2 is characterized in that: The risk coefficient of the corresponding monitoring area is generated according to the pipeline characteristic risk coefficient, the maintenance risk coefficient, the use risk coefficient and the handover risk coefficient, including: Obtain the pipeline characteristic risk coefficient, maintenance risk coefficient, use risk coefficient and handover risk coefficient in the monitoring area; and mark them as GTXi, WCXj, SFXi and JFXg respectively; GTXi is the pipeline characteristic risk coefficient corresponding to the pipeline numbered i in the monitoring area, i=1, 2,…, I; I is the total number of pipelines in the monitoring area; WCXj is the maintenance risk coefficient corresponding to the maintenance data numbered j, j=1, 2,…, J; J is the total number of maintenance data; SFXi is the use risk coefficient corresponding to the pipeline numbered i in the monitoring area; JFXg is the handover risk coefficient corresponding to the handover data numbered g, g=1, 2,…, G; G is the total number of monitoring data; By formula: The risk coefficient FX of the corresponding monitoring area is calculated; among them, α1, α2, α3 and α4 are proportional coefficients.

4. The gas leakage detection method of the gas extraction system based on laser radar according to claim 2 is characterized in that: The pipeline characteristic risk scoring model is obtained through artificial intelligence model training, including: Obtaining basic parameters of several pipelines and corresponding pipeline characteristic risk scores, wherein the basic parameters include pipeline thickness, pipeline diameter, pipeline structure and pipeline material; integrating the basic parameters of each pipeline and the corresponding pipeline characteristic risk score into several training data and test data; The artificial intelligence model is trained using training data, and the trained artificial intelligence model is tested using test data; an artificial intelligence model is obtained whose input is a number of basic parameters and whose output is a pipeline characteristic risk score; the pipeline characteristic risk score is output as a pipeline characteristic risk coefficient, and finally a pipeline characteristic risk scoring model is obtained whose input is a number of basic parameters and whose output is a pipeline characteristic risk coefficient; the artificial intelligence model includes a BP neural network model and an RBF neural network model.

5. The gas leakage detection method of the gas extraction system based on laser radar according to claim 2 is characterized in that: The maintenance risk scoring model is obtained through artificial intelligence model training, including: Obtain a number of pipeline maintenance data and their corresponding maintenance risk scores, and integrate the pipeline maintenance data and their corresponding maintenance risk scores into a number of training data and test data. The artificial intelligence model is trained using training data, and the trained artificial intelligence model is tested using test data; an artificial intelligence model is obtained whose input is pipeline maintenance data and whose output is maintenance risk score; the maintenance risk score is output as maintenance risk coefficient, and finally a maintenance risk score model is obtained whose input is pipeline maintenance data and whose output is maintenance risk coefficient; the artificial intelligence model includes a BP neural network model and an RBF neural network model.

6. The gas leakage detection method of the gas extraction system based on laser radar according to claim 2 is characterized in that: The generating a corresponding usage risk coefficient according to the usage data includes: Extract the pipeline pressure and transmission time from the usage data and mark them as GYk and CTk respectively, where k is the number of the usage data; By formula: The use risk score SFX is calculated; where ZY is the maximum bearing pressure of the corresponding pipeline; ST is the maximum service life of the corresponding pipeline; β1 is the adjustment coefficient; k = 1, 2, ..., K; K is the total number of use data.

7. The gas leakage detection method of the gas extraction system based on laser radar according to claim 2 is characterized in that: The generating of the handover risk coefficient according to the pipeline handover data includes: Extract the pipe diameter 1 and pipe diameter 2 of the two pipes to be connected in the pipe connection data, mark them as GZ1 and GZ2 respectively; and pipe thickness 1 and pipe thickness 2, mark them as GH1 and GH2 respectively; By formula: The handover risk coefficient JFX between pipelines is calculated, where β2 and β3 are adjustment coefficients.

8. The gas leakage detection method of the gas extraction system based on laser radar according to claim 1, characterized in that: The monitoring areas are grouped according to their risk coefficients to obtain a number of monitoring area groups, including: Get the maximum amount of data CD for one data transmission and the maximum amount of data ZD for monitoring data; by formula Calculate the transmission quantity ZD of the monitoring data transmitted each time; The monitoring areas are divided according to the number of transmissions in the order of risk coefficient from large to small, and several monitoring area groups are obtained.

9. The gas leakage detection method of the gas extraction system based on laser radar according to claim 1, characterized in that: The step of analyzing the monitoring data to obtain the pipeline detection status includes: The monitoring data is analyzed to obtain the gas concentration, and it is determined whether the gas concentration is greater than the set leakage threshold; if so, the pipeline detection state is set to gas leakage, and the point with the highest gas concentration is taken as the leakage point; if not, the pipeline detection state is set to normal.

10. The gas leakage detection method of a gas extraction system based on laser radar according to claim 1, characterized in that: The step 4 also includes obtaining a number of historical monitoring data of each laser radar, and generating a radar state of each laser radar according to the historical monitoring data; including: Extracting various performance parameters from historical monitoring data; fitting various performance parameters into performance parameter curves according to the time sequence of historical monitoring data, and obtaining the predicted performance parameters for the next stage through various performance parameter curves; It is determined in turn whether each predicted performance parameter is less than the corresponding performance parameter threshold; if yes, the radar state is set to an abnormal state; if not, the radar state is set to a normal state.

Images