Distributed energy pipeline natural gas transmission safety management method and Internet of Things system

Abstract

The present invention provides a distributed energy pipeline natural gas transmission safety management method and an Internet of Things system, including: obtaining a safety management area, in which a spot check site is set; based on a preset cycle, judging whether the monitoring sequence of the spot check site in a preset historical period meets a first preset condition; in response to not meeting the first preset condition, maintaining the first monitoring instruction of the spot check site; in response to meeting the first preset condition: based on the monitoring sequence and the inspection sequence in the preset historical period, adjusting the first monitoring instruction, and sending the adjusted second monitoring instruction to the monitoring perception control platform; based on the monitoring sequence, the inspection sequence and the adjacency matrix of the gas pipeline, determining the abnormal probability of the spot check site in a preset future period; based on the abnormal probability, generating a maintenance instruction, and sending the maintenance instruction to the maintenance perception control platform to deploy maintenance personnel. The present invention can improve the efficiency of pipeline monitoring, facilitate timely pipeline maintenance, and ensure the safe operation of the pipeline.

Description

Technical Field

[0001] The present invention relates to the field of pipeline natural gas transportation management, and in particular to a distributed energy pipeline natural gas transportation safety management method and an Internet of Things system. Background Art

[0002] With the development of society, natural gas, as a safe, efficient, clean and low-carbon energy, has gradually increased its share in China's primary energy consumption. Usually, there are some safety hazards in the transportation of natural gas, such as pipeline leakage and pipeline blockage, which requires a comprehensive and accurate assessment of the safety risks of natural gas transportation.

[0003] In response to the above-mentioned safety hazards in natural gas transportation, CN116957343B provides a natural gas transportation safety risk analysis method and system, which obtains the distribution information of the natural gas transmission pipeline and the booster stations on the natural gas transmission pipeline, analyzes the line-equipment failure probability set and the line-pipeline failure probability set, and then provides risk warnings for each path of the natural gas transmission pipeline. However, this method analyzes the booster stations, the locations of which are relatively fixed, and the relevant data obtained is still not comprehensive enough, resulting in the final risk judgment being still not accurate enough.

[0004] Therefore, it is necessary to provide a distributed energy pipeline natural gas transportation safety management method and Internet of Things system to conduct comprehensive monitoring and maintenance during the natural gas transportation process to ensure the safe operation of the pipeline. Summary of the invention

[0005] The invention content includes an Internet of Things system for the safe management of natural gas transmission in a distributed energy pipeline, wherein the Internet of Things system includes a distributed energy user platform, a distributed energy safety management platform, a distributed energy perception control platform, a distributed energy safety service platform and a distributed energy sensor network platform; wherein the distributed energy perception control platform includes a monitoring perception control platform and a maintenance perception control platform; the monitoring perception control platform is configured to receive monitoring instructions, and control monitoring equipment and crawling robots based on the monitoring instructions; the maintenance perception control platform is configured to receive maintenance instructions, and deploy maintenance personnel based on the maintenance instructions; the distributed energy safety service platform is configured to connect the distributed energy user platform with the distributed energy safety management platform for data transmission; the distributed energy safety management platform is configured to: obtain a safety management area, the A spot check site is set up in the safety management area; based on a preset cycle, it is determined whether the monitoring sequence of the spot check site within a preset historical period meets a first preset condition; the first preset condition is related to the number of monitoring data that are not within a normal range in the monitoring sequence; in response to not meeting the first preset condition, maintaining the first monitoring instruction of the spot check site; in response to meeting the first preset condition: based on the monitoring sequence and the inspection sequence within the preset historical period, adjusting the first monitoring instruction, and sending the adjusted second monitoring instruction to the monitoring perception control platform; based on the monitoring sequence, the inspection sequence and the adjacency matrix of the gas pipeline, determining the abnormal probability of the spot check site within a preset future period; based on the abnormal probability, generating the maintenance instruction, and sending the maintenance instruction to the maintenance perception control platform to deploy the maintenance personnel.

[0006] The invention content includes a distributed energy pipeline natural gas transmission safety management method, which is executed by a distributed energy safety management platform of a distributed energy pipeline natural gas transmission safety management Internet of Things system, and the method includes: obtaining a safety management area, in which a spot check site is set; based on a preset cycle, judging whether the monitoring sequence of the spot check site within a preset historical period meets a first preset condition; the first preset condition is related to the number of monitoring data that are not within a normal range in the monitoring sequence; in response to not meeting the first preset condition, maintaining the first monitoring instruction of the spot check site; in response to meeting the first preset condition: based on the monitoring sequence and the inspection sequence within the preset historical period, adjusting the first monitoring instruction, and sending the adjusted second monitoring instruction to the monitoring perception control platform; based on the monitoring sequence, the inspection sequence and the adjacency matrix of the gas pipeline, determining the abnormal probability of the spot check site within a preset future period; based on the abnormal probability, generating the maintenance instruction, and sending the maintenance instruction to the maintenance perception control platform to deploy the maintenance personnel.

[0007] The present invention can make the monitoring of monitoring equipment and crawling robots more targeted, effectively determine whether the current pipeline corresponding to the monitoring sequence is in a normal state and whether the current monitoring instructions need to be adjusted, and can promptly arrange maintenance personnel to perform repairs when an abnormality occurs in the pipeline to ensure the safe operation of the pipeline. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0009] Figure 1 It is a schematic diagram of the platform structure of the distributed energy pipeline natural gas transportation safety management Internet of Things system according to some embodiments of this specification;

[0010] Figure 2 is an exemplary flow chart of a method for managing the safe transportation of natural gas in a distributed energy pipeline according to some embodiments of this specification;

[0011] Figure 3 is an exemplary flow chart of adjusting the location of a spot check site and the distance interval at which a crawler robot sends sensor data during inspection according to some embodiments of this specification;

[0012] Figure 4 is an exemplary flow chart for determining the abnormal probability of a spot check site within a preset future time period according to some embodiments of this specification;

[0013] Figure 5 is an exemplary schematic diagram of a distribution map according to some embodiments of this specification;

[0014] Figure 6 is an exemplary schematic diagram of an analysis model according to some embodiments of the present specification. DETAILED DESCRIPTION

[0015] Safety supervision during natural gas pipeline transportation is an important part of ensuring the safe use of natural gas. CN116957343B only obtains fault warnings through information analysis of natural gas transmission pipelines and the booster stations on them, which is only applicable to situations where pipeline transportation problems occur due to booster station equipment failures. Therefore, in some embodiments of this specification, by reasonably setting up spot check sites, the intensity of spot checks in fault-prone areas can be increased in a targeted manner, reducing time waste in areas without faults or with fewer faults, and timely maintaining faulty areas to ensure safe operation of pipelines.

[0016] Figure 1It is a schematic diagram of the platform structure of the distributed energy pipeline natural gas transportation safety management Internet of Things system shown in some embodiments of this specification.

[0017] In some embodiments, Figure 1 As shown, the distributed energy pipeline natural gas transmission safety management Internet of Things system 100 may include a distributed energy user platform 110, a distributed energy safety service platform 120, a distributed energy safety management platform 130, a distributed energy sensor network platform 140 and a distributed energy perception control platform 150.

[0018] The distributed energy user platform 110 refers to a platform for interacting with users. In some embodiments, the distributed energy user platform 110 can interact bidirectionally with the distributed energy security service platform 120. For example, the distributed energy user platform 110 can receive maintenance information from the maintenance perception control platform 152 and adjust the gas usage time based on the maintenance information. Maintenance information refers to relevant information about pipeline maintenance by maintenance personnel. For example, maintenance information may include maintenance start time, maintenance end time, gas pipelines that need maintenance, maintenance personnel performing maintenance, etc. The distributed energy user platform 110 can be configured on a user terminal, such as a mobile terminal, a wearable device, etc.

[0019] The distributed energy security service platform 120 refers to a platform that provides distributed energy pipeline security services and information. In some embodiments, the distributed energy security service platform 120 connects the distributed energy user platform 110 and the distributed energy security management platform 130 for data transmission. The distributed energy security service platform 120 can interact with the distributed energy user platform 110 and the distributed energy security management platform 130 in a two-way manner.

[0020] The distributed energy security management platform 130 refers to a platform for executing the distributed energy pipeline natural gas transmission safety management method. In some embodiments, the distributed energy security management platform 130 can coordinate and coordinate the connection and collaboration between various functional platforms, and gather all the information of the Internet of Things to provide perception management and control management functions for the Internet of Things operation system.

[0021] In some embodiments, the distributed energy security management platform 130 is further configured with a processor. In some embodiments, the processor can process information and / or data related to the distributed energy pipeline natural gas transmission security management IoT system 100 to perform one or more functions described in this application. In some embodiments, the processor can include a central processing unit (CPU), an application specific integrated circuit (ASIC), etc., or any combination thereof.

[0022] In some embodiments, the distributed energy security management platform 130 can execute a distributed energy pipeline natural gas transmission security management method. For more information about this part, please refer to Figure 2-Figure 6 The corresponding description.

[0023] The distributed energy sensor network platform 140 refers to a functional platform for managing sensor communication. In some embodiments, the distributed energy sensor network platform 140 can be configured as a communication network or a gateway, etc. In some embodiments, the distributed energy sensor network platform 140 can realize the functions of sensing information sensor communication and controlling information sensor communication.

[0024] In some embodiments, the distributed energy security management platform 130 and the distributed energy perception control platform 150 may be connected to each other through the distributed energy sensor network platform 140. For example, the monitoring instructions issued by the distributed energy security management platform 130 may be sent to the distributed energy perception control platform 150 through the distributed energy sensor network platform 140. For detailed description of the monitoring instructions, please refer to Figure 2 The corresponding description.

[0025] The distributed energy sensing control platform 150 is a platform for generating sensing information and executing control instructions. In some embodiments, the distributed energy sensing control platform 150 can interact with the distributed energy sensor network platform 140 in a two-way manner.

[0026] The distributed energy perception control platform 150 includes a monitoring perception control platform 151 and a maintenance perception control platform 152 .

[0027] In some embodiments, the monitoring perception control platform 151 is configured to receive monitoring instructions and control the monitoring equipment and the crawling robot based on the monitoring instructions.

[0028] In some embodiments, the maintenance perception control platform 152 can receive maintenance instructions issued by the distributed energy security management platform 130, and deploy maintenance personnel based on the maintenance instructions, and generate corresponding maintenance information, and send the maintenance information to the distributed energy user platform 110. For detailed description of maintenance instructions, please refer to Figure 2 The corresponding description.

[0029] In some embodiments of the present specification, based on the distributed energy pipeline natural gas transmission safety management Internet of Things system 100, an information operation closed loop can be formed between various functional platforms, and coordinated and regularly operated under the unified management of the distributed energy safety management platform, thereby realizing the informatization and intelligence of the distributed energy pipeline natural gas transmission safety management.

[0030] Figure 2is an exemplary flow chart of a distributed energy pipeline natural gas transmission safety management method according to some embodiments of this specification. In some embodiments, process 200 can be executed by a distributed energy safety management platform 130. For more information about the distributed energy safety management platform 130, see Figure 1 The corresponding description.

[0031] In some embodiments, a distributed energy safety management platform can obtain a safety management area, in which a spot check site is set; based on a preset cycle, determine whether the monitoring sequence of the spot check site within a preset historical period meets a first preset condition; the first preset condition is related to the number of monitoring data in the monitoring sequence that is not within a normal range; in response to not meeting the first preset condition, maintain the first monitoring instruction of the spot check site; in response to meeting the first preset condition: based on the monitoring sequence and the inspection sequence within the preset historical period, adjust the first monitoring instruction, and send the adjusted second monitoring instruction to the monitoring perception control platform; based on the monitoring sequence, the inspection sequence and the adjacency matrix of the gas pipeline, determine the abnormal probability of the spot check site within a preset future period; based on the abnormal probability, generate a maintenance instruction, and send the maintenance instruction to the maintenance perception control platform to deploy maintenance personnel.

[0032] In some embodiments, Figure 2 As shown, process 200 includes the following steps.

[0033] Step 210, obtaining the security management area.

[0034] The safety management area refers to the area where gas transportation needs to be safely supervised, for example, the starting area of ​​gas transportation.

[0035] In some embodiments, the distributed energy safety management platform can divide the gas transportation range that needs to be inspected into multiple safety management areas based on geographical location. For example, the distributed energy safety management platform can divide the gas transportation range that needs to be inspected into multiple safety management areas according to longitude and latitude. The gas transportation range refers to the area through which the aboveground / underground gas pipelines that transport gas pass, including the area through which the aboveground / underground gas pipelines of multiple transportation paths from the starting point to the end point of gas transportation pass.

[0036] In some embodiments, spot check sites are set up in the security management area.

[0037] Spot check sites refer to the points set up in the gas pipeline for spot checks, such as the points where monitoring equipment is deployed and the starting points for the crawler robot to start inspection.

[0038] Monitoring equipment refers to related equipment for monitoring relevant information of gas pipelines. For example, monitoring equipment may include pressure sensors, flow sensors, and temperature sensors. The points for deploying monitoring equipment may be set by system default.

[0039] The crawling robot refers to a robot that can move in a gas pipeline and monitor the gas pipeline. In some embodiments, the crawling robot can be configured with the three sensors included in the above monitoring equipment. The starting point of the crawling robot can be set by the staff based on experience.

[0040] In some embodiments, the distributed energy safety management platform can determine the spot check site in a variety of ways. For example, the distributed energy safety management platform can divide the safety management area into multiple sub-areas according to a specific gas pipeline length, and set the center of each sub-area as a spot check site.

[0041] In some embodiments, the distributed energy safety management platform can determine the location of the spot check site and adjust it based on historical inspection data. For more information about this section, please refer to Figure 3 The corresponding description.

[0042] Step 220: Based on a preset period, determine whether the monitoring sequence of the spot-checked site within a preset historical period meets a first preset condition.

[0043] The preset period refers to the time interval between two adjacent judgment processes that are preset. The judgment process refers to judging whether the monitoring sequence meets the first preset condition. In some embodiments, in order to make the acquired monitoring sequence referenceable, the duration of the preset period is at least the duration corresponding to two preset historical time periods.

[0044] The preset historical period refers to a preset historical period, such as the last hour, the last day, etc.

[0045] In some embodiments, the preset period can be obtained in a variety of ways. For example, the preset period can be preset by a staff member based on experience.

[0046] In some embodiments, the preset period is negatively correlated to the adjustment frequency of the monitoring instruction.

[0047] Monitoring instructions refer to instructions related to monitoring of gas pipelines. For example, monitoring instructions may include the monitoring frequency of monitoring equipment and the inspection parameters of crawling robots.

[0048] Monitoring frequency refers to the number of times the gas pipeline is monitored per unit time, for example, 30 times / min.

[0049] Inspection parameters refer to the parameters related to the monitoring of the pipeline by the crawling robot. For example, the inspection parameters may include the inspection time interval and the distance interval for sending sensor data during the inspection. The time interval refers to the interval between two inspections. The distance interval refers to the interval between the positions when sending sensor data twice. Sensor data refers to the data obtained by the sensor, for example, the sensor data may include pressure sensor data, flow sensor data, temperature sensor data, etc.

[0050] The adjustment frequency of the monitoring instruction refers to the number of times the monitoring instruction is adjusted per unit time, for example, 2 times / h.

[0051] In some embodiments of the present specification, a preset period is set based on the adjustment frequency of the monitoring instructions. When the monitoring instructions are adjusted frequently, it means that the sensor data of the gas pipeline fluctuates more and there are more abnormal sensor data. At this time, the preset period can be appropriately shortened. When abnormal signs appear in the gas pipeline, the monitoring instructions can be adjusted in time to improve the safety of the system.

[0052] The monitoring sequence refers to a plurality of monitoring data related to the gas pipeline within a preset historical period, for example, pressure sensing data, flow sensing data, temperature sensing data, etc. In some embodiments, the monitoring sequence can be uploaded and obtained by the monitoring equipment of the spot check site.

[0053] The first preset condition refers to a condition for determining whether the first monitoring instruction needs to be adjusted. For the relevant description of the first monitoring instruction, please refer to the corresponding content in step 230.

[0054] In some embodiments, the first preset condition is related to the number of monitoring data in the monitoring sequence that is not within the normal range. For example, the first preset condition may be that the number of monitoring data in the monitoring sequence that is not within the normal range exceeds a preset number.

[0055] The normal range refers to the quantitative range of the pre-set monitoring data. In some embodiments, the distributed energy safety management platform can obtain the normal range by collecting statistics on the monitoring data of the gas pipeline during safe operation in the historical data.

[0056] The preset number refers to the preset maximum number of monitoring data that is not within the normal range. The preset number can be preset by the staff based on experience.

[0057] In some embodiments, the distributed energy safety management platform can determine whether each monitoring data in the monitoring sequence within a preset historical period is within a normal range, and the number of monitoring data that are not within the normal range. If the number of monitoring data that are not within the normal range exceeds a preset number, it is considered that the first preset condition is met.

[0058] Step 230: In response to the first preset condition not being met, maintaining the first monitoring instruction of the spot check site.

[0059] The first monitoring instruction refers to the monitoring instruction in the current preset cycle. The first monitoring instruction is generated by the distributed energy security management platform.

[0060] In some embodiments, when the first preset condition is not met, the number of monitoring data that is not within the normal range in the monitoring sequence does not exceed the preset number, which means that when executing the first monitoring instruction, the current monitoring data is relatively stable, and the situation at the spot check site is relatively safe. The distributed energy safety management platform can maintain the first monitoring instruction for the spot check site.

[0061] Step 240, in response to satisfying the first preset condition, executing the following steps 241 to 243:

[0062] Step 241, based on the monitoring sequence and the inspection sequence in the preset historical period, adjust the first monitoring instruction, and send the adjusted second monitoring instruction to the monitoring perception control platform.

[0063] The inspection sequence refers to a plurality of inspection data related to the gas pipeline within a preset historical period. The inspection data may include inspection time data, inspection location and sensor data. Inspection time data refers to time data related to the inspection of the crawling robot, such as the inspection start time point, the inspection end time point, the inspection duration, etc. The inspection location refers to the location data of the crawling robot during the inspection process, such as the initial location, the end location, etc. The sensor data included in the inspection data is the pressure sensor data, flow sensor data, temperature sensor data, etc. acquired by the inspection robot. In some embodiments, the inspection sequence can be uploaded and acquired by the crawling robot.

[0064] In some embodiments, the distributed energy security management platform can adjust the first monitoring instruction in a variety of ways based on the monitoring sequence and the inspection sequence in a preset historical period. For example, the distributed energy security management platform can adjust the first monitoring instruction by vector matching.

[0065] As an example only, the distributed energy security management platform can construct a first standard vector library. The first standard vector library can include multiple first standard vectors and multiple labels corresponding to them. Among them, the first standard vector can include a monitoring sequence and an inspection sequence, and the label can be a preferred monitoring frequency and inspection parameter.

[0066] In some embodiments, the distributed energy safety management platform can construct a first clustering vector based on historical monitoring sequences and historical inspection sequences in historical data, cluster multiple first clustering vectors, and construct a first standard vector based on the historical monitoring sequences and historical inspection sequences corresponding to the cluster centers formed by clustering, and use the monitoring frequencies and inspection parameters corresponding to the cluster centers that subsequently found the most abnormal data as the preferred monitoring frequencies and inspection parameters.

[0067] The historical monitoring sequence and historical inspection sequence refer to the monitoring sequence and inspection sequence within the historical preset period. Abnormal data refers to the monitoring data in the monitoring sequence that is not within the normal range.

[0068] In some embodiments, the distributed energy safety management platform can construct a first vector to be matched based on the monitoring sequence and inspection sequence in the current preset historical period, match the first vector to be matched with the first standard vector in the first standard vector library, calculate multiple vector similarities between the first vector to be matched and multiple first standard vectors, and determine the label corresponding to the first standard vector with the highest vector similarity as the adjusted monitoring frequency and inspection parameter. The vector similarity can be represented by cosine similarity, Euclidean distance, etc.

[0069] The second monitoring instruction refers to the monitoring instruction obtained after adjusting the first monitoring instruction. In some embodiments, the distributed energy safety management platform can obtain the adjusted monitoring frequency and inspection parameters as the second monitoring instruction based on the above-mentioned vector matching method.

[0070] In some embodiments, the distributed energy security management platform can send the second monitoring instruction to the monitoring perception control platform. For more information about the monitoring perception control platform, please refer to Figure 1 The corresponding description.

[0071] Step 242, based on the monitoring sequence, the inspection sequence and the adjacency matrix of the gas pipeline, determine the abnormal probability of the spot check site in a preset future time period.

[0072] The adjacency matrix of a gas pipeline refers to a two-dimensional matrix used to reflect the connection relationship between each gas pipeline. In some embodiments, the adjacency matrix of a gas pipeline can also indicate the direction of gas transportation. For example, assuming that the matrix is ​​D, gas pipeline 1 is connected to gas pipeline 2, and gas pipeline 1 is the upstream pipeline of gas pipeline 2, then D[gas pipeline 1][gas pipeline 2] = 1, D[gas pipeline 2][gas pipeline 1] = -1. In some embodiments, the adjacency matrix of a gas pipeline can be pre-set by a staff member based on a gas pipeline distribution map stored in the system.

[0073] The preset future time period refers to a preset time period in the future, such as the next day, the next week, etc. The preset future time period can be preset by the staff based on experience.

[0074] The abnormality probability refers to the probability of an abnormality occurring at the spot check site within a preset future period. It can be understood that the probability of an abnormality occurring at the spot check site refers to the probability of an abnormality occurring at the gas pipeline corresponding to the spot check site.

[0075] In some embodiments, the distributed energy safety management platform can determine the abnormal probability of the spot check site in a preset future period in a variety of ways based on the monitoring sequence, the inspection sequence and the adjacency matrix of the gas pipeline. For example, the distributed energy safety management platform can determine the abnormal probability of the spot check site in a preset future period by querying the first preset table. The first preset table refers to a table containing the corresponding relationship between the adjacency matrix of the monitoring sequence and the gas pipeline and the abnormal probability of the spot check site in the preset future period.

[0076] In some embodiments, the distributed energy safety management platform can construct a second clustering vector based on the monitoring sequence in the historical data and the adjacency matrix of the gas pipeline, cluster multiple second clustering vectors, count the number of gas pipelines in each type of cluster obtained by clustering that have abnormalities in their preset future time period, calculate the ratio of the number to the number of gas pipelines in the type of cluster, and use the ratio as the abnormality probability in the corresponding preset future time period, and include it in the first preset table.

[0077] An abnormal gas pipeline refers to the presence of abnormal data in the monitoring sequence corresponding to the gas pipeline.

[0078] In some embodiments, the distributed energy safety management platform can construct a distribution map of the safety management area based on the monitoring sequence and the adjacency matrix of the gas pipeline; based on the distribution map, the abnormal probability of the spot check site in the preset future period is determined through the analysis model. For more information about this part, see Figure 5 The corresponding description.

[0079] Step 243, based on the abnormal probability, a maintenance instruction is generated and sent to the maintenance perception control platform to deploy maintenance personnel.

[0080] Maintenance instructions refer to instructions related to pipeline maintenance. In some embodiments, the distributed energy security management platform can send maintenance instructions to the maintenance awareness control platform. For more information about the maintenance awareness control platform, please refer to Figure 1 The corresponding description.

[0081] In some embodiments, in response to the abnormality probability being greater than a first preset threshold, the distributed energy safety management platform may generate a maintenance instruction.

[0082] The first preset threshold refers to a preset maximum value of the abnormal probability. The first preset threshold can be preset by the staff based on experience.

[0083] In some embodiments of the present specification, by setting a first preset threshold for comparison with the abnormality probability, it is possible to quickly and efficiently determine whether a maintenance instruction needs to be generated, which is conducive to timely maintenance of abnormal pipelines.

[0084] In some embodiments, in response to receiving a maintenance instruction, the maintenance-aware control platform can deploy maintenance personnel who are currently idle to perform maintenance on the pipeline.

[0085] In some embodiments, process 200 further includes step 244 described below.

[0086] Step 244, generating equipment control instructions based on the abnormal probability; based on the equipment control instructions, adding and / or deleting spot check sites and / or crawling robots in the transport path segment.

[0087] Equipment control instructions refer to instructions for controlling the spot check sites and crawler robots, for example, adding a spot check site.

[0088] In some embodiments, in response to the abnormality probability exceeding a second preset threshold, the equipment control instruction is used to control the addition of a spot check site and / or a crawling robot in the transport path segment; in response to the abnormality probability being less than a third preset threshold and there are multiple spot check sites and / or crawling robots in the transport path segment, the equipment control instruction is used to control the deletion of a spot check site and / or a crawling robot in the transport path segment.

[0089] The second preset threshold and the third preset threshold are respectively preset probability thresholds for determining whether it is necessary to add and / or delete spot check sites and / or crawling robots. In some embodiments, the second preset threshold is greater than the third preset threshold. The second preset threshold and the third preset threshold can be preset by staff based on experience.

[0090] In some embodiments, the additional inspection site is located at: the center point between the location of the existing inspection site and the far end of the path segment corresponding to the existing inspection site. The far end of the path segment corresponding to the existing inspection site refers to the far end point of the transport path segment that is farther (e.g., farthest) from the existing inspection site among all the transport path segments corresponding to the inspection site. The inspection parameters of the additional crawling robot are the same as the inspection parameters of the existing crawling robot, and the first inspection of the additional crawling robot begins within a period of time after the existing crawling robot begins its inspection, and the period of time is less than the time taken by the crawling robot to conduct one inspection.

[0091] In some embodiments, the distributed energy safety management platform deletes spot check sites and / or crawling robots including: deleting the spot check sites with the smallest fluctuation in monitoring sequences within a preset historical period, and deleting the crawling robots with the smallest fluctuation in inspection sequences within a preset historical period.

[0092] The fluctuation of the monitoring sequence or inspection sequence is small, which means that the location of the spot check site or the sampling point of the crawling robot responds poorly to the changes in the gas pipeline. The setting of the spot check site or the crawling robot has a low feedback effect on the current transportation status of the gas pipeline and can be deleted to save costs.

[0093] In some embodiments of the present specification, based on experience, thresholds are set to judge the probability of abnormalities, and appropriate equipment control instructions can be determined to make the number of spot check sites and crawling robots more reasonable, so that when abnormalities occur in subsequent pipelines, they can be dealt with in a timely manner.

[0094] In some embodiments of the present specification, setting equipment control instructions based on abnormal probability can make the generated equipment control instructions more in line with the actual situation of the current pipeline, and by adding or deleting spot inspection sites and / or crawling robots, the impact of spot inspection sites and / or crawling robots on the pipeline can be reduced.

[0095] In some embodiments, process 200 further includes step 245 described below.

[0096] Step 245, based on the abnormal probability, adjust the second monitoring instruction of the spot check site, and send the third monitoring instruction obtained after adjustment to the monitoring perception control platform.

[0097] The third monitoring instruction refers to a monitoring instruction obtained by adjusting the second monitoring instruction.

[0098] In some embodiments, the distributed energy security management platform can adjust the second monitoring instruction in a variety of ways based on the abnormal probability. For example, the distributed energy security management platform can count the usage effects of the subsequent monitoring instructions actually used corresponding to the same abnormal probability in the historical data, and adjust the second monitoring instruction based on the subsequent monitoring instructions actually used with the best usage effect. The best usage effect means that the monitoring instruction actually used later obtains the least abnormal data in the monitoring sequence and inspection sequence and / or the number of the least abnormal data is greater.

[0099] In some embodiments, adjusting the second monitoring instruction of the spot-check site includes adjusting the monitoring frequency and inspection parameters of the second monitoring instruction.

[0100] In some embodiments, the monitoring frequency can be determined based on the first monitoring frequency, the second monitoring frequency and the abnormal probability. The first monitoring frequency refers to the lowest monitoring frequency that the sensor can reach. The second monitoring frequency refers to the highest monitoring frequency that the sensor can reach. Among them, the lowest monitoring frequency and the highest monitoring frequency can be determined by the performance of the sensor.

[0101] In some embodiments, the distributed energy security management platform may determine the monitoring frequency through a preset formula based on the first monitoring frequency, the second monitoring frequency, and the abnormal probability. For example, the preset formula may be represented by the following formula (1):

[0102] (1)

[0103] in, , and Respectively represent the monitoring frequency, the first monitoring frequency and the second monitoring frequency, Represents the probability of abnormality.

[0104] In some embodiments, the inspection parameters may include the time interval for the crawling robot to inspect and the distance interval for sending sensor data during inspection.

[0105] In some embodiments, the time interval can be determined based on the first time interval, the second time interval and the abnormal probability. The first time interval refers to the minimum time interval between two inspections of the crawling robot. The second time interval refers to the maximum time interval between two inspections of the crawling robot.

[0106] In some embodiments, the distance interval can be determined based on the first distance interval, the second distance interval and the abnormal probability. The first distance interval refers to the minimum distance interval between two inspections of the crawling robot. The second distance interval refers to the maximum distance interval between two inspections of the crawling robot.

[0107] Among them, the minimum time interval, the maximum time interval, the minimum distance interval, and the maximum distance interval can be preset by the staff based on experience. The method of calculating the time interval and the distance interval is similar to the method of calculating the monitoring frequency, which will not be repeated here.

[0108] In some embodiments of the present specification, the monitoring frequency of the sensor and the inspection parameters of the crawling robot are calculated based on the abnormal probability, so that the adjusted second monitoring instruction can be more accurate.

[0109] In some embodiments, the first distance intervals of different preset periods are different, and the second distance intervals of different preset periods are also different; the first distance interval and the second distance interval are determined based on the pressure fluctuation value, flow fluctuation value and robot accuracy coefficient of the monitoring sequence within the period.

[0110] The pressure fluctuation value refers to the pressure difference between the upper and lower limits of the pressure sensing data in the monitoring sequence. For example, the pressure fluctuation value is equal to the value of the upper limit of the pressure sensing data minus the lower limit.

[0111] The flow fluctuation value refers to the flow difference between the upper and lower limits of the flow sensor data in the monitoring sequence. For example, the flow fluctuation value is equal to the value of the upper limit of the flow sensor data minus the lower limit.

[0112] The robot accuracy coefficient refers to a coefficient reflecting the accuracy of the sensor carried by the crawler robot. The robot accuracy coefficient can be obtained based on the factory setting parameters of the crawler robot.

[0113] In some embodiments, the first distance interval is inversely proportional to the maximum pressure fluctuation value and the maximum flow fluctuation value, and is proportional to the robot precision coefficient. In some embodiments, the distributed energy safety management platform can determine the first distance interval by a preset formula based on the robot precision coefficient, the maximum pressure fluctuation value, and the maximum flow fluctuation value. Exemplarily, the preset formula can be expressed by the following formula (2):

[0114] (2)

[0115] in, represents the first distance interval, represents the robot accuracy coefficient, Indicates the maximum pressure fluctuation value, Indicates the maximum flow fluctuation value.

[0116] In some embodiments, the second distance interval is inversely proportional to the minimum pressure fluctuation value and the minimum flow fluctuation value, and is proportional to the robot precision coefficient. The method for calculating the second distance interval is the same as the method for calculating the first distance interval, and will not be repeated here.

[0117] In some embodiments of the present specification, by using the pressure fluctuation value, flow fluctuation value and robot precision coefficient within a preset period, the calculated first distance interval and second distance interval can be made to conform to the actual situation of the current preset period and be more accurate.

[0118] In some embodiments of the present specification, the second monitoring instruction is adjusted by abnormal probability so that the third monitoring instruction obtained after the adjustment can be more in line with the actual situation of the pipeline, and the corresponding monitoring equipment and crawling robot can respond to changes in the pipeline in a timely manner, which is conducive to the safe operation of the pipeline.

[0119] In some embodiments of the present specification, by reasonably setting up spot check sites, the monitoring of monitoring equipment and crawling robots can be made more targeted; by judging the monitoring sequence of the spot check site at preset intervals, it can be effectively determined whether the current pipeline corresponding to the monitoring sequence is in a normal state, and based on the judgment result, it can be determined whether the current monitoring instruction needs to be adjusted; based on the adjacency matrix of the monitoring sequence, the inspection sequence and the gas pipeline, the abnormal probability of the spot check site in a preset future time period can be effectively determined; based on the abnormal probability of the spot check site, the corresponding maintenance instruction is generated, and maintenance personnel can be arranged in time to perform repairs when an abnormality occurs in the pipeline, so as to ensure the safe operation of the pipeline and reduce the many inconveniences caused by pipeline abnormalities to people's lives.

[0120] Figure 3 This is an exemplary flow chart of adjusting the location of the spot check site and the distance interval of the crawler robot sending sensor data during inspection according to some embodiments of this specification. In some embodiments, process 300 can be executed by the distributed energy security management platform 130.

[0121] In some embodiments, the distributed energy safety management platform divides the safety management area into transportation path segments; based on historical inspection data, determines the location of the spot check site and the initial monitoring instructions within the transportation path segment; in response to the spot check site not meeting the second preset condition, based on the inspection sequence, adjusts the location of the spot check site and the distance interval for sending sensor data during the crawling robot inspection; wherein the second preset condition is related to the monitoring data of the spot check site.

[0122] In some embodiments, Figure 3 As shown, process 300 includes the following steps.

[0123] Step 310, divide the security management area into transportation path segments.

[0124] The transportation route segment refers to the route segment passed by some gas pipeline sections in the safety management area.

[0125] Each transport path in each security management area can be divided into one or more transport path segments.

[0126] The transportation route refers to the route that the above-ground / underground gas pipeline passes through from the starting point to the end point of gas transportation. The transportation route segment refers to the route segment that the corresponding part of the gas pipeline in the transportation route passes through.

[0127] In some embodiments, the distributed energy safety management platform may divide the multiple transportation paths included in each safety management area into multiple transportation path segments, wherein each transportation path may be divided into multiple transportation path segments.

[0128] In some embodiments, the distributed energy safety management platform divides all transportation paths within the safety management area into different numbers of candidate path segments according to the lengths of the transportation paths; selects several candidate path segments as transportation path segments of the safety management area according to the priority of the transportation paths and the coverage areas of multiple candidate path segments; wherein the priority of the transportation path is positively correlated to the length of the transportation path.

[0129] The candidate path segment refers to the path segment selected as a transportation path segment after preliminary division based on the transportation path within the safety management area.

[0130] In some embodiments, the distributed energy security management platform performs mathematical calculations on all transport paths within the security management area according to the length of the transport path, and divides transport paths of different lengths into different numbers of candidate path segments. For example, the distributed energy security management platform rounds up the ratio of the length of the transport path to the length of the standard transport path, thereby determining the number of candidate path segments corresponding to the transport path.

[0131] The standard transport path length refers to the value preset by the system to measure the path length.

[0132] In order to ensure the location of the spot check site, the inspection data obtained by the spot check site can cover the entire safety management area while avoiding setting up too many spot check sites to save costs. The distributed energy safety management platform distinguishes the transportation paths in the safety management area into main transportation paths and branch transportation paths, and divides the transportation path segments in different ways for the main transportation paths and branch transportation paths.

[0133] In some embodiments, all transportation paths within the security management area are divided into different numbers of candidate path segments according to the different lengths of the transportation paths, including: selecting the transportation path with the highest priority within the security management area as the main transportation path, and dividing the main transportation path into N candidate path segments; using transportation paths other than the main transportation path within the security management area as branch transportation paths, and the number of candidate path segments of the branch transportation paths is determined based on the N value, the length of the branch transportation path, and the length of the standard transportation path.

[0134] The main transport route refers to the transport route with the highest priority within the security management area, that is, the longest transport route within the security management area.

[0135] Priority refers to the priority of the transportation path being divided into transportation path segments. The distributed energy security management platform can set the corresponding priority based on the length of the transportation path. The longer the length of the transportation path, the higher the priority of the transportation path.

[0136] Each safety management area usually includes a main transportation path. If there are multiple longest transportation paths in a safety management area, further screening methods can be used to determine the priority of the transportation paths. For example, the distributed energy safety management platform can quantify the historical gas flow, historical accident frequency and other related factors corresponding to each transportation path, and perform weighted summation on them. The transportation path with the highest weighted sum is used as the main transportation path with the highest priority. Based on the descending order of the weighted sum of the remaining transportation paths, the priorities of the remaining transportation paths are set to the second, third, etc.

[0137] In some embodiments, the distributed energy security management platform may divide the main transportation path into N candidate path segments.

[0138] N refers to the number of candidate path segments into which the main transportation path is divided.

[0139] The N value is positively correlated to the number of gas pipelines involved in the main transportation path. For example, the greater the number of gas pipelines involved in the main transportation path, the greater the N value.

[0140] In some embodiments, the distributed energy security management platform can set the N value based on historical experience.

[0141] In some embodiments, the distributed energy security management platform may also set the N value according to the number of gas pipelines involved in the main transportation path. For example, the distributed energy security management platform may multiply the number of gas pipelines involved in the main transportation path by a preset proportionality coefficient to obtain the N value corresponding to the main transportation path. The preset proportionality coefficient may be preset by the system.

[0142] Branch transport routes refer to other transport routes other than the main transport route within the security management area.

[0143] In the same safety management area, the path length of the branch transport path is shorter than the path length of the main transport path. In the same safety management area, the number of branch transport paths corresponding to the same priority is usually 1. When there are multiple branch transport paths of the same length at the same time (for example, the number of transport paths with the second longest transport path length is greater than or equal to 2), the priority of the branch transport path can be determined in a similar way to the main transport path, which will not be repeated here.

[0144] In some embodiments, the distributed energy security management platform can determine the number of candidate path segments into which the branch transport path is divided based on the branch transport path length corresponding to the branch transport path, the standard transport path length, and the N value corresponding to the main transport path. For example, the distributed energy security management platform can calculate the ratio of the transport path length of the branch transport path to the standard transport path length, and then calculate the product of the ratio and N, and round up based on the product to an integer, thereby determining the number of candidate path segments into which the branch transport path is divided.

[0145] In some embodiments of the present specification, the greater the number of gas pipelines involved in the transportation path, the more complex the connection conditions on the transportation path. In this case, more spot check sites and more crawling robots need to be set up. By introducing the N value to divide the transportation path segments, the rationality of the setting of spot check sites and crawling robots can be further ensured, which helps to ensure the effectiveness of safety management while controlling costs.

[0146] In some embodiments, the distributed energy safety management platform may randomly select several candidate path segments as transportation path segments of the safety management area.

[0147] In some embodiments, the distributed energy safety management platform selects several candidate path segments as the transportation path segments of the safety management area according to the priority of the transportation path and the coverage area of ​​multiple candidate path segments, including one or more of the following steps S1 to S6.

[0148] Step S1: Select all candidate path segments on the transportation path with the highest priority as the first candidate transportation path segment in the safety management area.

[0149] Step S2: Calculate the coverage area of ​​the first candidate transport path segment within the safety management area.

[0150] The coverage area refers to the area corresponding to the circular area with the center of the transport path segment as the center.

[0151] In some embodiments, the distributed energy safety management platform can use the center point of each first candidate transport path segment (for example, the midpoint of the gas pipeline corresponding to each first candidate transport path segment) as the center of the circle, delineate a circular area with a preset diameter, calculate the area of ​​the circular area corresponding to each first candidate transport path segment, and add up the areas of the circular areas corresponding to all first candidate transport path segments and remove the overlapping parts as the coverage area of ​​the current first candidate transport path segment. The preset diameter can be set by the staff based on experience. The method for calculating the coverage area of ​​the second candidate transport path segment and other candidate transport path segments is similar to the method for calculating the coverage area of ​​the first candidate transport path segment, which will not be repeated here.

[0152] Step S3: In response to the coverage area of ​​the first candidate transport path segment being greater than a coverage area threshold, taking the first candidate transport path segment as the transport path segment.

[0153] The coverage area threshold can be set empirically.

[0154] The coverage area threshold is positively correlated with the number of candidate path segments. For example, the more candidate path segments there are, the larger the coverage area threshold is set.

[0155] In some embodiments of the present specification, the larger the number of candidate path segments, the larger the gas pipeline network as a whole may be and the more complex its structure may be. Associating the coverage area threshold with each other will help obtain more comprehensive and targeted monitoring and inspection data.

[0156] Step S4: In response to the coverage area of ​​the first candidate transport path segment being smaller than the coverage area threshold, the coverage areas of all candidate path segments on the transport path with the second priority within the safety management area are calculated.

[0157] The division method of the candidate path segments corresponding to the second priority transport path can refer to the division description of the branch transport path mentioned above. The calculation method of the coverage area of ​​the remaining candidate transport path segments other than the first candidate transport path segment is similar to the calculation method of the coverage area of ​​the first candidate transport path segment mentioned above, and will not be repeated here.

[0158] Step S5: Select the path segment on the transportation path with the second priority, whose coverage area overlaps the first candidate transportation path segment by the smallest rate, as the second candidate transportation path segment.

[0159] The overlap rate refers to the overlap ratio of the area involved in the coverage areas corresponding to the two groups of candidate transport path segments (eg, the first candidate transport path segment and the second candidate transport path segment) (referred to as the coverage area for short).

[0160] Step S51: In response to the sum of the coverage areas of the first candidate transport path segment and the second candidate transport path segment being greater than a coverage area threshold, the first candidate transport path segment and the second candidate transport path segment are used as transport path segments.

[0161] Step S52: In response to the sum of the coverage areas of the first candidate transport path segment and the second candidate transport path segment being less than the coverage area threshold, the path segment on the transport path with the second priority and having the second-lowest overlap rate with the coverage area of ​​the first candidate transport path segment is selected as the third candidate transport path segment.

[0162] Step S6, based on step S4, step S5, and so on, traverse to the current first candidate transport path segment, the second candidate transport path segment...the nth candidate transport path segment, and the sum of the coverage areas is greater than the coverage area threshold, and the current first candidate transport path segment, the second candidate transport path segment...the nth candidate transport path segment is taken as the final transport path segment.

[0163] In some embodiments of the present specification, through the above steps S1 to S6, the security management area can be divided more accurately and reasonably, and a more representative transportation path segment with a sufficiently large coverage area can be selected.

[0164] In some embodiments of the present specification, the transportation path segments of the safety management area are determined in combination with the priority of the transportation path and the coverage area of ​​multiple candidate path segments, which improves the representativeness and comprehensiveness of the monitoring data and inspection data, thereby improving the effectiveness of natural gas transportation safety management.

[0165] Step 320 , based on the historical inspection data, determine the location of the spot check site within the transport route segment and the initial monitoring instructions.

[0166] Historical inspection data refers to historical records of manual inspections and other related data. Historical inspection data may include potential hazards identified during manual inspections and / or locations of historical accidents and accident loss values.

[0167] Hidden danger points refer to points where gas pipelines are prone to failure.

[0168] In some embodiments, the distributed energy safety management platform can determine the potential danger point based on the historical record of the failure of the point and the historical record of the failure of the surrounding area of ​​the point. For example, although point A has never had a failure in the past, points B and C adjacent to point A have had failures in the past, so point A can be set as a potential danger point.

[0169] Potential danger points can also be set by inspectors based on inspection situations.

[0170] The accident loss value refers to data that characterizes the extent of the loss caused by the accident at the location. In some embodiments, the accident loss value can be represented by the weighted sum of the number of accidents at the location, the average gas outage area affected when the accident occurs, and the average gas outage duration when the accident occurs. The weight coefficient can be set by the staff based on experience.

[0171] The location of the sampling site refers to the geographical location of the sampling site. For more information about the sampling site, please refer to Figure 2 The corresponding description.

[0172] In some embodiments, the distributed energy safety management platform can construct a third clustering vector for each transportation path segment based on the above-mentioned hidden danger points and / or historical accident points through a clustering algorithm, cluster multiple third clustering vectors, determine the clustering center, and select the location corresponding to the clustering center of the cluster with the largest number of clustering points as the location of the spot check site in the gas transportation path segment. The clustering algorithm may include but is not limited to K-Means clustering and / or density-based clustering method (DBSCAN), etc. For more information about clustering, please refer to the relevant description of clustering elsewhere in this article, such as Figure 2 .

[0173] Initial monitoring instructions refer to the monitoring instructions initially set for the sampling site. Each sampling site has an initial monitoring instruction.

[0174] In some embodiments, the distributed energy security management platform can adjust the preset monitoring instructions corresponding to the security management area based on the adjustment ratio as the initial monitoring instructions for the path segment. The preset monitoring instructions can be the monitoring instructions set by the monitoring device at the factory.

[0175] In some embodiments, the distributed energy safety management platform can determine the adjustment ratio based on the average value of the accident loss value in the path segment in the historical data and the average value of the accident loss value in the safety management area where the path segment is located. For example, the distributed energy safety management platform can determine the adjustment ratio based on the following formula (3):

[0176] (3)

[0177] in, It means adjusting the proportion. It refers to the average value of accident loss in the path segment. It refers to the average value of accident loss values ​​in the safety management area where the path segment is located. In some embodiments, the distributed energy safety management platform can obtain the average value of accident loss values ​​in the path segment and the average value of accident loss values ​​in the safety management area where the path segment is located by calling historical data stored in the system.

[0178] If the average value of the accident loss value in the path segment is greater than the average value of the accident loss value in the safety management area where the path segment is located, the distributed energy safety management platform will increase the monitoring frequency, reduce the inspection time interval, and reduce the distance interval for sending sensor data during inspections according to the adjustment ratio. Conversely, if the average value of the accident loss value in the path segment is less than the average value of the accident loss value in the safety management area where the path segment is located, the distributed energy safety management platform will reduce the monitoring frequency, increase the inspection time interval, and increase the distance interval for sending sensor data during inspections according to the adjustment ratio.

[0179] For example, if the average value of the accident loss value within the path segment is greater than the average value of the accident loss value within the safety management area where the path segment is located, and the absolute value of the determined adjustment ratio is M%, then the adjustment of the preset monitoring instruction is: the monitoring frequency is increased by M% according to the adjustment ratio, the inspection time interval is reduced by M% according to the adjustment ratio, and the distance interval for sending sensor data during inspection is reduced by M% according to the adjustment ratio.

[0180] Step 330, in response to the spot check site not satisfying the second preset condition, based on the inspection sequence, adjusting the position of the spot check site and the distance interval for sending sensor data during the crawling robot inspection.

[0181] The second preset condition refers to a condition that must be met to determine whether to perform the adjustment process. The adjustment process refers to adjusting the location of the spot check site and the distance interval at which the crawler robot sends the sensor data during the inspection. Among them, the second preset condition is related to the monitoring data of the spot check site. For example, the second preset condition includes that after the location of the spot check site is determined, when an accident occurs in the current transportation path segment, the monitoring data sent by the spot check site is within a normal range.

[0182] In some embodiments, in response to the spot check site not meeting the second preset condition, the distributed energy safety management platform obtains multiple inspection sequences before the accident, and selects the center positions of multiple gas pipelines corresponding to multiple sensor data that are not within the normal range as the adjusted spot check site locations.

[0183] If in the inspection sequence, the previous and / or subsequent sensor data of the sensor data that is not within the normal range is within the normal range, the distributed energy safety management platform will reduce the distance interval for sending sensor data during the crawling robot inspection, for example, to 50% of the original.

[0184] It is understandable that when an accident occurs in a gas pipeline, sensor data at multiple points are abnormal. Therefore, when the sensor data before and / or after the sensor data that is not in the normal range is in the normal range, it means that the distance interval of the crawler sending sensor data is too large. At this time, the midpoint of the two sensor data transmissions can be taken and the sensor data can be sent again to ensure that enough abnormal data is collected. For more information about the distance interval of sending sensor data, please refer to Figure 2 The corresponding description.

[0185] In some embodiments of the present specification, the location of the spot check site and the distance interval for sending sensor data during the crawling robot inspection are adjusted based on historical inspection data and inspection sequences, thereby achieving a more reasonable spot check site location setting and a distance interval setting for the crawling robot, which helps to obtain more accurate and appropriate monitoring data and inspection data and improves the monitoring of abnormal situations.

[0186] Figure 4 This is an exemplary flow chart for determining the abnormal probability of a spot check site within a preset future period according to some embodiments of this specification. In some embodiments, process 400 can be executed by the distributed energy security management platform 130.

[0187] In some embodiments, the distributed energy safety management platform constructs a distribution map of the safety management area based on the monitoring sequence of each spot check site and the adjacency matrix of the gas pipeline; wherein the nodes of the distribution map are gas pipelines, the node features of the distribution map are the monitoring sequences of the spot check sites that meet the preset distance conditions, and the edges of the distribution map represent the connection relationship between the gas pipelines; according to the distribution map, the abnormal probability of the gas pipeline in a preset future time period is determined through the analysis model; wherein the analysis model is a machine learning model; based on the abnormal probability of the gas pipeline in the preset future time period, the abnormal probability of the spot check site in the preset future time period is determined.

[0188] In some embodiments, Figure 4 As shown, process 400 includes the following steps.

[0189] Step 410, constructing a distribution map of the safety management area based on the monitoring sequence of each spot check site and the adjacency matrix of the gas pipeline.

[0190] The nodes of the distribution map are gas pipelines. Figure 5 As shown, the distribution map 500 includes nodes A to E, which correspond to gas pipelines a to e (not shown in the figure), respectively.

[0191] The node feature of the distribution map is a monitoring sequence of the spot check sites that meet the preset distance condition. The preset distance condition includes that the transportation distance between the spot check site and the gas pipeline corresponding to the node is the shortest. The preset distance condition can be preset by the staff based on experience.

[0192] The nodes of the distribution map include root nodes and non-root nodes. For more information about root nodes and non-root nodes, see Figure 6 The corresponding description.

[0193] The edges of the distribution graph represent the connection relationship between the gas pipelines. If the gas pipelines corresponding to two nodes are connected, there is an edge connecting the two nodes. The direction of the edge is consistent with the direction of gas transportation, from the upstream gas pipeline to the downstream gas pipeline. Figure 5 As shown, the gas pipelines a and gas pipelines c corresponding to the nodes A and C in the distribution map 500 are connected to each other, and the gas pipeline a is the upstream pipeline and the gas pipeline c is the downstream pipeline, so there is an edge AC from the node A to the node C.

[0194] Step 420, based on the distribution map, by analyzing the model, determine the abnormal probability of the gas pipeline in a preset future time period.

[0195] The analytical model is a model used to predict the probability of abnormality of gas pipelines in a preset future period. The analytical model can be a machine learning model. For more information about analytical models, see Figure 6 Related instructions.

[0196] Step 430, based on the abnormal probability of the gas pipeline in the preset future time period, determine the abnormal probability of the spot check site in the preset future time period.

[0197] In some embodiments, the distributed energy safety management platform uses the spot check site with the shortest transportation distance from the gas pipeline as the spot check site corresponding to the gas pipeline; and uses the average of the abnormality probabilities of multiple gas pipelines corresponding to each spot check site in a preset future time period as the abnormality probability of each spot check site in the preset future time period.

[0198] In some embodiments of the present specification, by constructing a distribution map and combining it with an analysis model to determine the abnormal probability, the abnormal probability prediction can be performed more comprehensively and quickly, which helps to improve the intelligence and precision of energy security management.

[0199] It should be noted that the above descriptions of process 200, process 300 and process 400 are only for example and explanation, and do not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to process 200, process 300 and process 400 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

[0200] Figure 6 is an exemplary schematic diagram of an analysis model according to some embodiments of the present specification.

[0201] In some embodiments, Figure 6 As shown, the analysis model 600 includes a first neural network layer 610 and a second neural network layer 620, the first neural network layer corresponds to the root node in the distribution map, the second neural network layer corresponds to the non-root node in the distribution map, and the connection relationship between each neural network layer is determined based on the edge of the distribution map; the input of the first neural network layer 610 includes the monitoring sequence 631 of the spot check site corresponding to its corresponding root node, and the output includes the abnormal probability 640 of the gas pipeline corresponding to the root node within a preset future time period; the input of the second neural network layer includes the monitoring sequence 632 of the spot check site corresponding to its corresponding non-root node, and the abnormal probability set 641 of the node pointing to the non-root node within a preset future time period, and the output includes the abnormal probability 640 of the gas pipeline corresponding to the non-root node within a preset future time period.

[0202] Among them, the first neural network layer corresponds one-to-one to the root node in the distribution map, and the second neural network layer corresponds one-to-one to the non-root node in the distribution map.

[0203] A root node is a node that has no edges pointing to it from other nodes. Figure 5 Node A and Node B shown in .

[0204] A non-root node is a node that has edges pointing to it from other nodes. Figure 5 Node C, node D, and node E are shown.

[0205] The first neural network layer refers to a model for determining the abnormal probability corresponding to the root node. In some embodiments, the first neural network layer can be a machine learning model, such as a deep neural network (DNN) model.

[0206] In some embodiments, the monitoring sequence of the spot check site corresponding to the root node corresponding to the first neural network layer can be uploaded and obtained by the monitoring equipment of the spot check site.

[0207] The second neural network layer refers to a model for determining the abnormal probability corresponding to the non-root node. In some embodiments, the second neural network layer can be a machine learning model. For example, a deep neural network model, etc.

[0208] In some embodiments, the monitoring sequence of the spot check site corresponding to the non-root node corresponding to the second neural network layer can be uploaded and obtained by the monitoring equipment of the spot check site.

[0209] The abnormal probability set refers to the set of abnormal probabilities of gas pipelines corresponding to nodes pointing to non-root nodes. In some embodiments, the distributed energy safety management platform can traverse the distribution map to find all other nodes pointing to the non-root node, and then count the abnormal probabilities of gas pipelines corresponding to all other nodes in a preset future period, and determine the abnormal probability set in the preset future period.

[0210] Since the execution and training of the analysis model are gradually advanced along the direction of gas transportation, in the same batch training process, the output of the neural network layer corresponding to the upstream node is generated earlier than the downstream node. Since all other nodes pointing to the non-root node are located upstream of the non-root node, the distributed energy safety management platform can obtain the existing output of the analysis model for all other nodes in the same batch training process, and summarize the abnormal probability set of the gas pipeline corresponding to the node pointing to the non-root node in the preset future period.

[0211] like Figure 6As shown, the input of the second neural network layer also includes edge features 650 in the distribution map, wherein the edge features 650 in the distribution map refer to the edge features of the edges involved in the non-root nodes corresponding to the second neural network layer in the distribution map. The edge features include the influence of the upstream pipeline on the downstream pipeline.

[0212] The influence of the upstream pipeline on the downstream pipeline refers to the influence of the flow change of the upstream pipeline on the pressure of the downstream pipeline.

[0213] In some embodiments, the influence of the upstream pipeline on the downstream pipeline is related to the fluctuation range of the flow sensor data in the monitoring sequence of the upstream pipeline and the fluctuation range of the pressure sensor data in the monitoring sequence of the downstream pipeline. For example, the influence of the upstream pipeline on the downstream pipeline is inversely proportional to the fluctuation range of the flow sensor data in the monitoring sequence of the upstream pipeline and is directly proportional to the fluctuation range of the pressure sensor data in the monitoring sequence of the downstream pipeline. In some embodiments, the fluctuation range of the flow sensor data and the fluctuation range of the pressure sensor data can be represented by a flow fluctuation value and a pressure fluctuation value, respectively. For relevant descriptions of the flow fluctuation value and the pressure fluctuation value, please refer to Figure 2 Corresponding content.

[0214] In some embodiments, the distributed energy security management platform can determine the influence of the upstream pipeline on the downstream pipeline through a preset formula based on the fluctuation range of the flow sensor data in the monitoring sequence of the upstream pipeline and the fluctuation range of the pressure sensor data in the monitoring sequence of the downstream pipeline. Exemplarily, the preset formula can be expressed by the following formula (4):

[0215] (4)

[0216] in, Indicates the influence of the upstream pipeline on the downstream pipeline, represents the fluctuation range of the pressure sensing data in the monitoring sequence of the downstream pipeline, Represents the fluctuation range of the flow sensor data in the monitoring sequence of the upstream pipeline.

[0217] In some embodiments of the present specification, when predicting the abnormal probability of a pipeline, the accuracy of model prediction can be improved by considering the different degrees of influence of different upstream pipelines on the pipeline corresponding to the node.

[0218] In some embodiments, each neural network layer corresponds to a node one by one, that is, corresponds to each gas pipeline, and the edges of the distribution map represent the connection relationship between the gas pipelines. Based on the edges of the distribution map, the connection relationship between the neural network layers can be determined.

[0219] In some embodiments, the analysis model can be obtained through training in a variety of ways. For example, the distributed energy security management platform can obtain a trained analysis model based on overall training or local training.

[0220] Overall training refers to selecting the neural network layers corresponding to all nodes in the distribution map for training.

[0221] In some embodiments, the overall training of the analysis model can be performed by joint training.

[0222] The distributed energy security management platform can jointly train analysis models based on a large number of labeled training samples.

[0223] In some embodiments, the distributed energy safety management platform can input the first training sample into the initial first neural network layer corresponding to the root node to obtain the output of the initial first neural network layer; input the output of the initial first neural network layer and the second training sample corresponding to the non-root node connected to the root node into the initial second neural network layer corresponding to the non-root node to obtain the output of the initial second neural network layer; if the non-root node has other non-root nodes further connected, then further input the output of the initial second neural network layer and the second training sample corresponding to the other non-root node into the other initial neural network layer corresponding to the other non-root node to obtain the output of the other initial second neural network layer..., and so on, until all nodes that need to be trained for the analysis model are traversed, based on the output of the second neural network layer traversed last, and the second label corresponding to the second neural network layer, the analysis model is iteratively updated based on the loss function, and the training is completed when the preset conditions are met to obtain the trained analysis model. Among them, the preset conditions can be the convergence of the loss function, the number of iterations reaching a threshold, etc.

[0224] In some embodiments, the overall training of the analysis model can also be performed through layered training.

[0225] The distributed energy security management platform can hierarchically train analysis models based on a large number of labeled training samples.

[0226] In some embodiments, the distributed energy safety management platform can input the first training sample into the initial first neural network layer, obtain the output of the initial first neural network layer, construct a loss function based on the output of the initial first neural network layer and the first label, iteratively update the initial first neural network layer based on the loss function, and when the preset conditions are met, the training is completed to obtain a trained first neural network layer.

[0227] In some embodiments, the distributed energy safety management platform can train an initial second neural network layer based on the second training sample and the second label. The training process is similar to the hierarchical training process of the first neural network layer, which will not be repeated here.

[0228] In some embodiments, after the overall training is completed, the distributed energy security management platform can select some of the nodes for further local training as needed.

[0229] Local training refers to selecting the neural network layers corresponding to some nodes in the distribution map for training.

[0230] When conducting local training, the distributed energy security management platform can select some neural network layers in the analysis model for training based on the connection relationship between the neural network layers in the distribution map. For example, the neural network layer corresponding to the node whose sum of the absolute values ​​of the in-degree and out-degree is higher than the preset in-degree threshold is selected for local training. The preset in-degree threshold refers to the maximum value of the sum of the absolute values ​​of the in-degree and out-degree. The preset in-degree threshold can be preset by the staff.

[0231] By further local training of some nodes with relatively complex gas pipeline connection relationships, there is no need to repeatedly train the neural network layer corresponding to the entire distribution map (that is, there is no need to repeatedly train the entire analysis model), which helps save computing resources and improve prediction efficiency.

[0232] In some embodiments, local training of the analysis model can also be performed by joint training or layered training. The specific details can be similar to the corresponding description of the overall training above. The difference is that the training samples of the second neural network layer corresponding to the non-root nodes involved in the local training can also include the outputs of the second neural network corresponding to all nodes pointing to the non-root node obtained during the overall training process.

[0233] In some embodiments, for the first neural network layer, the distributed energy safety management platform may select the historical monitoring sequence of the spot check site corresponding to its corresponding root node as the first training sample, and use whether the gas pipeline corresponding to the group of first training samples actually has abnormalities later as the corresponding first label.

[0234] In some embodiments, for the second neural network layer, the distributed energy safety management platform may select the historical monitoring sequence of the spot check site corresponding to the non-root node and the subsequent actual abnormal situation set of the gas pipeline corresponding to the node pointing to the non-root node as the second training samples, and use whether the gas pipeline corresponding to the group of second training samples actually has abnormalities as the corresponding second label.

[0235] In some embodiments, the second training sample may further include the historical influence of the upstream pipeline of the non-root node on it.

[0236] It can be understood that the training samples of each single-layer neural network are only taken from the historical monitoring sequence of the spot check site corresponding to the node of the single-layer neural network, and do not include the historical monitoring sequence of other spot check sites.

[0237] In some embodiments, the distributed energy safety management platform can also construct a fourth clustering vector corresponding to the first neural network layer for each training sample corresponding to the first neural network layer in the training sample set. The fourth clustering vector includes the historical monitoring sequence of the spot check site corresponding to its corresponding root node. Cluster multiple fourth clustering vectors, and for each cluster cluster obtained by clustering, count the number of abnormalities corresponding to the training samples included therein in a subsequent period of time, and use the ratio of the number to the total number of training samples of this type as the first label corresponding to the first neural network layer. The abnormality in a subsequent period of time means that the gas pipeline corresponding to the spot check site corresponding to the training sample has an abnormality in a subsequent period of time.

[0238] In some embodiments, the distributed energy safety management platform can also construct the fifth clustering vector corresponding to each training sample corresponding to the second neural network layer in the training sample set. The fifth clustering vector includes the historical monitoring sequence of the spot check site corresponding to the non-root node, the subsequent actual abnormal situation set of the gas pipeline corresponding to the node pointing to the non-root node, and the historical influence of the upstream pipeline of the non-root node on it. Cluster multiple fifth clustering vectors, and for each cluster obtained by clustering, count the number of abnormalities corresponding to the training samples included in it in a subsequent period of time, and use the ratio of this number to the total number of training samples of this type as the second label corresponding to the second neural network layer.

[0239] In some embodiments, when training an analysis model based on a training data set, the training data set is used to train the neural network layer corresponding to each node, and the number of training samples in the training data set is greater than the sample size threshold corresponding to each node, and the sample size threshold is related to the in-degree and out-degree of each node.

[0240] The sample size threshold refers to the maximum number of training samples for a single neural network layer, for example, 100 training samples.

[0241] In-degree refers to the number of edges pointing to the node. Out-degree refers to the number of edges starting from the node and pointing to other nodes. In some embodiments, the in-degree and out-degree of a node can be statistically obtained when drawing a distribution graph.

[0242] In some embodiments, the sample size threshold is related to the in-degree and out-degree of the node. For example, the larger the sum of the in-degree and out-degree of the node, the larger the sample size threshold.

[0243] In some embodiments of the present specification, the in-degree and out-degree of the node are also considered when training the analysis model. The training data set can be adjusted according to the actual connection conditions of the gas pipeline corresponding to the node to improve the diversity of the training data set, which helps to train a more accurate model, thereby improving the reliability of the predicted abnormality probability.

[0244] In some embodiments, the training process of the analysis model may include: obtaining a training data set, the training data set including a number of training samples and a label corresponding to each training sample; performing multiple rounds of iterations, at least one round of iteration including: selecting one or more training samples from the training data set, inputting the one or more training samples into the initial analysis model, and obtaining the predicted output of the initial analysis model corresponding to the one or more training samples; substituting the predicted output of the initial analysis model corresponding to the one or more training samples and the labels corresponding to the one or more training samples into the formula of a predefined loss function to calculate the value of the loss function; based on the value of the loss function, reversely updating the model parameters in the initial analysis model; when the iteration end condition is met, ending the iteration to obtain a trained analysis model.

[0245] In some embodiments of the present specification, by reversely updating the model parameters of the analysis model through multiple rounds of iterations, the model prediction value can be made as close to the true value as possible, thereby improving the performance and accuracy of the model.

[0246] In some embodiments of the present specification, the data processing capability and adaptability of the model are utilized to predict the abnormal probability of each gas pipeline respectively through an analysis model composed of multiple neural network layers, which can make the predicted value of the abnormal probability more accurate and reduce the time wasted in determining the abnormal probability of multiple gas pipelines.

[0247] It should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, the alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A distributed energy pipeline natural gas transportation safety management Internet of Things system, characterized by: include: Distributed energy user platform, distributed energy security management platform, distributed energy perception control platform, distributed energy security service platform and distributed energy sensor network platform; wherein the distributed energy perception control platform includes a monitoring perception control platform and a maintenance perception control platform; The monitoring perception control platform is configured to receive monitoring instructions and control the monitoring device and the crawling robot based on the monitoring instructions; The maintenance perception control platform is configured to receive maintenance instructions and deploy maintenance personnel based on the maintenance instructions; The distributed energy security service platform is configured to connect the distributed energy user platform and the distributed energy security management platform for data transmission; The distributed energy security management platform is configured as follows: Acquire a safety management area, wherein a spot check site is set in the safety management area; Based on a preset period, determining whether the monitoring sequence of the spot check site within a preset historical period meets a first preset condition; the first preset condition is related to the number of monitoring data that are not within a normal range in the monitoring sequence; In response to the first preset condition not being met, maintaining the first monitoring instruction of the spot check site; In response to satisfying the first preset condition: Based on the monitoring sequence and the inspection sequence in the preset historical period, adjusting the first monitoring instruction, and sending the adjusted second monitoring instruction to the monitoring perception control platform; Based on the monitoring sequence, the inspection sequence and the adjacency matrix of the gas pipeline, the abnormal probability of the spot check site in a preset future time period is determined: The method comprises: constructing a distribution map of the safety management area according to the monitoring sequence of each spot check site and the adjacency matrix of the gas pipeline; wherein the nodes of the distribution map are gas pipelines, the node features of the distribution map are the monitoring sequence of the spot check site that meets a preset distance condition, and the edges of the distribution map represent the connection relationship between the gas pipelines; According to the distribution map, the abnormal probability of the gas pipeline in the preset future period is determined by an analysis model; wherein the analysis model is a machine learning model; Based on the abnormal probability of the gas pipeline in the preset future time period, the abnormal probability of the spot check site in the preset future time period is determined; wherein the adjacency matrix of the gas pipeline refers to a two-dimensional matrix used to reflect the connection relationship between the various gas pipelines; Based on the abnormal probability, the maintenance instruction is generated, and the maintenance instruction is sent to the maintenance perception control platform to deploy the maintenance personnel.

2. The system according to claim 1, characterized in that The preset period is negatively correlated to the adjustment frequency of the monitoring instruction.

3. The system according to claim 1, characterized in that The distributed energy security management platform is further configured as follows: In response to the abnormal probability being greater than a first preset threshold, the maintenance instruction is generated; wherein the first preset threshold is negatively correlated to the number of gas pipelines in the safety management area.

4. The system according to claim 1, characterized in that The distributed energy security management platform is further configured as follows: Dividing the security management area into transport path segments; Determining the location and initial monitoring instructions of the spot check sites within the transport path segment based on historical inspection data; In response to the random inspection site not satisfying a second preset condition, based on the inspection sequence, the position of the random inspection site and the distance interval at which the crawling robot sends sensor data during inspection are adjusted; wherein the second preset condition is related to the monitoring data of the random inspection site.

5. The system according to claim 1, characterized in that The analysis model includes a first neural network layer and a second neural network layer; the first neural network layer corresponds to the root node in the distribution graph, the second neural network layer corresponds to the non-root node in the distribution graph, and the connection relationship between the neural network layers is determined based on the edge of the distribution graph; The input of the first neural network layer includes the monitoring sequence of the spot check site corresponding to the root node, and the output includes the abnormal probability of the gas pipeline in the preset future time period; The input of the second neural network layer includes the monitoring sequence of the spot check site corresponding to the non-root node, the abnormal probability set of the nodes pointing to the non-root node within the preset future time period, and the output includes the abnormal probability of the gas pipeline within the preset future time period.

6. The system according to claim 5, characterized in that The input of the second neural network layer also includes edge features in the distribution map, and the edge features include the influence of the upstream pipeline on the downstream pipeline.

7. The system according to claim 1, characterized in that The distributed energy security management platform is further configured as follows: Based on the abnormal probability, generating equipment control instructions; Based on the equipment control instructions, the inspection stations and / or the crawling robots of the transport path segments are added and / or deleted.

8. The system according to claim 1, characterized in that The distributed energy security management platform is further configured as follows: Based on the abnormal probability, the second monitoring instruction of the spot check site is adjusted, and the third monitoring instruction obtained after the adjustment is sent to the monitoring perception control platform.

9. A distributed energy pipeline natural gas transportation safety management method, characterized in that: The method is executed by a distributed energy security management platform of a distributed energy pipeline natural gas transmission security management Internet of Things system, and the method includes: Acquire a safety management area, wherein a spot check site is set in the safety management area; Based on a preset period, determining whether the monitoring sequence of the spot check site within a preset historical period meets a first preset condition; the first preset condition is related to the number of monitoring data that are not within a normal range in the monitoring sequence; In response to the first preset condition not being met, maintaining the first monitoring instruction of the spot check site; In response to satisfying the first preset condition: Based on the monitoring sequence and the inspection sequence in the preset historical period, adjusting the first monitoring instruction, and sending the adjusted second monitoring instruction to the monitoring perception control platform; Based on the monitoring sequence, the inspection sequence and the adjacency matrix of the gas pipeline, the abnormal probability of the spot check site in a preset future time period is determined: The method comprises: constructing a distribution map of the safety management area according to the monitoring sequence of each spot check site and the adjacency matrix of the gas pipeline; wherein the nodes of the distribution map are gas pipelines, the node features of the distribution map are the monitoring sequence of the spot check site that meets a preset distance condition, and the edges of the distribution map represent the connection relationship between the gas pipelines; According to the distribution map, the abnormal probability of the gas pipeline in the preset future period is determined by an analysis model; wherein the analysis model is a machine learning model; Based on the abnormal probability of the gas pipeline in the preset future time period, the abnormal probability of the spot check site in the preset future time period is determined; wherein the adjacency matrix of the gas pipeline refers to a two-dimensional matrix used to reflect the connection relationship between the various gas pipelines; Based on the abnormal probability, a maintenance instruction is generated, and the maintenance instruction is sent to the maintenance perception control platform to deploy maintenance personnel.

Images