A smart gas accessory safety monitoring method, Internet of Things system and medium

Abstract

The present invention provides a smart gas accessory safety monitoring method, an Internet of Things system and a medium, and relates to the field of gas pipeline accessory facility monitoring. The method is executed by a smart gas accessory safety monitoring Internet of Things system, and includes: determining a first abnormal value; determining a target acquisition parameter based on the first abnormal value; determining an acquisition instruction based on the target acquisition parameter; determining a second abnormal value; and determining a maintenance instruction and / or a parameter adjustment instruction based on the second abnormal value. The Internet of Things system includes a government safety supervision management platform, a government safety supervision sensor network platform, a government safety supervision object platform, a gas company management platform, a gas company sensor network platform, a gas equipment object platform and a gas pipeline maintenance object platform. The method can be run after being read by a computer instruction stored in a computer-readable storage medium. The method can realize real-time monitoring and intelligent management of accessory components, making the detection results more comprehensive and accurate.

Description

Technical Field

[0001] This specification relates to the field of gas pipeline ancillary facility monitoring, and in particular to a smart gas ancillary component safety monitoring method, Internet of Things system and medium. Background Art

[0002] Promoting the development of smart gas systems is one of the important means to improve urban management efficiency and residents' quality of life. In smart gas systems, the operating status of auxiliary components (such as flanges, valves, compensators, drains, vent pipes, etc.) has an important impact on the safety and stability of the entire gas network. Abnormal conditions of auxiliary components will directly affect the transmission efficiency and safety of gas use.

[0003] The prior art CN110261437B uses infrared detection equipment to obtain infrared thermal images of natural gas station pressure equipment, and uses the infrared thermal images to process and identify the extracted test data, thus realizing an automated, rapid and low-cost defect survey of natural gas station pressure equipment. However, only infrared images are used for processing and identification, without combining multiple data for comprehensive analysis, resulting in limitations in test results and the risk of misjudgment.

[0004] Therefore, a smart gas accessory safety monitoring method and Internet of Things system are provided to combine multiple data to achieve real-time monitoring and intelligent management of accessory components, making the detection results more comprehensive and accurate. Summary of the invention

[0005] In order to solve the problem of how to improve the screening efficiency and accuracy of gas pipeline accessories, this specification provides a smart gas accessory safety monitoring method, Internet of Things system and medium.

[0006] The invention content includes a smart gas accessory component safety monitoring method. The smart gas accessory component safety monitoring method includes: determining a first abnormal value of the accessory component based on the operation data of the accessory component, the accessory component includes at least one of a flange, a valve, a compensator, a drainer and a vent pipe, the accessory component is located on a gas pipeline, and the operation data includes at least one of audio data, temperature data and vibration data; determining a target acquisition parameter based on the first abnormal value; determining an acquisition instruction based on the target acquisition parameter, and sending the acquisition instruction to an interactive device and / or a drone of a staff member to control an infrared detection device to collect infrared data of the accessory component with the target acquisition parameter, the infrared data including an infrared thermal map; determining a second abnormal value of the accessory component based on the infrared data, environmental data and the first abnormal value; and determining a maintenance instruction and / or a parameter adjustment instruction based on the second abnormal value, and sending it to a monitoring component and the interactive device, the maintenance instruction is configured to instruct the staff member to perform manual maintenance on the accessory component, the parameter adjustment instruction is configured to adjust the monitoring parameters of the monitoring component before completing the maintenance, and the monitoring component is configured to monitor the operation data of the accessory component.

[0007] The invention content includes a smart gas accessory component safety monitoring Internet of Things system, the smart gas accessory component safety monitoring Internet of Things system includes a government safety supervision management platform, a government safety supervision sensor network platform, a government safety supervision object platform, a gas company management platform, a gas company sensor network platform, a gas equipment object platform and a gas pipeline maintenance object platform, the smart gas accessory component safety monitoring Internet of Things system is configured as follows: the government safety supervision management platform obtains the operation data of the accessory component through the gas company management platform and the government safety supervision sensor network platform, the accessory component includes at least one of a flange, a valve, a compensator, a drainer, and a vent pipe, the accessory component is located on the gas pipeline, and the operation data includes at least one of audio data, temperature data and vibration data; through the government safety supervision management platform, based on the operation data, determine the first abnormal value of the accessory component; and based on the first abnormal value, determine the target acquisition parameter; and based on the target acquisition parameter, determine the acquisition instruction; send the acquisition instruction to the government The government safety supervision object platform is further sent to the gas pipeline maintenance object platform and / or the gas equipment object platform to control the infrared detection equipment to collect the infrared data of the accessory component with the target acquisition parameters, and the infrared data includes an infrared heat map; environmental data is obtained through the government safety supervision sensor network platform; and the government safety supervision management platform determines the second abnormal value of the accessory component based on the infrared data, the environmental data and the first abnormal value; and based on the second abnormal value, determines the maintenance instruction and / or the parameter adjustment instruction, and sends it to the gas company management platform of the government safety supervision object platform, and the gas company management platform further issues the maintenance instruction and / or the parameter adjustment instruction to the gas equipment object platform and the gas pipeline maintenance object platform, the maintenance instruction is configured to instruct the staff to perform manual maintenance on the accessory component, and the parameter adjustment instruction is configured to adjust the monitoring parameters of the monitoring component before completing the maintenance, and the monitoring component is configured to monitor the operating data of the accessory component.

[0008] The invention content includes a computer-readable storage medium, which stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes a smart gas accessory component safety monitoring method.

[0009] The beneficial effects brought about by the above invention include but are not limited to: (1) By analyzing the operating data of the accessory components, the screening range can be initially narrowed. Then, by obtaining the infrared data of the accessory components after the narrowing down, the abnormal operation conditions are further analyzed, and the abnormal accessory components are screened out. Then, by using drones or staff to conduct inspections, the maintenance screening problem of the accessory components is solved, and the screening efficiency and accuracy are improved. (2) By extracting temperature, audio, and vibration data related to the operation of the accessory components from the operating data, and jointly judging whether the operating status of the accessory components is abnormal from these three aspects, the error caused by considering only a single type of data is reduced, ensuring that the first abnormal value obtained by measurement can more accurately reflect the actual operation of the accessory components. (3) By determining the temperature center and its center characteristics, the temperature data with temperature abnormalities caused by the operation of the accessory components themselves can be screened out, thereby eliminating the interference data in the infrared data. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 It is a platform structure diagram of the smart gas accessory safety monitoring Internet of Things system shown in some embodiments of this specification;

[0011] Figure 2 is an exemplary flow chart of a method for safety monitoring of smart gas accessory components according to some embodiments of this specification;

[0012] Figure 3 is an exemplary flow chart for determining a first abnormal value according to some embodiments of this specification;

[0013] Figure 4 is a schematic diagram of a process for determining a second abnormal value according to some embodiments of this specification;

[0014] Figure 5 is an exemplary schematic diagram of determining a temperature center and its center characteristics according to some embodiments of the present specification. DETAILED DESCRIPTION

[0015] The following is a brief introduction to the drawings required for describing the embodiments. The drawings do not represent all implementation methods.

[0016] The "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0017] Unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0018] Figure 1 It is a platform structure diagram of the smart gas accessory safety monitoring Internet of Things system shown in some embodiments of this specification.

[0019] like Figure 1 As shown, the smart gas accessory safety monitoring Internet of Things system 100 may include a government safety supervision management platform 110, a government safety supervision sensor network platform 120, a government safety supervision object platform 130, a gas company management platform 140, a gas company sensor network platform 150, a gas equipment object platform 160 and a gas pipeline maintenance object platform 170.

[0020] The government security supervision management platform 110 refers to a comprehensive management platform for government management information.

[0021] In some embodiments, the government safety supervision management platform 110 may be configured to process and store safety monitoring data of smart gas accessory components. For example, the government safety supervision management platform 110 may determine a first abnormal value of an accessory component based on operating data. For another example, the government safety supervision management platform 110 may determine a target acquisition parameter based on the first abnormal value. For another example, the government safety supervision management platform 110 may determine a second abnormal value of an accessory component based on infrared data, environmental data, and the first abnormal value.

[0022] In some embodiments, the government security supervision management platform 110 may interact with the government security supervision sensor network platform 120 .

[0023] The government safety supervision sensor network platform 120 may be a functional platform for managing sensor communication. In some embodiments, the government safety supervision sensor network platform 120 may realize the functions of sensing information sensor communication and controlling information sensor communication. In some embodiments, the government safety supervision sensor network platform 120 may interact with the government safety supervision management platform 110 and the gas company management platform 140. For example, the government safety supervision management platform 110 may obtain operation data from the gas company management platform 140 based on the government safety supervision sensor network platform 120.

[0024] In some embodiments, the government safety regulatory sensor network platform 120 can obtain environmental data.

[0025] The government security supervision object platform 130 is a functional platform for monitoring and managing the execution of instructions.

[0026] The gas company management platform 140 is a platform for managing gas company and gas pipeline related parameters. In some embodiments, the gas company management platform 140 can interact with the gas company sensor network platform 150 and the government safety supervision sensor network platform 120.

[0027] The gas company sensor network platform 150 may be a functional platform for managing sensor communication. In some embodiments, the gas company sensor network platform 150 may implement the functions of sensing information sensor communication and controlling information sensor communication. In some embodiments, the gas company sensor network platform 150 may interact with the gas company management platform 140, the gas equipment object platform 160, and the gas pipeline maintenance object platform 170.

[0028] The gas equipment object platform 160 may be a functional platform for sensing the generation of gas equipment information and controlling the execution of gas equipment information.

[0029] In some embodiments, the gas equipment object platform 160 may be configured to monitor components, drones, and infrared detection equipment.

[0030] In some embodiments, the gas equipment object platform 160 may upload the operating data of the accessory components to the government safety supervision management platform 110 via the government safety supervision sensor network platform 120 .

[0031] The gas pipeline maintenance object platform 170 may be a functional platform for sensing the generation of gas user information and controlling the execution of gas user information.

[0032] In some embodiments, the gas pipeline maintenance object platform 170 may be configured as an interactive device for workers.

[0033] In some embodiments, the platform in the smart gas accessory safety monitoring IoT system 100 can be divided into a smart gas primary network and a smart gas secondary network. Among them, the smart gas primary network refers to the network where government users supervise the operation of the gas pipeline network, and the smart gas secondary network refers to the network where the gas pipeline network is operated. In some embodiments, the same platform in the smart gas accessory safety monitoring IoT system 100 can play different roles in the smart gas primary network and the smart gas secondary network.

[0034] In some embodiments, the smart gas primary network may include a smart gas primary network management platform, a smart gas primary network sensor network platform, and a smart gas primary network object platform. The smart gas primary network management platform may include a government safety supervision management platform 110, the smart gas primary network sensor network platform may include a government safety supervision sensor network platform 120, and the smart gas primary network object platform may include a government safety supervision object platform 130.

[0035] In some embodiments, the smart gas secondary network may include a smart gas secondary network management platform, a smart gas secondary network sensor network platform, and a smart gas secondary network object platform. The smart gas secondary network management platform may include a gas company management platform 140, the smart gas secondary network sensor network platform may include a gas company sensor network platform 150, and the smart gas secondary network object platform may include a gas equipment object platform 160 and / or a gas pipeline maintenance object platform 170.

[0036] For more details about accessory components, operating data, first abnormal value, target acquisition parameters, acquisition instructions, second abnormal value, maintenance instructions and parameter adjustment instructions, see Figure 2-Figure 5 and its related description.

[0037] In some embodiments of the present specification, the smart gas accessory safety monitoring Internet of Things system 100 can form an information operation closed loop between various platforms of the Internet of Things system, and coordinate and operate regularly under the unified management of the government safety supervision and management platform 110, thereby realizing the informatization and intelligence of the safety monitoring and maintenance management of gas accessories.

[0038] Figure 2 is an exemplary flow chart of a method for safety monitoring of smart gas accessory components according to some embodiments of this specification. Figure 2 As shown, the process 200 includes the following steps. In some embodiments, the process 200 can be executed by a smart gas accessory component safety monitoring Internet of Things system.

[0039] Step 210 : determining a first abnormal value of the accessory component based on the operating data of the accessory component.

[0040] The accessory components refer to equipment that ensures the safe operation of the smart gas accessory component safety monitoring Internet of Things system 100 and facilitates maintenance and overhaul. For example, the accessory components may include but are not limited to valves, compensators, drain pipes, vent pipes, etc.

[0041] In some embodiments, in order to ensure the safe operation of the pipeline network and take into account the needs of maintenance and wiring, auxiliary components can be located on the gas pipeline.

[0042] Operation data refers to data collected by the accessory component during operation. For example, the operation data may include but is not limited to audio data, temperature data, vibration data, etc.

[0043] Audio data refers to the sound signal generated by the accessory component when it is running. In some embodiments, the audio data can be represented by a signal wave.

[0044] Temperature data refers to the temperature of the accessory component when it is running. In some embodiments, the temperature data may include the operating temperature of the accessory component at at least one time. For example, the temperature data may be expressed as [(T1, t1), (T2, t2), ... (T n , t n )] represents. Among them, t n is the nth moment, T n is the operating temperature of the accessory component at the nth moment, where n is a positive integer.

[0045] Vibration data refers to the vibration signal generated when the accessory component is running. In some embodiments, the vibration data can be represented by a signal wave.

[0046] In some embodiments, the operation data of the accessory components can be obtained in a variety of ways. For example, the government safety supervision management platform 110 can obtain the operation data through the monitoring component. Among them, the monitoring component refers to a device or sensor used to monitor the operation status of the accessory components in real time in the smart gas accessory component safety monitoring Internet of Things system 100. The monitoring component may include an audio data acquisition device, a temperature data acquisition device, and a vibration data acquisition device.

[0047] The first abnormal value refers to a numerical value, a letter, etc. used to reflect the degree of abnormal operation of the accessory component. For example, the first abnormal value can be represented by a numerical value less than 1.

[0048] In some embodiments, the higher the first abnormal value is, the higher the possibility that an accident occurs in the pipeline where the accessory component is located.

[0049] In some embodiments, the government security supervision management platform 110 may determine the first abnormal value based on the operation data of the auxiliary component in a variety of ways. For example, the government security supervision management platform 110 may determine the first abnormal value based on the following ways:

[0050] S211, the government security supervision management platform 110 determines reference operation data based on the attached components.

[0051] The reference operation data refers to the operation data collected by the accessory component under normal operation. For example, the reference transportation data may include but is not limited to reference audio data, reference temperature data, reference vibration data, etc.

[0052] In some embodiments, the reference operating data may be determined from historical data.

[0053] S212, calculating the similarity between the operating data of the accessory component and the corresponding reference operating data.

[0054] In some embodiments, the government safety supervision management platform 110 can quantize the reference operation data and operation data of the accessory components into vectors. For example, since the temperature data is sequence data, the government safety supervision management platform 110 can directly quantize it into a vector. For another example, the audio data is a continuous signal wave, and the government safety supervision management platform 110 can randomly sample the signal wave and select a series of sampling points to construct a vector containing multiple sampling times and their corresponding sound signals. The vibration data quantization method is similar to that of the audio data and will not be repeated here.

[0055] In some embodiments, the similarity between the reference operating data and the operating data may be negatively correlated to the vector distance between the reference operating data and the operating data, and the vector distance may be determined based on the cosine distance, etc. In some embodiments, the government safety supervision management platform 110 may calculate the vector distance between the reference operating data and the operating data, and then determine the similarity between the reference operating data and the operating data by looking up a table.

[0056] The similarity refers to the similarity between the reference operation data and the actual operation data of the accessory component. For example, the similarity can be a value between 0 and 1, wherein the closer the value is to 1, the higher the similarity.

[0057] S213, the first abnormal value of the subsidiary component=1-the maximum value of the similarity.

[0058] It can be understood that if the number of reference operation data is multiple, the number of corresponding similarities is also multiple. Because the greater the similarity, the greater the possibility of normal operation, and the smaller the first abnormal value, therefore, the first abnormal value can be determined based on the maximum value of the similarity.

[0059] In some embodiments, the government safety supervision management platform 110 may also determine a reference interval based on the operation data; determine an operation characteristic based on the reference interval; determine an operation intensity based on the operation characteristic; and determine a first abnormal value based on the operation intensity. For more information about reference intervals, operation characteristics, operation intensity, and how to determine the first abnormal value, see Figure 3 and its related description.

[0060] Step 220: Determine target acquisition parameters based on the first abnormal value.

[0061] The target acquisition parameters refer to the order in which infrared data are collected in the field.

[0062] In some embodiments, the government security supervision management platform 110 can determine the target acquisition parameters in a variety of ways. For example, the government security supervision management platform 110 can filter out the auxiliary components whose first abnormal value is greater than the first preset threshold, and arrange them according to the size of the first abnormal value, and use the arrangement order as the acquisition order to obtain the target acquisition parameters.

[0063] The first preset threshold refers to a threshold used to determine whether the first abnormal value exceeds the normal safety standard. In some embodiments, the first preset threshold can be set by professional technicians or by system default. For example, the first preset threshold can be 10%.

[0064] Step 230, based on the target acquisition parameters, determine the acquisition instructions, and send the acquisition instructions to the staff's interactive device and / or the drone to control the infrared detection device to collect infrared data of the accessory components.

[0065] The collection instruction refers to the instruction for collecting infrared data that can be recognized by the smart gas accessory safety monitoring Internet of Things system 100.

[0066] In some embodiments, the government security supervision management platform 110 may encode the target acquisition parameters to obtain acquisition instructions, wherein the encoding may include but is not limited to binary code, ASCII code, etc.

[0067] In some embodiments, the government safety supervision management platform 110 can send the collection instruction to the government safety supervision object platform 130, and further send it to the staff's interactive device. In response to receiving the notification from the interactive device, the staff operates the infrared detection device to detect the infrared data of the accessory components and manually uploads the infrared data to the smart gas accessory component safety monitoring Internet of Things system 100.

[0068] In some embodiments, the government safety supervision management platform 110 can also send a collection instruction to the gas equipment object platform 160, and further send it to the drone to control the infrared detection device to collect infrared data of the accessory components with the target collection parameters. For example, the infrared detection device can be mounted on the drone, and the infrared data detection of the accessory components can be realized by controlling the drone or the autonomous movement of the drone, and the infrared data can be sent to the smart gas accessory component safety monitoring Internet of Things system 100 through copying or wireless network.

[0069] The interactive device refers to a tool that enables the smart gas accessory safety monitoring IoT system 100 to exchange information with the user. For example, the interactive device may be an application on a staff member's work device (such as a mobile phone, computer, etc.).

[0070] Infrared detection equipment refers to instruments used to detect infrared radiation (i.e. thermal radiation). For example, infrared detection equipment can be infrared thermal imagers, infrared cameras, etc.

[0071] Infrared data refers to infrared radiation information collected by infrared detection equipment, which is used to analyze and monitor the thermal distribution, temperature changes and potential thermal anomalies of attached components. For example, infrared data may include but is not limited to infrared heat maps.

[0072] Step 240 , determining a second abnormal value of the accessory component based on the infrared data, the environmental data and the first abnormal value.

[0073] Environmental data refers to data related to the environment in which the accessory is located. For example, environmental data may include but is not limited to ambient temperature, humidity, ambient dust concentration, and ventilation level.

[0074] In some embodiments, the government safety supervision management platform 110 may acquire environmental data based on the government safety supervision sensor network platform 120 .

[0075] The second abnormal value refers to a numerical value or letter etc. after adjusting the first abnormal value of the accessory component, and is used to reflect the degree of abnormal operation of the accessory component. For example, the second abnormal value can be represented by a numerical value less than 1.

[0076] In some embodiments, the government safety supervision management platform 110 can determine the second abnormal value in a variety of ways based on the infrared data, environmental data and the first abnormal value.

[0077] In some embodiments, the government safety supervision management platform 110 can perform clustering based on infrared data, environmental data, first abnormal values ​​in historical data, and subsequent actual second abnormal values ​​to obtain multiple cluster centers. The types of clustering algorithms can include multiple types, for example, clustering algorithms can include K-Means (K mean) clustering, density-based clustering method (DBSCAN), etc. There is no limitation on the types of clustering algorithms.

[0078] In some embodiments, the government safety supervision management platform 110 can construct a standard vector based on the infrared data, environmental data, and first abnormal value corresponding to the cluster center. The standard second abnormal value corresponding to the standard vector is the actual accident frequency of the corresponding accessory component in the historical period.

[0079] When the second abnormal value of the auxiliary component needs to be determined, a vector to be matched is constructed according to the current infrared data, environmental data and the first abnormal value of the auxiliary component.

[0080] In some embodiments, the second outlier corresponding to the vector to be matched is determined by calculating the similarity between the vector to be matched and the standard vector. For example, the second outlier corresponding to the vector to be matched is the standard second outlier with the highest similarity to the standard vector, that is, the shortest vector distance. The similarity between the reference vector and the feature vector may be negatively correlated to the vector distance between the reference vector and the feature vector, and the vector distance may be determined based on the cosine distance, etc. For example, the similarity may be the inverse of the vector distance.

[0081] In some embodiments, the government safety supervision management platform 110 can also determine the temperature center and its center characteristics of the infrared data based on the infrared data; determine the temperature characteristics of the auxiliary components based on the temperature center and its center characteristics; and determine the second abnormal value based on the temperature characteristics, environmental data and the first abnormal value. For more information about the temperature center and its center characteristics, temperature characteristics, and how to determine the second abnormal value, please refer to Figure 4 and its related description.

[0082] Step 250: Based on the second abnormal value, a maintenance instruction and / or a parameter adjustment instruction is determined and sent to the monitoring component and the interactive device.

[0083] The maintenance instruction refers to an instruction for performing maintenance on an accessory component. In some embodiments, the maintenance instruction can be used to instruct a worker to perform manual maintenance on an accessory component.

[0084] In some embodiments, the government safety supervision management platform 110 may determine the accessory component whose second abnormal value is greater than the second preset threshold as a component that needs to be repaired, and integrate the information of the accessory component that needs to be repaired to generate a maintenance instruction that can be recognized by the smart gas accessory component safety monitoring Internet of Things system 100. For example, the maintenance instruction may be text-based, binary-coded, or in other machine-readable formats. In some embodiments, the maintenance instruction includes specific information of the accessory component that needs to be repaired, such as the accessory component number, location, etc.

[0085] The second preset threshold refers to a threshold used to determine whether the accessory component needs to be repaired. In some embodiments, the second preset threshold can be set by a professional technician or by the system by default. For example, the second preset threshold can be 5%.

[0086] The parameter adjustment instruction is an instruction used to adjust the monitoring parameters of the monitoring component to optimize the monitoring effect before completing the maintenance. For more information about the monitoring component, please refer to the above and its related description.

[0087] Monitoring parameters refer to various parameters used to adjust the monitoring effect of the monitoring component. For example, monitoring parameters may include but are not limited to monitoring frequency.

[0088] In some embodiments, the government security supervision management platform 110 may determine the parameter adjustment instruction based on the second abnormal value by querying the first preset relationship table.

[0089] In some embodiments, the first preset relationship table may include a correspondence between the second abnormal value of the accessory component and the parameter adjustment instruction. In some embodiments, for a second abnormal value, the government safety supervision management platform 110 may monitor the second abnormal value at different monitoring frequencies, and calculate the monitoring frequency corresponding to the maximum reduction of the second abnormal value, thereby constructing the first preset relationship table.

[0090] In some embodiments, the government safety supervision management platform 110 can send maintenance instructions and / or parameter adjustment instructions to the gas company management platform 140 of the government safety supervision object platform 130, and the gas company management platform 140 can further issue maintenance instructions and / or parameter adjustment instructions to the monitoring components of the gas equipment object platform 160 and the interactive devices of the gas pipeline maintenance object platform 170.

[0091] In some embodiments of the present specification, the screening range can be initially narrowed by analyzing the operating data of the accessory components. Then, by acquiring the infrared data of the accessory components after the narrowing down, the abnormal operation is further analyzed, and the abnormal accessory components are screened out. Then, the maintenance screening problem of the accessory components is solved by inspection by drones or staff, and the screening efficiency and accuracy are improved. At the same time, this method reduces the consumption of manpower and material resources while ensuring the safety of gas use.

[0092] It should be noted that the above description of the process 200 is only for example and illustration, and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to the process 200 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

[0093] Figure 3 is an exemplary flow chart for determining a first abnormal value according to some embodiments of this specification. Figure 3 As shown, the process 300 includes the following steps: In some embodiments, the process 300 may be executed by the government security supervision management platform 110 .

[0094] Step 310, determining a reference interval based on the operating data.

[0095] The reference interval refers to the range interval where the reference operation data is located. For example, the reference interval may include but is not limited to a reference audio interval, a reference temperature interval, and a reference vibration interval.

[0096] Among them, the reference audio interval is the interval range where the audio data is located; the reference temperature interval refers to the interval range where the temperature data is located; and the reference vibration interval refers to the interval where the vibration data is located.

[0097] In some embodiments, since different environments and working conditions will generate their own unique reference audio data, multiple sections of reference audio data can be obtained. In some embodiments, the multiple sections of reference audio data can be stored in a database.

[0098] In some embodiments, the government security supervision management platform 110 can calculate the sequence similarity between the current audio data of the auxiliary component and multiple reference audio data segments, and determine the reference audio data with the largest sequence similarity as the reference audio interval. The specific steps are as follows:

[0099] S311, for each section of reference audio data, a sliding selection is performed according to a sliding window of a preset step length, thereby obtaining multiple candidate reference audio intervals. For each candidate reference audio interval, the sequence similarity between the audio data and the candidate reference audio interval is calculated. Wherein, the length of the sliding window, the length of the candidate reference audio interval and the length of the audio data are all the same. In some embodiments, the method for calculating the similarity may include but is not limited to the Manhattan distance method, the SVM method, etc.

[0100] S312, perform the above operation on multiple reference audio data to obtain multiple sequence similarities. Select the candidate reference audio interval with the largest sequence similarity and determine it as the reference audio interval. For example, assuming there are 20 reference audio data, each audio data can be divided into 30 sliding windows according to the preset step size and sliding window size, then a total of 20×30 comparisons are performed to obtain 600 sequence similarities, and then the largest corresponding candidate reference audio interval is found from the 600 sequence similarities as the reference audio interval.

[0101] Sequence similarity refers to the similarity between audio data and reference audio data.

[0102] In some embodiments, the similarity between the audio data and the reference audio data may be negatively correlated to the vector distance between the audio data and the reference audio data, and the vector distance may be determined based on the cosine distance, etc. For example, the similarity may be the inverse of the vector distance.

[0103] The method for determining the reference temperature range and the reference vibration range is similar to the method for determining the reference audio range, and will not be described in detail here.

[0104] Step 320, determining the operating characteristics based on the reference interval.

[0105] The operation characteristics refer to characteristics used to describe the abnormal operation state of the accessory component. For example, the operation characteristics may include but are not limited to audio abnormality vectors, temperature abnormality vectors, vibration abnormality vectors, etc.

[0106] The audio anomaly vector is a vector consisting of abnormal audio difference values ​​and their time points. For example, the audio anomaly vector can be represented by [(t1, a1), (t2, a2), … (t n , a n )], where t1, t2, ..., t n is the time point of the audio anomaly, a1, a2, ..., a n is the audio difference value at the corresponding time point.

[0107] The audio difference value refers to the difference between the audio data when the accessory component operates abnormally and the reference audio data.

[0108] In some embodiments, the government security supervision management platform 110 can construct an audio anomaly vector based on the following method: since the length of the audio data is the same as the length of the reference audio interval, the number of time points in the audio data and the reference audio interval is the same. The audio data and the reference audio interval are compared in order at each time point, and the time point at which the audio difference value is greater than the third preset threshold is determined as the audio anomaly time point, and the difference value of the time point is determined as the audio difference value of the audio anomaly time point. Thus, multiple audio anomaly time points and their corresponding audio difference values ​​are combined into a vector to obtain an audio anomaly vector.

[0109] The third preset threshold refers to a threshold used to determine whether the audio data is abnormal. In some embodiments, the third preset threshold can be set by professional technicians or by system default.

[0110] The temperature anomaly vector refers to a vector composed of abnormal temperature difference values ​​and their time points; the vibration anomaly vector refers to a vector composed of abnormal vibration difference values ​​and their time points.

[0111] The method of constructing the temperature anomaly vector and the vibration anomaly vector is similar to that of the audio anomaly vector and will not be described in detail here.

[0112] Step 330, determining the operation intensity based on the operation characteristics.

[0113] The operation intensity refers to a value that can reflect the degree of abnormal operation of the accessory component. For example, the operation intensity may include but is not limited to the abnormal audio intensity, abnormal temperature intensity, and abnormal vibration intensity.

[0114] The audio abnormality intensity refers to the intensity value related to the audio data when the accessory components are operating abnormally; the temperature abnormality intensity refers to the intensity value related to the temperature data when the accessory components are operating abnormally; the vibration abnormality intensity refers to the intensity value related to the vibration data when the accessory components are operating abnormally.

[0115] In some embodiments, the government security supervision management platform 110 can calculate the audio abnormality intensity by the following formula (1) based on the relationship that the audio abnormality intensity is positively correlated with the number of audio abnormality time points in the reference audio interval and the average value of the audio abnormality value in the reference audio interval:

[0116] (1)

[0117] in, Abnormal audio intensity, is the weight coefficient of the number of abnormal audio time points in the reference audio interval, is the weight coefficient of the average value of the audio difference value, is the number of abnormal audio time points in the reference audio interval, is the average value of the audio anomalies in the reference audio interval. In some embodiments, , It is a positive number and can be obtained by default.

[0118] In some embodiments, the government safety supervision management platform 110 can calculate the temperature anomaly intensity by the following formula (2) based on the relationship that the temperature anomaly intensity is positively correlated with the number of temperature anomaly time points in the reference temperature interval and the average value of the temperature anomaly value in the reference temperature interval:

[0119] (2)

[0120] in, is the temperature anomaly intensity, is the weight coefficient of the number of abnormal temperature time points in the reference temperature interval, is the weight coefficient of the average value of temperature difference, is the number of abnormal temperature time points in the reference temperature interval, is the average value of the temperature anomaly values ​​in the reference temperature interval. In some embodiments, , It is a positive number and can be obtained by default.

[0121] In some embodiments, the government safety supervision management platform 110 can calculate the abnormal vibration intensity by the following formula (3) based on the relationship that the abnormal vibration intensity is positively correlated with the number of abnormal vibration time points in the reference vibration interval and the average value of the abnormal vibration value in the reference vibration interval:

[0122] (3)

[0123] in, Abnormal vibration intensity, is the weight coefficient of the number of abnormal vibration time points in the reference vibration interval, is the weight coefficient of the average value of the vibration difference value, is the number of abnormal vibration time points in the reference vibration interval, is the average value of the vibration abnormality values ​​in the reference vibration interval. In some embodiments, , It is a positive number and can be obtained by default.

[0124] Step 340: determining a first abnormal value based on the operating intensity.

[0125] In some embodiments, the government safety supervision management platform 110 may also determine the first abnormal value based on the operation intensity in a variety of ways. For example, the government safety supervision management platform 110 may calculate the first abnormal value based on the positive correlation between the first abnormal value and the audio abnormal intensity, the temperature abnormal intensity, and the vibration abnormal intensity by the following formula (4):

[0126] (4)

[0127] in, is the first outlier, Abnormal audio intensity, is the temperature anomaly intensity, Abnormal vibration intensity, , , They are the weight coefficients of audio abnormality intensity, temperature abnormality intensity, and vibration abnormality intensity respectively.

[0128] In some embodiments, , , It can be preset based on historical experience data. For example, the government safety supervision management platform 110 can show based on the experience knowledge obtained from historical data that if the vibration abnormality has a higher accuracy in determining the fault, then its corresponding weight coefficient larger to enhance its importance in determining faults.

[0129] In some embodiments, the government safety supervision management platform 110 may determine the amplification coefficient based on a reference audio interval, a reference temperature interval, and a reference vibration interval.

[0130] The increase coefficient refers to a value used to measure the similarity between operating characteristics and reference operating data within a specific time period.

[0131] In some embodiments, the government safety supervision management platform 110 may also determine the amplification coefficient based on the negative correlation between the amplification coefficient and the average value of the distance between two intervals in the reference audio interval, the reference temperature interval, and the reference vibration interval by the following formula (5):

[0132] (5)

[0133] in, is the amplification factor, is the average value of the distance between two intervals in the reference audio interval, reference temperature interval, and reference vibration interval. is a weight coefficient. In some embodiments, It is a positive number and can be obtained by default.

[0134] Understandably, due to , are all positive numbers, so It is greater than 1, so it can play an amplifying role.

[0135] In some embodiments, the distance between any two intervals in the reference audio interval, the reference temperature interval, and the reference vibration interval refers to the time difference between any two intervals in the reference audio interval, the reference temperature interval, and the reference vibration interval.

[0136] For example, as shown in Table 1, when the reference audio interval is [a y1 , a y2 , a y3 ], the reference temperature range is [b y7 , b y8 , b y9 ], the reference vibration range is [c y5 , c y6 , c y7 ], then the distance between the reference audio interval and the reference temperature interval is t y7 -t y1 , the distance between the reference audio interval and the reference vibration interval is t y5 -t y1 , the distance between the reference temperature range and the reference vibration range is t y7 -t y5 .

[0137] Table 1

[0138]

[0139] The time interval refers to the interval formed by the time of collecting the operation data. In some embodiments, the size of the time interval can be the actual time interval where the reference audio interval, the reference temperature interval and the reference vibration interval are located.

[0140] In some embodiments, the government safety supervision management platform 110 may also determine the first abnormal value by the following formula (6) based on the positive correlation between the first abnormal value and the operation intensity and the increase coefficient:

[0141] (6)

[0142] in, is the first outlier, is the amplification factor, Abnormal audio intensity, is the temperature anomaly intensity, Abnormal vibration intensity, , , They are the weight coefficients of audio anomaly intensity, temperature anomaly intensity, and vibration anomaly intensity respectively.

[0143] In some embodiments, , , It can be set by professional technicians or by system default.

[0144] In some embodiments, the reference audio interval, the reference temperature interval, and the reference vibration interval overlap (for example, when the reference audio interval is [a y1 , a y2 , a y3 ], the reference temperature range is [b y1 , b y2 , b y3 ], the reference vibration range is [c y1 , c y2 , c y3 ]), it means that the current operation characteristics are most suitable for the overlap time interval (i.e., t y1 ~t y3 ), the abnormal audio intensity, abnormal temperature intensity, and abnormal vibration intensity can reflect the current operating characteristics and the time interval (i.e., t y1 ~t y3 ) degree of difference.

[0145] In some embodiments, in response to the possible causal relationship between the abnormal audio intensity, abnormal temperature intensity, and abnormal vibration intensity (for example, abnormal temperature may cause abnormal audio and vibration, or a specific reason may cause these three abnormalities at the same time), the smart gas accessory safety monitoring IoT system 100 has a greater possibility of real failure. Therefore, by introducing an amplification coefficient, a greater weight is given to the abnormal audio intensity, abnormal temperature intensity, and abnormal vibration intensity, and a certain degree of amplification can be performed. And the degree of amplification should match the degree of overlap of the reference audio interval, the reference temperature interval, and the reference vibration interval.

[0146] In some embodiments of the present specification, the government safety supervision management platform 110 can reflect the correlation between the abnormal audio intensity, the abnormal temperature intensity, and the abnormal vibration intensity by considering the amplification coefficient, and can more effectively identify real faults rather than false positives caused by errors in monitoring data. In addition, the government safety supervision management platform 110 can further improve the accuracy and reliability of fault monitoring by introducing the amplification coefficient to determine the first abnormal value.

[0147] In some embodiments, the government security supervision management platform 110 may determine the first abnormal value based on the operation intensity and the impact value.

[0148] The impact value is a numerical value used to measure the degree of harm caused by abnormalities in attached components.

[0149] In some embodiments, the government security supervision management platform 110 may determine the impact value of the subsidiary component based on querying the second preset relationship table.

[0150] In some embodiments, the second preset relationship table may include a corresponding relationship between the subsidiary components and the impact values. In some embodiments, the second preset relationship table may be constructed based on historical data.

[0151] In some embodiments, the impact value is related to the pipeline parameters of the reference pipeline and the number of upstream and downstream users of the reference pipeline.

[0152] The reference pipe is the gas pipe where the accessory components are located.

[0153] Pipeline parameters refer to relevant parameters of the gas pipeline. For example, pipeline parameters may include but are not limited to the diameter of the pipeline, the flatness of the pipeline, the service life of the pipeline, etc.

[0154] The smoothness of the pipeline refers to a numerical value used to reflect the impact and influence of gas flow on the attached components.

[0155] In some embodiments, the government safety supervision management platform 110 can measure the inclination angle of the pipeline when constructing the pipeline, and obtain the average impact force on the pipeline wall by performing multiple experiments on the pipeline at the inclination angle, thereby obtaining the flatness of the pipeline.

[0156] In some embodiments, the government safety supervision management platform 110 may determine the impact value based on a variety of methods. For example, the government safety supervision management platform 110 may determine the impact value based on pipeline parameters and the number of upstream and downstream users of the reference pipeline.

[0157] In some embodiments, the government safety supervision management platform 110 can perform clustering based on the pipeline parameters in the historical data, the number of upstream and downstream users of the reference pipeline, and the actual impact value of the candidate to obtain multiple cluster centers. The types of clustering algorithms can include multiple types, for example, the clustering algorithm can include K-Means (K mean) clustering, density-based clustering method (DBSCAN), etc. There is no limitation on the type of clustering algorithm.

[0158] In some embodiments, the government safety supervision management platform 110 can construct multiple standard vectors based on the pipeline parameters of the subsidiary components corresponding to the cluster center and the number of users upstream and downstream of the reference pipeline. The standard impact value corresponding to the standard vector is the number of people affected by the accident related to the corresponding subsidiary component in the subsequent period of time multiplied by the length of time affected by the accident to determine the impact value; if there are multiple accidents, the number of people affected by all accidents multiplied by the length of time affected by the accident is added up, and the sum is the impact value.

[0159] Among them, the length of time an accident affects refers to the time from the occurrence of the accident to its complete resolution.

[0160] In some embodiments, the government safety supervision management platform 110 constructs a vector to be matched based on the pipeline parameters of the current attached component and the number of users upstream and downstream of the reference pipeline.

[0161] In some embodiments, the second outlier influence value corresponding to the vector to be matched is determined by calculating the similarity between the vector to be matched and a plurality of standard vectors. For example, the standard influence value corresponding to the vector with the highest similarity to the standard vector is used as the influence value corresponding to the vector to be matched. The similarity between the reference vector and the feature vector may be negatively correlated to the vector distance between the reference vector and the feature vector, and the vector distance may be determined based on the cosine distance, etc. For example, the similarity may be the inverse of the vector distance.

[0162] In some embodiments of the present specification, the government safety supervision and management platform 110 takes into account the relevant parameters of the gas pipeline where the accessory is located (such as diameter, flatness, and service life) and the number of users upstream and downstream of the pipeline. The impact value can more comprehensively reflect the importance and potential impact of the accessory, thereby helping to give priority to those accessories that have a greater impact on the safety and users of the smart gas accessory safety monitoring Internet of Things system 100 when determining the collection order and priority maintenance, thereby further improving the overall operating efficiency and safety of the smart gas accessory safety monitoring Internet of Things system 100.

[0163] In some embodiments, the government safety supervision management platform 110 may determine the first abnormal value by the following formula (7) based on the positive correlation between the first abnormal value and the operation intensity and the impact value:

[0164] (7)

[0165] in, is the first outlier, Abnormal audio intensity, is the temperature anomaly intensity, Abnormal vibration intensity, is the impact value, , , , They are the weight coefficients of audio anomaly intensity, temperature anomaly intensity, vibration anomaly intensity and impact value respectively.

[0166] In some embodiments, , , , It can be set by professional technicians or by system default.

[0167] In some embodiments, the impact value is more important than the first abnormal value. For example, in response to the impact value of accessory A being 1, the impact value of accessory B is 10. Even if accessory A fails, since the first abnormal value of accessory A is low, the possibility of an accident occurring in the pipeline where accessory A is located is low. On the contrary, even if accessory B is only slightly damaged, but the first abnormal value of accessory B is high, the possibility of an accident occurring in the pipeline where accessory B is located is high.

[0168] In some embodiments of the present specification, the government safety supervision management platform 110 can ensure the comparability of the first abnormal values ​​between accessories of different importance by considering the impact values ​​of the accessories, which in turn helps to determine the collection sequence and the maintenance sequence, thereby improving the reliability and safety of the operation of the smart gas accessory safety monitoring Internet of Things system 100.

[0169] In some embodiments of the present specification, the government safety supervision management platform 110 extracts temperature, audio, and vibration data related to the operation of the accessory components from the operating data, and jointly determines whether there is any abnormality in the operating status of the accessory components from these three aspects, thereby reducing the error caused by considering only a single type of data, and ensuring that the measured first abnormal value can more accurately reflect the actual operating conditions of the accessory components.

[0170] It should be noted that the above description of process 300 is only for example and illustration, and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to process 300 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

[0171] Figure 4 FIG. 1 is a schematic diagram of a process for determining a second abnormal value according to some embodiments of this specification. Figure 4 As shown, the process 400 includes the following steps: In some embodiments, the process 400 may be executed by the government security supervision management platform 110 .

[0172] Step 410, based on the infrared data, determine the temperature center and its center characteristics of the infrared data.

[0173] The temperature center refers to the center where the temperature is locally highest or lowest on the infrared thermal image. The temperature of the temperature center is the same. For example, Figure 5 As shown, the temperature center can include a low temperature center of 4°C and a high temperature center of 28°C.

[0174] In some embodiments, the government safety supervision management platform 110 can perform isotherm analysis on the infrared data to obtain the following: Figure 5 The same infrared thermal image can then be used to obtain multiple temperature centers.

[0175] Isotherms are lines formed by connecting points with the same temperature on an infrared thermal image. Isotherms allow you to clearly see the area of ​​temperature distribution.

[0176] The center feature refers to a numerical value that can reflect the property of the temperature center. For example, the center feature can include the temperature of the temperature center and the range of the temperature center.

[0177] In some embodiments, the government safety supervision management platform 110 can read the isotherms to obtain the temperature of the center. In some embodiments, the government safety supervision management platform 110 can calculate the range of the temperature center by the distance on the map combined with the scale when shooting.

[0178] Step 420 , based on the temperature center and its center characteristics, determine the temperature characteristics of the attached components.

[0179] A temperature feature refers to data that can reflect the temperature properties of an attached component. For example, a temperature feature can include temperature.

[0180] In some embodiments, the government safety supervision management platform 110 can determine the temperature characteristics of the attached components based on the temperature center and its center characteristics.

[0181] For example, the government safety supervision management platform 110 may determine the temperature characteristics of the accessory components through steps S421 - S422.

[0182] In step S421, the government safety supervision management platform 110 can determine whether the accessory component belongs to a low-temperature center or a high-temperature center according to the type of the accessory component through a preset table.

[0183] The preset table may include a correspondence between the accessory component type and the temperature center. The government safety supervision management platform 110 may obtain the preset table based on the experimental data by counting the experimental data with the most occurrences under each accessory component type into a table. The experimental data may include low temperature center data and high temperature center data. For example, if, in 10 experiments of a certain accessory component, 7 experimental data are low temperature center data and 3 experimental data are high temperature center data, then the accessory component is a low temperature center.

[0184] The experimental data is data obtained by conducting field experiments on various types of accessory components. For example, the government safety supervision management platform 110 can detect the operating temperature of each type of accessory component, and determine whether each type of accessory component is a high temperature center or a low temperature center according to the operating temperature and the surrounding data, and obtain low temperature center data and high temperature center data.

[0185] Step S422, determining the temperature characteristics of the accessory components.

[0186] In some embodiments, if the accessory component belongs to a low-temperature center and there is only one low-temperature center data, the temperature of the low-temperature center is determined as the temperature of the accessory component; otherwise, all low-temperature center data are retained, all high-temperature center data are eliminated, and all infrared data corresponding to the retained low-temperature center data are determined as candidate temperature data.

[0187] In some embodiments, if the accessory component belongs to a high temperature center and there is only one high temperature center data, the temperature of the high temperature center is determined as the temperature of the accessory component; otherwise, all high temperature center data are retained, all low temperature center data are eliminated, and all infrared data corresponding to the retained high temperature center data are determined as candidate temperature data.

[0188] In some embodiments, after determining the candidate temperature data, the range of the temperature center of the candidate temperature data is compared with the range of the accessory component. If the range difference is lower than the range threshold, the candidate temperature data is determined as the temperature feature of the accessory component. The range difference may be the distance between the geometric center of the range of the temperature center and the geometric center of the range of the accessory component. The range threshold refers to a threshold used to determine whether the candidate temperature data is a temperature feature of the accessory component. In some embodiments, the range threshold may be set by a professional technician or by system default.

[0189] Step 430, based on the temperature characteristics, environmental data and the first abnormal value, determine the second abnormal value through the prediction model.

[0190] In some embodiments, the government safety supervision management platform 110 can determine the second abnormal value through a vector matching method based on the temperature characteristics of the auxiliary components, environmental data, and the first abnormal value.

[0191] Exemplarily, the government safety supervision management platform 110 can cluster the temperature characteristics, environmental data, first abnormal values ​​of the accessory components in the historical data, and the subsequent actual second abnormal values ​​of the accessory components to form a preset number of cluster centers, construct the pipeline parameters of the gas pipeline (i.e., the reference pipeline) where the accessory components corresponding to the cluster centers are located, and the number of users upstream and downstream of the pipeline into a standard vector, and determine the actual second abnormal value of the accessory components as the label of the standard vector. The actual second abnormal value is the actual accident frequency of the corresponding accessory components in the historical period.

[0192] The government safety supervision management platform 110 can construct the temperature characteristics of the accessory components, environmental data, and the first abnormal value in the current historical data into a vector to be matched, and calculate the vector similarity based on the standard vector and the vector to be matched, and determine the label of the standard vector with the highest similarity as the second abnormal value of the accessory component. The vector similarity can be negatively correlated with the distance between the vectors, and the vector distance can be determined based on the cosine distance, etc. For example, the vector similarity can be the inverse of the vector distance. For more information about the first abnormal value, please refer to Figure 2 and its related description.

[0193] In some embodiments, the government safety supervision management platform 110 can determine the second abnormal value through a prediction model based on the temperature characteristics, environmental data and the first abnormal value.

[0194] The prediction model is a model for predicting the second outlier. The prediction model can be a machine learning model, for example, at least one of a convolutional neural network (CNN) model, a deep neural network model (DNN), a neural network model (NN), and the like.

[0195] In some embodiments, the input of the prediction model may include temperature characteristics, environmental data, and a first abnormal value, and the output may be a second abnormal value.

[0196] In some embodiments, the government security supervision management platform 110 can perform multiple rounds of iterative training on the initial prediction model based on multiple labeled training samples to obtain a trained prediction model. For example, the government security supervision management platform 110 can first obtain a training data set, select one or more samples from the training data set, input these training samples into the prediction model, and obtain the prediction output of the model; construct a loss function based on the output and label of the prediction model, and iteratively and reversely update the parameters of the prediction model based on the loss function. When the loss function meets the preset conditions, the iteration ends, the model training is completed, and a trained prediction model is obtained. Among them, the preset conditions can be that the loss function converges, the number of iterations reaches a threshold, etc. Various methods can be used to update the parameters in the prediction model. For example, the parameters in the prediction model can be updated based on the gradient descent method.

[0197] In some embodiments, the training samples may include sample temperature characteristics, sample environmental data, and sample first abnormal values ​​of accessory components at historical moments in the historical data. In some embodiments, the label may be the actual frequency of accidents that occurred in a subsequent historical period of time at the historical moment.

[0198] In some embodiments, the prediction model can be obtained after training based on a training set, verifying based on a validation set, and testing based on a test set; the training set, the test set, and the validation set are data sets consisting of sample temperature characteristics, sample environmental data, and sample first outliers of training samples in historical data.

[0199] In some embodiments, the data volume of the training set, the test set, and the validation set constitutes a preset ratio. In some embodiments, the government security supervision management platform 110 can divide the available data set into three parts according to a preset ratio: a training set, a validation set, and a test set.

[0200] In some embodiments, the preset ratio is 8:1:1. That is, the training set accounts for 80% of the data set and is used to train the model. This part of the data is used to adjust the weights and parameters of the model. The validation set accounts for 10% of the data set and is used for validation during the model training process. The test set accounts for 10% of the data set and is used to finally evaluate the performance of the model.

[0201] In some embodiments, there is no data overlap among the training set, the test set, and the validation set. Data overlap means that when dividing the data set, each data sample is ensured to belong to only one of the training set, the validation set, or the test set.

[0202] In some embodiments, the statistical difference of the training samples in the training set is greater than a preset difference threshold, and the preset difference threshold is related to the variance of the influence values ​​corresponding to the training samples in the training set. The statistical difference of the training samples refers to data that can reflect the diversity of the samples in the training set. The greater the diversity of the samples, the greater the statistical difference of the training samples.

[0203] Exemplarily, the government safety supervision management platform 110 can quantize the sample temperature characteristics, sample environmental data, and sample first outlier of each training sample in the training set into a vector. Then, the government safety supervision management platform 110 can calculate the vector distance between any two training samples for all training samples in the training set, thereby obtaining multiple vector distances. Then, the government safety supervision management platform 110 can calculate the variance of the multiple vector distances and determine the variance as the statistical difference of the training samples. The larger the variance, the greater the diversity of the samples and the greater the statistical difference of the training samples. For more information on quantized vectors, see Figure 2 and its related description.

[0204] The difference threshold refers to a reference value used to determine whether the statistical difference of the training samples in the training set meets the requirements. In some embodiments, the preset difference threshold is positively correlated to the variance of the influence values ​​corresponding to the training samples in the training set. For example, the preset difference threshold can be the variance of the influence values ​​corresponding to the training samples in the training set. For more information about the influence value, see Figure 3 and its related description.

[0205] Since the larger the variance of the impact value, the more uncertain the gas accident is and the more potential impacts are involved, the preset difference threshold can be increased to ensure that the diversity of the training set samples covers more actual situations, so that the prediction model can learn to predict the target more accurately.

[0206] In some embodiments of the present specification, by introducing statistical differences in training samples, the prediction model can increase the diversity of training data during the training process, ensuring that the prediction model can still maintain good performance when facing different situations and noisy data, making the model more robust and preventing the model from overfitting, rather than relying on specific sample features.

[0207] In some embodiments, the prediction model input may also include infrared data features within a preset time period.

[0208] The infrared data feature refers to the temperature distribution feature in the infrared data. For example, the infrared data feature may include temperature distribution data.

[0209] In some embodiments, the government safety supervision management platform 110 can obtain temperature distribution data through the infrared data reading software provided by the infrared detection device or other software that can read infrared data (for example, FLIR Tools), and then determine the temperature distribution data as infrared data features.

[0210] In some embodiments of the present specification, the accuracy of the predicted second abnormal value can be further improved by taking the infrared data features within a preset time period as the input of the prediction model.

[0211] In some embodiments of the present specification, by introducing a prediction model, the second abnormal value can be automatically determined, and the prediction accuracy and prediction efficiency of the second abnormal value are improved, thereby improving the reliability of the overall smart gas accessory safety monitoring Internet of Things system 100. Thus, the smart gas accessory safety monitoring Internet of Things system 100 can more timely and accurately identify and warn of potential abnormal situations, and ensure the safe operation of the smart gas accessory safety monitoring Internet of Things system 100.

[0212] In some embodiments of the present specification, since the accessory components and the gas pipeline are very close to each other, the measured infrared data may be interfered by various factors (i.e., interference from other high temperature sources or low temperature sources in the environment in addition to the temperature anomaly caused by the operation of the accessory components themselves). By determining the temperature center and its center characteristics, the temperature data of the temperature anomaly caused by the operation of the accessory components themselves can be screened out, thereby eliminating the interference data in the infrared data.

[0213] It should be noted that the above description of process 400 is only for example and explanation, and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to process 400 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

[0214] Some embodiments of the present specification provide a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed by a computer, the smart gas accessory component safety monitoring method described in any one of the embodiments of the present specification is implemented.

[0215] The embodiments in this specification are only for illustration and description, and do not limit the scope of application of this specification. For those skilled in the art, various modifications and changes that can be made under the guidance of this specification are still within the scope of this specification.

[0216] Certain features, structures or characteristics of one or more embodiments of this specification may be appropriately combined.

[0217] Finally, 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, 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 smart gas accessory safety monitoring method, characterized in that: The method is performed by a smart gas accessory component safety monitoring Internet of Things system, and includes: Determining a first abnormal value of the accessory component based on operation data of the accessory component, the accessory component comprising at least one of a flange, a valve, a compensator, a drain, and a bleed pipe, the accessory component being located on a gas pipeline, the operation data comprising at least one of audio data, temperature data, and vibration data; Based on the first abnormal value, determining a target acquisition parameter, wherein the target acquisition parameter is an order of collecting infrared data in the field; Based on the target acquisition parameters, determine an acquisition instruction, and send the acquisition instruction to the staff's interactive device and / or drone to control the infrared detection device to collect the infrared data of the accessory component with the target acquisition parameters, wherein the infrared data includes an infrared thermal map; Based on the infrared data, the environmental data and the first abnormal value, a second abnormal value of the accessory component is determined, wherein the environmental data is data related to the environment in which the accessory component is located, and the second abnormal value is a numerical value or letter after adjusting the first abnormal value of the accessory component; wherein the determining the second abnormal value of the accessory component based on the infrared data, the environmental data and the first abnormal value specifically includes: Based on the infrared data, determining the temperature center and the center characteristics of the infrared data; determining a temperature characteristic of the accessory component based on the temperature center and the center characteristic; Based on the temperature feature, the environmental data and the first abnormal value, the second abnormal value is determined by a prediction model; wherein the prediction model is a machine learning model; and Based on the second abnormal value, a maintenance instruction and / or a parameter adjustment instruction is determined and sent to the monitoring component and the interactive device, the maintenance instruction is configured to instruct the staff to perform manual maintenance on the accessory component, and the parameter adjustment instruction is configured to adjust the monitoring parameters of the monitoring component before completing the maintenance, and the monitoring component is configured to monitor the operating data of the accessory component.

2. The method according to claim 1, characterized in that The determining, based on the operation data of the accessory component, the first abnormal value of the accessory component comprises: Based on the operation data, determining a reference interval, the reference interval comprising at least one of a reference audio interval, a reference temperature interval, and a reference vibration interval; Based on the reference interval, determining an operation feature, the operation feature comprising at least one of an audio anomaly vector, a temperature anomaly vector, and a vibration anomaly vector; Based on the operation characteristics, determining an operation intensity, the operation intensity comprising at least one of an audio abnormality intensity, a temperature abnormality intensity, and a vibration abnormality intensity; and Based on the operating intensity, the first abnormal value is determined.

3. The method according to claim 2, characterized in that The determining the first abnormal value based on the operation intensity includes: determining an amplification factor based on the reference audio interval, the reference temperature interval, and the reference vibration interval; and The first abnormal value is determined based on the operating intensity and the amplification coefficient.

4. The method according to claim 2, characterized in that Based on the operation intensity, determining the first abnormal value includes: The first abnormality value is determined based on the operation intensity and the impact value, wherein the impact value represents the degree of harm caused by the abnormality of the accessory component.

5. The method according to claim 4, characterized in that The impact value is related to the pipeline parameters of the reference pipeline and the number of upstream and downstream users of the reference pipeline. The reference pipeline refers to the gas pipeline where the accessory component is located.

6. A smart gas accessory safety monitoring Internet of Things system, characterized in that: The smart gas accessory safety monitoring IoT system includes a government safety supervision management platform, a government safety supervision sensor network platform, a government safety supervision object platform, a gas company management platform, a gas company sensor network platform, a gas equipment object platform and a gas pipeline maintenance object platform. The smart gas accessory safety monitoring IoT system is configured as follows: The government safety supervision management platform obtains operation data of accessory components through the gas company management platform and the government safety supervision sensor network platform, wherein the accessory components include at least one of flanges, valves, compensators, drainers, and vent pipes, and the accessory components are located on the gas pipeline, and the operation data includes at least one of audio data, temperature data, and vibration data; Determining, by the government safety supervision management platform, a first abnormal value of the accessory component based on the operation data; And, based on the first abnormal value, determine the target acquisition parameter, the target acquisition parameter is the order of collecting infrared data on the spot; and based on the target acquisition parameter, determine the acquisition instruction; send the acquisition instruction to the government safety supervision object platform, and further send it to the gas pipeline maintenance object platform and / or the gas equipment object platform to control the infrared detection equipment to collect the infrared data of the accessory component with the target acquisition parameter, the infrared data including the infrared heat map; Acquiring environmental data through the government safety supervision sensor network platform; as well as The government safety supervision management platform determines the second abnormal value of the accessory component based on the infrared data, the environmental data and the first abnormal value, wherein the environmental data is data related to the environment in which the accessory component is located, and the second abnormal value is a numerical value or letter after adjusting the first abnormal value of the accessory component; wherein the determining the second abnormal value of the accessory component based on the infrared data, the environmental data and the first abnormal value specifically includes: determining the temperature center and the center characteristics of the infrared data based on the infrared data; determining the temperature characteristics of the accessory component based on the temperature center and the center characteristics; determining the temperature characteristics of the accessory component based on the temperature characteristics, the environmental data and the first abnormal value ... Abnormal value, determine the second abnormal value through a prediction model; wherein the prediction model is a machine learning model; and based on the second abnormal value, determine the maintenance instruction and / or the parameter adjustment instruction, and send it to the gas company management platform of the government safety supervision object platform, and the gas company management platform further issues the maintenance instruction and / or the parameter adjustment instruction to the gas equipment object platform and the gas pipeline maintenance object platform, the maintenance instruction is configured to instruct the staff to perform manual maintenance on the accessory component, the parameter adjustment instruction is configured to adjust the monitoring parameters of the monitoring component before completing the maintenance, and the monitoring component is configured to monitor the operating data of the accessory component.

7. The smart gas accessory safety monitoring Internet of Things system according to claim 6 is characterized in that: The gas equipment object platform is configured as the monitoring component, the drone and the infrared detection device; The gas pipeline maintenance object platform is configured as an interactive device for the staff.

8. A computer-readable storage medium, characterized in that: The storage medium stores computer instructions. When the computer reads the computer instructions in the storage medium, the computer executes the smart gas accessory safety monitoring method as described in claim 1.

Images