A gas leakage prediction method based on multi-dimensional sensor data of a gas pipe network

Abstract

The present application relates to a kind of based on gas pipe network multidimensional sensor data's leak prediction method, comprising the following steps: using Transformer model, according to the use habit of each gas equipment, the reasonable interval of each gas equipment is screened out;Wherein, the use habit is the duration of each use of gas equipment;The detection result of flow sensor configured in the main transport pipeline of gas pipe network is obtained;When the detection result is transport state, the duration of each gas equipment is obtained;Based on the duration of each gas equipment and the reasonable interval corresponding to each gas equipment, data processing is carried out, and whether early warning and leak alarm are judged according to data processing result and gas sensor;When the detection result is non-transport state, then according to gas sensor, whether early warning is judged.The present application can improve the accuracy and timeliness of gas leakage prediction.

Description

Technical Field

[0001] This invention relates to the field of gas leak early warning technology, and in particular to a leak prediction method based on multi-dimensional sensor data of gas pipeline networks. Background Technology

[0002] With the acceleration of urbanization, natural gas, as an important energy source, is widely used in residential life and industrial production. Therefore, developing efficient methods for predicting and monitoring gas leaks is particularly important.

[0003] Traditional gas leak detection methods mainly rely on manual inspections and simple sensor monitoring. These methods suffer from drawbacks such as slow response time and insufficient accuracy, making them unsuitable for the safety requirements of modern urban gas pipeline networks. In recent years, with the rapid development of IoT technology and big data analytics, the application of multi-dimensional sensor data has provided new insights into gas leak monitoring.

[0004] Multidimensional sensors in gas pipeline networks can collect flow data in real time, providing a wealth of information for leak prediction. However, how to effectively process and analyze this data to achieve accurate monitoring of the operating status of gas equipment and leak early warning remains a pressing technical challenge. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a leak prediction method based on multi-dimensional sensor data of gas pipeline network, which can improve the accuracy and timeliness of gas leak prediction.

[0006] The technical solution adopted by this invention to solve its technical problem is: to provide a leakage prediction method based on multi-dimensional sensor data of gas pipeline network, including the following steps:

[0007] Using the Transformer model, a reasonable range for each gas appliance is selected based on its usage habits; where the usage habits refer to the duration of each use of the gas appliance.

[0008] Obtain the detection results of the flow sensor configured in the main transport pipeline of the gas pipeline network;

[0009] When the detection result indicates a transportation status, the duration of each gas device is obtained;

[0010] Data is processed based on the duration of each gas appliance and the corresponding reasonable range for each gas appliance, and the results of the data processing and gas sensors are used to determine whether to issue a warning or a leak alarm.

[0011] If the detection result indicates a non-transportation status, a warning reminder will be issued based on the gas sensor readings.

[0012] The use of the Transformer model to filter out a reasonable range for each gas appliance based on its usage habits includes:

[0013] Data collection for all elements within the gas equipment set:

[0014] Randomly select any element from the set of gas equipment as the target equipment;

[0015] The time of each startup of the target device within the current month is obtained by using the Transformer model and is denoted as the startup interval, which includes the start time and the end time.

[0016] According to the order in which the gas appliances were started, the start-up intervals were entered into the first time set;

[0017] If the number of starts in the current month is less than or equal to the start quantity threshold, then obtain all start intervals of the previous month and enter them into the first time set according to the order of gas equipment start time.

[0018] Select elements from the first time set as the first judgment element in sequence;

[0019] Obtain the termination time from the first element to be judged as the first time;

[0020] Get all adjacent elements of the first judgment element in the first time set, and select the adjacent element with the latest gas equipment start time as the second judgment element.

[0021] Use the start time within the second judgment element as the second time;

[0022] Calculate the difference between the second time and the first time, and record it as the time difference;

[0023] If the time difference is less than or equal to the time interval threshold, the start time of the first judgment element is taken as the start time of the second judgment element, and the first judgment element in the first time set is removed.

[0024] Reselect an element from the first time set as the new first judgment element for judgment;

[0025] If the time difference is greater than the time interval threshold, the first judgment element is entered into the second time set, and the first judgment element in the first time set is removed.

[0026] Reselect an element from the first time set as the new first judgment element for judgment;

[0027] When the number of elements in the first time set is zero, calculate the difference between the end time and the start time of each element in the second time set, record it as the duration, and enter it into the third time set.

[0028] Obtain the minimum and maximum elements within the third time set, form a reasonable interval for the target device, and enter the interval set;

[0029] The minimum value in the reasonable interval is the minimum element in the third time set, and the maximum value in the reasonable interval is the maximum element in the third time set.

[0030] Reselect elements within the device set and obtain a reasonable range for the current element through data collection;

[0031] Data collection stops once all elements in the device set have been traversed.

[0032] The process involves data processing based on the duration of each gas appliance and the corresponding reasonable timeframe for each appliance, and determining whether to issue a warning or leak alarm based on the data processing results and gas sensor readings. Specifically, this includes:

[0033] Obtain the operating status of all elements within the device set; wherein, the operating status includes start status and stop status;

[0034] Add the elements in the startup state from the device set to the startup set;

[0035] If the number of elements in the set is zero, a leak alarm will be triggered;

[0036] If the number of elements in the activation set is not zero, the gas data of each gas sensor is acquired in real time and recorded in the safety set; wherein, the gas data is the gas concentration in the space where the gas sensor is located;

[0037] When there is an element in the safety set that is greater than or equal to the gas leak concentration threshold, the element is marked as a hazardous element, and a leak alarm is triggered on the gas equipment corresponding to the hazardous element.

[0038] Randomly select elements from the start set as monitoring elements;

[0039] The duration for which the monitored element is in the active state is acquired in real time and recorded as the target duration;

[0040] Within the interval set, obtain the reasonable interval corresponding to the monitored element, and denote it as the target reasonable interval;

[0041] If the duration of the monitored element is less than the minimum value in the reasonable range of the target, then select an element from the start set as the monitored element and continue monitoring.

[0042] When a monitored element changes from an active state to a stopped state, and the target duration is less than the minimum value in the reasonable range of the target, the target duration is recorded into the minimum set corresponding to the current monitored element.

[0043] Remove the currently monitored elements from the start set, select new elements from the start set as new monitored elements, and determine whether to issue a warning or a leak alarm.

[0044] The leakage prediction method based on multi-dimensional sensor data of gas pipeline network also includes the step of adjusting the reasonable range of the corresponding gas equipment according to the minimum set, specifically including:

[0045] Get the number of elements in the smallest set of currently monitored elements;

[0046] If the number of elements in the current minimum set of monitored elements is greater than one-third of the start-up threshold, then obtain the maximum value in the current minimum set of monitored elements.

[0047] If the above maximum value is less than the minimum value in the reasonable range corresponding to the current monitored element, then the above maximum value will replace the minimum value in the reasonable range corresponding to the current monitored element.

[0048] If the target duration is within a reasonable range, then reselect an element from the start set as a monitoring element and continue monitoring.

[0049] If the target duration is greater than the maximum value in the reasonable range of the target, then calculate the difference between the target duration and the time fluctuation threshold, and record it as the first difference.

[0050] If the first difference is less than or equal to the maximum value in the target reasonable range, then reselect an element from the starting set as the monitoring element and continue monitoring;

[0051] If the first difference is greater than the maximum value in the target reasonable range, then a warning reminder and leakage alarm should be issued based on the monitoring set.

[0052] The process of determining whether to issue early warnings and leak alarms based on the monitoring set specifically includes:

[0053] Get the smallest element in the monitoring set corresponding to the current monitored element, and denote it as the warning value;

[0054] When the duration of the target equals the warning value, an alert is issued and user feedback is obtained.

[0055] If the user reports that the monitored element is in use, then select another element from the start set as the monitored element and continue monitoring.

[0056] If the user reports that the monitored element is not in use, a leak alarm will be triggered;

[0057] When a monitored element changes from an active state to a stopped state, the target duration is recorded in the monitoring set corresponding to the current monitored element.

[0058] Remove the currently monitored elements from the start set, select new elements from the start set as new monitored elements, and determine whether to issue a warning or a leak alarm.

[0059] If the number of elements in the set is zero, then obtain the result displayed by the flow sensor;

[0060] If the flow sensor displays a transport status, a leak alarm will be triggered.

[0061] The leakage prediction method based on multi-dimensional sensor data of gas pipeline network also includes:

[0062] Get the number of elements in the current monitored element's monitoring set;

[0063] If the number of elements in the current monitoring element's monitoring set is greater than one-third of the activation threshold, then obtain the minimum value in the current monitoring element's monitoring set.

[0064] If the minimum value mentioned above is greater than the maximum value in the reasonable range corresponding to the current monitored element, then the minimum value mentioned above will be replaced with the maximum value in the reasonable range corresponding to the current monitored element.

[0065] The process of determining whether to issue a warning based on the gas sensor specifically includes:

[0066] Real-time acquisition of gas data from all gas sensors, and recording of the gas data into the normal set;

[0067] If there are elements in the normal set that have a concentration greater than the gas leak concentration threshold, they are marked as leaking elements;

[0068] The gas sensor corresponding to the leaking element provides a leak warning.

[0069] Beneficial effects

[0070] By adopting the above-mentioned technical solution, this invention has the following advantages and positive effects compared with the prior art: This invention improves the accuracy and timeliness of gas leak prediction by deeply analyzing the usage habits of gas equipment and combining real-time sensor data. This method can not only monitor the operating status of gas equipment in real time, but also automatically adjust the reasonable usage range through intelligent algorithms, thereby achieving effective early warning of potential leak risks and providing more reliable technical support for ensuring gas safety. Attached Figure Description

[0071] Figure 1 This is a flowchart of an embodiment of the present invention. Detailed Implementation

[0072] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims.

[0073] This invention relates to a leakage prediction method based on multi-dimensional sensor data from a gas pipeline network. In this embodiment, the gas pipeline network includes a main transport pipeline and multiple sub-transport pipelines. The main transport pipeline is equipped with a flow sensor. The transmission end of the main transport pipeline connects to multiple sub-transport pipelines, each sub-transport pipeline corresponding to a gas device. Each sub-transport pipeline corresponding to a gas device is equipped with a gas sensor. Figure 1 As shown, the leakage prediction method based on multi-dimensional sensor data of gas pipeline network in this embodiment specifically includes the following steps:

[0074] Using the Transformer model, a reasonable range for each gas appliance is selected based on its usage habits; where the usage habits refer to the duration of each use of the gas appliance.

[0075] Obtain the detection results of the flow sensor configured in the main transport pipeline of the gas pipeline network;

[0076] When the detection result indicates a transportation status, the duration of each gas device is obtained;

[0077] Data is processed based on the duration of each gas appliance and the corresponding reasonable range for each gas appliance, and the results of the data processing and gas sensors are used to determine whether to issue a warning or a leak alarm.

[0078] If the detection result indicates a non-transportation status, a warning reminder will be issued based on the gas sensor readings.

[0079] This implementation method uses a Transformer model based on multi-dimensional sensor data to accurately identify gas pipeline leaks. By filtering reasonable ranges and analyzing real-time data from monitoring sets, it can issue timely warnings or leak alarms when equipment malfunctions, effectively preventing gas leaks caused by equipment failures or pipeline problems.

[0080] This implementation uses the Transformer model to filter out reasonable ranges for each gas appliance based on its usage habits, specifically including:

[0081] Data collection for all elements within the gas equipment set:

[0082] Randomly select any element from the set of gas equipment as the target equipment;

[0083] The time of each startup of the target device within the current month is obtained by using the Transformer model and is denoted as the startup interval, which includes the start time and the end time.

[0084] According to the order in which the gas appliances were started, the start-up intervals were entered into the first time set;

[0085] Set a startup data threshold 'a'. This startup quantity threshold 'a' is a constant and can be adjusted according to the actual situation to select a sufficient amount of historical data.

[0086] If the number of starts in the current month is less than or equal to the start quantity threshold 'a', then obtain all start intervals from the previous month and enter them into the first time set according to the order of gas equipment start time.

[0087] Set a time interval threshold b, which is a constant and can be adjusted according to the user's actual situation. This threshold is used to determine the size of the interval between adjacent startup times.

[0088] Select elements from the first time set as the first judgment element in sequence;

[0089] Obtain the termination time from the first element to be judged as the first time;

[0090] Get all adjacent elements of the first judgment element in the first time set, and select the adjacent element with the latest gas equipment start time as the second judgment element.

[0091] Use the start time within the second judgment element as the second time;

[0092] Calculate the difference between the second time and the first time, and record it as the time difference;

[0093] If the time difference is less than or equal to the time interval threshold b, then the start time of the first judgment element is taken as the start time of the second judgment element, and the first judgment element in the first time set is removed.

[0094] By sequentially selecting elements from the first time set as the first judgment element and obtaining their termination time, and then finding the element with the latest start time among all its adjacent elements as the second judgment element, the time difference between the two is calculated. If the time difference is less than or equal to the time interval threshold b, the two are merged, and the first judgment element is removed; otherwise, the first judgment element is entered into the second time set. By merging adjacent elements whose time interval is less than the time interval threshold b, this method can better reflect the actual use of gas equipment and reduce the error caused by frequent starts in a short period of time.

[0095] Reselect an element from the first time set as the new first judgment element for judgment;

[0096] If the time difference is greater than the time interval threshold b, the first judgment element is entered into the second time set, and the first judgment element in the first time set is removed.

[0097] Reselect an element from the first time set as the new first judgment element for judgment;

[0098] When the number of elements in the first time set is zero, calculate the difference between the end time and the start time of each element in the second time set, record it as the duration, and enter it into the third time set.

[0099] Obtain the minimum and maximum elements within the third time set, form a reasonable interval for the target device, and enter the interval set;

[0100] The minimum value in the reasonable interval is the minimum element in the third time set, and the maximum value in the reasonable interval is the maximum element in the third time set.

[0101] Reselect elements within the device set and obtain a reasonable range for the current element through data collection;

[0102] Data collection stops once all elements in the device set have been traversed.

[0103] Therefore, by analyzing the historical data of each gas appliance to obtain its usage habits, and establishing corresponding reasonable intervals based on these habits, the operating status of each appliance can be traced. By monitoring the usage time of each gas appliance in real time, accurate monitoring of the operating time of different appliances can be ensured, reducing false alarms or omissions in the data.

[0104] In this embodiment, data processing is performed based on the duration of each gas device and the corresponding reasonable range for each gas device. The data processing results and gas sensor readings are then used to determine whether to issue a warning or leak alarm. Specifically, this includes:

[0105] Get the running status of all elements in the device set; where running status includes start status and stop status;

[0106] Add the elements in the startup state from the device set to the startup set;

[0107] If the number of elements in the set is zero, a leak alarm will be triggered;

[0108] If the number of elements in the activation set is not zero, the gas data of each gas sensor is acquired in real time and recorded in the safety set; wherein, the gas data is the gas concentration in the space where the gas sensor is located;

[0109] Set a gas leakage concentration threshold d, which is a constant and is used to define the maximum allowable gas concentration in the space.

[0110] When there is an element in the safety set that is greater than or equal to the gas leak concentration threshold d, the element is marked as a dangerous element, and a leak alarm is triggered on the gas equipment corresponding to the dangerous element.

[0111] Randomly select elements from the start set as monitoring elements;

[0112] The duration for which the monitored element is in the active state is acquired in real time and recorded as the target duration;

[0113] Within the interval set, obtain the reasonable interval corresponding to the monitored element, and denote it as the target reasonable interval;

[0114] Set a time fluctuation threshold c, which is a constant and can be adjusted according to the actual situation. It is used to determine the maximum allowable duration of the subsequent fluctuation.

[0115] If the duration of the monitored element is less than the minimum value in the reasonable range of the target, then select an element from the start set as the monitored element and continue monitoring.

[0116] When a monitored element changes from an active state to a stopped state, and the target duration is less than the minimum value in the reasonable range of the target, the target duration is recorded into the minimum set corresponding to the current monitored element.

[0117] Remove the currently monitored elements from the start set, select new elements from the start set as new monitored elements, and determine whether to issue a warning or a leak alarm.

[0118] It is easy to see that by combining the data from flow sensors and gas sensors, the system can more accurately monitor the operation of gas equipment in both transport and non-transport states. This not only reduces the possibility of users forgetting to turn off gas equipment, but also detects and alerts users in a timely manner when leaks occur, significantly improving the overall safety of the gas pipeline network.

[0119] This implementation also includes the step of adjusting the reasonable range of the corresponding gas equipment according to the minimum set, specifically including:

[0120] Get the number of elements in the smallest set of currently monitored elements;

[0121] If the number of elements in the current minimum set of monitored elements is greater than one-third of the start-up threshold (i.e., a / 3), then obtain the maximum value in the current minimum set of monitored elements.

[0122] If the above maximum value is less than the minimum value in the reasonable range corresponding to the current monitored element, then the above maximum value will replace the minimum value in the reasonable range corresponding to the current monitored element.

[0123] By dynamically adjusting the optimal operating range of gas equipment based on its actual usage, and through minimal set data analysis, the system can update the optimal operating range of equipment in real time, making gas equipment management more flexible and intelligent.

[0124] In this embodiment, when determining whether to issue a warning and leak alarm based on the data processing results and the gas sensor, the following is also included:

[0125] If the target duration is within a reasonable range, then reselect an element from the start set as a monitoring element and continue monitoring.

[0126] If the target duration is greater than the maximum value in the reasonable range of the target, then calculate the difference between the target duration and the time fluctuation threshold c, and record it as the first difference.

[0127] If the first difference is less than or equal to the maximum value in the target reasonable range, then reselect an element from the starting set as the monitoring element and continue monitoring;

[0128] If the first difference is greater than the maximum value in the target reasonable range, then a warning reminder and leakage alarm should be issued based on the monitoring set.

[0129] Specifically, the determination of whether to issue early warnings and leakage alarms based on the monitoring set is as follows:

[0130] Get the smallest element in the monitoring set corresponding to the current monitored element, and denote it as the warning value;

[0131] When the duration of the target equals the warning value, an alert is issued and user feedback is obtained.

[0132] If the user reports that the monitored element is in use, then select another element from the start set as the monitored element and continue monitoring.

[0133] If the user reports that the monitored element is not in use, a leak alarm will be triggered;

[0134] When a monitored element changes from an active state to a stopped state, the target duration is recorded in the monitoring set corresponding to the current monitored element.

[0135] Remove the currently monitored elements from the start set, select new elements from the start set as new monitored elements, and determine whether to issue a warning or a leak alarm.

[0136] If the number of elements in the set is zero, then obtain the result displayed by the flow sensor;

[0137] If the flow sensor displays a transport status, a leak alarm will be triggered.

[0138] This implementation method employs a multi-layered early warning mechanism. When the device's operating duration exceeds a reasonable range, an early warning is issued first, followed by further assessment based on user feedback to determine if a leak has occurred. This human-machine interactive approach improves alarm accuracy and reduces false alarm rates.

[0139] It is worth mentioning that this implementation method also includes:

[0140] Get the number of elements in the current monitored element's monitoring set;

[0141] If the number of elements in the current monitoring element's monitoring set is greater than one-third of the activation threshold (i.e., a / 3), then obtain the minimum value in the current monitoring element's monitoring set.

[0142] If the minimum value mentioned above is greater than the maximum value in the reasonable range corresponding to the current monitored element, then the minimum value mentioned above will be replaced with the maximum value in the reasonable range corresponding to the current monitored element.

[0143] Therefore, this implementation method can also update the reasonable range of the equipment in real time through data analysis of the monitoring set, and dynamically adjust the reasonable range of the equipment according to the actual use of the gas equipment, making the management of gas equipment more flexible and intelligent.

[0144] In this embodiment, determining whether to issue a warning based on the gas sensor specifically includes:

[0145] If the flow sensor displays a non-transportation status, the gas data from all gas sensors will be acquired in real time and entered into the normal set.

[0146] If there is an element in the normal set that is greater than the gas leak concentration threshold d, then it is marked as a leak element;

[0147] The gas sensor corresponding to the leaking element provides a leak warning;

[0148] When the system detects an abnormal gas concentration in a gas appliance even when it is not in operation, it will promptly issue a leak alarm, greatly reducing the safety hazards that may be caused by gas leaks.

[0149] It is easy to see that this invention improves the accuracy and timeliness of gas leak prediction by deeply analyzing the usage habits of gas appliances and combining real-time sensor data. This method not only monitors the operating status of gas appliances in real time, but also automatically adjusts the reasonable usage range through intelligent algorithms, thereby achieving effective early warning of potential leak risks and providing more reliable technical support for ensuring gas safety.

Claims

1. A method for leak prediction based on multi-dimensional sensor data of a gas network, characterized in that, The method comprises the following steps: Using a Transformer model, filtering a reasonable interval for each gas device according to the use habit of each gas device, specifically including: Collecting data on all elements in the gas device set: Randomly selecting any element in the gas device set as the target device; Obtaining the time of each start of the target device in the current month using the Transformer model, denoted as the start interval, which includes the start time and the end time; According to the chronological order of the start of the gas device, the start interval is recorded in the first time set; If the number of starts in the current month is less than or equal to the start number threshold, obtain all start intervals of the last month and record them in the first time set according to the chronological order of the start of the gas device; Selecting elements in the first time set as first judgment elements in turn; Obtaining the end time in the first judgment element as the first time; Obtaining all adjacent elements of the first judgment element in the first time set, and selecting the adjacent element with the latest start time of the gas device as the second judgment element; Taking the start time in the second judgment element as the second time; Calculate the difference between the second time and the first time, denoted as the time difference; If the time difference is less than or equal to the time interval threshold, the start time of the first judgment element is taken as the start time of the second judgment element, and the first judgment element in the first time set is removed; Re-selecting elements in the first time set as new first judgment elements for judgment; If the time difference is greater than the time interval threshold, the first judgment element is recorded in the second time set, and the first judgment element in the first time set is removed; Re-selecting elements in the first time set as new first judgment elements for judgment; When the number of elements in the first time set is zero, calculate the difference between the end time and the start time of each element in the second time set, denoted as the duration, and record it in the third time set; Obtaining the minimum element and the maximum element in the third time set to form a reasonable interval for the target device, and recording it in the interval set; The minimum value in the reasonable interval is the minimum element in the third time set, and the maximum value in the reasonable interval is the maximum element in the third time set; Re-selecting elements in the device set and obtaining the reasonable interval of the current element through data collection; When all elements in the device set are traversed, data collection is stopped; The use habit is the duration of each use of the gas device; Obtaining the detection result of the flow sensor configured in the main transport pipeline of the gas pipe network; When the detection result is the transport state, obtaining the duration of each gas device; Based on the duration of each gas device and the reasonable interval corresponding to each gas device, data processing is performed, and according to the data processing result and the gas sensor, whether to perform early warning and leakage alarm is determined, specifically including: Obtaining the running state of all elements in the device set; wherein, the running state includes the start state and the stop state; Recording the elements in the start state in the device set in the start set; If the number of elements in the start set is zero, perform leakage alarm; If the number of elements in the activation set is not zero, the gas data of each gas sensor is acquired in real time and recorded in the safety set; wherein, the gas data is the gas concentration in the space where the gas sensor is located; When there is an element in the safety set that is greater than or equal to the gas leak concentration threshold, the element is marked as a hazardous element, and a leak alarm is triggered on the gas equipment corresponding to the hazardous element. Randomly select elements from the start set as monitoring elements; The duration for which the monitored element is in the active state is acquired in real time and recorded as the target duration; Within the interval set, obtain the reasonable interval corresponding to the monitored element, and denote it as the target reasonable interval; If the duration of the monitored element is less than the minimum value in the reasonable range of the target, then select an element from the start set as the monitored element and continue monitoring. When a monitored element changes from an active state to a stopped state, and the target duration is less than the minimum value in the reasonable range of the target, the target duration is recorded into the minimum set corresponding to the current monitored element. Remove the currently monitored elements from the start set, select new elements from the start set as new monitored elements, and determine whether to issue a warning or a leak alarm. If the detection result indicates a non-transportation status, a warning reminder will be issued based on the gas sensor readings.

2. The method of claim 1, wherein, It also includes the step of adjusting the reasonable range of the corresponding gas equipment based on the minimum set, specifically including: Get the number of elements in the smallest set of currently monitored elements; If the number of elements in the current minimum set of monitored elements is greater than one-third of the start-up threshold, then obtain the maximum value in the current minimum set of monitored elements. If the above maximum value is less than the minimum value in the reasonable range corresponding to the current monitored element, then the above maximum value will replace the minimum value in the reasonable range corresponding to the current monitored element.

3. The leakage prediction method based on multi-dimensional sensor data of gas pipeline network according to claim 1, characterized in that, If the target duration is within a reasonable range, then reselect an element from the start set as a monitoring element and continue monitoring. If the target duration is greater than the maximum value in the reasonable range of the target, then calculate the difference between the target duration and the time fluctuation threshold, and record it as the first difference. If the first difference is less than or equal to the maximum value in the target reasonable range, then reselect an element from the starting set as the monitoring element and continue monitoring; If the first difference is greater than the maximum value in the target reasonable range, then a warning reminder and leakage alarm should be issued based on the monitoring set.

4. The gas network multi-dimensional sensor data based leak prediction method according to claim 3, characterized in that, The process of determining whether to issue early warnings and leak alarms based on the monitoring set specifically includes: Get the smallest element in the monitoring set corresponding to the current monitored element, and denote it as the warning value; When the duration of the target equals the warning value, an alert is issued and user feedback is obtained. If the user reports that the monitored element is in use, then select another element from the start set as the monitored element and continue monitoring. If the user reports that the monitored element is not in use, a leak alarm will be triggered; When a monitored element changes from an active state to a stopped state, the target duration is recorded in the monitoring set corresponding to the current monitored element. Remove the currently monitored elements from the start set, select new elements from the start set as new monitored elements, and determine whether to issue a warning or a leak alarm. If the number of elements in the set is zero, then obtain the result displayed by the flow sensor; If the flow sensor displays a transport status, a leak alarm will be triggered.

5. The method for leak prediction based on multi-dimensional sensor data of gas network of claim 1, wherein, Also includes: Get the number of elements in the current monitored element's monitoring set; If the number of elements in the current monitoring element's monitoring set is greater than one-third of the activation threshold, then obtain the minimum value in the current monitoring element's monitoring set. If the minimum value mentioned above is greater than the maximum value in the reasonable range corresponding to the current monitored element, then the minimum value mentioned above will be replaced with the maximum value in the reasonable range corresponding to the current monitored element.

6. The method of claim 3, wherein the method further comprises: The process of determining whether to issue a warning based on the gas sensor specifically includes: Real-time acquisition of gas data from all gas sensors, and recording of the gas data into the normal set; If there are elements in the normal set that have a concentration greater than the gas leak concentration threshold, they are marked as leaking elements; The gas sensor corresponding to the leaking element provides a leak warning.

Images