Gas pipe network leakage early warning system applying artificial intelligence

Abstract

The invention relates to the field of gas pipe network leakage early warning, in particular to a gas pipe network leakage early warning system applying artificial intelligence. Comprising the following steps: acquiring sensor data, and processing the sensor data through a data fusion algorithm to obtain fused sensor data; on the basis of the sensor data and the fused sensor data, a fuel gas leakage score is obtained through calculation by adopting a multi-level spatio-temporal evolutionary algorithm; based on the fuel gas leakage score, a fuel gas leakage trend analysis algorithm is used for calculation to obtain a fuel gas leakage trend value; external environment data are obtained, and based on the gas leakage score, the gas leakage trend value and the external environment data, an early warning priority evaluation algorithm is used for calculation to obtain an early warning priority value; and setting a low-level early warning threshold value and a high-level early warning threshold value, dividing early warning levels, and formulating emergency response strategies according to different early warning levels. The technical problems that the change of the leakage risk cannot be captured in time in the trend analysis of the gas leakage and the sensitivity to the score change is insufficient are solved.

Description

Technical Field

[0001] The present invention relates to the field of gas pipeline network leakage warning, and in particular to a gas pipeline network leakage warning system using artificial intelligence. Background Art

[0002] Gas pipeline network is an extremely important part of urban infrastructure, mainly used to transport gas energy such as natural gas. However, with the acceleration of urbanization and the continuous increase in gas demand, the scale and complexity of gas pipeline network are also expanding, which brings convenience to energy supply but also increases the risk of pipeline leakage. Gas leakage will not only lead to energy waste, but more importantly, once the leaked gas encounters a fire source, it will cause a fire or even an explosion, causing serious safety hazards.

[0003] Traditional gas pipeline leakage detection methods mainly include manual inspection and fixed monitoring equipment. Manual inspection requires a lot of manpower and time, is inefficient and prone to missed detection. Although fixed monitoring equipment can achieve 24-hour monitoring, its coverage is limited and it is easily affected by environmental factors, resulting in false alarms and missed alarms. Therefore, establishing a gas pipeline leakage early warning system using artificial intelligence is of great significance to ensure public safety and save energy.

[0004] By introducing artificial intelligence technology, comprehensive and real-time monitoring of gas pipeline networks can be achieved, and the accuracy and response speed of leak detection can be improved, reducing the risk of safety accidents. It has broad application prospects and important practical significance.

[0005] However, the existing gas pipeline leakage warning methods have the following technical problems: relying on a single type of sensor leads to incomplete data collection and cannot accurately reflect the gas leakage situation; lack of comprehensive consideration of time and space dimensions makes it difficult to accurately assess the severity and trend of gas leakage; gas leakage trend analysis cannot timely capture changes in leakage risks, and is not sensitive enough to score changes, which can easily miss potential leakage risks and affect the response speed and accuracy of the gas pipeline leakage warning system; it is easy to ignore external environmental factors and cannot comprehensively assess the severity and urgency of gas leakage risks. Summary of the invention

[0006] The present invention provides a gas pipeline leakage early warning system using artificial intelligence to solve the problems that the existing system relies on a single type of sensor, resulting in incomplete data collection and inability to accurately reflect the gas leakage situation; lacks comprehensive consideration of time and space dimensions, making it difficult to accurately assess the severity and trend of gas leakage; the trend analysis of gas leakage cannot timely capture changes in leakage risks, and is insufficiently sensitive to changes in scores, easily missing potential leakage risks and affecting the response speed and accuracy of the gas pipeline leakage early warning system; and easily ignores external environmental factors and is unable to comprehensively assess the severity and urgency of gas leakage risks.

[0007] The present invention provides a gas pipeline network leakage early warning system using artificial intelligence, which specifically includes the following technical solutions:

[0008] A gas pipeline network leakage early warning method using artificial intelligence comprises the following steps:

[0009] S1: Collect sensor data, process the sensor data through data fusion algorithm, and obtain fused sensor data; based on the sensor data and fused sensor data, use multi-level spatiotemporal evolution algorithm to calculate the gas leakage score;

[0010] S2: Based on the gas leakage score, the gas leakage trend value is calculated using the gas leakage trend analysis algorithm;

[0011] S3: Obtain external environmental data, and calculate the warning priority value using the warning priority evaluation algorithm based on the gas leakage score, the gas leakage trend value and the external environmental data;

[0012] S4: Set low-level warning thresholds and high-level warning thresholds, compare them with the warning priority values, divide the warning levels, and formulate emergency response strategies according to different warning levels.

[0013] Preferably, the S1 specifically includes:

[0014] In the process of implementing the multi-level spatiotemporal evolutionary algorithm, the spatial level and the temporal level are co-evolved. Through data fusion and layer-by-layer optimization in different dimensions, gas pipeline network leaks are identified and gas leak scores are obtained.

[0015] Preferably, the S1 specifically includes:

[0016] The co-evolution of the spatial level is to obtain a spatial level score by weighted averaging the sensor data for each spatial level; the co-evolution of the temporal level is to obtain a temporal level score by weighted averaging the sensor data of the monitoring position within the time window set for each temporal level and classifying the sensor data within a period of time into the same set.

[0017] Preferably, the S1 specifically includes:

[0018] The specific calculation formula of spatial hierarchy score is:

[0019]

[0020] Among them, S i (t) represents the spatial hierarchy score of the sensor at the ith position at time t; represents the weighted sum of the scores of the L spatial levels of the sensor at the i-th position; w k represents the weight of the kth spatial level, satisfying Represents the size of the neighborhood set of the sensor at the i-th position at the k-th spatial level; represents the sum of sensor data of all neighborhood sets of the sensor at the i-th position at the k-th spatial level; Di, j (t) represents the sensor data of the ,th sensor at the ,th position at time t;

[0021] The specific calculation formula for time level scoring is:

[0022]

[0023] Among them, T i (t) represents the temporal hierarchical score of the sensor at the ith position at time t; represents the weighted sum of the scores of the H time levels of the sensor at the i-th position; v m Represents the weight of the mth time level, satisfying Represents the size of the time window set of the sensor at the i-th position at the m-th time level; It means summing up the sensor data of the sensor at the i-th position in all time windows at the m-th time level; D f，i (t) represents the sensor data of the sensor at the i-th position after fusion at time t.

[0024] Preferably, the S2 specifically includes:

[0025] In the process of implementing the gas leakage trend analysis algorithm, the change rate of the gas leakage score is calculated to evaluate the change trend of the gas leakage situation. A nonlinear adjustment factor is introduced to make nonlinear adjustments to the changes in gas leakage scores of different degrees to obtain the gas leakage trend value. The specific implementation formula is:

[0026]

[0027] Among them, P i (t) represents the gas leakage trend value of the sensor at the i-th position at time t; F i (t) represents the gas leakage score of the sensor at the i-th position at time t; ΔF indicates the rate of change of the gas leakage score of the sensor at the i-th position in the time period [t-τ, t]; i (t-τ, t) represents the change in the gas leakage score of the sensor at the ith position within the time period [t-τ, t]; τ represents the length of the time period; e represents the base of the natural logarithm; F′ represents the gas leakage score threshold; Represents the nonlinear adjustment factor.

[0028] Preferably, the S3 specifically includes:

[0029] In the implementation process of the early warning priority evaluation algorithm, the gas leakage score and the logarithm of the gas leakage trend value are multiplied, and combined with the environmental hazard coefficient, the early warning priority value is calculated through a nonlinear combination of the logarithmic function and the power function. The specific implementation formula is:

[0030]

[0031] Among them, U i (t) represents the warning priority value of the sensor at the ith position at time t; F i (t) represents the gas leakage score of the sensor at the ith position at time t; P i (t) represents the gas leakage trend value of the sensor at the i-th position at time t; ln(P i (t)+1) represents the gas leakage trend value P i (t) add 1 and take the natural logarithm; a i represents the population density of the area around the sensor at the i-th location; b i represents the building density around the sensor at the ith location; c i represents the influence coefficient of important facilities near the sensor at the i-th position; a i +b i +c i Represents the power exponent.

[0032] Preferably, the S4 specifically includes:

[0033] Based on the warning priority value, the warning levels are divided into:

[0034] Low-level warning: U i (t)<α

[0035] Intermediate warning: α≤U i (t)<β

[0036] Advanced Warning: U i (t)≥β

[0037] Among them, α represents the low-level warning threshold; β represents the high-level warning threshold; U i (t) represents the warning priority value of the sensor at the i-th position at time t.

[0038] Preferably, the S4 specifically includes:

[0039] Based on the warning level, formulate an emergency response strategy, which is:

[0040] Low-level warning: monitor and record gas leaks, and regularly check sensors and gas pipe networks;

[0041] Intermediate warning: notify maintenance personnel to conduct on-site inspections and perform necessary repairs and maintenance;

[0042] Advanced warning: Immediately activate the emergency plan, evacuate personnel, close relevant pipelines, and notify relevant departments to conduct comprehensive inspection and handling.

[0043] A gas pipeline network leakage early warning system using artificial intelligence includes the following parts:

[0044] Sensor data acquisition module, data processing module, gas leakage scoring module, gas leakage trend analysis module, external data acquisition module, warning priority assessment module, warning level classification and emergency response module;

[0045] Sensor data acquisition module: collects sensor data and outputs the collected sensor data to the data processing module and the gas leakage scoring module;

[0046] Data processing module: receives sensor data from the sensor data acquisition module, processes the collected sensor data through the data fusion algorithm to obtain fused sensor data, and outputs the fused sensor data to the gas leakage scoring module;

[0047] Gas leakage scoring module: receives sensor data from the sensor data acquisition module and fused sensor data from the data processing module, uses a multi-level spatiotemporal evolutionary algorithm to calculate the gas leakage score, and outputs the gas leakage score to the gas leakage trend analysis module and the warning priority assessment module;

[0048] Gas leakage trend analysis module: receives the gas leakage score from the gas leakage scoring module, calculates the gas leakage trend value using the gas leakage trend analysis algorithm, and outputs the gas leakage trend value to the early warning priority evaluation module;

[0049] External data acquisition module: obtain external environment data and output the external environment data to the early warning priority assessment module;

[0050] Early warning priority evaluation module: receives the gas leakage score from the gas leakage scoring module, the gas leakage trend value from the gas leakage trend analysis module and the external environment data from the external data acquisition module, calculates the early warning priority value using the early warning priority evaluation algorithm, and outputs the early warning priority value to the early warning level classification and emergency response module;

[0051] Warning level classification and emergency response module: Receive the warning priority value from the warning priority assessment module, set the low-level warning threshold and the high-level warning threshold, compare them with the warning priority value, classify the warning level, and formulate emergency response strategies according to different warning levels.

[0052] The beneficial effects of the technical solution of the present invention are:

[0053] 1. Deploy different types of sensors at different locations in the gas pipeline network, such as gas concentration sensors and acoustic wave sensors, to form a distributed monitoring network, which can ensure the comprehensiveness and real-time nature of monitoring.

[0054] 2. Based on sensor data and fused sensor data, a multi-level spatiotemporal evolution algorithm is used to calculate the gas leakage score, which is co-evolved at different spatial and temporal levels. It can reflect the changes of gas leakage at different locations and times, and improve the accuracy and timeliness of gas pipeline leakage detection.

[0055] 3. Based on the gas leakage score, the gas leakage trend analysis algorithm is used to calculate the gas leakage trend value. By introducing a nonlinear adjustment factor, a sensitive analysis of the gas leakage score change is performed, which improves the response speed and accuracy of the gas pipeline leakage warning system and reduces the incidence of sudden accidents.

[0056] 4. The warning priority assessment algorithm enables the gas pipeline leakage warning system to not only focus on the gas leakage itself, but also comprehensively consider the impact of environmental factors on the gas leakage risk, ensuring that gas leaks in high-risk areas and near important facilities are given priority, thereby improving the rationality and effectiveness of emergency response.

[0057] 5. According to the warning priority value, the warning levels are divided into low, medium and high warnings, and emergency response strategies are formulated according to the warning levels. A clear emergency response strategy ensures that when a gas leak occurs, appropriate measures can be taken quickly and accurately to control the development of the situation, reduce accident losses, and improve the safety and operation efficiency of the entire gas pipeline system. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1This is a structural diagram of a gas pipeline network leakage early warning system using artificial intelligence according to the present invention;

[0059] Figure 2 This is a flow chart of a gas pipeline network leakage early warning method using artificial intelligence according to the present invention. DETAILED DESCRIPTION

[0060] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0062] The following is a detailed description of a specific scheme of a gas pipeline network leakage warning system using artificial intelligence provided by the present invention in conjunction with the accompanying drawings.

[0063] Refer to the attached Figure 1 , which shows a structural diagram of a gas pipeline network leakage early warning system using artificial intelligence provided by an embodiment of the present invention, the system includes the following parts:

[0064] Sensor data acquisition module, data processing module, gas leakage scoring module, gas leakage trend analysis module, external data acquisition module, warning priority assessment module, warning level classification and emergency response module;

[0065] Sensor data acquisition module: deploy different types of sensors at different locations in the gas pipeline network to collect sensor data in real time, and output the collected sensor data to the data processing module and the gas leakage scoring module;

[0066] Data processing module: receives sensor data from the sensor data acquisition module, processes the collected sensor data through the data fusion algorithm to obtain fused sensor data, and outputs the fused sensor data to the gas leakage scoring module;

[0067] Gas leakage scoring module: receives sensor data from the sensor data acquisition module and fused sensor data from the data processing module, uses a multi-level spatiotemporal evolutionary algorithm to calculate the gas leakage score, and outputs the gas leakage score to the gas leakage trend analysis module and the warning priority assessment module;

[0068] Gas leakage trend analysis module: receives the gas leakage score from the gas leakage scoring module, calculates the gas leakage trend value using the gas leakage trend analysis algorithm, and outputs the gas leakage trend value to the early warning priority evaluation module;

[0069] External data acquisition module: access external data sources, obtain external environmental data in real time, such as population density, building density and distribution of important facilities, and output the external environmental data to the early warning priority assessment module;

[0070] Early warning priority evaluation module: receives the gas leakage score from the gas leakage scoring module, the gas leakage trend value from the gas leakage trend analysis module and the external environment data from the external data acquisition module, calculates the early warning priority value using the early warning priority evaluation algorithm, and outputs the early warning priority value to the early warning level classification and emergency response module;

[0071] Warning level classification and emergency response module: Receive the warning priority value from the warning priority assessment module, set the low-level warning threshold and the high-level warning threshold, compare them with the warning priority value, classify the warning level, and formulate emergency response strategies according to different warning levels.

[0072] Refer to the attached Figure 2 , which shows a flow chart of a gas pipeline network leakage early warning method using artificial intelligence provided by an embodiment of the present invention, the method comprising the following steps:

[0073] S1. Collect sensor data, process the sensor data through a data fusion algorithm, and obtain fused sensor data; based on the sensor data and the fused sensor data, use a multi-level spatiotemporal evolution algorithm to calculate the gas leak score;

[0074] By deploying different types of sensors at different locations in the gas pipeline network, such as gas concentration sensors and acoustic wave sensors, a distributed monitoring network is formed to collect sensor data in real time. The collected sensor data is processed through a data fusion algorithm to obtain fused sensor data. The specific calculation formula is:

[0075]

[0076] Among them, D f，i (t) represents the sensor data of the sensor at the i-th position after fusion at time t; represents the weighted summation of n sensor data of the sensor at the i-th position; w i，j represents the weight of the jth sensor at the ith position, which is used to measure the importance of sensor data; D i，j (t) represents the sensor data of the ,th sensor at the ,th position at time t;

[0077] Based on the sensor data and the fused sensor data, a multi-level spatiotemporal evolutionary algorithm is used to calculate the gas leak score;

[0078] The core of the multi-level spatiotemporal evolutionary algorithm lies in the co-evolution of spatial and temporal levels, and the identification of gas pipeline network leaks is achieved through data fusion and layer-by-layer optimization in different dimensions.

[0079] The co-evolution of the spatial hierarchy is to perform weighted averaging of sensor data of all neighborhoods of the location at each spatial hierarchy. The weight distribution is based on the contribution value of each sensor in the neighborhood set to ensure that each sensor data can be reasonably used. The neighborhood set is determined based on the geographical location relationship between sensors to ensure that each monitoring location is scored at the spatial hierarchy.

[0080] The co-evolution of the time level is to perform weighted averaging of the sensor data of the monitoring location within the time window set for each time level. The determination of the time window set is based on the division of time. The sensor data within a period of time are classified into the same set, and different weights are assigned according to timeliness to ensure that the time level score can reflect the changes of the monitoring location in different time periods.

[0081] The specific calculation formula for gas leak score is:

[0082] F i (t) = S i (t)·T i (t)

[0083] Among them, F i (t) represents the gas leakage score of the sensor at the ith position at time t, reflecting the degree of gas leakage; S i (t) represents the spatial hierarchical score of the sensor at the ith position at time t; T i (t) represents the temporal hierarchical score of the sensor at the ith position at time t; the specific calculation formula of the spatial hierarchical score is:

[0084]

[0085] in, represents the weighted sum of the scores of the L spatial levels of the sensor at the i-th position; w k Represents the weight of the k-th spatial level, which is used to measure the importance of the sensor data at the k-th spatial level, satisfying The size of the neighborhood set of the sensor at the i-th position at the k-th spatial level can be set according to the specific implementation scenario and is not limited here; represents the sum of the sensor data of all neighborhood sets of the sensor at the i-th position at the k-th spatial level; D i，j (t) represents the sensor data of the jth sensor at the i-th position at time t;

[0086] The specific calculation formula for time level scoring is:

[0087]

[0088] in, represents the weighted sum of the scores of the H time levels of the sensor at the i-th position; v m Represents the weight of the mth time level, which is used to measure the importance of the sensor data at the mth time level, satisfying |T i，m | represents the size of the time window set of the sensor at the i-th position at the m-th time level, which can be set according to the specific implementation scenario and is not limited here; It represents the sum of the sensor data of the sensor at the i-th position in all time window sets at the m-th time level; Df,i(t) represents the sensor data of the sensor at the i-th position after fusion at time t;

[0089] S2. Based on the gas leakage score, a gas leakage trend value is calculated using a gas leakage trend analysis algorithm;

[0090] Based on the gas leakage score, the gas leakage trend value is calculated using the gas leakage trend analysis algorithm to predict the potential gas leakage risk;

[0091] The gas leakage trend analysis algorithm evaluates the changing trend of the gas leakage status by calculating the changing rate of the gas leakage score in the past period of time, introduces a nonlinear adjustment factor, and makes nonlinear adjustments to the changes of gas leakage scores of different degrees, thereby ensuring that it is sensitive to the changes of gas leakage scores close to the gas leakage score threshold, timely capturing potential risks, and improving the response speed and accuracy of the gas pipeline network leakage early warning system;

[0092] The specific calculation formula for the gas leakage trend value is:

[0093]

[0094] Among them, P i (t) represents the gas leakage trend value of the sensor at the i-th position at time t, which is used to predict the potential gas leakage risk; F i (t) represents the gas leakage score of the sensor at the i-th position at time t, reflecting the degree of gas leakage; represents the rate of change of the gas leakage score of the sensor at the ith position within the time period [t-τ, t]; ΔF i (t-τ, t) represents the change in the gas leakage score of the sensor at the ith position within the time period [t-τ, t]; τ represents the length of the time period; e represents the base of the natural logarithm; F′ represents the gas leakage score threshold, which can be set according to the specific implementation scenario and is not limited here; Represents the nonlinear adjustment factor used to adjust the gas leak score by the amount of change ΔF i (t-τ, t) is scaled so that when the gas leakage score is close to the gas leakage score threshold, the change in the gas leakage score has a significant impact on the gas leakage trend value;

[0095] The gas leakage trend value calculated by the gas leakage trend analysis algorithm can effectively reduce the false alarm rate, significantly improve the reliability of the gas pipeline network leakage warning system, which is of great significance for the daily operation and maintenance of the gas pipeline network, can reduce unnecessary shutdown and inspection costs, and improve the overall operation efficiency of the gas pipeline network;

[0096] S3. Obtain external environmental data, and calculate the warning priority value using the warning priority evaluation algorithm based on the gas leakage score, the gas leakage trend value and the external environmental data;

[0097] In order to comprehensively assess the risk of gas leakage, it is necessary to consider the environmental factors around the leakage point and obtain external environmental data in real time by accessing external data sources, such as population density, building density, and distribution of important facilities.

[0098] Based on the gas leak score, gas leak trend value and external environment data, the warning priority value is calculated using the warning priority evaluation algorithm;

[0099] The warning priority evaluation algorithm multiplies the gas leakage score and the logarithm of the gas leakage trend value, and then combines it with the environmental hazard coefficient, and calculates the warning priority value through a nonlinear combination of logarithmic function and power function to ensure the most timely and accurate response to gas leakage risks in different scenarios;

[0100]

[0101] Among them, U i (t) represents the warning priority value of the sensor at the ith position at time t, reflecting the severity and urgency of the gas leakage risk, and is used to determine the response measures of the gas pipeline network leakage warning system; F i (t) represents the gas leakage score of the sensor at the ith position at time t, reflecting the degree of gas leakage; P i(t) represents the gas leakage trend value of the sensor at the i-th position at time t, which is used to predict the potential gas leakage risk; ln(P i (t)+1) represents the gas leakage trend value P i (t) add 1 and take the natural logarithm to convert the gas leakage trend value into a logarithmic scale to enhance the response sensitivity and ensure the validity of the logarithmic operation; a i represents the population density of the area around the sensor at the ith position, reflecting the potential impact of gas leakage on the regional population; b i represents the building density around the sensor at the ith position, reflecting the potential impact of gas leakage on regional buildings; c i represents the influence coefficient of important facilities near the sensor at the ith position, such as hospitals and schools; i +b i +c i Represents the power index, which is used to amplify or reduce the impact of the gas leakage score and gas leakage trend value;

[0102] By considering the environmental factors around the gas leak point, the warning priority assessment algorithm not only focuses on the gas leak itself, but also assesses the potential hazards that the gas leak may bring, ensuring that gas leaks near densely populated areas and important facilities receive priority attention and timely response;

[0103] S4. Set low-level warning thresholds and high-level warning thresholds, compare them with the warning priority values, divide the warning levels, and formulate emergency response strategies according to different warning levels;

[0104] Set the low-level warning threshold α and high-level warning threshold β, compare them with the warning priority value, divide the warning level, and clarify the severity and urgency of the gas leakage risk. The specific division is as follows:

[0105] Low-level warning: U i (t)<α

[0106] Intermediate warning: α≤U i (t)<β

[0107] Advanced Warning: U i (t)≥β

[0108] The low-level warning threshold and the high-level warning threshold can be set according to the specific implementation scenario and are not limited here; the warning level is used to guide the emergency response strategy;

[0109] The emergency response strategy is as follows:

[0110] Low-level warning: monitor and record gas leaks, and regularly check sensors and gas pipe networks;

[0111] Intermediate warning: notify maintenance personnel to conduct on-site inspections and perform necessary repairs and maintenance;

[0112] Advanced warning: Immediately activate the emergency plan, evacuate nearby personnel, shut down relevant pipelines, and notify relevant departments to conduct comprehensive inspection and handling;

[0113] Formulating emergency response strategies according to different warning levels can ensure that appropriate measures are taken quickly when gas leaks are discovered, control the development of the situation, avoid the expansion and spread of accidents, improve emergency efficiency, ensure personnel safety, reduce downtime caused by gas leaks, and ensure the stability and continuity of gas supply.

[0114] In summary, a gas pipeline leakage early warning system using artificial intelligence has been completed.

[0115] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0116] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0117] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A gas pipeline network leakage early warning method using artificial intelligence, characterized in that: The following steps are involved: S1: Collect sensor data, process the sensor data through data fusion algorithm, and obtain fused sensor data; based on the sensor data and fused sensor data, use multi-level spatiotemporal evolution algorithm to calculate the gas leakage score; S2: Based on the gas leakage score, the gas leakage trend value is calculated using the gas leakage trend analysis algorithm; S3: Obtain external environmental data, and calculate the warning priority value using the warning priority evaluation algorithm based on the gas leakage score, the gas leakage trend value and the external environmental data; S4: Set low-level warning thresholds and high-level warning thresholds, compare them with the warning priority values, divide the warning levels, and formulate emergency response strategies according to different warning levels.

2. According to the method of claim 1, the gas pipeline network leakage early warning method using artificial intelligence is characterized in that: The S1 specifically includes: In the process of implementing the multi-level spatiotemporal evolutionary algorithm, the spatial level and the temporal level are co-evolved. Through data fusion and layer-by-layer optimization in different dimensions, gas pipeline network leaks are identified and gas leak scores are obtained.

3. The method for early warning of gas pipeline leakage using artificial intelligence according to claim 2 is characterized in that: The S1 specifically includes: The co-evolution of the spatial level is to obtain a spatial level score by weighted averaging the sensor data for each spatial level; the co-evolution of the temporal level is to obtain a temporal level score by weighted averaging the sensor data of the monitoring position within the time window set for each temporal level and classifying the sensor data within a period of time into the same set.

4. The method for early warning of gas pipeline leakage using artificial intelligence according to claim 3 is characterized in that: The S1 specifically includes: The specific calculation formula of spatial hierarchy score is: Among them, S i (t) represents the spatial hierarchy score of the sensor at the ith position at time t; represents the weighted sum of the scores of the L spatial levels of the sensor at the i-th position; w k represents the weight of the kth spatial level, satisfying Represents the size of the neighborhood set of the sensor at the i-th position at the k-th spatial level; represents the sum of sensor data of all neighborhood sets of the sensor at the i-th position at the k-th spatial level; Di, j (t) represents the sensor data of the ,th sensor at the ,th position at time t; The specific calculation formula for time level scoring is: Among them, T i (t) represents the temporal level score of the sensor at the ith position at time t; represents the weighted sum of the scores of the H time levels of the sensor at the i-th position; v m Represents the weight of the mth time level, satisfying |T i，m | represents the size of the time window set of the sensor at the i-th position at the m-th time level; It means summing the sensor data of the sensor at the i-th position in all time windows at the m-th time level; D f，i (t) represents the sensor data of the sensor at the i-th position after fusion at time t.

5. The method for early warning of gas pipeline leakage using artificial intelligence according to claim 1, characterized in that: The S2 specifically includes: In the process of implementing the gas leakage trend analysis algorithm, the change rate of the gas leakage score is calculated to evaluate the change trend of the gas leakage situation. A nonlinear adjustment factor is introduced to make nonlinear adjustments to the changes in gas leakage scores of different degrees to obtain the gas leakage trend value. The specific implementation formula is: Among them, P i (t) represents the gas leakage trend value of the sensor at the i-th position at time t; F i (t) represents the gas leakage score of the sensor at the i-th position at time t; represents the rate of change of the gas leakage score of the sensor at the ith position within the time period [t-τ, t]; ΔF i (t-τ, t) represents the change in the gas leakage score of the sensor at the ith position within the time period [t-τ, t]; τ represents the length of the time period; e represents the base of the natural logarithm; F′ represents the gas leakage score threshold; Represents the nonlinear adjustment factor.

6. The method for early warning of gas pipeline leakage using artificial intelligence according to claim 1, characterized in that: The S3 specifically includes: In the implementation process of the early warning priority evaluation algorithm, the gas leakage score and the logarithm of the gas leakage trend value are multiplied, and combined with the environmental hazard coefficient, the early warning priority value is calculated through a nonlinear combination of the logarithmic function and the power function. The specific implementation formula is: Among them, U i (t) represents the warning priority value of the sensor at the ith position at time t; F i (t) represents the gas leakage score of the sensor at the ith position at time t; P i (t) represents the gas leakage trend value of the sensor at the i-th position at time t; ln(P i (t)+1) represents the gas leakage trend value P i (t) add 1 and take the natural logarithm; a i represents the population density of the area around the sensor at the i-th location; b i represents the building density around the sensor at the i-th location; c i represents the influence coefficient of important facilities near the sensor at the i-th position; a i +b i +c i Represents the power exponent.

7. The method for early warning of gas pipeline leakage using artificial intelligence according to claim 1, characterized in that: The S4 specifically includes: Based on the warning priority value, the warning levels are divided into: Low-level warning: U i (t)<α Intermediate warning: α≤U i (t)<β Advanced Warning: U i (t)≥β Among them, α represents the low-level warning threshold; β represents the high-level warning threshold; U i (t) represents the warning priority value of the sensor at the i-th position at time t.

8. The method for early warning of gas pipeline leakage using artificial intelligence according to claim 7 is characterized in that: The S4 specifically includes: Based on the warning level, formulate an emergency response strategy, which is: Low-level warning: monitor and record gas leaks, and regularly check sensors and gas pipe networks; Intermediate warning: notify maintenance personnel to conduct on-site inspections and perform necessary repairs and maintenance; Advanced warning: Immediately activate the emergency plan, evacuate personnel, close relevant pipelines, and notify relevant departments to conduct comprehensive inspection and handling.

9. A gas pipeline network leakage early warning system using artificial intelligence, applied to the gas pipeline network leakage early warning method using artificial intelligence as claimed in claim 1, characterized in that: Includes the following sections: Sensor data acquisition module, data processing module, gas leakage scoring module, gas leakage trend analysis module, external data acquisition module, warning priority assessment module, warning level classification and emergency response module; Sensor data acquisition module: collects sensor data and outputs the collected sensor data to the data processing module and the gas leakage scoring module; Data processing module: receives sensor data from the sensor data acquisition module, processes the collected sensor data through the data fusion algorithm to obtain fused sensor data, and outputs the fused sensor data to the gas leakage scoring module; Gas leakage scoring module: receives sensor data from the sensor data acquisition module and fused sensor data from the data processing module, uses a multi-level spatiotemporal evolutionary algorithm to calculate the gas leakage score, and outputs the gas leakage score to the gas leakage trend analysis module and the warning priority assessment module; Gas leakage trend analysis module: receives the gas leakage score from the gas leakage scoring module, calculates the gas leakage trend value using the gas leakage trend analysis algorithm, and outputs the gas leakage trend value to the early warning priority evaluation module; External data acquisition module: obtain external environment data and output the external environment data to the early warning priority assessment module; Early warning priority evaluation module: receives the gas leakage score from the gas leakage scoring module, the gas leakage trend value from the gas leakage trend analysis module and the external environment data from the external data acquisition module, calculates the early warning priority value using the early warning priority evaluation algorithm, and outputs the early warning priority value to the early warning level classification and emergency response module; Warning level classification and emergency response module: Receive the warning priority value from the warning priority assessment module, set the low-level warning threshold and the high-level warning threshold, compare them with the warning priority value, classify the warning level, and formulate emergency response strategies according to different warning levels.

Images