A gas data risk early warning method and system

Abstract

The present invention discloses a gas data risk warning method and system, which comprises the following steps: collecting gas pipeline leakage-related risk data and performing data preprocessing; inputting the preprocessed gas pipeline leakage-related risk data into a recurrent neural network model to predict the gas pipeline leakage risk data value; judging whether the predicted gas pipeline leakage risk data value exceeds a preset threshold value, if so, performing a gas pipeline leakage risk warning and enabling an optimal backup gas pipeline transmission path at the same time; if not, performing a historical gas pipeline leakage monitoring data value analysis for a connected time period to judge whether a linear growth trend is presented; if the predicted gas pipeline leakage risk data value for a connected time period has a linear growth trend, proceeding to the next step, if there is no linear growth trend, terminating the entire gas data risk warning process; judging that the gas pipeline leakage has an increasing trend, and continuously paying attention to the gas pipeline nodes where the gas pipeline leakage has an increasing trend.

Description

Technical Field

[0001] The present invention belongs to the technical field of gas pipeline leakage risk warning, and in particular, relates to a gas data risk warning method and system. Background Art

[0002] The urban gas distribution system is an extension of the natural gas transmission system. It is a system that supplies gas to users through various levels of gas distribution pipelines at urban gate stations. The urban gas distribution system includes urban gate stations, gas distribution pipelines, pressure regulating facilities and monitoring systems. Gas pipeline failures can be comprehensively classified into two categories: continuous leakage and instantaneous leakage.

[0003] Existing gas pipeline leakage risk warning methods usually use real-time data collected by the SCADA system for detection and analysis, and directly transmit the faulty gas pipeline nodes to the operation and maintenance personnel; however, the faulty gas pipeline nodes are in a completely paralyzed state from the gas data risk warning to the maintenance completion stage, causing a lot of economic losses and life inconveniences to the faulty gas pipeline nodes. Summary of the invention

[0004] In view of the problem that in the existing gas data risk warning system, the faulty gas pipeline node is in a completely paralyzed state from the gas data risk warning to the maintenance completion stage, and it is impossible to provide risk prediction and realize the call of spare gas pipelines, the present invention provides a gas data risk warning method and system.

[0005] In order to achieve the above technical objectives, the technical solution adopted by the present invention is as follows:

[0006] A gas data risk early warning method comprises the following steps:

[0007] S1. Collect risk data related to gas pipeline leakage and perform data preprocessing;

[0008] S2, inputting the pre-processed gas pipeline leakage related risk data into the recurrent neural network model to predict the gas pipeline leakage risk data value;

[0009] S3. Compare the above-mentioned predicted gas pipeline leakage risk data value with the preset threshold value to determine whether the predicted gas pipeline leakage risk data value exceeds the preset threshold value. If it exceeds the preset threshold value, proceed to step S4; if it does not exceed the preset threshold value, proceed to step S5.

[0010] S4. Provide a gas pipeline leakage risk warning, and traverse the gas pipeline nodes based on the gas pipeline structure characteristics to enable the optimal backup gas pipeline transmission path;

[0011] S5. Analyze the historical gas pipeline leakage monitoring data values ​​in the connected time period to determine whether there is a linear growth trend. If the gas pipeline leakage risk data values ​​predicted in the connected time period have a linear growth trend, proceed to step S6. If there is no linear growth trend, end the entire gas data risk warning process;

[0012] S6. Determine that there is an increasing trend in gas pipeline leakage, and continue to pay attention to gas pipeline nodes where there is an increasing trend in gas pipeline leakage.

[0013] Furthermore, the risk data related to gas pipeline leakage include methane gas concentration, methane gas pressure, methane gas flow, pipeline length between detection nodes, pipeline transmission time between detection nodes, and pipeline transmission cost between detection nodes.

[0014] Furthermore, the detailed steps of data preprocessing include:

[0015] S101. Supplement missing values ​​in gas pipeline leakage-related risk data through filling method;

[0016] S102, deleting the entire data of outliers in the gas pipeline node by using an abnormal data value elimination method;

[0017] S103. Convert the data in the gas pipeline leakage related risk data into data with the same dimension through normalization processing.

[0018] Further, the recurrent neural network model includes an input layer, an output layer, and a hidden layer;

[0019] The calculation formulas for the output layer and hidden layer are:

[0020] O t =g(V·S t )

[0021] S t =f(U·X t +W·S t-1 )

[0022] Among them, t is the output result at time t, S t represents the hidden layer state at time t, X t is the value of the input layer at time t, U, V, W are the weight parameters to be trained, f and g are activation functions, and different activation functions are selected according to different prediction requirements.

[0023] The activation function is preset according to the historical data change trend of methane gas concentration, methane gas pressure and methane gas flow in the gas pipeline node. For example, methane gas pressure is proportional to methane gas concentration and methane gas flow.

[0024] Furthermore, in step S4, the gas pipeline structure includes a main gas transmission node, a branch gas transmission node (detection point or gas pipeline node) and a transmission pipeline. The gas pipeline structure characteristics are regarded as multi-branch tree structure characteristics. At the same time, except that the terminal end transmission pipeline has no second transmission pipeline, at least one second transmission pipeline is provided on other main gas transmission nodes and branch gas transmission nodes.

[0025] Furthermore, the gas pipeline structure characteristics are used to traverse the gas pipeline nodes and enable the optimal backup gas pipeline transmission path. Detailed steps:

[0026] S4011. According to the path characteristics of gas transportation in a multi-branch tree structure of the gas pipeline, any one of the three types of data, including the length of the pipeline between the detection nodes, the pipeline transmission time between the detection nodes, and the pipeline transmission cost between the detection nodes, in the gas pipeline leakage-related risk data is selected as the optimal solution target for the gas pipeline transportation path;

[0027] S4012, traversing other urban gas distribution systems connected to the second transmission pipelines on the currently cut-off gas transmission main node and the branch gas transmission node;

[0028] S4013. Collect all possible transport paths according to the optimal solution target of the gas pipeline transport path, and sum the data values ​​of all possible transport paths;

[0029] S4014, the data values ​​after summing up all the transportable paths are obtained by a bubbling program, and the maximum value or the minimum value obtained by the bubbling program is obtained as the optimal solution for the gas pipeline transportation path;

[0030] S4015. According to the obtained optimal solution of the gas pipeline transmission path, the corresponding gas pipeline transmission path (i.e., the gas transmission node and the transmission pipeline) is obtained; and the optimal backup gas pipeline is enabled to transmit the gas to the user end of the faulty node.

[0031] Furthermore, the gas pipeline structure characteristics are used to traverse the gas pipeline nodes and enable the optimal backup gas pipeline transmission path. Detailed steps:

[0032] S4021. According to the path characteristics of gas transportation in a multi-branch tree structure of the gas pipeline, the pipeline length between the detection nodes in the gas pipeline leakage risk data is preset as L, the pipeline transmission time between the detection nodes is preset as T, and the pipeline transmission cost between the detection nodes is preset as S; corresponding weight coefficients W are preset for L, T and S. L , W T and W S , the objective function value of the optimal solution of the gas pipeline transmission path: X = L*W L +T*W T +S*W S ;

[0033] S4022, traversing other urban gas distribution systems connected to the second transmission pipelines on the currently cut-off gas transmission main node and the branch gas transmission node;

[0034] S4023, collecting all possible transport paths according to the optimal solution objective function of the gas pipeline transport path, and obtaining all optimal solution objective function values ​​for all possible transport paths;

[0035] S4024, using a bubbling program for all the optimal solution objective function values ​​obtained, and obtaining the minimum value obtained by the bubbling program as the optimal solution for the gas pipeline transmission path;

[0036] S4025. According to the obtained optimal solution of the gas pipeline transmission path, the corresponding gas pipeline transmission path (i.e., the gas transmission node and the transmission pipeline) is obtained; and the optimal backup gas pipeline is enabled to transmit the gas to the user end of the faulty node.

[0037] Furthermore, the method for determining whether the historical gas pipeline leakage monitoring data values ​​in the connected time period present a linear growth trend is as follows:

[0038] Select the gas pipeline measured data value of the historical connected time period corresponding to the predicted gas pipeline leakage risk data value not exceeding the preset threshold value and bring it into the linear trend analysis formula, the exponential smoothing trend analysis formula or the polynomial trend analysis formula;

[0039] Determine whether the measured data values ​​of the gas pipeline in the historical connected time period correspond to the growth law of the composite linear trend analysis formula, the exponential smoothing trend analysis formula or the polynomial trend analysis formula. If it conforms to the growth law of any of the three analysis formulas, it is determined that the gas pipeline leakage risk data value predicted in the connected time period has a linear growth trend; if not, the entire gas data risk warning process is terminated.

[0040] A gas data risk early warning system includes a gas pipeline leakage related risk data acquisition module, a data preprocessing module, a gas pipeline leakage risk data value prediction module, an over-limit judgment module, a standby gas pipeline transmission path activation module, a growth trend judgment module, a gas pipeline node attention module and a data storage module;

[0041] The gas pipeline leakage related risk data collection module is connected to the data preprocessing module for collecting gas pipeline leakage related risk data;

[0042] The data preprocessing module is in communication connection with the gas pipeline leakage risk data value prediction module, and is used to preprocess the collected gas pipeline leakage related risk data;

[0043] The gas pipeline leakage risk data value prediction module is in communication connection with the over-limit judgment module, and is used to input the pre-processed gas pipeline leakage related risk data into the recurrent neural network model to predict the gas pipeline leakage risk data value;

[0044] The over-limit judgment module is provided with communication connection with the backup gas pipeline transmission path activation module and the growth trend judgment module, and is used to compare the predicted gas pipeline leakage risk data value with the preset threshold value to judge whether the predicted gas pipeline leakage risk data value exceeds the preset threshold value;

[0045] The backup gas pipeline transmission path activation module is used to perform gas pipeline node traversal based on the predicted gas pipeline leakage risk data value exceeding the preset threshold value through gas pipeline structural characteristic analysis, and activate the optimal backup gas pipeline transmission path;

[0046] The growth trend judgment module is connected to the gas pipeline node attention module for analyzing the historical gas pipeline leakage monitoring data values ​​in the connected time period to determine whether a linear growth trend is present;

[0047] The gas pipeline node attention module is used to determine if there is an increasing trend in gas pipeline leakage and to continuously pay attention to gas pipeline nodes where there is an increasing trend in gas pipeline leakage;

[0048] The data storage module is communicatively connected with the gas pipeline leakage related risk data collection module, and is used for storing the collected gas pipeline leakage related risk data.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] Through the collection, preprocessing, prediction, threshold comparison, activation of the optimal backup gas pipeline transmission path and linear growth trend analysis of gas pipeline leakage-related risk data, the faulty gas pipeline node can be directly transmitted to the operation and maintenance personnel. At the same time, the prediction, judgment and acquisition of the optimal backup gas pipeline transmission path for the gas pipeline leakage-related risk data can be implemented to minimize the economic losses of the faulty gas pipeline node from the gas data risk warning to the maintenance completion stage, and improve the convenience of life for the end users of the faulty gas pipeline node. It also realizes the purpose of paying attention to the pipeline nodes with an increasing trend of gas pipeline leakage in advance to avoid the occurrence of gas leakage. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is an overall flow chart of a gas data risk warning method in an embodiment of the present invention;

[0052] Figure 2 A flowchart of the detailed steps of data preprocessing in an embodiment of the present invention;

[0053] Figure 3This is an overall structural block diagram of a gas data risk early warning system in an embodiment of the present invention;

[0054] Figure 4 It is a structural block diagram of the city gas distribution system in an embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to facilitate the understanding of those skilled in the art, the present invention is further described below in conjunction with embodiments and drawings. The contents mentioned in the implementation modes are not intended to limit the present invention.

[0056] like Figure 1 As shown, this embodiment provides a gas data risk warning method, including the steps of:

[0057] S1. Collect risk data related to gas pipeline leakage and perform data preprocessing;

[0058] S2, inputting the pre-processed gas pipeline leakage related risk data into the recurrent neural network model to predict the gas pipeline leakage risk data value;

[0059] S3. Compare the above-mentioned predicted gas pipeline leakage risk data value with the preset threshold value to determine whether the predicted gas pipeline leakage risk data value exceeds the preset threshold value. If it exceeds the preset threshold value, proceed to step S4; if it does not exceed the preset threshold value, proceed to step S5.

[0060] S4. Provide a gas pipeline leakage risk warning, and traverse the gas pipeline nodes based on the gas pipeline structure characteristics to enable the optimal backup gas pipeline transmission path;

[0061] S5. Analyze the historical gas pipeline leakage monitoring data values ​​in the connected time period to determine whether there is a linear growth trend. If the gas pipeline leakage risk data values ​​predicted in the connected time period have a linear growth trend, proceed to step S6. If there is no linear growth trend, end the entire gas data risk warning process;

[0062] S6. Determine that there is an increasing trend in gas pipeline leakage, and continue to pay attention to gas pipeline nodes where there is an increasing trend in gas pipeline leakage.

[0063] The risk data related to gas pipeline leakage include methane gas concentration, methane gas pressure, methane gas flow, pipeline length between detection nodes, pipeline transmission time between detection nodes and pipeline transmission cost between detection nodes.

[0064] like Figure 2 As shown in Figure 2, the detailed steps of data preprocessing include:

[0065] S101. Supplement missing values ​​in gas pipeline leakage-related risk data through filling method;

[0066] The missing values ​​are filled by selecting the average value of 5-10 adjacent data of the same type in the gas pipeline leakage risk data;

[0067] The average value calculation formula is:

[0068]

[0069] in: It is the sum of n adjacent data of the same type in the gas pipeline leakage related risk data; n is the number of data of the same type.

[0070] S102, deleting the entire data of outliers in the gas pipeline node by using an abnormal data value elimination method;

[0071] If one type of data or multiple types of data in the methane gas concentration, methane gas pressure and methane gas flow in the gas pipeline node are abnormal, all types of data of the methane gas concentration, methane gas pressure and methane gas flow collected at the gas pipeline node at the current moment will be deleted.

[0072] S103. Convert the data in the gas pipeline leakage related risk data into data with the same dimension through normalization processing.

[0073] In the normalization process, deviation standardization is used to standardize the risk data related to gas pipeline leakage. The calculation formula is as follows:

[0074]

[0075] Where: x min is the minimum value in the data sample, x max is the maximum value in the data sample, and x* is the normalized data. After normalization, the data values ​​of the risk related to gas pipeline leakage fall between [0,1].

[0076] Further, the recurrent neural network model includes an input layer, an output layer, and a hidden layer;

[0077] The calculation formulas for the output layer and hidden layer are:

[0078]

[0079] Among them, t is the output result at time t, S t represents the hidden layer state at time t, X t is the value of the input layer at time t, U, V, W are the weight parameters to be trained, f and g are activation functions, and different activation functions are selected according to different prediction requirements.

[0080] The activation function is preset according to the historical data change trend of methane gas concentration, methane gas pressure and methane gas flow in the gas pipeline node.

[0081] Furthermore, in step S4, the gas pipeline structure includes a main gas transmission node, a branch gas transmission node (detection point or gas pipeline node) and a transmission pipeline. The gas pipeline structure characteristics are regarded as multi-branch tree structure characteristics. At the same time, except for the terminal end transmission pipeline which does not have a second transmission pipeline, other main gas transmission nodes and branch gas transmission nodes are provided with at least a second transmission pipeline. By setting at least a second transmission pipeline on other main gas transmission nodes and branch gas transmission nodes, in coordination with the city gas distribution system, after a leak occurs between the current nodes, the original transmission pipeline is cut off by a safety valve or a gas switch valve, and the second transmission pipeline is dispatched to transmit gas to the current main gas transmission node and the branch gas transmission node. It prevents the original transmission pipeline from being cut off and before the maintenance is completed, and there is a second transmission pipeline to transmit gas.

[0082] Furthermore, the gas pipeline structure characteristics are used to traverse the gas pipeline nodes and enable the optimal backup gas pipeline transmission path. Detailed steps:

[0083] S4011. According to the path characteristics of gas transportation in a multi-branch tree structure of the gas pipeline, any one of the three types of data, including the length of the pipeline between the detection nodes, the pipeline transmission time between the detection nodes, and the pipeline transmission cost between the detection nodes, in the gas pipeline leakage-related risk data is selected as the optimal solution target for the gas pipeline transportation path;

[0084] S4012, traversing other urban gas distribution systems connected to the second transmission pipelines on the currently cut-off gas transmission main node and the branch gas transmission node;

[0085] S4013. Collect all possible transport paths according to the optimal solution target of the gas pipeline transport path, and sum the data values ​​of all possible transport paths;

[0086] S4014, the data values ​​after summing up all the transportable paths are obtained by a bubbling program, and the maximum value or the minimum value obtained by the bubbling program is obtained as the optimal solution for the gas pipeline transportation path;

[0087] S4015. According to the obtained optimal solution of the gas pipeline transmission path, the corresponding gas pipeline transmission path (i.e., the gas transmission node and the transmission pipeline) is obtained; and the optimal backup gas pipeline is enabled to transmit the gas to the user end of the faulty node.

[0088] Furthermore, the gas pipeline structure characteristics are used to traverse the gas pipeline nodes and enable the optimal backup gas pipeline transmission path. Detailed steps:

[0089] S4021. According to the path characteristics of gas transportation in a multi-branch tree structure of the gas pipeline, the pipeline length between the detection nodes in the gas pipeline leakage risk data is preset as L, the pipeline transmission time between the detection nodes is preset as T, and the pipeline transmission cost between the detection nodes is preset as S; corresponding weight coefficients W are preset for L, T and S. L , W T and W S , the objective function value of the optimal solution of the gas pipeline transmission path: X = L*W L +T*W T +S*W S ;

[0090] S4022, traverse other city gas distribution systems that are currently disconnected from the second transmission pipeline on the main gas transmission node and the branch gas transmission node; the existing city gas distribution system is as follows: Figure 4 shown.

[0091] S4023, collecting all possible transport paths according to the optimal solution objective function of the gas pipeline transport path, and obtaining all optimal solution objective function values ​​for all possible transport paths;

[0092] S4024, using a bubbling program for all the optimal solution objective function values ​​obtained, and obtaining the minimum value obtained by the bubbling program as the optimal solution for the gas pipeline transmission path;

[0093] S4025. According to the obtained optimal solution of the gas pipeline transmission path, the corresponding gas pipeline transmission path (i.e., the gas transmission node and the transmission pipeline) is obtained; and the optimal backup gas pipeline is enabled to transmit the gas to the user end of the faulty node.

[0094] Furthermore, the method for determining whether the historical gas pipeline leakage monitoring data values ​​in the connected time period present a linear growth trend is as follows:

[0095] Select the gas pipeline measured data value of the historical connected time period corresponding to the predicted gas pipeline leakage risk data value not exceeding the preset threshold value and bring it into the linear trend analysis formula, the exponential smoothing trend analysis formula or the polynomial trend analysis formula;

[0096] Linear trend analysis formula: Y = a + b * X;

[0097] Among them, Y is the variable to be predicted, X is the independent variable, a is the intercept, and b is the slope. The most appropriate values ​​of a and b can be obtained by the least squares method.

[0098] Exponential smoothing trend analysis formula: Y(t+1)=α*X(t)+(1-α)*Y(t);

[0099] Where Y(t+1) is the variable to be predicted, X(t) is the current observation, Y(t) is the predicted value of the previous period, and α is a smoothing constant that controls the weights of past observations and past predicted values.

[0100] Polynomial trend analysis formula: Y = a0 + a1*X + a2*X 2 +...+an*X n ;

[0101] Among them, Y is the variable to be predicted, X is the independent variable, and a0 to an are the coefficients of polynomial regression. The specific judgment method can be to select at least three historical connected time periods of gas pipeline measured data values ​​and bring them into the above analysis formula to calculate whether the two sides of the equation are equal. If they are equal, it conforms to the growth law of the analysis formula.

[0102] Determine whether the measured data values ​​of the gas pipeline in the historical connected time period correspond to the growth law of the composite linear trend analysis formula, the exponential smoothing trend analysis formula or the polynomial trend analysis formula. If it conforms to the growth law of any of the three analysis formulas, it is determined that the gas pipeline leakage risk data value predicted in the connected time period has a linear growth trend; if not, the entire gas data risk warning process is terminated.

[0103] like Figure 3 As shown, a gas data risk early warning system includes a gas pipeline leakage related risk data acquisition module, a data preprocessing module, a gas pipeline leakage risk data value prediction module, an over-limit judgment module, a backup gas pipeline transmission path activation module, a growth trend judgment module, a gas pipeline node attention module and a data storage module;

[0104] The gas pipeline leakage related risk data collection module is connected to the data preprocessing module for collecting gas pipeline leakage related risk data;

[0105] The data preprocessing module is in communication connection with the gas pipeline leakage risk data value prediction module, and is used to preprocess the collected gas pipeline leakage related risk data;

[0106] The gas pipeline leakage risk data value prediction module is in communication connection with the over-limit judgment module, and is used to input the pre-processed gas pipeline leakage related risk data into the recurrent neural network model to predict the gas pipeline leakage risk data value;

[0107] The over-limit judgment module is provided with communication connection with the backup gas pipeline transmission path activation module and the growth trend judgment module, and is used to compare the predicted gas pipeline leakage risk data value with the preset threshold value to judge whether the predicted gas pipeline leakage risk data value exceeds the preset threshold value;

[0108] The backup gas pipeline transmission path activation module is used to perform gas pipeline node traversal based on the predicted gas pipeline leakage risk data value exceeding the preset threshold value through gas pipeline structural characteristic analysis, and activate the optimal backup gas pipeline transmission path;

[0109] The growth trend judgment module is connected to the gas pipeline node attention module for analyzing the historical gas pipeline leakage monitoring data values ​​in the connected time period to determine whether a linear growth trend is present;

[0110] The gas pipeline node attention module is used to determine if there is an increasing trend in gas pipeline leakage and to continuously pay attention to gas pipeline nodes where there is an increasing trend in gas pipeline leakage;

[0111] The data storage module is communicatively connected with the gas pipeline leakage related risk data collection module, and is used for storing the collected gas pipeline leakage related risk data.

[0112] Compared with the prior art, the present invention has the following beneficial effects:

[0113] Through the collection, preprocessing, prediction, threshold comparison, activation of the optimal backup gas pipeline transmission path and linear growth trend analysis of gas pipeline leakage-related risk data, the faulty gas pipeline node can be directly transmitted to the operation and maintenance personnel. At the same time, the prediction, judgment and acquisition of the optimal backup gas pipeline transmission path for the gas pipeline leakage-related risk data can be implemented to minimize the economic losses of the faulty gas pipeline node from the gas data risk warning to the maintenance completion stage, and improve the convenience of life for the end users of the faulty gas pipeline node. It also realizes the purpose of paying attention to the pipeline nodes with an increasing trend of gas pipeline leakage in advance to avoid the occurrence of gas leakage.

[0114] The above is a detailed introduction to a gas data risk warning method and system provided by the present application. The description of the specific embodiment is only used to help understand the method and its core idea of ​​the present application. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of the present application.

Claims

1. A gas data risk early warning method, characterized in that: Includes steps: S1. Collect risk data related to gas pipeline leakage and perform data preprocessing; S2, inputting the pre-processed gas pipeline leakage related risk data into the recurrent neural network model to predict the gas pipeline leakage risk data value; S3, comparing the above-mentioned predicted gas pipeline leakage risk data value with a preset threshold value, and determining whether the predicted gas pipeline leakage risk data value exceeds the preset threshold value, if it exceeds the preset threshold value, proceeding to step S4; if it does not exceed the preset threshold value, proceeding to step S5; S4. Provide a gas pipeline leakage risk warning, and traverse the gas pipeline nodes based on the gas pipeline structure characteristics to enable the optimal backup gas pipeline transmission path; S5. Analyze the historical gas pipeline leakage monitoring data values ​​in the connected time period to determine whether there is a linear growth trend. If the gas pipeline leakage risk data values ​​predicted in the connected time period have a linear growth trend, proceed to step S6. If there is no linear growth trend, end the entire gas data risk warning process; S6. Determine that there is an increasing trend in gas pipeline leakage, and continue to pay attention to gas pipeline nodes where there is an increasing trend in gas pipeline leakage.

2. A gas data risk early warning method according to claim 1, characterized in that: The risk data related to gas pipeline leakage include methane gas concentration, methane gas pressure, methane gas flow, pipeline length between detection nodes, pipeline transmission time between detection nodes and pipeline transmission cost between detection nodes.

3. A gas data risk early warning method according to claim 2, characterized in that: The detailed steps of data preprocessing include: S101. Supplement missing values ​​in gas pipeline leakage-related risk data through filling method; S102, deleting the entire data of outliers in the gas pipeline node by using an abnormal data value elimination method; S103. Convert the data in the gas pipeline leakage related risk data into data with the same dimension through normalization processing.

4. A gas data risk early warning method according to claim 3, characterized in that: The recurrent neural network model includes an input layer, an output layer, and a hidden layer; The calculation formulas for the output layer and hidden layer are: O t =g(V·S t ) S t =f(U·X t +W·S t-1 ) Among them, t is the output result at time t, S t represents the hidden layer state at time t, X t is the value input to the input layer at time t, U, V, W are the weight parameters to be trained, f and g are activation functions, and the activation function is preset according to the historical data change trend of methane gas concentration, methane gas pressure and methane gas flow in the gas pipeline node.

5. A gas data risk early warning method according to claim 4, characterized in that: In step S4, the gas pipeline structure includes a main gas transmission node, a branch gas transmission node and a transmission pipeline. The gas pipeline structure characteristics are regarded as multi-branch tree structure characteristics. At the same time, except that the terminal end transmission pipeline has no second transmission pipeline, at least one second transmission pipeline is provided on other main gas transmission nodes and branch gas transmission nodes.

6. A gas data risk early warning method according to claim 5, characterized in that: The detailed steps of traversing gas pipeline nodes and activating the optimal backup gas pipeline transmission path are as follows: S4011. According to the path characteristics of gas transportation in a multi-branch tree structure of the gas pipeline, any one of the three types of data, including the length of the pipeline between the detection nodes, the pipeline transmission time between the detection nodes, and the pipeline transmission cost between the detection nodes, in the gas pipeline leakage-related risk data is selected as the optimal solution target for the gas pipeline transportation path; S4012, traversing other urban gas distribution systems connected to the second transmission pipelines on the currently cut-off gas transmission main node and the branch gas transmission node; S4013. Collect all possible transport paths according to the optimal solution target of the gas pipeline transport path, and sum the data values ​​of all possible transport paths; S4014, the data values ​​after summing up all the transportable paths are obtained by a bubbling program, and the maximum value or the minimum value obtained by the bubbling program is obtained as the optimal solution for the gas pipeline transportation path; S4015. According to the obtained optimal solution of the gas pipeline transmission path, the corresponding gas pipeline transmission path is obtained; and the optimal backup gas pipeline is enabled to transmit gas to the user end of the fault node.

7. A gas data risk early warning method according to claim 5, characterized in that: The detailed steps of traversing gas pipeline nodes and activating the optimal backup gas pipeline transmission path are as follows: S4021. According to the path characteristics of gas transportation in a multi-branch tree structure of the gas pipeline, the pipeline length between the detection nodes in the gas pipeline leakage risk data is preset as L, the pipeline transmission time between the detection nodes is preset as T, and the pipeline transmission cost between the detection nodes is preset as S; Preset corresponding weight coefficients W for L, T and S L , W T and W S , the objective function value of the optimal solution of the gas pipeline transmission path: X = L*W L +T*W T +S*W S ; S4022, traversing other urban gas distribution systems connected to the second transmission pipelines on the currently cut-off gas transmission main node and the branch gas transmission node; S4023, collecting all possible transport paths according to the optimal solution objective function of the gas pipeline transport path, and obtaining all optimal solution objective function values ​​for all possible transport paths; S4024, using a bubbling program for all the optimal solution objective function values ​​obtained, and obtaining the minimum value obtained by the bubbling program as the optimal solution for the gas pipeline transmission path; S4025. According to the obtained optimal solution of the gas pipeline transmission path, the corresponding gas pipeline transmission path is obtained; and the optimal backup gas pipeline is enabled to transmit gas to the user end of the faulty node.

8. A gas data risk early warning method according to claim 6 or 7, characterized in that: Method for judging whether the historical gas pipeline leakage monitoring data values ​​in a connected time period show a linear growth trend: Select the gas pipeline measured data value of the historical connected time period corresponding to the predicted gas pipeline leakage risk data value not exceeding the preset threshold value and bring it into the linear trend analysis formula, the exponential smoothing trend analysis formula or the polynomial trend analysis formula; Determine whether the measured data values ​​of the gas pipeline in the historical connected time period correspond to the growth law of the composite linear trend analysis formula, the exponential smoothing trend analysis formula or the polynomial trend analysis formula. If it conforms to the growth law of any of the three analysis formulas, it is determined that the gas pipeline leakage risk data value predicted in the connected time period has a linear growth trend; if not, the entire gas data risk warning process is terminated.

9. A gas data risk early warning system, characterized in that: It includes a gas pipeline leakage related risk data collection module, a data preprocessing module, a gas pipeline leakage risk data value prediction module, an over-limit judgment module, a backup gas pipeline transmission path activation module, a growth trend judgment module, a gas pipeline node attention module and a data storage module; The gas pipeline leakage related risk data collection module is connected to the data preprocessing module for collecting gas pipeline leakage related risk data; The data preprocessing module is in communication connection with the gas pipeline leakage risk data value prediction module, and is used to preprocess the collected gas pipeline leakage related risk data; The gas pipeline leakage risk data value prediction module is in communication connection with the over-limit judgment module, and is used to input the pre-processed gas pipeline leakage related risk data into the recurrent neural network model to predict the gas pipeline leakage risk data value; The over-limit judgment module is provided with communication connection with the backup gas pipeline transmission path activation module and the growth trend judgment module, and is used to compare the predicted gas pipeline leakage risk data value with the preset threshold value to judge whether the predicted gas pipeline leakage risk data value exceeds the preset threshold value; The backup gas pipeline transmission path activation module is used to perform gas pipeline node traversal based on the predicted gas pipeline leakage risk data value exceeding the preset threshold value through gas pipeline structural characteristic analysis, and activate the optimal backup gas pipeline transmission path; The growth trend judgment module is connected to the gas pipeline node attention module for analyzing the historical gas pipeline leakage monitoring data values ​​in the connected time period to determine whether a linear growth trend is present; The gas pipeline node attention module is used to determine if there is an increasing trend in gas pipeline leakage and to continuously pay attention to gas pipeline nodes where there is an increasing trend in gas pipeline leakage; The data storage module is communicatively connected with the gas pipeline leakage related risk data collection module, and is used for storing the collected gas pipeline leakage related risk data.

Images