Enterprise data automatic construction method and system for data fusion

Abstract

The present application belongs to the technical field of natural gas pipeline network intelligent monitoring, and provides an enterprise data automatic construction method and system facing data fusion, comprising: constructing a pipeline network static GIS topology and collecting SCADA real-time operation data, discriminating multi-source gas mixing events through a flow threshold value, accurately solving a real-time spatial position of a gas mixing interface in combination with a gas transient flow equation, introducing a gas mixing interface virtual node and dynamically updating a coordinate, completing coordinate confidence degree screening through a pipeline inventory deviation check, calculating an interface moving speed to construct a dynamic pipeline network topology model, binding gas source heat value component data, realizing accurate alignment of operation data and gas quality data through space-time deduction, and finally constructing an enterprise pipeline network data atlas. The present application realizes dynamic visual tracking of a natural gas pipeline network mixing boundary and multi-source heterogeneous data fusion, and can provide reliable data support for gas metering, dispatching and operation and maintenance.

Description

Technical Field

[0001] This invention belongs to the field of intelligent monitoring technology for natural gas pipeline networks, specifically a method and system for automated construction of enterprise data oriented towards data fusion. Background Technology

[0002] City gas companies generally form multi-source complementary gas supply networks, with gas sources including conventional pipeline natural gas, imported LNG regasification gas, and shale gas. Some networks also conduct small-scale hydrogen blending and transportation tests. The calorific value, composition, and physical properties of different gas sources vary significantly. Under conditions of multi-source mixed supply, gas source switching, and pipeline peak shaving, a continuously and dynamically migrating mixed gas interface will form inside the pipeline network. The calorific value and composition properties of the media on both sides of the interface are significantly different.

[0003] Currently, the data automation construction systems of urban gas companies mainly rely on static GIS pipeline topology, SCADA real-time operation data, and business billing data to complete the fusion of multi-source data, construct an enterprise data asset map, and support business operations such as pipeline scheduling, energy billing, and operation and maintenance analysis. Existing technologies all use the fixed physical topology of the pipeline network as the sole data mapping benchmark, and the association between pipeline segments and data is fixed and unchanging in the long term.

[0004] However, in multi-source gas mixing conditions, while the physical structure of the pipeline network is fixed, the calorific value boundaries of the internal medium change dynamically. Traditional automated data construction methods using static topology mapping cannot synchronously update the dynamic virtual calorific value interface, leading to mismatches between SCADA flow measurement data and fixed calorific value parameters in the billing system. This results in spatiotemporal offsets and metering distortions in the fused dataset, reducing the consistency between pipeline operation data and revenue settlement data. Existing publicly available data fusion and automated construction solutions do not design a dedicated dynamic topology mapping mechanism for the dynamic mixing interface of multi-source pipeline networks, failing to meet the business requirements of high-precision gas data fusion. Therefore, this invention provides an automated enterprise data construction method and system for data fusion. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0006] The technical solution adopted by this invention to solve its technical problem is:

[0007] On the one hand, this invention provides a method for automated construction of enterprise data oriented towards data fusion, including:

[0008] Construct static GIS spatial topology data of the natural gas pipeline network and collect real-time SCADA flow data of each physical node of the natural gas pipeline network;

[0009] Analyze real-time flow data to determine whether a multi-source gas mixing event has occurred. If it has, calculate the real-time spatial location of the interface between different gas sources based on the pressure and flow parameters in the SCADA real-time flow data and the transient gas flow equation.

[0010] In static GIS spatial topology data, real-time spatial locations are marked as virtual nodes of the gas mixing interface, and the position coordinates of the virtual nodes of the gas mixing interface are updated over time. Based on the pressure, temperature, and natural gas flow collected from each physical node of the upstream and downstream natural gas pipeline network, a confidence analysis is performed on the position coordinates of the updated virtual nodes of the gas mixing interface to filter valid position coordinates. Based on the valid position coordinates, the moving speed of the virtual nodes of the gas mixing interface is calculated to construct a dynamic pipeline network topology model.

[0011] Obtain the gas source calorific value component data from the natural gas revenue billing system, and bind the gas source calorific value component data with the virtual node of the gas mixing interface in the dynamic pipeline topology model;

[0012] In the dynamic pipeline topology model, the expected timestamps of the virtual nodes at the gas mixing interface reaching each physical node of the downstream natural gas pipeline are calculated based on the moving speed of the virtual nodes at the gas mixing interface. Using the expected timestamps as a benchmark, the real-time flow data of the downstream physical nodes are spatiotemporally aligned with the bound gas source calorific value component data to construct the enterprise pipeline data map.

[0013] As a further improvement of the present invention, the specific process of constructing static GIS spatial topology data of the natural gas pipeline network is as follows:

[0014] The spatial coordinates, diameter, material, wall thickness, and location information of each physical node of the natural gas pipeline network are obtained. The physical nodes are used as vertices of the graph structure, and the pipelines connecting adjacent physical nodes are used as edges of the graph structure. The physical length and elevation difference of each edge are calculated based on the spatial coordinates of the pipeline. The physical length, elevation difference, diameter, material, and wall thickness are used as static attributes of the corresponding edges, and the location information is used as static attributes of the corresponding vertices, thus constructing static GIS spatial topology data.

[0015] As a further improvement of the present invention, the specific process of collecting real-time SCADA flow data from each physical node of the natural gas pipeline network is as follows:

[0016] The SCADA system of the natural gas pipeline network acquires the sensor operating parameters at each physical node. The sensor operating parameters include real-time pressure, real-time temperature, real-time natural gas flow, and valve opening status. The sensor operating parameters are then timestamped and outlier removed. The processed sensor operating parameters with uniform timestamps are used as SCADA real-time stream data.

[0017] As a further improvement of the present invention, the specific process for calculating the real-time spatial position of the interface between different gas sources is as follows:

[0018] Based on static GIS spatial topology data, identify the physical nodes of the flow that are connected to two or more air intake pipelines, and obtain the real-time natural gas flow and valve opening status in the SCADA real-time flow data corresponding to the physical nodes of the flow.

[0019] Real-time monitoring of the status of each gas inlet pipeline at the same gas inlet physical node. If the real-time natural gas flow of two or more gas inlet pipelines at the same gas inlet physical node is greater than the effective gas inlet flow threshold at the same time, and the corresponding valve opening status is open; or the valve opening status of gas inlet pipelines connected to different gas sources changes from closed to open, and the real-time natural gas flow of the gas inlet pipelines connected to different gas sources is greater than the effective gas inlet flow threshold, then a multi-source gas mixing event is determined to have occurred.

[0020] Obtain the initial moment of the multi-source gas mixing event and use the position coordinates of the corresponding confluence physical node as the initial spatial position of the gas mixing interface;

[0021] Obtain the real-time pressure and real-time natural gas flow rate from the SCADA real-time flow data corresponding to the pipeline where the gas-mixing interface is located, as well as the pipe diameter parameters from the static GIS spatial topology data. The pipe diameter parameters include the outer diameter and wall thickness of the pipeline. Calculate the internal cross-sectional area of ​​the pipeline based on the pipe diameter parameters, and calculate the actual volumetric flow rate by combining the real-time pressure. Divide the actual volumetric flow rate by the internal cross-sectional area to obtain the real-time flow velocity of the gas in the pipeline.

[0022] Using the initial moment as the starting point of time integration, the real-time flow velocity is integrated along the time axis to obtain the cumulative movement distance of the gas-mixed interface inside the pipeline. The cumulative movement distance is then mapped to static GIS spatial topology data to obtain the real-time spatial location of different gas source interfaces.

[0023] As a further improvement of the present invention, the specific process of updating the position coordinates of the virtual node at the gas-mixture interface is as follows:

[0024] At the real-time spatial location, a new vertex is inserted and marked as a virtual node of the gas-mixing interface. The original pipeline edge where the virtual node of the gas-mixing interface is currently located is deleted. Using the virtual node of the gas-mixing interface as the intermediate connection point, a first dynamic sub-pipeline edge connecting the virtual node of the gas-mixing interface with the adjacent upstream physical node and a second dynamic sub-pipeline edge connecting the virtual node of the gas-mixing interface with the adjacent downstream physical node are generated to establish an initial virtual topology. The latest real-time spatial location is received and used to cover the current coordinates of the virtual node of the gas-mixing interface. The lengths of the first dynamic sub-pipeline edge and the second dynamic sub-pipeline edge are adjusted, and the position coordinates of the virtual node of the gas-mixing interface are updated.

[0025] As a further improvement of the present invention, the specific process of filtering valid position coordinates is as follows:

[0026] Obtain the real-time pressure, real-time temperature, and real-time natural gas flow rate from the SCADA real-time flow data of the upstream and downstream physical nodes adjacent to the virtual node at the gas mixing interface at the corresponding time.

[0027] Based on the real-time pressure and temperature of upstream and downstream physical nodes, and combined with the gas state equation, the actual gas volume in the current pipe segment is calculated.

[0028] Based on the updated position coordinates of the virtual nodes at the gas-mixing interface, and combined with the pipe diameter parameters, calculate the theoretical gas pipe inventory corresponding to the updated position coordinates of the virtual nodes at the gas-mixing interface.

[0029] Calculate the relative deviation between the actual gas tube inventory and the theoretical gas tube inventory, and determine whether the relative deviation is less than the confidence deviation threshold. If it is less than the threshold, mark the currently updated position coordinates as valid position coordinates; if it is greater than or equal to the threshold, retain the valid position coordinates from the previous time.

[0030] As a further improvement of the present invention, the specific process for calculating the moving speed of the virtual node at the gas-mixture interface is as follows:

[0031] Construct a historical trajectory sequence containing timestamps and valid location coordinates. In the historical trajectory sequence, obtain the valid location coordinates at the current moment and the valid location coordinates at the previous adjacent moment. In the static GIS spatial topology data, calculate the actual pipeline length difference between the valid location coordinates at the current moment and the valid location coordinates at the previous adjacent moment along the physical direction of the pipeline.

[0032] Divide the actual difference in pipe length by the time interval between the current moment and the previous adjacent moment to obtain the instantaneous speed;

[0033] The instantaneous moving speed is obtained for multiple consecutive calculation cycles. The arithmetic mean of the instantaneous moving speed for multiple consecutive calculation cycles is calculated using a moving average filtering algorithm. The smoothed result is used as the moving speed of the virtual node at the gas-mixing interface.

[0034] As a further improvement of the present invention, the specific process of binding the gas source calorific value component data with the virtual node of the gas-mixing interface is as follows:

[0035] Static GIS spatial topology data is used as the underlying physical framework, the selected effective location coordinates are used as spatial attributes, and the calculated movement speed is used as motion attributes. These are all encapsulated in the data structure of the virtual node of the gas-mixing interface. The encapsulated virtual node of the gas-mixing interface is injected into the underlying physical framework, and dynamic edge connection relationships between adjacent physical nodes are established. By integrating the above graph structure containing static physical nodes, dynamic virtual nodes, and dynamic edges that scale over time, a dynamic pipeline topology model that can reflect the movement state of the fluid boundary inside the pipeline network in real time is generated.

[0036] Obtain the latest gas quality analysis report for each intake air source, and extract the gas source calorific value component data from the gas quality analysis report. The gas source calorific value component data includes higher heating value, lower heating value, Wobbe index, and the molar percentage content of methane, ethane, propane, and carbon dioxide.

[0037] In the dynamic pipeline topology model, locate the virtual node of the gas-mixing interface and identify the gas source type corresponding to the upstream side of the virtual node. Add a dynamic attribute field to the data structure of the virtual node of the gas-mixing interface and write the gas source calorific value component data that matches the gas source type into the dynamic attribute field.

[0038] Establish a mapping table between the calorific value component data of the gas source and the unique identifier of the virtual node of the gas-mixing interface, and complete the data binding.

[0039] As a further improvement of the present invention, the specific process of aligning the real-time flow data of downstream physical nodes with the bound gas source calorific value component data in time and space based on the expected timestamp to construct the enterprise pipeline network data map is as follows:

[0040] In the dynamic pipeline topology model, a path tracing algorithm is used to determine the natural gas transmission path from the current location of the virtual node at the gas mixing interface to the target downstream physical node;

[0041] Along the natural gas transmission path, the physical lengths of each pipeline side are accumulated to obtain the remaining transmission distance from the virtual node at the gas-mixing interface to the target downstream physical node.

[0042] Divide the remaining transmission distance by the moving speed of the virtual node at the gas-mixing interface to calculate the remaining time for natural gas transmission.

[0043] Add the current system time to the remaining time to obtain the expected timestamp of the virtual node at the gas-mixing interface reaching the downstream physical node; if there are branches in the pipeline network resulting in multiple downstream physical nodes, repeat the above calculation process to obtain the expected timestamps corresponding to each downstream physical node.

[0044] When the system time reaches the expected timestamp, the time matching tolerance window is expanded before and after the expected timestamp, and the real-time stream data of the corresponding downstream physical node is obtained within the time matching tolerance window;

[0045] Obtain the gas source calorific value component data bound to the virtual nodes of the gas mixing interface in the dynamic pipeline network topology model;

[0046] The captured real-time flow data and gas source calorific value component data are spliced ​​together along the expected timestamp dimension to generate a multi-dimensional fused data record that includes spatial location, time node, physical flow rate and chemical calorific value.

[0047] By using downstream physical nodes as entity nodes and multi-dimensional fused data records as entity attributes, or by using multi-dimensional fused data records as independent data nodes and establishing time-dimensional association edges with downstream physical nodes, the graph mapping of heterogeneous data is completed, and the enterprise pipeline network data graph is constructed.

[0048] On the other hand, the present invention provides an automated enterprise data construction system for data fusion, comprising:

[0049] Static topology and data acquisition module: Constructs static GIS spatial topology data of the natural gas pipeline network and acquires real-time SCADA flow data of each physical node of the natural gas pipeline network;

[0050] Gas mixing determination module: Analyzes real-time flow data to determine whether a multi-source gas mixing event has occurred. If it has occurred, it calculates the real-time spatial location of the interface between different gas sources based on the pressure and flow parameters in the SCADA real-time flow data and the transient gas flow equation.

[0051] Dynamic topology construction module: In static GIS spatial topology data, real-time spatial locations are marked as virtual nodes of the gas mixing interface, and the position coordinates of the virtual nodes of the gas mixing interface are updated over time. Based on the pressure, temperature, and natural gas flow collected from each physical node of the upstream and downstream natural gas pipeline network, the confidence analysis of the position coordinates of the updated virtual nodes of the gas mixing interface is performed to filter valid position coordinates. Based on the valid position coordinates, the moving speed of the virtual nodes of the gas mixing interface is calculated to construct a dynamic pipeline network topology model.

[0052] Component binding module: Obtains gas source calorific value component data from the natural gas revenue billing system, and binds the gas source calorific value component data to the virtual node of the gas mixing interface in the dynamic pipeline topology model;

[0053] Map construction module: In the dynamic pipeline topology model, the expected timestamps of the virtual nodes at the gas mixing interface reaching each physical node of the downstream natural gas pipeline are calculated based on the moving speed of the virtual nodes at the gas mixing interface. Using the expected timestamps as a benchmark, the real-time flow data of the downstream physical nodes are spatiotemporally aligned with the bound gas source calorific value component data to construct the enterprise pipeline network data map.

[0054] The beneficial effects of this invention are as follows:

[0055] This invention relies on static topology data of GIS pipeline network and high-frequency real-time flow data of SCADA, and combines gas transient flow equations to complete the dynamic solution of gas-mixing interface. It introduces a dynamic topology mechanism of virtual nodes at the gas-mixing interface, and eliminates abnormal coordinates and corrects the moving speed through a pipeline stock mass conservation deviation verification mechanism. This solves the defects of traditional static pipeline network models that cannot represent gas source mixing and fluid boundary dynamic migration, improves the continuity and accuracy of spatial position tracking of multi-gas source mixing interface, and avoids problems such as interface tracking lag, jump, and distortion.

[0056] This invention breaks down the barriers between three types of heterogeneous data: GIS topology data, SCADA operation data, and revenue billing gas composition data. It uniquely binds the gas source calorific value, composition attributes, and dynamic virtual nodes, and achieves precise spatiotemporal alignment of operating condition data and gas composition data by relying on path tracing algorithms and timestamp extrapolation mechanisms. This changes the traditional situation where pipeline network system operation monitoring, gas composition analysis, and billing traceability are independent and cannot be linked, and realizes the integrated fusion of physical operating conditions and chemical attributes. Attached Figure Description

[0057] The invention will now be further described with reference to the accompanying drawings.

[0058] Figure 1 This is a schematic diagram illustrating the evolution of the multi-source gas mixing and dynamic pipeline topology model in an embodiment of the present invention;

[0059] Figure 2 This is a schematic diagram illustrating the principle of path tracing and expected timestamp calculation based on a dynamic pipeline network topology model in an embodiment of the present invention.

[0060] Figure 3 This is a schematic diagram of the system process of an enterprise data automation construction system oriented towards data fusion according to the present invention. Detailed Implementation

[0061] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0062] Example 1

[0063] like Figure 1As shown in the embodiment of the present invention, an automated enterprise data construction method for data fusion includes:

[0064] Step 1: Construct static GIS spatial topology data of the natural gas pipeline network and collect real-time SCADA flow data of each physical node of the natural gas pipeline network;

[0065] In some embodiments, the specific process of constructing static GIS spatial topology data of the natural gas pipeline network in step one is as follows:

[0066] By using the enterprise geographic information system interface, the spatial coordinates, diameter, material, wall thickness, and location information of each physical node of the natural gas pipeline network can be obtained.

[0067] The physical nodes are used as vertices of the graph structure, and the pipelines connecting adjacent physical nodes are used as edges of the graph structure.

[0068] The physical length and elevation difference of each edge are calculated based on the spatial coordinates of the pipeline. The physical length, elevation difference, pipe diameter, pipe material and wall thickness are used as the static attributes of the corresponding edge, and the location information is used as the static attributes of the corresponding vertex. The static GIS spatial topology data is then constructed by combining these attributes.

[0069] In some embodiments, the specific process of collecting real-time SCADA flow data of each physical node in the natural gas pipeline network in step one is as follows:

[0070] SCADA system connected to natural gas pipeline network via industrial control protocols;

[0071] According to the preset sampling frequency, the sensor operating parameters at each physical node are collected in real time, including real-time pressure, real-time temperature, real-time natural gas flow rate, and valve opening status.

[0072] The collected sensor operating parameters are timestamped and outlier removed. The processed sensor operating parameters with uniform timestamps are then stored and output as the SCADA real-time stream data.

[0073] It should be noted that the preset sampling frequency is specifically set based on the transient characteristics of gas flow in the natural gas pipeline network and the communication bandwidth of the SCADA system, typically between 1 and 5 seconds. This preset sampling frequency setting aims to ensure that pressure waves and flow rate changes caused by switching between multiple gas sources or adjusting valves are captured. This provides high temporal resolution data support for the subsequent accurate calculation of the real-time spatial position of the gas-mixture interface using the transient gas flow equation, avoiding severe lag or spatial jumps in interface position tracking due to excessively large sampling intervals.

[0074] Step 2: Analyze the real-time flow data to determine whether a multi-source gas mixing event has occurred. If it has, calculate the real-time spatial location of the interface between different gas sources based on the pressure and flow parameters in the SCADA real-time flow data and the transient gas flow equation.

[0075] In some embodiments, the specific process of analyzing real-time stream data to determine whether a multi-source gas mixing event has occurred in step two is as follows:

[0076] Based on the static GIS spatial topology data, the confluence physical nodes connected to two or more air intake pipelines are identified as potential mixing monitoring points.

[0077] Obtain the real-time natural gas flow rate and valve opening status from the SCADA real-time flow data corresponding to the merge physical node;

[0078] Set the effective intake airflow threshold;

[0079] It should be noted that the effective intake flow rate threshold is used to filter out invalid flow caused by background noise in the pipeline network and minor leaks in valves. Specifically, it is set based on the minimum baseline flow rate during the historical normal operation of the busbar physical node, and is usually set to 5% to 10% of the full scale. The setting of the effective intake flow rate threshold aims to eliminate misjudgments caused by sensor zero-point drift or valves not being closed tightly, ensuring that only genuine large-scale air source injection will be recognized by the system.

[0080] The status of each gas inlet pipeline at the same gas inlet physical node is monitored in real time. If the real-time natural gas flow rate of two or more gas inlet pipelines at the same gas inlet physical node is greater than the effective gas inlet flow rate threshold at the same time, and the corresponding valve opening status is open; or if the valve opening status of the gas inlet pipeline connected to different gas sources is detected to switch from closed to open, and the real-time natural gas flow rate of the gas inlet pipeline connected to different gas sources is greater than the effective gas inlet flow rate threshold, then a multi-source gas mixing event is determined to have occurred.

[0081] In some embodiments, if this occurs in step two, the specific process of calculating the real-time spatial location of different gas source interfaces using the gas transient flow equation based on the pressure and flow parameters in the SCADA real-time flow data is as follows:

[0082] The initial moment when the multi-source gas mixing event is determined is obtained, and the position coordinates of the corresponding confluence physical node are used as the initial spatial position of the gas mixing interface.

[0083] Obtain the real-time pressure and real-time natural gas flow rate from the SCADA real-time flow data corresponding to the pipeline where the gas-mixing interface is currently located, as well as the pipe diameter parameters from the static GIS spatial topology data;

[0084] The pipe diameter parameters include: the outer diameter of the pipe and the wall thickness of the pipe;

[0085] The internal cross-sectional area of ​​the pipeline is calculated based on the pipe diameter parameters, and the real-time pressure is combined with the actual gas state equation to convert the real-time natural gas flow rate under standard conditions into the actual volumetric flow rate under pipeline operating conditions.

[0086] For example, the formula for calculating the actual volumetric flow rate is: ,in, This is the actual volumetric flow rate. This refers to the real-time natural gas flow rate under standard conditions. and The absolute pressure and thermodynamic temperature under standard conditions. and Real-time pressure and temperature under pipeline operating conditions. The compressibility factor of natural gas;

[0087] Divide the actual volumetric flow rate by the internal cross-sectional area to obtain the real-time gas velocity in the pipeline;

[0088] Using the initial moment as the starting point of time integration, the continuity equation in the one-dimensional gas transient flow equation is used to integrate the real-time flow velocity along the time axis to obtain the cumulative movement distance of the gas-mixed interface inside the pipeline.

[0089] The cumulative movement distance is mapped to the static GIS spatial topology data, and coordinate extrapolation is performed along the spatial direction of the corresponding pipeline to obtain the real-time spatial location of different gas source interfaces.

[0090] Step 3: In the static GIS spatial topology data, mark the real-time spatial location as a virtual node of the gas mixing interface, and update the position coordinates of the virtual node of the gas mixing interface over time. Based on the pressure, temperature and natural gas flow collected from each physical node of the upstream and downstream natural gas pipeline network, perform confidence analysis on the position coordinates of the updated virtual node of the gas mixing interface, filter the valid position coordinates, calculate the moving speed of the virtual node of the gas mixing interface based on the valid position coordinates, and construct a dynamic pipeline network topology model.

[0091] In some embodiments, the specific process of marking real-time spatial locations as virtual nodes of the gas-mixing interface in static GIS spatial topology data and updating the position coordinates of the virtual nodes of the gas-mixing interface over time in step three is as follows:

[0092] In the graph structure of the static GIS spatial topology data constructed in step one, a new vertex is inserted at the real-time spatial location calculated in step two, and the new vertex is marked as the virtual node of the gas-mixing interface.

[0093] In the static GIS spatial topology data, the original pipeline edge where the gas-mixing interface virtual node is currently located is deleted, and the first dynamic sub-pipeline edge connecting the gas-mixing interface virtual node and the adjacent upstream physical node is generated with the gas-mixing interface virtual node as the intermediate connection point, and the second dynamic sub-pipeline edge connecting the gas-mixing interface virtual node and the adjacent downstream physical node is generated to establish the initial virtual topology relationship.

[0094] According to the preset calculation cycle, the latest real-time spatial position output in step two is continuously received. The latest real-time spatial position is used to cover the current coordinates of the virtual node of the gas-mixing interface, and the lengths of the first dynamic sub-pipeline side and the second dynamic sub-pipeline side are adjusted synchronously to realize the dynamic update of the position coordinates of the virtual node of the gas-mixing interface over time.

[0095] It should be noted that the preset calculation period is synchronized with the preset sampling frequency in step one, and is usually set to 1 to 5 seconds to ensure that the update frequency of the topology model is consistent with the refresh frequency of the underlying SCADA data, so as to achieve smooth dynamic tracking.

[0096] In some embodiments, the specific process of performing confidence analysis on the position coordinates of the updated virtual nodes at the gas mixing interface based on the pressure, temperature, and natural gas flow rate collected from each physical node of the upstream and downstream natural gas pipeline network, and then filtering the valid position coordinates, is as follows:

[0097] Obtain the real-time SCADA flow data of the upstream and downstream physical nodes adjacent to the virtual node of the gas mixing interface at the corresponding time, and obtain the real-time pressure, real-time temperature and real-time natural gas flow rate in the real-time SCADA flow data of the upstream and downstream physical nodes adjacent to the virtual node of the gas mixing interface at the corresponding time.

[0098] Based on the real-time pressure and temperature of the upstream and downstream physical nodes, and combined with the gas state equation, the actual gas volume in the current pipe segment is calculated.

[0099] Simultaneously, based on the updated position coordinates of the virtual node at the gas-mixing interface, and combined with the pipe diameter parameters, the theoretical gas pipe inventory corresponding to the updated position coordinates of the virtual node at the gas-mixing interface is calculated.

[0100] The relative deviation between the actual gas tube inventory and the theoretical gas tube inventory is calculated and used as a confidence index for the location coordinates.

[0101] Set a confidence deviation threshold;

[0102] It should be noted that the confidence deviation threshold is calibrated based on the comprehensive measurement error of the pipeline network instruments, and is usually set to 3% to 5%.

[0103] If the relative deviation value is less than the confidence deviation threshold, it is determined that the currently updated position coordinates conform to the law of conservation of mass in fluid dynamics, and the confidence level is high. The currently updated position coordinates are retained and marked as the valid position coordinates. If the relative deviation value is greater than or equal to the threshold, it is determined that the confidence level is low, and the coordinates are discarded, while the valid position coordinates of the previous moment are retained.

[0104] In some embodiments, the specific process of calculating the moving speed of the virtual node at the gas-mixture interface based on the effective position coordinates in step three is as follows:

[0105] Construct a historical trajectory sequence that includes timestamps and the effective location coordinates;

[0106] In the historical trajectory sequence, obtain the valid position coordinates at the current moment and the valid position coordinates at the previous adjacent moment;

[0107] In the static GIS spatial topology data, the difference in actual pipeline length between the effective position coordinates at the current moment and the effective position coordinates at the previous adjacent moment is calculated along the physical direction of the pipeline.

[0108] Divide the actual pipe length difference by the time interval between the current moment and the previous adjacent moment to obtain the instantaneous moving speed;

[0109] To eliminate data fluctuations and obtain the instantaneous moving speed over multiple consecutive calculation cycles, a moving average filtering algorithm is used to calculate the arithmetic mean of the instantaneous moving speed over multiple consecutive calculation cycles. The smoothed result is then used as the moving speed of the virtual node at the gas-mixing interface.

[0110] In some embodiments, the specific process of constructing the dynamic pipeline network topology model in step three is as follows:

[0111] The static GIS spatial topology data is used as the underlying physical framework;

[0112] The selected valid position coordinates are used as spatial attributes, and the calculated moving speed is used as motion attributes. Both are encapsulated into the data structure of the virtual node of the gas-mixing interface.

[0113] The encapsulated virtual node of the gas-mixing interface is injected into the underlying physical framework, and a dynamic edge connection relationship is established between it and adjacent physical nodes.

[0114] By integrating the graph structure that includes static physical nodes, dynamic virtual nodes, and dynamic edges that scale over time, a dynamic pipeline topology model is generated that can reflect the movement state of fluid boundaries within the pipeline network in real time.

[0115] Combination Figure 1As shown, in the static GIS spatial topology data, the intake pipelines of gas source A and gas source B converge at the confluence physical node and connect to the downstream physical node. At this time, the pipeline network consists of fixed physical nodes and static pipeline edges. When a multi-source gas mixing event is determined in step two, the calculated real-time spatial location is instantiated as a virtual node of the gas mixing interface in the dynamic pipeline network topology model. The virtual node deletes the original static pipeline edges in the topology graph and dynamically generates a first dynamic sub-pipeline edge and a second dynamic sub-pipeline edge. As the fluid advances, the virtual node continuously moves to the right in the topology graph, and the length attributes of the two dynamic sub-pipeline edges change inversely over time (the first dynamic sub-pipeline edge lengthens, and the second dynamic sub-pipeline edge shortens), thus accurately mapping the physical movement state of the fluid boundary inside the pipeline network at the graph structure level.

[0116] Step 4: Obtain the gas source calorific value component data from the natural gas revenue billing system, and bind the gas source calorific value component data to the virtual node of the gas mixing interface in the dynamic pipeline topology model;

[0117] In some embodiments, the specific process of obtaining the gas source calorific value component data from the natural gas revenue billing system and binding the gas source calorific value component data with the virtual node of the gas mixing interface in the dynamic pipeline topology model is as follows:

[0118] Through the enterprise data bus interface, the latest gas quality analysis reports of each gas source are obtained by periodically accessing the natural gas revenue billing system or the gas source handover metering system.

[0119] The gas source calorific value component data are obtained from the gas quality analysis report. The gas source calorific value component data includes higher calorific value, lower calorific value, Wobbe index, and the molar percentage content of methane, ethane, propane, and carbon dioxide.

[0120] In the dynamic pipeline topology model, the virtual node of the gas mixing interface is located, and the gas source type corresponding to the upstream side of the virtual node is identified;

[0121] The gas source type refers to the specific gas source of the injected pipeline network (e.g., imported LNG, conventional natural gas or shale gas), and different gas source types correspond to different proportions of components such as methane, ethane, propane, and carbon dioxide.

[0122] A dynamic attribute field is added to the data structure of the virtual node at the gas-mixing interface, and the calorific value component data of the gas source that matches the gas source type is written into the dynamic attribute field;

[0123] A mapping table is established between the calorific value component data of the gas source and the unique identifier of the virtual node of the gas-mixing interface, and the data binding is completed so that when the virtual node of the gas-mixing interface moves in the dynamic pipeline topology model, it can carry and transmit the calorific value and component characteristics of the corresponding gas source on the upstream side.

[0124] Step 5: In the dynamic pipeline topology model, calculate the expected timestamps of the virtual nodes at the gas mixing interface reaching each physical node of the downstream natural gas pipeline network based on their moving speed. Using the expected timestamps as a benchmark, align the real-time flow data of the downstream physical nodes with the bound gas source calorific value component data in time and space to construct the enterprise pipeline network data map.

[0125] In some embodiments, the specific process of calculating the expected timestamps of the virtual nodes at the gas-mixing interface reaching the physical nodes of the downstream natural gas pipeline network in step five, based on the moving speed of the virtual nodes at the gas-mixing interface in the dynamic pipeline topology model, is as follows:

[0126] In the dynamic pipeline topology model, based on the connectivity of the graph structure and the gas flow direction, a path tracing algorithm is used to determine the natural gas transmission path from the current position of the virtual node at the gas-mixing interface to the target downstream physical node.

[0127] Along the natural gas transmission path, the physical lengths of each pipeline side are accumulated to obtain the remaining transmission distance from the virtual node of the gas-mixing interface to the target downstream physical node.

[0128] The remaining transmission distance is divided by the moving speed of the virtual node at the gas-mixing interface to calculate the remaining time required for natural gas transmission.

[0129] Add the current system time to the remaining time to obtain the expected timestamp of the virtual node at the gas-mixing interface reaching the downstream physical node; if there are branches in the pipeline resulting in multiple downstream physical nodes, repeat the above calculation process to obtain the set of expected timestamps corresponding to each downstream physical node.

[0130] Combination Figure 2 As shown, the dynamic pipeline topology model fully leverages the connectivity advantages of the graph structure when calculating the expected timestamps. When there are pipeline branches (such as branch physical nodes) downstream of the virtual node at the gas-mixing interface, a path tracing algorithm is used to obtain all possible natural gas transmission paths. For downstream physical node 1, its remaining transmission distance is L1+L2; for downstream physical node 2, its remaining transmission distance is L1+L3. By dividing the above accumulated distances by the moving speed of the virtual node, the remaining time for the fluid to reach the end of different branches can be calculated in parallel, thus obtaining the set of expected timestamps corresponding to each downstream physical node. and This solved the problem of spatiotemporal alignment of multiple nodes in complex ring or branch pipe networks.

[0131] In some embodiments, the specific process of constructing an enterprise pipeline network data map in step five, by spatiotemporally aligning the real-time flow data of downstream physical nodes with the bound gas source calorific value component data based on the expected timestamp, is as follows:

[0132] Set the time matching tolerance window;

[0133] It should be noted that the time matching tolerance window is used to absorb the network transmission delay of the SCADA system and the small fluctuations in the flow rate calculation process. It is usually set to 10 to 30 seconds before or after the expected timestamp.

[0134] When the system time reaches the expected timestamp, the time matching tolerance window is expanded before and after the expected timestamp, and the real-time stream data of the corresponding downstream physical node is obtained within the time matching tolerance window;

[0135] Synchronously acquire the gas source calorific value component data bound to the virtual nodes of the gas mixing interface in the dynamic pipeline topology model;

[0136] The captured real-time stream data and the gas source calorific value component data are spliced ​​together along the dimension of the expected timestamp to generate a multi-dimensional fused data record containing spatial location, time node, physical flow rate and chemical calorific value.

[0137] The spatiotemporal alignment results based on the dynamic pipeline network topology model are persistently stored. Specifically, in an enterprise-level graph database, the downstream physical nodes are used as entity nodes, and the multidimensional fused data records are used as entity attributes. Alternatively, the multidimensional fused data records can be used as independent data nodes and time-dimensional association edges can be established with the downstream physical nodes to complete the graph mapping of heterogeneous data. Finally, the enterprise pipeline network data graph with dynamic spatiotemporal traceability capability is constructed.

[0138] It should be noted that the dynamic pipeline topology model is a real-time simulation engine running in memory, while the enterprise pipeline data map is an external storage system (such as Neo4j) that persistently stores the merged data. By constructing the enterprise pipeline data map, the transient calculation results of the dynamic pipeline topology model are transformed into historical traceability ledgers that can be queried by the natural gas revenue billing system at any time, realizing a data closed loop from real-time calculation to business application.

[0139] Example 2

[0140] like Figure 3 As shown, based on the specific implementation process of Embodiment 1, the present invention provides an enterprise data automation construction system for data fusion, comprising:

[0141] Static topology and data acquisition module: Constructs static GIS spatial topology data of the natural gas pipeline network and acquires real-time SCADA flow data of each physical node of the natural gas pipeline network;

[0142] Gas mixing determination module: Analyzes real-time flow data to determine whether a multi-source gas mixing event has occurred. If it has occurred, it calculates the real-time spatial location of the interface between different gas sources based on the pressure and flow parameters in the SCADA real-time flow data and the transient gas flow equation.

[0143] Dynamic topology construction module: In static GIS spatial topology data, real-time spatial locations are marked as virtual nodes of the gas mixing interface, and the position coordinates of the virtual nodes of the gas mixing interface are updated over time. Based on the pressure, temperature, and natural gas flow collected from each physical node of the upstream and downstream natural gas pipeline network, the confidence analysis of the position coordinates of the updated virtual nodes of the gas mixing interface is performed to filter valid position coordinates. Based on the valid position coordinates, the moving speed of the virtual nodes of the gas mixing interface is calculated to construct a dynamic pipeline network topology model.

[0144] Component binding module: Obtains gas source calorific value component data from the natural gas revenue billing system, and binds the gas source calorific value component data to the virtual node of the gas mixing interface in the dynamic pipeline topology model;

[0145] Map construction module: In the dynamic pipeline topology model, the expected timestamps of the virtual nodes at the gas mixing interface reaching each physical node of the downstream natural gas pipeline are calculated based on the moving speed of the virtual nodes at the gas mixing interface. Using the expected timestamps as a benchmark, the real-time flow data of the downstream physical nodes are spatiotemporally aligned with the bound gas source calorific value component data to construct the enterprise pipeline network data map.

[0146] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for automating enterprise data construction oriented towards data fusion, characterized in that: include: Construct static GIS spatial topology data of the natural gas pipeline network and collect real-time SCADA flow data of each physical node of the natural gas pipeline network; Analyze real-time flow data to determine whether a multi-source gas mixing event has occurred. If it has, calculate the real-time spatial location of the interface between different gas sources based on the pressure and flow parameters in the SCADA real-time flow data and the transient gas flow equation. In static GIS spatial topology data, real-time spatial locations are marked as virtual nodes of the gas mixing interface, and the position coordinates of the virtual nodes of the gas mixing interface are updated over time. Based on the pressure, temperature, and natural gas flow collected from each physical node of the upstream and downstream natural gas pipeline network, a confidence analysis is performed on the position coordinates of the updated virtual nodes of the gas mixing interface to filter valid position coordinates. Based on the valid position coordinates, the moving speed of the virtual nodes of the gas mixing interface is calculated to construct a dynamic pipeline network topology model. Obtain the gas source calorific value component data from the natural gas revenue billing system, and bind the gas source calorific value component data with the virtual node of the gas mixing interface in the dynamic pipeline topology model; In the dynamic pipeline topology model, the expected timestamps of the virtual nodes at the gas mixing interface reaching each physical node of the downstream natural gas pipeline are calculated based on the moving speed of the virtual nodes at the gas mixing interface. Using the expected timestamps as a benchmark, the real-time flow data of the downstream physical nodes are spatiotemporally aligned with the bound gas source calorific value component data to construct the enterprise pipeline data map.

2. The enterprise data automation construction method for data fusion as described in claim 1, characterized in that, The specific process for constructing the static GIS spatial topology data of the natural gas pipeline network is as follows: The spatial coordinates, diameter, material, wall thickness, and location information of each physical node of the natural gas pipeline network are obtained. The physical nodes are used as vertices of the graph structure, and the pipelines connecting adjacent physical nodes are used as edges of the graph structure. The physical length and elevation difference of each edge are calculated based on the spatial coordinates of the pipeline. The physical length, elevation difference, diameter, material, and wall thickness are used as static attributes of the corresponding edges, and the location information is used as static attributes of the corresponding vertices, thus constructing static GIS spatial topology data.

3. The enterprise data automation construction method for data fusion as described in claim 1, characterized in that, The specific process for collecting real-time SCADA flow data from each physical node of the natural gas pipeline network is as follows: The SCADA system of the natural gas pipeline network acquires the sensor operating parameters at each physical node. The sensor operating parameters include real-time pressure, real-time temperature, real-time natural gas flow, and valve opening status. The sensor operating parameters are then timestamped and outlier removed. The processed sensor operating parameters with uniform timestamps are used as SCADA real-time stream data.

4. The enterprise data automation construction method for data fusion as described in claim 1, characterized in that, The specific process for calculating the real-time spatial location of the interfaces between different gas sources is as follows: Based on static GIS spatial topology data, identify the confluence physical nodes that are connected to two or more air intake pipelines, and obtain the real-time natural gas flow and valve opening status in the SCADA real-time flow data corresponding to the confluence physical nodes. Real-time monitoring of the status of each gas inlet pipeline at the same gas inlet physical node. If the real-time natural gas flow of two or more gas inlet pipelines at the same gas inlet physical node is greater than the effective gas inlet flow threshold at the same time, and the corresponding valve opening status is open; or the valve opening status of gas inlet pipelines connected to different gas sources changes from closed to open, and the real-time natural gas flow of the gas inlet pipelines connected to different gas sources is greater than the effective gas inlet flow threshold, then a multi-source gas mixing event is determined to have occurred. Obtain the initial moment of the multi-source gas mixing event and use the position coordinates of the corresponding confluence physical node as the initial spatial position of the gas mixing interface; Obtain the real-time pressure and real-time natural gas flow rate from the SCADA real-time flow data corresponding to the pipeline where the gas-mixing interface is located, as well as the pipe diameter parameters from the static GIS spatial topology data. The pipe diameter parameters include the outer diameter and wall thickness of the pipeline. Calculate the internal cross-sectional area of ​​the pipeline based on the pipe diameter parameters, and calculate the actual volumetric flow rate by combining the real-time pressure. Divide the actual volumetric flow rate by the internal cross-sectional area to obtain the real-time flow velocity of the gas in the pipeline. Using the initial moment as the starting point of time integration, the real-time flow velocity is integrated along the time axis to obtain the cumulative movement distance of the gas-mixed interface inside the pipeline. The cumulative movement distance is then mapped to static GIS spatial topology data to obtain the real-time spatial location of different gas source interfaces.

5. The enterprise data automation construction method for data fusion as described in claim 1, characterized in that, The specific process for updating the position coordinates of the virtual nodes at the gas-mixture interface is as follows: At the real-time spatial location, a new vertex is inserted and marked as a virtual node of the gas-mixing interface. The original pipeline edge where the virtual node of the gas-mixing interface is currently located is deleted. Using the virtual node of the gas-mixing interface as the intermediate connection point, a first dynamic sub-pipeline edge connecting the virtual node of the gas-mixing interface with the adjacent upstream physical node and a second dynamic sub-pipeline edge connecting the virtual node of the gas-mixing interface with the adjacent downstream physical node are generated to establish an initial virtual topology. The latest real-time spatial location is received and used to cover the current coordinates of the virtual node of the gas-mixing interface. The lengths of the first dynamic sub-pipeline edge and the second dynamic sub-pipeline edge are adjusted, and the position coordinates of the virtual node of the gas-mixing interface are updated.

6. The enterprise data automation construction method for data fusion as described in claim 1, characterized in that, The specific process for filtering valid location coordinates is as follows: Obtain the real-time pressure, real-time temperature, and real-time natural gas flow rate from the SCADA real-time flow data of the upstream and downstream physical nodes adjacent to the virtual node at the gas mixing interface at the corresponding time. Based on the real-time pressure and temperature of upstream and downstream physical nodes, and combined with the gas state equation, the actual gas volume in the current pipe segment is calculated. Based on the updated position coordinates of the virtual nodes at the gas-mixing interface, and combined with the pipe diameter parameters, calculate the theoretical gas pipe inventory corresponding to the updated position coordinates of the virtual nodes at the gas-mixing interface. Calculate the relative deviation between the actual gas tube inventory and the theoretical gas tube inventory, and determine whether the relative deviation is less than the confidence deviation threshold. If it is less than the threshold, mark the currently updated position coordinates as valid position coordinates; if it is greater than or equal to the threshold, retain the valid position coordinates from the previous time.

7. The enterprise data automation construction method for data fusion as described in claim 1, characterized in that, The specific process for calculating the moving speed of the virtual node at the gas-mixture interface is as follows: Construct a historical trajectory sequence containing timestamps and valid location coordinates. In the historical trajectory sequence, obtain the valid location coordinates at the current moment and the valid location coordinates at the previous adjacent moment. In the static GIS spatial topology data, calculate the actual pipeline length difference between the valid location coordinates at the current moment and the valid location coordinates at the previous adjacent moment along the physical direction of the pipeline. Divide the actual difference in pipe length by the time interval between the current moment and the previous adjacent moment to obtain the instantaneous speed; The instantaneous moving speed is obtained for multiple consecutive calculation cycles. The arithmetic mean of the instantaneous moving speed for multiple consecutive calculation cycles is calculated using a moving average filtering algorithm. The smoothed result is used as the moving speed of the virtual node at the gas-mixing interface.

8. The enterprise data automation construction method for data fusion as described in claim 1, characterized in that, The specific process of binding the gas source calorific value component data with the virtual node of the gas-mixing interface is as follows: Static GIS spatial topology data is used as the underlying physical framework, the selected effective location coordinates are used as spatial attributes, and the calculated movement speed is used as motion attributes. These are all encapsulated in the data structure of the virtual node of the gas-mixing interface. The encapsulated virtual node of the gas-mixing interface is injected into the underlying physical framework, and dynamic edge connection relationships between adjacent physical nodes are established. By integrating the above graph structure containing static physical nodes, dynamic virtual nodes, and dynamic edges that scale over time, a dynamic pipeline topology model that can reflect the movement state of the fluid boundary inside the pipeline network in real time is generated. Obtain the latest gas quality analysis report for each intake air source, and extract the gas source calorific value component data from the gas quality analysis report. The gas source calorific value component data includes higher heating value, lower heating value, Wobbe index, and the molar percentage content of methane, ethane, propane, and carbon dioxide. In the dynamic pipeline topology model, locate the virtual node of the gas-mixing interface and identify the gas source type corresponding to the upstream side of the virtual node. Add a dynamic attribute field to the data structure of the virtual node of the gas-mixing interface and write the gas source calorific value component data that matches the gas source type into the dynamic attribute field. Establish a mapping table between the calorific value component data of the gas source and the unique identifier of the virtual node of the gas-mixing interface, and complete the data binding.

9. The enterprise data automation construction method for data fusion as described in claim 1, characterized in that, The specific process of constructing an enterprise pipeline network data map by spatiotemporally aligning the real-time flow data of downstream physical nodes with the bound gas source calorific value component data based on the expected timestamp is as follows: In the dynamic pipeline topology model, a path tracing algorithm is used to determine the natural gas transmission path from the current location of the virtual node at the gas mixing interface to the target downstream physical node; Along the natural gas transmission path, the physical lengths of each pipeline side are accumulated to obtain the remaining transmission distance from the virtual node at the gas-mixing interface to the target downstream physical node. Divide the remaining transmission distance by the moving speed of the virtual node at the gas-mixing interface to calculate the remaining time for natural gas transmission. Add the current system time to the remaining time to get the expected timestamp of the virtual node at the gas-mixing interface reaching the downstream physical node; If the pipeline has branches resulting in multiple downstream physical nodes, repeat the above calculation process to obtain the expected timestamps for each downstream physical node. When the system time reaches the expected timestamp, the time matching tolerance window is expanded before and after the expected timestamp, and the real-time stream data of the corresponding downstream physical node is obtained within the time matching tolerance window; Obtain the gas source calorific value component data bound to the virtual nodes of the gas mixing interface in the dynamic pipeline network topology model; The captured real-time flow data and gas source calorific value component data are spliced ​​together along the expected timestamp dimension to generate a multi-dimensional fused data record that includes spatial location, time node, physical flow rate and chemical calorific value. By using downstream physical nodes as entity nodes and multi-dimensional fused data records as entity attributes, or by using multi-dimensional fused data records as independent data nodes and establishing time-dimensional association edges with downstream physical nodes, the graph mapping of heterogeneous data is completed, and the enterprise pipeline network data graph is constructed.

10. An automated enterprise data construction system for data fusion, applied to the method described in any one of claims 1-9, characterized in that: include: Static topology and data acquisition module: Constructs static GIS spatial topology data of the natural gas pipeline network and acquires real-time SCADA flow data of each physical node of the natural gas pipeline network; Gas mixing determination module: Analyzes real-time flow data to determine whether a multi-source gas mixing event has occurred. If it has occurred, it calculates the real-time spatial location of the interface between different gas sources based on the pressure and flow parameters in the SCADA real-time flow data and the transient gas flow equation. Dynamic topology construction module: In static GIS spatial topology data, real-time spatial locations are marked as virtual nodes of the gas mixing interface, and the position coordinates of the virtual nodes of the gas mixing interface are updated over time. Based on the pressure, temperature, and natural gas flow collected from each physical node of the upstream and downstream natural gas pipeline network, the confidence analysis of the position coordinates of the updated virtual nodes of the gas mixing interface is performed to filter valid position coordinates. Based on the valid position coordinates, the moving speed of the virtual nodes of the gas mixing interface is calculated to construct a dynamic pipeline network topology model. Component binding module: Obtains gas source calorific value component data from the natural gas revenue billing system, and binds the gas source calorific value component data to the virtual node of the gas mixing interface in the dynamic pipeline topology model; Map construction module: In the dynamic pipeline topology model, the expected timestamps of the virtual nodes at the gas mixing interface reaching each physical node of the downstream natural gas pipeline are calculated based on the moving speed of the virtual nodes at the gas mixing interface. Using the expected timestamps as a benchmark, the real-time flow data of the downstream physical nodes are spatiotemporally aligned with the bound gas source calorific value component data to construct the enterprise pipeline network data map.

Images