Long-distance water delivery pipeline flow monitoring method with fused terrain dynamic compensation

Abstract

The present application belongs to the technical field of pipeline flow monitoring, and specifically relates to a long-distance water conveying pipeline flow monitoring method fusing terrain dynamic compensation, comprising the following steps: selecting a long-distance water conveying pipeline reference point, synchronously collecting reference flow and static pressure; based on a multi-source data fusion technology, constructing a pipeline terrain elevation parameter model, generating a three-dimensional terrain matrix, and calculating elevation difference and slope 、 radius of curvature, constructing a terrain correction matrix; based on Darcy formula correction, slope section kinetic energy suppression, and downhill section pressure compensation algorithm, correcting water head loss, suppressing potential energy conversion error, and eliminating water hammer effect interference; adopting edge computing to correct flow in the pipeline, performing global flow balance verification; and based on an integrated visual platform, assisting in positioning a burst pipe or a leakage point. The method effectively improves the measurement accuracy of pipeline flow under complex terrain, and provides a high-precision, full-terrain self-adaptive intelligent monitoring and early warning platform for long-distance water conveying pipelines.

Description

TECHNICAL FIELD

[0001] The present application mainly relates to the technical field of pipeline flow monitoring, in particular to a long-distance water conveying pipeline flow monitoring method fusing terrain dynamic compensation, which is especially suitable for eliminating the flow measurement distortion problem caused by pipeline slope, elevation difference and terrain undulation, and realizing the whole-line flow consistency monitoring of water conveying pipeline network under complex terrain conditions. BACKGROUND

[0002] As the core facility of water conservancy projects, the flow monitoring accuracy of long-distance water conveying pipeline directly affects the water resource scheduling efficiency and operation safety. However, the existing flow measurement technology (traditional ultrasonic wave, electromagnetic flowmeter, etc.) does not consider the influence of pipeline longitudinal slope change on fluid energy state during installation: for example, in the downhill pipeline section, the continuous conversion of fluid potential energy into kinetic energy leads to the virtual high flow rate, and the error is even higher than 15%.

[0003] The existing correction algorithm mostly adopts a static pressure compensation model, which neither establishes a dynamic correlation equation of terrain elevation-fluid energy state, nor quantitatively analyzes the local turbulent energy loss caused by pipeline undulation. Especially in the complex terrain area, the vortex generated by the separation of turbulent boundary layer will exacerbate the mechanical energy loss, and the constant friction coefficient method adopted in the existing standard cannot reflect such transient hydraulic characteristics, resulting in the deviation of pressure compensation value from the actual working condition, and further expanding the flow data deviation. Therefore, it is urgent to build a dynamic correction system fusing terrain elevation compensation and turbulent energy loss, to solve the systematic measurement error problem of long-distance water conveying pipeline network caused by terrain undulation. SUMMARY

[0004] In order to solve the deficiencies of the prior art, the present application combines the prior art and starts from practical application, and provides a long-distance water conveying pipeline flow monitoring method fusing terrain dynamic compensation, which effectively improves the measurement accuracy of pipeline flow under complex terrain, and provides a high-precision, full-terrain self-adaptive intelligent monitoring and early warning platform for long-distance water conveying pipeline.

[0005] The technical scheme of the present application is as follows:

[0006] The long-distance water conveying pipeline flow monitoring method fusing terrain dynamic compensation comprises the following steps:

[0007] S1, selecting a long-distance water conveying pipeline reference point, and synchronously collecting reference flow by a pressure sensor and a multi-channel ultrasonic flowmeter Q 0 and static pressure P 0;

[0008] S2, based on multi-source data fusion technology, constructing a pipeline terrain elevation parameter model by GIS spatial analysis and LiDAR point cloud processing, generating a three-dimensional terrain matrix, and calculating the elevation difference Δ H、 slope θ,Curvature radius R, build terrain correction matrix K terrain ;

[0009] S3, based on Darcy formula correction, downhill segment kinetic energy inhibition and downhill segment pressure compensation algorithm correction head loss, inhibition of potential energy conversion error and elimination of water hammer effect interference;

[0010] S4, using edge computing to correct the flow in the pipeline, the corrected flow is integrated into the platform for global flow balance verification;

[0011] S5, based on the integrated visualization platform, real-time display of the corrected flow, terrain parameters and pressure gradient of each node, generate pressure gradient distribution map and pressure gradient heat map to assist in locating the burst pipe or leak point.

[0012] Further, in step S1, the reference point is preferentially set in a flat area with a terrain slope ≤1% and a straight pipe segment length ≥30D, where D is the pipe diameter;

[0013] The pressure sensor is vertically installed on the top of the pipe and orthogonal to the fluid flow direction, and the temperature fluctuation of the medium is eliminated through the temperature compensation module to eliminate the influence of the temperature fluctuation of the medium on the pressure measurement.

[0014] Further, the specific process of step S2 includes:

[0015] S21, data preprocessing: generate a digital elevation model DEM using LiDAR point cloud data, separate ground points and non-ground points through a random forest filtering algorithm, eliminate interference sources, and obtain a high-fidelity terrain surface;

[0016] S22, spatial parameter calculation: calculate the elevation difference Δ H、 slope θ, Curvature radius R;

[0017] S23, correction coefficient modeling: build a terrain correction matrix K terrain , introduce a dynamic weight factor, supervise learning through multiple sets of measured data, and determine the final parameter combination using cross-validation method.

[0018] Further, in step S22, the elevation difference Δ H The calculation method is: using Delaunay triangulation algorithm to establish the spatial topological relationship between the pipeline axis and the terrain surface, calculating the absolute elevation difference of each node relative to the reference point based on DEM data;

[0019] slope θ The calculation method is: using the slope tool of the ArcGIS hydrological analysis module, combining with the Z factor correction to eliminate coordinate projection error, calculating the slope change rate of the pipeline axis projection surface, and the formula (1) is as follows:

[0020] θ =arctan(Δ H / Δ L )×180 / π (1)

[0021] In the formula, Δ L The length of the adjacent node.

[0022] The radius of curvature R is calculated as follows: based on the quadratic surface fitting model, the parameters are iteratively optimized using the Levenberg-Marquardt algorithm, and the geometric characteristics of the bend are quantified using the differential geometry formula R=1 / K. Local fine sampling is performed on the sharp bend with a radius ≤5D, where K is the curvature and D is the pipe diameter.

[0023] Furthermore, in step S23, after introducing a dynamic weighting factor, the terrain correction matrix... K terrain Formula (2) is as follows:

[0024] K terrain = α ·Δ H / H 0+ β ·sinθ+ γ ·D / R (2)

[0025] In the formula, α, β, γ The weight coefficients are optimized using a BP neural network. H 0 represents the elevation of the reference point, and D represents the pipe diameter.

[0026] Furthermore, in step S3, Darcy's formula is modified as follows:

[0027] Embedding terrain correction matrices into the traditional Darcy-Wiesbach equations K terrain The head loss term is corrected, and the modified formula (3) is as follows:

[0028] h fm =( fL / D+ K terrain )· v 2 / (2g)(3)

[0029] In the formula: h fm This is the corrected head loss; f D is the Darcy friction coefficient, reflecting the effect of pipe wall roughness on water flow resistance; D is the pipe diameter. L This refers to the length of the pipe. vV is the flow velocity in the pipe, g is the acceleration of gravity;

[0030] The kinetic energy of the uphill section is inhibited as follows:

[0031] In view of the abnormal conversion of the potential energy of the fluid to kinetic energy in the uphill section, a flow correction equation based on energy conservation is proposed, and formula (4) is as follows:

[0032] Q i = Q raw [1- η tan( θ )·Δ H / L ](4)

[0033] In the formula: Q i Q is the corrected flow rate; Q raw Q0 is the measured flow rate without terrain correction; η ζ is a terrain attenuation factor, reflecting the attenuation effect of terrain on flow rate, and is positively correlated with slope θ and elevation difference Δ H ; Δ H / L is the elevation change rate per unit pipe length, reflecting the potential energy gradient strength;

[0034] The pressure fluctuation compensation of the downhill section is as follows:

[0035] In view of the pressure drop and water hammer effect caused by gravity acceleration in the downhill section, a flow rate compensation model driven by pressure gradient is established, and formula (5) is as follows:

[0036] v real = v +ζ(K / ρ )· P / t(5)

[0037] In the formula: ζ is a pressure gradient response factor, with a value range of 0.05-0.1; K K is the water body elastic modulus; P / t is the change amount of pressure per unit time, reflecting the transient fluctuation strength; v real Q is the real flow rate after multi-factor dynamic compensation, ρ and ρ is the density of water.

[0038] Further, step S4 is specifically as follows:

[0039] The edge computing is utilized to perform formula (3) and formula (4) on the measured flow of the flowmeter on the pipeline, the corrected water head loss is calculated through formula (3), and the corrected flow is output through formula (4) Q i , and the influence of delay on transient pressure fluctuation is reduced

[0040] The flow data integrated into the cloud platform are subjected to global flow balance verification, an error threshold is set, when the flow deviation of a node exceeds the threshold, the platform calls terrain parameters (H ΔH, θ, R ) and datum point data of the region Q 0 and P 0, and recalculates the terrain correction matrix (H K terrain Corresponding weight coefficients.

[0041] Further, step S5 is specifically as follows:

[0042] The BIM+GIS+IoT fusion architecture is integrated, the corrected flow of each node, terrain parameters and pressure gradient are realized in real time, and the pressure gradient distribution map and the pressure gradient heat map are generated through the Kriging interpolation algorithm with static pressure (H P 0 as the datum.

[0043] Advantages of the present application:

[0044] The present application establishes a terrain elevation-fluid energy dynamic compensation system, for the first time, couples the parameters such as pipeline elevation difference (H H、 Slope θ, Radius of curvature R and fluid kinetic energy-potential energy conversion effect modeling, and develops a multi-modal correction algorithm to effectively improve the measurement accuracy of pipeline flow under complex terrain. Through the visualization platform integrated with BIM+GIS, the pressure gradient heat map can be generated in real time, which can assist in locating the burst pipe or leakage point, and provide a high-precision, full-terrain adaptive intelligent monitoring and early warning platform for long-distance water conveying pipelines. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 The detailed flowchart of the present application. DETAILED DESCRIPTION

[0046] The present application will be further described in conjunction with the drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present application and not to limit the scope of the present application. In addition, it should be understood that after reading the content taught by the present application, those skilled in the art can make various modifications or modifications to the present application, and these equivalent forms also fall within the scope defined by the present application.

[0047] This embodiment provides a method for monitoring flow in long-distance water transmission pipelines by integrating dynamic terrain compensation. To eliminate flow measurement distortion caused by pipeline slope, elevation difference, and terrain undulations, this embodiment focuses on selecting benchmark points, modeling terrain parameters, developing a dynamic terrain-flow compensation algorithm, constructing a multi-node collaborative verification platform, and establishing a visualization platform. Benchmark point selection is fundamental, serving as a reference for subsequent flow correction. Terrain parameter modeling generates parameters such as elevation difference, slope, and radius of curvature using GIS and LiDAR data, quantifying the impact of terrain on flow and pressure. The dynamic terrain-flow compensation algorithm is the core, addressing flow deviations by modifying the Darcy formula to handle flow along and downhill sections. The multi-node collaborative verification platform ensures consistent flow data across the entire pipeline network, while the visualization platform is used for real-time monitoring and fault location. The process is as follows: Figure 1 As shown, the specific monitoring methods are as follows.

[0048] 1. Selection of reference points

[0049] Selecting appropriate reference points in long-distance water pipelines is fundamental to flow correction. The selection of reference points requires comprehensive consideration of terrain features, fluid dynamics, and the compatibility of the measuring equipment.

[0050] The benchmark should preferably be set in a gently sloping area with a terrain slope of ≤1% and a straight pipe section length of ≥30D (D is the pipe diameter). Such areas can effectively avoid local turbulence interference sources such as bends and valves, ensuring that the fluid fully develops into a stable laminar flow state. In practice, the pipeline is arranged according to the terrain undulations, with a terrain slope of ≤1%, or the pipeline slope can be ≤1%.

[0051] In terms of measurement equipment, a dual-redundant 0.1%FS accuracy pressure sensor and a multi-channel ultrasonic flow meter are used to synchronously acquire the reference flow rate. Q 0 and static pressure P 0, wherein the pressure sensor should be installed vertically on the top of the pipe and orthogonal to the fluid flow direction, and the influence of medium temperature fluctuations on pressure measurement should be eliminated through a temperature compensation module.

[0052] 2. Terrain elevation parameter modeling

[0053] Based on multi-source data fusion technology, terrain elevation parameter modeling constructs a 3D terrain elevation model of the pipeline through GIS spatial analysis and LiDAR point cloud processing. The specific process includes:

[0054] (1) Data preprocessing: A 1cm resolution digital elevation model (DEM) is generated using LiDAR point cloud data. Ground points and non-ground points are separated by random forest filtering algorithm to eliminate interference sources such as vegetation and buildings, and a high-fidelity terrain surface is obtained.

[0055] (2) Spatial parameter calculation:

[0056] Elevation difference Δ H : The spatial topological relationship between pipeline axis and terrain surface is established by Delaunay triangulation algorithm, and the absolute elevation difference of each node relative to the reference point is calculated based on DEM data, with an accuracy of ±1 cm;

[0057] Slope θ : The slope change rate of pipeline axis projection surface is calculated by the slope tool of ArcGIS hydrological analysis module, combined with Z factor correction to eliminate coordinate projection error, and the formula (1) is as follows:

[0058] θ =arctan(Δ H / Δ L )×180 / π (1)

[0059] Curvature radius R: Based on the quadratic surface fitting model, the Levenberg-Marquardt (LM) algorithm is used to iteratively optimize the parameters, and the differential geometry formula R=1 / K (K is the curvature) is used to quantify the geometric characteristics of the elbow section. The local sampling is encrypted for the sharp bend section with a radius ≤5D (pipe diameter).

[0060] (3) Correction coefficient modeling:

[0061] Constructing terrain correction matrix K terrain When the dynamic weight factor is introduced, the terrain correction matrix K terrain is expressed as formula (2). Through supervised learning of multiple sets of measured data (including water hammer pressure wave speed and turbulent energy dissipation rate), the cross-validation method is used to determine the final parameter combination. For complex terrain (such as slope sudden change area), the RANS equation and RNG k-ε turbulence model are integrated to establish a local energy loss compensation sub-model, which reduces the correction coefficient error from ±15% of traditional method to ±3.2%.

[0062] K terrain = α ·Δ H / H 0+ β ·sinθ+ γ ·D / R (2)

[0063] In the formula: α, β, γ is the weight coefficient optimized by BP neural network, Δ H is the elevation difference of each node, which is calculated and generated by GIS and LiDAR data, reflecting the influence of terrain undulation on fluid potential energy (m); θ is the slope, indicating the pipeline slope angle (°); RR is the radius of curvature, which is a parameter to quantify the degree of pipeline bending; H 0 is the benchmark point elevation.

[0064] Terrain correction matrix K terrain Through supervised learning of multiple sets of measured data, the final parameter combination is determined by cross-validation method, and the specific method is as follows.

[0065] Data acquisition: Select multiple sets of measured data in different terrain scenes (such as plains, mountains, and hills), and record the flow Q, pressure P, terrain parameters (elevation difference, slope angle, and curvature radius), and dynamic parameters such as water hammer pressure wave speed synchronously. ΔH, θ, R ) and water hammer pressure wave speed synchronously.

[0066] BP neural network optimization: Taking the measured flow deviation as the objective function, the initial value (the initial range is set to 0.1~0.9) is adjusted through BP neural network iteration α , β , γ The mean square error (MSE) of the model predicted flow and the measured flow is minimized.

[0067] Cross-validation: Using multiple-fold cross-validation method, taking 5-fold cross-validation method as an example, the data set is divided into 5 parts, and each time 4 parts are trained and 1 part is validated, and the average error is repeated 5 times to determine the optimal parameter combination.

[0068] The weight coefficient α is the elevation difference ΔH weight, used to guide the influence of elevation difference on potential energy conversion, the larger α, the greater the flow correction amount of elevation difference.

[0069] The weight coefficient β is the pipe slope angle θ weight, used to quantify the inhibitory effect of slope on abnormal conversion of kinetic energy, β is positively correlated with the flow correction strength of the uphill section.

[0070] The weight coefficient γ is the radius of curvature R weight, used to reflect the influence of bend curvature on turbulent energy consumption, the larger γ, the greater the flow correction amount of sharp bend section (≤5D). R

[0071] 3. Terrain-flow dynamic compensation algorithm

[0072] The algorithm consists of Darcy formula correction, kinetic energy inhibition in uphill section, and pressure compensation in downhill section, which are respectively used to correct head loss, inhibit potential energy conversion error, and eliminate water hammer effect interference, to solve the flow measurement distortion problem caused by terrain fluctuation changes in long-distance water conveying pipeline.

[0073] (1) Darcy formula correction and turbulent energy dynamic compensation:

[0074] Embedding terrain correction matrix in traditional Darcy-Weisbach equation​K terrain , the modified water head loss term, the modified formula (3) is as follows:

[0075] h fm = ( fL / D+ K terrain )· v 2 / (2g)(3)

[0076] In the formula: h fm is the corrected water head loss; f is the Darcy friction coefficient, reflecting the influence of pipe wall roughness on water flow resistance; D is the inner diameter of the pipeline (m); L is the length of the pipeline (m); v is the water flow velocity in the pipeline (m / s); K terrain is used to correct the local turbulent loss and energy gradient deviation caused by the terrain undulation;

[0077] (2) Kinetic energy suppression in the downhill section:

[0078] For the abnormal conversion of fluid potential energy to kinetic energy in the downhill section ( θ > 3°), the flow correction equation based on energy conservation is proposed, and the formula (4) is as follows:

[0079] Q i = Q raw [1- η tan( θ )·Δ H / L ](4)

[0080] In the formula: Q i is the corrected flow (m 3 / s); Q raw is the measured flow value without terrain correction (m 3 / s); η is the terrain attenuation factor, reflecting the attenuation effect of terrain on flow velocity, which is positively related to the slope θ and the elevation difference Δ H ; Δ H / L is the elevation change rate per unit length of pipe, reflecting the potential energy gradient intensity.

[0081] The terrain attenuation factor η ranges from 0.1 to 0.5, which is determined by fitting the measured data. The specific calculation formula is: η =k1· θ+k2· ΔH wherein k1, k2 are fitting coefficients, obtained by regression analysis of historical data.

[0082] (3) Pressure fluctuation compensation in downhill section:

[0083] For the pressure drop and water hammer effect caused by gravity acceleration in the downhill section (0°< downhill < -2°), a flow rate compensation model driven by pressure gradient is established, and formula (5) is as follows: θ

[0084] v real = v +ζ( K / ρ )· P / t(5)

[0085] In the formula: ζ is a pressure gradient response factor (0.05~0.1), which represents the sensitivity of the pipeline system to pressure fluctuations, such as 0.07 for steel pipes and 0.09 for plastic pipes, which is determined by water hammer effect experiment; K is the elastic modulus of water; P / t is the change of pressure per unit time, reflecting the intensity of transient fluctuation (Pa / s); v real is the true flow rate after multi-factor dynamic compensation (m / s), ρ is the density of water, v is the flow rate of water in the pipeline (m / s), which can be directly collected by the ultrasonic flow meter.

[0086] 4. Multi-node collaborative verification platform

[0087] Through edge computing and cloud collaborative mechanism, the consistency and real-time performance of pipeline flow data are ensured. Among them, edge computing is used to correct the flow in the pipeline, and the corrected flow is integrated into the platform for global flow balance verification, as follows:

[0088] Local flow correction based on edge computing: use edge computing to execute formula (3) and formula (4) on the measured flow of the flow meter on the pipeline, calculate the corrected head loss through formula (3), and output the corrected flow through formula (4) Q i , reducing the influence of delay on transient pressure fluctuation. Formula (3) is used to correct the head loss caused by terrain undulation, ensuring the accuracy of the Darcy formula; formula (4) is based on the principle of energy conservation, inhibiting the abnormal conversion of fluid potential energy to kinetic energy in the downhill section, and further correcting the flow value. Formula (3) and formula (4) work together to achieve comprehensive correction of flow from two angles of head loss and energy conversion.​

[0089] Global flow balance check: Perform global flow balance check on the flow data integrated into the cloud platform, set the error threshold to ±2%, if the deviation of adjacent nodes >3% trigger curvature parameter recalibration, that is, dynamic correction formula (2). Specifically, when the flow deviation of a node exceeds the threshold, the platform calls the terrain parameters of the region ΔH, θ, R ) and the reference point data Q 0 and P 0, recalculate α, β, γ to minimize the flow deviation of the region, that is, recalculate the terrain correction matrix K terrain corresponding weight coefficient α, β, γ.

[0090] 5, visualization platform

[0091] Integrate BIM+GIS+IoT fusion architecture to realize real-time display of flow after correction of each node, terrain parameters and pressure gradient. Take P 0 as the reference, generate pressure gradient distribution map through Kriging interpolation algorithm, generate pressure gradient heat map to assist in positioning burst pipe or leakage point.

Claims

1. A method for monitoring flow in long-distance water transmission pipelines integrating dynamic terrain compensation, characterized in that, Includes the following steps: S1. Select a reference point for a long-distance water transmission pipeline, and simultaneously collect the reference flow rate using a pressure sensor and a multi-channel ultrasonic flow meter. Q 0 and static pressure P 0; S2. Based on multi-source data fusion technology, a pipeline terrain elevation parameter model is constructed through GIS spatial analysis and LiDAR point cloud processing, generating a three-dimensional terrain matrix and calculating the elevation difference Δ. H、 slope θ、 Constructing the terrain correction matrix based on the radius of curvature R. K terrain ; S3. Based on Darcy's formula correction, downhill section kinetic energy suppression and downhill section pressure compensation algorithm, head loss is corrected, potential energy conversion error is suppressed and water hammer effect interference is eliminated; S4. Edge computing is used to correct the flow in the pipeline. The corrected flow is then integrated into the platform for global flow balance verification. S5. Based on an integrated visualization platform, the corrected flow rate, terrain parameters and pressure gradient of each node are displayed in real time, and pressure gradient distribution map and pressure gradient heat map are generated to help locate the pipe burst or leak point. In step S3, Darcy's formula is modified as follows: Embedding terrain correction matrices into the traditional Darcy-Wiesbach equations K terrain The head loss term is corrected, and the modified formula (3) is as follows: h fm =( fL / D+ K terrain )· v 2 / (2g)(3) In the formula: h fm This is the corrected head loss; f D is the Darcy friction coefficient, reflecting the effect of pipe wall roughness on water flow resistance; D is the pipe diameter. L This refers to the length of the pipe. v Let g be the velocity of the water flow in the pipe, and g be the acceleration due to gravity. The kinetic energy suppression along the slope is as follows: To address the anomalous conversion of fluid potential energy to kinetic energy in the downslope section, a flow correction equation based on energy conservation is proposed, as shown in equation (4): Q i = Q raw [1- η tan( θ )·Δ H / L ](4) In the formula: Q i The corrected flow rate; Q raw The measured flow rate is the value before terrain correction. η This is the topographic attenuation factor, reflecting the attenuation effect of topography on flow velocity, and is related to slope. θ and elevation difference Δ H Positive correlation; Δ H / L The elevation change rate per unit pipe length reflects the potential energy gradient intensity; The pressure fluctuation compensation for downhill sections is as follows: To address the sudden pressure drop and water hammer effect caused by gravity acceleration on the downhill section, a velocity compensation model driven by pressure gradient is established, and formula (5) is as follows: v real = v +ζ(K / ρ ); P / t(5) In the formula: ζ is the pressure gradient response factor, with a value range of 0.05~0.1; K The elastic modulus of water; P / t represents the change in pressure per unit time, reflecting the intensity of transient fluctuations; v real The actual flow velocity after dynamic compensation for multiple factors. ρ This is the density of water.

2. The method for monitoring the flow rate of long-distance water transmission pipelines with integrated terrain dynamic compensation according to claim 1, characterized in that, In step S1, the benchmark point is preferably set in a flat area with a terrain slope of ≤1% and a straight pipe section length of ≥30D, where D is the pipe diameter; The pressure sensor is installed vertically at the top of the pipe and orthogonal to the fluid flow direction, and the influence of medium temperature fluctuations on pressure measurement is eliminated through a temperature compensation module.

3. The method for monitoring the flow rate of long-distance water transmission pipelines with integrated terrain dynamic compensation according to claim 1, characterized in that, Step S2 includes the following specific steps: S21. Data preprocessing: Use LiDAR point cloud data to generate a digital elevation model (DEM), separate ground points from non-ground points using a random forest filtering algorithm, eliminate interference sources, and obtain a high-fidelity terrain surface. S22. Spatial parameter calculation: Calculate the elevation difference Δ H、 slope θ、 radius of curvature R; S23. Correction Coefficient Modeling: Constructing the Terrain Correction Matrix K terrain A dynamic weighting factor is introduced, and supervised learning is performed using multiple sets of measured data. The final parameter combination is determined by cross-validation.

4. The method for monitoring the flow rate of long-distance water transmission pipelines with integrated terrain dynamic compensation according to claim 3, characterized in that, In step S22, the elevation difference Δ H The calculation method is as follows: the Delaunay triangulation algorithm is used to establish the spatial topological relationship between the pipeline axis and the terrain surface, and the absolute elevation difference of each node relative to the reference point is calculated based on the DEM data; slope θ The calculation method is as follows: using the slope tool in the ArcGIS hydrological analysis module, combined with Z-factor correction to eliminate coordinate projection errors, the slope change rate of the pipeline axis projection surface is calculated, and the formula (1) is as follows: θ =arctan(Δ H / Δ L )×180 / π (1) In the formula, Δ L The length of the adjacent node; The radius of curvature R is calculated as follows: based on the quadratic surface fitting model, the parameters are iteratively optimized using the Levenberg-Marquardt algorithm, and the geometric characteristics of the bend are quantified using the differential geometry formula R=1 / K. Local fine sampling is performed on the sharp bend with a radius ≤5D, where K is the curvature and D is the pipe diameter.

5. The method for monitoring the flow rate of long-distance water conveyance pipelines with integrated terrain dynamic compensation according to claim 3, characterized in that, In step S23, after introducing the dynamic weighting factor, the terrain correction matrix... K terrain Formula (2) is as follows: K terrain = α ·D H / H 0+ β ·sinθ+ γ ·D / R(2) In the formula, α, β, γ The weight coefficients are optimized using a BP neural network. H 0 represents the elevation of the reference point, and D represents the pipe diameter.

6. The method for monitoring the flow of long-distance water transmission pipelines with integrated terrain dynamic compensation as described in claim 1, characterized in that, Step S4 is as follows: Edge computing is used to apply formulas (3) and (4) to the measured flow rate of the flow meter on the pipeline. The corrected head loss is calculated using formula (3), and the corrected flow rate is output using formula (4). Q i This reduces the impact of delay on transient pressure fluctuations; The traffic data integrated into the cloud platform is subjected to global traffic balancing verification, and an error threshold is set. When the traffic deviation of a node exceeds the threshold, the platform calls the terrain parameters of that area. ΔH, θ, R and benchmark data Q 0 and P 0, recalculate the terrain correction matrix K terrain The corresponding weighting coefficients.

7. The method for monitoring the flow rate of long-distance water conveyance pipelines with integrated terrain dynamic compensation according to claim 1, characterized in that, Step S5 is as follows: The integrated BIM+GIS+IoT architecture enables real-time display of corrected flow rates, terrain parameters, and pressure gradients at each node, using static pressure... P Using 0 as the baseline, pressure gradient distribution map and pressure gradient heat map are generated using the Kriging interpolation algorithm.

Images