Automatic rapid modeling method for underground water supply and drainage pipe network assembly

Abstract

The invention relates to the technical field of data modeling, in particular to an automatic rapid modeling method for an underground water supply and drainage pipe network assembly. According to the method, a liquid level monitoring time sequence is analyzed, the estimated sectional area of each height is estimated by using a liquid level climbing rate, and sectional area change rate and section fluctuation degree characteristics are extracted. And then, matching the sectional area and the change rate with standard component characteristics, constructing a difference degree matrix, and calculating an optimal component type sequence through dynamic planning in combination with a state transition matrix containing component topology constraints. And finally, on the basis of the difference between the actually measured sectional area and the optimal component and the fluctuation characteristics of the cross section, distinguishing entity attributes, correcting the open area and generating a water level-volume relation table. According to the method, the absolute physical size can be automatically calibrated and the well bottom disease type can be accurately identified in the absence of inflow data, so that the accuracy of the pipe network hydraulic model is improved.

Description

Technical Field

[0001] This invention relates to the field of data modeling technology, and specifically to an automated and rapid modeling method for underground water supply and drainage network components. Background Technology

[0002] Hydraulic models of urban drainage networks are the core data foundation for assessing flood control capacity, formulating flood prevention strategies, and optimizing network scheduling. In the construction of hydraulic models, manholes, as key nodes in the network system, directly determine the system's instantaneous storage capacity and overflow risk through their internal geometric volume. Precast concrete manholes are typically assembled from standard components with various cross-sections, such as bases, manhole rings, and tapered ends. During long-term operation, the effective storage capacity at the bottom of the manhole is easily reduced due to silt deposition, or the flow cross-section may be altered due to engineering measures such as lining repairs. Traditional modeling methods often simplify manholes to standard cylinders based on design drawings, ignoring the complex geometric variations and siltation conditions that actually exist, severely affecting the accuracy of hydraulic simulations.

[0003] Currently, using sensor data to invert downhole conditions has become a major trend in low-cost surveys. However, the equipment cost and maintenance difficulty of simultaneously deploying high-precision inflow flow meters at tens of thousands of manhole nodes throughout the city are prohibitive for municipal management, and most monitoring points only have basic liquid level monitoring capabilities. Under the condition of a "single-dimensional blind source" with only liquid level data, when inverting the cross-sectional area using the principle of fluid continuity, the lack of flow boundary conditions will lead to unknown proportional deviations in the calculation results, making it impossible to obtain the true physical dimensions. In addition, existing inversion methods usually only focus on the shrinkage of the geometric profile and cannot distinguish between the smooth shrinkage caused by lining repair and the rough shrinkage caused by siltation. Although both of these situations manifest as a reduction in flow area, their impact on water flow resistance and maintenance requirements are completely different. Summary of the Invention

[0004] To address the problems of high inspection costs for localized drainage network manholes in existing technologies, and the inability of data-driven inversion methods to effectively determine entity attributes at various height levels, resulting in poor final modeling quality, this invention aims to provide an automated and rapid modeling method for underground water supply and drainage network components. The specific technical solution adopted is as follows: This invention proposes an automated and rapid modeling method for underground water supply and drainage network components, the method comprising: Obtain the liquid level monitoring time series in the target inspection well, estimate the cross-sectional area of ​​the target inspection well at each height based on the liquid level rise rate at each time point, and obtain the height-cross-sectional area series; obtain the cross-sectional area change rate and cross-sectional fluctuation degree at each height based on the height-cross-sectional area series; Based on the estimated cross-section and cross-sectional area change rate corresponding to the height, the feature of the standard component at the corresponding height is matched to obtain the matching difference degree between each height and each standard component; Construct a state transition matrix, where the rows and columns represent the number of standard components, and the elements are the allowable probabilities of transitioning from a lower-level component to an upper-level component. Based on the allowable probabilities and the matching difference, calculate from the lowest height to the highest height to obtain the optimal component type corresponding to each height. Based on the difference in cross-sectional area between each height and the optimal component type, as well as the difference in the degree of cross-sectional fluctuation, the entity attribute to which each height belongs is determined; the effective flow area of ​​each height is determined based on the entity attribute, and a water level-volume relationship table is generated based on the effective flow area.

[0005] Furthermore, the method for calculating the estimated cross-sectional area includes: The liquid level rise rate of each liquid level is obtained based on the difference between adjacent liquid levels; if the liquid level rise rate is less than a preset rate threshold, the liquid level rise rate is set to the preset rate threshold; the ratio of the preset reference inflow rate to the remaining liquid level rise rate is used as the estimated cross-sectional area at the height of the inspection well where the corresponding liquid level is located.

[0006] Furthermore, the method for obtaining the degree of cross-sectional fluctuation includes: In the height-cross-sectional area sequence, for each height, a preset height range is constructed with the height as the center, and the variance of the estimated cross-sectional area is calculated within the preset height range to obtain the degree of cross-sectional fluctuation.

[0007] Furthermore, the method for obtaining the matching difference includes: Use the height of the standard component as the local height; For each local height, a first difference is obtained between the estimated cross-sectional area of ​​the inspection well at that height and the standard cross-sectional area of ​​the standard component at that local height; a second difference is obtained between the rate of change of the cross-sectional area at that height and the rate of change of the standard cross-sectional area of ​​the standard component at that local height; the first difference and the second difference are weighted and summed using a preset threshold to obtain the initial difference degree between each local height and the current height of the inspection well; The minimum initial difference among all local heights is taken as the matching difference between the current inspection well height and the standard component.

[0008] Furthermore, the allowed probability is a preset binary probability, including 0 and positive infinity. 0 indicates that the lower-level height component corresponding to the current element is allowed to transition to the upper-level height component corresponding to the current element; positive infinity indicates that the lower-level height component corresponding to the current element is not allowed to transition to the upper-level height component corresponding to the current element.

[0009] Furthermore, the method for calculating the optimal component type includes: The cumulative cost between each height and each standard component is obtained using a cumulative cost calculation formula, which includes: ;in This represents the cumulative cost between the k-th height and the j-th standard component; This represents the cumulative cost between component i and the previous height k-1. Let i be the allowed probability in the state transition matrix when the lower-level height component is i and the upper-level height component is j. This means iterating through all components i and taking the minimum value; For the kth height The degree of matching difference with the j-th standard component; Obtain the minimum cumulative cost corresponding to the highest height, and backtrack using the minimum cumulative cost to obtain the optimal component type corresponding to all heights.

[0010] Furthermore, the method for determining the entity attributes includes: The cross-sectional area difference is the difference between the estimated cross-sectional area at the current height and the standard cross-sectional area of ​​the optimal component type; If the difference in cross-sectional area is less than a preset first cross-sectional area difference threshold, and the difference in the degree of cross-sectional fluctuation is greater than a preset cross-sectional fluctuation difference threshold, then the current height is determined to be a sediment accumulation layer. If the difference in cross-sectional area is less than a preset first cross-sectional area difference threshold, and the difference in the degree of cross-sectional fluctuation is less than or equal to a preset threshold for the degree of cross-sectional fluctuation, then the current height is determined to be a structural diameter reduction. If the difference in cross-sectional area is greater than the preset second cross-sectional area difference threshold, then the current height is determined to be a non-standard straight cylinder expansion section; If the cross-sectional area difference is greater than or equal to a preset first cross-sectional area difference threshold and less than or equal to a second cross-sectional area difference threshold, it is determined to be a standard component; The preset first cross-sectional area difference threshold is the opposite of the preset second cross-sectional area difference threshold.

[0011] Furthermore, the method for obtaining the effective flow area includes: If the current height is a sediment accumulation layer, then the effective flow area is set to the estimated cross-sectional area; If the current height is a structurally narrowed diameter, then the effective flow area is set to the estimated cross-sectional area; If the current height is a non-standard straight cylinder expansion section, then the effective flow area is set to the estimated cross-sectional area; If the current height is a standard component, then the effective flow area will be set to the standard cross-sectional area of ​​the corresponding optimal component type.

[0012] Furthermore, after obtaining the estimated cross-sectional area, the process also includes: The discrete points of each height and the corresponding estimated cross-sectional area in a two-dimensional coordinate system are obtained, where the horizontal axis of the two-dimensional coordinate system is the height and the vertical axis is the estimated cross-sectional area. The discrete points are interpolated using a linear interpolation method to obtain the estimated cross-sectional area corresponding to each height on the horizontal axis.

[0013] Furthermore, after obtaining the height-cross-sectional area sequence, the process further includes: Based on the rate of change of cross-sectional area and the degree of cross-sectional fluctuation at each height, a standard straight cylindrical reference segment is selected from the height-cross-sectional area sequence. The standard cross-sectional area of ​​the standard straight cylindrical component is used as the numerator, and the average estimated cross-sectional area of ​​the standard straight cylindrical reference segment is used as the denominator to obtain the correction coefficient. The correction coefficient is multiplied by the estimated cross-sectional area in the height-cross-sectional area sequence to obtain the updated height-cross-sectional area sequence.

[0014] The present invention has the following beneficial effects: This invention, based on liquid level time-series data, utilizes the liquid level rise rate to invert the height-cross-sectional area sequence. Under the "blind source" condition of unknown flow baseline, it innovatively leverages the scaling-invariant characteristic of the cross-sectional area change rate to achieve feature matching with a standard component library. By constructing a state transition matrix with topological constraints for component sequence deduction, it can automatically reconstruct the physical framework of the manhole conforming to engineering specifications, solving the problem of reliance on manual measurement in traditional methods. Furthermore, by combining the dual criteria of cross-sectional area difference and cross-sectional fluctuation, it can accurately distinguish entity attributes, effectively overcoming the limitation of single geometric dimensions in identifying lining repair and siltation. Ultimately, it generates a hydraulic model of the manhole with effective water level-volume information and clear entity attributes at various heights. This invention relies solely on low-cost liquid level monitoring to generate a high-precision water level-volume relationship table containing real flow boundaries in a low-cost and automated manner, significantly improving the accuracy of pipeline hydraulic models. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of an automated rapid modeling method for underground water supply and drainage network components, provided as an embodiment of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an automated rapid modeling method for underground water supply and drainage network components proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] The following describes in detail, with reference to the accompanying drawings, a specific scheme of the automated rapid modeling method for underground water supply and drainage network components provided by the present invention.

[0020] Please see Figure 1 The diagram illustrates a flowchart of an automated rapid modeling method for underground water supply and drainage network components according to an embodiment of the present invention. The method includes: Step S1: Obtain the liquid level monitoring time series in the target inspection well, estimate the estimated cross-sectional area of ​​each inspection well at each height based on the liquid level rise rate at each time point, and obtain the height-cross-sectional area series; obtain the cross-sectional area change rate and cross-sectional fluctuation degree at each height based on the height-cross-sectional area series.

[0021] In the data monitoring process of drainage pipe network inspection wells, liquid level sensors are used to sample at fixed time intervals, recording the liquid level monitoring time series. Since the rate of liquid level rise is affected by cross-sectional changes, the change in liquid level can indirectly reflect the size of the cross-section, thus avoiding the need to deploy additional sensors or other acquisition measures to collect the cross-sectional area at various height levels on the inspection well during use. Therefore, this embodiment of the invention uses the liquid level monitoring time series as the data basis, extracting data that meets the conditions for fluid dynamics inversion from the continuous liquid level data stream.

[0022] Liquid level sensors record a large amount of low water level data during the dry season or non-full pipe flow data during long-term operation. However, only the full-section geometric characteristics of the wellbore caused by short-term heavy rainfall can reflect the full-pipe filling process. Therefore, the liquid level monitoring time series collected in this embodiment of the invention should be a monotonically increasing sequence with a significant increase. Simultaneously, to ensure that the analysis object is the target monitoring well, the lowest liquid level point in the liquid level monitoring time series should be the lowest point of the target monitoring well, and the highest liquid level point should be the highest point of the target monitoring well. That is, after obtaining a complete initial liquid level monitoring time series, a first sequence is extracted based on the height range of the target inspection well. Then, the total rise in liquid level in the first sequence is calculated. If the total rise in liquid level is greater than 50% of the total depth of the wellbore, it indicates that the first sequence is the liquid level monitoring time series to be analyzed in this embodiment of the invention, indicating that the target of the data is the target inspection well and the time period is the full-pipe filling process caused by short-term heavy rainfall. It should be noted that the short-term heavy rainfall process can be regarded as a process of constant flow input to the inspection well, i.e., the theoretical design flow rate of the pipeline under full-pipe conditions.

[0023] Because the liquid level monitoring time series is a data sequence of liquid level height changing over time, it can be regarded as two-dimensional discrete data with time on the horizontal axis and liquid level on the vertical axis. Therefore, for each liquid level, the liquid level rise rate can be obtained by statistically analyzing the change in liquid level from the previous moment to the current moment. Since the data acquisition in this embodiment of the invention is at a fixed time frequency, the short-term heavy rainfall process can be regarded as a process of constant flow input to the inspection well. Therefore, if the liquid level rise rate is larger, it indicates that the cross-section of the inspection well at the corresponding liquid level height is smaller, and there will be a faster liquid level rise under a fixed flow input. Therefore, the estimated cross-sectional area at each liquid level can be estimated based on the liquid level rise rate at each moment, that is, the estimated cross-sectional area at each height of the target inspection well is obtained.

[0024] After calculating the estimated cross-sectional area at each height, the time-level data was converted into height-estimated cross-sectional area data, completing the conversion from time-series data to spatial data. This data can then be used for subsequent fluid dynamics inversion.

[0025] To achieve fluid dynamics inversion, the rate of change of cross-sectional area and the degree of cross-sectional fluctuation at each height on the inspection well can characterize the fluid characteristics at that height. The rate of change of cross-sectional area can decouple the geometry of the cross-section from the flow rate baseline at the current height; the degree of cross-sectional fluctuation can distinguish between rigid components and flexible deposits under fluid boundary layer theory. Therefore, these two characteristics are significant under fluid dynamics and can be used for component matching and entity property identification in subsequent steps.

[0026] It should be noted that the estimated cross-sectional area in this embodiment of the invention is only an estimate, intended to characterize the size of the cross-section through data size, and is not the actual area size. Because subsequent cross-sectional comparisons with standard components are required, and the liquid level rise rate of the standard components under a fixed flow rate input is also known, the estimated cross-sectional area corresponding to the standard components can be obtained. In this embodiment of the invention, the estimated cross-sectional area of ​​the standard components is denoted as the standard cross-sectional area. The standard component is any component in the prefabricated component parameter library. This parameter library contains the standard cross-sectional area of ​​each standard component, the rate of change of cross-sectional area at each height under a fixed flow rate input, and the degree of cross-sectional fluctuation. In subsequent steps, the parameters pre-stored in this parameter library can be directly called for calculation.

[0027] Preferably, in this embodiment of the invention, the method for estimating the cross-sectional area includes: The liquid level rise rate of each liquid level is obtained based on the difference between adjacent liquid levels. If the liquid level rise rate is less than the preset rate threshold, it indicates that the liquid level rises slowly and the cross-section is large. In subsequent calculations, the cross-sectional result may be impossible to estimate. Therefore, these liquid level rise rates can be directly set to the rate threshold, i.e., the rate threshold is a significantly small value. The ratio of the preset reference inflow rate to the remaining liquid level rise rate is used as the estimated cross-sectional area at the height of the inspection well where the corresponding liquid level is located.

[0028] It should be noted that the reference inflow rate is the theoretical design flow rate of the pipe under full-pipe conditions described in the above embodiments. Its purpose is only to provide a dimensional reference for calculation. That is, the calculation does not require the instantaneous flow rate under actual operating conditions; it only needs to ensure that the estimation algorithm for all cross-sectional areas is the same and that the same reference inflow rate is used. In other words, the estimated cross-sectional area obtained in the embodiments of the present invention is not an area in the conventional physical sense, but rather reflects the size of the cross-sectional area, which is a characteristic value. Only the numerical value is used in the process.

[0029] It should be noted that the rate threshold can be set according to the specific parameters of the inspection well, that is, it can be set according to the benchmark inflow rate. In this embodiment of the invention, it can be set to the data of the last 20% of the liquid level rise rate at each time point after being arranged from large to small. This embodiment of the invention will not elaborate or limit it.

[0030] Preferably, in this embodiment of the invention, the method for obtaining the liquid level rise rate is as follows: the difference between the current liquid level height and the previous liquid level height is used as the numerator, and the time interval between the current liquid level time and the previous time is used as the denominator to obtain the liquid level rise rate.

[0031] It should be noted that, in other embodiments of the present invention, the liquid level monitoring time series can also be mapped to a two-dimensional time-domain coordinate system, i.e., the horizontal axis of the two-dimensional time-domain coordinate system is time and the vertical axis is liquid level. A Savitzky-Golay digital filter is used for time-domain smoothing to preserve low-frequency trends and suppress high-frequency noise. A curve is obtained, and then the time derivative of the smoothed curve is calculated to obtain the liquid level rise rate at each time point. The specific method is a well-known technique to those skilled in the art and will not be elaborated here.

[0032] Preferably, in this embodiment of the invention, considering that the obtained height-cross-sectional area data is significantly discrete because it is converted from time-level height data, resulting in a higher coefficient of data points in areas with a narrower wellbore and a higher density of data points in areas with a wider wellbore, a unified vertical spatial coordinate system must be established to map the data onto a fixed height grid in order for subsequent algorithms to analyze structural features layer by layer. Therefore, after obtaining the estimated cross-sectional area, this embodiment of the invention further includes: The discrete points of each height and the corresponding estimated cross-sectional area in a two-dimensional coordinate system are obtained, where the horizontal axis of the two-dimensional coordinate system is the height and the vertical axis is the estimated cross-sectional area. The discrete points are interpolated using a linear interpolation method to obtain the estimated cross-sectional area corresponding to each height on the horizontal axis.

[0033] It should be noted that linear interpolation is a technique well-known to those skilled in the art, and it can be used to calculate any desired difference in height. For example, the estimated cross-sectional area is expressed by the formula: ;in Height to be interpolated The estimated cross-sectional area below, Distance to the height to be interpolated Most recent and less than The height corresponding to the dispersion, for The height corresponding to the discrete point at the next height of the corresponding discreteness. for The estimated cross-sectional area, for The estimated cross-sectional area.

[0034] The formula for obtaining the estimated cross-sectional area of ​​the above interpolation is a well-known technique in the art. It is based on a previous known data point, obtains the change in data between the previous and subsequent known data points, then determines the weights based on the positional distribution between the interpolation point and the two known data points, and adjusts the base data by weighting the data changes using these weights to obtain the final interpolation result. The specific logic will not be elaborated upon.

[0035] It should be noted that in the above interpolation process, the embodiments of the present invention divide the height range of the target inspection well into multiple height points with a step size of 0.05 meters. That is, the interpolation object is each of the divided height points. If there is a corresponding discrete point under the height point, it means that no interpolation is needed at that position.

[0036] Preferably, in this embodiment of the invention, the estimated cross-sectional area is considered to be in an uncalibrated state due to the lack of real inflow data. To ensure that the estimated cross-sectional area more accurately represents the cross-sectional area information, considering that municipal inspection wells all contain a standard straight-tube manhole ring with a height greater than 0.5 meters, this prior engineering knowledge can be used to lock onto a stable segment in the height-cross-sectional area sequence to calibrate other data. Specifically, this includes: Based on the rate of change of cross-sectional area and the degree of cross-sectional fluctuation at each height, a standard straight cylindrical reference segment is selected from the height-cross-sectional area sequence. In this embodiment of the invention, data segments with a rate of change of cross-sectional area less than a preset rate of change threshold and a degree of cross-sectional fluctuation less than a preset degree of fluctuation threshold are selected as standard straight cylindrical reference segments.

[0037] Because the standard straight-tube well ring is a standard straight tube, based on the statistical characteristics of the smooth concrete well wall, the area fluctuation of the smooth wall surface during a steady rise in liquid level mainly originates from sensor quantization noise, which has extremely low variance. However, when flowing through an irregular sediment layer, turbulence will cause significant fluctuations in area calculation. This threshold is usually taken as 2-3 times the noise floor. Therefore, the rate of change of the cross-sectional area of ​​this standard component should be a minimum value close to zero. Thus, the rate of change threshold can be set to a small value close to zero, such as 0.05. The value range can be set between 0.05 and 0.1, which is not limited or elaborated here. Similarly, the fluctuation degree threshold is also a small value. In this embodiment of the invention, it is set to 0.002, which can be set between 0.002 and 0.005, which is not limited or elaborated here.

[0038] Using the standard cross-sectional area of ​​the standard straight cylindrical assembly as the numerator and the average estimated cross-sectional area of ​​the standard straight cylindrical reference segment as the denominator, a correction coefficient is obtained. This correction coefficient can be characterized as a proportional multiplication correction applied to all obtained estimated cross-sectional areas, that is, multiplying the correction coefficient by the estimated cross-sectional area in the height-cross-sectional area sequence to obtain the updated height-cross-sectional area sequence.

[0039] It should be noted that if multiple standard straight-tube reference segments are available, the longest segment should be selected as the standard straight-tube reference segment for subsequent analysis. If no standard straight-tube reference segment is available, no further correction will be performed, and the current data should be recorded as low-confidence data, as the data reliability may be reduced due to random factors such as acquisition errors. This set of data may be deleted. Furthermore, in other implementations of this invention, to ensure the accuracy of the standard straight-tube reference segment, a height search range can be set on the inspection well. That is, the standard straight-tube well ring should be located near the top of the well. By pre-dividing the search range, the search volume can be reduced. In this embodiment, the height can be set to one meter from the top, and further details are omitted.

[0040] Preferably, in this embodiment of the invention, the method for obtaining the degree of cross-sectional fluctuation includes: In the height-cross-sectional area sequence, for each height, a preset height range is constructed centered on that height. Within this preset height range, the variance of the estimated cross-sectional area is calculated to obtain the degree of cross-sectional fluctuation. The height range is defined as the range of three height points above and below the target height, containing a total of seven height points. A larger variance indicates a more chaotic cross-sectional area distribution within the range centered on the target height. This feature is based on fluid boundary layer theory. When fluid flows through a rough and irregularly shaped sediment layer, local turbulence and micro-fluid level fluctuations occur, leading to high-frequency fluctuations in the calculated cross-sectional area. Conversely, when flowing through the smooth inner wall of a precast concrete component, the liquid level changes steadily. Therefore, this feature can be used to evaluate fluid properties at the target height.

[0041] Step S2: Based on the estimated cross-section and cross-sectional area change rate corresponding to the height, match it with the features of the standard component at the corresponding height to obtain the matching difference between each height and each standard component.

[0042] This invention aims to evaluate the corresponding component type at each height by analyzing the fluid dynamics of a manhole. Because manholes can change shape due to blockages, maintenance, and other factors during actual use, it's unclear whether the shape at each height corresponds to the original component during modeling. Those skilled in the art cannot reconstruct the absolute physical dimensions and structural details of a node based solely on liquid level data in the absence of real inflow data (a "blind source"). Therefore, this invention matches each standard component in the prefabricated component parameter library for each height, using the estimated cross-section and cross-sectional area change rate as the matching criteria to obtain the matching difference between each height and each standard component.

[0043] It should be noted that, in a specific implementation of this invention, a matching difference matrix can be constructed, where the rows of the matching difference matrix represent each height and the columns represent each standard component. Each element in the matrix represents the matching difference between the corresponding height and the corresponding standard component.

[0044] Preferably, in this embodiment of the invention, considering that a standard component may have multiple heights, as in a specific implementation of this embodiment, the height division method described above can be used to divide the component in steps of 0.05 meters to obtain multiple heights on the standard component. In this embodiment of the invention, these are referred to as local heights. That is, each local height will have a matching result with the current target height of the inspection well. The specific method for obtaining the matching difference includes: Use the height of the standard component as the local height; For each local height, a first difference is obtained between the estimated cross-sectional area of ​​the inspection well at that height and the standard cross-sectional area of ​​the standard component at that local height; a second difference is obtained between the rate of change of cross-sectional area at that height and the corresponding rate of change of standard cross-sectional area of ​​the standard component at that local height; the first difference and the second difference are weighted and summed using a preset threshold to obtain the initial difference degree between each local height and the current height of the inspection well. That is, the first difference is a size difference, used to constrain the consistency of the two cross-sectional sizes; the second difference is a morphological difference, used to constrain the consistency of the shape change trend. For example, the standard component is a tapered cone, which has a varying height and a varying rate of change of cross-sectional area, therefore the second difference is needed to compare the morphological differences at each height. In this embodiment of the invention, the weight of the first difference is set to 0.6, and the weight of the second difference is set to 0.4, indicating that this embodiment of the invention focuses more on size differences. In other embodiments of this invention, the weights can be set according to actual needs, which will not be elaborated or limited here.

[0045] The minimum initial difference among all local heights is taken as the matching difference between the current inspection well height and the standard component.

[0046] It should be noted that, in the embodiments of the present invention, the first difference and the second difference are both the squared differences between two corresponding features, which can be expressed by the formula: ;in Represents the k-th height The degree of matching difference with the j-th standard component, For the first difference, For height The estimated cross-sectional area below, Let H be the standard cross-sectional area of ​​the j-th standard component at height H. For height Rate of change of cross-sectional area Let be the rate of change of the standard cross-sectional area of ​​the j-th standard component at height H. This means finding the minimum value among all the initial differences obtained from H.

[0047] It should be noted that this matching difference is used for subsequent selection of the optimal component type. The smaller the value, the better the match. Therefore, there is no need to consider the unit of the matching difference during the calculation process; only the numerical value needs to be retained for subsequent processing. In other implementations of this invention, normalization processing can also be performed. The normalization step is a common mathematical algorithm used by those skilled in the art, and will not be elaborated or limited here.

[0048] Step S3: Construct a state transition matrix. The rows and columns of the state transition matrix represent the number of standard components, and the elements are the allowable probabilities of transitioning from a lower-level component to an upper-level component. Based on the allowable probabilities and the matching difference, calculate from the lowest height to the highest height to obtain the optimal component type corresponding to each height.

[0049] The stacking of components must comply with civil engineering assembly specifications (e.g., manhole covers cannot be placed directly on the manhole base, and a manhole ring or cover must be connected above the tapered section). To eliminate non-compliant combinations, this embodiment of the invention employs the Viterbi algorithm to find the global optimum under topological constraints. First, a state transition matrix needs to be constructed. The rows and columns of the state transition matrix represent the number of standard components, and the elements are the permissible probabilities of transitioning from a lower-level component to an upper-level component. For example, in the state transition matrix... This represents the allowed probability of transitioning from component i at a lower height to component j at an upper height.

[0050] Preferably, in this embodiment of the invention, the allowed probability is a preset binary probability, including 0 and positive infinity. 0 indicates that the lower-level height component corresponding to the current element is allowed to transition to the upper-level height component corresponding to the current element; positive infinity indicates that the lower-level height component corresponding to the current element is not allowed to transition to the upper-level height component corresponding to the current element. Because the height in this embodiment of the invention is a discrete, subdivided height, there may be multiple consecutive heights belonging to the same component. Therefore, taking the three standard components—base, well ring, and concave cone—as examples, this embodiment of the invention allows transitions from base to base, base to well ring, well ring to well ring, well ring to concave cone, and concave cone to concave cone. However, it does not allow transitions from base to concave cone or from concave cone to well ring, because these are not in accordance with engineering specifications. Therefore, the state transition matrix obtained using these three standard components as examples can be expressed as: That is, the matrix has three rows and three columns, and the ∞ in the elements represents the infinity symbol.

[0051] After constructing the state transition matrix, and combining the matching difference results from the previous steps, inference is performed from the bottom layer to the top layer, i.e., from the lowest height to the highest height, to obtain the optimal component type for each height. It should be noted that because the state matrix represents the permissible distribution between lower-level components and upper-level components, and since there is no lower layer for the bottom layer, the inference for the bottom layer should directly calculate the matching difference with the standard bottom-level components, without needing to calculate with other components.

[0052] Preferably, in this embodiment of the invention, the method for calculating the optimal component type includes: The cumulative cost between each height and each standard component is obtained using a cumulative cost calculation formula, which includes: ;in This represents the cumulative cost between the k-th height and the j-th standard component; This represents the cumulative cost between component i and the previous height k-1. Let i be the allowed probability in the state transition matrix when the lower-level height component is i and the upper-level height component is j. This means iterating through all components i and taking the minimum value; For the kth height The degree of matching difference with the j-th standard component; Obtain the minimum cumulative cost corresponding to the highest height, and backtrack using the minimum cumulative cost to obtain the optimal component type corresponding to all heights.

[0053] It should be noted that the process of the Viterbi algorithm finding the global optimal solution under topological constraints is a well-known process in the field. However, in the cumulative cost formula, this invention innovatively combines the matching difference degree and the allowable probability of the state matrix. The formula utilizes the result of the minimum value function, so that the optimal result corresponding to the bottom layer height can be clearly and effectively known during the final backtracking process.

[0054] As an example, taking 5 height layers (i.e., k=[1 2 3 4 5]) and three standard components as an example, the process of finding the optimal matching difference includes: (1) k=1: At the bottom level, because there is no lower level, You can get , , Three possible outcomes; (2) k=2: The calculation is performed according to the cumulative cost formula described above. , , ; (3) k=3: The same logic applies as above when k equals 2, so I will not repeat it here. (4) k=4: The same principle applies as above when k equals 2, so I will not repeat it here.

[0055] (5) k=5: The same logic applies as above when k equals 2, so I won't repeat it further. The final result is the minimum... By backtracking forward, the minimum value selected by the minimum value function in the cumulative cost calculation formula can be determined, and the optimal component type can be obtained. For example, [1, 1, 2, 2, 3] means that the first two height layers are base components, the middle two height layers are well ring components, and the top height layer is conical component.

[0056] Step S4: Determine the entity attribute of each height based on the difference in cross-sectional area between each height and the optimal component type, as well as the difference in cross-sectional fluctuation. Determine the effective flow area of ​​each height based on the entity attribute, and generate a water level-volume relationship table based on the effective flow area.

[0057] Once the optimal component type is determined, the component type corresponding to each height of the current target inspection well can be directly visualized and labeled during the modeling process. Furthermore, considering that a simple geometric skeleton cannot be directly used for hydraulic calculations, and that standard design drawings often fail to reflect sedimentation during operation or non-standard changes during construction, this embodiment of the invention needs to further determine the entity attributes of each height. These entity attributes include sediment accumulation layers, structural diameter reductions, non-standard straight-tube expansion sections, and standard components. After further determining the entity attributes of each layer, accurate modeling can be performed based on the specific entity attributes, and a water level-volume relationship table can be generated based on the effective flow area for model calculations.

[0058] Preferably, in this embodiment of the invention, the method for determining entity attributes includes: The difference in cross-sectional area is the difference between the estimated cross-sectional area at the current height and the standard cross-sectional area of ​​the optimal component type. This difference can be positive or negative. If it is negative and small, it means that the estimated cross-sectional area at the current height is significantly smaller than the standard cross-sectional area; if it is positive and large, it means that it is significantly larger than the standard cross-sectional area.

[0059] If the difference in cross-sectional area is less than a preset first cross-sectional area difference threshold, and the difference in cross-sectional fluctuation degree is greater than a preset cross-sectional fluctuation degree difference threshold, it indicates that the estimated cross-sectional area at the current height is significantly less than the standard cross-sectional area, that is, the measured area is less than the standard area, and the flow fluctuation is significantly higher than the threshold reflected by the smooth wall surface, then the current height is determined to be a sediment accumulation layer. If the difference in cross-sectional area is less than a preset first cross-sectional area difference threshold, and the difference in cross-sectional fluctuation degree is less than or equal to a preset cross-sectional fluctuation degree difference threshold, it indicates that the estimated cross-sectional area of ​​the current height is significantly smaller, but the flow is smooth, and the current height is judged to be a structural diameter reduction. If the difference in cross-sectional area is greater than the preset second cross-sectional area difference threshold, it indicates that the estimated cross-sectional area at the current height is significantly larger, and the current height is determined to be a non-standard straight cylinder expansion section. If the cross-sectional area difference is greater than or equal to a preset first cross-sectional area difference threshold and less than or equal to a second cross-sectional area difference threshold, it is determined to be a standard component; The preset first cross-sectional area difference threshold is the opposite of the preset second cross-sectional area difference threshold.

[0060] In this embodiment of the invention, the second cross-sectional area difference threshold is set to 0.03, specifically in the range of 0.03 to 0.05, or 5% to 10% of the standard cross-sectional area, to represent a deviation tolerance.

[0061] The threshold for cross-sectional fluctuation can be set between 0.01 and 0.02. In this embodiment of the invention, it is set to 0.01. This threshold is used to distinguish between smooth walls and rough deposits. In addition, experimental observations show that the calculation variance caused by local turbulence generated when fluid flows through a sediment deposit surface is usually more than three times that when flowing through a smooth pipe wall. Therefore, this threshold can be set to 2-4 times the fluctuation threshold of the smooth standard straight-tube well ring described in step S1.

[0062] Preferably, based on the above-mentioned entity attribute discrimination, the method for obtaining the effective flow area in this embodiment of the invention includes: If the current height is a sediment accumulation layer, then the effective flow area is set to the estimated cross-sectional area; If the current height is a structurally narrowed diameter, then the effective flow area is set to the estimated cross-sectional area; If the current height is a non-standard straight cylinder expansion section, then the effective flow area is set to the estimated cross-sectional area; If the current height is a standard component, the effective flow area is set to the standard cross-sectional area of ​​the corresponding optimal component type. That is, in the case of sediment accumulation layers, structural diameter reduction, and non-standard straight cylinder expansion sections, the standard cross-sectional area marked on the drawings when initially deploying the inspection well, as well as the standard cross-sectional area of ​​the optimal component matched by the embodiment of the present invention, are not meaningful for reference. The actual measured cross-sectional area should be used as the reference, that is, the estimated cross-sectional area at the current height should be used as the reference.

[0063] Ultimately, in this embodiment of the invention, drainage network hydraulic simulation software such as SWMM can be used to calculate the water storage volume of nodes at different water levels by querying the water level-storage capacity curve. Based on the aforementioned effective flow area, the trapezoidal integral method is used to generate the curve corresponding to the final water level-volume relationship table. Specifically, the cumulative volume is first initialized. ( (where M is the bottom elevation of the well), calculate the cumulative volume layer by layer from the bottom of the well k=1 to K=M, where M represents the index of the top layer. The formula is as follows: ;in This represents the cumulative volume at the k-th height. The cumulative volume at height k-1. Let be the effective flow area at the k-th height. The effective flow area at the (k-1)th height. Let k be the height difference between the k-th height and the k-1 height.

[0064] It should be noted that the above formula is a commonly used formula for calculating cumulative volume by those skilled in the art. The formula's logic is to adjust the height difference based on the difference in effective flow area to obtain the volume change relative to the lower layer's height. Adding this volume change to the lower layer's height yields the cumulative volume of the current layer. The calculation results can then be compiled into a water level-volume relationship table containing multiple data pairs. This table accurately describes the actual storage capacity of the inspection well at different water levels and can be directly input as parameters into the hydraulic model, achieving a closed-loop application from monitoring data to model parameters.

[0065] In summary, this invention analyzes liquid level monitoring time series, estimates the cross-sectional area at each height using the liquid level rise rate, and extracts the cross-sectional area change rate and cross-sectional fluctuation characteristics. Subsequently, it matches the cross-sectional area and change rate with standard component characteristics to construct a difference matrix, and combines this with a state transition matrix containing component topological constraints to dynamically calculate the optimal component type sequence. Finally, based on the difference between the measured cross-sectional area and the optimal component, and the cross-sectional fluctuation characteristics, it identifies entity attributes, corrects the flow area, and generates a water level-volume relationship table. This invention can automatically calibrate absolute physical dimensions and accurately identify wellbore defects even in the absence of inflow data, improving the accuracy of pipeline hydraulic models.

[0066] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0067] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. An automated and rapid modeling method for underground water supply and drainage network components, characterized in that, The method includes: Obtain the liquid level monitoring time series in the target inspection well, estimate the cross-sectional area of ​​the target inspection well at each height based on the liquid level rise rate at each time point, and obtain the height-cross-sectional area series; obtain the cross-sectional area change rate and cross-sectional fluctuation degree at each height based on the height-cross-sectional area series; Based on the estimated cross-section and cross-sectional area change rate corresponding to the height, the feature of the standard component at the corresponding height is matched to obtain the matching difference degree between each height and each standard component; Construct a state transition matrix, where the rows and columns represent the number of standard components, and the elements are the allowable probabilities of transitioning from a lower-level component to an upper-level component. Based on the allowable probabilities and the matching difference, calculate from the lowest height to the highest height to obtain the optimal component type corresponding to each height. Based on the difference in cross-sectional area between each height and the optimal component type, as well as the difference in the degree of cross-sectional fluctuation, the entity attribute to which each height belongs is determined; the effective flow area of ​​each height is determined based on the entity attribute, and a water level-volume relationship table is generated based on the effective flow area.

2. The automated rapid modeling method for underground water supply and drainage network components according to claim 1, characterized in that, The method for calculating the estimated cross-sectional area includes: The liquid level rise rate of each liquid level is obtained based on the difference between adjacent liquid levels; if the liquid level rise rate is less than a preset rate threshold, the liquid level rise rate is set to the preset rate threshold; the ratio of the preset reference inflow rate to the remaining liquid level rise rate is used as the estimated cross-sectional area at the height of the inspection well where the corresponding liquid level is located.

3. The automated rapid modeling method for underground water supply and drainage network components according to claim 1, characterized in that, The method for obtaining the degree of cross-sectional fluctuation includes: In the height-cross-sectional area sequence, for each height, a preset height range is constructed with the height as the center, and the variance of the estimated cross-sectional area is calculated within the preset height range to obtain the degree of cross-sectional fluctuation.

4. The automated rapid modeling method for underground water supply and drainage network components according to claim 1, characterized in that, The method for obtaining the matching difference includes: Use the height of the standard component as the local height; For each local height, a first difference is obtained between the estimated cross-sectional area of ​​the inspection well at that height and the standard cross-sectional area of ​​the standard component at that local height; a second difference is obtained between the rate of change of the cross-sectional area at that height and the rate of change of the standard cross-sectional area of ​​the standard component at that local height; the first difference and the second difference are weighted and summed using a preset threshold to obtain the initial difference degree between each local height and the current height of the inspection well; The minimum initial difference among all local heights is taken as the matching difference between the current inspection well height and the standard component.

5. The automated rapid modeling method for underground water supply and drainage network components according to claim 1, characterized in that, The allowed probability is a preset binary probability, including 0 and positive infinity. 0 means that the lower-level height component corresponding to the current element is allowed to transition to the upper-level height component corresponding to the current element; positive infinity means that the lower-level height component corresponding to the current element is not allowed to transition to the upper-level height component corresponding to the current element.

6. The automated rapid modeling method for underground water supply and drainage network components according to claim 1, characterized in that, The method for calculating the optimal component type includes: The cumulative cost between each height and each standard component is obtained using a cumulative cost calculation formula, which includes: ;in This represents the cumulative cost between the k-th height and the j-th standard component; This represents the cumulative cost between component i and the previous height k-1. Let i be the allowed probability in the state transition matrix when the lower-level height component is i and the upper-level height component is j. This means iterating through all components i and taking the minimum value; For the kth height The degree of matching difference with the j-th standard component; Obtain the minimum cumulative cost corresponding to the highest height, and backtrack using the minimum cumulative cost to obtain the optimal component type corresponding to all heights.

7. The automated rapid modeling method for underground water supply and drainage network components according to claim 1, characterized in that, The methods for determining the entity attributes include: The cross-sectional area difference is the difference between the estimated cross-sectional area at the current height and the standard cross-sectional area of ​​the optimal component type; If the difference in cross-sectional area is less than a preset first cross-sectional area difference threshold, and the difference in the degree of cross-sectional fluctuation is greater than a preset cross-sectional fluctuation difference threshold, then the current height is determined to be a sediment accumulation layer. If the difference in cross-sectional area is less than a preset first cross-sectional area difference threshold, and the difference in the degree of cross-sectional fluctuation is less than or equal to a preset threshold for the degree of cross-sectional fluctuation, then the current height is determined to be a structural diameter reduction. If the difference in cross-sectional area is greater than the preset second cross-sectional area difference threshold, then the current height is determined to be a non-standard straight cylinder expansion section; If the cross-sectional area difference is greater than or equal to a preset first cross-sectional area difference threshold and less than or equal to a second cross-sectional area difference threshold, it is determined to be a standard component; The preset first cross-sectional area difference threshold is the opposite of the preset second cross-sectional area difference threshold.

8. The automated rapid modeling method for underground water supply and drainage network components according to claim 7, characterized in that, The method for obtaining the effective flow area includes: If the current height is a sediment accumulation layer, then the effective flow area is set to the estimated cross-sectional area; If the current height is a structurally narrowed diameter, then the effective flow area is set to the estimated cross-sectional area; If the current height is a non-standard straight cylinder expansion section, then the effective flow area is set to the estimated cross-sectional area; If the current height is a standard component, then the effective flow area will be set to the standard cross-sectional area of ​​the corresponding optimal component type.

9. The automated rapid modeling method for underground water supply and drainage network components according to claim 1, characterized in that, After obtaining the estimated cross-sectional area, the process also includes: The discrete points of each height and the corresponding estimated cross-sectional area in a two-dimensional coordinate system are obtained, where the horizontal axis of the two-dimensional coordinate system is the height and the vertical axis is the estimated cross-sectional area. The discrete points are interpolated using a linear interpolation method to obtain the estimated cross-sectional area corresponding to each height on the horizontal axis.

10. The automated rapid modeling method for underground water supply and drainage network components according to claim 1, characterized in that, After obtaining the height-cross-sectional area sequence, the process further includes: Based on the rate of change of cross-sectional area and the degree of cross-sectional fluctuation at each height, a standard straight cylindrical reference segment is selected from the height-cross-sectional area sequence. The standard cross-sectional area of ​​the standard straight cylindrical component is used as the numerator, and the average estimated cross-sectional area of ​​the standard straight cylindrical reference segment is used as the denominator to obtain the correction coefficient. The correction coefficient is multiplied by the estimated cross-sectional area in the height-cross-sectional area sequence to obtain the updated height-cross-sectional area sequence.

Images