A multi-cutting point sewage distribution method and device considering the spatial heterogeneity of sewage pipe network

Abstract

The present invention discloses a multi-cutting point sewage distribution method and device that considers the spatial heterogeneity of sewage pipe networks. The method comprises the following steps: obtaining sewage pipe network data and historical flow monitoring data for a target area and cleaning them; establishing a sewage pipe network topology map based on the cleaned sewage pipe network data, and generating a water collection range map for each pump station; calculating the initial value of the average inflow rate of a single node in each administrative district based on the water collection range map of each pump station, combined with the per capita daily sewage discharge and the total population of each administrative district; calculating the theoretical total water collection rate based on the initial value of the average inflow rate of the single node, and preliminarily calibrating the average inflow rate of the single node based on the ratio of the actual total inflow rate to the theoretical total water collection rate; identifying the control domain of each cut point and the independent control domain of each node of interest; calibrating the theoretical inflow rate of each control domain, and then obtaining the average inflow rate of the single node after secondary calibration. The method has higher accuracy than the traditional DWF estimation method.

Description

Technical Field

[0001] The present invention relates to urban drainage technology, and in particular to a multi-cutting point sewage distribution method and equipment taking into account the spatial heterogeneity of a sewage pipe network. Background Art

[0002] In the hydrological and hydraulic modeling of urban drainage systems, accurately estimating daily sewage discharge at each node is fundamental to building high-precision models. Such models are not only used to simulate pipe network operation but also play a key role in drainage system performance assessment, pipe network capacity analysis, water pollution load forecasting, and urban infrastructure optimization design (such as system renovation and overflow control). Daily sewage discharge in the absence of rainfall is commonly referred to as dry weather flow (DWF). Its primary sources include residential sewage, commercial service water, institutional discharge, and a certain proportion of industrial water. DWF forms the "baseline" of the drainage system flow process. Superimposing seepage water and rainfall runoff on this basis forms the complete operational process of the pipe network system. Because DWF is a crucial background flow, inaccurate estimation of its spatial distribution will directly affect the accuracy and operability of the model in dry weather load analysis, overflow tracing, and storage capacity assessment.

[0003] Traditional DWF estimation methods are typically based on population, calculating the average daily sewage discharge at each node simply by combining service area divisions with sewage generation parameters per unit population. This method is simple to operate and requires only a single source of data, making it widely used in practical projects. However, given the complex and diverse nature of urban land use, current pipeline network models based on population estimation have significant limitations:

[0004] 1) Heterogeneous drainage behavior in multifunctional land leads to biased applicability of the population method: The residential population cannot represent the actual water use behavior in commercial and industrial areas. For example, areas with dense concentrations of commercial office buildings and restaurants, despite having a sparse population, have significantly higher drainage loads than ordinary residential areas. Drainage in industrial areas primarily comes from process water and production wastewater, and its discharge intensity is almost unrelated to population. The traditional population coefficient method struggles to reflect the emission characteristics of these areas, which can easily lead to biased estimation of system source terms. Institutional areas such as schools, hospitals, and office parks serve a population with strong temporal fluctuations. Fixed population estimation parameters are difficult to adapt to their drainage variations, leading to underestimation of peak loads and misjudgment of valley flows. Cities are home to a large number of mixed-use areas, including residential, commercial, and office areas. The drainage intensities of different functional areas vary significantly, and averaging the population can easily mask true spatial differences, affecting the spatial accuracy of the model.

[0005] 2) Ignoring groundwater infiltration and illegal access: Traditional methods often fail to account for background flows such as groundwater infiltration and illegal access during rainy days, leading to inaccurate flow estimates. These errors are particularly significant in older pipe networks or areas with high groundwater levels. Aging pipes, loose joints, and inadequate maintenance can easily lead to persistent groundwater infiltration, resulting in a stable baseflow. If this is not accounted for, the model is prone to systematically underestimate dry-weather flows. Ignoring this baseflow not only leads to a significant underestimation of dry-weather baseline flows but also further impacts the accuracy of key indicators such as pipeline load, overflow frequency, and storage capacity in subsequent simulations.

[0006] 3) Lack of data closed-loop verification and correction mechanism, limited model accuracy: Most existing population estimation models use static parameter settings. When expressing node sewage inflow, they mainly assign values ​​based on limited data. They fail to fully consider heterogeneous factors such as the water inflow of sewage treatment plants, the water collection area upstream of the pumping station, differences in the service population, and the structure of the drainage network. It is difficult to dynamically adjust and iteratively calibrate the flow distribution according to actual operating conditions, resulting in limited simulation accuracy and difficulty in truly reflecting the spatial differences in the sewage generation and transportation process. Only the population center of gravity is used as a distribution reference, which can easily lead to inaccurate simulation of the sewage transmission path in space, thereby limiting the model's ability to reflect the actual operating laws and easily causing distortion at key locations such as bifurcations and the front end of the storage tank.

[0007] In summary, the traditional DWF estimation method has large distortion and low accuracy. It is necessary to improve the spatial resolution and rationality of sewage source allocation to enhance the reliability and engineering applicability of model simulation results, and provide strong support for the healthy operation of urban drainage systems and pollution prevention and control. Summary of the Invention

[0008] In view of the problems existing in the prior art, the purpose of the present invention is to provide a more accurate multi-cutting point sewage distribution method and equipment that takes into account the spatial heterogeneity of the sewage network.

[0009] In order to achieve the above-mentioned object of the invention, the present invention provides the following technical solutions:

[0010] A multi-cutting point sewage distribution method considering the spatial heterogeneity of sewage pipe networks includes the following steps:

[0011] (1) Obtain sewage pipe network data and historical flow monitoring data in the target area and clean the data;

[0012] (2) Based on the cleaned sewage pipe network data, establish a sewage pipe network topology map and generate a water collection range map for each pumping station;

[0013] (3) Based on the administrative districts included in the water collection range map of each pumping station, combined with the per capita daily sewage discharge and the total population of each administrative district, calculate the initial value of the average inflow flow of each single node in each administrative district;

[0014] (4) Calculate the theoretical total water intake of the target area based on the initial value of the average inflow of a single node, calculate the actual total inflow of the target area based on the historical flow monitoring data, and use the ratio of the actual total inflow to the theoretical total water intake as the first calibration coefficient to perform a preliminary calibration on the initial value of the average inflow of a single node;

[0015] (5) Identify all the cut-off points and the control domain of each cut-off point in the sewage network topology diagram, as well as the independent control domain of each node of interest, wherein the cut-off point is the intersection of the common water collection path of at least two nodes of interest, and there are at least two independent and non-overlapping branch paths in the downstream path, each branch path reaches a different node of interest, and the node of interest is a node with flow monitoring function;

[0016] (6) For each concerned node, the independent control domain of the concerned node and the control domain of each upstream split point are respectively used as a flow monitoring related control domain of the concerned node;

[0017] (7) With the goal of ensuring that the difference between the theoretical total inflow and the monitored flow of each node of interest in the target area is within the preset threshold range, the secondary calibration coefficient is calculated to perform secondary calibration on the average inflow of a single node after the preliminary calibration. The theoretical total inflow is the sum of the theoretical inflows of all flow monitoring-related control domains corresponding to the node of interest after calibration using the secondary calibration coefficient.

[0018] Furthermore, after step (7), the following steps are further included:

[0019] (8) The average inflow of a single node after secondary calibration is weighted according to the sewage discharge pattern curve of the population land use type of the control domain, and the sewage discharge time series data of a single node in each time period throughout the day are obtained; wherein, the sewage discharge pattern curve is a set of time distribution coefficients that describe the change law of sewage discharge in each time period throughout the day, and the sum of the time distribution coefficients is 1.

[0020] Furthermore, step (2) specifically includes the following steps:

[0021] (2.1) Based on the cleaned sewage pipe network data, establish a sewage pipe network topology map with sewage pipe lines as edges and sewage pipe end points as nodes;

[0022] (2.2) Based on the sewage network topology, the upstream nodes and sewage pipelines of each pumping station are identified through a depth-first search algorithm, and a water collection range map of each pumping station is generated based on the upstream nodes and sewage pipelines.

[0023] Furthermore, step (3) specifically includes the following steps:

[0024] (3.1) Spatially overlay the administrative district boundary map of the target area with the water collection range map of each pumping station, and calculate the proportion of area within each administrative district that falls within the water collection range of the pumping station;

[0025] (3.2) Spatially overlay the administrative district boundary map, sewage network topology map, and pump station water collection area map of the target area, and count the number of sewage pipelines within the pump station water collection area in each administrative district;

[0026] (3.3) For each administrative district, multiply the total population of the administrative district by the corresponding area percentage and divide the result by the number of sewer lines within the pumping station in the administrative district to obtain the number of people served by each sewer pipe in the administrative district;

[0027] (3.4) For each administrative district, obtain the average daily sewage discharge per capita in the administrative district and multiply the average daily sewage discharge per capita by the number of people served by a single sewer pipe to obtain the initial value of the average inflow per node.

[0028] Furthermore, step (4) specifically includes the following steps:

[0029] (4.1) For each pumping station, the theoretical total water intake of the pumping station is the sum of the initial average inflow of individual nodes in all administrative districts within the pumping station's water intake range and the product of the number of sewage pipelines in the administrative district within the pumping station's water intake range;

[0030] (4.2) Add the theoretical total water collection capacity of all pumping stations in the target area to obtain the theoretical total water collection capacity of the target area;

[0031] (4.3) Calculate the actual total inflow to the target area based on the historical flow monitoring data of all pumping stations in the target area;

[0032] (4.4) Divide the actual total inflow of the target area by the theoretical total water intake to obtain the first calibration coefficient of the target area;

[0033] (4.5) Multiply the initial value of the average inflow of a single node in each administrative area by the first calibration coefficient to obtain the average inflow of a single node after preliminary calibration.

[0034] Furthermore, step (5) specifically includes the following steps:

[0035] (5.1) Identify all split points in the sewage network topology, where the split point is the intersection of at least two common catchment paths of the nodes of interest, and the downstream path has at least two independent, non-overlapping branch paths, each branch path reaching a different node of interest;

[0036] (5.2) Identify the control domain of each split point, where the control domain of a split point is the area between the current split point and its upstream adjacent split point, and the control domain of the upstreammost split point is the entire area upstream of the split point;

[0037] (5.3) Identify the independent control domain of each concerned node, wherein the independent control domain of the concerned node is the area between the concerned node and the adjacent upstream split point.

[0038] Furthermore, step (6) specifically includes:

[0039] (6.1) For each node of interest, starting from the current node of interest, use the depth-first search method to traverse upstream in reverse order to obtain the set U of all upstream nodes that can reach the current node of interest;

[0040] (6.2) Obtain all the split points to form a split point set C;

[0041] (6.3) Calculate the intersection of the split point set C and the upstream node set U of the current concerned node as the upstream split point set C′ of the current concerned node;

[0042] (6.4) The control domain of each split point in the upstream split point set C′ of the current concerned node, as well as the independent control domain of the current concerned node, are used as the flow monitoring related control domain of the current concerned node.

[0043] Furthermore, step (7) specifically includes the following steps:

[0044] (7.1) For each flow monitoring-related control domain, the average inflow of each node after preliminary calibration of all its inflow nodes is summarized as the initial theoretical inflow;

[0045] (7.2) Establish a calibration coefficient optimization model, where the objective function of the calibration coefficient optimization model is to minimize the difference between the theoretical total inflow and the monitored inflow for each node of interest. The optimization parameter is the second calibration coefficient of each flow monitoring-related control domain. The theoretical total inflow is the sum of the calibrated theoretical inflows of all flow monitoring-related control domains corresponding to the node of interest, and the calibrated theoretical inflow of a flow monitoring-related control domain is the product of its initial theoretical inflow and the corresponding secondary calibration coefficient.

[0046] (7.3) Solving the calibration coefficient optimization model to obtain the second calibration coefficient of each flow monitoring related control domain;

[0047] (7.4) Multiply the average inflow of each single node after preliminary calibration by the second calibration coefficient of the relevant control domain of the flow monitoring to which it belongs to obtain the average inflow of the single node after secondary calibration.

[0048] Furthermore, the calibration coefficient optimization model described in step (7.2) is specifically:

[0049] ,

[0050] ,

[0051] ,

[0052] ,

[0053] ,

[0054] st ,

[0055] Where, represents the second calibration coefficient vector, represents the second calibration coefficient of the flow monitoring related control domain y, where Y represents the number of all flow monitoring related control domains in the target area. represents the monitoring traffic matrix of the concerned node, represents the monitoring traffic of the concerned node x, X represents the number of concerned nodes in the target area, Represents the initial estimated inflow matrix, specifically an X-row Y-column matrix, express The xth row and yth column element of Indicates whether the traffic monitoring related control domain y belongs to The indicator variable, represents the initial estimated inflow of the control domain y related to flow monitoring, Represents the set of all traffic monitoring related control domains concerned with node x.

[0056] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above method.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] 1. Improving the spatial representation accuracy of node inflow boundary conditions: Utilizing multi-source data such as pipe network topology, population data, monitoring data, and pump station water collection areas, we scientifically characterize the spatial heterogeneity of sewage generation at each node, refine sewage distribution to the level of individual cut-off points, and combine dynamic calibration of the cut-off point control domain to significantly improve the rationality and physical consistency of the distribution of inflow boundary conditions in the model, providing a more realistic input basis for subsequent simulation calculations and improving overall accuracy.

[0059] 2. Enhance the model's adaptability to time-varying operating conditions and development scenarios: The model introduces a first calibration coefficient and a second calibration coefficient for dynamic adjustment, enabling responsiveness to temporal changes in node sewage inflow and planning development trends (such as population growth, water use structure adjustments, and drainage pattern evolution), effectively supporting system prediction and assessment under multiple scenarios.

[0060] 3. Enhance the model's adaptability to complex pipe network structures: By introducing an algorithm that automatically identifies the control domain of the pipe network's split points, the control domain of the pipe network can be dynamically divided. This method can construct a flow topology structure that is more in line with the actual operation logic while ensuring hydraulic continuity, and accurately depict the sewage flow path and control relationship. This move significantly improves the model's adaptability in urban sewage pipe networks with complex structures and diverse functions, ensuring that even in areas with dense water confluence, mixed connection areas, and dense discharge nodes, the calculation accuracy and topological consistency can be maintained, effectively supporting overflow path identification, pollution source analysis, and control strategy formulation;

[0061] 4. Reduced reliance on monitoring equipment, improving the economic efficiency and scalability of model applications: This method, while relying on only a small number of monitoring points, achieves intelligent estimation of inflow distribution through spatial feature-driven and dynamic deduction, reducing reliance on high-cost and high-maintenance flow monitoring equipment. This significantly improves the model's applicability and scalability in areas with limited data.

[0062] 5. Achieve multi-scale and multi-path closure verification of the sewage system: The introduction of multi-source data breaks the traditional "isolated point calculation" method. It can implement flow conservation constraints at the system scale through upstream and downstream closure relationships, enhance the overall physical consistency of the model, and form a truly "evidence-based" modeling logic;

[0063] 6. Optimize dynamic calibration efficiency and improve modeling efficiency and versatility: Prioritize identifying nodes of interest and dynamically calibrating them locally to avoid ineffective global parameter adjustments, reduce manual intervention intensity and computational costs, effectively shorten model convergence time, improve parameter optimization accuracy, and enhance the method's portability and engineering practicality in large-scale, complex pipe network systems.

[0064] 7. Provide a high-reliability model foundation for smart drainage systems: The highly precisely calibrated model can be directly used as a core component of the intelligent operation and maintenance system, supporting a number of advanced applications such as drainage system operation and scheduling optimization, water pollution load prediction, and flood warning analysis, laying the foundation for the digital transformation of urban drainage systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 This is a flow chart of an embodiment of a multi-cutting point sewage distribution method considering the spatial heterogeneity of a sewage pipe network provided by the present invention;

[0066] Figure 2 It is a schematic diagram of the cutting point provided by the present invention;

[0067] Figure 3 This is a flow chart of another embodiment of the multi-cutting point sewage distribution method according to the present invention that takes into account the spatial heterogeneity of the sewage network;

[0068] Figure 4 It is a structural schematic diagram of the computer device provided by the present invention. DETAILED DESCRIPTION

[0069] The technical solutions in the embodiments of the present invention will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present invention.

[0070] Example 1

[0071] The embodiment of the present invention provides a multi-cutting point sewage distribution method considering the spatial heterogeneity of the sewage network. Figure 1 As shown, the following steps are included:

[0072] S101. Obtain sewage pipe network data and historical flow monitoring data in the target area, and clean the data.

[0073] Among them, the drainage network data is first obtained. The drainage network data includes the type, shape, length, drainage type, size, topological relationship of each drainage pipeline in the target area, as well as the type, depth, and ground elevation data of the node. Then, the drainage pipeline type of sewage pipeline is extracted to form the sewage network data; the historical monitoring data includes the flow time series data of the pump station, the time series data of the water inlet of the sewage treatment plant, and the flow time series data of each monitoring point.

[0074] Due to errors in manual editing or data integration, the acquired pipe network data may contain duplicate pipeline and node names, missing node names, and overlapping pipe network topology relationships. Therefore, it is necessary to first clean it using the corresponding algorithm. The steps for cleaning sewage pipe network data include:

[0075] 1) First, duplicate names need to be removed using a hash check + automatic numbering algorithm. This algorithm uses a Python dictionary structure to traverse and count node or pipeline names. If a name appears more than once, a number suffix (such as "_1" or "_2") is automatically appended to it to ensure global uniqueness.

[0076] 2) Secondly, the issue of missing node names is addressed using an algorithm developed using a naming logic strategy based on unique identification of spatial data. The algorithm identifies nodes with missing names by determining whether a valid name already exists for the node's latitude and longitude. For nodes that are missing names but have a valid spatial location, the algorithm automatically assigns them a unique name. Specifically, the algorithm adds a custom prefix, such as "Added," to these nodes, and combines this prefix with an auto-incrementing number to generate a unique name. This name is then assigned to the corresponding node, achieving automatic binding and consistency between spatial location and name.

[0077] 3) For topological overlap, the algorithm accurately compares the spatial geometric features of pipeline paths, systematically classifying overlaps into three categories and adopting corresponding processing strategies: complete overlap with identical pipeline directions, complete overlap with opposite pipeline directions, and partial overlap.

[0078] a) For completely overlapping pipelines with the same direction, Shapely's equals() method is used to identify them, performing geometric equality checks and identifying completely overlapping pipeline features through spatial relationship analysis. During this process, one of the pipelines is retained and the other redundant pipelines are automatically deleted.

[0079] b) To address the issue of complete overlap of pipelines running in opposite directions, Shapely combines geometric judgment with "coordinate order reversal comparison" logic to accurately identify pipeline features that overlap but run in opposite directions through reverse comparison of vertex coordinate order, and the algorithm then marks them.

[0080] c) For the problem of partially overlapping pipelines, use Shapely's intersects() method for preliminary identification and mark the partially overlapping pipelines.

[0081] For cases where pipelines are completely overlapped or partially overlapped in opposite directions, the algorithm will correct this type of data based on the rationality of the marked pipelines and the actual pipeline network layout.

[0082] Through the above cleaning process, high-quality pipeline network data with standardized structure, clear topology and unique naming can be finally obtained, providing a reliable data foundation.

[0083] Due to errors in the monitoring equipment or human recording operations in the flow monitoring data, outliers may appear in the monitoring data, thus affecting the subsequent calibration steps. Therefore, these outliers need to be cleaned. The following are three methods for determining outliers:

[0084] 1) The first judgment method is to set the abnormal judgment threshold range. According to the measurement range of the monitoring equipment, the physical limit value and the historical experience value, the minimum flow threshold L is set. min and the maximum threshold Lmax ,For observations outside this range, they are preliminarily judged as physically illegal outliers.

[0085] 2) The second method is statistical outlier identification, which uses the interquartile range method (IQR method) to perform statistical analysis on data distribution to identify outlier data points such as device jumps and observation errors. The specific steps are as follows:

[0086] Calculate the first quartile Q1 and the third quartile Q3 of the monitored value, and obtain IQR = Q3-Q1;

[0087] Set the abnormal threshold range: lower limit = Q1-1.5×IQR, upper limit = Q3+1.5×IQR;

[0088] Data that falls outside this interval are marked as statistical outliers.

[0089] 3) The third determination method is change rate anomaly identification (sliding change detection). Based on the temporal continuity characteristics of traffic, the sliding change rate method is used to detect sudden change anomalies. The details are as follows:

[0090] Calculate the monitoring data L of any two consecutive monitoring moments b and b-1 b 、L b-1 The rate of change ΔL b =|L b -L b-1 |;

[0091] If ΔL b If the mutation threshold Trate is exceeded, L b Or adjacent points are marked as jump outliers.

[0092] The threshold Trate can be set based on the average change of historical data or an empirical multiple (such as 3 times the median change).

[0093] After determining outliers using the three methods above, create an outlier flag variable for each data point. Mark outliers that meet any of the following conditions as outliers. Delete the marked outliers to obtain reliable and usable monitoring data:

[0094] A.L. b < L min or L b > L max ;

[0095] B.L b IQR abnormal threshold interval;

[0096] C, ΔL b >Trate.

[0097] S102. Based on the cleaned sewage pipe network data, a sewage pipe network topology map is established, and a water collection range map of each pumping station is generated.

[0098] This step specifically includes:

[0099] S1021. Based on the cleaned sewage pipe network data, a sewage pipe network topology map is established with sewage pipes as edges and sewage pipe end points as nodes. Specifically, a vector sewage pipe network topology map can be established on a GIS platform.

[0100] S1022. Based on the sewage pipe network topology map, identify the upstream nodes and sewage pipelines of each pumping station through a depth-first search algorithm, and generate a water collection range map of each pumping station based on the upstream nodes and sewage pipelines.

[0101] The depth-first search algorithm identifies the upstream water collection area of ​​a pump station. Starting from a specified pump station node, it recursively searches for all connected upstream nodes and sewage pipelines along the "reverse flow direction" of the pipeline network, and finally delineates its water collection area. It then generates a vector layer (shp) containing the names of all pipelines and nodes in the pump station water collection area as the pump station water collection range map, named xx_pump_upstream, where xx_pump is the name of the pump station.

[0102] The DFS algorithm specifically includes the following steps:

[0103] 1) Background setting: The sewage pipe network topology is abstracted into a directed graph G = (V, E), where V is a set of nodes (such as inspection wells, pumping stations); E = {e ij} is the edge set, edge e ij =(v i ,v j ) indicates that the water flows from node v i Flow to node v j , that is, v i v j Upstream of the given pump station node v pump ∈V, the goal is to construct its upstream subgraph G up ⊆G, G up Contains all the possible pump The nodes and edges of the pump station are the water collection range diagram;

[0104] 2) Initialization: For all nodes v∈V, assign visited(v) to False. visited(v) marks whether the node v has been visited. If so, assign it to True.

[0105] 3) Perform the following calculation, recursively starting from the downstream pump station node v pumpBacktrack to all reachable upstream nodes u:

[0106] ,

[0107] Where, DFS up (v pump ) represents DFS operation, Predecessor(v pump )={u∈V|(u, v pump )∈E} means all pointing upstream node;

[0108] 4) The recursion ends and the upstream region node set is obtained: V up ={v∈V|visited(v)=True}, get the upstream pipeline set: E up ={(u,v)∈E|u,v∈V up}.

[0109] Then, a pump station water collection range map is generated based on the upstream area node set and the upstream area pipeline set.

[0110] S103. Calculate an initial value of the average inflow rate of a single node in each administrative district based on the administrative districts included in the water collection range map of each pump station and the per capita daily sewage discharge and total population of each administrative district.

[0111] This step specifically includes:

[0112] S1031. Spatially superimpose the administrative district boundary map of the target area and the water collection range map of each pumping station, and calculate the area ratio within the water collection range of each administrative district.

[0113] Since each pumping station may cover multiple administrative districts, but may not cover all administrative districts, but may cover parts of multiple administrative districts, for each pumping station, the administrative district boundary map of the target area (for example, a town boundary map) and the pumping station water collection range map are imported into the GIS. After ensuring that the projections of the two layers are consistent, a spatial overlay analysis method (overlay operation Intersect) is used to extract the overlapping area of ​​each administrative district (for example, a town) and the pumping station water collection area. The following formula is then used to calculate the area share of each administrative district in the pumping station water collection area:

[0114] α i = (A i ∩BY) / A i ,

[0115] Where, α i A represents the proportion of administrative district i in the water collection area BY of the pump station, i represents the area of ​​administrative district i.

[0116] S1032. Spatially superimpose the administrative district boundary map, sewage pipe network topology map, and water collection range map of each pumping station in the target area, and count the number of sewage pipelines within the water collection range of the pumping station in each administrative district.

[0117] S1033. For each administrative district, multiply the total population of the administrative district by the corresponding area ratio, and then divide the result by the number of sewage pipelines within the water collection range of the pumping station in the administrative district to obtain the number of people served by a single sewage pipe in the administrative district.

[0118] The specific calculation formula for the number of people served by a single sewage pipe is:

[0119] p i =(P i* α i ) / n i ,

[0120] Among them, p i is the population served by a single sewer pipe in administrative district i, n i is the number of sewage pipelines within the water collection area of ​​the pumping station in administrative district i, P i is the total population of administrative district i, which can be obtained from documents published by administrative agencies.

[0121] S1034. For each administrative district, obtain the per capita daily sewage discharge of the administrative district, and multiply the per capita daily sewage discharge by the number of people served by a single sewage pipe to obtain an initial value of the average inflow of a single node.

[0122] The average daily discharge per capita of each administrative district can be obtained from published documents, such as the water resources bulletin of the administrative agency, or by dividing the total discharge by the total population. The initial value of the average inflow of a single node is calculated as follows:

[0123] q i = p i *o i ,

[0124] Where q i is the initial value of the average inflow of a single node in administrative area i, in L / day, i is the average daily sewage discharge per capita in administrative district i, measured in L / person / day. The number of people served by a single sewage pipe is the number of people served by that pipe's inflow node. Therefore, multiplying the average daily sewage discharge per capita by the number of people served by that pipe yields the initial average inflow per node.

[0125] S104. Calculate the theoretical total water intake of the target area based on the initial value of the average inflow of a single node, calculate the actual total inflow of the target area based on the historical flow monitoring data, and use the ratio of the actual total inflow to the theoretical total water intake as the first calibration coefficient to preliminarily calibrate the initial value of the average inflow of a single node.

[0126] Taking into account the various uncertainties in demographic data, including the mismatch between census results and the study period, the spatiotemporal mobility of the population, the inconsistency between the statistical area for per capita daily sewage discharge and the study area, and other inflows such as groundwater infiltration and illegal rainwater access that may be included in the monitoring data, this step performs a preliminary calibration of the initial value of the average inflow at a single node. This preliminary calibration can initially eliminate the impact of population errors, inaccuracies in per capita daily sewage discharge, and other inflow conditions on the accuracy of the results, achieving dynamic calibration of the sewage system across the entire region and improving the overall reliability and engineering applicability of the model.

[0127] This step specifically includes the following steps:

[0128] S1041. For each pumping station, the theoretical total water intake of the pumping station is calculated as the sum of the initial average inflow values ​​of individual nodes in all administrative districts within the pumping station's water intake range and the product of the number of sewage pipelines in the administrative district within the pumping station's water intake range:

[0129] ,

[0130] Where, It represents the theoretical total water intake of pump station k, in L / day. represents the number of administrative districts within the water collection range of pump station k.

[0131] S1042. Add the theoretical total water collection capacity of all pumping stations in the target area to obtain the theoretical total water collection capacity of the target area. :

[0132] ,

[0133] Where N represents the number of pumping stations in the target area.

[0134] S1043. Calculate the actual total inflow Q of the target area based on the historical flow monitoring data of all pumping stations in the target area. s,all .

[0135] Specifically, the actual total inflow Q of the target area can be obtained by accumulating the theoretical total inflow of all sewage treatment plants in the historical flow monitoring data of the target area. s,all .

[0136] S1044. Divide the actual total inflow of the target area by the theoretical total water intake to obtain a first calibration coefficient for the target area:

[0137] β=Q s,all / Q l,all ,

[0138] If β>1, it means that the initial sewage volume is underestimated and the upstream population needs to be increased proportionally.

[0139] If β<1, it means that the initial sewage volume estimate is too high and the upstream population needs to be reduced proportionally.

[0140] S1045. Multiply the initial value of the average inflow of a single node in each administrative area by the first calibration coefficient β to obtain the average inflow of a single node after preliminary calibration:

[0141] q i,c =q i ×β,

[0142] Where q i,c represents the average inflow of a single node after preliminary calibration in administrative area i.

[0143] In this step, if there are population errors, inaccurate per capita daily sewage discharge, or other inflow conditions, there will be an error between the actual total inflow and the theoretical total water intake. Therefore, the ratio of the actual total inflow to the theoretical total water intake can be used to calibrate the average inflow of each node, preliminarily eliminating the impact of population errors, inaccurate per capita daily sewage discharge, or other inflow conditions. This allows for dynamic and accurate calibration of the entire sewage system.

[0144] S105: Identify all the division points and the control domain of each division point in the sewage pipe network topology diagram, as well as the independent control domain of each node of interest.

[0145] This step specifically includes:

[0146] S1051. Identify all the dividing points in the sewage pipe network topology diagram.

[0147] Among them, the nodes of interest are hydraulic structures with monitoring data, such as pumping stations, sewage treatment plants and monitoring points, which are nodes with flow monitoring functions. The split point is the intersection of the common water collection path of at least two nodes of interest, and there are at least two independent non-overlapping branch paths in the downstream path, and each branch path reaches a different node of interest. These split points are located in the intersection section of the common water collection path of the nodes of interest, and have common contributions to multiple nodes of interest on the hydraulic path. Therefore, their impact can be observed by multiple downstream nodes of interest, and they are key points for model verification, pollution tracing, scheduling analysis and other tasks. Figure 2As shown, pumping station A and pumping station B are nodes of interest, and the red nodes are the dividing points between pumping station A and pumping station B.

[0148] S1052: Identify the control domain of each segmentation point.

[0149] The control domain of a split point is the area between the current split point and its upstream adjacent split point. The control domain of the most upstream split point is the entire area upstream of that split point. Dividing the control domain can help create smaller calibration units when calibrating sewage volume, thereby making the calibration of sewage volume distribution more accurate. The control domain is a surface unit spatially enclosed by several sewage pipelines and nodes. In the GIS system, it is represented as a vector surface object, containing the spatial region derived from the node set (Domain_Nodes) and the pipeline set (Domain_Edges), which serves as the basic unit for sewage inflow estimation and calibration.

[0150] S1053: Identify the independent control domain of each concerned node.

[0151] The independent control domain of the concerned node is the area between the concerned node and the adjacent upstream split point.

[0152] S106 : For each concerned node, the independent control domain of the concerned node and the control domain of each upstream split point are respectively used as a flow monitoring related control domain of the concerned node.

[0153] This step specifically includes:

[0154] S1061. For each node of interest, starting from the current node of interest, use a depth-first search method to traverse upstream in reverse order to obtain a set U of all upstream nodes that can reach the current node of interest.

[0155] S1062. Obtain all the segmentation points to form a segmentation point set C;

[0156] S1063. Calculate the intersection of the split point set C and the upstream node set U of the current concerned node as the upstream split point set C′ of the current concerned node.

[0157] S1064: The control domain of each split point in the upstream split point set C′ of the current concerned node and the independent control domain of the current concerned node are used as the flow monitoring related control domain of the current concerned node.

[0158] For example, assume the set of split points C = {C1, C2, C3, C4}; the corresponding set of control domains is {DQ1, DQ2, DQ3, DQ4}; the upstream split point set of a node of interest A2 is C′2 = {C3, C4}; then the upstream split point control domain set is U′ = {DQ3, DQ4}. Assuming A2's independent control domain is dQ2, the control domains related to A2's flow monitoring are DQ3, DQ4, and dQ2, respectively.

[0159] S107 , with the goal that the difference between the theoretical total inflow and the monitored inflow of each node of interest in the target area is within a preset threshold range, calculate a secondary calibration coefficient, thereby secondary calibrating the average inflow of a single node after the preliminary calibration.

[0160] This step specifically includes:

[0161] S1071. For each flow monitoring-related control domain, the average inflow of each inflow node after preliminary calibration is summarized as the initial theoretical inflow:

[0162] ,

[0163] Where, represents the initial theoretical inflow of the control domain y related to flow monitoring, is the average inflow rate of a single node after preliminary calibration of inflow node j, Indicates the inflow node belonging to the flow monitoring related control domain y.

[0164] S1072. Establish a calibration coefficient optimization model. The objective function of the calibration coefficient optimization model is to minimize the difference between the theoretical total inflow and the monitored flow for each node of interest. The optimization parameter is the second calibration coefficient of each flow monitoring-related control domain. The theoretical total inflow is the sum of the calibrated theoretical inflows of all flow monitoring-related control domains corresponding to the node of interest. The calibrated theoretical inflow of a flow monitoring-related control domain is the product of its initial theoretical inflow and the secondary calibration coefficient. The calibration coefficient optimization model is specifically as follows:

[0165] ,

[0166] ,

[0167] ,

[0168] ,

[0169] ,

[0170] st ,

[0171] Where, represents the second calibration coefficient vector, represents the second calibration coefficient of the flow monitoring related control domain y, where Y represents the number of all flow monitoring related control domains in the target area. represents the monitoring traffic matrix of the concerned node, represents the monitoring traffic of the concerned node x, X represents the number of concerned nodes in the target area, Represents the initial estimated inflow matrix, specifically an X-row Y-column matrix, express The xth row and yth column element of Indicates whether the traffic monitoring related control domain y belongs to The indicator variable, represents the initial estimated inflow of the control domain y related to flow monitoring, Represents the set of all traffic monitoring related control domains concerned with node x.

[0172] S1073: Solve the calibration coefficient optimization model to obtain the second calibration coefficient of each flow monitoring-related control domain.

[0173] S1074. Multiply the average inflow of each single node after the initial calibration by the second calibration coefficient of the control domain related to the flow monitoring to obtain the average inflow of the single node after the secondary calibration:

[0174] ,

[0175] Where, is the average inflow of a single node after secondary calibration of node j in the flow monitoring related control domain y.

[0176] This step ensures that each adjustment is based on monitoring data, has a basis and does not cause feedback errors, which helps to improve the overall stability and reliability of subsequent models.

[0177] In other embodiments, step S108 may be added after step S107, such as Figure 3 As shown, step S108 is specifically as follows: weighting the average inflow of a single node after secondary calibration according to the sewage discharge pattern curve of the population land type to which the control domain belongs, and obtaining the sewage discharge time series data of a single node in each time period throughout the day.

[0178] Among them, the sewage discharge pattern curve a(t) is a set of time distribution coefficients that describe the sewage discharge variation pattern in each time period throughout the day, and the sum of the time distribution coefficients is 1, which is used to characterize the discharge pattern of the current population land type in different time periods throughout the day. a(t) corresponds to different values ​​in different time periods t of each day. For example, each hour is a time period, and the cumulative sum of the 24-hour values ​​is 1. Therefore, the sewage discharge pattern curve a(t) is multiplied by the average inflow of a single node to obtain the sewage discharge of a single node in different time periods t of each day. The sewage discharge pattern curve a(t) corresponds to the population land type. Different population land types correspond to different sewage discharge pattern curve parameters. a(t) can be obtained through experience or by fitting the emission data of different population land types.

[0179] In this step, when identifying the population land type corresponding to the control domain, the drainage change trend reflected in the acquired monitoring data and the land use planning map of the area (including the division of population land types such as residential land, commercial land, and industrial land) can be combined to estimate the possible main population land type in the area. The sewage discharge pattern curve of the population land type can be used to preliminarily obtain the sewage discharge time series data of the area, and further refine the time series distribution.

[0180] Example 2

[0181] Figure 4 This is a schematic diagram of the structure of a computer device provided by an embodiment of the present invention, which provides services for implementing the method of the first embodiment of the present invention. Figure 4 As shown, the device may include: a memory 201 storing a computer executable program; a processor 202 coupled to the memory 201; the processor 202 calls the computer executable program stored in the memory 201 to execute the steps in the method described in embodiment 1.

[0182] Memory 201 may include computer-readable media in the form of volatile memory, such as random access memory (RAM) and / or cache memory. The device may further include other removable / non-removable, volatile / non-volatile computer-system storage media. By way of example only, memory 201 may be used to read and write to non-removable, non-volatile magnetic media (commonly referred to as a "hard drive"). A program / utility having a set (at least one) of program modules may be stored, for example, in memory 201. Such program modules include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each of these examples, or some combination thereof, may include an implementation of a network environment. The computer-executable programs of the program modules generally perform the functions and / or methods described in the embodiments of the present invention.

[0183] The processor 202 executes various functional applications and data processing by running the programs stored in the memory 201, such as implementing the method provided in the first embodiment of the present invention.

[0184] The code of the computer executable program can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages.

[0185] It should be understood that the above embodiments and descriptions only describe the principles, main features and advantages of the present invention. Without departing from the spirit and scope of the present invention, the present invention may be subject to various changes and improvements, and these changes and improvements all fall within the scope of protection of the present invention.

Claims

1. A multi-cutting point sewage distribution method considering the spatial heterogeneity of sewage pipe network, characterized by: The steps include: (1) Obtain sewage pipe network data and historical flow monitoring data in the target area and clean the data; (2) Based on the cleaned sewage pipe network data, establish a sewage pipe network topology map and generate a water collection range map for each pumping station; (3) Based on the administrative districts included in the water collection range map of each pumping station, combined with the per capita daily sewage discharge and the total population of each administrative district, calculate the initial value of the average inflow flow of each single node in each administrative district; (4) Calculate the theoretical total water intake of the target area based on the initial value of the average inflow of a single node, calculate the actual total inflow of the target area based on the historical flow monitoring data, and use the ratio of the actual total inflow to the theoretical total water intake as the first calibration coefficient to perform a preliminary calibration on the initial value of the average inflow of a single node; (5) Identify all the cut-off points and the control domain of each cut-off point in the sewage network topology diagram, as well as the independent control domain of each node of interest, wherein the cut-off point is the intersection of the common water collection path of at least two nodes of interest, and there are at least two independent and non-overlapping branch paths in the downstream path, each branch path reaches a different node of interest, and the node of interest is a node with flow monitoring function; (6) For each concerned node, the independent control domain of the concerned node and the control domain of each upstream split point are respectively used as a flow monitoring related control domain of the concerned node; (7) With the goal that the difference between the theoretical total inflow and the monitored flow of each node of interest in the target area is within the preset threshold range, calculate the secondary calibration coefficient, thereby recalibrating the average inflow of a single node after the initial calibration, where the theoretical total inflow is the sum of the theoretical inflows of all flow monitoring-related control domains corresponding to the node of interest after calibration using the secondary calibration coefficient; (8) The average inflow of a single node after secondary calibration is weighted according to the sewage discharge pattern curve of the population land use type of the control area, and the sewage discharge time series data of a single node in each time period throughout the day are obtained; wherein the sewage discharge pattern curve is a set of time distribution coefficients that describe the change law of sewage discharge in each time period throughout the day, and the sum of the time distribution coefficients is 1; Step (3) specifically includes the following steps: (3.1) Spatially overlay the administrative district boundary map of the target area with the water collection range map of each pumping station, and calculate the proportion of area within each administrative district that falls within the water collection range of the pumping station; (3.2) Spatially overlay the administrative district boundary map, sewage network topology map, and pump station water collection area map of the target area, and count the number of sewage pipelines within the pump station water collection area in each administrative district; (3.3) For each administrative district, multiply the total population of the administrative district by the corresponding area percentage and divide the result by the number of sewer lines within the pumping station in the administrative district to obtain the number of people served by each sewer pipe in the administrative district; (3.4) For each administrative district, obtain the average daily sewage discharge per capita in that district and multiply it by the number of people served by a single sewer pipe to obtain the initial value of the average inflow per node. Step (7) specifically includes the following steps: (7.1) For each flow monitoring-related control domain, the average inflow of each node after preliminary calibration of all its inflow nodes is summarized as the initial theoretical inflow; (7.2) Establish a calibration coefficient optimization model, where the objective function of the calibration coefficient optimization model is to minimize the difference between the theoretical total inflow and the monitored inflow for each node of interest. The optimization parameter is the second calibration coefficient of each flow monitoring-related control domain. The theoretical total inflow is the sum of the calibrated theoretical inflows of all flow monitoring-related control domains corresponding to the node of interest, and the calibrated theoretical inflow of a flow monitoring-related control domain is the product of its initial theoretical inflow and the corresponding secondary calibration coefficient. (7.3) Solving the calibration coefficient optimization model to obtain the second calibration coefficient of each flow monitoring related control domain; (7.4) Multiply the average inflow of each single node after preliminary calibration by the second calibration coefficient of the relevant control domain of the flow monitoring to which it belongs to obtain the average inflow of the single node after secondary calibration.

2. A multi-cutting point sewage distribution method considering the spatial heterogeneity of sewage pipe network according to claim 1, characterized in that: Step (2) specifically includes the following steps: (2.1) Based on the cleaned sewage pipe network data, establish a sewage pipe network topology map with sewage pipe lines as edges and sewage pipe end points as nodes; (2.2) Based on the sewage network topology, the upstream nodes and sewage pipelines of each pumping station are identified through a depth-first search algorithm, and a water collection range map of each pumping station is generated based on the upstream nodes and sewage pipelines.

3. The multi-cutting point sewage distribution method considering the spatial heterogeneity of the sewage pipe network according to claim 1 is characterized in that: Step (4) specifically includes the following steps: (4.1) For each pumping station, the theoretical total water intake of the pumping station is the sum of the initial average inflow of individual nodes in all administrative districts within the pumping station's water intake range and the product of the number of sewage pipelines in the administrative district within the pumping station's water intake range; (4.2) Add the theoretical total water collection capacity of all pumping stations in the target area to obtain the theoretical total water collection capacity of the target area; (4.3) Calculate the actual total inflow to the target area based on the historical flow monitoring data of all pumping stations in the target area; (4.4) Divide the actual total inflow of the target area by the theoretical total water intake to obtain the first calibration coefficient of the target area; (4.5) Multiply the initial value of the average inflow of a single node in each administrative area by the first calibration coefficient to obtain the average inflow of a single node after preliminary calibration.

4. The multi-cutting point sewage distribution method considering the spatial heterogeneity of the sewage network according to claim 1 is characterized in that: Step (5) specifically includes the following steps: (5.1) Identify all split points in the sewage network topology, where the split point is the intersection of at least two common catchment paths of the nodes of interest, and the downstream path has at least two independent, non-overlapping branch paths, each branch path reaching a different node of interest; (5.2) Identify the control domain of each split point, where the control domain of a split point is the area between the current split point and its upstream adjacent split point, and the control domain of the upstreammost split point is the entire area upstream of the split point; (5.3) Identify the independent control domain of each concerned node, wherein the independent control domain of the concerned node is the area between the concerned node and the adjacent upstream split point.

5. The multi-cutting point sewage distribution method considering the spatial heterogeneity of the sewage network according to claim 1 is characterized in that: Step (6) specifically includes: (6.1) For each node of interest, starting from the current node of interest, use the depth-first search method to traverse upstream in reverse order to obtain the set U of all upstream nodes that can reach the current node of interest; (6.2) Obtain all the split points to form a split point set C; (6.3) Calculate the intersection of the split point set C and the upstream node set U of the current concerned node as the upstream split point set C′ of the current concerned node; (6.4) The control domain of each split point in the upstream split point set C′ of the current concerned node, as well as the independent control domain of the current concerned node, are used as the flow monitoring related control domain of the current concerned node.

6. The multi-cutting point sewage distribution method considering the spatial heterogeneity of the sewage network according to claim 1 is characterized in that: The calibration coefficient optimization model described in step (7.2) is specifically: ， ， ， ， ， s.t. , Where, represents the second calibration coefficient vector, represents the second calibration coefficient of the flow monitoring related control domain y, Y represents the number of all flow monitoring related control domains in the target area, represents the monitoring traffic matrix of the concerned node, represents the monitoring traffic of the concerned node x, X represents the number of concerned nodes in the target area, Represents the initial estimated inflow matrix, specifically an X-row Y-column matrix, express The xth row and yth column element of Indicates whether the traffic monitoring related control domain y belongs to The indicator variable, represents the initial estimated inflow of the control domain y related to flow monitoring, Represents the set of all traffic monitoring related control domains concerned with node x.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: The processor executes the computer program to implement the method according to any one of claims 1 to 6.

Images