Urban road intelligent drainage scheduling system and method based on Internet of Things

Abstract

The invention discloses an urban road intelligent drainage scheduling system and method based on the Internet of Things. The method comprises the following steps that an urban road drainage Internet of Things sensing network is constructed, and a multi-source heterogeneous data set corresponding to an urban road target area is collected; performing space-time alignment and noise filtration processing on the multi-source heterogeneous data set, and meanwhile, performing ponding increase rate and pipe network pressure bearing evaluation to obtain urban ponding increase rates and drainage pipe network pressure bearing degrees corresponding to different urban road subnetworks; calculating drainage pipe network negative pressure indexes corresponding to the different urban road section sub-networks according to the urban water accumulation increasing rates and the drainage pipe network pressure bearing degrees corresponding to the different urban road section sub-networks; and acquiring an urban road network topological structure, performing drainage hierarchical scheduling processing, issuing the urban road network topological structure to drainage facilities in corresponding areas to execute drainage scheduling and feed back scheduled drainage effect data, and re-executing closed-loop scheduling if the drainage effect data is not accordant with the drainage effect data. According to the invention, accurate drainage scheduling of the urban road area can be realized.

Description

Technical Field

[0001] This invention relates to the field of drainage scheduling technology, and more specifically, to an intelligent urban road drainage scheduling system and method based on the Internet of Things. Background Technology

[0002] Internet of Things-based smart drainage technology, with its unique advantages such as multi-node real-time sensing, data interconnection, and remote precise control, is gradually replacing the traditional drainage mode that relies on manual inspection and fixed threshold control. It has become an important technical means to alleviate urban flooding, optimize water resource recycling, and improve the level of intelligent urban water management.

[0003] Currently, by deploying level sensors, flow sensors, and water quality monitoring equipment at road flooding points and key nodes of drainage networks, and using low-power wide-area communication technologies such as LoRa and NB-IoT to transmit the collected monitoring data to a cloud management platform, and combining this with preset drainage scheduling rules to remotely control pump start-up and shutdown, gate opening, etc., the existing system has insufficient ability to fuse and analyze multi-source data. It is easily affected by factors such as sensor signal interference under extreme weather conditions and data synchronization delays of monitoring equipment in different areas, which can lead to errors in the prediction of flooding risks. At the same time, traditional scheduling strategies are mostly based on local data from a single area, making it difficult to achieve coordinated scheduling between different network nodes and failing to fully utilize the maximum efficiency of existing drainage facilities, thereby reducing the drainage efficiency of the network. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an intelligent urban road drainage scheduling method based on the Internet of Things, comprising the following steps: Step S1: Construct an urban road drainage IoT sensing network within the target area of ​​urban roads, and collect multi-source heterogeneous datasets containing real-time rainfall data, road water accumulation data, drainage network operation data, and water pressure data within the target area of ​​urban roads based on the urban road drainage IoT sensing network; perform spatiotemporal alignment and noise filtering on the multi-source heterogeneous datasets to generate a standardized urban drainage monitoring dataset. Step S2: Based on the target area of ​​urban roads, divide the rainfall data, water accumulation data, drainage operation data and pipe pressure data corresponding to different urban road sub-networks from the urban standardized drainage monitoring dataset. Based on the rainfall data, water accumulation data, drainage operation data and pipe pressure data corresponding to different urban road sub-networks, conduct water accumulation increase rate and pipe network pressure assessment to obtain the urban water accumulation increase rate and drainage pipe network pressure degree corresponding to different urban road sub-networks. Step S3: Calculate the negative pressure index of the drainage pipe network corresponding to different urban road segment subnets based on the urban waterlogging increase rate and drainage pipe network pressure bearing capacity of different urban road segment subnets; obtain the urban road network topology, and generate an urban road area drainage pressure distribution map based on the drainage pipe network negative pressure index corresponding to different urban road segment subnets and the urban road network topology. Step S4: Based on the urban road area drainage pressure distribution map, obtain the real-time drainage pressure difference between different urban road subnets and perform drainage hierarchical scheduling to generate an urban road drainage hierarchical scheduling strategy; distribute the urban road drainage hierarchical scheduling strategy to the drainage facilities in the corresponding area to perform drainage scheduling and provide feedback on the drainage effect data after scheduling; if the drainage effect data does not meet the requirements, return to step S3 to perform closed-loop scheduling.

[0005] Furthermore, the present invention also provides an Internet of Things-based intelligent drainage scheduling system for urban roads, including a processor, a memory, and a computer program stored in the memory and executable on the processor, for performing the Internet of Things-based intelligent drainage scheduling method for urban roads as described above.

[0006] The beneficial effects of this application are as follows: Firstly, the construction of an IoT sensing network enables comprehensive collection of multi-dimensional data, including real-time rainfall, road flooding, drainage network operation, and pipe water pressure. This breaks through the limitations of traditional monitoring, which relies on a single data source and has limited coverage. It ensures the capture of the overall operational status of the urban road drainage system. Secondly, the spatiotemporal alignment processing of multi-source heterogeneous datasets effectively solves the problem of data synchronization delays between monitoring devices in different areas. By using unified timestamps and spatial coordinates, data collected from different road sections and devices are accurately matched, avoiding analytical biases caused by misaligned data time or location. Thirdly, noise filtering eliminates abnormal data generated by sensor signal interference under extreme weather conditions, ensuring data authenticity and completely changing the situation where poor data quality leads to inaccurate flood risk predictions in existing systems, thus improving the accuracy and reliability of drainage system analysis. Finally, by dividing the data into urban road segment subnets, the overall drainage system is decomposed into multiple independently analyzable local units. This avoids the problem of local risks being masked by the "one-size-fits-all" approach in traditional analysis, allowing for precise focus on the drainage characteristics of each road segment (such as large-diameter main road drainage networks and high load on residential area networks). By assessing the rate of water accumulation, the real-time trend of water accumulation in each sub-network can be monitored, while the network pressure assessment can reveal the pipeline's operational load. Combining these two methods creates a comprehensive assessment of the drainage risk of each sub-network. This step dynamically captures the development trend of risks and the network's carrying capacity, providing accurate regional risk data for the subsequent generation of drainage pressure distribution maps. Then, the drainage network negative pressure index integrates the rate of water accumulation and the network pressure level. Through quantitative indicators, it intuitively reflects the drainage pressure of each sub-network. By combining this with the urban road network topology to generate a pressure distribution map, the negative pressure index of each sub-network is linked to the actual road network location and pipeline connections, clearly presenting the pressure correlation between different road segment sub-networks. This breaks through the limitations of traditional local data analysis. This visualized pressure distribution map allows managers to quickly identify high-pressure core areas and pressure transmission paths, clarifying the overall drainage risk pattern. This step achieves a leap from local risk identification to global pressure control, ensuring that scheduling strategies can consider both local and overall interests. Finally, by analyzing the real-time drainage pressure differences between different subnets for hierarchical scheduling, the limitations of traditional single-area scheduling are broken, avoiding resource waste caused by overloading some facilities and leaving others idle. After the scheduling strategy is issued, closed-loop adjustments are made based on the feedback drainage effect data. If the water accumulation in a certain area is still not relieved after scheduling, the pressure distribution is reassessed and the strategy is optimized to ensure that the scheduling measures can dynamically adapt to the actual drainage situation. This hierarchical collaborative + closed-loop optimization scheduling mode can maximize the overall effectiveness of existing drainage facilities and significantly improve the emergency response capability and operational efficiency of the urban drainage system. Attached Figure Description

[0007] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart of the IoT-based intelligent drainage scheduling method for urban roads in this embodiment; Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S1. Detailed Implementation

[0008] The following drawings disclose several embodiments of the present invention. For clarity, many practical details will be described in the following description. However, it should be understood that these practical details are not intended to limit the invention. That is, in some embodiments of the invention, these practical details are not essential. Furthermore, for the sake of simplicity, some conventional structures and components will be shown in the drawings in a simple schematic manner.

[0009] To further understand the invention's content, features, and effects, the following embodiments are provided, and detailed descriptions are given below in conjunction with the accompanying drawings: Reference Figure 1 , Figure 1 This is a flowchart of the IoT-based intelligent drainage scheduling method for urban roads in this embodiment. The IoT-based intelligent drainage scheduling method for urban roads in this embodiment includes the following steps: Step S1: Construct an urban road drainage IoT sensing network within the target area of ​​urban roads, and collect multi-source heterogeneous datasets containing real-time rainfall data, road water accumulation data, drainage network operation data, and water pressure data within the target area of ​​urban roads based on the urban road drainage IoT sensing network; perform spatiotemporal alignment and noise filtering on the multi-source heterogeneous datasets to generate a standardized urban drainage monitoring dataset. In this embodiment of the invention, an urban road drainage Internet of Things sensing network is constructed in the target area (5 square kilometers, including 3 road segment subnets). The LoRaWAN communication protocol (transmission rate 0.3-50kbps, communication distance 1-3 kilometers) is adopted, and 12 sensing device points are deployed (6 road sections prone to water accumulation, 2 pipeline nodes, and 4 low-lying areas). Each location is equipped with multi-dimensional sensing devices: a rainfall intensity sensor (measurement range 0-200mm / h, sampling frequency 5 minutes / time) to collect real-time rainfall data (e.g., 15.2mm / h rainfall on Binhe Road at 08:00); a water level sensor (measurement range 0-100cm, sampling frequency 1 minute / time) to collect road water accumulation data (e.g., 3.2cm water depth on Xincheng Road at 08:02); a flow meter (measurement range 0-500m³ / h, sampling frequency 2 minutes / time) to collect drainage network operation data (e.g., 120m³ / h flow rate at node J35 at 08:05); and a pressure sensor (measurement range 0-1.6MPa, sampling frequency 2 minutes / time) to collect water pressure data in the pipeline (e.g., 0.32MPa water pressure at node J35 at 08:05), generating a multi-source heterogeneous dataset. A UTM spatiotemporal coordinate system is established based on GPS positioning information (accuracy ≤ 10 meters) and timestamps (accurate to the second) from sensor devices, mapping the data to a unified spatiotemporal dimension. An adaptive wavelet threshold denoising algorithm (wavelet basis db4, decomposition level 3) is used to filter fluctuation noise (e.g., ±0.05MPa interference in water pressure data of 0.32MPa is corrected to 0.32MPa), and the data format is unified (numerical values ​​are retained to 1 decimal place, timestamp "yyyy-MM-ddHH:mm:ss") to generate a standardized urban drainage monitoring dataset.

[0010] Step S2: Based on the target area of ​​urban roads, divide the rainfall data, water accumulation data, drainage operation data and pipe pressure data corresponding to different urban road sub-networks from the urban standardized drainage monitoring dataset. Based on the rainfall data, water accumulation data, drainage operation data and pipe pressure data corresponding to different urban road sub-networks, conduct water accumulation increase rate and pipe network pressure assessment to obtain the urban water accumulation increase rate and drainage pipe network pressure degree corresponding to different urban road sub-networks. In this embodiment of the invention, data is divided from the urban standardized drainage monitoring dataset into three sub-networks based on the target area (sub-network 1: Binhe Road, sub-network 2: Xincheng Road, sub-network 3: Keji Road-Wenhua Road): Sub-network 1 acquires rainfall data (J1-J3 sensors), water accumulation data (J1-J3 sensors), drainage operation data (Binhe Road pumping station), and pipe water pressure data (J31-J35 nodes); Sub-network 2 acquires data from the corresponding Xincheng Road sensors and pumping stations; Sub-network 3 acquires data from the corresponding Keji Road-Wenhua Road sensors and pumping stations. The data missing rate of each sub-network is calculated (preset threshold 5%). The water accumulation data missing rate of sub-network 2 is 5.21%, which is filled in using the spatial difference method between adjacent sub-networks (e.g., the 08:10 data is filled in as 3.14cm). Based on the completed data, the calculations for subnet 1 are as follows: Urban waterlogging increase rate = (rainfall infiltration 108 m³ - pipe network discharge 67.32 m³) / waterlogged area 4500 m² ≈ 0.90 cm. The pipe network pressure resistance is calculated using the formula: "Overload frequency 20% × 0.4 + Fluctuation amplitude 15% × 0.3 + Lifespan decay 10% × 0.3 = 15.5". Subnet 2 has a waterlogging increase rate of 1.16 cm and a pressure resistance of 12.13. Subnet 3 has a waterlogging increase rate of 0.40 cm and a pressure resistance of 8.4. All calculations are based on standardized data, and the results match the subnet risk levels without deviation.

[0011] Step S3: Calculate the negative pressure index of the drainage pipe network corresponding to different urban road segment subnets based on the urban waterlogging increase rate and drainage pipe network pressure bearing capacity of different urban road segment subnets; obtain the urban road network topology, and generate an urban road area drainage pressure distribution map based on the drainage pipe network negative pressure index corresponding to different urban road segment subnets and the urban road network topology. In this embodiment of the invention, the index of each sub-network is calculated by using the weighted formula "Drainage network negative pressure index = contribution of water accumulation rate risk × 0.4 + contribution of network pressure risk × 0.6": Sub-network 1 has a water accumulation rate risk contribution of 0.0315 and a pressure risk contribution of 0.225, with a negative pressure index ≈ 0.1476; Sub-network 2 has contributions of 0.0481 and 0.1913, with an index ≈ 0.1340; Sub-network 3 has contributions of 0.0140 and 0.16, with an index ≈ 0.1016. The road network topology was obtained from the city's geographic information platform (subnets 1 and 2 intersect at N5-N6, and subnets 2 and 3 intersect at N9-N10). An ArcGIS rendering tool was used to generate a drainage pressure distribution map: color grading was set (0.10-0.12 green, 0.12-0.14 yellow, 0.14-0.16 red). Subnet 1 (0.1476) was rendered in red, subnet 2 (0.1340) in yellow, and subnet 3 (0.1016) in green. Subnet boundaries, manholes, and pump station locations were marked. Red areas were marked "high pressure, it is recommended to increase pump station power". The map intuitively reflects the pressure differences in each area and there are no rendering errors.

[0012] Step S4: Based on the urban road area drainage pressure distribution map, obtain the real-time drainage pressure difference between different urban road subnets and perform drainage hierarchical scheduling to generate an urban road drainage hierarchical scheduling strategy; distribute the urban road drainage hierarchical scheduling strategy to the drainage facilities in the corresponding area to perform drainage scheduling and provide feedback on the drainage effect data after scheduling; if the drainage effect data does not meet the requirements, return to step S3 to perform closed-loop scheduling.

[0013] In this embodiment of the invention, the real-time pressure difference between adjacent subnets is calculated based on the pressure distribution map: the difference between subnets 1 and 2 is 0.0136, and the difference between subnets 2 and 3 is 0.0324. A preset first-level threshold of 0.03 and a safety threshold of 0.015 are used. The scheduling level is determined: subnets 1 and 2 are subject to second-level mitigation scheduling (adjusting the pumping station frequency of subnet 1 to 45Hz and the gate opening / closing degree to 70%); subnets 2 and 3 are subject to first-level emergency scheduling (activating emergency pumping stations, fully opening gates, and shutting down non-essential water use). The strategy is distributed via a LoRaWAN encrypted link (AES-128 algorithm, latency ≤30 seconds). Data feedback after 10 minutes shows that the pressure difference between subnets 2 and 3 has decreased to 0.028 (still higher than the threshold), which does not meet the requirements. Returning to step S3, a pressure distribution map is generated based on the new data (subnet 2 index 0.128, subnet 3 0.105, difference 0.023). The strategy is adjusted (emergence power of subnet 2 emergency pump station is increased to 25kW). After execution again, the pressure difference drops to 0.022 (meets requirements), and closed-loop scheduling is stopped. The entire process is characterized by stable communication, accurate strategy adjustment, and no execution deviation.

[0014] Furthermore, refer to Figure 2 , Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S1 in this embodiment. Step S1 in this embodiment includes the following steps: Step S11: Obtain the corresponding urban road planning layout and drainage pipe network distribution through the target urban road area, and plan the deployment points of sensing devices in the corresponding waterlogged road sections, drainage pipe network nodes and low-lying areas within the target urban road area based on the urban road planning layout and drainage pipe network distribution, so as to form a drainage Internet of Things sensing network topology covering the target area. In this embodiment of the invention, the urban road planning layout (road width 15-40 meters, number of lanes 2-6) and drainage pipe network distribution (pipe diameter 300-1200 mm, pipe network nodes 80, average spacing 50 meters) of the target area (area 5 square kilometers, including 10 main roads and 20 secondary roads) are obtained through the urban geographic information system. Hydraulic simulation software was used to analyze road sections prone to flooding: When the simulated rainfall was 50 mm / h, the water depth of the main roads "Binhe Road (K0+200-K0+500 section)" and "Xincheng Road (K1+100-K1+300 section)" exceeded 15 cm, and were identified as road sections prone to flooding; among the drainage pipe network nodes, "Node J35 (intersection of Binhe Road and Wenhua Road)" and "Node J62 (intersection of Xincheng Road and Keji Road)" had a flow load exceeding 80%, and were identified as critical nodes; the low-lying areas were "low-lying area of ​​Keji Road (elevation 28.5-29.0 meters)" and "low-lying area of ​​Wenhua Road (elevation 29.2-29.5 meters)". The planned deployment locations of the sensing devices are as follows: one point is set every 100 meters in sections prone to water accumulation (6 points in total), one point is set at each key pipeline node (2 points in total), and one point is set every 5,000 square meters in low-lying areas (4 points in total), for a total of 12 points, forming a drainage IoT sensing network topology covering the target area. The topology adopts a star structure, with the central node being the regional data aggregation gateway, ensuring that the signal transmission distance of each point is ≤1,000 meters and there are no signal blind spots.

[0015] Step S12: Construct a corresponding urban road drainage IoT sensing network based on the topology of the drainage IoT sensing network, and deploy corresponding multi-dimensional sensing devices at the corresponding sensing device deployment points through the urban road drainage IoT sensing network to detect data in real time, including rainfall intensity sensors to collect real-time rainfall data, water level sensors to collect road water accumulation data, including the water accumulation depth and water accumulation range of each road, flow meters to collect drainage pipe network operation data, including the flow rate and flow velocity in the drainage pipe network, and pressure sensors to collect water pressure data in the pipes; In this embodiment of the invention, an urban road drainage IoT sensing network is constructed based on a sensing network topology and using the LoRaWAN wireless communication protocol (transmission rate 0.3-50kbps, communication distance 1-3 km, power consumption as low as 1μA). The network includes one regional data aggregation gateway (deployed on the roof of a municipal building in the center of the target area) and 12 sensor terminal devices. Multi-dimensional sensing devices are deployed at the 12 deployment points: rainfall intensity sensors (measurement range 0-200mm / h, accuracy ±0.1mm / h, sampling frequency 5 minutes / time) and water level sensors (measurement range 0-100cm, accuracy ±0.5cm, sampling frequency 1 minute / time) are deployed at points prone to water accumulation. Real-time rainfall data and road water accumulation data are collected (water depth accurate to 0.1cm, water accumulation range estimated by the distance between points, e.g., the water accumulation range of the Binhe Road section K0+200-K0+300 is 10...). The system is 0m x 15m; flow meters (measurement range 0-500m³ / h, accuracy ±2%, sampling frequency 2 minutes / time) and pressure sensors (measurement range 0-1.6MPa, accuracy ±0.01MPa, sampling frequency 2 minutes / time) are deployed at pipeline node points to collect drainage network operation data (flow rate accurate to 1m³ / h, flow velocity calculated from pipe diameter and flow rate, e.g., a flow rate of 180m³ / h corresponds to a flow velocity of 1.77m / s for a pipe diameter of 600mm) and water pressure data within the pipeline (accurate to 0.01MPa). All sensing devices have IP68 anti-corrosion and waterproof ratings, adapting to humid road environments. The equipment is powered by a combination of solar energy and lithium batteries (endurance ≥72 hours without sunlight), ensuring 24-hour uninterrupted operation.

[0016] Step S13: The data from each sensing device, including real-time rainfall data, road water accumulation data, drainage network operation data and water pressure data in the pipeline, are aggregated through the urban road drainage Internet of Things sensing network. The data is then preprocessed in real time using edge computing nodes to filter out abnormal data and generate the corresponding multi-source heterogeneous dataset. In this embodiment of the invention, through the drainage Internet of Things sensing network, each sensing device transmits data to the regional data aggregation gateway at a preset sampling frequency (transmission delay ≤ 30 seconds, data packet loss rate ≤ 0.1%). The gateway summarizes the data every 5 minutes to form a raw dataset containing real-time rainfall data (e.g., rainfall intensity of 15.2 mm / h at 08:00 on 2024-07-10), road waterlogging data (water depth of 3.2 cm at K0+200 on Binhe Road), pipeline operation data (flow rate of 120 m³ / h at node J35), and water pressure data (water pressure of 0.32 MPa at node J35). Edge computing nodes (using ARM Cortex-A53 processors, computing power 1.2 TOPS, memory 2GB) are deployed on the gateway side for real-time preprocessing: the 3σ criterion is used to filter out abnormal data, such as a water level sensor collecting 150cm (outside the 0-100cm measurement range), which is judged as an outlier and removed; the mean imputation method is used to supplement missing data, such as a flow meter missing one data point due to a temporary communication interruption, which is filled with the average of the previous 5 data points (118m³ / h), and finally a multi-source heterogeneous dataset containing 12 points, 1 record every 5 minutes, and a total of 288 records / day is generated, with no invalid data.

[0017] Step S14: Obtain the GPS positioning information and timestamps corresponding to each sensing device, and establish a spatiotemporal coordinate system based on the corresponding GPS positioning information and timestamps to map the corresponding multi-source heterogeneous datasets to a unified spatiotemporal dimension, generating a multi-source heterogeneous spatiotemporal aligned dataset. In this embodiment of the invention, GPS positioning information output by the GPS positioning module (positioning accuracy ≤ 10 meters, update frequency 1 hour / time) built into each sensing device is obtained, such as the positioning coordinates of the point at K0+200 on Binhe Road (30°15′22.3″N, 120°05′18.6″E), and the timestamp of each data collection is obtained (accurate to the second, such as 2024-07-10 08:00:00). Based on location information and timestamps, a spatiotemporal coordinate system is established using the UTM projected coordinate system (X-axis for east, Y-axis for north, and T-axis for time). Each data point in the multi-source heterogeneous dataset is mapped to this coordinate system. For example, "2024-07-10 08:00:00, 30°15′22.3″ N (X=3356782.1m), 120°05′18.6″ E (Y=548921.3m), rainfall intensity 15.2mm / h" is mapped to (3356782.1, 548921.3, 20240710080000, 15.2). All data are mapped according to this rule to ensure that data from different locations and of different types are aligned in the same spatiotemporal dimension, generating a multi-source heterogeneous spatiotemporally aligned dataset with a spatiotemporal deviation of ≤1 minute and 10 meters, and no spatiotemporal misalignment.

[0018] Step S15: Perform noise filtering on the multi-source heterogeneous spatiotemporal aligned dataset to identify and eliminate fluctuation noise points caused by equipment failure and electromagnetic interference using adaptive wavelet threshold denoising. At the same time, perform format standardization conversion to generate a standardized urban drainage monitoring dataset.

[0019] In this embodiment of the invention, noise filtering is performed on a multi-source heterogeneous spatiotemporally aligned dataset using an adaptive wavelet threshold denoising algorithm (wavelet basis function selected as db4, decomposition level 3, threshold calculated according to the Birgé-Massart strategy): For fluctuation noise in water level data caused by vehicle vibration (e.g., a sudden change from 3.2cm to 5.8cm followed by rapid recovery), wavelet decomposition is used to decompose the signal into low-frequency approximate components and high-frequency detail components. The high-frequency detail components are processed with a threshold of 0.3cm, and after noise removal, the signal is reconstructed and corrected to 3.3cm. Similarly, for high-frequency noise in water pressure data caused by electromagnetic interference (e.g., fluctuations of ±0.05MPa at 0.32MPa), the same algorithm is used to remove it and correct it to 0.32MPa. After noise filtering, the signal-to-noise ratio of the data is improved to over 30dB, and the fluctuation error is ≤0.5%. The data was then standardized: all numerical data were retained to one decimal place (e.g., rainfall intensity 15.2 mm / h, flow rate 120.0 m³ / h), timestamps were standardized to "yyyy-MM-ddHH:mm:ss" format, coordinates were standardized to UTM projected coordinates (retaining one decimal place), data format was standardized to JSON, and field names were named according to "location ID-data type-value-timestamp-coordinates", generating a standardized urban drainage monitoring dataset, which can be directly used for subsequent drainage scheduling analysis.

[0020] Furthermore, step S2 includes the following steps: Step S21: Obtain the corresponding road network data and drainage pipe network GIS data through the target area of ​​urban roads. The road network data includes road segment name, start and end nodes, number of lanes and road surface elevation. The drainage pipe network GIS data includes pipe network direction, pipe diameter, manhole location and pump station affiliation. Based on the road network data and drainage pipe network GIS data, establish the spatial relationship between road segments and drainage pipe networks corresponding to the target area. In this embodiment of the invention, road network data and drainage pipe network GIS data of a target area (5 square kilometers, including 10 main roads and 20 secondary roads) are obtained from an urban geographic information platform. In the road network data, Binhe Road (road segment name) has start and end nodes N1-N5 (eastern starting point N1, western ending point N5), 6 lanes, and a road surface elevation of 29.6-29.8m; Xincheng Road has start and end nodes N6-N9, 4 lanes, and a road surface elevation of 29.4-29.5m; and Keji Road-Wenhua Road has start and end nodes N10-N13, 2 lanes, and a road surface elevation of 29.2-29.3m. In the drainage network GIS data, the pipeline route corresponding to Binhe Road is east-west (laid along the north side of the road), with a pipe diameter of 800mm, manhole locations J31-J35 (spaced 200 meters apart), and the pumping station belongs to Binhe Road Pumping Station; the pipeline route corresponding to Xincheng Road is north-south, with a pipe diameter of 600mm, manholes J41-J48, and the pumping station belongs to Xincheng Road Pumping Station; the pipeline route corresponding to Keji Road-Wenhua Road is east-west, with a pipe diameter of 500mm, manholes J61-J65, and the pumping station belongs to Keji Road Pumping Station. Using ArcGIS spatial overlay analysis tools, road network data and drainage pipe network GIS data are overlaid. Spatial relationships are established based on the rule that "the distance between the center line of the road and the center line of the pipe network is ≤ 5 meters": Binhe Road is associated with pipe networks J31-J35, Xincheng Road with pipe networks J41-J48, and Keji Road-Wenhua Road with pipe networks J61-J65. This forms a spatial association table of "road segment name-pipe network number-pump station affiliation", ensuring that each road segment is accurately matched with the corresponding pipe network without any misalignment.

[0021] Step S22: Obtain road functional attributes, drainage network hierarchy, and historical waterlogging risk level. The road functional attributes include arterial roads, secondary arterial roads, and branch roads. The drainage network hierarchy includes main pipes, branch pipes, and service pipes. Based on the spatial relationship between road segments and drainage networks, combined with the road functional attributes, drainage network hierarchy, and historical waterlogging risk level, the urban road target area is divided into sub-networks for each road segment to generate multiple target area road segment sub-networks. Each target area road segment sub-network contains one or more continuous roads and corresponding drainage networks. In this embodiment of the invention, the road functional attributes, drainage network level, and historical waterlogging risk level of each road segment are obtained: Binhe Road is a "main road", the corresponding pipeline is a "main pipe" (diameter ≥ 800 mm), and the historical waterlogging risk level in the past 3 years is "high" (average waterlogging ≥ 3 times per year); Xincheng Road is a "secondary road", the corresponding pipeline is a "branch pipe" (diameter 600-800 mm), and the risk level is "medium" (average waterlogging 1-2 times per year); Keji Road-Wenhua Road is a "branch road", the corresponding pipeline is a "household connection pipe" (diameter ≤ 600 mm), and the risk level is "low" (average waterlogging ≤ 1 time per year). Based on the spatial relationship established by S21, road segments are divided into subnets according to the consistency principle of "functional attribute - pipeline level - risk level": Binhe Road, which is "main road + main pipeline + high risk", is divided into subnet 1; Xincheng Road, which is "secondary road + branch pipeline + medium risk", is divided into subnet 2; and Keji Road-Wenhua Road, which is "branch road + household connection pipeline + low risk", is divided into subnet 3. Each subnet contains one continuous road and a corresponding complete pipeline: Subnet 1 includes Binhe Road (N1-N5) and J31-J35 pipelines; Subnet 2 includes Xincheng Road (N6-N9) and J41-J48 pipelines; and Subnet 3 includes Keji Road-Wenhua Road (N10-N13) and J61-J65 pipelines. After the division, the subnet boundaries are clear, and there are no roads or pipelines that overlap.

[0022] Step S23: Based on multiple target area road segment subnets, divide the rainfall data, water accumulation data, drainage operation data and pipe water pressure data corresponding to different urban road segment subnets from the urban standardized drainage monitoring dataset. Calculate the missing rate of the corresponding type of data for each urban road segment subnet, and if the missing rate exceeds a preset threshold, fill in the missing data by using the spatial difference of the same type of data from adjacent road segment subnets. In this embodiment of the invention, data is divided into three subnets based on the urban standardized drainage monitoring dataset (1 record every 5 minutes, 288 records / day): Subnet 1 acquires rainfall data (rainfall intensity sensors J1-J3), water accumulation data (water level sensors J1-J3), drainage operation data (Binhe Road pumping station), and pipe pressure data (J31-J35); Subnet 2 acquires data from the corresponding sensors and Xincheng Road pumping station; Subnet 3 acquires data from the corresponding sensors and Keji Road pumping station. The missing data rate is calculated using the formula "missing rate = number of missing records / total number of records × 100%", with a preset missing rate threshold of 5%: Subnet 1 has 2 missing rainfall data records (out of 288), a missing rate of 0.69% (<5%, no need to complete); Subnet 2 has 15 missing water accumulation data records, a missing rate of 5.21% (>5%, need to complete); Subnet 3 has 8 missing pipe pressure data records, a missing rate of 2.78% (<5%, no need to complete). For the 15 missing water accumulation data in subnet 2, the spatial difference method of similar data in adjacent subnets was used to complete them: based on the water accumulation depth of 3.2-3.8cm in subnet 1 (closest to subnet 2) and 2.8-3.2cm in subnet 3 during the same period, the water accumulation depth in subnet 2 was calculated according to "water accumulation depth in subnet 2 = (subnet 1 + subnet 3) / 2 × distance correction coefficient" (the distance between subnet 2 and subnet 1 is 1 km, and the correction coefficient is 0.95). For example, the missing data at 08:10 was completed as (3.5+3.0) / 2×0.95≈3.14cm. After completion, the missing data rate of water accumulation in subnet 2 was reduced to 0%, and the data integrity met the standard.

[0023] Step S24: Based on the rainfall data, water accumulation data, drainage operation data and pipe pressure data corresponding to different urban road sub-networks, conduct water accumulation increase rate and pipe network pressure assessment to obtain the urban water accumulation increase rate and drainage pipe network pressure degree corresponding to different urban road sub-networks.

[0024] In this embodiment of the invention, based on the complete monitoring data after the division of each subnet, the water accumulation rate and the pressure bearing capacity of the pipe network are calculated according to the methods described in S241-S245 above. Subnet 1: Rainfall data (2.53 mm rainfall per unit time, lasting 40 minutes), water accumulation data (depth 3.2-3.8 cm, expansion rate 20 m / min), drainage operation data (flow rate 120 m³ / h, power 30 kW), water pressure data (static pressure 0.32-0.42 MPa), the calculated water accumulation rate is 0.90 cm and the pipe network pressure bearing capacity is 15.5 (high pressure bearing capacity); Subnet 2: Rainfall data (2.12 mm rainfall per unit time), water accumulation data (depth 2.8-3.4 cm, expansion rate 18 m / min), drainage operation data, and water pressure data (static pressure 0.32-0.42 MPa), the calculated water accumulation rate is 0.90 cm and the pipe network pressure bearing capacity is 15.5 (high pressure bearing capacity); Based on the data (flow rate 100 m³ / h, power 25 kW) and water pressure data (static pressure 0.28-0.36 MPa), the calculated water accumulation rate is 1.16 cm and the pressure rating is 12.13 (medium pressure). For subnet 3: rainfall data (rainfall per unit time 1.85 mm), water accumulation data (depth 2.2-2.8 cm, expansion rate 15 m / min), drainage operation data (flow rate 90 m³ / h, power 22 kW), and water pressure data (static pressure 0.25-0.31 MPa), the calculated water accumulation rate is 0.40 cm and the pressure rating is 8.4 (low pressure). All calculations use the same formulas and parameter standards as the subsequent calculations. The results match the subnet functional attributes and risk levels (high-risk subnets have higher water accumulation rates and pressure ratings), with no calculation deviations, and can be directly used for subsequent drainage scheduling decisions.

[0025] Furthermore, step S24 includes the following steps: Step S241: Obtain the rainfall per unit time, rainfall duration and rainfall recurrence area through rainfall data corresponding to different urban road segment subnets; obtain the time-series changes in water accumulation depth and the expansion rate of water accumulation area corresponding to different monitoring points through water accumulation data; obtain the real-time flow rate of the pipe network and the operating power of the pumping station through drainage operation data; obtain the static pressure value, dynamic pressure value and pressure fluctuation frequency corresponding to different nodes of the pipe through pipe pressure data, so as to generate multi-dimensional features of road drainage corresponding to different urban road segment subnets; In this embodiment of the invention, the target area is divided into three urban road sub-networks (sub-network 1: Binhe Road, sub-network 2: Xincheng Road, sub-network 3: Keji Road-Wenhua Road), and data from each sub-network is extracted from the standardized drainage monitoring dataset. Subnet 1 rainfall data: Rainfall per unit time (cumulative 2.53mm from 08:00 to 08:05) and rainfall duration (40 minutes from 07:30 to 08:10) were obtained through rainfall intensity sensors. The rainfall recurrence area was determined to be a "once-in-two-years" event by comparing with the historical rainfall database. Water accumulation data: Water level sensors recorded the temporal changes in water depth at monitoring points J1-J3 (J1 08:00 3.2cm, 08:01 3.5cm). The expansion rate of the water accumulation area was calculated to be 20m / min by using the distance between adjacent monitoring points (100m) and the time difference of water accumulation (5 minutes). Drainage operation data: The flow meter obtained the real-time flow rate of the pipeline network at 120m³ / h, and the pump station monitoring system obtained the operating power at 30kW. Pipeline water pressure data: Pressure sensors recorded the static pressure value of node J35 at 0.32-0.42MPa and the dynamic pressure value at 0.26-0.32MPa. The pressure fluctuation frequency was calculated to be 3 times / minute based on the number of pressure changes within 1 hour. Using this method, 12 core indicators from 3 subnets were extracted to generate multidimensional features for road drainage. The data time granularity was uniformly set to 1 minute, with no indicators omitted.

[0026] Step S242: Obtain the pavement slope and pavement permeability coefficient corresponding to different urban road segment subnets, and calculate the rainfall infiltration based on the pavement slope and pavement permeability coefficient combined with the rainfall per unit time, rainfall duration and rainfall recurrence section, to obtain the pavement rainfall infiltration corresponding to different urban road segment subnets; In this embodiment of the invention, the pavement parameters of each sub-network are obtained from the urban road design archive: sub-network 1 pavement slope 0.3%, permeability coefficient 2.5×10⁻ 6 m / s (asphalt pavement); Subnet 2 slope 0.2%, permeability coefficient 2.2×10⁻ 6 m / s; Subnet 3 slope 0.1%, permeability coefficient 2.0×10⁻ 6 m / s. The Horton infiltration formula is used to calculate rainfall infiltration, which is: "Rainfall infiltration = Permeability coefficient × Road surface area × Rainfall duration × Slope influence coefficient". The road surface area is calculated based on the actual road area of ​​the subnet (subnet 1 is 1000m × 15m = 15000 square meters). The slope influence coefficient is set according to the slope value (0.3% corresponds to 1.2, 0.2% corresponds to 1.1, and 0.1% corresponds to 1.0). The rainfall duration is converted to seconds (40 minutes = 2400 seconds). Rainfall infiltration in subnet 1 = 2.5 × 10⁻⁻⁻⁶ 6 ×15000×2400×1.2=108m³; Subnet 2=2.2×10⁻ 6×12000×2400×1.1≈84.29m³; Subnet 3=2.0×10⁻ 6 ×18000×2400×1.0=86.4m³. The calculation result is rounded to two decimal places. Each parameter corresponds to the actual road surface characteristics, and there is no calculation deviation.

[0027] Step S243: Calculate the pipeline discharge volume based on the real-time pipeline flow and pump station operating power corresponding to different urban road segment sub-networks to obtain the pipeline discharge volume corresponding to different urban road segment sub-networks; In this embodiment of the invention, the discharge volume of the pipeline is calculated based on the real-time flow rate of each sub-network and the operating power of the pump station. The conversion formula is "pipeline discharge volume = real-time flow rate of the pipeline × pump station efficiency coefficient × operating time correction coefficient". The pump station efficiency coefficient is fixed at 0.85 according to the nameplate parameters (the efficiency range of a 30kW pump station is 0.8-0.9). The operating time correction coefficient is the ratio of the actual running time to 1 hour (sub-network 1 runs for 40 minutes, correction coefficient = 40 / 60≈0.67). Subnet 1 has a real-time flow rate of 120 m³ / h, and its discharge volume is approximately 67.32 m³ (120 × 0.85 × 0.67). Subnet 2 has a real-time flow rate of 100 m³ / h and operates for 35 minutes, with a discharge volume of approximately 49.58 m³ (100 × 0.85 × (35 / 60)). Subnet 3 has a real-time flow rate of 90 m³ / h and operates for 45 minutes, with a discharge volume of approximately 57.38 m³ (90 × 0.85 × (45 / 60)). The real-time flow rate was calculated using a 5-minute average to ensure data stability. The results are accurate to 0.01 m³, with no conversion errors.

[0028] Step S244: Based on the temporal changes in water depth and the rate of water expansion at different monitoring points, calculate the water expansion to obtain the waterlogged area of ​​different urban road sub-networks; calculate the corresponding urban waterlogging increase rate based on the road surface rainfall infiltration and pipe network discharge corresponding to different urban road sub-networks, specifically: urban waterlogging increase rate = (road surface rainfall infiltration - pipe network discharge) / road section waterlogged area; In this embodiment of the invention, the water accumulation area of ​​the road section is calculated based on the water accumulation data of each subnet monitoring point: Subnet 1 monitoring points J1-J3 (100 meters apart), at 08:02, the J1-J3 section had water accumulation (length 300 meters), the road width was 15 meters, and the water accumulation area = 300 × 15 = 4500 square meters; Subnet 2 had water accumulation length of 250 meters and width of 12 meters, with an area of ​​3000 square meters; Subnet 3 had water accumulation length of 400 meters and width of 18 meters, with an area of ​​7200 square meters. The urban waterlogging increase rate is calculated using the formula: "Urban waterlogging increase rate = (Road surface rainfall infiltration - Pipeline discharge) / Waterlogged area of ​​road section": Sub-network 1 = (108 - 67.32) / 4500 ≈ 40.68 / 4500 ≈ 0.0090 m (i.e., 0.90 cm); Sub-network 2 = (84.29 - 49.58) / 3000 ≈ 34.71 / 3000 ≈ 0.0116 m (1.16 cm); Sub-network 3 = (86.4 - 57.38) / 7200 ≈ 29.02 / 7200 ≈ 0.0040 m (0.40 cm). The increase rate results are rounded to four decimal places. Positive values ​​indicate an increase in waterlogging, and negative values ​​indicate a decrease in waterlogging, directly reflecting the waterlogging trend of each sub-network and eliminating calculation errors.

[0029] Step S245: Based on the static pressure value, dynamic pressure value, and pressure fluctuation frequency of different nodes in the pipeline within different urban road subnets, combined with the corresponding pump station operating power, perform pipeline pressure assessment calculation to obtain the pressure bearing capacity of the drainage pipeline network corresponding to different urban road subnets.

[0030] In this embodiment of the invention, the pressure bearing capacity of the pipeline network is assessed based on the water pressure data of each sub-network and the power of the pumping station. The assessment formula is "pressure bearing capacity of pipeline network = (static pressure overload frequency × 0.4 + pressure fluctuation amplitude × 0.3 + life decay amplitude × 0.3)". Subgrid 1 has a static pressure overload frequency of 20%, pressure fluctuation amplitude of 15%, and lifespan attenuation amplitude of 10%, with a pressure bearing capacity of (20×0.4+15×0.3+10×0.3)=15.5; Subgrid 2 has an overload frequency of 13.33%, fluctuation amplitude of 17.14%, and lifespan attenuation amplitude of 5.54%, with a pressure bearing capacity of approximately (13.33×0.4+17.14×0.3+5.54×0.3)≈12.13; Subgrid 3 has an overload frequency of 6.67%, fluctuation amplitude of 16.67%, and lifespan attenuation amplitude of 2.43%, with a pressure bearing capacity of approximately (6.67×0.4+16.67×0.3+2.43×0.3)≈8.4. The subnet is classified into three levels: "high (>12), medium (8-12), and low (<8)". Subnet 1 is classified as high pressure, subnet 2 as medium pressure, and subnet 3 as low pressure. The evaluation weights are set according to the degree of influence of the indicators on the safety of the pipeline network. The results can be directly used for drainage scheduling decisions.

[0031] Furthermore, the calculation of rainfall infiltration based on road surface slope and pavement permeability coefficient combined with rainfall per unit time, rainfall duration, and rainfall recurrence zone as described in step S242 includes the following steps: The rainfall per unit time is converted into the corresponding hourly rainfall intensity, and the length of the rainfall recurrence segment corresponding to the current rainfall is determined based on the rainfall recurrence segment. In this embodiment of the invention, the unit-time rainfall is extracted from the standardized drainage monitoring dataset for three urban road segment sub-networks and converted into hourly rainfall intensity (unit: mm / h). The conversion rule is "hourly rainfall intensity = unit-time rainfall × 60 / unit time (minutes)", with the unit time uniformly set to 5 minutes (consistent with the sensor sampling frequency). Sub-network 1 (Binhe Road area) had a unit-time rainfall of 2.53 mm (5-minute cumulative value) from 08:00 to 08:05, and an hourly rainfall intensity of 2.53 mm × 60 / 5 = 30.36 mm / h; Sub-network 2 (Xincheng Road area) had a unit-time rainfall of 2.12 mm during the same period, and an hourly rainfall intensity of 2.12 × 12 = 25.44 mm / h; Sub-network 3 (Keji Road-Wenhua Road area) had a unit-time rainfall of 1.85 mm during the same period, and an hourly rainfall intensity of 1.85 × 12 = 22.2 mm / h. Based on the historical rainfall database (rainfall records of the past 10 years), the rainfall recurrence intervals are determined: hourly rainfall intensity of 30-35 mm / h corresponds to a "2-year return period" recurrence interval (i.e., rainfall of this intensity occurs on average once every 2 years), 25-30 mm / h corresponds to a "1.5-year return period", and 20-25 mm / h corresponds to a "1-year return period". Based on this, the recurrence interval length of subnet 1 is determined to be 2 years, that of subnet 2 is 1.5 years, and that of subnet 3 is 1 year, ensuring that the recurrence interval length accurately matches the actual rainfall intensity.

[0032] Furthermore, the rainfall retention probability is obtained, and the rainfall recurrence attenuation is estimated based on the hourly rainfall intensity and the rainfall retention probability to the length of the rainfall recurrence segment corresponding to the current rainfall, so as to obtain the rainfall recurrence segment attenuation coefficient. In this embodiment of the invention, rainfall retention data from the past five years provided by the urban meteorological department was used to statistically analyze the rainfall retention probability (the probability of rainwater not forming runoff during rainfall) for each road sub-network: Sub-network 1 (asphalt pavement, low pavement aging) had a rainfall retention probability of 15% (i.e., 15% of rainfall was absorbed or evaporated by the pavement), Sub-network 2 (asphalt pavement, moderately aged) had a probability of 12%, and Sub-network 3 (cement pavement, high aging) had a probability of 8%. The attenuation coefficient was calculated using the formula "Rainfall recurrence section attenuation coefficient = 1 - (hourly rainfall intensity × rainfall retention probability / 1000)", which is based on the logical premise that "the greater the rainfall intensity, the more significant the impact of retention probability on the recurrence section length". Subnet 1 attenuation coefficient = 1 - (30.36 × 15 / 1000) = 1 - 0.4554 = 0.5446; Subnet 2 = 1 - (25.44 × 12 / 1000) = 0.6947; Subnet 3 = 1 - (22.2 × 8 / 1000) = 1 - 0.1776 = 0.8224. The attenuation coefficient ranges from 0 to 1. The smaller the value, the greater the impact of retention on the recurrence section length. The calculation results are rounded to four decimal places, and there are no outliers outside the range.

[0033] Furthermore, the corresponding pavement material type is obtained through the urban road segment subnet, and the urban road segment subnet is divided into multiple segment boundary calculation units based on the pavement permeability coefficient, rainfall per unit time, rainfall duration, pavement slope, and rainfall recurrence section attenuation coefficient of each unit, and the components are summed to obtain the pavement rainfall infiltration amount corresponding to different urban road segment subnets.

[0034] In this embodiment of the invention, by obtaining the road surface material type of each subnet (subnets 1 and 2 are asphalt pavement, and subnet 3 is cement pavement), and combining the road surface slope to divide the segment boundary calculation unit: Subnet 1 has a slope of 0.3% (one-way slope), and is divided into 3 units according to the slope change inflection point (no inflection point) and road intersection (unit 1: K0+000-K0+300, area 4500m²; unit 2: K0+300-K0+700, area 6000m²; unit 3: K0+700-K1+000, area 4500m²), totaling 15000m² (consistent with the total area of ​​the subnet); Subnet 2 has a slope of 0.2%, and is divided into 3 units (total area 12000m²); Subnet 3 has a slope of 0.1%, and is divided into 4 units (total area 18000m²). The formula "Unit rainfall infiltration component = pavement permeability coefficient × unit area × rainfall duration × pavement slope influence coefficient × rainfall recurrence section attenuation coefficient" is used to calculate each unit component. The pavement permeability coefficient and slope influence coefficient are based on the S242 parameters (subnet 1 permeability coefficient 2.5 × 10⁻⁻⁴). 6(Rainfall speed m / s, slope influence coefficient 1.2), duration 40 minutes (2400s). Subnet 1 unit 1 component = 2.5 × 10⁻ 6 ×4500×2400×1.2×0.5446≈2.5×10⁻ 6 ×4500×2400×1.2×0.5446≈17.57m³; Unit 2≈2.5×10⁻ 6 ×6000×2400×1.2×0.5446≈23.43m³; Unit 3≈17.57m³; Total =17.57+23.43+17.57≈58.57m³ (i.e., the infiltration of rainfall on the road surface of subnet 1). Using the same method, the infiltration of subnet 2 is approximately 49.82m³ and subnet 3 is approximately 63.15m³. Each unit component calculation is associated with material, slope, and attenuation coefficient, with results accurate to 0.01m³, showing no calculation deviation.

[0035] Furthermore, the calculation of water accumulation expansion based on the temporal changes in water depth and the rate of water accumulation expansion at different monitoring points, as described in step S244, includes the following steps: The latitude and longitude coordinates and altitude of different monitoring points are obtained, and the expansion rate of the water accumulation area is decomposed vertically and horizontally based on the latitude and longitude coordinates and altitude to obtain the water accumulation expansion rate components in the vertical and horizontal directions. In this embodiment of the invention, the latitude and longitude coordinates and altitude of each monitoring point in three road sub-networks are obtained from the sensing devices of the urban road drainage Internet of Things sensing network: monitoring points J1 (30°15′22.3″N, 120°05′18.6″E, altitude 29.8m), J2 (30°15′25.1″N, 120°05′22.4″E, altitude 29.7m), and J3 (30°15′27.9″N, 120°05′26.2″E, altitude 29.6m) in sub-network 1 (Binhe Road); monitoring points J4-J6 in sub-network 2 (Xincheng Road) have an altitude of 29.5-29.4m; and monitoring points J7-J9 in sub-network 3 (Keji Road-Wenhua Road) have an altitude of 29.3-29.2m. Based on the latitude and longitude, the horizontal distance (J1-J2 horizontal distance 100m) and the altitude difference (J1-J2 altitude difference 0.1m) are calculated. The expansion rate of the water accumulation area (previously calculated as 20m / min for subnet 1) is decomposed into vertical and horizontal components: Vertical component = expansion rate × (altitude difference / horizontal distance) = 20m / min × (0.1m / 100m) = 0.02m / min (i.e. 2cm / min); Horizontal component = expansion rate × √[1-(altitude difference / horizontal distance)²] ≈ 20 × 0.9999 ≈ 19.998m / min (approximately 20m / min). Subnet 2 expands at a speed of 18 m / min, with an elevation difference of 0.08 m / 100 m, a vertical component of 0.0144 m / min, and a horizontal component of approximately 17.999 m / min; Subnet 3 expands at a speed of 15 m / min, with an elevation difference of 0.05 m / 100 m, a vertical component of 0.0075 m / min, and a horizontal component of approximately 14.999 m / min. The component decomposition is based on the Pythagorean theorem for right triangles, and the results are rounded to four decimal places.

[0036] Furthermore, based on the temporal variation of water depth at different monitoring points and combined with the water expansion velocity component in the vertical direction, the rate of change of water depth per unit time is calculated, and the water submersion height in the vertical direction at different monitoring points is determined by combining the corresponding road elevation data. In this embodiment of the invention, the rate of change of water depth per unit time is calculated based on the temporal variation of water depth at each monitoring point. The water depth at monitoring point J1 in subnet 1 is 3.2cm at 08:00, 3.5cm at 08:01, and 3.8cm at 08:02. The rate of change of depth is (3.8-3.2)cm / 2 minutes = 0.3cm / minute. Combining this with the vertical expansion velocity component of 0.02m / minute (2cm / minute), the rate of change of water depth per unit time is superimposed with the vertical component, and the rate of change of depth remains the core indicator (the vertical component has a negligible impact). Road surface elevation data for each monitoring point are obtained from the urban road design archives: J1 road surface elevation 29.8m (curbstone elevation 30.0m), J2 road surface elevation 29.7m, and J3 road surface elevation 29.6m. The floodwater height is calculated as follows: Floodwater height = road surface elevation + water depth. For J108:02, the floodwater height is 29.8m + 0.038m = 29.838m (not exceeding the curb elevation). For J208:02, with a water depth of 3.6cm, the floodwater height is 29.7m + 0.036m = 29.736m. For J308:02, with a water depth of 3.4cm, the floodwater height is 29.6m + 0.034m = 29.634m. The water depth is consistently converted to meters during calculation, and the floodwater height is rounded to three decimal places to ensure accuracy with the road surface elevation.

[0037] Furthermore, the road topology corresponding to each city road segment subnet is obtained, and it is determined whether there are slope changes and obstacles. If the road has no obvious obstacles and the slope is gentle, the horizontal expansion of water accumulation is calculated as a circular or elliptical diffusion. The diffusion radius is obtained by multiplying the corresponding water expansion velocity component in the horizontal direction by time with the initial water accumulation point as the center. At the same time, the horizontal expansion range in this state is determined based on the diffusion radius. If the road has slope or obstacles, the horizontal expansion of water accumulation is calculated by correction along the slope direction. The correction coefficient for the expansion velocity in the downward slope direction is greater than 1, and the correction coefficient for the upward slope direction is less than 1. The obstacle area is marked as the expansion boundary constraint, and the horizontal expansion range in this state is determined based on the expansion velocity correction coefficient combined with the expansion boundary constraint and the water expansion velocity component. In this embodiment of the invention, the road topology of each subnet is obtained: Subnet 1 (Binhe Road) is a two-way 6-lane road with no obstacles and a road surface slope of 0.3% (one-way gentle slope, no turns), and is determined to be "no obstacles + gentle slope". The horizontal expansion of water accumulation is calculated as an elliptical diffusion (the major axis is along the road direction, and the minor axis is perpendicular to the road direction). Taking J1 as the initial water accumulation point (water accumulation starts at 08:00), the horizontal expansion velocity component is 20m / minute, the diffusion time at 08:02 is 2 minutes, the diffusion radius (major axis) = 20m / minute × 2 minutes = 40m, the minor axis = major axis × 0.5 (road width 15m, minor axis matches road width) = 20m, and the horizontal expansion range is an elliptical area with a major axis of 40m and a minor axis of 20m (area ≈ π × 20 × 10 ≈ 628.32m²). Subnet 2 (Xincheng Road) has two manhole cover obstacles (50m apart), and the road surface slope is 0.2% (one-way slope). It is judged as "obstacle + slope". The expansion speed correction factor in the downward slope direction (westward) is set to 1.2 (accelerated expansion), and the correction factor in the upward slope direction (eastward) is set to 0.8 (decelerated expansion). The obstacle is marked as the expansion boundary (cannot be crossed). The horizontal component is 17.999m / min. At 08:02, the downward diffusion radius is 17.999×2×1.2≈43.2m, and the upward diffusion radius is 17.999×2×0.8≈28.8m. After deducting the 10m area occupied by the obstacle, the horizontal expansion range is trapezoidal (upper base 28.8m, lower base 43.2m, and road width 12m), with an area ≈ (28.8+43.2)×12 / 2≈432m². Subnet 3 (Keji Road - Wenhua Road) has a median barrier with a slope of 0.1%. The correction factor is the same as that of subnet 2. The expansion range is calculated as a polygon (excluding the area of ​​the median barrier), with an area of ​​approximately 510 m². The correction factor is set according to the slope (the greater the slope, the greater the difference in the correction factor). The boundary of the barrier is determined by actual coordinates.

[0038] Furthermore, the water accumulation area at different monitoring points is calculated by superimposing the water inundation height in the vertical direction and the horizontal expansion range in the horizontal direction, so as to obtain the water accumulation area of ​​different urban road subnets.

[0039] In this embodiment of the invention, the area of ​​the flooding height and horizontal expansion range of each subnet monitoring point is calculated by superimposing the area: Subnet 1 has three monitoring points J1, J2 and J3. At 08:02, the expansion range of J1 is 628.32m², the expansion range of J2 (diffusion time 1 minute, area ≈π×10×5≈157.08m²) and the expansion range of J3 (diffusion time 0.5 minutes, area ≈π×5×2.5≈39.27m²). The three areas do not overlap (distance 100m). The flooded area of ​​the road section in subnet 1 = 628.32 + 157.08 + 39.27 ≈ 824.67m². Subnet 2 has three monitoring points: J4, J5, and J6. J4 has an expansion range of 432 m², J5 (with obstacles) has an expansion range of 380 m², and J6 has an expansion range of 320 m². J4 and J5 overlap by 10 m². The water accumulation area of ​​subnet 2 is approximately 1122 m² (432 + 380 + 320 - 10). Subnet 3 has three monitoring points: J7, J8, and J9, with expansion ranges of 510 m², 480 m², and 450 m² respectively. There is no overlap, and the water accumulation area is approximately 1440 m² (510 + 480 + 450). The overlay calculation uses a GIS spatial overlay tool, matching the area locations by the monitoring point coordinates. Overlapping areas are merged according to the "largest value" principle (retaining the area corresponding to the larger water depth). The area result is rounded to two decimal places to ensure consistency with the actual water coverage area and to prevent double counting or omissions.

[0040] Furthermore, step S245 includes the following steps: Obtain the pipeline design pressure values ​​for pipelines within different urban road subnets, and statistically analyze the percentage of times the static pressure value at different pipeline nodes exceeds the pipeline design pressure value, in order to determine the corresponding pressure overload frequency. In this embodiment of the invention, the design pressure values ​​of the pipes in three road sub-networks are obtained from the urban drainage pipe network design archives: Sub-network 1 (Binhe Road) uses DN800 reinforced concrete pipes with a design pressure value of 0.4MPa; Sub-network 2 (Xincheng Road) uses DN600 reinforced concrete pipes with a design pressure value of 0.35MPa; Sub-network 3 (Keji Road-Wenhua Road) uses DN500 reinforced concrete pipes with a design pressure value of 0.3MPa. From the standardized drainage monitoring dataset, static pressure monitoring records (1 record every 2 minutes, 30 records in total) for each sub-network pipeline node within one hour (08:00-09:00) were extracted: Sub-network 1 node J35 had 6 records with static pressure exceeding 0.4 MPa (e.g., 08:10 0.42 MPa, 08:25 0.41 MPa); Sub-network 2 node J48 had 4 records with static pressure exceeding 0.35 MPa; Sub-network 3 node J62 had 2 records with static pressure exceeding 0.3 MPa. The overload frequency was calculated using the formula "Pressure overload frequency = number of overload records / total number of records × 100%": Sub-network 1 = 6 / 30 × 100% = 20%; Sub-network 2 = 4 / 30 × 100% ≈ 13.33%; Sub-network 3 = 2 / 30 × 100% ≈ 6.67%. The frequency results are retained to two decimal places, which intuitively reflects the frequency of static pressure overload in each subnetwork pipeline, with no statistical bias.

[0041] Furthermore, the ratio between the maximum fluctuation difference of the dynamic pressure value corresponding to different nodes of the pipeline and the pipeline design pressure value is calculated to obtain the corresponding pressure fluctuation amplitude. In this embodiment of the invention, the maximum fluctuation difference (the difference between the maximum and minimum dynamic pressure values) is calculated based on the dynamic pressure value monitoring records of each subnet pipeline node within one hour: The dynamic pressure value recording range of subnet node J35 is 0.26-0.32MPa (minimum 0.26MPa at 08:05, maximum 0.32MPa at 08:30), and the maximum fluctuation difference is 0.32-0.26=0.06MPa; the dynamic pressure value range of subnet node J48 is 0.22-0.28MPa, and the maximum fluctuation difference is 0.06MPa; the dynamic pressure value range of subnet node J62 is 0.18-0.23MPa, and the maximum fluctuation difference is 0.05MPa. The pressure fluctuation amplitude is calculated using the formula "Pressure fluctuation amplitude = Maximum fluctuation difference / Pipeline design pressure value × 100%": Subnet 1 = 0.06 / 0.4 × 100% = 15%; Subnet 2 = 0.06 / 0.35 × 100% ≈ 17.14%; Subnet 3 = 0.05 / 0.3 × 100% ≈ 16.67%. The amplitude results are rounded to two decimal places. Larger values ​​indicate more drastic changes in pipeline dynamic pressure. This ensures accurate calculation and guarantees a correct proportional relationship with the design pressure value.

[0042] Furthermore, based on the pressure fluctuation frequency of different nodes in the pipeline within different urban road sub-networks and the corresponding pump station operating power, the corresponding drainage network life attenuation amplitude is calculated. In addition, the pressure bearing capacity of the pipeline is assessed by combining the pressure overload frequency and pressure fluctuation amplitude, so as to obtain the pressure bearing capacity of the drainage network corresponding to different urban road sub-networks.

[0043] In this embodiment of the invention, the lifespan attenuation amplitude is calculated using the formula "drainage network lifespan attenuation amplitude = pressure fluctuation frequency × pump station operating power correction coefficient × 0.5", where the pump station operating power correction coefficient is set according to the ratio of actual power to standard power (30kW): Sub-network 1 pump station power 30kW (correction coefficient 1.0), lifespan attenuation amplitude = 20% × 1.0 × 0.5 = 10%; Sub-network 2 pump station power 25kW (correction coefficient 0.83), lifespan attenuation amplitude = 13.33% × 0.83 × 0.5 ≈ 5.54%; Sub-network 3 pump station power 22kW (correction coefficient 0.73), lifespan attenuation amplitude = 6.67% × 0.73 × 0.5 ≈ 2.43%. Then, using the comprehensive evaluation formula "Pipeline pressure bearing capacity = (Pressure overload frequency × 0.4 + Pressure fluctuation amplitude × 0.3 + Life decay amplitude × 0.3)" to calculate: Sub-network 1 = (20 × 0.4 + 15 × 0.3 + 10 × 0.3) = 8 + 4.5 + 3 = 15.5; Sub-network 2 = (13.33 × 0.4 + 17.14 × 0.3 + 5.54 × 0.3) ≈ 5.33 + 5.14 + 1.66 = 12.13; Sub-network 3 = (6.67 × 0.4 + 16.67 × 0.3 + 2.43 × 0.3) ≈ 2.67 + 5.00 + 0.73 = 8.4. The subnet is classified into three levels: "high (>12), medium (8-12), and low (<8)". Subnet 1 is classified as high pressure bearing capacity, subnet 2 as medium pressure bearing capacity, and subnet 3 as low pressure bearing capacity. The evaluation weights are set according to the degree of impact of each indicator on the safety of the pipeline network. The results can be directly used to guide the adjustment of pump station power and pipeline network maintenance plans.

[0044] Furthermore, step S3 includes the following steps: Step S31: Based on the urban waterlogging increase rate corresponding to different urban road segment subnets, obtain the corresponding real-time waterlogging increase rate, waterlogging increase rate time series slope, and waterlogging increase rate deviation rate from the historical average for the same period. Then, multiply the real-time waterlogging increase rate by the waterlogging increase rate time series slope and divide by the waterlogging increase rate deviation rate from the historical average for the same period to obtain the corresponding waterlogging increase rate risk contribution. In this embodiment of the invention, relevant indicators of water accumulation rate are extracted from the monitoring data of each road segment sub-network: the real-time water accumulation rate of sub-network 1 (Binhe Road) is 0.90cm (data at 08:02), and the slope of the water accumulation rate over time is calculated from the data of the past 10 minutes (08:02 0.90cm, 08:07 0.95cm, 08:12 1.00cm), slope = (1.00-0.90)cm / 10 minutes = 0.01cm / minute; the average water accumulation rate of the same period in history (08:00-08:12 on July 10 of the past 3 years) is 0.70cm, and the deviation rate = (real-time increase rate - historical average) / historical average × 100% = (0.90-0.70) / 0.70 × 100% ≈ 28.57%. The risk contribution of water accumulation rate is calculated using the formula "Water accumulation rate increase risk contribution = (real-time water accumulation rate increase × water accumulation rate increase time series slope) / water accumulation rate increase deviation rate": Subnet 1 = (0.90 × 0.01) / 28.57% ≈ 0.009 / 0.2857 ≈ 0.0315. Subnet 2 (Xincheng Road): Real-time increase rate 1.16cm, time-series slope 0.012cm / minute, historical average 0.90cm, deviation rate 28.89%, risk contribution = (1.16×0.012) / 28.89%≈0.0139 / 0.2889≈0.0481; Subnet 3 (Keji Road-Wenhua Road): Real-time increase rate 0.40cm, time-series slope 0.005cm / minute, historical average 0.35cm, deviation rate 14.29%, risk contribution = (0.40×0.005) / 14.29%≈0.002 / 0.1429≈0.0140. Calculation results are rounded to four decimal places; larger values ​​indicate higher water accumulation risk contribution. There is no calculation bias.

[0045] Step S32: Calculate the corresponding pressure bearing risk contribution of the drainage pipe network based on the pressure bearing degree of the drainage pipe network corresponding to different urban road sub-networks, where the pressure bearing risk contribution of the pipe network = 1 - pressure bearing degree of drainage pipe network / preset pressure bearing threshold. In this embodiment of the invention, a preset pressure-bearing threshold for the drainage network is established based on network design standards and safety redundancy: the preset pressure-bearing threshold for the main pipe (sub-network 1) is 20 (corresponding to the pressure-bearing level classification mentioned above, the upper limit of high pressure), the threshold for the branch pipe (sub-network 2) is 15 (the upper limit of medium pressure), and the threshold for the service pipe (sub-network 3) is 10 (the upper limit of low pressure). The risk contribution of network pressure is calculated using the formula "Network pressure risk contribution = 1 - Drainage network pressure level / Preset pressure threshold": Sub-network 1 pressure level 15.5, risk contribution = 1 - 15.5 / 20 = 1 - 0.775 = 0.225; Sub-network 2 pressure level 12.13, risk contribution = 1 - 12.13 / 15 ≈ 1 - 0.8087 = 0.1913; Sub-network 3 pressure level 8.4, risk contribution = 1 - 8.4 / 10 = 1 - 0.84 = 0.16. The calculation results are rounded to four decimal places. The larger the value, the higher the risk of pressure on the pipeline. For example, subnet 1 has a higher risk contribution than other subnets because its pressure level is close to the threshold. This is consistent with the actual safety risk assessment logic and there is no parameter setting deviation.

[0046] Step S33: Calculate the drainage network negative pressure index corresponding to different urban road sub-networks based on the risk contribution of water accumulation rate and the risk contribution of pipeline pressure. In this embodiment of the invention, the negative pressure index of the drainage network is calculated using a weighted summation formula. The weights are set according to the degree of impact of risks on drainage safety: the contribution weight of the risk of water accumulation rate is 0.4 (water accumulation directly affects road traffic), and the contribution weight of the risk of network pressure is 0.6 (the risk of network rupture is more dangerous). The formula is "Drainage network negative pressure index = contribution of water accumulation rate risk × 0.4 + contribution of network pressure risk × 0.6". Subnet 1 = 0.0315 × 0.4 + 0.225 × 0.6 ≈ 0.0126 + 0.135 = 0.1476; Subnet 2 = 0.0481 × 0.4 + 0.1913 × 0.6 ≈ 0.0192 + 0.1148 = 0.1340; Subnet 3 = 0.0140 × 0.4 + 0.16 × 0.6 ≈ 0.0056 + 0.096 = 0.1016. The negative pressure index ranges from 0 to 1. The higher the value, the higher the overall pressure of the drainage system. Subnet 1 has the highest negative pressure index due to the high risk of pressure on the pipe network, while subnet 3 has the lowest pressure. The results match the risk level of the subnets and there is no subjective bias in weight allocation.

[0047] Step S34: Obtain the urban road network topology and generate an urban road area drainage pressure distribution map based on the drainage pipe network negative pressure index corresponding to different urban road segment subnets and the urban road network topology.

[0048] In this embodiment of the invention, the urban road network topology is obtained from the urban geographic information platform, including vector data such as the centerline, intersections, and lane widths of all roads within the target area. Binhe Road (subnet 1) and Xincheng Road (subnet 2) intersect at nodes N5-N6, and Xincheng Road intersects with Keji Road-Wenhua Road (subnet 3) at nodes N9-N10, forming an "L"-shaped road network connection. Based on the negative pressure index of each subnet's drainage network, an ArcGIS map rendering tool is used to generate a drainage pressure distribution map: a color grading standard is set (0.10-0.12 is green, 0.12-0.14 is yellow, and 0.14-0.16 is red). Subnet 1 (0.1476) is rendered as red, subnet 2 (0.1340) as yellow, and subnet 3 (0.1016) as green. The map marks the boundaries of each subnetwork, the locations of manholes (J31-J35, J41-J48, etc.), and the locations of pump stations. Red areas are marked "High pressure, pump station power recommended," yellow areas "Medium pressure, monitoring needs to be strengthened," and green areas "Low pressure, normal operation." The generated distribution map perfectly matches the road network topology, and the color grading intuitively reflects pressure differences. It can be directly used by dispatchers to formulate targeted measures, with no map rendering bias or information omissions.

[0049] Furthermore, step S4 includes the following steps: Step S41: Calculate the drainage pressure difference between each pair of adjacent urban road segment subnets based on the urban road area drainage pressure distribution map, so as to obtain the real-time drainage pressure difference between different urban road segment subnets. In this embodiment of the invention, based on the previously generated drainage pressure distribution map, the negative pressure index of the drainage network for the three road segment subnets is extracted (subnet 1: 0.1476, subnet 2: 0.1340, subnet 3: 0.1016). The real-time drainage pressure difference between each pair of adjacent subnets is calculated, and the pressure difference is taken as the absolute value of the negative pressure index of the two subnets. Subnet 1 and subnet 2 are adjacent (intersecting at nodes N5-N6), and the real-time drainage pressure difference is |0.1476-0.1340|=0.0136; subnet 2 and subnet 3 are adjacent (intersecting at nodes N9-N10), and the pressure difference is |0.1340-0.1016|=0.0324; subnet 1 and subnet 3 are not directly adjacent, so no pressure difference needs to be calculated. Preset pressure difference thresholds: Level 1 emergency dispatch threshold 0.03 (excessive pressure difference can easily cause pipeline overload), safety threshold 0.015 (pressure difference is within a stable range). The calculation results are retained to 4 decimal places, which clarifies the degree of pressure difference between adjacent subnets and provides data basis for subsequent dispatch level determination. There is no calculation deviation.

[0050] Step S42: Obtain the corresponding drainage pipe network connection relationship between each pair of adjacent city road segment subnets, and determine the corresponding pressure transmission path and the corresponding schedulable drainage flow overflow value based on the corresponding drainage pipe network connection relationship between each pair of adjacent city road segment subnets; In this embodiment of the invention, the pipe network connection relationship between adjacent sub-networks is obtained from the drainage pipe network GIS data: Sub-network 1 (Binhe Road, DN800 main pipe) and Sub-network 2 (Xincheng Road, DN600 branch pipe) are connected by a 600mm diameter connecting pipe (located between manholes J35 and J41 at the N5-N6 intersection); Sub-network 2 and Sub-network 3 (Keji Road-Wenhua Road, DN500 service pipe) are connected by a 500mm diameter connecting pipe (located between manholes J48 and J61 at the N9-N10 intersection). Based on the connection relationship, the pressure transmission path is determined: the pressure of Sub-network 1 can be transmitted to Sub-network 2 through the connecting pipe, and the pressure of Sub-network 2 can be transmitted bidirectionally to Sub-network 1 and Sub-network 3. Calculate the overflow value of the dispatchable drainage flow: Overflow value = (Design flow of high-pressure sub-network - Real-time flow) × Connecting pipe flow coefficient (0.8 for DN600 connecting pipe, 0.7 for DN500): Sub-network 1 real-time flow 120 m³ / h, design flow 200 m³ / h, overflow value to sub-network 2 = (200-120)×0.8 = 64 m³ / h; Sub-network 2 real-time flow 100 m³ / h, design flow 150 m³ / h, overflow value to sub-network 3 = (150-100)×0.7 = 35 m³ / h, overflow value to sub-network 1 = (150-100)×0.8 = 40 m³ / h; Sub-network 3 real-time flow 90 m³ / h, design flow 120 m³ / h, no overflow value (lowest pressure). The overflow value clearly defines the upper limit of the dispatchable flow for each sub-network, ensuring that dispatching does not exceed the network's carrying capacity and that there is no parameter setting deviation.

[0051] Step S43: Based on the pressure transmission path and the overflow value of the dispatchable drainage flow, and combined with the real-time drainage pressure difference between different urban road sub-networks, drainage is classified and dispatched. When the real-time drainage pressure difference exceeds the preset threshold, it is determined to be a Level 1 emergency dispatch, and the goal is to quickly relieve pressure, including starting the emergency pumping station of the high pressure difference sub-network, opening the gate of the connecting pipe to achieve cross-sub-network diversion, and temporarily shutting down the non-essential water facilities of the low pressure difference sub-network to free up drainage capacity. When the real-time drainage pressure difference is between the preset threshold and the safety threshold, it is determined to be a Level 2 mitigation dispatch, with the goal of balancing pressure, including adjusting the operating frequency of the pumping station within it, optimizing the gate opening degree to control the drainage flow distribution, and prioritizing the drainage of potential pipe blockages in the direction of pressure difference. When the real-time drainage pressure difference is lower than the safety threshold, it is determined to be a Level 3 maintenance dispatch, with the goal of maintaining stable operation, including maintaining the basic operating parameters of the existing drainage facilities, increasing the frequency of water accumulation monitoring in key road sections, and pre-storing emergency dispatch resources, to generate a graded dispatch strategy for urban road drainage. In this embodiment of the invention, the scheduling level is determined and a strategy is generated based on the pressure difference and scheduling rules: the pressure difference between subnet 1 and subnet 2 is 0.0136 (between the safety threshold of 0.015 and the first-level threshold of 0.03), which is determined to be a second-level relief scheduling. The target is to balance the pressure: adjust the operating frequency of the pump station in subnet 1 (from 50Hz to 45Hz, and the flow rate from 120m³ / h to 108m³ / h), optimize the opening and closing degree of the gate valve of the J35-J41 connecting pipe (from 50% to 70%, and increase the diversion flow to subnet 2 to 15m³ / h), and check whether there is any blockage in the J35 inspection well on the transmission path from subnet 1 to subnet 2 (if there is no blockage, no additional treatment is required). The pressure difference between subnet 2 and subnet 3 is 0.0324 (exceeding the first-level threshold of 0.03), indicating a first-level emergency dispatch. The target is rapid pressure relief: activate the emergency pumping station in subnet 2 (20kW standby power, additional flow rate of 30m³ / h), fully open the gate valves of the J48-J61 connecting pipe (100%, increasing the diversion flow rate to 35m³ / h), and temporarily shut down non-essential water facilities in subnet 3 (such as greening irrigation and municipal watering, freeing up 10m³ / h of drainage capacity). Subnet 1 has no other high pressure difference and maintains basic monitoring. The third-level maintenance dispatch targets areas without high pressure differences (such as within subnet 3), maintaining the pumping station power at 22kW and the gate opening degree at 50%, increasing the water accumulation monitoring frequency from 5 minutes / time to 2 minutes / time, and pre-stocking emergency pumping equipment (2 units, each with a flow rate of 50m³ / h). The dispatch strategy covers pressure regulation, facility control, and emergency preparedness, with no measures omitted.

[0052] Step S44: The urban road drainage hierarchical scheduling strategy is sent to the drainage facilities in the corresponding area through the IoT encrypted communication link to perform drainage scheduling and feedback the drainage effect data after scheduling. If the drainage effect data does not meet the requirements, the process returns to step S3 to perform closed-loop scheduling. Based on the new drainage pressure distribution map, the pressure difference is adjusted to calculate and generate an optimized scheduling strategy, which is then executed again until the drainage effect meets the requirements.

[0053] In this embodiment of the invention, the scheduling strategy is distributed to the corresponding facilities via a LoRaWAN encrypted communication link (encryption algorithm AES-128, transmission rate 50kbps, latency ≤30 seconds): Subnet 1 pump station receives the "reduce frequency to 45Hz" command, and J35 gate receives the "opening degree 70%" command; Subnet 2 emergency pump station receives the "start" command, J48 gate receives the "fully open" command, and subnet 3 water facilities receive the "shut down" command. After each facility executes the command, the drainage effect data is fed back through the IoT sensing network: after 10 minutes, the flow rate of subnet 1 is 108 m³ / h, the flow rate of subnet 2 is 145 m³ / h (including 30 m³ / h of the emergency pump station), and the flow rate of subnet 3 is 90 m³ / h. The pressure difference between subnet 2 and subnet 3 drops to 0.028 (still higher than the safety threshold), which does not meet the effect requirement of "pressure difference below 0.03". Returning to step S3, a pressure distribution map is generated based on the new monitoring data (subnet 2 negative pressure index 0.128, subnet 3 0.105, pressure difference 0.023). The scheduling strategy is adjusted: the power of the emergency pump station in subnet 2 is increased to 25kW (adding an additional flow rate of 38m³ / h), and the execution is reissued. Data feedback after 5 minutes shows: subnet 2 flow rate 153m³ / h, pressure difference reduced to 0.022 (below the first-level threshold of 0.03), meeting the performance requirements. Closed-loop scheduling is then stopped. Throughout the process, communication was stable, data feedback was timely, strategy adjustments were precise, and there were no execution deviations.

[0054] Furthermore, the present invention also provides an Internet of Things-based intelligent drainage scheduling system for urban roads, including a processor, a memory, and a computer program stored in the memory and executable on the processor, for performing the Internet of Things-based intelligent drainage scheduling method for urban roads as described above.

[0055] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. An Internet of Things-based intelligent urban road drainage scheduling method, characterized in that, The method comprises the following steps: Step S1: constructing a city road drainage Internet of Things sensing network in a target city road area, and collecting a multi-source heterogeneous data set corresponding to real-time rainfall data, road waterlogging data, drainage pipe network operation data, and pipe water pressure data in the target city road area based on the city road drainage Internet of Things sensing network; performing time-space alignment and noise filtering processing on the multi-source heterogeneous data set to generate a city standardized drainage monitoring data set; Step S2: dividing rainfall data, waterlogging data, drainage operation data, and pipe water pressure data corresponding to different city road segment subnets from the city standardized drainage monitoring data set based on the target city road area, and performing waterlogging increase rate and pipe network pressure bearing evaluation based on the rainfall data, waterlogging data, drainage operation data, and pipe water pressure data corresponding to different city road segment subnets to obtain city waterlogging increase rates and drainage pipe network pressure bearing degrees corresponding to different city road segment subnets; Step S3: calculating drainage pipe network negative pressure indexes corresponding to different city road segment subnets according to the city waterlogging increase rates and drainage pipe network pressure bearing degrees corresponding to different city road segment subnets; obtaining a city road network topology structure, and generating a city road area drainage pressure distribution map based on the drainage pipe network negative pressure indexes corresponding to different city road segment subnets and the city road network topology structure; Step S4: performing drainage hierarchical scheduling processing based on real-time drainage pressure differences between different city road segment subnets obtained based on the city road area drainage pressure distribution map to generate a city road drainage hierarchical scheduling strategy; issuing the city road drainage hierarchical scheduling strategy to drainage facilities in a corresponding area to perform drainage scheduling and feed back drainage effect data after scheduling, and returning to step S3 for closed-loop scheduling if the drainage effect data does not meet the requirements. 2.The Internet of Things based urban road intelligent drainage scheduling method according to claim 1, characterized in that, Step S1 comprises the following steps: Step S11: obtaining a corresponding city road planning layout and drainage pipe network distribution in a target city road area, and planning sensor device deployment points in corresponding waterlogging-prone road segments, drainage pipe network nodes, and low-lying areas in the target city road area based on the city road planning layout and the drainage pipe network distribution to form a corresponding drainage Internet of Things sensing network topology covering the target area; Step S12: constructing a corresponding city road drainage Internet of Things sensing network based on the drainage Internet of Things sensing network topology, and deploying corresponding multi-dimensional sensing devices at corresponding sensor device deployment points through the city road drainage Internet of Things sensing network to detect data in real time, including real-time rainfall data collected by a rainfall intensity sensor, road waterlogging data collected by a water level sensor, which contains waterlogging depth and waterlogging range of each road, drainage pipe network operation data collected by a flowmeter, which contains flow and flow rate in the drainage pipe network, and pipe water pressure data collected by a pressure sensor; Step S13: collecting real-time rainfall data, road waterlogging data, drainage pipe network operation data, and pipe water pressure data of each sensing device through the city road drainage Internet of Things sensing network, and performing real-time preprocessing by an edge computing node to filter out abnormal data to generate a corresponding multi-source heterogeneous data set; Step S14: Obtain the GPS positioning information and timestamp corresponding to each sensing device, and establish a space-time coordinate system based on the corresponding GPS positioning information and timestamp to map the corresponding multi-source heterogeneous data set to a unified space-time dimension, and generate a multi-source heterogeneous space-time alignment data set; Step S15: Perform noise filtering processing on the multi-source heterogeneous space-time alignment data set to identify and eliminate fluctuating noise points corresponding to device failure and electromagnetic interference using adaptive wavelet threshold denoising, and perform format standardization conversion to generate a city standardized drainage monitoring data set. 3.The Internet of Things based urban road intelligent drainage scheduling method according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Obtain the corresponding road network data and drainage pipe network GIS data in the target area of the city road, wherein the road network data includes road segment name, start and end node, lane number and road surface elevation, and the drainage pipe network GIS data includes pipe network direction, pipe diameter, inspection well position and pump station attribution, and the road segment-drainage pipe network spatial correlation of the target area is established based on the road network data and the drainage pipe network GIS data; Step S22: Obtain the road function attribute, drainage pipe network level and historical waterlogging risk level, wherein the road function attribute includes trunk road, secondary trunk road and branch road, and the drainage pipe network level includes main pipe, branch pipe and service pipe, and the city road target area is divided into each road segment subnetwork based on the road segment-drainage pipe network spatial correlation combined with the road function attribute, the drainage pipe network level and the historical waterlogging risk level to generate a plurality of target area road segment subnetworks, wherein each target area road segment subnetwork includes one or more continuous roads and corresponding drainage pipe network; Step S23: Based on the plurality of target area road segment subnetworks, the rainfall data, waterlogging data, drainage operation data and pipe water pressure data corresponding to different city road segment subnetworks are divided from the city standardized drainage monitoring data set, wherein the missing rate of each type of data corresponding to each city road segment subnetwork is calculated, and when the missing rate exceeds the preset threshold, the same type of data of the adjacent road segment subnetwork is used to complete the missing data through spatial difference value; Step S24: Based on the rainfall data, waterlogging data, drainage operation data and pipe water pressure data corresponding to different city road segment subnetworks, waterlogging increase rate and pipe network pressure bearing evaluation are performed to obtain the city waterlogging increase rate and drainage pipe network pressure bearing degree corresponding to different city road segment subnetworks.

4. The Internet of Things-based urban road intelligent drainage scheduling method according to claim 3, characterized in that, Step S24 includes the following steps: Step S241: Obtain the unit time rainfall, rainfall duration and rainfall recurrence section through the rainfall data corresponding to different city road segment subnetworks, obtain the waterlogging depth time sequence change and waterlogging range expansion speed through the waterlogging data corresponding to different monitoring points, obtain the pipe network real-time flow and pump station operation power through the drainage operation data, and obtain the static pressure value, dynamic pressure value and pressure fluctuation frequency of different nodes of the pipe through the pipe water pressure data, to generate the road drainage multi-dimensional features corresponding to different city road segment subnetworks; Step S242: Obtain the road surface slope and road surface permeability coefficient corresponding to different city road segment subnetworks, and calculate the rainfall infiltration amount based on the road surface slope and road surface permeability coefficient combined with the unit time rainfall, rainfall duration and rainfall recurrence section to obtain the road surface rainfall infiltration amount corresponding to different city road segment subnetworks; Step S243: Based on the real-time flow of the pipe network corresponding to different urban road segment subnets and the operating power of the pump station, the pipe network discharge is converted to obtain the pipe network discharge corresponding to different urban road segment subnets; Step S244: Based on the time series change of the water depth corresponding to different monitoring points and the water range expansion speed, the water expansion is calculated to obtain the road water area corresponding to different urban road segment subnets; Based on the road surface rainfall infiltration amount and the pipe network discharge corresponding to different urban road segment subnets, the corresponding urban water increase rate is calculated by combining the corresponding road water area, that is, urban water increase rate=(road surface rainfall infiltration amount-pipe network discharge) / road water area; Step S245: Based on the static pressure value, dynamic pressure value and pressure fluctuation frequency corresponding to different nodes in the pipe network in different urban road segment subnets, the pipe network pressure bearing evaluation is calculated by combining the operating power of the pump station to obtain the pressure bearing degree of the drainage pipe network corresponding to different urban road segment subnets. 5.The Internet of Things based urban road intelligent drainage scheduling method according to claim 4, characterized in that, The rainfall infiltration amount calculation based on the road surface slope and the road surface permeability coefficient in step S242 includes the following steps: The unit time rainfall is converted into corresponding hour rainfall intensity, and the corresponding rainfall recurrence section length of the current rainfall is determined based on the rainfall recurrence section. The rainfall retention probability is obtained, and the rainfall recurrence section length corresponding to the current rainfall is estimated based on the hour rainfall intensity and the rainfall retention probability to obtain the rainfall recurrence section attenuation coefficient. The corresponding road surface material type is obtained through the urban road segment subnet, and the urban road segment subnet is divided into multiple segmented boundary calculation units by combining the road surface slope. Based on the multiple segmented boundary calculation units, the rainfall infiltration component of each unit is calculated based on the road surface permeability coefficient, unit time rainfall, rainfall duration, road surface slope and rainfall recurrence section attenuation coefficient of each unit, and the sum is calculated to obtain the road surface rainfall infiltration amount corresponding to different urban road segment subnets. 6.The IoT-based urban road intelligent drainage scheduling method according to claim 4, characterized in that, The water expansion calculation based on the time series change of the water depth corresponding to different monitoring points and the water range expansion speed in step S244 includes the following steps: The latitude and longitude coordinates and the altitude corresponding to different monitoring points are obtained, and the vertical and horizontal direction decomposition of the water range expansion speed is performed based on the latitude and longitude coordinates and the altitude to obtain the corresponding water expansion speed components in the vertical and horizontal directions; Based on the time series change of the water depth corresponding to different monitoring points and combining the corresponding water expansion speed components in the vertical direction, the water depth change rate corresponding to the unit time is calculated, and the water submerged height corresponding to the water depth in the vertical direction at different monitoring points is determined based on the corresponding road surface elevation data; Obtain the road topology structure corresponding to each city road segment subnet, and determine whether there is a slope turning point and an obstacle blockage. If there is no obvious obstacle and the slope is gentle, the water level expansion is calculated in a circular or elliptical diffusion manner, and the corresponding water expansion speed component in the horizontal direction is multiplied by time to obtain the diffusion radius with the initial water accumulation point as the center. At the same time, the horizontal expansion range in this state is determined based on the diffusion radius. If there is a slope or an obstacle, the water level expansion is calculated along the slope direction. The expansion speed correction coefficient in the downward slope direction is greater than 1, the expansion speed correction coefficient in the upward slope direction is less than 1, and the obstacle area is marked as an expansion boundary constraint. The horizontal expansion range in this state is determined based on the expansion speed correction coefficient, the expansion boundary constraint and the water expansion speed component. The water area of different monitoring points in the vertical direction corresponding to the water submergence height and the horizontal direction corresponding to the horizontal expansion range is superimposed to obtain the road water area corresponding to different city road segment subnets. 7.The IoT-based intelligent urban road drainage scheduling method according to claim 4, characterized in that, Step S245 includes the following steps: Obtain the pipe design pressure value corresponding to the pipe in different city road segment subnets, and count the proportion of the number of times that the static pressure value corresponding to different nodes of the pipe exceeds the pipe design pressure value, to obtain the corresponding pressure overload frequency; Calculate the ratio between the maximum fluctuation difference of the dynamic pressure value corresponding to different nodes of the pipe and the pipe design pressure value, to obtain the corresponding pressure fluctuation amplitude; Based on the pressure fluctuation frequency corresponding to different nodes of the pipe in different city road segment subnets, the running power of the corresponding pump station is calculated to obtain the drainage pipe network life attenuation amplitude, and the pressure overload frequency and the pressure fluctuation amplitude are combined for pipe network pressure bearing evaluation calculation, to obtain the drainage pipe network pressure bearing degree corresponding to different city road segment subnets. 8.The Internet of Things based urban road intelligent drainage scheduling method according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Based on the city water increase rate corresponding to different city road segment subnets, the real-time water increase rate, the water increase rate time sequence slope and the deviation rate of the water increase rate and the historical same period average value are obtained, and the water increase rate risk contribution degree is obtained by multiplying the water increase rate time sequence slope by the real-time water increase rate and dividing by the deviation rate of the water increase rate and the historical same period average value; Step S32: Based on the drainage pipe network pressure bearing degree corresponding to different city road segment subnets, the pipe network pressure bearing risk contribution degree is calculated, wherein the pipe network pressure bearing risk contribution degree = 1-drainage pipe network pressure bearing degree / preset pressure bearing threshold; Step S33: Based on the water increase rate risk contribution degree and the pipe network pressure bearing risk contribution degree corresponding to different city road segment subnets, the drainage pipe network negative pressure index corresponding to different city road segment subnets is calculated; Step S34: Obtain the city road network topology structure, and generate a city road area drainage pressure distribution map based on the drainage pipe network negative pressure index corresponding to different city road segment subnets and the city road network topology structure. 9.The Internet of Things based urban road intelligent drainage scheduling method according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Based on the city road area drainage pressure distribution map, the drainage pressure difference between two adjacent city road segment subnets is calculated to obtain the real-time drainage pressure difference between different city road segment subnets. Step S42: Obtain the corresponding drainage pipe network connection relationship between each two adjacent urban road section subnets, and determine the corresponding pressure conduction path and schedulable corresponding drainage flow overflow value based on the corresponding drainage pipe network connection relationship between each two adjacent urban road section subnets; Step S43: Based on the pressure conduction path and schedulable corresponding drainage flow overflow value, and combined with the corresponding real-time drainage pressure difference between different urban road section subnets, the drainage hierarchical scheduling processing is performed, when the real-time drainage pressure difference exceeds the preset threshold value, it is determined as a first emergency scheduling, and combined with the pressure conduction path and schedulable corresponding drainage flow overflow value, the rapid pressure relief is taken as the target, including starting the emergency pump station of the high pressure difference subnet, opening the communication pipe gate to realize cross-subnet shunting, temporarily shutting down unnecessary water facilities in the low pressure difference subnet to release the drainage capacity; when the real-time drainage pressure difference is between the preset threshold value and the safety threshold value, it is determined as a second relief scheduling, and the pressure balance is taken as the target, including adjusting the operation frequency of the pump station therein, optimizing the gate opening degree to control the drainage flow distribution, and preferentially dredging the pipe network blockage hidden danger in the pressure difference direction; when the real-time drainage pressure difference is lower than the safety threshold value, it is determined as a third maintenance scheduling, and the stable operation is taken as the target, including maintaining the basic operation parameters of the existing drainage facilities, strengthening the waterlogging monitoring frequency of the key road section, and pre-storing the emergency scheduling resources, to generate the urban road drainage hierarchical scheduling strategy; Step S44: The urban road drainage hierarchical scheduling strategy is issued to the corresponding regional drainage facilities through the Internet of Things encryption communication link to execute the drainage scheduling and feedback the drainage effect data after scheduling, if the drainage effect data does not meet, return to step S3 to execute the closed-loop scheduling, to adjust the pressure difference calculation based on the new drainage pressure distribution map to generate the optimized scheduling strategy and execute again, until the drainage effect meets.

10. An Internet of Things-based intelligent urban road drainage scheduling system, characterized in that, The computer program stored on the memory and executable on the processor is used to execute the intelligent urban road drainage scheduling method based on the Internet of Things. The computer program stored on the memory and executable on the processor is used to execute the intelligent urban road drainage scheduling method based on the Internet of Things.

Images