Intelligent control method and system for rainwater overflow and reutilization of viaduct

Abstract

The invention provides an intelligent control method and system for rainwater overflow and reutilization of a viaduct, and relates to the field of self-adaptive control, and the method comprises the steps: carrying out the intelligent control of the rainwater overflow and reutilization of the viaduct based on the historical rainwater recovery data of a plurality of porous drainage pipes and the historical irrigation data of a green belt of the viaduct; the arrangement of the plurality of rainwater storage assemblies, the communication relation between the plurality of porous drainage pipes and the plurality of rainwater storage assemblies, the arrangement of the plurality of greening irrigation assemblies and the communication relation between the plurality of rainwater storage assemblies and the plurality of greening irrigation assemblies are optimized; according to the communication relation between the multi-hole drainage pipes and the rainwater storage assemblies and the real-time rainwater recycling data of the multi-hole drainage pipes, the operation of the rainwater storage assemblies and the opening degree of the overflow valve are controlled; according to the real-time soil state data of the green belt and the communication relation between the multiple rainwater storage assemblies and the multiple greening irrigation assemblies, the multiple greening irrigation assemblies are controlled to operate, and the system has the advantages that cyclic utilization of water resources is achieved, and the pressure of a municipal pipe network is reduced.

Description

Technical Field

[0001] The present invention relates to the field of adaptive control, and in particular to an intelligent control method and system for rainwater overflow and reuse on viaducts. Background Art

[0002] With the rapid growth of urban population density, urban traffic congestion is becoming more and more serious. In order to improve the traffic capacity of urban areas, save space and reduce construction costs, more and more cities are starting to build viaducts in urban areas.

[0003] In the existing technology, the rainwater system of the viaduct mostly adopts a drainage system, which leads the rainwater to the drainage system laid under the ground auxiliary road through the rainwater pipe. On the one hand, the green plants in the green belt under the viaduct lack rainwater irrigation due to the obstruction of the viaduct, and often require additional municipal water irrigation, which increases the cost of labor and water use and does not fully utilize the rainwater on rainy days; on the other hand, the rainwater on the bridge deck directly flows into the road drainage system, and the rainwater on the bridge is not fully utilized, resulting in a waste of natural resources and increasing the pressure on the municipal pipeline network.

[0004] Therefore, it is necessary to provide an intelligent control method and system for rainwater overflow and reuse on elevated bridges to achieve the recycling of water resources and reduce the pressure on the municipal pipeline network. Summary of the Invention

[0005] The present invention provides an intelligent control system for rainwater overflow and reuse of an elevated bridge, which is applied to a rainwater recycling and reuse system arranged on an elevated bridge. The rainwater recycling and reuse system includes an elevated bridge rainwater recycling component, multiple rainwater storage components and multiple greening irrigation components. The elevated bridge rainwater recycling component includes multiple porous drainage pipes, an overflow valve is provided between the rainwater storage component and the municipal rainwater pipe, and an irrigation pipe is provided between the rainwater storage component and the greening irrigation component. The system includes: a data acquisition module for collecting historical rainwater recycling data of the multiple porous drainage pipes and historical soil status data of the green belt of the elevated bridge; a structure optimization ... The historical irrigation data of the green belt of the bridge is used to optimize the settings of multiple rainwater storage components, the connectivity relationship between multiple porous drainage pipes and multiple rainwater storage components, the settings of multiple green irrigation components, and the connectivity relationship between multiple rainwater storage components and multiple green irrigation components; the overflow control module includes multiple overflow control units, each rainwater storage component corresponds to an overflow control unit, and the overflow control unit is used to control the operation of the rainwater storage components and the opening of the overflow valve according to the connectivity relationship between the multiple porous drainage pipes and the multiple rainwater storage components and the real-time rainwater recovery data of the multiple porous drainage pipes; the irrigation control module is used to control the operation of multiple green irrigation components according to the real-time soil status data of the green belt.

[0006] Furthermore, the historical rainwater recovery data of the porous drainage pipes include the water levels and flow rates of the multiple porous drainage pipes at multiple historical rainfall time points during multiple historical rainfall processes; the historical soil state data of the green belt of the viaduct include the soil temperature and humidity of multiple soil monitoring points at multiple historical time points; the structural optimization module optimizes the settings of multiple rainwater storage components, the connectivity relationship between the multiple porous drainage pipes and the multiple rainwater storage components, the settings of multiple greening irrigation components, and the connectivity relationship between the multiple rainwater storage components and the multiple greening irrigation components based on the historical rainwater recovery data of the multiple porous drainage pipes and the historical irrigation data of the green belt of the viaduct, including: Based on the historical rainwater recovery data of porous drainage pipes, the rainwater recovery-related characteristics of multiple porous drainage pipes are determined; based on the soil temperature and humidity of multiple soil monitoring points at multiple historical time points, the green belt of the viaduct is divided into multiple irrigation areas, and multiple greening irrigation components are set up; based on the historical irrigation data of multiple greening irrigation components, the irrigation demand-related characteristics of multiple irrigation areas are determined; based on the rainwater recovery-related characteristics of multiple porous drainage pipes and the irrigation demand-related characteristics of multiple irrigation areas, the setting of multiple rainwater storage components, the connectivity relationship between multiple porous drainage pipes and multiple rainwater storage components, and the connectivity relationship between multiple rainwater storage components and multiple greening irrigation components are optimized.

[0007] Furthermore, the rainwater recovery related characteristics of the multiple porous drainage pipes include rainwater recovery correlation coefficients of any two porous drainage pipes; and the irrigation demand related characteristics of the multiple irrigation areas include irrigation demand correlation coefficients of any two irrigation areas.

[0008] Furthermore, the structural optimization module optimizes the settings of multiple rainwater storage components, the connectivity relationships between multiple porous drainage pipes and multiple rainwater storage components, and the connectivity relationships between multiple rainwater storage components and multiple greening irrigation components based on the rainwater recovery-related characteristics of multiple porous drainage pipes and the irrigation demand-related characteristics of multiple irrigation areas, including: optimizing the settings of multiple rainwater storage components, the connectivity relationships between multiple porous drainage pipes and multiple rainwater storage components, and the connectivity relationships between multiple rainwater storage components and multiple greening irrigation components based on the rainwater recovery-related characteristics of multiple porous drainage pipes and the irrigation demand-related characteristics of multiple irrigation areas through an improved particle swarm algorithm.

[0009] Furthermore, based on the rainwater recovery related characteristics of multiple porous drainage pipes and the irrigation demand related characteristics of multiple irrigation areas, the settings of multiple rainwater storage components, the connectivity relationship between multiple porous drainage pipes and multiple rainwater storage components, and the connectivity relationship between multiple rainwater storage components and multiple greening irrigation components are optimized through an improved particle swarm algorithm, including: determining multiple rainfall simulation scenarios based on the historical rainwater recovery data of the porous drainage pipes; determining multiple irrigation demand scenarios based on the historical irrigation data of multiple greening irrigation components; based on multiple rainfall simulation scenarios and multiple irrigation demand scenarios; based on the rainwater recovery related characteristics of multiple porous drainage pipes, the irrigation demand related characteristics of multiple irrigation areas, multiple rainfall simulation scenarios and multiple irrigation demand scenarios, the settings of multiple rainwater storage components, the connectivity relationship between multiple porous drainage pipes and multiple rainwater storage components, and the connectivity relationship between multiple rainwater storage components and multiple greening irrigation components are optimized through an improved particle swarm algorithm.

[0010] Furthermore, the viaduct rainwater recovery component also includes a fully transparent pavement arranged on both sides of the viaduct and a pebble layer arranged under the fully transparent pavement, a permeable geotextile is arranged between the fully transparent pavement and the pebble layer, and a porous drainage pipe is laid on the pebble layer; the rainwater storage component includes a distribution pool, a water storage tank and a vacuum device connected to the connecting pipe, an overflow plate is provided in the water storage tank, and the overflow plate is used to divide the space inside the water storage tank into a first water storage area and a second water storage area, the first water storage area is located above the distribution pool, a filling area is provided in the distribution pool, the first water storage area is connected to the filling area, and the vacuum equipment is provided on the water storage tank; the first water storage area is provided with a first water level sensor, and the second water storage area is provided with a second water level sensor.

[0011] Furthermore, the overflow control unit controls the operation of the rainwater storage assembly according to the connectivity relationship between the multiple porous drainage pipes and the multiple rainwater storage assemblies and the real-time rainwater recovery data of the multiple porous drainage pipes, including: judging whether to turn on the vacuum device according to the water level of the first water storage area collected by the first water level sensor at multiple consecutive time points; predicting the water level of the second water storage area at multiple future time points according to the real-time rainwater recovery data of the porous drainage pipes connected to the rainwater storage assembly and the water level of the second water storage area collected by the second water level sensor at multiple consecutive time points; and determining the opening time of the overflow valve according to the predicted water level of the second water storage area at multiple future time points.

[0012] Furthermore, the overflow control unit controls the opening of the overflow valve based on the connectivity between the multiple porous drainage pipes and the multiple rainwater storage components and the real-time rainwater recovery data of the multiple porous drainage pipes, including: determining the initial opening of the overflow valve based on the predicted water levels of the second water storage area at multiple future time points; after the overflow valve is opened, adjusting the opening of the overflow valve based on the real-time rainwater recovery data of the porous drainage pipes connected to the rainwater storage components and the water level of the second water storage area collected by the second water level sensor at multiple consecutive time points.

[0013] Furthermore, the distribution tank is connected to a municipal sewage pipe on the side away from the filling area, and a sewage valve is provided between the distribution tank and the municipal sewage pipe; the overflow control unit is also used to: determine whether to open the sewage valve based on the water level of the first water storage area collected by the first water level sensor at multiple consecutive time points.

[0014] The present invention provides an intelligent control method for rainwater overflow and reuse of an elevated bridge, which is applied to the above-mentioned intelligent control system for rainwater overflow and reuse of an elevated bridge, comprising: collecting historical rainwater recovery data of multiple porous drainage pipes and historical soil status data of the green belt of the elevated bridge; optimizing the arrangement of multiple rainwater storage components, the connectivity relationship between multiple porous drainage pipes and multiple rainwater storage components, the arrangement of multiple greening irrigation components, and the connectivity relationship between multiple rainwater storage components and multiple greening irrigation components based on the historical rainwater recovery data of the multiple porous drainage pipes and multiple rainwater storage components and the real-time rainwater recovery data of the multiple porous drainage pipes, controlling the operation of the rainwater storage components and the opening of the overflow valve; and controlling the operation of the multiple greening irrigation components based on the real-time soil status data of the green belt and the connectivity relationship between the multiple rainwater storage components and the multiple greening irrigation components.

[0015] Compared with the existing technology, the intelligent control method and system for rainwater overflow and reuse of viaducts provided by the present invention have at least the following beneficial effects: By real-time monitoring of the rainwater recovery flow rate and water level of the rainwater storage components in the porous drainage pipes, the overflow valve opening is dynamically adjusted to ensure that rainwater is preferentially collected by the storage components and overflows into the municipal pipe network only when storage capacity is reached. Based on historical rainwater recovery data, the layout of the storage components and the connectivity of the drainage pipes are optimized to improve the overall rainwater collection efficiency of the system, reduce ineffective overflows, and alleviate the load on the urban drainage system. Each rainwater storage component is equipped with an independent overflow control unit that can adjust the valve opening based on local rainfall intensity, storage water level, and drainage pipe flow, avoiding resource waste or overflow risks caused by "one-size-fits-all" control. Through a closed-loop design of data perception, structural optimization, and intelligent control, the elevated bridge achieves efficient collection, storage, and reuse of rainwater resources, while simultaneously addressing green irrigation needs and urban drainage safety, providing a replicable and scalable intelligent solution for urban stormwater management and green infrastructure construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] This specification will be further described in the form of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, like numbers represent like structures, wherein: Figure 1 is a schematic structural diagram of an elevated bridge rainwater storage device according to some embodiments of this specification; Figure 2 is a schematic structural diagram of an elevated bridge rainwater recycling assembly according to some embodiments of this specification; Figure 3 is a schematic structural diagram of a rainwater storage assembly according to some embodiments of this specification; Figure 4 is a module diagram of an intelligent control system for rainwater overflow and reuse on an elevated bridge according to some embodiments of this specification; Figure 5 It is a flow chart of an intelligent control method for rainwater overflow and reuse on an elevated bridge according to some embodiments of this specification.

[0017] In the figure, 1. Overpass; 2. Overpass rainwater recovery component; 21. Fully permeable pavement; 22. Pebble layer; 23. Permeable geotextile; 24. Porous drainage pipe; 3. Rainwater storage component; 31. Distribution pool; 311. Filling area; 32. Water storage tank; 321. First water storage area; 322. Second water storage area; 323. Overflow plate; 33. Vacuum equipment; 4. Connecting pipe; 5. Municipal rainwater pipe; 6. Municipal sewage pipe; 7. Irrigation water pipe. DETAILED DESCRIPTION

[0018] To more clearly illustrate the technical solutions of the embodiments of this specification, the following briefly describes the drawings required for describing the embodiments. Obviously, the drawings described below are merely examples or embodiments of this specification. Those skilled in the art can apply this specification to other similar scenarios based on these drawings without inventive effort. Unless otherwise apparent from the context or otherwise noted, the same reference numerals in the figures represent the same structure or operation.

[0019] Figure 1 is a schematic structural diagram of an elevated bridge rainwater storage device according to some embodiments of this specification, such as Figure 1 As shown, the viaduct rainwater storage equipment includes an viaduct rainwater recovery component 2 and multiple rainwater storage components 3.

[0020] Figure 2 is a schematic structural diagram of an elevated bridge rainwater recovery assembly 2 according to some embodiments of this specification, such as Figure 2 As shown, the viaduct rainwater recovery component 2 includes a plurality of porous drainage pipes 24. The viaduct rainwater recovery component 2 also includes a fully transparent pavement 21 arranged on both sides of the viaduct 1 and a pebble layer 22 arranged under the fully transparent pavement 21. A permeable geotextile 23 is arranged between the fully transparent pavement 21 and the pebble layer 22. The porous drainage pipes 24 are laid on the pebble layer 22. The porous drainage pipes 24 are provided with a plurality of water inlet holes on the side facing the fully transparent pavement 21.

[0021] Figure 3 is a schematic structural diagram of a rainwater storage assembly 3 according to some embodiments of this specification, such as Figure 3 As shown, a discharge valve is installed between the rainwater storage assembly 3 and the municipal rainwater pipe 5. The rainwater storage assembly 3 includes a distribution tank 31 connected to the connecting pipe 4, a water storage tank 32, and a vacuum device 33. An overflow plate 323 is installed in the water storage tank 32, which divides the space inside the water storage tank 32 into a first water storage area 321 and a second water storage area 322. The first water storage area 321 is located above the distribution tank 31. The distribution tank 31 is provided with a filling area 311, and the first water storage area 321 and the filling area 311 are connected. The vacuum device 33 is installed on the water storage tank 32. A first water level sensor is installed in the first water storage area 321, and a second water level sensor is installed in the second water storage area 322. The side of the distribution tank 31 away from the filling area 311 is also connected to the municipal sewage pipe 6. A drain valve is installed between the distribution tank 31 and the municipal sewage pipe 6. An irrigation water pipe 7 is installed between the water storage tank 32 and the landscaping irrigation assembly.

[0022] The operating principle of the elevated bridge rainwater storage system is as follows: Rainwater from the viaduct 1's deck drains through the road's cross slope onto the permeable pavement at the roadside. Under the influence of the permeable pavement, the rainwater enters a porous drainage pipe 24 located beneath the permeable pavement. Rainwater from the viaduct 1 flows through the drainage pipe into a distribution tank 31. High-density impurities settle at the bottom of the distribution tank 31, while low-density impurities are intercepted by a filler area 311. A permeable geotextile 23 is placed between the filler area 311 and the water storage tank 32. After passing through the distribution tank 31 and filler area 311, the rainwater enters the first water storage area 321. When the water level in the first water storage area 321 rises to a certain level, the vacuum system 33 at the top of the water storage tank 32 activates. Due to the reduced atmospheric pressure above, the rainwater enters the first water storage area 321 more rapidly under the influence of atmospheric pressure. When the water level in the first water storage area 321 reaches the level of the overflow plate 323, the clean rainwater flows over the overflow plate 323 and enters the second water storage area 322. A hose is connected to the bottom of the second water storage area 322, allowing clean rainwater from the second water storage area 322 to be used for irrigation during dry periods on sunny days. Furthermore, a rainwater overflow pipe is located at the bottom of the second water storage area 322. Once the water depth exceeds a certain level, excess rainwater flows through the overflow pipe into the municipal rainwater system for discharge. When water is needed on a sunny day, the drain valve connected to the distribution tank 31 is opened, breaking the vacuum in the water storage tank 32. The rainwater stored in the first water storage area 321 quickly flushes the filling area 311 and the distribution tank 31, removing any accumulated impurities. The rainwater stored in the second water storage area 322 remains unaffected and can be used for irrigation.

[0023] Figure 4 is a module diagram of an intelligent control system for rainwater overflow and reuse of an elevated bridge according to some embodiments of this specification, such as Figure 4 As shown, the intelligent control system for rainwater overflow and reuse of viaducts can include a data acquisition module, a structure optimization module, an overflow control module and an irrigation control module.

[0024] The data collection module is used to collect historical rainwater recovery data from multiple porous drainage pipes and historical soil status data from the green belts of the viaduct.

[0025] Specifically, historical rainwater recovery data from porous drain pipes includes water levels and flow rates at multiple historical rainfall points during multiple rainfall events. Water level sensors and flow rate sensors are installed in these pipes to monitor changes in water level and flow rate. Water level sensors can be float, pressure, or ultrasonic, accurately measuring water levels within the pipes. Flow rate sensors utilize principles such as electromagnetic induction and the ultrasonic Doppler effect to measure the flow rate of rainwater.

[0026] The historical soil condition data for the viaduct's green belt includes soil temperature and humidity data from multiple soil monitoring points over multiple historical time periods. For each historical time period, soil temperature and humidity data can be obtained at multiple historical time points within that time period. The time difference between the start time point of the historical time period and the completion time of the most recent irrigation session must be greater than a time difference threshold, for example, 1 hour.

[0027] The structural optimization module is used to optimize the setting of multiple rainwater storage components, the connectivity between multiple porous drainage pipes and multiple rainwater storage components, the setting of multiple greening irrigation components, and the connectivity between multiple rainwater storage components and multiple greening irrigation components based on the historical rainwater recovery data of multiple porous drainage pipes and the historical irrigation data of the green belts of the viaduct.

[0028] Specifically include: Based on historical rainwater recovery data of porous drain pipes, determine the rainwater recovery related characteristics of multiple porous drain pipes; Based on the soil temperature and humidity at multiple soil monitoring points at multiple historical time points, the green belt of the viaduct is divided into multiple irrigation areas and multiple green irrigation components are set up; Determine irrigation demand-related characteristics for multiple irrigation areas based on historical irrigation data from multiple green irrigation components; Based on the rainwater recovery-related characteristics of multiple porous drainage pipes and the irrigation demand-related characteristics of multiple irrigation areas, the settings of multiple rainwater storage components, the connectivity between multiple porous drainage pipes and multiple rainwater storage components, and the connectivity between multiple rainwater storage components and multiple greening irrigation components are optimized.

[0029] Specifically, the rainwater recovery related characteristics of the plurality of porous drainage pipes include a rainwater recovery correlation coefficient between any two porous drainage pipes.

[0030] For any two porous drainage pipes, the water levels of the two pipes at multiple historical rainfall points can be substituted into a correlation coefficient calculation formula (e.g., Pearson correlation coefficient, Spearman rank correlation coefficient, etc.) to obtain the water level correlation coefficients for the two pipes corresponding to the historical rainfall. The flow velocities of the two pipes at multiple historical rainfall points can be substituted into a correlation coefficient calculation formula (e.g., Pearson correlation coefficient, Spearman rank correlation coefficient, etc.) to obtain the flow velocity correlation coefficients for the two pipes corresponding to the historical rainfall. The water level correlation coefficients for each historical rainfall event in the two pipes can be averaged to obtain the mean water level correlation coefficients. The flow velocity correlation coefficients for each historical rainfall event in the two pipes can be averaged to obtain the mean flow velocity correlation coefficients for the two pipes. The rainwater recovery correlation coefficients for the two pipes can be calculated by taking the weighted sum of the mean water level correlation coefficients and the mean flow velocity correlation coefficients.

[0031] Based on the soil temperature and moisture of multiple soil monitoring points at multiple historical time points, the Euclidean distance of soil temperature and moisture between any two soil monitoring points is calculated. Specifically, the Euclidean distance of soil temperature and soil moisture between any two soil monitoring points can be calculated based on the soil temperature and moisture of the multiple soil monitoring points at multiple historical time points. The Euclidean distance of soil temperature and soil moisture between the two soil monitoring points is weighted and summed to obtain the Euclidean distance of soil temperature and moisture between the two soil monitoring points. A clustering algorithm (e.g., K-Means, hierarchical clustering, etc.) is used to cluster the multiple soil monitoring points based on the Euclidean distance of soil temperature and moisture between any two soil monitoring points to obtain multiple soil monitoring point clusters. It is understood that the smaller the Euclidean distance of soil temperature and moisture between two soil monitoring points, the greater the probability that the two soil monitoring points are assigned to the same soil monitoring point cluster. For each soil monitoring point cluster, the irrigation area corresponding to the soil monitoring point cluster is determined based on the monitoring area covered by the soil monitoring points included in the cluster (e.g., a circular area with a radius of 1 meter). Each irrigation area corresponds to a soil monitoring point cluster among the multiple irrigation areas. A greening irrigation component can be set at the center of the irrigation area corresponding to the soil monitoring point cluster, and the parameters of the greening irrigation component can be adjusted according to the size of the irrigation area corresponding to the soil monitoring point cluster. For example, the nozzle range can be adjusted so that the greening irrigation component can irrigate most (for example, 90%) of the irrigation area corresponding to the soil monitoring point cluster.

[0032] The irrigation demand related characteristics of the plurality of irrigation areas include an irrigation demand correlation coefficient between any two irrigation areas.

[0033] The historical irrigation data of multiple green irrigation components can include the water consumption of each green irrigation component for irrigation over multiple historical time periods. The water consumption of the green irrigation components for irrigation over the historical time periods can be obtained through instantaneous flow integration (e.g., real-time data collection via a flow meter). The green irrigation components use an automatic sensing method for irrigation. That is, if the temperature and humidity of the soil in the irrigation area where the green irrigation component is located (e.g., the soil 10 cm below the surface) meet the irrigation conditions (e.g., temperature > 5°C and soil humidity < 25%), the green irrigation component automatically performs a watering operation.

[0034] For any two irrigation areas, the water consumption of the two greening irrigation components for irrigation in multiple historical time periods can be substituted into the correlation coefficient calculation formula (for example, Pearson correlation coefficient, Spearman rank correlation coefficient, etc.) to calculate the irrigation demand correlation coefficient of the two irrigation areas.

[0035] In some embodiments, the structural optimization module optimizes the arrangement of the plurality of rainwater storage assemblies, the connectivity between the plurality of porous drainage pipes and the plurality of rainwater storage assemblies, and the connectivity between the plurality of rainwater storage assemblies and the plurality of landscaping irrigation assemblies based on rainwater recovery-related characteristics of the plurality of porous drainage pipes and irrigation demand-related characteristics of the plurality of irrigation areas, including: Through an improved particle swarm optimization algorithm based on the rainwater recovery-related characteristics of multiple porous drainage pipes and the irrigation demand-related characteristics of multiple irrigation areas, the settings of multiple rainwater storage components, the connectivity between multiple porous drainage pipes and multiple rainwater storage components, and the connectivity between multiple rainwater storage components and multiple greening irrigation components are optimized.

[0036] Specifically, the following processes may be involved: S11. Based on rainwater recovery-related characteristics of the multiple porous drainage pipes and irrigation demand-related characteristics of the multiple irrigation areas, generate a particle swarm, where each particle represents a structural optimization scheme, wherein the structural optimization scheme includes the number of rainwater storage components, the capacity of each rainwater storage component, the connectivity relationship between the multiple porous drainage pipes and the multiple rainwater storage components, and the connectivity relationship between the multiple rainwater storage components and the multiple greening irrigation components; S12, initialize the particle speed; S13, establishing a multi-objective fitness function; S14. Calculate the multi-objective fitness value of each particle through the multi-objective fitness function; S14. For each particle, calculate a position guidance value of the particle based on rainwater recovery-related characteristics of the multiple porous drainage pipes, irrigation demand-related characteristics of the multiple irrigation areas, and the multi-objective fitness value of the particle; S15. Determine the global optimal position and the historical optimal position of each particle based on the position guidance value of each particle; S16, determining whether the end condition is met; if so, optimizing the arrangement of the multiple rainwater storage components, the connectivity between the multiple porous drainage pipes and the multiple rainwater storage components, and the connectivity between the multiple rainwater storage components and the multiple greening irrigation components based on the particle with the largest multi-objective fitness value in the current particle swarm; if not, executing S17; S17. Update the position and velocity of each particle based on the global optimal position and the historical optimal position of each particle, and execute S14.

[0037] Specifically, each particle represents a structural optimization solution and needs to encode the following information: Number of rainwater storage components N s , which can be a sampling value within a preset number range (for example, 5-10, etc.), wherein the preset number range can be determined based on the historical rainwater recovery data of the porous drain pipe. If the historical rainwater recovery volume of the porous drain pipe is large, the upper and lower limits of the preset number can be increased.

[0038] The connectivity relationship between the porous drainage pipe and the storage component (represented by matrix A, A j,i =1 indicates that the porous drainage pipe j is connected to the rainwater storage component i).

[0039] The connectivity relationship between the storage component and the irrigation component (represented by matrix B, B i,k =1 indicates that rainwater storage component i is connected to greening irrigation component k).

[0040] The generation of particles needs to meet the following constraints: 1. Each porous drainage pipe must be connected to a rainwater storage component; 2. Each green irrigation component must be connected to a rainwater storage component; 3. Each rainwater storage component must be connected to at least one porous drainage pipe and at least one green irrigation component; 4. The rainwater recovery correlation coefficient of any two porous drainage pipes connected to the same rainwater storage assembly is less than a preset rainwater recovery correlation coefficient threshold (e.g., 0.5), where the preset rainwater recovery correlation coefficient threshold can be determined based on experimental data; 5. The irrigation demand correlation coefficient of any two greening irrigation components connected to the same rainwater storage component is less than a preset irrigation demand correlation coefficient threshold (for example, 0.5, etc.), wherein the preset irrigation demand correlation coefficient threshold can be determined through experimental data.

[0041] As can be understood, Constraint 1 ensures that rainwater from all drainage pipes is recycled, avoiding resource waste. Constraint 2 guarantees the water needs of all irrigation zones, avoiding irrigation blind spots. Constraint 3 prevents "idle" storage modules (connected only to drainage pipes or only to irrigation modules), ensuring efficient use of system resources. Constraint 4 distributes the rainwater collection from highly correlated drainage pipes to prevent overflowing of the rainwater storage module due to simultaneous surges in rainwater collection. That is, if the collection rates of two drainage pipes are highly correlated, their collection rates will peak simultaneously. If they are connected to the same rainwater storage module, this may cause excessive overflow of that module, reducing rainwater collection efficiency. By limiting the correlation coefficient of rainwater collection and assigning low-correlation porous drainage pipes to the same module, the temporal distribution of collection rates can be balanced and the overflow probability reduced. Constraint 5 distributes the water demand of highly correlated irrigation modules to prevent depletion of rainwater storage modules due to simultaneous water shortages. That is, if the water demands of two irrigation zones are highly correlated, their water demand will surge simultaneously. If they are connected to the same rainwater storage module, this may cause rapid depletion of that module, affecting irrigation effectiveness. By limiting the irrigation demand correlation coefficient and allocating low-correlation irrigation areas to the same irrigation demand correlation coefficient, water demand can be smoothed and the water supply time of the rainwater storage component can be extended.

[0042] The velocity of a particle has the same dimensions as the particle's position vector, including: Number of rainwater storage components Ns The amount of adjustment; Changes in the elements of the connectivity matrix A between the porous drain pipe and the storage component; Changes in the elements of the connectivity matrix B between the storage component and the irrigation component.

[0043] Randomly generate velocity values ​​within a reasonable range (e.g., 1-2) and the number of rainwater storage components. Ns ; For discrete variables (such as connectivity relationships A, B), the connection state is randomly flipped according to probability.

[0044] A multi-objective fitness function needs to balance multiple objectives, including: 1. Cost minimization: System construction and maintenance costs are related to the number of rainwater storage components; 2. Maximize efficiency: Rainwater utilization rate (such as the ratio of effective irrigation water to rainfall); 3. Maximizing reliability: The ability to guarantee water supply for green irrigation (e.g., the matching degree between the total capacity of storage components and demand); 4. Storage balance: The balance of rainwater stored in each rainwater storage component can be represented by the standard deviation of the amount of rainwater stored in the rainwater storage component.

[0045] In some embodiments, the improved particle swarm algorithm calculates the multi-objective fitness value of particles, including: Determine multiple rainfall simulation scenarios based on historical rainwater recovery data from porous drains; Determine multiple irrigation demand scenarios based on historical irrigation data of multiple green irrigation components; Based on multiple rainfall simulation scenarios and multiple irrigation demand scenarios, the multi-objective fitness values ​​of particles are calculated.

[0046] Specifically, key features (such as total rainfall, peak intensity, and rainfall distribution type) are extracted from each rainfall event. K-means or hierarchical clustering is used to classify rainfall events into several categories (such as "short-term heavy rainfall," "long-term light rainfall," and "uniform rainfall"). One rainfall event from each category is selected as a simulation scenario, or synthetic scenarios are generated through statistical methods (such as Monte Carlo simulation). Multiple sets of discrete rainfall simulation scenarios are obtained, for example: Rainfall simulation scenario 1: short-term heavy rainfall (total rainfall 50 mm, peak intensity 20 mm / h, lasting 1 hour); Rainfall simulation scenario 2: long light rain (total rainfall 30 mm, evenly distributed, lasting 6 hours); Rainfall simulation scenario 3: extreme rainfall (total rainfall 100 mm, peak intensity 40 mm / h, lasting 2 hours).

[0047] Use K-means or hierarchical clustering to cluster the historical time periods based on the water consumption of each greening irrigation component over multiple time periods. The time periods are then divided into several categories (e.g., high water demand, medium water demand, low water demand, etc.). For each category, select one historical time period as a simulation scenario, or use statistical methods (e.g., Monte Carlo simulation) to generate synthetic scenarios. This yields multiple irrigation demand scenarios, each of which can include the water consumption required for each greening irrigation component.

[0048] Build a simulation model of the rainwater recycling system in simulation software (e.g., SWMM (Storm Water Management Model)). For each rainfall scenario, the simulation model can determine the total amount of rainwater overflow from the particles and the water level of each rainwater storage component at the end of the rainfall.

[0049] For each rainwater storage component, the average of the water levels of the rainwater storage component at the end of rainfall in each rainfall simulation scenario is calculated as the initialization water level of the rainwater storage component.

[0050] For each irrigation demand scenario, the simulation model is updated based on the initial water level of the rainwater storage component. For each irrigation demand scenario, the updated simulation model is used to determine the total irrigation water shortage for the particle under each irrigation demand scenario and the water level of each rainwater storage component at the end of the irrigation demand scenario.

[0051] The multi-objective fitness value of the particle is calculated based on the total rainwater overflow of the particle in each rainfall simulation scenario and the water level of each rainwater storage component at the end of the rainfall, the total irrigation water shortage of the particle in each irrigation demand scenario and the water level of each rainwater storage component at the end of the irrigation demand scenario.

[0052] Specifically, the cost of a single rainwater storage component may be multiplied by the number of rainwater storage components corresponding to a particle to obtain the cost value corresponding to the particle.

[0053] For each rainfall simulation scenario, the ratio of the total rainfall in the simulation scenario minus the total rainwater overflow of the particle in that simulation scenario to the total rainfall in the simulation scenario can be calculated as the rainwater recovery efficiency for that rainfall simulation scenario. The rainwater recovery efficiency for each rainfall simulation scenario is averaged to obtain the average rainwater recovery efficiency for the particle.

[0054] For each rainfall simulation scenario, the standard deviation of the water level of each rainwater storage component at the end of the rainfall is calculated to obtain the water level standard deviation corresponding to the rainfall simulation scenario. The water level standard deviation corresponding to each rainfall simulation scenario is averaged to obtain the first water level standard deviation mean.

[0055] For each irrigation demand scenario, the total irrigation water shortage of the particles under each irrigation demand scenario is averaged to obtain the mean irrigation water shortage.

[0056] For each irrigation demand scenario, the standard deviation of the water level of each rainwater storage component at the end of the irrigation demand scenario is calculated to obtain the water level standard deviation corresponding to the irrigation demand scenario. The mean of the water level standard deviations corresponding to each irrigation demand scenario is calculated to obtain the second water level standard deviation mean.

[0057] The cost value, mean value of rainwater recycling efficiency, mean value of the first water level standard deviation, mean value of irrigation water shortage, and mean value of the second water level standard deviation corresponding to the particles are normalized to generate the normalized cost value, mean value of rainwater recycling efficiency, mean value of the first water level standard deviation, mean value of irrigation water shortage, and mean value of the second water level standard deviation. The multi-objective fitness value of the particle is calculated based on the normalized cost value, mean value of rainwater recycling efficiency, mean value of the first water level standard deviation, mean value of irrigation water shortage, and mean value of the second water level standard deviation.

[0058] For example, the multi-objective fitness function can be:

[0059] in, is the multi-objective fitness value, and is the preset weight, and greater than 0, ,For example, =0.1, =0.3, =0.3, =0.3, is the normalized cost value, is the normalized mean value of rainwater recovery efficiency, is the normalized mean of the first water level standard deviation, is the normalized mean of the second water level standard deviation, is the normalized mean value of irrigation water deficit.

[0060] Understandably, is the normalized cost value, which is directly related to the number of rainwater storage components (the more components, the higher the cost). By taking the reciprocal, the cost minimization problem is transformed into maximizing the sub-item value. The normalized mean value of rainwater recycling efficiency directly reflects the system's ability to utilize rainwater. The smaller the normalized mean value of irrigation water shortage, the stronger the water supply guarantee capacity. By taking the reciprocal, the water shortage minimization problem is transformed into maximizing the sub-item value. By summing , comprehensively measure the imbalance under the two demands, then take the inverse, and transform the standard deviation minimization problem into maximizing the sub-item value. The smaller the mean of the normalized first water level standard deviation and the mean of the normalized second water level standard deviation are, the better the balance is and the larger the sub-item value is.

[0061] For each rainwater storage component, the mean of the rainwater recovery correlation coefficients of any two porous drainage pipes connected to the rainwater storage component (referred to as the mean rainwater recovery correlation coefficient corresponding to the rainwater storage component) and the mean of the irrigation demand correlation coefficients of any two irrigation areas connected to the rainwater storage component (referred to as the mean irrigation demand correlation coefficient of the rainwater storage component) can be calculated. The mean of the mean rainwater recovery correlation coefficients for each rainwater storage component (referred to as the mean rainwater recovery correlation coefficient corresponding to the particle) and the mean of the mean irrigation demand correlation coefficients for each rainwater storage component (referred to as the mean irrigation demand correlation coefficient corresponding to the particle) can be calculated. Based on the mean rainwater recovery correlation coefficients, the mean irrigation demand correlation coefficients, and the multi-objective fitness value of the particle, the position guidance value of the particle is calculated. The smaller the mean rainwater recovery correlation coefficient, the smaller the mean irrigation demand correlation coefficient, and the larger the multi-objective fitness value of the particle, the greater the position guidance value of the particle. For example, the first value is obtained by adding 1 to the mean value of the rainwater recovery correlation coefficient corresponding to the particle and taking the inverse thereof. The second value is obtained by adding 1 to the mean value of the irrigation demand correlation coefficient corresponding to the particle and taking the inverse thereof. The first value, the second value and the multi-objective fitness value of the particle are weightedly summed to obtain the position guidance value of the particle.

[0062] The global optimal position can be the position corresponding to the particle with the largest position guidance value in the current iteration. The historical optimal position of a particle can be the position with the largest position guidance value during the search process since the particle was initialized.

[0063] As can be understood, by combining the rainwater recycling correlation coefficient and the irrigation demand correlation coefficient to calculate the position guidance value, particles are guided to assign low-correlated porous drains to the same rainwater storage component, reducing overflow risk. Simultaneously, low-correlated irrigation areas are connected to the same component, smoothing water demand and preventing system efficiency degradation caused by supply-demand imbalances (synchronous overflow or depletion) in storage components. The correlation-based guidance value calculation allows particle velocity updates to focus on reducing system conflicts (such as the centralized connection of low-correlated drains / irrigation areas), accelerating convergence to a Pareto optimal solution that balances efficiency and stability, and reducing ineffective search paths.

[0064] The termination conditions may include the number of iterations reaching the maximum number of iterations, the fitness value being stable (for example, when the multi-objective fitness value of the particle swarm (such as the coverage of the Pareto front or the super volume index) is continuously N The change in the iteration is less than the preset threshold (like =0.001), indicating that the solution space has been fully explored and the algorithm has converged), particle position stagnation (for example, if the position update amplitude (such as Euclidean distance) of all particles is continuously M iterations are less than δ (like δ =0.01×search space range), indicating that the particle swarm falls into a local optimum and the iteration can be terminated).

[0065] The existing particle swarm algorithm can be used to update the position and velocity of particles, and the position and velocity of each particle can be updated according to the global optimal position and the historical optimal position of each particle.

[0066] The overflow control module includes multiple overflow control units, each rainwater storage component corresponds to an overflow control unit, and the overflow control unit is used to control the operation of the rainwater storage component and the opening of the overflow valve based on the connectivity between the multiple porous drainage pipes and the multiple rainwater storage components and the real-time rainwater recovery data of the multiple porous drainage pipes.

[0067] In some embodiments, the overflow control unit controls the operation of the rainwater storage assembly based on the connectivity between the plurality of porous drainage pipes and the plurality of rainwater storage assemblies and the real-time rainwater recovery data of the plurality of porous drainage pipes, including: determining whether to start the vacuum device based on the water level of the first water storage area collected by the first water level sensor at multiple consecutive time points; predicting the water level of the second water storage area at multiple future time points based on real-time rainwater recovery data from the porous drainage pipe connected to the rainwater storage assembly and the water level of the second water storage area collected by the second water level sensor at multiple consecutive time points; The opening time of the overflow valve is determined according to the predicted water levels of the second water storage area at multiple future time points.

[0068] Specifically, the water levels of the first water storage area collected by the first water level sensor at multiple consecutive time points are averaged to obtain a water level average. If the water level average is greater than a first preset water level, it is determined to turn on the vacuum equipment, wherein the first preset water level can be determined based on manual or experimental data.

[0069] A first water level prediction model (for example, an LSTM (Long Short-Term Memory Network) neural network or an ARIMA (Autoregressive Integrated Moving Average Model) model) is used to predict the water level of the second water storage area at multiple future time points based on real-time rainwater recovery data (for example, water level, flow rate, etc.) of the porous drainage pipe connected to the rainwater storage component and the water level of the second water storage area collected by the second water level sensor at multiple consecutive time points. If the water level at a certain future time point is predicted to be greater than a second preset water level, the future time is used as the opening time of the overflow valve, wherein the second preset water level can be determined based on manual or experimental data.

[0070] In some embodiments, the overflow control unit controls the opening of the overflow valve according to the connectivity between the plurality of porous drainage pipes and the plurality of rainwater storage assemblies and the real-time rainwater recovery data of the plurality of porous drainage pipes, including: determining an initial opening of the overflow valve according to the predicted water levels of the second water storage area at multiple future time points; After the overflow valve is opened, the opening of the overflow valve is adjusted according to the real-time rainwater recovery data of the porous drainage pipe connected to the rainwater storage component and the water level of the second water storage area collected by the second water level sensor at multiple consecutive time points.

[0071] Specifically, the total overflow water volume is calculated based on the predicted water level at the opening time of the overflow valve and the water level at the last future time point. The average flow rate required by the overflow valve is calculated based on the total overflow water volume and the time lengths of multiple future times. The initial opening of the overflow valve is determined based on the average flow rate required by the overflow valve. For example, a correlation function between the flow rate and the opening of the overflow valve can be established in advance. The average flow rate required by the overflow valve is substituted into the correlation function to calculate the initial opening of the overflow valve.

[0072] A second water level prediction model (e.g., an LSTM (Long Short-Term Memory Network) neural network model) is used to predict the water levels of the second water storage area at multiple future time points after the overflow valve is opened based on the real-time rainwater recovery data of the porous drainage pipe connected to the rainwater storage component and the water level of the second water storage area collected by the second water level sensor at multiple consecutive time points. If there is at least one future time point where the water level is greater than the second preset water level, the water level at the last future time point and the second preset water level are calculated, and the replenishment overflow water volume is calculated. Based on the replenishment overflow water volume and the time lengths of multiple future times, the replenishment flow required to be provided by the overflow valve is calculated. The replenishment flow required to be provided by the overflow valve and the average flow required to be provided by the overflow valve corresponding to the initial opening are summed and then inserted into the correlation function to obtain the adjusted opening of the overflow valve, thereby adjusting the opening of the overflow valve.

[0073] In some embodiments, the overflow control unit is further configured to: Whether to open the sewage valve is determined based on the water level of the first water storage area collected by the first water level sensor at multiple consecutive time points.

[0074] Specifically, if the standard deviation of the water level in the first water storage area collected by the first water level sensor at multiple consecutive time points is less than a standard deviation threshold (for example, 0.2), it is determined that the drain valve is opened, and the rainwater stored in the first water storage area can quickly flush the filling area and the distribution tank, taking away the deposited impurities in the filling area and the distribution tank.

[0075] The irrigation control module is used to control the operation of multiple greening irrigation components based on the real-time soil status data of the green belt.

[0076] Specifically, if the soil temperature and humidity in the irrigation area (e.g., 10 cm below the surface) of the greening irrigation component meet the irrigation conditions (e.g., temperature > 5°C and soil moisture < 25%), the greening irrigation component will automatically perform a watering operation. The time interval between two consecutive watering operations must be greater than a preset time interval threshold (e.g., 10 minutes).

[0077] Figure 5 is a flow chart of an intelligent control method for rainwater overflow and reuse of an elevated bridge according to some embodiments of this specification, such as Figure 5 As shown, the intelligent control method for viaduct rainwater overflow and reuse may include the following steps.

[0078] Collect historical rainwater recovery data from multiple porous drainage pipes and historical soil condition data from the green belts of the viaduct; Based on the historical rainwater recovery data of multiple porous drainage pipes and the historical irrigation data of the green belt of the viaduct, the arrangement of multiple rainwater storage components, the connectivity relationship between multiple porous drainage pipes and multiple rainwater storage components, the arrangement of multiple greening irrigation components, and the connectivity relationship between multiple rainwater storage components and multiple greening irrigation components are optimized; Controlling the operation of the rainwater storage components and the opening of the overflow valve according to the connectivity between the multiple porous drainage pipes and the multiple rainwater storage components and the real-time rainwater recovery data of the multiple porous drainage pipes; The operation of the multiple greening irrigation components is controlled according to the real-time soil status data of the green belt and the connectivity relationship between the multiple rainwater storage components and the multiple greening irrigation components.

[0079] The intelligent control method for viaduct rainwater overflow and reuse can be applied to the above-mentioned intelligent control system for viaduct rainwater overflow and reuse, which will not be described in detail here.

[0080] Finally, it should be understood that the embodiments described in this specification are intended only to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly described and illustrated in this specification.

Claims

1. The intelligent control system for rainwater overflow and reuse on viaduct is characterized by: A rainwater recycling and reuse system is applied to a viaduct. The system includes a viaduct rainwater recycling assembly, multiple rainwater storage assemblies, and multiple green irrigation assemblies. The viaduct rainwater recycling assembly includes multiple porous drainage pipes, an overflow valve is provided between the rainwater storage assembly and the municipal rainwater pipe, and an irrigation pipe is provided between the rainwater storage assembly and the green irrigation assembly. The system includes: Data collection module, used to collect historical rainwater recovery data from multiple porous drainage pipes and historical soil status data from the green belts of the viaduct; A structural optimization module for optimizing the arrangement of multiple rainwater storage components, the connectivity between the multiple porous drainage pipes and the multiple rainwater storage components, the arrangement of multiple greening irrigation components, and the connectivity between the multiple rainwater storage components and the multiple greening irrigation components based on historical rainwater recovery data of the multiple porous drainage pipes and historical irrigation data of the green belt of the viaduct; An overflow control module includes a plurality of overflow control units, one for each rainwater storage assembly. The overflow control units are used to control the operation of the rainwater storage assemblies and the opening of the overflow valves based on the connectivity between the plurality of porous drainage pipes and the plurality of rainwater storage assemblies and the real-time rainwater recovery data of the plurality of porous drainage pipes. The irrigation control module is used to control the operation of multiple greening irrigation components based on the real-time soil status data of the green belt.

2. The intelligent control system for rainwater overflow and reuse of viaduct according to claim 1 is characterized in that: The historical rainwater recovery data of porous drain pipes includes the water level and flow rate of multiple porous drain pipes at multiple historical rainfall time points during multiple historical rainfall processes; The historical soil status data of the green belt of the viaduct includes soil temperature and humidity at multiple soil monitoring points at multiple historical time points; The structural optimization module optimizes the arrangement of multiple rainwater storage components, the connectivity between the multiple porous drainage pipes and the multiple rainwater storage components, the arrangement of multiple greening irrigation components, and the connectivity between the multiple rainwater storage components and the multiple greening irrigation components based on historical rainwater recovery data of multiple porous drainage pipes and historical irrigation data of the viaduct's green belt, including: Based on historical rainwater recovery data of porous drain pipes, determine the rainwater recovery related characteristics of multiple porous drain pipes; Based on the soil temperature and humidity at multiple soil monitoring points at multiple historical time points, the green belt of the viaduct is divided into multiple irrigation areas and multiple green irrigation components are set up; Determine irrigation demand-related characteristics for multiple irrigation areas based on historical irrigation data from multiple green irrigation components; Based on the rainwater recovery-related characteristics of multiple porous drainage pipes and the irrigation demand-related characteristics of multiple irrigation areas, the settings of multiple rainwater storage components, the connectivity between multiple porous drainage pipes and multiple rainwater storage components, and the connectivity between multiple rainwater storage components and multiple greening irrigation components are optimized.

3. The intelligent control system for rainwater overflow and reuse of viaduct according to claim 2 is characterized in that: The rainwater recovery related characteristics of the plurality of porous drainage pipes include a rainwater recovery correlation coefficient between any two porous drainage pipes; The irrigation demand related characteristics of the plurality of irrigation areas include an irrigation demand correlation coefficient between any two irrigation areas.

4. The intelligent control system for rainwater overflow and reuse of viaduct according to claim 3 is characterized in that: The structural optimization module optimizes the arrangement of multiple rainwater storage components, the connectivity between the multiple porous drainage pipes and the multiple rainwater storage components, and the connectivity between the multiple rainwater storage components and the multiple greening irrigation components based on the rainwater recovery-related characteristics of the multiple porous drainage pipes and the irrigation demand-related characteristics of the multiple irrigation areas, including: Through an improved particle swarm optimization algorithm based on the rainwater recovery-related characteristics of multiple porous drainage pipes and the irrigation demand-related characteristics of multiple irrigation areas, the settings of multiple rainwater storage components, the connectivity between multiple porous drainage pipes and multiple rainwater storage components, and the connectivity between multiple rainwater storage components and multiple greening irrigation components are optimized.

5. The intelligent control system for rainwater overflow and reuse of viaduct according to claim 4 is characterized in that: An improved particle swarm optimization algorithm is used to optimize the configuration of multiple rainwater storage components, the connectivity between multiple porous drainage pipes and multiple rainwater storage components, and the connectivity between multiple rainwater storage components and multiple greening irrigation components based on the rainwater recovery-related characteristics of multiple porous drainage pipes and the irrigation demand-related characteristics of multiple irrigation areas. The optimization includes: Determine multiple rainfall simulation scenarios based on historical rainwater recovery data from porous drains; Determine multiple irrigation demand scenarios based on historical irrigation data of multiple green irrigation components; Based on multiple rainfall simulation scenarios and multiple irrigation demand scenarios; Through an improved particle swarm algorithm, based on the rainwater recovery-related characteristics of multiple porous drainage pipes, the irrigation demand-related characteristics of multiple irrigation areas, multiple rainfall simulation scenarios and multiple irrigation demand scenarios, the settings of multiple rainwater storage components, the connectivity between multiple porous drainage pipes and multiple rainwater storage components, and the connectivity between multiple rainwater storage components and multiple greening irrigation components are optimized.

6. The intelligent control system for rainwater overflow and reuse of an elevated bridge according to any one of claims 1 to 5, characterized in that: The viaduct rainwater recovery assembly also includes a fully permeable pavement arranged on both sides of the viaduct and a pebble layer arranged under the fully permeable pavement, a permeable geotextile is arranged between the fully permeable pavement and the pebble layer, and a porous drainage pipe is laid in the pebble layer; The rainwater storage assembly includes a distribution pool connected to a connecting pipe, a water storage tank, and a vacuum device. The water storage tank is provided with an overflow plate, which is used to divide the space inside the water storage tank into a first water storage area and a second water storage area. The first water storage area is located above the distribution pool. A filling area is provided in the distribution pool. The first water storage area is connected to the filling area. The vacuum device is provided on the water storage tank. The first water storage area is provided with a first water level sensor, and the second water storage area is provided with a second water level sensor.

7. The intelligent control system for rainwater overflow and reuse of viaduct according to claim 6 is characterized in that: The overflow control unit controls the operation of the rainwater storage assembly according to the communication relationship between the multiple porous drainage pipes and the multiple rainwater storage assemblies and the real-time rainwater recovery data of the multiple porous drainage pipes, including: determining whether to start the vacuum device based on the water level of the first water storage area collected by the first water level sensor at multiple consecutive time points; predicting the water level of the second water storage area at multiple future time points based on real-time rainwater recovery data from the porous drainage pipe connected to the rainwater storage assembly and the water level of the second water storage area collected by the second water level sensor at multiple consecutive time points; The opening time of the overflow valve is determined according to the predicted water levels of the second water storage area at multiple future time points.

8. The intelligent control system for rainwater overflow and reuse of viaduct according to claim 7 is characterized in that: The overflow control unit controls the opening of the overflow valve according to the connectivity between the multiple porous drainage pipes and the multiple rainwater storage components and the real-time rainwater recovery data of the multiple porous drainage pipes, including: determining an initial opening of the overflow valve according to the predicted water levels of the second water storage area at multiple future time points; After the overflow valve is opened, the opening of the overflow valve is adjusted according to the real-time rainwater recovery data of the porous drainage pipe connected to the rainwater storage component and the water level of the second water storage area collected by the second water level sensor at multiple consecutive time points.

9. The intelligent control system for rainwater overflow and reuse of viaduct according to claim 6 is characterized in that: The distribution tank is further connected to a municipal sewage pipe on one side away from the filling area, and a sewage valve is provided between the distribution tank and the municipal sewage pipe; The overflow control unit is further used for: Whether to open the sewage valve is determined based on the water level of the first water storage area collected by the first water level sensor at multiple consecutive time points.

10. Intelligent control method for rainwater overflow and reuse on viaduct, characterized in that: The intelligent control system for rainwater overflow and reuse of an elevated bridge as claimed in claim 1 comprises: Collect historical rainwater recovery data from multiple porous drainage pipes and historical soil condition data from the green belts of the viaduct; Based on the historical rainwater recovery data of multiple porous drainage pipes and the historical irrigation data of the green belt of the viaduct, the arrangement of multiple rainwater storage components, the connectivity relationship between multiple porous drainage pipes and multiple rainwater storage components, the arrangement of multiple greening irrigation components, and the connectivity relationship between multiple rainwater storage components and multiple greening irrigation components are optimized; Controlling the operation of the rainwater storage components and the opening of the overflow valve according to the connectivity between the multiple porous drainage pipes and the multiple rainwater storage components and the real-time rainwater recovery data of the multiple porous drainage pipes; The operation of the multiple greening irrigation components is controlled according to the real-time soil status data of the green belt and the connectivity relationship between the multiple rainwater storage components and the multiple greening irrigation components.

Images