Urban drainage network optimization method and device based on big data

Abstract

The present invention discloses a method and device for optimizing an urban drainage network based on big data, which relates to the field of data processing technology. The method includes: constructing multiple characteristic rainfall scenarios based on differential clustering of historical rainfall events, evaluating the drainage network's drainage response compliance under each scenario, screening and generating target rainfall scenarios, combining above-ground and underground data for adaptive layout adjustment, establishing a layout adjustment space, performing multi-degree-of-freedom optimization based on a network loss prediction model, forming an optimization strategy, and performing state compensation through real-time monitoring data, ultimately generating a dynamically adaptive drainage network optimization solution. The present invention solves the technical problems of existing urban drainage networks in responding to complex rainfall scenarios, such as lack of systematic evaluation and multi-dimensional optimization mechanisms, insufficient drainage response capabilities, and low operating efficiency. The present invention achieves the technical effect of effectively improving the drainage network's drainage response capabilities and operating efficiency through multi-scenario optimization and real-time monitoring compensation.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method and device for optimizing an urban drainage network based on big data. Background Art

[0002] In the current context of rapid urbanization, drainage systems, as a crucial component of urban infrastructure, have a direct impact on a city's flood control and drainage capabilities through their planning, design, and operational efficiency. However, traditional drainage network planning methods are mostly based on static historical rainfall data and single-design rainstorm intensity formulas. These methods lack systematic analysis of complex rainfall patterns and are therefore unable to cope with the increasingly frequent and intense extreme rainfall events of recent years.

[0003] In addition, the existing drainage system generally has problems such as lagging design standards, rigid spatial layout, and slow operation and maintenance response. Especially when the rainfall intensity is concentrated in a short period of time or unevenly distributed in space, it is easy to cause local waterlogging or even urban waterlogging disasters. Summary of the Invention

[0004] This application provides a method and device for optimizing urban drainage networks based on big data, which is used to solve the technical problems that existing urban drainage networks lack systematic evaluation and multi-dimensional optimization mechanisms when dealing with complex rainfall scenarios, and have insufficient drainage response capabilities and low operating efficiency.

[0005] The first aspect of the present application provides an urban drainage network optimization method based on big data, the method comprising: performing differential clustering on a target city's historical rainfall event set to obtain a plurality of characteristic rainfall scenarios; performing drainage response compliance calculations on the target city's drainage network based on the plurality of characteristic rainfall scenarios to obtain drainage response compliance coefficients for each scenario; performing compliance inspections on the plurality of characteristic rainfall scenarios based on the drainage response compliance coefficients for each scenario to generate a target rainfall scenario; obtaining above-ground and underground data sets of the target city, and adaptively adjusting the layout of the drainage network in combination with the target rainfall scenario to establish a network layout adjustment space; performing multi-degree-of-freedom optimization on the network layout adjustment space based on a network loss prediction model to generate a first network layout optimization strategy; obtaining multi-dimensional data of network monitoring of the drainage network, and performing network status compensation optimization on the first network layout optimization strategy based on the multi-dimensional data of network monitoring to obtain a second network optimization strategy.

[0006] The second aspect of the present application provides an urban drainage network optimization device based on big data, the device comprising: a rainfall scenario combing module, the rainfall scenario combing module is used to perform differential clustering on the historical rainfall event set of the target city to obtain multiple characteristic rainfall scenarios; a drainage response compliance calculation module, the drainage response compliance calculation module is used to perform drainage response compliance calculation on the drainage network of the target city according to the multiple characteristic rainfall scenarios to obtain drainage response compliance coefficients for each scenario; a compliance inspection module, the compliance inspection module is used to perform compliance inspection on the multiple characteristic rainfall scenarios according to the drainage response compliance coefficients for each scenario to generate a target rainfall scenario; A layout adjustment module, the pipe network layout adjustment module is used to obtain the above-ground and underground data sets of the target city, perform adaptive layout adjustment on the drainage pipe network in combination with the target rainfall scenario, and establish a pipe network layout adjustment space; a multi-degree-of-freedom optimization module, the multi-degree-of-freedom optimization module is used to perform multi-degree-of-freedom optimization on the pipe network layout adjustment space according to the pipe network loss prediction model, and generate a first pipe network layout optimization strategy; a pipe network state compensation module, the pipe network state compensation module is used to obtain multi-dimensional pipe network monitoring data of the drainage pipe network, and perform pipe network state compensation optimization on the first pipe network layout optimization strategy according to the multi-dimensional pipe network monitoring data to obtain a second pipe network optimization strategy.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The urban drainage network optimization method and device based on big data provided in this application relate to the field of data processing technology. Characteristic rainfall scenarios are constructed through differential clustering, the drainage response compliance is evaluated to generate target scenarios, the pipeline network layout is adjusted in combination with above-ground and underground data, the scheme is optimized based on the loss prediction model, and compensation correction is performed by integrating monitoring data to form a dynamically optimized drainage network strategy. The method and device solve the technical problems of the existing urban drainage network lacking systematic evaluation and multi-dimensional optimization mechanism when dealing with complex rainfall scenarios, insufficient drainage response capability and low operating efficiency. The method achieves the technical effect of effectively improving the drainage response capability and operating efficiency of the drainage network through multi-scenario optimization and real-time monitoring compensation. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A flow chart of a method for optimizing an urban drainage network based on big data provided in an embodiment of the present application;

[0011] Figure 2 Schematic diagram of the structure of the urban drainage network optimization device based on big data provided in an embodiment of the present application.

[0012] Explanation of the accompanying symbols: rainfall scenario sorting module 11, drainage response compliance calculation module 12, compliance inspection module 13, pipe network layout adjustment module 14, multi-degree-of-freedom optimization module 15, pipe network status compensation module 16. DETAILED DESCRIPTION

[0013] This application provides a method and device for optimizing urban drainage networks based on big data, which is used to solve the technical problems that existing urban drainage networks lack systematic evaluation and multi-dimensional optimization mechanisms when dealing with complex rainfall scenarios, and have insufficient drainage response capabilities and low operating efficiency.

[0014] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0015] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices.

[0016] Example 1, as Figure 1 As shown, this application provides an urban drainage network optimization method based on big data, which includes:

[0017] P10: Perform differential clustering on the historical rainfall event set of the target city to obtain multiple characteristic rainfall scenes.

[0018] Furthermore, step P10 in the embodiment of the present application further includes:

[0019] P11: Perform pairwise difference evaluation based on the historical rainfall event set to obtain a first differential detection distribution; P12: Classify the historical rainfall event set based on the first differential detection distribution to obtain multiple rainfall event classes; P13: Perform inter-class difference evaluation based on the multiple rainfall event classes to obtain a second differential detection distribution, and perform inter-class fusion on the multiple rainfall event classes based on the second differential detection distribution to obtain multiple rainfall event areas; P14: Perform rainfall feature recognition and fusion on each rainfall event area within the multiple rainfall event areas to generate the multiple characteristic rainfall scenes.

[0020] It should be understood that in order to achieve efficient optimization of urban drainage networks under variable climate conditions, it is first necessary to extract the historical rainfall event set of the target city and perform differential clustering to obtain multiple characteristic rainfall scenes that can represent different rainfall patterns.

[0021] Specifically, a pairwise dissimilarity assessment is first performed on all events in the historical rainfall event set to construct a quantitative index of relative heterogeneity between rainfall events. This dissimilarity assessment can be carried out based on multidimensional meteorological characteristic parameters, including but not limited to total rainfall, rainfall duration, maximum one-hour rainfall intensity, rainfall peak time, rainfall center of gravity, and spatial distribution pattern. By constructing a comprehensive dissimilarity function, such as weighted Euclidean distance or Mahalanobis distance, the difference values ​​between all events are calculated, and then the first difference detection distribution is obtained. This distribution reflects the similarities and distinguishability between events in the historical rainfall event set.

[0022] After obtaining the first differential detection distribution, historical rainfall events are classified using an unsupervised clustering method based on dissimilarity. Examples of clustering methods include DBSCAN, spectral clustering, or hierarchical clustering. Based on the difference values, events are divided into several rainfall event clusters. Within each cluster, events are highly similar in terms of rainfall intensity, temporal characteristics, and spatial features.

[0023] After the initial clustering is completed, the differences between these rainfall event classes are further evaluated between classes. By calculating the distance or distribution overlap between the average eigenvectors of each class, a second differential detection distribution is obtained. The evaluation of inter-class differences can also be performed from multiple dimensions such as rainfall intensity, rainfall duration, and rainfall distribution. For example, the difference in average rainfall intensity and average rainfall duration between two rainfall event classes can be calculated. Through the second differential detection distribution, it is possible to identify which rainfall event classes have significant differences and which classes have smaller differences, thereby providing a basis for further inter-class fusion. This process is a key step in the differential clustering algorithm, which determines the degree of similarity between rainfall event classes. Through reasonable inter-class difference evaluation, it is possible to avoid over-segmentation or merging of inappropriate rainfall event classes, ensuring that the final rainfall event area has reasonable representativeness and discrimination.

[0024] Finally, based on the second differential detection distribution, multiple rainfall event classes are fused between classes. Rainfall event classes with smaller differences are merged into larger rainfall event areas, while rainfall event classes with significant differences are retained as independent rainfall event areas. After forming multiple rainfall event areas, rainfall feature identification and fusion are further performed within each rainfall event area. For example, by analyzing rainfall events within each rainfall event area, representative rainfall features such as average rainfall intensity, typical rainfall duration, and primary rainfall distribution areas are extracted. These rainfall features are fused to generate the final characteristic rainfall scenarios. These characteristic rainfall scenarios can fully cover the main characteristics of historical rainfall events in the target city, providing accurate rainfall condition simulation for subsequent drainage network optimization analysis.

[0025] Rainfall feature identification and fusion are key steps in generating characteristic rainfall scenarios. By extracting and fusing rainfall features within a rainfall event area, concise and representative characteristic rainfall scenarios can be generated. These scenarios not only reflect the main characteristics of actual rainfall events, but also provide effective input data for the drainage response compliance calculation of drainage networks.

[0026] P20: Calculate the waterlogging response compliance of the drainage network of the target city according to the multiple characteristic rainfall scenarios to obtain a waterlogging response compliance coefficient for each scenario.

[0027] Furthermore, step P20 in the embodiment of the present application further includes:

[0028] P21: Extract the first characteristic rainfall scene based on the multiple characteristic rainfall scenes; P22: Search the drainage response time of the drainage network based on the first characteristic rainfall scene to obtain the drainage time set of the first rainfall scene; P23: Select the drainage time set of the first rainfall scene based on the standard drainage response time to obtain a first standard drainage time space that is less than or equal to the standard drainage response time; P24: Calculate the number of samples for the first rainfall scene drainage time set and the first standard drainage time space to obtain the number of first scene time samples and the number of first standard time samples; P25: Calculate the proportion of the first standard time sample number and the first scene time sample number to obtain the first scene drainage response compliance coefficient, and add the first scene drainage response compliance coefficient to the drainage response compliance coefficient of each scene.

[0029] Optionally, based on the identified multiple characteristic rainfall scenarios, the drainage capacity of the target city's existing drainage network system can be quantitatively evaluated to obtain a drainage response compliance coefficient for each scenario. This compliance coefficient reflects the efficiency and reliability of the drainage network in completing its drainage task under the corresponding rainfall scenario and is an important indicator for the subsequent formulation of network optimization strategies.

[0030] First, a first characteristic rainfall scenario to be evaluated is sequentially extracted from the plurality of characteristic rainfall scenarios to establish a drainage performance evaluation model corresponding to the scenario. The scenario can represent a rainfall pattern with specific intensity, spatial distribution, or temporal characteristics, and has a clear evaluation objective.

[0031] Then, based on the first characteristic rainfall scenario, the system retrieved historical drainage operation data for the urban drainage network corresponding to this rainfall scenario and extracted the corresponding drainage response duration records to form the drainage duration set for the first rainfall scenario. This set includes multiple historical drainage response durations, reflecting the drainage network's actual drainage efficiency during different historical rainfall events.

[0032] Next, the drainage duration set for the first rainfall scenario is screened based on the preset standard drainage response time. The standard drainage response time is determined based on urban planning or relevant regulations and measures the maximum time it takes for a drainage network to complete drainage tasks under specified rainfall conditions. Within the drainage duration set for the first rainfall scenario, historical drainage response times that are less than or equal to the standard drainage response time are selected to form the first standard drainage time space. The number of samples within this space is the number of times the drainage network achieved the standard drainage response time under this characteristic rainfall scenario.

[0033] Furthermore, the sample counts were calculated for the first rainfall scenario drainage duration set and the first standard-compliant drainage duration space. The total number of samples in the first rainfall scenario drainage duration set was counted, denoted as the first scenario duration sample count. Simultaneously, the number of samples in the first standard-compliant drainage duration space was counted, denoted as the first standard-compliant duration sample count. These two sample counts provided the foundation for the subsequent standard-compliant coefficient calculation.

[0034] Finally, the ratio of the number of samples of the first standard-reaching duration to the number of samples of the first scenario duration is calculated. By dividing the number of samples of the first standard-reaching duration by the number of samples of the first scenario duration, the first scenario drainage response compliance coefficient is obtained. This coefficient reflects the probability that the drainage network will meet the standard drainage response duration under the first characteristic rainfall scenario, that is, the drainage response compliance capacity of the drainage network. The calculation formula is: First scenario drainage response compliance coefficient = Number of samples of the first standard-reaching duration / Number of samples of the first scenario duration. The calculated first scenario drainage response compliance coefficient will be added to the set of drainage response compliance coefficients for each scenario, providing a key quantitative indicator for subsequent drainage network optimization.

[0035] This process not only makes full use of historical data and preset standards, but also provides a scientific basis for the optimization of the drainage network through quantitative analysis, avoiding subjective evaluation methods that rely on single simulations or expert experience, and ensuring the objectivity, repeatability and engineering applicability of the evaluation results.

[0036] P30: Conduct compliance checks on the multiple characteristic rainfall scenarios based on the drainage response compliance coefficients of the scenarios to generate a target rainfall scenario.

[0037] Furthermore, step P30 in the embodiment of the present application further includes:

[0038] P31: Determine whether the drainage response compliance coefficient of each scenario is less than the drainage response compliance threshold, and obtain multiple drainage response compliance test results; P32: Filter the multiple characteristic rainfall scenarios according to the multiple drainage response compliance test results, and obtain a screening set of rainfall scenarios with unqualified responses; P33: Fusion the screening set of rainfall scenarios with unqualified responses to obtain the target rainfall scenario.

[0039] Specifically, based on the drainage response compliance coefficients corresponding to the aforementioned multiple characteristic rainfall scenarios, a compliance judgment operation is performed to screen out key rainfall scenarios whose drainage capacity has not yet met the standards, thereby forming a target rainfall scenario that has constraining guiding significance for drainage network optimization.

[0040] Specifically, the compliance of the drainage response compliance coefficient corresponding to each characteristic rainfall scenario is first determined. The judgment logic is to compare the compliance coefficient of each scenario with the preset drainage response compliance threshold. This threshold is usually determined by urban planning or industry standards and is used to determine whether the pipeline network has compliant drainage capacity. For example, if the threshold is set to 0.6, it means that at least 60% of historical drainage samples in this scenario meet the standard response time for the drainage capacity in this scenario to be considered to meet the standard. The judgment results are recorded as Boolean values ​​or labels, forming multiple drainage response compliance test results.

[0041] Next, based on the compliance test results, characteristic rainfall scenarios that failed the compliance test were further screened. For all scenarios whose compliance coefficients fell below the set threshold, their corresponding characteristic rainfall scenarios were marked as unqualified and aggregated into a screening set of unqualified rainfall scenarios. This screening set represents the set of risk scenarios within the current drainage network and serves as the core basis for subsequent optimization strategy design. By centralizing these unqualified scenarios, we can more systematically analyze and address these problematic scenarios, providing clear targets for subsequent optimization strategies.

[0042] Finally, in step P33, the filtered set of unsatisfactory rainfall scenarios is fused to generate a target rainfall scenario. This fusion can be performed using strategies such as superposition of primary feature frequencies between scenarios, weighted averaging, and reconstruction of representative samples. This extracts representative and structurally feasible rainfall input conditions, which are used to drive the input boundaries of the subsequent drainage system optimization model. The target rainfall scenario is not simply a superposition of the original feature scenarios, but rather a composite rainfall-driven model formed by focusing and abstracting the system's weak responses.

[0043] Through this compliance inspection and target scenario extraction process, not only a closed-loop drainage capacity diagnosis mechanism was established, but also precise environmental input conditions were provided for layout adjustment and optimization paths in subsequent steps, thereby improving the targeting and engineering value of the overall optimization strategy.

[0044] P40: Obtain the above-ground and underground datasets of the target city, perform adaptive layout adjustment on the drainage network in combination with the target rainfall scenario, and establish a network layout adjustment space.

[0045] Furthermore, step P40 in this embodiment of the present application further includes:

[0046] P41: Modeling is performed according to the pipe network layout plan of the drainage pipe network to obtain a three-dimensional pipe network model, and the three-dimensional pipe network model is expanded according to the above-ground and underground data sets to generate an urban drainage pipe network model; P42: Simulating drainage of the urban drainage pipe network model according to the target rainfall scenario to obtain pipe network simulation drainage data; P43: Performing risk identification based on the pipe network simulation drainage data to obtain pipe network drainage risk identification results; P44: Adjusting the pipe network layout plan according to the pipe network drainage risk identification results to generate the pipe network layout adjustment space.

[0047] Optionally, the spatial layout of the existing drainage network can be intelligently adjusted by combining the target city's comprehensive spatial information with key rainfall scenarios. This creates a network layout adjustment space with both adjustable feasibility and structural optimization potential. This space serves as the input for subsequent optimization solutions, reflecting the drainage performance of different layout options under the target rainfall scenario.

[0048] In specific implementation, an initial 3D model of the drainage network is first constructed based on existing design blueprints or operational data. This model is a spatial geometric representation of structural elements such as pipeline routes, node locations, and elevation information. Subsequently, this 3D model is expanded by incorporating data from the target city's above- and underground areas, including topography, surface building layout, underground pipeline routes, and underground space usage. This results in an urban drainage network model that incorporates multi-source data. This model serves as the foundation for subsequent simulations and layout assessments.

[0049] Next, based on the generated urban drainage network model, the previously extracted parameters of the target rainfall scenario are input to conduct a dynamic drainage simulation. This simulation can be performed using a hydrodynamic simulation tool such as the Storm Water Management Model (SWMM). The target rainfall characteristics (such as rainfall intensity, duration, and distribution) are used as input to simulate the drainage process of the drainage network under these rainfall conditions. This generates simulated drainage data for each drainage unit under this scenario, including indicators such as water level changes, flow velocity in pipe sections, and node overflow.

[0050] Next, the simulated drainage data is subjected to structured analysis and anomaly identification. By analyzing data such as water level changes and flow distribution at each node during the simulated drainage process, potential risk points in the drainage network are identified. For example, certain areas may experience problems such as excessively high water levels or prolonged drainage times, indicating potential layout or design flaws in the drainage network. The risk identification results will detail the location, type, and severity of these potential issues, providing clear guidance for subsequent adjustments to the network layout.

[0051] Finally, based on the risk identification results, the existing drainage network layout plan is automatically or semi-automatically adjusted to construct a number of network layout adjustment schemes with differentiated structural characteristics. For example, this may require increasing pipe diameters, adjusting pipe slopes, adding drainage pumping stations, or changing pipe connection methods. Through these adjustments, multiple network layout adjustment schemes are generated, forming a network layout adjustment space. These adjustment schemes provide multiple options for subsequent optimization strategies, ensuring that the drainage network can drain water more effectively under target rainfall scenarios and reducing the risk of waterlogging.

[0052] This process not only makes full use of above-ground and underground data and target rainfall scenario information, but also provides a scientific basis for the optimization of the drainage network through simulation and risk assessment, ensuring the scientific nature and effectiveness of the entire optimization method.

[0053] P50: Perform multi-degree-of-freedom optimization on the pipeline network layout adjustment space according to the pipeline network loss prediction model to generate a first strategy for pipeline network layout optimization.

[0054] Furthermore, step P50 in the embodiment of the present application further includes:

[0055] P51: Optimize the drainage response of the pipeline layout adjustment space according to the standard drainage response time to establish the first optimization space for pipeline layout adjustment; P52: Optimize the pipeline loss prediction of the first optimization space for pipeline layout adjustment according to the pipeline loss prediction model to obtain the second optimization space for pipeline layout adjustment; P53: Perform weight allocation based on the pipeline loss prediction multi-dimensional indicators of the pipeline loss prediction model to obtain a pipeline comprehensive loss calculation model; P54: Optimize the pipeline comprehensive loss minimization of the second optimization space for pipeline layout adjustment according to the pipeline comprehensive loss calculation model to obtain the first strategy for pipeline layout optimization.

[0056] It should be understood that, with the core goal of reducing drainage system operational losses, a systematic, multi-dimensional optimization analysis of the previously established pipe network layout adjustment space was conducted, ultimately outputting a first pipe network layout optimization strategy that can be used for project deployment. This strategy, while ensuring drainage response efficiency, further accounts for the various structural and economic losses incurred during the long-term operation of the pipe network, thereby achieving optimal overall performance at the layout level.

[0057] In practice, the standard drainage response time is used as the fundamental evaluation metric to initially screen the drainage performance of various schemes within the network layout adjustment space. Schemes with significantly excessive response times or insufficient regulation effects are eliminated, thereby constructing a first optimization space for network layout adjustment that meets basic drainage requirements. This space retains only candidate schemes with strong drainage capabilities, serving as the foundation for subsequent optimization.

[0058] Next, a pipeline network loss prediction model was introduced to analyze the losses of the solutions in the first optimization space. This prediction model, trained and generated based on historical operational data, engineering statistics, and hydraulic simulation results, quantitatively evaluates the performance of different layout solutions under target rainfall scenarios, outputting multiple prediction metrics, including node waterlogging losses, structural settlement trends, and pipeline damage rates. By performing pipeline network loss predictions on each solution in the first optimization space, solutions with significantly excessive loss predictions or high overall risks were eliminated, forming a second optimization space after preliminary loss prediction screening.

[0059] On this basis, a multi-dimensional loss trade-off mechanism is introduced to configure the weights of several loss dimensions included in the pipeline network loss prediction model to form a unified pipeline network comprehensive loss calculation model. In this application, the multi-dimensional indicators of the pipeline network loss prediction model include drainage waterlogging loss, drainage groundwater subsidence loss and drainage pipe loss. These indicators respectively reflect the potential losses of the drainage network in different aspects. In order to comprehensively evaluate the pros and cons of each layout scheme, it is necessary to assign weights to these multi-dimensional indicators. The distribution of weights can be adjusted according to the specific needs and priorities of the city. For example, if the city pays special attention to the risk of waterlogging, it can give a higher weight to drainage waterlogging loss. Through weight distribution, the pipeline network comprehensive loss calculation model is obtained, and its formula is:

[0060] in, is the comprehensive loss of the pipeline network, To drain waterlogging losses, is the drainage subsidence loss, For drainage pipe losses, 、 、 They are the weight coefficients for drainage waterlogging loss, drainage ground subsidence loss and drainage pipe loss, and can be configured according to the actual urban operation center or safety-sensitive area.

[0061] Finally, using this comprehensive loss model as the objective function, all options within the second optimization space are optimized to minimize the comprehensive loss of the pipeline network. Heuristic algorithms (such as genetic algorithms and particle swarm optimization) or exact algorithms (such as integer programming) can be used to perform a global search of multiple degrees of freedom variables. By calculating the comprehensive loss value of each layout option while ensuring drainage response constraints, a set of layout adjustment options with the lowest loss is selected, thereby obtaining the first strategy for pipeline network layout optimization. This strategy not only considers the drainage response time requirement but also comprehensively considers the potential losses in various aspects of the pipeline network, ensuring that the optimized pipeline network layout achieves optimal overall performance.

[0062] Furthermore, step P51 of the embodiment of the present application further includes:

[0063] P51-1: Based on the target rainfall scenario, the drainage response duration is predicted for each pipeline layout adjustment scheme in the pipeline layout adjustment space to obtain multiple predicted drainage response durations; P51-2: Determine whether the multiple predicted drainage response durations are less than or equal to the standard drainage response duration to obtain multiple drainage response judgment results; P51-3: Filter the pipeline layout adjustment space according to the multiple drainage response judgment results to obtain the first optimization space for the pipeline layout adjustment.

[0064] In a possible embodiment of the present application, in order to improve the efficiency of accurate screening of the drainage capacity of the pipeline network, the process of optimizing the drainage response of the pipeline network layout adjustment space based on the standard drainage response time can be further refined.

[0065] Specifically, the drainage response time is first predicted for each layout adjustment scheme within the pipe network adjustment space based on characteristic parameters such as rainfall intensity, duration, and spatial distribution of the target rainfall scenario. This prediction can be implemented using a hydraulic model (such as the SWMM model) or an integrated machine learning model (such as a regression prediction network trained on historical simulation data). The output is the drainage response time for each scheme under the current rainfall input conditions. This is the time it takes to simulate the entire process from rainfall onset to the elimination of waterlogging at the node, which is recorded as the predicted drainage response time.

[0066] Subsequently, the above-mentioned multiple predicted drainage response times are compared with the preset standard drainage response time. This standard is usually set according to the urban drainage design specifications or regional specific drainage performance indicators, for example, it is set to 3 hours or 2.5 hours, which means that in the target scenario, the drainage system should complete the task of draining accumulated water within a limited time. The comparison result is output as a logical judgment, that is, whether the predicted drainage response time is less than or equal to the standard drainage response time, thereby obtaining multiple drainage response judgment results. For example, if a certain predicted drainage response time is less than or equal to the standard drainage response time, it is considered that the pipe network layout adjustment plan meets the requirements in terms of drainage response time.

[0067] Finally, based on the above judgment results, all the solutions in the original pipe network layout adjustment space are filtered, and only those that meet the drainage response standard are retained. Specifically, all solutions whose corresponding predicted drainage response time is less than or equal to the standard value are screened out and added to the first optimization space of the pipe network layout adjustment. This space represents a set of solutions with feasible structures and basic drainage compliance capabilities, and is the solution set foundation for subsequent multi-dimensional loss prediction and comprehensive optimization. In this way, each solution in the first optimization space of the pipe network layout adjustment has the ability to complete the drainage task within the standard drainage response time, thus providing a solid foundation for the subsequent pipe network loss prediction optimization.

[0068] Furthermore, step P52 of the embodiment of the present application further includes:

[0069] P52-1: According to the first optimization space of the pipeline layout adjustment, the nth pipeline layout adjustment scheme is extracted, where n is a positive integer; P52-2: The nth pipeline layout adjustment scheme and the target rainfall scenario are input into the pipeline loss prediction model to obtain the nth pipeline loss prediction result, which includes the nth drainage waterlogging loss coefficient, the nth drainage underground subsidence loss coefficient and the nth drainage pipe loss coefficient; P52-3: According to the multidimensional indicators of the pipeline loss prediction, multidimensional constraints for the pipeline loss prediction are set; P52-4: Determine whether the nth pipeline loss prediction result meets the multidimensional constraints for the pipeline loss prediction; P52-5: If the nth pipeline loss prediction result meets the multidimensional constraints for the pipeline loss prediction, add the nth pipeline layout adjustment scheme to the second optimization space of the pipeline layout adjustment.

[0070] Optionally, in order to further evaluate and control the operation risks of the drainage system while ensuring the efficiency of drainage response, prediction and constraint screening operations are carried out for the multi-dimensional losses in the operation of the pipeline network.

[0071] First, the nth pipeline layout adjustment scheme is extracted from the first optimization space of pipeline layout adjustment, where n is a positive integer. This process serves as the starting point for a step-by-step evaluation of pipeline layout adjustment schemes. By extracting each scheme one by one, it provides a concrete target for subsequent loss prediction. Each pipeline layout adjustment scheme includes specific parameters of the pipeline network, such as pipe diameter, slope, and connection method. These parameters serve as input conditions for the subsequent loss prediction model.

[0072] Next, the extracted nth pipeline network layout adjustment scheme, along with the target rainfall scenario parameters, is input into the pipeline network loss prediction model. This prediction model, driven by multi-source data, operates through a combination of the following core modules: a hydraulic simulation module, which predicts the risk of node flooding and overflow intensity; a structural response analysis module, which simulates surface settlement caused by flow concentration and foundation disturbance; and a material fatigue and service life model, which assesses pipeline structural damage trends due to stress, corrosion, and aging. These submodules work together to output loss prediction results for the nth scheme under a given rainfall scenario. These include the nth drainage waterlogging loss coefficient (reflecting the risk of node flooding); the nth drainage groundwater settlement loss coefficient (reflecting the potential for structural settlement); and the nth drainage pipeline loss coefficient (reflecting the potential for system failure or maintenance costs).

[0073] Subsequently, multidimensional constraints for pipeline loss prediction are set based on the multidimensional indicators. These constraints are based on requirements for urban planning, environmental protection, and engineering safety, ensuring that the optimized pipeline network layout meets minimum standards in all aspects. For example, the drainage waterlogging loss coefficient can be set to no more than 0.5, ground subsidence to no more than 0.3, and pipeline loss to no more than 0.2, forming specific threshold vectors.

[0074] Next, the nth pipeline network loss prediction result is compared with the set multi-dimensional constraints to determine whether it simultaneously meets the control requirements of all dimensions. The judgment logic is full indicator compliance, that is, all three loss coefficients must be lower than or equal to the set threshold for the constraints to be considered satisfied. For example, the nth drainage waterlogging loss coefficient, the nth drainage groundwater subsidence loss coefficient, and the nth drainage pipeline loss coefficient are compared with the corresponding constraints respectively. If all loss coefficients meet the corresponding constraints, the pipeline network layout adjustment plan is considered feasible in terms of loss prediction.

[0075] If the nth pipeline loss prediction result satisfies the multidimensional constraints of pipeline loss prediction, the nth pipeline network layout adjustment scheme is added to the second optimization space for pipeline network layout adjustment. All schemes in this space demonstrate good drainage response performance and pass multiple safety assessments for flooding risk, structural settlement, and maintenance loss, forming the candidate solution set for the subsequent comprehensive optimization stage.

[0076] Through the implementation of this step, the solution selection no longer relies solely on drainage efficiency indicators. Instead, it introduces the loss assessment logic during system operation, constructs a multi-dimensional and refined evaluation closed loop, and makes the optimization path more in line with the life cycle management objectives and sustainable operation and maintenance requirements of the urban drainage system.

[0077] P60: Obtaining multi-dimensional data of pipe network monitoring of the drainage pipe network, and performing pipe network status compensation optimization on the first pipe network layout optimization strategy based on the multi-dimensional data of pipe network monitoring to obtain a second pipe network optimization strategy.

[0078] Furthermore, step P60 of the embodiment of the present application further includes:

[0079] P61: Performing status anomaly detection on the drainage network based on the multi-dimensional data of the network monitoring to obtain the results of the network status anomaly detection; P62: Performing compensatory optimization on the first strategy for optimizing the network layout based on the results of the network status anomaly detection to generate the second optimization strategy for the network optimization.

[0080] It should be understood that the introduction of a real-time monitoring data feedback mechanism allows for dynamic correction and engineering adaptation of the first pipeline layout optimization strategy previously obtained through model prediction and simulation analysis, thereby generating a second pipeline optimization strategy that is more in line with the actual operating status and has higher robustness and feasibility.

[0081] Specifically, the drainage network is first detected for state anomalies based on its multidimensional monitoring data to obtain network state anomaly detection results. Multidimensional network monitoring data typically includes, but is not limited to, water level monitoring data, flow rate monitoring data, pipeline pressure monitoring data, water quality monitoring data, and equipment operating status monitoring data. This data comprehensively reflects various status information of the drainage network during actual operation. Based on this data, anomaly detection algorithms (such as K-means clustering, isolation forest algorithm, sliding window threshold detection, etc.) are applied to identify anomalies in the entire network's operating status. The network state anomaly detection results are output, including the anomaly location (node ​​or pipe section), anomaly type (waterlogging, turbulent flow, structural deformation, siltation, etc.), time label, and severity score.

[0082] Next, based on the aforementioned test results, the constructed first strategy for pipe network layout optimization is optimized for state compensation. This compensation process is executed using two types of logic: First, structural redundancy compensation. For high-risk or high-frequency anomaly areas, additional drainage units, overflow channels, or expanded drainage pipes are dynamically added. Second, strategy redirection compensation. If some existing layout paths are affected by abnormal conditions and are not functioning smoothly, the drainage path, diversion direction, or node sharing mechanism is adjusted. Compensation strategies can be rule-driven or incorporate actual state constraints into existing optimization models, forming an iterative optimization solution model based on real-time constraint feedback.

[0083] Ultimately, the results of integrating anomaly detection with structural adjustments are output as the second optimization strategy for the pipeline network. This strategy not only inherits the theoretical optimality of the first strategy in the model space, but also fully incorporates data feedback from the operating site, ensuring the system's stability, emergency response capabilities, and ease of maintenance in complex real-world environments.

[0084] In summary, the embodiments of the present application have at least the following technical effects:

[0085] This application constructs multiple characteristic rainfall scenarios based on differential clustering of historical rainfall events, evaluates the drainage response compliance of the drainage network under each scenario, screens and generates target rainfall scenarios, performs adaptive layout adjustment based on above-ground and underground data, establishes a layout adjustment space, performs multi-degree-of-freedom optimization based on the pipeline loss prediction model, forms an optimization strategy, and performs state compensation through real-time monitoring data, ultimately generating a dynamically adaptive drainage network optimization plan.

[0086] The technical effect of effectively improving the drainage response capability and operating efficiency of the drainage network was achieved through multi-scenario optimization and real-time monitoring compensation.

[0087] Example 2 is based on the same inventive concept as the urban drainage network optimization method based on big data in the above embodiment. Figure 2As shown, this application provides an urban drainage network optimization device based on big data. The device and method embodiments in the embodiments of this application are based on the same inventive concept. The device includes:

[0088] The rainfall scene combing module 11 is used to perform differential clustering on the historical rainfall event set of the target city to obtain multiple characteristic rainfall scenes.

[0089] The drainage response compliance calculation module 12 is used to perform drainage response compliance calculation on the drainage network of the target city according to the multiple characteristic rainfall scenarios, and obtain the drainage response compliance coefficient for each scenario.

[0090] The compliance inspection module 13 is used to perform compliance inspection on the multiple characteristic rainfall scenarios according to the drainage response compliance coefficients of the scenarios to generate a target rainfall scenario.

[0091] The pipe network layout adjustment module 14 is used to obtain the above-ground and underground data sets of the target city, perform adaptive layout adjustment on the drainage pipe network in combination with the target rainfall scenario, and establish a pipe network layout adjustment space.

[0092] The multi-degree-of-freedom optimization module 15 is used to perform multi-degree-of-freedom optimization on the pipeline layout adjustment space according to the pipeline loss prediction model, and generate a first strategy for optimizing the pipeline layout.

[0093] The pipe network state compensation module 16 is used to obtain the pipe network monitoring multi-dimensional data of the drainage pipe network, and perform pipe network state compensation optimization on the first pipe network layout optimization strategy based on the pipe network monitoring multi-dimensional data to obtain the second pipe network optimization strategy.

[0094] Furthermore, the rainfall scene combing module 11 is further configured to perform the following steps:

[0095] A pairwise difference evaluation is performed on the historical rainfall event set to obtain a first differential detection distribution; the historical rainfall event set is classified according to the first differential detection distribution to obtain a plurality of rainfall event classes; an inter-class difference evaluation is performed on the plurality of rainfall event classes to obtain a second differential detection distribution, and the plurality of rainfall event classes are inter-class fused according to the second differential detection distribution to obtain a plurality of rainfall event areas; rainfall feature recognition and fusion are performed on each rainfall event area within the plurality of rainfall event areas to generate the plurality of characteristic rainfall scenes.

[0096] Furthermore, the drainage response compliance calculation module 12 is further configured to perform the following steps:

[0097] Based on the multiple characteristic rainfall scenes, a first characteristic rainfall scene is extracted; based on the first characteristic rainfall scene, the drainage response time of the drainage network is retrieved to obtain a first rainfall scene drainage time set; based on the standard drainage response time, the first rainfall scene drainage time set is selected to obtain a first standard drainage time space that is less than or equal to the standard drainage response time; the number of samples of the first rainfall scene drainage time set and the first standard drainage time space is calculated to obtain the number of first scene time samples and the number of first standard time samples; the proportion of the first standard time sample number and the first scene time sample number is calculated to obtain the first scene drainage response standard coefficient, and the first scene drainage response standard coefficient is added to the drainage response standard coefficients of each scene.

[0098] Furthermore, the compliance inspection module 13 is further configured to perform the following steps:

[0099] Determine whether the drainage response compliance coefficient of each scenario is less than the drainage response compliance threshold, and obtain multiple drainage response compliance test results; filter the multiple characteristic rainfall scenarios according to the multiple drainage response compliance test results to obtain a screening set of rainfall scenarios with unqualified responses; and fuse the screening set of rainfall scenarios with unqualified responses to obtain the target rainfall scenario.

[0100] Furthermore, the pipe network layout adjustment module 14 is further configured to perform the following steps:

[0101] Modeling is performed according to the pipe network layout plan of the drainage pipe network to obtain a three-dimensional pipe network model, and the three-dimensional pipe network model is expanded according to the above-ground and underground data sets to generate an urban drainage pipe network model; drainage of the urban drainage pipe network model is simulated according to the target rainfall scenario to obtain pipe network simulation drainage data; risk identification is performed based on the pipe network simulation drainage data to obtain pipe network drainage risk identification results; the pipe network layout plan is adjusted according to the pipe network drainage risk identification results to generate the pipe network layout adjustment space.

[0102] Furthermore, the multi-degree-of-freedom optimization module 15 is further configured to perform the following steps:

[0103] The drainage response of the pipeline layout adjustment space is optimized according to the standard drainage response time to establish the first optimization space for pipeline layout adjustment; the pipeline loss prediction of the first optimization space for pipeline layout adjustment is optimized according to the pipeline loss prediction model to obtain the second optimization space for pipeline layout adjustment; weights are allocated according to the pipeline loss prediction multi-dimensional indicators of the pipeline loss prediction model to obtain a pipeline comprehensive loss calculation model; the pipeline comprehensive loss minimization of the second optimization space for pipeline layout adjustment is optimized according to the pipeline comprehensive loss calculation model to obtain the first strategy for pipeline layout optimization.

[0104] Furthermore, the multi-degree-of-freedom optimization module 15 is further configured to perform the following steps:

[0105] Based on the target rainfall scenario, the drainage response duration is predicted for each pipeline layout adjustment scheme in the pipeline layout adjustment space to obtain multiple predicted drainage response durations; it is determined whether the multiple predicted drainage response durations are less than or equal to the standard drainage response duration to obtain multiple drainage response judgment results; the pipeline layout adjustment space is screened according to the multiple drainage response judgment results to obtain the first optimization space for the pipeline layout adjustment.

[0106] Furthermore, the multi-degree-of-freedom optimization module 15 is further configured to perform the following steps:

[0107] According to the first optimization space for pipe network layout adjustment, an nth pipe network layout adjustment scheme is extracted, where n is a positive integer; the nth pipe network layout adjustment scheme and the target rainfall scenario are input into the pipe network loss prediction model to obtain an nth pipe network loss prediction result, wherein the nth pipe network loss prediction result includes the nth drainage waterlogging loss coefficient, the nth drainage underground subsidence loss coefficient and the nth drainage pipe loss coefficient; according to the multidimensional indicators for pipe network loss prediction, a multidimensional constraint for pipe network loss prediction is set; it is determined whether the nth pipe network loss prediction result satisfies the multidimensional constraint for pipe network loss prediction; if the nth pipe network loss prediction result satisfies the multidimensional constraint for pipe network loss prediction, the nth pipe network layout adjustment scheme is added to the second optimization space for pipe network layout adjustment.

[0108] Furthermore, the pipe network state compensation module 16 is further configured to perform the following steps:

[0109] The drainage network is detected for abnormal state according to the multi-dimensional data of the network monitoring to obtain the abnormal state detection result of the network; the first strategy for optimizing the network layout is compensated and optimized according to the abnormal state detection result of the network to generate the second optimization strategy for the network optimization.

[0110] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0111] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

[0112] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. The urban drainage network optimization method based on big data is characterized by: include: Perform differential clustering on the historical rainfall event set of the target city to obtain multiple characteristic rainfall scenes; Performing drainage response compliance calculations on the drainage network of the target city according to the multiple characteristic rainfall scenarios to obtain drainage response compliance coefficients for each scenario; Perform compliance tests on the multiple characteristic rainfall scenarios according to the drainage response compliance coefficients of the scenarios to generate a target rainfall scenario; Obtaining the above-ground and underground datasets of the target city, adaptively adjusting the layout of the drainage network based on the target rainfall scenario, and establishing a network layout adjustment space; Performing multi-degree-of-freedom optimization on the pipe network layout adjustment space according to the pipe network loss prediction model to generate a first pipe network layout optimization strategy; The multi-dimensional data of pipe network monitoring of the drainage pipe network is obtained, and the first pipe network layout optimization strategy is optimized for pipe network status compensation based on the multi-dimensional data of pipe network monitoring to obtain a second pipe network optimization strategy.

2. The urban drainage network optimization method based on big data according to claim 1, characterized in that: Differential clustering is performed on the historical rainfall event set of the target city to obtain multiple characteristic rainfall scenarios, including: Performing pairwise difference evaluation based on the historical rainfall event set to obtain a first differential detection distribution; classifying the historical rainfall event set according to the first differential detection distribution to obtain a plurality of rainfall event classes; performing inter-class difference evaluation on the multiple rainfall event classes to obtain a second differential detection distribution, and performing inter-class fusion on the multiple rainfall event classes based on the second differential detection distribution to obtain multiple rainfall event areas; Rainfall feature recognition and fusion are performed on each rainfall event area in the multiple rainfall event areas to generate the multiple characteristic rainfall scenes.

3. The urban drainage network optimization method based on big data according to claim 1, characterized in that: Calculating the flood drainage response compliance of the drainage network of the target city according to the multiple characteristic rainfall scenarios to obtain a flood drainage response compliance coefficient for each scenario includes: Extracting a first characteristic rainfall scene according to the multiple characteristic rainfall scenes; Searching the drainage response duration of the drainage pipe network according to the first characteristic rainfall scenario to obtain a drainage duration set of the first rainfall scenario; Selecting the first rainfall scenario drainage duration set according to the standard drainage response duration to obtain a first standard drainage duration space that is less than or equal to the standard drainage response duration; Calculating the number of samples for the first rainfall scenario drainage duration set and the first standard-reaching drainage duration space to obtain the number of first scenario duration samples and the number of first standard-reaching duration samples; The proportion of the number of samples of the first compliance time and the number of samples of the first scene duration is calculated to obtain the first scene drainage response compliance coefficient, and the first scene drainage response compliance coefficient is added to the drainage response compliance coefficients of each scene.

4. The urban drainage network optimization method based on big data according to claim 1, characterized in that: Obtaining the above-ground and underground datasets of the target city, adaptively adjusting the layout of the drainage network in combination with the target rainfall scenario, and establishing a network layout adjustment space, including: Modeling is performed according to the pipe network layout plan of the drainage pipe network to obtain a three-dimensional pipe network model, and the three-dimensional pipe network model is expanded according to the above-ground and underground data sets to generate an urban drainage pipe network model; Simulating drainage of the urban drainage pipe network model according to the target rainfall scenario to obtain simulated drainage data of the pipe network; Perform risk identification based on the simulated drainage data of the pipe network to obtain a pipe network drainage risk identification result; The pipe network layout plan is adjusted according to the pipe network drainage risk identification result to generate the pipe network layout adjustment space.

5. The urban drainage network optimization method based on big data according to claim 1, characterized in that: Performing multi-degree-of-freedom optimization on the pipeline network layout adjustment space according to the pipeline network loss prediction model to generate a first pipeline network layout optimization strategy, including: Optimizing the drainage response of the pipe network layout adjustment space according to the standard drainage response time to establish a first optimization space for pipe network layout adjustment; Performing pipeline loss prediction optimization on the first optimization space for pipeline network layout adjustment according to the pipeline network loss prediction model to obtain a second optimization space for pipeline network layout adjustment; weight allocation is performed according to the multi-dimensional indicators of pipeline network loss prediction of the pipeline network loss prediction model to obtain a comprehensive pipeline network loss calculation model; The pipeline network comprehensive loss calculation model is used to minimize the pipeline network comprehensive loss in the second optimization space of the pipeline network layout adjustment to obtain the first strategy for optimizing the pipeline network layout.

6. The urban drainage network optimization method based on big data according to claim 5, characterized in that: Optimizing the drainage response of the pipe network layout adjustment space according to the standard drainage response time to establish a first optimization space for pipe network layout adjustment, including: Based on the target rainfall scenario, predicting the drainage response time for each pipe network layout adjustment scheme within the pipe network layout adjustment space to obtain multiple predicted drainage response times; Determining whether the multiple predicted drainage response time periods are less than or equal to the standard drainage response time periods, and obtaining multiple drainage response determination results; The pipe network layout adjustment space is screened according to the multiple drainage response judgment results to obtain the first optimization space for pipe network layout adjustment.

7. The urban drainage network optimization method based on big data according to claim 5, characterized in that: Performing pipeline loss prediction optimization on the first optimization space for pipeline network layout adjustment according to the pipeline network loss prediction model to obtain a second optimization space for pipeline network layout adjustment includes: According to the first optimization space for pipe network layout adjustment, extracting an nth pipe network layout adjustment scheme, where n is a positive integer; Inputting the nth pipe network layout adjustment plan and the target rainfall scenario into the pipe network loss prediction model to obtain an nth pipe network loss prediction result, wherein the nth pipe network loss prediction result includes an nth drainage waterlogging loss coefficient, an nth drainage underground settlement loss coefficient, and an nth drainage pipeline loss coefficient; Setting multidimensional constraints for pipeline network loss prediction based on the multidimensional indicators for pipeline network loss prediction; Determining whether the nth pipeline network loss prediction result satisfies the pipeline network loss prediction multidimensional constraint; If the nth pipeline loss prediction result satisfies the pipeline loss prediction multi-dimensional constraint, the nth pipeline network layout adjustment scheme is added to the second optimization space for pipeline network layout adjustment.

8. The urban drainage network optimization method based on big data according to claim 1, characterized in that: The first pipeline layout optimization strategy is optimized for pipeline status compensation based on the pipeline network monitoring multi-dimensional data to obtain a second pipeline network optimization strategy, including: Performing state abnormality detection on the drainage pipe network according to the pipe network monitoring multi-dimensional data to obtain a pipe network state abnormality detection result; The first pipeline network layout optimization strategy is compensated and optimized according to the pipeline network status abnormality detection result to generate the second pipeline network optimization strategy.

9. The urban drainage network optimization method based on big data according to claim 1, characterized in that: The plurality of characteristic rainfall scenarios are subjected to compliance inspection according to the drainage response compliance coefficients of the scenarios to generate a target rainfall scenario, including: Determine whether the drainage response compliance coefficient of each scenario is less than the drainage response compliance threshold, and obtain multiple drainage response compliance test results; Filtering the plurality of characteristic rainfall scenarios according to the plurality of drainage response compliance test results to obtain a filtered set of rainfall scenarios with unqualified responses; The target rainfall scene is obtained by fusing the screening set of rainfall scenes with unqualified responses.

10. The urban drainage network optimization device based on big data is characterized by: The device is used to execute the urban drainage network optimization method based on big data according to any one of claims 1 to 9, and the device comprises: A rainfall scene sorting module is used to perform differential clustering on the historical rainfall event set of the target city to obtain multiple characteristic rainfall scenes; a drainage response compliance calculation module, configured to perform drainage response compliance calculation on the drainage network of the target city according to the plurality of characteristic rainfall scenarios, and obtain a drainage response compliance coefficient for each scenario; a compliance inspection module, configured to perform compliance inspections on the plurality of characteristic rainfall scenarios according to the drainage response compliance coefficients of the scenarios to generate a target rainfall scenario; a pipe network layout adjustment module, the pipe network layout adjustment module being used to obtain above-ground and underground datasets of the target city, perform adaptive layout adjustment on the drainage pipe network in combination with the target rainfall scenario, and establish a pipe network layout adjustment space; a multi-degree-of-freedom optimization module, configured to perform multi-degree-of-freedom optimization on the pipe network layout adjustment space according to a pipe network loss prediction model, and generate a first pipe network layout optimization strategy; The pipe network status compensation module is used to obtain the pipe network monitoring multi-dimensional data of the drainage pipe network, and perform pipe network status compensation optimization on the first pipe network layout optimization strategy based on the pipe network monitoring multi-dimensional data to obtain the second pipe network optimization strategy.

Images