A method and system for operation and maintenance management of urban drainage facilities based on digital twins

Abstract

This invention relates to the field of urban drainage facility operation and maintenance management technology, specifically a method and system for urban drainage facility operation and maintenance management based on digital twins. The method includes: reconstructing multi-source operational data of the drainage system into event-based data, identifying the continuous impact relationships between events, forming a drainage memory chain, and constructing a digital twin foundation; identifying the residual impacts left by historical events, constructing a residual impact field, generating the facility's inertial state through coupling analysis, and merging it with the real-time digital twin state to form an inertial digital twin; further conducting potential risk screening, constructing a comprehensive risk index and establishing a risk evolution chain, and calculating operation and maintenance priorities; digitally twinning operation and maintenance resources, identifying key propagation nodes to generate minimum intervention paths, autonomously orchestrating tasks, and optimizing based on execution feedback loops. This invention realizes the transformation of urban drainage facilities from presenting the current state to predicting future trends and perceiving potential risks, providing predictive decision-making basis for operation and maintenance scheduling.

Description

Technical Field

[0001] This invention relates to the field of urban drainage facility operation and maintenance management technology, specifically to a method and system for urban drainage facility operation and maintenance management based on digital twins. Background Technology

[0002] As the scale and complexity of urban drainage facilities continue to increase, their operational status directly affects urban safety and residents' quality of life; in recent years, digital twin technology has been increasingly widely used in the field of urban water affairs.

[0003] Chinese invention patent application CN117132264A discloses a digital twin-based urban drainage facility operation and maintenance management system, including a real-world scenario and equipment, a digital twin model, a data platform, a business scenario operation platform, and a basic platform. The real-world scenario and equipment consist of drainage pipes, the surrounding working environment, corresponding monitoring equipment, and pipe network detection equipment. The digital twin model includes a pipe digital twin model and a pipe robot digital twin model, which can be divided into a three-dimensional information model, a principle calculation model, and a behavior evolution model. The data platform provides the digital twin system with accurate, large-scale, comprehensive, and timely data for data storage, analysis, and decision support. The business scenario operation platform provides customers with detailed pipe conditions and maintenance guidance. The basic platform connects the real-world scenario and the data platform, transmits decision-making and execution instructions from the business scenario operation platform, and collects and transmits corresponding detection and monitoring data.

[0004] The aforementioned digital twin applications can effectively reflect the real-time operational status of drainage facilities, providing technical support for daily inspections and emergency responses. However, in the actual operation of urban drainage systems, multiple historical events such as rainstorm erosion, siltation, and equipment overload can have a continuous and cumulative impact. How to systematically integrate the long-term inheritance effects of these historical impacts into the digital twin model and use it to predict future hidden risks and optimize operation and maintenance decisions remains a direction worthy of further exploration. Based on this, how to deeply integrate the impact transmission patterns of historical events with real-time twin models to build an operation and maintenance management system with memory and inertial cognitive capabilities has become an important issue in improving the intelligence level of urban drainage facilities. Summary of the Invention

[0005] The purpose of this invention is to address the problems existing in the background technology by proposing a method and system for the operation and maintenance management of urban drainage facilities based on digital twins.

[0006] The technical solution of this invention: a method for operation and maintenance management of urban drainage facilities based on digital twins, comprising the following specific implementation steps: S1. Reconstruct the multi-source operation data of the drainage system through event-based methods, identify the continuous impact relationships between events, form a drainage memory chain with inheritance attributes, integrate real-time status and historical impact information, and construct a digital twin foundation. S2. Based on the digital twin, identify the residual impact of historical events and quantify it as residual impact sources, construct the event residual impact field, generate the facility inertial state through coupling analysis, and merge the facility inertial state with the real-time digital twin state to form an inertial digital twin. S3. Based on the inertial digital twin, potential risks are screened, a comprehensive risk index is constructed and a risk evolution chain is established, operation and maintenance priorities are calculated, and priority ranking and blocking strategies for operation and maintenance resource allocation are formed. S4. Based on the risk evolution chain, digital twins of operation and maintenance resources are generated to identify key propagation nodes and generate minimal intervention paths. Operation and maintenance tasks are autonomously arranged, and the risk chain and resource capabilities are updated through execution feedback to achieve adaptive optimization.

[0007] Preferably, step S1 specifically includes: Perform time alignment and standardization on multi-source data, divide independent drainage behavior event units according to the characteristics of facility operation status changes, and construct standardized event objects; Based on the pipeline topology and facility connection relationships, the continuous impact transmission relationship between drainage behavior events is analyzed, and events with significant transmission relationships are transformed into memory nodes to form an event network structure. Starting from the current state of the facilities, we trace back to historical memory nodes and construct a drainage memory chain formed by linking multiple historical events according to their impact and contribution. By weighted and fused with the real-time operating status of the facility and the historical inherited information extracted from the drainage memory chain, a digital twin foundation is constructed.

[0008] Preferably, in the process of constructing standardized event objects, an event impact intensity index is constructed based on the flow fluctuation amplitude, water level fluctuation amplitude, siltation change and facility recovery time, combined with the flow impact weight coefficient, water level impact weight coefficient, siltation impact weight coefficient and recovery time weight coefficient. Among them, the flow rate impact weight coefficient comes from the statistical regression analysis of historical rainstorm events or the hydraulic sensitivity calibration results of different regions; the water level impact weight coefficient comes from the analysis of historical waterlogging and overflow events; the siltation impact weight coefficient comes from the statistical data of long-term pipeline blockage and dredging; and the recovery time weight coefficient comes from the statistical data of operation and maintenance response efficiency. In the process of analyzing the transmission of lasting impact, the event transmission impact value is constructed based on the event correlation coefficient and time interval; Among them, the event correlation coefficient is derived from the drainage network topology, hydraulic path analysis results, or historical flow data; In the process of constructing the drainage memory chain, the influence retention coefficient is introduced to calculate the memory inheritance value; the influence retention coefficient is dynamically adjusted according to the facility type, maintenance frequency, pipe diameter characteristics and historical treatment effects.

[0009] Preferably, step S2 specifically includes: The drainage memory chain nodes are analyzed to identify the residual effects that still exist after the event ends. These residual effects are then decomposed into hydraulic residual effect factors, sedimentary residual effect factors, structural residual effect factors, and operation and maintenance residual effect factors. The residual effect retention values ​​are calculated. The residual impact sources are mapped to the drainage network space, and the residual impact field is constructed based on the pipe network topology, water flow direction and confluence area. The impact field intensity of each facility unit is then calculated. Coupled analysis of multiple residual influence fields is performed to calculate the inertial state value of the facility and quantify the cumulative effect and potential risk trend of different historical influences; By fusing the inertial state of the facility with the real-time digital twin state, an inertial digital twin is generated, forming a future state representation model.

[0010] Preferably, the calculation of the residual effect retention value takes into account the memory inheritance value, the interval from the end of the event to the current time, and the residual effect decay coefficient; The residual effect attenuation coefficient is determined by the facility type, pipe diameter, maintenance frequency, and environmental conditions; The calculation of the field strength takes into account the propagation weight, which is determined by the network connectivity coefficient, hydraulic capacity coefficient, and local resistance coefficient. Among them, the connectivity coefficient of the pipeline network structure is determined by whether the topological relationship is directly connected and the connection level, the hydraulic capacity coefficient is calculated by historical flow capacity and pipe diameter parameters, and the local resistance coefficient is jointly determined by sediment thickness, number of bends and slope changes. In the coupling analysis, a coupling enhancement effect is constructed for any two types of residual influence fields, and a coupling coefficient is introduced to characterize their interaction relationship. The coupling coefficient is obtained by statistical analysis of historical operating data. The calculation of the facility's inertial state value integrates the intensity of various residual influence fields and coupled influences, introduces single influence weight coefficients and coupled influence weights, and performs time evolution analysis on the inertial state value to obtain the facility's inertial evolution rate.

[0011] Preferably, step S3 specifically includes: Based on inertial digital twins, potential risks of drainage facilities are screened. A comprehensive risk index is constructed by introducing inertial growth trend, deposition evolution rate and structural fatigue accumulation factors, and risk level classification and ranking of each facility node is carried out. Based on the pipeline network topology, hydraulic flow direction, and residual distribution of inertial influence, the dependencies between risk nodes are analyzed, a chain propagation path from the risk source node to the terminal manifestation node is constructed, and the risk transmission intensity between nodes is quantified through a propagation weight model to form a risk evolution chain. Accumulated risk calculations are performed on each node in the risk evolution chain, and the growth rate and trend of the risk are analyzed in conjunction with time series analysis to determine whether the risk has entered a critical state. Operation and maintenance priorities are calculated based on cumulative risk value and growth rate. Key propagation nodes in the risk chain are identified as priority intervention targets, and blocking operation plans are generated in combination with operation and maintenance resource constraints.

[0012] Preferably, the construction of the comprehensive risk index incorporates sedimentation growth trend factors, hydraulic response hysteresis factors, and structural fatigue accumulation factors, and combines the weight coefficients corresponding to each factor. The weighting coefficients are obtained from historical operation and maintenance samples. The calculation of propagation weights takes into account the facility inertial digital twin value, hydraulic path distance, pipeline network structure coupling coefficient and standardization coefficient. The facility inertial digital twin value is obtained by fusing the digital twin base state and the inertial state, and the hydraulic path distance reflects the actual drainage path length between nodes; The calculation of the cumulative risk value takes into account the set of all affected nodes. The cumulative risk value is calculated on multiple time slices to form a risk evolution curve. When the growth rate of the cumulative risk value exceeds the historical average or multiple upstream nodes rise synchronously within a continuous time period, it is determined that the node has entered a critical evolution state and an early warning is triggered. Operation and maintenance priorities are calculated based on cumulative risk value and evolution speed, combined with weighting coefficients.

[0013] Preferably, step S4 specifically includes: The system digitally twins personnel, vehicles, equipment, and storage resources in urban drainage operation and maintenance, records real-time status and historical operational capabilities, calculates resource capability indices, and maps them to facility risk nodes. The risk blocking contribution of each node is calculated based on the risk evolution chain, key propagation nodes are identified, and the minimum intervention path is generated, while node intervention strategies are planned. Based on the minimum intervention path and resource capacity index, the system automatically generates operation and maintenance tasks, calculates task priorities, completes resource matching, job sequence planning and route arrangement, and dynamically adjusts them according to real-time data during task execution. Real-time collection of task execution data, assessment of risk reduction and resource consumption, and feedback of results to the digital twin enable closed-loop optimization and adaptive adjustment.

[0014] Preferably, the calculation of the resource capacity index considers the unit time operation capacity, historical operation efficiency index and real-time schedulable capacity, and combines their respective weight coefficients. The weight coefficients are calibrated from historical operation and maintenance performance data. The unit time operation capacity includes the dredging volume per hour, the historical operation efficiency index includes the task completion rate to the project duration ratio, and the real-time schedulable capacity considers the current location, availability and standby capacity of the resources. The risk blocking contribution is calculated based on the risk propagation weight from node to downstream node. By ranking the contribution and resource availability, the minimum intervention path is automatically generated to achieve the maximum risk propagation blocking effect with the least resource input. The task priority index is calculated based on the operation and maintenance priority index, risk blocking contribution and corresponding weight coefficients. Factors considered include operation time window, traffic conditions, weather impact and equipment availability. During execution, the task order and resource allocation are dynamically adjusted by accessing pipeline water level data, flow data and sedimentation change data in real time.

[0015] The technical solution of this invention: A digital twin-based operation and maintenance management system for urban drainage facilities, comprising: Memory; processor; A computer program stored in the memory and capable of running on the processor; When the processor executes the computer program, it implements the above-described method for the operation and maintenance management of urban drainage facilities based on digital twins.

[0016] Compared with the prior art, the above-mentioned technical solution of the present invention has the following beneficial technical effects: This invention designs a method and system for the operation and maintenance management of urban drainage facilities based on digital twins. By constructing a drainage memory chain structure, the traditional digital twin model, which only focuses on real-time status, is extended to a comprehensive model that integrates the transmission of historical event impacts and state inheritance. This allows the system to trace the complete formation process of the facility's current state, significantly improving the ability to identify hidden risk sources and avoiding blind spots in operation and maintenance caused by ignoring historical cumulative effects. By introducing residual influence fields and inertial digital twins, the four types of residual influences—hydraulic, sedimentary, structural, and operational—are spatially expressed and coupled for analysis. This enables the early detection of inertial risks that have not yet manifested as faults, achieving proactive prediction of the facility's future operational trends, rather than passive response. Based on the risk evolution chain, comprehensive risk index calculation and propagation path analysis can accurately identify key propagation nodes in the risk chain, thereby formulating a minimum intervention path under limited resource conditions, effectively blocking risk spread, and improving the pertinence and efficiency of operation and maintenance decisions. By digitally twinning operation and maintenance resources such as personnel, vehicles, and equipment, and combining them with the risk evolution chain, autonomous task orchestration, dynamic resource matching, and execution feedback loops are achieved, forming a data-driven proactive operation and maintenance system. This continuously optimizes future strategies, reduces overall operation and maintenance costs, extends the service life of drainage facilities, and enhances the resilience and reliability of urban drainage systems in complex environments. This invention significantly improves the intelligence, foresight, and refinement of urban drainage facility operation and maintenance management. Attached Figure Description

[0017] Figure 1 This is a flowchart of a digital twin-based operation and maintenance management method for urban drainage facilities proposed in this invention. Detailed Implementation

[0018] Example 1, as Figure 1 As shown, the present invention proposes a method for the operation and maintenance management of urban drainage facilities based on digital twins, which includes the following specific implementation steps: S1. By expanding the traditional digital twin model centered on real-time status to a three-layer structure of historical events, impact transmission, and status inheritance: Multi-source operational data of the drainage system are reconstructed using event-based methods to identify the continuous impact relationships between events, forming a drainage memory chain with inheritance attributes. Based on this, real-time status and historical impact information are integrated to construct a digital twin foundation that reflects the formation process of facility status. The specific implementation process is as follows: S11. Perform time alignment and standardization on multi-source data such as flow, water level, rainfall, and operation and maintenance records in the urban drainage system. Based on the characteristics of changes in facility operation status, divide the data into independent drainage behavior event units, extract key indicators such as flow fluctuations, water level changes, siltation growth, and recovery time, and construct standardized event objects with a unified expression form. Specifically: Establish a data access system covering the entire life cycle of urban drainage facilities, and unify the data from pipeline flow monitoring equipment, water level monitoring equipment, pump station control systems, level gauges, rain gauges, inspection terminals, maintenance work order systems, and historical accident archives into the management platform, and process them synchronously according to a unified time benchmark; Since different data sources have different collection frequencies, for example, traffic data is usually updated on a minute-by-minute basis, while inspection records may be updated on a daily or weekly basis. Therefore, the data of various types are first time-aligned and status-compensated to form a continuous facility operation sequence dataset. After obtaining continuous operational data, identify the key processes that truly affect the operational status of the facility from an operations and maintenance management perspective. Specifically, based on changes in facility status, when there are significant changes in flow rate, water level, siltation degree, pump station load, or discharge capacity, the corresponding operating interval is automatically divided and defined as a drainage behavior event unit. For example, if during a heavy rainfall event, the flow rate in a certain pipe section increases rapidly, the water level rises continuously, and remains at a high level for a long time after the rainfall ends, then the entire process is identified as an independent drainage event; or if a manhole's drainage capacity decreases due to garbage accumulation, and this has a lasting impact before subsequent cleaning is completed, then it is defined as a siltation evolution event. To quantify the impact of an event on the facility's condition, an event impact intensity index is constructed: ; In the formula, It represents the intensity of the impact of a drainage event, describes the comprehensive impact of the i-th drainage behavior event on the facility's operational status, and is used to uniformly quantify the disturbance level of different types of events (rainstorms, siltation, overflows, emergency repairs, etc.) on drainage capacity; This represents the amplitude of flow fluctuation, describing the degree of flow change in the drainage network or pumping station during the i-th event. Indicates the amplitude of water level fluctuations, describing the magnitude of changes in the water level inside the facility during the event; Indicates the amount of sedimentation change, describing the degree of change in sediment (mud, garbage, etc.) in pipelines or inspection wells during an event; Indicates facility recovery time, describing the time required for the drainage system to recover from an event-induced impact state to a stable operating state; The flow rate impact weighting coefficient represents the importance of flow rate fluctuations in the intensity of an event's impact. It is derived from: statistical regression analysis of historical rainstorm events and calibration results of hydraulic sensitivity in different regions. The water level influence weighting coefficient represents the weight of water level changes on facility operation, and is derived from the analysis of historical flooding and overflow events. The weighting coefficient for siltation influence represents the weight of sediment changes on long-term drainage capacity. It is derived from long-term pipeline blockage and dredging data statistics. The recovery time weighting coefficient represents the weight of the system's recovery capability in the impact of the event, and is derived from: Operation and maintenance response efficiency statistics; S12. Based on the pipeline topology and facility connection relationships, analyze the continuous impact transmission relationship between drainage behavior events. By calculating the relationship between event impact intensity and time decay, identify the contribution of preceding events to subsequent events, and transform events with significant transmission relationships into memory nodes to form an event network structure with influence inheritance attributes. Specifically: After completing the construction of the event units, further analysis is conducted to determine whether there are any lasting impact relationships between the events, with a focus on identifying the long-term impact transmission relationships between the events. Specifically, each drainage behavior event unit is taken as the analysis object. Combining the pipeline topology, drainage flow direction relationship, facility connection relationship and historical maintenance records, the contribution of the preceding event to the subsequent event is analyzed. When a certain historical event has a lasting impact on the subsequent event, the inheritance relationship between the two is established. To characterize this influence relationship, an event propagation influence value is constructed: ; In the formula, This represents the impact value of event propagation, describing the degree of sustained influence of the i-th event on the j-th event; The correlation coefficient represents the degree of structural or functional correlation between two events and is derived from: drainage network topology, hydraulic path analysis results, and historical flow data. This represents the time interval, i.e., the time difference between event i and event j; It should be noted that the degree of correlation is not simply spatial distance, but a result that takes into account pipeline connections, water flow paths, common confluence areas, and historical patterns of influence propagation. For example, two adjacent manholes may be close in distance, but if there is no direct drainage connection, their degree of correlation is low; while facilities that are far apart but located on the same confluence path may have a higher degree of correlation. When the transmitted impact value reaches a preset threshold, the corresponding event is transformed into a drainage memory node, and its impact source, affected object and impact path are recorded to form a drainage memory network with transmission attributes. S13. Using the current facility status as the endpoint, trace back the historical memory nodes to construct a drainage memory chain formed by linking multiple historical events according to their impact contribution. Introduce an impact retention coefficient to quantify the degree of continued influence of historical events in the current status, thereby forming a historical inheritance trajectory structure that can express the formation path of the facility status, specifically: After the drainage memory node is formed, a drainage memory chain is further established according to the direction of influence transmission: starting from the current facility status, all historical events that contributed to the current status are traced back and a complete inheritance trajectory is established. For example, if a certain area is at risk of flooding, not only is the current flooding phenomenon recorded, but the historical process that led to the formation of the risk is also traced back, including factors such as the accumulation of silt caused by previous rainfall, local blockages that were not dealt with in time, and the decline in the discharge capacity of upstream facilities. These events are then linked together according to their impact transmission relationship to form a complete memory chain. To measure the extent to which historical influences are retained in the current state, a memory inheritance value is constructed: ; In the formula, This represents the memory inheritance value, which is the cumulative degree of historical impact contained in the current state of the drainage facility; The impact retention coefficient represents the degree to which the impact of historical events remains in the current system. It is derived from: operation and maintenance records (dredging, repair, and renovation) and changes in facility structure; n represents the number of historical events, i.e., the total number of historical events that participated in the formation of the current state. It should be noted that the retention factor is affected The retention coefficient is dynamically adjusted based on facility type, maintenance frequency, pipe diameter characteristics, and historical treatment effects. For example, after dredging is completed, some historical impacts are removed, resulting in a decrease in the retention coefficient. If maintenance is neglected for a long period, the retention coefficient remains at a high level. S14. The real-time operational status of the facility is weighted and fused with historical inherited information extracted from the drainage memory chain to construct a digital twin foundation that combines real-time performance with the ability to express historical impact. This allows the model to not only reflect the current operational status but also simultaneously present the state formation process and the sources of historical impact, providing a unified state expression basis for subsequent operation and maintenance decisions. Specifically: After obtaining the drainage memory chain and historical inheritance trajectory, we began to build the digital twin foundation, introducing the historical inheritance state as an important component of the digital twin model, so that the model has both real-time mapping capability and historical cognition capability. Specifically, real-time monitoring status, facility structural attributes, historical operation and maintenance information, and inherited status data will be integrated and calculated to establish a unified facility status expression model. This will construct a digital twin of the facility's basic status values ​​(which can not only display real-time information such as current flow, water level, and equipment operating status, but also simultaneously present the historical sources, paths, and degree of inheritance of the influence that led to the current status): ; In the formula, It represents the basic state value of the digital twin, that is, the comprehensive digital twin expression after integrating real-time state and historical memory; This indicates the real-time operating status value, that is, the actual operating status of the facility at the current moment; This represents the weighting factor of the real-time state, which is the fusion ratio of the real-time state and historical memory in the twin model. It is set according to the facility type and operating characteristics.

[0019] S2. Based on the drainage memory chain-driven digital twin foundation constructed in step S1, the hidden risks left over from historical events are identified and quantified as residual impact sources. An event residual impact field is constructed, and the coupling effect of different impacts is analyzed to generate the facility inertial state. Finally, the real-time digital twin state is fused to form an inertial digital twin, realizing the prediction of the future operation trend of urban drainage facilities and the perception of potential risks. The specific implementation process is as follows: S21. Analyze the drainage memory chain nodes formed in step S1, identify the residual effects that still exist after the event ends, and decompose them into four categories of influencing factors: hydraulic, sedimentary, structural, and operational factors. Quantify the continuing effect of historical events through the residual effect retention value, specifically: For the digital twin foundation formed in step S1, and for each memory node, analyze its legacy effects: For each event node in the drainage memory chain, identify its impact that still exists at the current time point based on the event type, duration, intensity, and maintenance records; For example, if the sediment at the bottom of the pipe is not cleaned after a rainstorm, its sedimentation effect will continue to reduce the drainage capacity of the pipe; after a pumping station is overloaded, its mechanical parts will still have wear accumulation; and incomplete cleaning of local blockages will cause slight changes in the water flow path and affect drainage efficiency in the long term. The residual effects are broken down into four basic impact factors: (1) Residual hydraulic factors: reflect the continuous impact of historical events on flow rate, water level and pipeline flow structure, such as water stagnation, head increase or velocity decrease; (2) Influencing factors of sedimentary residue: reflecting the long-term occupation of drainage cross-sectional area, flow velocity and silt accumulation by historical sedimentation; (3) Residual structural factors: reflecting the impact of long-term loads, scouring and material fatigue on the structural degradation of facilities such as pipelines and pumping stations; (4) Residual impact factors of operation and maintenance: reflecting potential risks that have not been completely eliminated by historical maintenance work, such as insufficient dredging, equipment not being properly inspected or missed during inspections; For the lingering impact of each historical event, calculate the residual impact retention value: ; In the formula, This represents the current residual impact value of the i-th historical event; This represents the memory inheritance value calculated in step S1; Indicates the interval between the end of the event and the current time; The residual effect attenuation coefficient is determined by the facility type, pipe diameter, maintenance frequency, and environmental conditions. S22. Map the residual impact sources to the drainage network space, construct the residual impact field based on the pipe network topology, water flow direction, and confluence area, and calculate the impact field intensity of each facility unit to achieve spatial and quantitative expression of the impact, specifically: After identifying the sources of residual impact, they are mapped onto the drainage network space to construct a complete residual impact field: Centered on each residual impact source node, the impact is propagated to relevant facility units based on the pipeline network topology, water flow direction, pipe diameter changes, and confluence area, forming a spatial distribution map. This mapping can reflect the local concentration and diffusion of the impact of historical events and reveal potential risk areas. Define the influence field intensity: ; ; In the formula, This indicates the intensity of the residual impact of historical event i on facility unit j; The propagation weight is determined by comprehensively considering hydraulic connections, flow direction, pipe diameter attenuation, and confluence relationships. This represents the connectivity coefficient of the pipeline network structure, which is determined by the topological relationship (whether it is directly connected and the connection level). This represents the hydraulic capacity coefficient, which is calculated from historical flow capacity and pipe diameter parameters. This represents the local resistance coefficient, which is determined by the deposition thickness, the number of bends, and the slope variation. Based on the deconstruction results of the influencing factors, hydraulic, sedimentary, structural, and operational residual influence fields are generated respectively. S23. Perform coupled analysis on multiple residual influence fields, calculate the facility's inertial state value, quantify the cumulative effect and potential risk trend of different historical influences, reveal latent risks that have not yet manifested as failures through inertial state analysis, provide a basis for subsequent future state prediction and operation and maintenance decisions, and realize dynamic management of the cumulative effect of historical events, specifically: Based on the various residual influence fields formed in step S2, an in-situ analysis is performed on the superposition state of the same facility unit in different influence fields. Taking drainage facilities (pipe sections, inspection wells, pumping station units) as the basic analysis unit, the hydraulic residual influence field, sedimentary residual influence field, structural residual influence field, and operation and maintenance residual influence field are mapped and superimposed in the same spatial coordinate system to identify their overlapping areas. The following three types of key coupling areas are identified in particular: Strong overlap region: A region where multiple influencing fields have high intensity and are spatially consistent; Dominant bias region: The region where a certain type of influencing field is clearly dominant but is continuously disturbed by other influencing fields; Weakly coupled diffusion zone: A region where a single influence dominates but has a potential diffusion trend; Further analysis of the interaction relationships between different residual effects reveals the introduction of an effect enhancement factor to describe the mutual reinforcement or inhibition relationships between different effects. For any two types of residual effect fields... and Construct their interaction relationships: ; In the formula, This indicates the enhanced coupling effect between influencing fields a and b; This represents the coupling coefficient, used to characterize the interaction between two types of influences; The coupling coefficient was obtained from historical operational data. For example, the relationship between sedimentary and hydraulic effects is usually positive feedback (decreased flow velocity → increased sedimentation); the relationship between structural and hydraulic effects is one of hysteresis-enhancing (high water level → accumulated structural fatigue); and operational and maintenance effects have a weakening effect on the other three types of effects (dredging or repair reduces the intensity of the effect). After obtaining the intensity of various individual influences and their interactive enhancement effects, all influences are integrated to construct the facility-level inertial state quantity and define the comprehensive inertial influence function: ; In the formula, This represents the inertial state value of facility k; Indicates the intensity of the m-th type of residual influence field; This represents the weighting coefficient for a single influence. This indicates the coupling effect weight; n represents the number of effect types (4 types in this example); Furthermore, a time evolution analysis is performed on the inertial state values, by analyzing the values ​​within a continuous time window. The changing trend is calculated to obtain the facility's inertial evolution rate, which is used to identify the rate of risk accumulation. For example: if A steady increase indicates that the facility is entering a phase of risk accumulation; if A sudden increase indicates the presence of multiple sources influencing sudden coupling; if A continuous decline indicates that the operation and maintenance intervention has been effective. S24. The inertial state of the facility is fused with its real-time digital twin state to generate an inertial digital twin, forming a future state representation model. This model predicts potential siltation growth, water level changes, structural degradation, and maintenance needs, enabling the digital twin system to shift from presenting the current state to perceiving the future state. Specifically: By integrating inertial states into the digital twin foundation, an inertial digital twin with the ability to perceive future states is formed: The digital twin base state generated in step S1 With respect to the inertial state in step 23 Fusion, computing facility inertial digital twin values: ; In the formula, Represents the digital twin value of facility inertia; This indicates the real-time status weight, reflecting the degree to which the current monitoring data contributes to future trends; Based on inertial digital twins, a future operational profile is generated, including: potential areas of siltation growth, locations where water levels may exceed limits in the future, structural degradation trends, and early warnings of maintenance needs.

[0020] S3. Based on the inertial digital twin formed in step S2, systematically identify and model the potential risks of urban drainage facilities. By constructing a risk indicator system to screen risk nodes and establishing a risk evolution chain in conjunction with the pipe network topology, further analyze the cumulative growth and propagation trends of risks over time. This leads to priority ranking and blocking strategies for operation and maintenance resource allocation, transforming the approach from single-point risk assessment to chain-like risk evolution decision-making. This provides predictive decision-making basis for the operation and maintenance scheduling of urban drainage systems. The specific implementation process is as follows: S31. Based on inertial digital twins, potential risks of drainage facilities are screened. This involves not only considering the current abnormal state but also introducing multi-dimensional factors such as inertial growth trends, depositional evolution rates, and structural fatigue accumulation to construct a comprehensive risk index. This index is used to classify and rank the risk levels of each facility node, thereby identifying key nodes with future risk evolution potential. Specifically: In inertial digital twin Based on this, a risk screening of the entire drainage network is conducted, introducing an inertial residual superposition triggering mechanism, which assumes that the risk does not arise suddenly, but gradually manifests after long-term accumulation of impact: Calculate the inertial digital twin value for each facility unit. In conjunction with fluctuations in local historical residual effects, facility nodes with a continuous upward trend are identified. Even if the current operating status is normal, if the inertia value continues to increase, they are also included in the potential risk set. Based on the selected nodes, a comprehensive risk index is constructed by incorporating structural degradation trends, sediment growth trends, and hydraulic response hysteresis. ; In the formula, This represents the comprehensive risk index of facility k; Factors indicating the depositional growth trend (reflecting the rate of sediment accumulation); Indicates the hydraulic response hysteresis factor (reflecting the degree of decline in drainage recovery capacity); This represents the structural fatigue accumulation factor (derived from the effects of long-term load). , , and This represents the weighting coefficient, which is obtained from historical operation and maintenance samples. Combined with the set risk thresholds, and according to Based on distribution characteristics, nodes are divided into three categories: "latent risk", "growing risk" and "critical risk", and their risk sensitivity in the network is marked. S32. Based on the pipeline network topology, hydraulic flow direction, and residual distribution of inertial influence, analyze the dependencies between risk nodes, construct a chain propagation path from the risk source node to the terminal manifest node, and quantify the risk transmission intensity between nodes through a propagation weight model, thereby forming a risk evolution chain structure that can express the risk diffusion process, specifically: After identifying potential risk nodes, we further analyze the transmission relationship between risks, construct a risk evolution chain, and analyze whether there is a causal dependency relationship between different risk nodes based on drainage topology, water flow direction and residual distribution of inertial influence. For example, increased upstream sedimentation can gradually lead to a decrease in downstream flow velocity, and the decrease in downstream flow velocity may in turn exacerbate upstream backflow, thus forming a two-way influence structure. Connect the risk nodes that have dependencies according to the direction of their impact to form a risk evolution chain: risk source node → intermediate propagation node → terminal manifestation node; Define propagation weight modeling: ; In the formula, This indicates the intensity of risk propagation from node i to node j; Represents the digital twin value of facility inertia; Indicates the hydraulic path distance; Indicates the coupling coefficient of the pipeline network structure; Represents the standardized coefficient; The principle of prioritizing dominant propagation paths is introduced, retaining only the propagation paths that contribute the most to terminal risks; S33. Accumulated risk calculation is performed on each node in the risk evolution chain, and its growth rate and trend are analyzed in conjunction with time series data. By identifying risk growth acceleration points and inflection points, it is determined whether the risk has entered a critical state, thereby achieving dynamic prediction and trend extrapolation analysis of the risk evolution process. Specifically: After constructing the risk evolution chain, further cumulative calculations are performed on the risks within the chain over time to determine whether the risks are accelerating their formation. Cumulative risks of compute nodes: ; In the formula, This represents the cumulative risk value of node j; Represents the set of all affected nodes; Calculations on multiple time slices This generates a risk evolution curve and analyzes its rate of change. If the curve is flat, it indicates that the risk is in a stable accumulation phase; if the slope of the curve is rising, it indicates that the risk is accelerating; if an inflection point appears, it indicates that the risk is about to erupt. An alert is triggered when the following conditions are met: within a continuous time period. If the growth rate exceeds the historical average or multiple upstream nodes rise simultaneously, then the node is determined to have entered a critical evolutionary state. S34. Calculate maintenance priorities based on risk accumulation and growth rate, identify key propagation nodes in the risk chain as priority intervention targets, and generate blocking operation plans in conjunction with maintenance resource constraints to achieve optimal resource scheduling and risk path cutoff, specifically: Based on the risk evolution chain and trend analysis results, the analysis proceeds to the operations and maintenance decision-making level, where the risk analysis results are transformed into specific operations and maintenance strategies. Priority is calculated based on a combination of cumulative risk value and evolution rate: ; In the formula, Indicates the operation and maintenance priority index; Indicates the weighting coefficient; Identify key propagation nodes in the risk chain, i.e. facilities that have the greatest impact on the overall risk spread, and prioritize interventions at these nodes. For example, upstream dredging can block the risk of downstream water accumulation. By combining personnel, vehicles, equipment, and operation time windows, the optimal execution path is generated to minimize resource input and maximize risk suppression; After the operation and maintenance are completed, the inertia value and risk propagation intensity of the corresponding node are recalculated, and the risk chain structure is updated to enable the model to continuously adapt and optimize.

[0021] S4. Based on the risk evolution chain generated in step S3, the operation and maintenance resources are digitally twinned, key propagation nodes are identified, minimal intervention paths are generated, and operation and maintenance tasks are autonomously orchestrated. At the same time, the risk chain and resource capabilities are updated through execution feedback to achieve adaptive optimization. The specific implementation process is as follows: S41. Digitally twinnize personnel, vehicles, equipment, and storage resources in urban drainage operation and maintenance, record real-time status and historical operational capabilities, calculate resource capacity indices, and map them to facility risk nodes, specifically: Real-time collection of information on all available resources in the urban drainage operation and maintenance system, including dredging vehicles, vacuum vehicles, pipeline inspection robots, inspection personnel, pump station maintenance teams, emergency equipment warehouses, and standby pumps; each resource not only records its current location, availability, and maintenance status, but also acquires historical operation data, such as average processing speed, operation coverage, task completion rate, special working condition handling capacity, and response latency. By combining historical data with real-time status, a capability index is constructed for each resource: ; In the formula, This represents the overall operational capability index of resource i; Indicates the work capacity per unit time (e.g., dredging volume / hour); This represents the historical work efficiency index (such as the ratio of task completion rate to project duration). This indicates real-time schedulable capability (considering location, status, and backup capability). , and This represents the weighting coefficient, which is determined by historical operation and maintenance performance data. Map the operation and maintenance resource capabilities one by one with the risk nodes and evolution chain nodes in step S3 to ensure that each risk node has available resources to support it. S42. Based on the risk evolution chain in step S3, calculate the risk blocking contribution of each node, identify key propagation nodes, and generate the minimum intervention path to achieve the maximum risk blocking effect with limited resources at the fewest intervention nodes. Simultaneously, plan node intervention strategies, specifically: Analyze the risk evolution chain generated in step S3 and calculate the contribution of each node to the risk propagation of the entire chain: ; In the formula, This represents the risk blocking contribution of node k; This represents the risk propagation weight from node k to downstream node j; This represents the set of downstream nodes of node k, i.e., all subsequent nodes that the risk may propagate from node k to affect in the drainage network. Based on contribution ranking and resource availability, the minimum intervention path is automatically generated, which means that the overall risk chain is blocked to the greatest extent by intervening in a few key nodes with limited resources. It should be noted that the minimum intervention path refers to the operation and maintenance path that achieves the maximum risk propagation blocking effect with the least amount of resources invested. For example, if there are originally 5 risk nodes that need to be addressed, but more than 80% of the risk can be eliminated by dealing with just one critical node, then this solution is the preferred option. Targeted intervention measures are generated at each key node, such as dredging of upstream inspection wells, advance maintenance of pumping stations and pumps, or pressure release of local pipeline sections, to ensure that the spread of risk is blocked at the source rather than just consuming resources at the end. S43. Based on the minimum intervention path and resource capacity index, automatically generate operation and maintenance tasks, calculate task priorities, complete resource matching, job sequence planning and route arrangement, and dynamically adjust according to real-time on-site data during task execution to achieve risk-driven autonomous scheduling, specifically: Based on the minimum intervention path generated in step S42, the task priority is calculated according to the node's cumulative risk value, growth rate, and resource capability index: ; In the formula, Indicates task priority index; This indicates the maintenance priority index (output of step S3). Indicates the contribution to risk prevention; , and Indicates the weighting coefficient; Automatically allocate the most suitable resource units to each task, generate job sequence and route planning, taking into account factors such as job time window, traffic conditions, weather impact and equipment availability; During the execution process, real-time access to on-site data, such as changes in pipeline water level, flow rate, and sediment volume, is used to dynamically adjust the task sequence and resource allocation, ensuring the maximum effectiveness of risk intervention while optimizing resource utilization efficiency. For example, for areas that are about to form high-risk water accumulation, the system automatically generates the following steps: clean the manholes in the morning, inspect the pipe sections in the afternoon, and complete the local repairs the next day, thereby blocking the flow of water before the risk is formed. S44. Real-time collection of task execution data, assessment of risk reduction and resource consumption, and feedback of results to the digital twin to achieve closed-loop optimization and adaptive adjustment. This enables continuous optimization of future operation and maintenance strategies and more precise risk intervention, constructing a data-driven proactive operation and maintenance closed loop, specifically: Detailed records are kept of the execution of each operation and maintenance task, including operation time, personnel, equipment usage, changes in risk index, and handling of abnormal events; The execution results and on-site data are fed back into the inertial digital twin to update the facility status, risk index, and residual impact value; The operation and maintenance execution results, risk evolution trends, and scheduling effects are compiled into an operation and maintenance experience library, which is used to: automatically adjust future risk weights, optimize task decomposition and resource matching strategies, and provide training data for machine learning models to improve prediction accuracy.

[0022] Example 2: The present invention proposes a digital twin-based urban drainage facility operation and maintenance management system, which is used to execute the digital twin-based urban drainage facility operation and maintenance management method proposed in Example 1, comprising: Memory; processor; A computer program stored in the memory and capable of running on the processor; The processor executes a computer program to implement the digital twin-based urban drainage facility operation and maintenance management method described in Embodiment 1 above.

[0023] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A digital-twin-based urban drainage facility operation and maintenance management method, characterized by, The specific implementation steps include the following: S1. Reconstruct the multi-source operation data of the drainage system through event-based methods, identify the continuous impact relationships between events, form a drainage memory chain with inheritance attributes, integrate real-time status and historical impact information, and construct a digital twin foundation. S2. Based on the digital twin, identify the residual impact of historical events and quantify it as residual impact sources, construct the event residual impact field, generate the facility inertial state through coupling analysis, and merge the facility inertial state with the real-time digital twin state to form an inertial digital twin. S3. Based on the inertial digital twin, potential risks are screened, a comprehensive risk index is constructed and a risk evolution chain is established, operation and maintenance priorities are calculated, and priority ranking and blocking strategies for operation and maintenance resource allocation are formed. S4. Based on the risk evolution chain, digital twins of operation and maintenance resources are generated to identify key propagation nodes and generate minimal intervention paths. Operation and maintenance tasks are autonomously arranged, and the risk chain and resource capabilities are updated through execution feedback to achieve adaptive optimization.

2. The urban drainage facility operation and maintenance management method based on digital twinning according to claim 1, characterized in that, Step S1 specifically includes: Perform time alignment and standardization on multi-source data, divide independent drainage behavior event units according to the characteristics of facility operation status changes, and construct standardized event objects; Based on the pipeline topology and facility connection relationships, the continuous impact transmission relationship between drainage behavior events is analyzed, and events with significant transmission relationships are transformed into memory nodes to form an event network structure. Starting from the current state of the facilities, we trace back to historical memory nodes and construct a drainage memory chain formed by linking multiple historical events according to their impact and contribution. By weighted and fused with the real-time operating status of the facility and the historical inherited information extracted from the drainage memory chain, a digital twin foundation is constructed.

3. The urban drainage facility operation and maintenance management method based on digital twinning according to claim 2, characterized in that, In the process of constructing standardized event objects, an event impact intensity index is constructed based on the amplitude of flow fluctuation, the amplitude of water level fluctuation, the amount of siltation change, and the facility recovery time, combined with the weighting coefficients of flow impact, water level impact, siltation impact, and recovery time. Among them, the flow rate impact weight coefficient comes from the statistical regression analysis of historical rainstorm events or the hydraulic sensitivity calibration results of different regions; the water level impact weight coefficient comes from the analysis of historical waterlogging and overflow events; the siltation impact weight coefficient comes from the statistical data of long-term pipeline blockage and dredging; and the recovery time weight coefficient comes from the statistical data of operation and maintenance response efficiency. In the process of analyzing the transmission of lasting impact, the event transmission impact value is constructed based on the event correlation coefficient and time interval; Among them, the event correlation coefficient is derived from the drainage network topology, hydraulic path analysis results, or historical flow data; In the process of constructing the drainage memory chain, the influence retention coefficient is introduced to calculate the memory inheritance value; the influence retention coefficient is dynamically adjusted according to the facility type, maintenance frequency, pipe diameter characteristics and historical treatment effects.

4. The urban drainage facility operation and maintenance management method based on digital twinning according to claim 3, characterized in that, Step S2 specifically includes: The drainage memory chain nodes are analyzed to identify the residual effects that still exist after the event ends. These residual effects are then decomposed into hydraulic residual effect factors, sedimentary residual effect factors, structural residual effect factors, and operation and maintenance residual effect factors. The residual effect retention values ​​are calculated. The residual impact sources are mapped to the drainage network space, and the residual impact field is constructed based on the pipe network topology, water flow direction and confluence area. The impact field intensity of each facility unit is then calculated. Coupled analysis of multiple residual influence fields is performed to calculate the inertial state value of the facility and quantify the cumulative effect and potential risk trend of different historical influences; By fusing the inertial state of the facility with the real-time digital twin state, an inertial digital twin is generated, forming a future state representation model.

5. The urban drainage facility operation and maintenance management method based on digital twinning according to claim 4, characterized in that, The calculation of residual effect retention value takes into account memory inheritance value, the interval from the end of the event to the current time, and residual effect decay coefficient; The residual effect attenuation coefficient is determined by the facility type, pipe diameter, maintenance frequency, and environmental conditions; The calculation of the field strength takes into account the propagation weight, which is determined by the network connectivity coefficient, hydraulic capacity coefficient, and local resistance coefficient. Among them, the connectivity coefficient of the pipeline network structure is determined by whether the topological relationship is directly connected and the connection level, the hydraulic capacity coefficient is calculated by historical flow capacity and pipe diameter parameters, and the local resistance coefficient is jointly determined by sediment thickness, number of bends and slope changes. In the coupling analysis, a coupling enhancement effect is constructed for any two types of residual influence fields, and a coupling coefficient is introduced to characterize their interaction relationship. The coupling coefficient is obtained by statistical analysis of historical operating data. The calculation of the facility's inertial state value integrates the intensity of various residual influence fields and coupled influences, introduces single influence weight coefficients and coupled influence weights, and performs time evolution analysis on the inertial state value to obtain the facility's inertial evolution rate.

6. The urban drainage facility operation and maintenance management method based on digital twinning according to claim 5, characterized in that, Step S3 specifically includes: Based on inertial digital twins, potential risks of drainage facilities are screened. A comprehensive risk index is constructed by introducing inertial growth trend, deposition evolution rate and structural fatigue accumulation factors, and risk level classification and ranking of each facility node is carried out. Based on the pipeline network topology, hydraulic flow direction, and residual distribution of inertial influence, the dependencies between risk nodes are analyzed, a chain propagation path from the risk source node to the terminal manifestation node is constructed, and the risk transmission intensity between nodes is quantified through a propagation weight model to form a risk evolution chain. Accumulated risk calculations are performed on each node in the risk evolution chain, and the growth rate and trend of the risk are analyzed in conjunction with time series analysis to determine whether the risk has entered a critical state. Operation and maintenance priorities are calculated based on cumulative risk value and growth rate. Key propagation nodes in the risk chain are identified as priority intervention targets, and blocking operation plans are generated in combination with operation and maintenance resource constraints.

7. The urban drainage facility operation and maintenance management method based on digital twinning according to claim 6, characterized in that, The construction of the comprehensive risk index incorporates sedimentation growth trend factors, hydraulic response hysteresis factors, and structural fatigue accumulation factors, and combines the weight coefficients corresponding to each factor. The weighting coefficients are obtained from historical operation and maintenance samples. The calculation of propagation weights takes into account the facility inertial digital twin value, hydraulic path distance, pipeline network structure coupling coefficient and standardization coefficient. The facility inertial digital twin value is obtained by fusing the digital twin base state and the inertial state, and the hydraulic path distance reflects the actual drainage path length between nodes; The calculation of the cumulative risk value takes into account the set of all affected nodes. The cumulative risk value is calculated on multiple time slices to form a risk evolution curve. When the growth rate of the cumulative risk value exceeds the historical average or multiple upstream nodes rise synchronously within a continuous time period, it is determined that the node has entered a critical evolution state and an early warning is triggered. Operation and maintenance priorities are calculated based on cumulative risk value and evolution speed, combined with weighting coefficients.

8. The urban drainage facility operation and maintenance management method based on digital twinning according to claim 7, characterized in that, Step S4 specifically includes: The system digitally twins personnel, vehicles, equipment, and storage resources in urban drainage operation and maintenance, records real-time status and historical operational capabilities, calculates resource capability indices, and maps them to facility risk nodes. The risk blocking contribution of each node is calculated based on the risk evolution chain, key propagation nodes are identified, and the minimum intervention path is generated, while node intervention strategies are planned. Based on the minimum intervention path and resource capacity index, the system automatically generates operation and maintenance tasks, calculates task priorities, completes resource matching, job sequence planning and route arrangement, and dynamically adjusts them according to real-time data during task execution. Real-time collection of task execution data, assessment of risk reduction and resource consumption, and feedback of results to the digital twin enable closed-loop optimization and adaptive adjustment.

9. The urban drainage facility operation and maintenance management method based on digital twinning according to claim 8, characterized in that, The calculation of the resource capacity index considers the unit time operation capacity, historical operation efficiency index and real-time schedulable capacity, and combines their respective weight coefficients. The weight coefficients are calibrated from historical operation and maintenance performance data. The unit time operation capacity includes the dredging volume per hour, the historical operation efficiency index includes the task completion rate to the project duration ratio, and the real-time schedulable capacity considers the current location, availability and standby capacity of the resources. The risk blocking contribution is calculated based on the risk propagation weight from node to downstream node. By ranking the contribution and resource availability, the minimum intervention path is automatically generated to achieve the maximum risk propagation blocking effect with the least resource input. The task priority index is calculated based on the operation and maintenance priority index, risk blocking contribution and corresponding weight coefficients. Factors considered include operation time window, traffic conditions, weather impact and equipment availability. During execution, the task order and resource allocation are dynamically adjusted by accessing pipeline water level data, flow data and sedimentation change data in real time.

10. A digital-twin-based urban drainage facility operation and maintenance management system, characterized by, include: Memory; processor; A computer program stored in the memory and capable of running on the processor; When the processor executes the computer program, it implements a digital twin-based urban drainage facility operation and maintenance management method as described in any one of claims 1 to 9.

Images