A waste transmission pipe network digital twin state mapping and fault positioning method

Abstract

The application discloses a kind of garbage transmission pipe network digital twin state mapping and fault location method, it is related to digital twin technical field, including: establishing garbage transmission pipe network digital twin topology, according to garbage deposit end, suction power equipment and valve state determine effective transport sub-topology and map real-time pressure and equipment state;Based on state change source and pressure measuring point division propagation edge, call the diagnostic propagation path formed by matching propagation edge characteristics, to determine the fault candidate area with first mismatch edge;Through independent diagnostic event repeated verification forms stable propagation abnormal section, and is combined with entity state evidence, limited diagnostic perturbation and actual relationship verification determines fault section, fault entity and model mismatch.By dynamically matching actual transport topology and propagation response, the fault of complex branch pipe network is segmented and positioned, and entity fault and digital twin model mismatch are distinguished, to improve fault location accuracy, stability and model operation reliability.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology, specifically to a method for digital twin state mapping and fault location of a waste transport pipeline network. Background Technology

[0002] Waste pipeline transportation systems typically utilize suction power equipment to create negative pressure airflow within the pipeline network, transporting waste from the waste disposal point through pipe sections, valves, and collection pipelines to centralized processing locations. As the scale of the waste transportation pipeline network expands, the number of pipe sections, branches, and valves increases, causing changes in the network's operational status due to the start / stop of waste disposal, the operation of the suction power equipment, and the open / closed status of valves. Pipeline blockages, abnormal valve operation, equipment malfunctions, and pressure detection anomalies can all reduce transportation efficiency, and in severe cases, lead to waste accumulation and system shutdown. Therefore, real-time monitoring of the pipeline network's operational status and timely identification of anomalies are necessary.

[0003] Existing waste transport network monitoring systems typically collect data such as pipeline pressure, valve status, and suction equipment operating parameters, which are then centrally displayed through a monitoring platform. With the development of digital twin technology, some systems map the network structure, equipment spatial location, and real-time monitoring data onto a virtual network model, reflecting the on-site operating status through color changes, status indicators, and alarm information. However, this approach primarily focuses on physical state mapping and visualization, underutilizing the actual current connectivity of the network and the propagation process of pressure changes within it. When the combined status of the waste disposal end, valves, and suction power equipment changes, the actual pipe segment involved in waste transport is only a portion of the entire network's conductive path. If judgments are made directly based on a fixed topology or the threshold of a single pressure measurement point, they are easily affected by different operating conditions and changes in branch status.

[0004] On the other hand, blockages and other anomalies in waste transport networks typically manifest as changes in the propagation direction, response time, amplitude transmission, and steady-state pressure relationship of pressure changes between adjacent locations. Existing alarm methods using pressure upper and lower limits, single-point duration, and pressure difference between adjacent points can detect some anomalies, but it is difficult to further determine in which propagation path the anomaly first occurred. For pipelines with multi-branch structures, the same state change may propagate along multiple conduction directions, and the response sequence and response amplitude of different pressure points change with the effective transport path, equipment operating conditions, and state change amplitude. Using only a uniform threshold is insufficient to reliably distinguish between normal operating condition changes and actual propagation anomalies.

[0005] In addition, another type of problem that can easily be confused with actual pipeline faults exists in actual operation: inconsistencies between the digital twin model and the physical relationship on site. For example, failure to update the topology connections in a timely manner after pipeline modifications, inconsistencies between the actual installation location of pressure measuring points and the model's binding location, incorrect valve feedback signal associations, and errors in recording equipment topology locations can all cause discrepancies between the pressure propagation path predicted by the digital twin model and the on-site response. If fault location is directly based on the model's prediction results, it is easy to misjudge model relationship mismatches as physical faults such as pipeline blockages or valve malfunctions. Conversely, if the original propagation characteristics are continued to be used after the model relationships have changed, the reliability of subsequent location results will also be reduced.

[0006] Meanwhile, under certain fault conditions, relying solely on pressure changes generated during natural waste transport is insufficient to obtain adequate evidence for localization. For example, an abnormal section may repeatedly exhibit propagation mismatch during multiple operations, but lacks equipment status evidence that directly points to a specific entity. Directly identifying the faulty entity in this situation carries the risk of misjudgment; maintaining only the alarm status makes it difficult to further narrow down the fault range. Therefore, how to supplement diagnostic evidence with controllable operational changes without altering the current pipeline continuity, and how to distinguish between evidence of physical faults and evidence of model relationship mismatch, are also problems that existing pipeline fault diagnosis needs to address. Summary of the Invention

[0007] Based on the shortcomings of the prior art described above, the purpose of this invention is to provide a digital twin state mapping and fault location method for waste transport pipelines to solve the above-mentioned technical problems.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a method for digital twin state mapping and fault location of a waste transport network, comprising: Establish a digital twin topology for the waste transport network to record the connections and binding relationships between pipe sections, waste disposal ends, valves, suction power equipment, and pressure measuring points; Based on the working status of the waste disposal end, the conveying condition of the suction power equipment, and the valve feedback conduction status, an effective conveying sub-topology is determined. Real-time pressure and equipment status are mapped to the topology, and diagnostic events whose status change sources can be determined and whose changes have been confirmed by feedback are selected. The unique conductive section from the state change source entity to the first pressure measuring point in each direction, and the unique conductive section between adjacent pressure measuring points are determined as propagation edges. Propagation edge features that match the sub-topology, operating conditions, state change source, propagation edge topology relationship and change conditions are called to form diagnostic propagation paths from the state change source along each direction. The first mismatched edge coverage section of each path is determined as a fault candidate area. By repeatedly verifying independent diagnostic events, a stable propagation anomaly segment is formed when the first mismatch repeatedly covers the same segment; the fault segment is identified when the evidence of the state of an independent entity is consistent with the topological location and direction of influence of the anomaly, and the fault entity is identified when the evidence points to a specific entity; when the evidence is insufficient and there is a diagnostic execution device that meets the conditions, a restricted diagnostic perturbation that keeps the sub-topology unchanged is implemented; when the perturbation mismatch occurs, evidence of entity fault reinforcement is formed; when the location that the model predicts should not respond repeatedly forms an associated response in independent diagnostic events, candidate evidence of model mismatch is formed. For stable propagation anomaly segments that lack independent entity state evidence even after perturbation, acquire actual relationship verification data; update the relationship when the actual relationship is inconsistent with the topology; determine the fault segment when the actual relationship is consistent and the perturbation forms evidence of entity fault reinforcement; verify the applicability of propagation edge features after verification, and continue to locate based on the verified topology.

[0009] The present invention is further configured to establish a digital twin topology for a waste transport network, including: Acquire pipeline network structure data, pipeline geometric parameters, and equipment spatial locations, and assign a unique number to each entity; Pipe segments record geometric parameters and connection relationships; valves record connected pipe segments, control status and actual feedback status; pressure measuring points record the bound entities. The topology is formed based on the physical connections of the entities for path search and state propagation analysis, and the topological positions of valves, suction power equipment and pressure measuring points are recorded.

[0010] The present invention is further configured to determine an effective transport sub-topology, including: The working status of the waste disposal end, the conveying condition of the suction power equipment, and the valve feedback conduction status are detected, and the period during which the combined status is maintained is determined as a waste conveying event. The current event ends when any of the following occurs: the working status of the waste disposal end changes, the valve conduction relationship changes, or the suction power equipment exits the conveying mode. Changes in the operating parameters of the suction power equipment during the conveying mode do not change the event boundary. Searching along the guiding path from the waste disposal end to the suction power equipment, entities that meet the guiding conditions and pressure measuring points are combined to form an effective transport sub-topology; when the measuring points outside the sub-topology form an effective response that is temporally correlated with the diagnostic event, evidence of abnormal relationship in the model prediction is formed.

[0011] The present invention is further configured such that the establishment of the propagation edge and its features includes: When there are no other configured pressure measurement points between the source entity of the state change and the first pressure measurement point in a certain propagation direction, and there is a unique continuous physical path, the corresponding entity and segment are determined as the source propagation edge; when there are no other configured pressure measurement points between adjacent configured pressure measurement points, and there is a unique continuous physical path, the corresponding segment is determined as the measurement point propagation edge. When there are two or more simultaneously conducting paths between the aforementioned endpoints, the area formed by the corresponding paths is defined as the composite propagation segment; Under the condition that the entity connection, measurement point binding and state association are verified to be consistent with the actual situation, the relevant entities are fault-free and the measurement point data is valid, the baseline diagnostic event that can be determined by the state change source is established according to the effective transmission sub-topology, operating conditions, the relationship between the state change source and the propagation edge topology, as well as the change type, direction and amplitude range. The normal propagation edge characteristics characterize the direction, time and amplitude relationship between the effective change on the front side and the pressure response on the back side. The measurement point propagation edge also characterizes the response order of adjacent measurement points and the steady-state effective pressure difference range. Under the aforementioned conditions, a single diagnostic execution device is used to implement a preset perturbation, and diagnostic propagation edge characteristics are established according to the topological relationship between the diagnostic execution device and the propagation edge, as well as the perturbation type, direction, and amplitude range. When the validity check of the data involved in the determination fails or the current conditions lack matching propagation edge features, the corresponding propagation edge will be determined as an undeterminable propagation edge.

[0012] The present invention is further configured such that the determination of diagnostic events and independent diagnostic events includes: During the effective delivery sub-topology conduction relationship maintenance, either the actual operating parameter change of the suction power equipment confirmed by feedback or the actual opening degree change of the valve maintaining conduction will be regarded as a diagnostic event; Record the source of state change, time of occurrence, type of change, direction of change, and magnitude of change; determine the conduction propagation direction of pressure measurement points from the source of state change along the effective transport sub-topology. When multiple state change sources exist within the same response time window and the response attribution cannot be distinguished, the event is not used to perform the first propagation mismatch localization; when the valve action changes the conduction relationship, the current waste conveying event is terminated and the effective conveying sub-topology is re-determined; When the state change sources of adjacent diagnostic events can be separately attributed, the response time windows do not overlap, and the dynamic response of the previous event ends and enters a stable state for a preset duration, the adjacent diagnostic events are determined to be independent diagnostic events, and the continuous responses generated by the same state change are not split and counted.

[0013] The present invention is further configured such that the determination of the fault candidate region includes: The diagnostic propagation path is formed by sequentially connecting the source end propagation edge and the measuring point propagation edge along each conduction propagation direction from the source of state change; For the source-end propagation edge, the actual change confirmed by the feedback of the state change source is taken as the effective change of the front side; for the measurement point propagation edge, the effective response of the measurement point before the reference response order is taken as the effective change of the front side. If a valid change is established on the front side and the data on the back side is valid, and the response on the back side meets the characteristics of a matching propagation edge, then the edge is determined as a propagation consistent edge; if no valid response is formed within the response time range or at least one of the response characteristics that should be judged does not meet the corresponding characteristics, then the edge is determined as a propagation mismatch edge. The segment covered by the mismatched edge during the first propagation of each path is identified as a candidate fault area; when an undeterminable propagation edge is encountered, the localization of the corresponding path is stopped; when a composite propagation segment is encountered, the current round of edge-by-edge localization is stopped and marked as a segment to be distinguished.

[0014] The present invention is further configured such that the formation of evidence of the state of independent entities and evidence of anomalies in measurement point data includes: The physical status is determined by at least one of the following: the pipe section physical verification results obtained independently of the pressure propagation mismatch judgment, valve feedback, and equipment operating parameters; When the corresponding data indicates an entity anomaly, the entity is located in the fault candidate region, and the direction of its influence along the diagnostic propagation path is consistent with the actual propagation mismatch direction, independent entity status evidence is formed; when this evidence can independently point to a specific entity, the faulty entity is identified. When a pressure measurement point repeatedly exhibits at least one of the following in multiple independent diagnostic events: exceeding the measurement range or abnormal timestamps, it constitutes evidence of abnormal measurement point data. When a pressure measurement point maintains a fixed value across multiple independent diagnostic events, and adjacent valid pressure measurement points generate a pressure response consistent with the corresponding diagnostic event, this constitutes evidence of abnormal measurement point data.

[0015] The present invention is further configured such that repeated verification includes: Filter out multiple independent diagnostic events whose diagnostic propagation paths pass through the same fault candidate region, and call the propagation edge features that match their respective effective delivery sub-topology, operating conditions, state change sources, propagation edge topology relationships, and change conditions; If the data used for the determination is valid and the initial propagation mismatch edge repeatedly covers the same physical segment in multiple independent diagnostic events, the segment is identified as a stable propagation abnormal segment. If the segment to be distinguished forms a unique conductive path and has matching propagation edge characteristics in subsequent independent diagnostic events, it continues to be located according to the propagation edge; if the aforementioned conditions are not met, it remains in the state of being to be distinguished. When the location where the model predicts should not respond repeatedly forms a valid response that is temporally correlated with the source of state change in multiple independent diagnostic events, it forms candidate evidence of model mismatch.

[0016] The present invention is further configured such that the restricted diagnostic perturbation includes: When a stable propagation anomaly segment lacks independent entity state evidence to explain the propagation anomaly and can correspond to a propagation edge, determine whether there is a diagnostic execution device located outside the anomaly segment that is operating normally, provides consistent feedback, and whose diagnostic response can pass through the anomaly segment; When a diagnostic execution device that meets the conditions exists, a perturbation is implemented by selecting one of the suction power equipment or the continuous opening regulating valve; the suction power equipment changes its operating parameters within a range that does not trigger operating condition switching, protection, or shutdown, and the valve changes its opening within a range that maintains the conduction relationship unchanged; Invoke diagnostic propagation edge features that match the diagnostic execution device, propagation edge topology, and perturbation conditions; when a candidate propagation edge forms a valid change on the front side and the data on the back side is valid, but the back side does not form a valid response that satisfies the corresponding features, evidence of physical fault reinforcement is formed. After each perturbation, the diagnostic execution device is reset, and the next perturbation is implemented after the dynamic response ends and the device enters a stable state for a preset duration. When the external measurement point of the subtopology repeatedly forms an effective response that is temporally correlated with the perturbation in multiple diagnostic perturbations, it forms candidate evidence of model mismatch. When there is no diagnostic execution device that meets the conditions or when there is a lack of matching diagnostic propagation edge features, the corresponding perturbation is not implemented and the stable propagation abnormal segment is transferred to relation verification.

[0017] The present invention is further configured such that the model mismatch processing includes: For model mismatch candidate evidence, stable propagation anomaly sections that do not meet perturbation execution conditions and still lack independent entity state evidence after perturbation, at least one of the following relationships to be verified is formed: pipe segment topology connection, pressure measurement point binding, valve state association, and equipment topology location. Acquire data on actual pipeline connections, measurement point installation locations, valve status, feedback signal sources, and actual equipment locations, and compare them with the digital twin topology; When the verification relationship is inconsistent, the corresponding relationship is updated according to the actual data, and the affected valid transport sub-topology and propagation edge are re-determined; when the existing propagation edge features cannot be confirmed to be applicable to the update relationship, the corresponding edge is determined as an undeterminable propagation edge, and the matching features are re-established. When a stable propagation anomaly segment with consistent verification relationships forms evidence of physical fault reinforcement in a restricted diagnostic perturbation, the corresponding fault segment is identified; when a stable propagation anomaly segment with consistent verification relationships does not form evidence of physical fault reinforcement, the stable propagation anomaly segment is retained and further verified using subsequent independent diagnostic events; when unexplained propagation conflicts still exist after model updates, the verified and confirmed update relationships are retained and the remaining conflicts are further verified.

[0018] This invention provides a digital twin state mapping and fault location method for waste transport pipeline networks. By establishing a digital twin topology for the waste transport pipeline network, an effective transport sub-topology is determined based on the actual state of the waste disposal end, suction power equipment, and valves, and real-time pressure and equipment status are mapped. The unique conductive section from the source of state change to the first pressure measuring point in each direction and between adjacent pressure measuring points is determined as a propagation edge. A diagnostic propagation path is formed based on the characteristics of the matched propagation edges, and the first mismatch edge of each path is used to determine candidate fault regions. Stable propagation anomaly sections are formed through repeated verification of independent diagnostic events. Combined with independent entity state evidence, constrained diagnostic perturbations, and actual relationship verification, entity faults and model mismatches are distinguished. The topology is updated and propagation edge characteristics are verified when model relationships change. The beneficial effects include: 1. By dynamically determining the effective conveying sub-topology based on the working status of the waste disposal end, the conveying condition of the suction power equipment, and the valve feedback conduction status, and mapping the real-time pressure and equipment status to the digital twin topology, fault diagnosis is based on the current actual conduction relationship of the pipeline network, reducing the interference of fixed topology and changes in operating conditions on the diagnostic results and improving the accuracy of status mapping. 2. By dividing the source of state change to the first pressure measuring point and the adjacent pressure measuring points into propagation edges, and combining the characteristics of the propagation edges that match the operating conditions and state change conditions to analyze the pressure response edge by edge, and by repeatedly verifying the first propagation mismatch and independent diagnostic events to determine the stable propagation anomaly section, the anomaly can be further located from a single-point alarm to a specific pipeline section, thereby improving the accuracy and stability of fault location in complex branch pipeline networks. 3. By combining independent entity state evidence, constrained diagnostic perturbations, and on-site actual relationship verification, the system distinguishes between entity faults and digital twin model relationship mismatches. After changes in model relationships, the topology is updated and propagation edge features are re-verified to avoid misjudging topology connections, measurement point bindings, and state associations as entity faults, thereby improving the reliability of the digital twin model and the credibility of fault location during long-term operation.

[0019] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1The flowchart illustrates a digital twin state mapping and fault location method for a waste transport network, which is an exemplary embodiment of the present invention. Detailed Implementation

[0021] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0022] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0023] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0024] A digital twin state mapping and fault location method for waste transport pipeline networks, such as Figure 1 As shown, it includes: Establish a digital twin topology for the waste transport network to record the connections and binding relationships between pipe sections, waste disposal ends, valves, suction power equipment, and pressure measuring points; Based on the working status of the waste disposal end, the conveying condition of the suction power equipment, and the valve feedback conduction status, an effective conveying sub-topology is determined. Real-time pressure and equipment status are mapped to the topology, and diagnostic events whose status change sources can be determined and whose changes have been confirmed by feedback are selected. The unique conductive section from the state change source entity to the first pressure measuring point in each direction, and the unique conductive section between adjacent pressure measuring points are determined as propagation edges. Propagation edge features that match the sub-topology, operating conditions, state change source, propagation edge topology relationship and change conditions are called to form diagnostic propagation paths from the state change source along each direction. The first mismatched edge coverage section of each path is determined as a fault candidate area. By repeatedly verifying independent diagnostic events, a stable propagation anomaly segment is formed when the first mismatch repeatedly covers the same segment; the fault segment is identified when the evidence of the state of an independent entity is consistent with the topological location and direction of influence of the anomaly, and the fault entity is identified when the evidence points to a specific entity; when the evidence is insufficient and there is a diagnostic execution device that meets the conditions, a restricted diagnostic perturbation that keeps the sub-topology unchanged is implemented; when the perturbation mismatch occurs, evidence of entity fault reinforcement is formed; when the location that the model predicts should not respond repeatedly forms an associated response in independent diagnostic events, candidate evidence of model mismatch is formed. For stable propagation anomaly segments that lack independent entity state evidence even after perturbation, acquire actual relationship verification data; update the relationship when the actual relationship is inconsistent with the topology; determine the fault segment when the actual relationship is consistent and the perturbation forms evidence of entity fault reinforcement; verify the applicability of propagation edge features after verification, and continue to locate based on the verified topology.

[0025] Specifically, the digital twin topology of the waste transport network is used to represent the physical connections and data binding relationships between on-site waste transport pipelines and their associated equipment. When establishing the digital twin topology, the network structure data, pipeline geometric parameters, and equipment spatial location data of the waste transport network are acquired. The network structure data can be derived from at least one of the following: existing CAD network diagrams, GIS network data, 3D network models, and on-site equipment ledgers. Unique entity numbers are assigned to each on-site entity—pipe segment, waste disposal end, valve, suction power equipment, and pressure measurement point—ensuring that the same on-site entity uses a consistent entity identifier in the structural model, real-time database, and operational status records.

[0026] Specifically, for each pipe segment, the system records at least the starting point, ending point, length, diameter, slope, bends, and connections to other pipe segments; for each waste disposal end, it records the pipe segment it connects to and its operating status; for each valve, it records the connecting pipe segments on both sides of the valve, its control status, set opening degree, and actual feedback status; for each suction power device, it records the device's connection location, operating status, operating parameters, and feedback parameters; and for each pressure measuring point, it records the pressure sensor number, actual installation location, and binding relationship with the pipe segment entity. Thus, a digital twin topology is constructed using entities and their connections to create an executable path search, enabling the system to search from any waste disposal end along the current connectivity to the suction power device, and to determine which pipe segment, branch, and topological position of any pressure measuring point relative to other pressure measuring points are located.

[0027] Real-time data is acquired through a data acquisition device and written to the corresponding digital twin entity according to the entity number. In practice, each pressure sensor, valve controller, and suction power equipment controller can connect to the edge data acquisition terminal via a field communication network. The edge data acquisition terminal timestamps the data according to a unified time base and sends the pressure value, valve feedback status, equipment operating status, and operating parameters to the status mapping module. Real-time data can be acquired according to a preset sampling period, such as a second-level acquisition period. The data can be simultaneously stored in a time-series database and a local cache for real-time judgment and historical diagnostic event backtracking. When checking the validity of pressure data, at least the continuity of timestamps, whether the sampled value exceeds the sensor's range, and whether the data remains constant over a long period should be checked. For transient isolated jumps, filtering can be performed based on adjacent sampled values, but the filtered fictitious response should not replace the original response for judgment.

[0028] During real-time operation, the current effective transport sub-topology is dynamically determined based on the working status of the waste disposal end, whether the suction power equipment is in waste transport mode, and the actual conduction status reported by each valve. Specifically, the waste disposal end in operation is used as the starting point for path search. The search proceeds along the pipeline where the valves are actually in a conduction state towards the suction power equipment, incorporating pipe sections, valves, related equipment, and associated pressure measuring points that can form an actual transport path into the effective transport sub-topology. Pipe sections and measuring points not covered by the current actual conduction path are not included in the ordinary propagation mismatch judgment as the current normal pressure propagation path.

[0029] This embodiment further determines the waste transport event boundary based on the waste disposal end status, the conveying condition of the suction power equipment, and the actual valve conduction relationship. When the above combined states remain unchanged, it is considered to be in the same waste transport event; when any of the following occurs: a change in the waste disposal end's working state, a change in the actual valve conduction relationship, or the suction power equipment exiting the conveying condition, the current waste transport event ends, and the effective transport sub-topology is recalculated based on the changed site conditions. Changes in the rotational speed, frequency, and other operating parameters of the suction power equipment during the waste transporting condition do not alter the waste transport event boundary, allowing these operating parameter changes to be used as pressure propagation diagnostic events within the same effective transport sub-topology.

[0030] The real-time status mapping in the digital twin topology includes not only pressure values, but also valve feedback status, suction power equipment operating status, and waste disposal terminal working status. In practice, the status attributes of the corresponding digital twin entities can be updated according to normal, abnormal, pending judgment, and mismatch statuses, ensuring that the topology display status and subsequent fault diagnosis use the same entity number and the same valid transport sub-topology, avoiding object correspondence deviations between the display model and the diagnostic model.

[0031] Diagnostic events are state change events that generate identifiable pressure changes during the effective transmission sub-topology continuity period. Specifically, these include changes in the operating parameters of the suction power equipment confirmed by actual feedback, and changes in the actual opening degree of valves maintaining continuity. For suction power equipment, for example, after the controller issues a frequency change command, a change is only considered a diagnostic event if the equipment's feedback frequency, actual rotational speed, or feedback parameters characterizing the actual operating state do indeed change accordingly. For continuously adjustable valves, a diagnostic event is only considered if the actual opening feedback changes and the valve remains continuous. A control command without actual change in feedback is not considered a valid diagnostic event, thus avoiding misjudging a lack of actual equipment operation as a propagation anomaly in subsequent pipe sections.

[0032] For each diagnostic event, the source of the state change, the time of occurrence, the type of change, the direction of change, and the magnitude of change are recorded. The type of change distinguishes between changes in the operating parameters of the suction power equipment and changes in valve opening. The direction of change indicates whether the corresponding parameter increases or decreases, and the magnitude of change indicates the degree of change of the actual feedback quantity relative to the stable value before the change. Based on the topological position of the state change source in the effective delivery sub-topology, diagnostic propagation directions are established from the state change source along the conduction direction of each pressure measurement point.

[0033] To avoid a positioning blind spot between the state change source and the first pressure measuring point, this embodiment divides the propagation edge into source-end propagation edges and measuring point propagation edges. When there are no other configured pressure measuring points between the state change source entity and the first pressure measuring point in a certain propagation direction, and only one currently connected continuous physical path exists, the state change source entity and this continuous physical path are jointly defined as the source-end propagation edge. When there are no other configured pressure measuring points between two adjacent configured pressure measuring points, and only one currently connected continuous physical path exists, the corresponding pipe segment and related entities between the two pressure measuring points are defined as the measuring point propagation edge.

[0034] Therefore, starting from the source of state change, a source-end propagation edge and one or more subsequent measurement point propagation edges are sequentially connected along each conduction propagation direction to form the diagnostic propagation path corresponding to that propagation direction. For the source-end propagation edge, the effective change before it is the actual change confirmed by feedback from the state change source, and the response after it is the pressure response generated by the first pressure measurement point; for the measurement point propagation edge, the effective change before it is the effective response of the pressure measurement point located ahead according to the baseline response order, and the response after it is the pressure response of the next pressure measurement point.

[0035] If there are two or more simultaneously conducting continuous paths between the source of the state change and the first pressure measuring point, or between adjacent pressure measuring points, it is impossible to determine which internal path the anomaly is located on based solely on the endpoint pressure response. Therefore, the area covered by multiple paths is defined as a composite propagation section. In the current event, instead of directly identifying a certain pipe segment as a propagation mismatch edge based on the data at both ends of the composite propagation section, it is marked as a segment to be distinguished. When a subsequent waste transport event causes a path in this composite propagation section to become the only conducting path due to changes in the state of other valves, and a usable propagation edge characteristic exists under the corresponding conditions, then edge-by-edge positioning is performed according to the ordinary propagation edge method.

[0036] Propagation edge features are established under a baseline condition confirming the correctness of the on-site physical relationships. Specifically, before establishing the baseline, the pipe segment connection relationships, pressure measuring point binding relationships, and valve status association relationships are verified to ensure that the digital twin topology is consistent with the actual physical relationships; simultaneously, it is confirmed that the relevant entities involved in the baseline establishment are in normal operating condition, the data of the participating measuring points are valid, and the source of the status change of the baseline diagnostic event can be identified. Under this condition, based on historical operating data and preset baseline test data, normal propagation edge features are established according to the topological relationship, change type, change direction, and change amplitude range of the effective transport sub-topology, operating condition, and the source of status change with the propagation edge. In this embodiment, the operating condition is jointly identified by the current operating parameter range of the suction power equipment, the set of working waste disposal ends, and the actual conduction relationship of the valves. When the effective transport sub-topology is the same but the operating parameters of the suction power equipment fall into different operating parameter ranges, propagation edge features are established separately as different operating conditions. The operating parameter range is divided according to the adjustable range and amplitude level of the equipment control parameters, without crossing the equipment transport condition switching boundary.

[0037] Normal propagation edge features are established according to the effective transport sub-topology, operating conditions, state change source, propagation edge, change type, change direction, and change amplitude level. Under the same combination of conditions, baseline diagnostic events are collected that have been confirmed to be fault-free, whose digital twin relationships are consistent with the field, and whose data are valid. In this embodiment, after obtaining no less than 20 valid baseline diagnostic events under the same combination of conditions, corresponding propagation edge features are established. If there are fewer than 20, propagation edge features for that combination of conditions are not generated, and the corresponding propagation edge is determined to be missing a matching feature during diagnosis. The change amplitude level is divided according to the controllable parameters of the diagnostic execution equipment itself. In this embodiment, when the suction power equipment uses the operating frequency as the state change parameter, a 2Hz actual frequency change amplitude is used as one amplitude level; when the continuous opening regulating valve uses the actual opening as the state change parameter, a 5 percentage point change in the actual opening is used as one amplitude level. When the actual change amplitude falls between the boundaries of two adjacent levels, it is assigned to the amplitude level containing that actual value; when it exceeds the existing level range, existing propagation edge features are not invoked, and a new amplitude level feature is established after a sample that meets the baseline conditions is formed.

[0038] For each baseline diagnostic event, record the time of the effective change occurring on the front side and the time when the pressure measuring point on the rear side first forms an effective response; record the difference between the two as the response time for that event. For the measuring point propagation edge, record the pressure change amplitudes of the front and rear pressure measuring points and calculate the amplitude transfer ratio between the rear and front pressure change amplitudes. After the pressure reaches a steady state, record the steady-state pressure difference between the two ends of the propagation edge. For the source-end propagation edge, use the change amplitude of the actual feedback parameter of the state change source as the front-side change amount, and record the correspondence between this change amount and the rear-side pressure change amount.

[0039] For each response time obtained under the same combination of conditions, arranged in ascending order of value, the 10th and 90th percentile values ​​are taken as the lower and upper limits of the response time sample, respectively. The allowable range of response time is obtained by subtracting the time measurement tolerance from the lower limit and adding the time measurement tolerance to the upper limit. The time measurement tolerance is determined based on the pressure data sampling period and the allowable time synchronization error of each acquisition device, and is taken as the sum of one sampling period and the maximum allowable time synchronization error. For example, if the sampling period is 1 second and the allowable time synchronization error of the acquisition system is 0.1 seconds, the time measurement tolerance is 1.1 seconds.

[0040] The amplitude transfer ratios are arranged in the same manner, and the 10th and 90th percentile values ​​are used to form the amplitude transfer sample interval. Then, amplitude transfer margins corresponding to the pressure measurement tolerance are added to both ends of the interval to obtain the amplitude transfer range. The pressure measurement tolerance is obtained from the current valid verification certificate, calibration record, or equipment technical parameters of the pressure sensor. When both pressure measurement points are involved in the calculation, the sum of the absolute values ​​of the allowable pressure measurement errors of the two measurement points is used as the measurement tolerance for the pressure difference calculation.

[0041] The steady-state pressure difference samples are arranged in ascending order of value. The 10th percentile and 90th percentile values ​​are taken respectively, and the sum of the absolute values ​​of the allowable measurement errors of the pressure measurement points before and after are subtracted and added respectively to form the steady-state effective pressure difference range. Thus, the response time range, amplitude transfer range, and steady-state effective pressure difference range are all determined by the fault-free reference samples under the corresponding operating conditions and the known measurement errors, rather than being arbitrarily set manually.

[0042] To determine the effective response of the pressure measurement points, a continuous and stable sampling window is selected as the baseline window before the diagnostic event occurs. In this embodiment, the baseline window contains no fewer than 20 continuous sampling points. The pressure values ​​within the baseline window are arranged in numerical order, and the median value is taken as the pre-event reference pressure. The absolute deviation between each sampled pressure value and the median value is calculated, and the 95th percentile of the absolute deviation is taken as the on-site pressure fluctuation. The allowable measurement error of the pressure sensor is obtained from the currently valid verification certificate, calibration record, or equipment technical parameters. The larger value between the on-site pressure fluctuation and the allowable measurement error of the pressure sensor is determined as the pressure response identification threshold.

[0043] After a diagnostic event occurs, if the absolute value of the change in pressure at a pressure measurement point relative to the pre-event baseline pressure exceeds the pressure response identification threshold, and this change persists for at least three consecutive sampling points, the pressure change is determined as a valid response. The time corresponding to the first sampling point that meets the continuity determination condition is determined as the response arrival time. In this embodiment, when a 1-second sampling period is used, three consecutive sampling points correspond to three consecutive seconds of valid change. When the sampling period changes, at least three consecutive sampling points are used as the continuity determination condition.

[0044] If multiple state change sources are detected within the same response time window, and it is impossible to distinguish which state change source caused the pressure response based on the occurrence time, topological location, and feedback change, then this event will not participate in the initial propagation mismatch localization to avoid incorrect attribution caused by the superposition of multiple pressure excitations. When a valve action changes the actual conduction relationship, the propagation path before the action will not be used for diagnosis. Instead, the current waste transport event will be terminated, the effective transport sub-topology will be recalculated, and subsequent diagnosis will be performed.

[0045] In this embodiment, the stable state is determined based on the range of pressure changes. For each pressure measurement point participating in the current diagnostic propagation path, data within a stable determination window is continuously acquired; the length of the stable determination window is the larger of the maximum value among the upper limits of the allowable intervals of the response times of each propagation side in the current diagnostic propagation path and the value among 10 sampling periods. When the difference between the maximum and minimum pressure values ​​of each pressure measurement point within the stable determination window does not exceed twice the respective pressure response identification threshold, and the actual feedback parameters of the state change source remain unchanged within the same window, it is determined that the dynamic response of the previous diagnostic event has ended and entered a stable state.

[0046] The stable state is only used to distinguish the dynamic response of adjacent diagnostic events and does not require the pressure to return to the fault-free baseline value; therefore, even when the pipeline network is continuously blocked and enters a new abnormal stable pressure level, subsequent independent diagnostic events can still be formed.

[0047] Independent diagnostic events are used to rule out single, occasional fluctuations. Specifically, the source of state change for two adjacent diagnostic events should be identifiable, their response time windows should not overlap, and the dynamic pressure response generated by the previous diagnostic event should have ended, with the pipeline entering a stable state that lasts for a preset duration. Only then will the subsequent diagnostic event be considered a new independent diagnostic event for repeated verification. This stable state does not require a return to the normal pressure value under fault-free conditions; it only requires that the pressure change enter a sustained stable phase. Therefore, even if there is persistent blockage in the pipeline and it stabilizes at an abnormal pressure level, subsequent independent diagnostic events can still be obtained. Continuous pressure oscillations or continuous responses generated by the same state change are not broken down into multiple independent diagnostic events.

[0048] After the propagation path is diagnosed, the actual response is compared with the characteristics of the propagation edge along each side of the path from the source of state change. For the source-end propagation edge, it is first confirmed that the change in state change source is valid through actual feedback; for the measurement point propagation edge, it is first confirmed that the pressure measurement point ahead of the baseline response sequence has formed a valid response. When the valid change on the front side is valid and the data participating in the judgment on the back side is valid, it is determined whether the actual response on the back side conforms to the characteristics of the corresponding propagation edge.

[0049] If the direction, response time, and amplitude transmission characteristics of the rear pressure response are all within the matching range, and the steady-state effective pressure difference is also within the corresponding range when a steady-state response has been formed, then the propagation edge is determined as the propagation consistent edge, and the judgment continues along the current diagnostic propagation path.

[0050] If a valid change has been established on the front side and the data on the back side is valid, but no valid pressure response is formed on the back side within the corresponding response time range, then this propagation edge is identified as a propagation mismatch edge. Similarly, if a pressure response is formed on the back side, but at least one of the following—pressure change direction, response time, amplitude transmission relationship, or other characteristics that should be included in the current judgment—does not meet the characteristics of a matching propagation edge, then this propagation edge is also identified as a propagation mismatch edge. For the diagnostic stage where a steady state has not yet been formed, the steady-state effective pressure difference is not used as a necessary judgment condition to avoid misjudgment caused by using non-existent steady-state data.

[0051] The propagation edges are sequentially determined from the state change source along the same diagnostic propagation path. When a propagation mismatch edge appears for the first time, the entity and physical segment covered by the mismatch edge are identified as the candidate fault region corresponding to that path. The search for the so-called first fault further away is not continued across the mismatch edge. The above determination is performed separately for multiple propagation directions. Therefore, when there are multiple conducting branches for the same state change source, the first mismatch result for each propagation direction can be obtained separately.

[0052] When the data validity check of a propagation edge fails, that propagation edge is designated as an undecidable propagation edge, and fault location on the current path is stopped at that location. The inference of subsequent faults is not performed beyond the invalid propagation edge. Similarly, if the current operating conditions lack matching normal propagation edge features, the corresponding propagation edge is also designated as an undecidable propagation edge. This process avoids forcibly interpreting uncertain states as faults due to missing data or insufficient baselines.

[0053] The validity of measurement point data can be determined by combining real-time data and multiple independent diagnostic events. When a pressure measurement point repeatedly generates out-of-range data or timestamp anomalies in multiple independent diagnostic events, it constitutes evidence of measurement point data anomaly. Similarly, when a pressure measurement point maintains a fixed value across multiple independent diagnostic events, while its adjacent valid pressure measurement points all show pressure changes consistent with the corresponding diagnostic events, this also constitutes evidence of measurement point data anomaly. The absence of a pressure response from a pressure measurement point in a single diagnostic event does not directly indicate an anomaly, as actual pipeline blockage can also prevent downstream measurement points from showing pressure changes.

[0054] After obtaining the candidate fault region, multiple independent diagnostic events are used for repeated verification. Independent diagnostic events whose diagnostic propagation paths all pass through the same candidate fault region are selected, and their corresponding propagation edge features are invoked based on each event's valid transport sub-topology, operating conditions, state change source, propagation edge topology relationship, and change conditions. Provided the data used for judgment is valid, if the initial propagation mismatch edges of multiple independent diagnostic events repeatedly cover the same physical segment, then that physical segment is identified as a stable propagation anomaly segment. If the initial propagation mismatch positions cannot stably cover the same physical segment, the candidate fault state is retained, and a stable anomaly position is not directly determined based on a single mismatch.

[0055] For sections with stable propagation anomalies, further obtain physical status information independent of the pressure propagation mismatch judgments mentioned above. Pipeline segment physical status information can come from on-site inspections, maintenance records, internal pipe section inspection results, and data that independently reflects the pipe segment's status; valve physical status information can include actual valve opening feedback, actuator status, and related feedback signals; suction power equipment status information can include actual speed, operating frequency, current, load, and controller operating status. The above evidence should be obtained independently of the current pressure propagation mismatch result to avoid using the same pressure anomaly to both prove propagation mismatch and conversely prove physical anomalies.

[0056] When independent data indicates an anomaly in an entity, further analysis is needed to determine whether the entity is located in the physical region corresponding to the stable propagation anomaly segment, and whether the direction of the pressure impact generated by the anomaly along the diagnostic propagation path is consistent with the actual propagation mismatch. For example, if the actual valve opening deviates from the set state, and the valve is located within the initial propagation mismatch edge, its actual state can lead to a weakening of the subsequent pressure response. In this case, the valve state is consistent with the propagation mismatch in both topological location and direction of influence, thus forming independent entity state evidence. If an entity anomaly exists but its location is not within the candidate region, or if the expected direction of the propagation impact is opposite to the current actual mismatch, this evidence is not used to confirm the current fault.

[0057] When the evidence of an independent entity's condition is consistent with the topological location and direction of influence of a stable propagation anomaly segment, the corresponding faulty segment can be identified. When this evidence can further uniquely point to a specific pipe section, valve, or suction power equipment, the corresponding entity is identified as the faulty entity. If the existing evidence of an independent entity's condition is insufficient to identify a specific entity, the stable propagation anomaly segment is maintained, and it is determined whether the conditions for restricted diagnostic perturbation are met.

[0058] Constrained diagnostic perturbation is used to artificially generate controllable pressure excitation without altering the current effective transport subtopology, to further verify anomalous sections. First, it checks whether a stable propagation anomalous section corresponds to a propagation edge with established diagnostic propagation edge characteristics. Then, it checks whether there exists a diagnostic execution device located outside the anomalous section, operating normally, providing consistent feedback, and whose diagnostic response can pass through the anomalous section. The diagnostic execution device can be a suction power device or a valve with continuous opening adjustment capability.

[0059] Before determining the diagnostic perturbation amplitude, first read the currently valid operating limit parameters of the diagnostic execution equipment. For suction power equipment, obtain the current operating parameters, the lower limit of allowable conveying conditions, the upper limit of allowable conveying conditions, the lower limit of protection action, the upper limit of protection action, the shutdown trigger value, and the minimum adjustment step size of the controller. These parameters are preferentially read from the currently operating inverter, PLC, or equipment controller's valid configuration. When the control system cannot read them automatically, they are entered from the currently effective equipment commissioning parameter table or equipment operating parameter configuration table, and the corresponding parameter version is recorded. For continuously opening regulating valves, obtain the current actual opening, the minimum opening to maintain conduction, the maximum allowable opening, the minimum adjustment step size of the actuator, and the valve protection limit parameters.

[0060] Before each diagnostic perturbation, the above parameters are reread, and the current actual feedback value is used as the starting point for the perturbation. Historical setpoints are not used to replace the currently effective protection parameters. When the protection parameters in the control system are inconsistent with the equipment configuration data, the protection parameters currently in effect in the actual controller are used as the basis for this perturbation calculation.

[0061] For the suction power equipment, the current actual operating parameters are denoted as X0, the upper limit of the current conveying condition is denoted as XH, the lower limit of the current conveying condition is denoted as XL, the upper limit of the protection action is denoted as PH, the lower limit of the protection action is denoted as PL, and the minimum adjustment step size of the equipment is denoted as S. A safety margin G is set, which is 2S in this embodiment.

[0062] When performing diagnostic perturbations in the direction of increasing execution parameters, the upper limit of the perturbation amplitude, Umax, is determined by the smaller value between XH-X0 and PH-X0-G; when performing diagnostic perturbations in the direction of decreasing execution parameters, the upper limit of the perturbation amplitude, Umax, is determined by the smaller value between X0-XL and X0-PL-G. A calculated Umax not greater than 0 indicates that there are no executable diagnostic perturbations in the current direction.

[0063] For continuously adjustable valves, the current actual opening is Y0, the minimum opening to maintain conduction is YL, the maximum allowable opening is YH, and the minimum valve adjustment step is Sy, with 2Sy taken as the opening safety margin. The upper limit of the perturbation in the direction of increasing the opening is determined according to YH-Y0-2Sy; the upper limit of the perturbation in the direction of decreasing the opening is determined according to Y0-YL-2Sy. When the calculation result is not greater than 0, the corresponding direction is not used to diagnose the perturbation.

[0064] In this way, the upper limit of the perturbation amplitude is determined by the actual operating point of the current equipment, the boundary of the conveying condition, the current effective protection threshold, and the control resolution of the actuator. Therefore, it can be determined whether the perturbation value to be used will exceed the condition switching boundary, protection boundary, and shutdown boundary before the perturbation is implemented.

[0065] When using suction power equipment to implement perturbation, the operating parameters are slightly changed while maintaining the waste conveying condition. The perturbation range should not trigger equipment condition switching, protection actions, or shutdowns, and the resulting actual parameter changes should be verifiable through feedback. When using continuously adjustable valves to implement perturbation, the actual opening is changed only within the range where the valve's conduction relationship remains unchanged, thus ensuring that the effective conveying subtopology remains unchanged before and after the perturbation. After determining the upper limit of the perturbation amplitude Umax, the lower limit of the perturbation amplitude Umin that can generate a recognizable diagnostic response is further determined. During the diagnostic propagation side feature establishment stage, under the conditions of no physical faults, correct digital twin verification, and valid pressure measurement data, starting from the minimum adjustment step size of the diagnostic execution equipment, the perturbation amplitude is increased step by step according to one minimum adjustment step size. After each level of perturbation is completed, the diagnostic execution equipment is reset, and the next level of testing is performed only after the pipeline network returns to a stable state.

[0066] For each level of testing, it is checked whether the actual feedback change from the diagnostic execution device is valid, and whether the pressure measurement point located on the front side of the propagation edge to be verified on the diagnostic propagation path generates an effective response exceeding its pressure response identification threshold. The perturbation amplitude that first stably generates an effective pressure response is determined as the identifiable perturbation lower limit Umin corresponding to the diagnostic execution device, the effective transport sub-topology, and the operating condition. To avoid single random fluctuations, the same amplitude is repeated at least 3 times. Only when all 3 times generate an effective response on the front side that conforms to the characteristics of the diagnostic propagation edge is the amplitude recorded as Umin.

[0067] Then, Umin and Umax are compared. When Umin is less than or equal to Umax, it is determined that there is an executable diagnostic perturbation range, which is an amplitude range not less than Umin and not greater than Umax. In actual execution, the minimum amplitude not less than Umin is selected from the discrete adjustment step size achievable by the diagnostic execution device as the amplitude of this diagnostic perturbation to reduce the impact on the normal waste transportation process. When Umin is greater than Umax, it indicates that there is no perturbation amplitude that simultaneously satisfies identifiability and operational safety under the current operating conditions. In this case, the restricted diagnostic perturbation is not implemented, and the stable propagation anomaly segment is directly transferred to the actual relationship verification according to the aforementioned rules of this embodiment.

[0068] Therefore, this embodiment does not presuppose that "the identifiable lower limit and the safety upper limit must have an intersection", but instead determines whether there is an intersection between them by calculating Umin and Umax respectively; it only performs perturbation when the intersection exists, and enters the relationship verification process when there is no intersection.

[0069] Diagnostic propagation edge features are pre-established under conditions where no faults are confirmed, the model relationships are consistent with the field, and the measurement point data is valid. When establishing diagnostic propagation edge features, the Umax of each test direction is first determined according to the aforementioned calculation method for the safe upper limit of perturbation amplitude. The test amplitude is then increased incrementally starting from the minimum adjustment step size of the diagnostic execution device, implementing a baseline perturbation within the range not exceeding Umax. For each perturbation level, the actual feedback change of the diagnostic execution device, the pressure response direction of the measurement points before and after each propagation edge, the response arrival time, and the pressure change amplitude are recorded. For perturbation amplitudes that can stably form an effective response, diagnostic propagation edge features are established according to perturbation type, perturbation direction, and amplitude level; perturbation amplitudes that cannot form an identifiable response are not included in the diagnostic propagation edge feature library.

[0070] When calling the diagnostic propagation edge features in subsequent actual diagnoses, only features that match the current diagnostic execution device, the current effective transport sub-topology, the relationship between the execution device and the propagation edge topology, the perturbation direction, and the actual perturbation amplitude level are called. Perturbation responses of other amplitude levels are not used as the basis for judging the current propagation edge.

[0071] After implementing the restricted diagnostic perturbation, the actual feedback change of the diagnostic execution device is first confirmed, and then candidate propagation edges are determined along the propagation path of the diagnostic response. If a valid change has been formed on the front side of a candidate propagation edge and the data on the back side is valid, but the back side does not form a valid response that meets the characteristics of the diagnostic propagation edge within the corresponding time range, then evidence of physical fault reinforcement is formed. If a response is formed on the back side, but its direction, time, and amplitude transmission characteristics that should be used for judgment do not meet the characteristics of the corresponding diagnostic propagation edge, then perturbation propagation mismatch is also constituted.

[0072] After each perturbation, the diagnostic execution equipment is restored to its pre-perturbation state and waits for the pipeline dynamic response to end and for the system to enter a stable state for a preset duration before implementing the next independent perturbation. This avoids the same pressure dynamic process being counted repeatedly. In multiple diagnostic perturbations, if the model predicts that locations should not be affected by the perturbation, especially pressure measurement points outside the current effective transport sub-topology, repeatedly generate effective pressure responses with a clear temporal correlation to the perturbation's occurrence time, then this constitutes candidate evidence of model mismatch.

[0073] When a stable propagation anomaly segment lacks a suitable diagnostic execution device or the current state lacks matching diagnostic propagation edge features, active perturbation is not enforced. Instead, the stable propagation anomaly segment is directly transferred to actual relationship verification. This avoids interrupting the diagnostic process due to the absence of available execution devices on-site.

[0074] Model mismatch candidate evidence doesn't only originate from constrained diagnostic perturbations. In normal independent diagnostic events, if the digital twin topology predicts that a pressure gauge should not be affected by the current state change source, but this pressure gauge repeatedly produces valid responses with temporal correlation to the state change source in multiple independent diagnostic events, this also constitutes model mismatch candidate evidence. Especially when a pressure gauge outside the current valid delivery sub-topology exhibits a stable correlated response, it indicates that there may be unrecorded connections in the digital twin model, pressure gauge binding deviations, or valve state correlation errors in the field.

[0075] For candidate evidence of model mismatch, stable propagation anomaly segments that do not meet the perturbation execution conditions, and stable propagation anomaly segments that still lack sufficient independent entity state evidence to determine the faulty entity after limited diagnostic perturbation, a verification relationship is formed. The verification relationship can include at least pipe segment topology connection relationship, pressure measurement point binding relationship, valve state association relationship, and equipment topology location.

[0076] Subsequently, on-site verification data was obtained. Specifically, pipe segment topology connections were confirmed through as-built documentation, on-site pipeline inspections, and maintenance records; pressure measurement point binding relationships were confirmed through sensor installation locations, numbers, and corresponding connected pipe segments; valve status relationships were confirmed through valve actual locations, actual opening degrees, and feedback signal sources; and equipment topology locations were confirmed through equipment actual installation locations and connecting pipe segments. The on-site verification results were then compared item by item with the corresponding relationships in the digital twin topology.

[0077] When the actual on-site relationships differ from the digital twin topology, the corresponding relationships in the digital twin topology are updated using data verified on-site. For example, if it is confirmed on-site that a pressure sensor is actually installed on another pipe segment, its binding entity is updated; if it is confirmed that a valve feedback signal actually corresponds to another valve, the valve status association is updated; if it is confirmed that the connection direction has changed after on-site pipeline modification, the pipe segment connection relationship is updated. After updating, a valid transport sub-topology search is re-executed, and the source-end propagation edges, measuring-point propagation edges, and composite propagation sections affected by the relationship change are re-determined.

[0078] After the digital twin topology is updated, the original propagation edge features do not automatically remain valid. The system checks whether the existing normal propagation edge features and diagnostic propagation edge features still correspond to the same actual physical propagation relationship based on the updated entity connections, measurement point bindings, state associations, and operating conditions. After the model relationship is updated, the system compares the front-end entity, back-end pressure measurement point, actual physical connection path, state change source and propagation edge topology relationship, and pressure measurement point binding relationship of the propagation edge before and after the update. If none of the above items have changed, and the update has not changed the effective transmission path of the propagation edge, the original propagation edge features remain valid; if any of these items have changed, the original propagation edge features are not used, the corresponding propagation edge is marked as undecidable, and matching features are regenerated according to the aforementioned methods for establishing normal and diagnostic propagation edge features.

[0079] When the on-site relationship verification results are consistent with the digital twin topology, and the stable propagation anomaly segment has already formed evidence of entity failure reinforcement in the restricted diagnostic perturbation, the explanation that the propagation mismatch is caused by the current model relationship to be verified is ruled out, and the corresponding stable propagation anomaly segment can be identified as a fault segment. If the on-site relationship verification is consistent but the restricted diagnostic perturbation has not formed evidence of entity failure reinforcement, then the entity failure is not forcibly identified. Instead, the stable propagation anomaly segment is retained, and new propagation evidence and entity status evidence are obtained by continuing to use subsequent independent diagnostic events.

[0080] If, after the model relationship is verified to be incorrect and updated on-site, other unexplained propagation conflicts still exist using the updated digital twin topology, the old model relationship that has been verified to be incorrect on-site will not be restored due to the continued existence of conflicts. Instead, the updated result after on-site verification will be retained, and entity fault verification and other model relationship verification will continue to be performed on the remaining conflicts. This allows for the simultaneous existence of model relationship mismatch and actual entity faults in the same pipeline network.

[0081] For example, in a waste transport network with a main conveying pipeline and two branch pipelines, waste disposal end A is connected to the main conveying pipeline via valve V1. The main conveying pipeline is subsequently connected to the two branch pipelines via branch valves V2 and V3, respectively. A suction power unit F is located at the collection end, and pressure measuring points P1, P2, P3, and P4 are arranged along each pipeline. When waste disposal end A is in operation, suction power unit F is in conveying mode, V1 and V2 are in a conducting state, and V3 is in a non-conducting state, the current effective conveying sub-topology is formed based on field feedback. If the actual operating frequency of suction power unit F is confirmed by feedback to have adjusted from a first stable value to a second stable value, this change is considered a diagnostic event. Based on the position of F in the effective conveying sub-topology, the source-end propagation edge and subsequent measuring point propagation edge corresponding to each conducting direction are determined.

[0082] Assuming that under a confirmed fault-free baseline condition, after a change in operating parameters of the same magnitude occurs in F, P4, P3, P2, and P1 sequentially generate pressure changes in the expected direction within their respective baseline response time ranges. The corresponding response sequence, response time, and amplitude propagation relationship are then preserved as the normal propagation edge characteristics under this operating condition. When the feedback change in F occurs during actual operation, P4 and P3 both generate valid responses conforming to the baseline, while P2 does not generate a valid response within its corresponding time range. Furthermore, P2's data acquisition is normal. Therefore, the propagation edge corresponding to P3 and P2 becomes the first propagation mismatch edge on the current diagnostic propagation path. The pipe segment covered by this propagation edge and the valve entity within it are identified as a fault candidate area.

[0083] In subsequent independent diagnostic events, if the initial propagation mismatch repeatedly falls between P3 and P2, a stable propagation anomaly segment is formed. If independent feedback is obtained from the field indicating an abnormal actual opening of V2, and V2 is located within this stable propagation anomaly segment, and the direction of the pressure impact caused by its abnormal opening is consistent with the direction of the propagation mismatch where P3 responds normally while P2's response is weakened, then this feedback constitutes independent entity status evidence. The corresponding area can be identified as the fault segment, and V2 can be further identified as the fault entity.

[0084] If V2 feedback, the status of the suction power equipment, and other independent entity data cannot confirm a specific entity anomaly, then a restricted diagnostic perturbation is implemented on the suction power equipment F, located outside the stable propagation anomaly section and capable of allowing the diagnostic response to pass through that section. If, after a small, pre-set change in the actual operating parameters of F and confirmation through feedback, P3 generates an effective response conforming to the diagnostic propagation side characteristics, while P2 still fails to generate an effective response conforming to the diagnostic propagation side characteristics, then evidence of entity fault reinforcement is formed. Subsequently, the actual installation locations of P2 and P3, the actual status correlation of V2, and the actual connection relationships of the corresponding pipe sections are verified. If the verification results are consistent with the digital twin topology, then the corresponding section between P3 and P2 can be further confirmed as the fault section.

[0085] If, during the aforementioned perturbation process, another non-conducting branch measuring point besides P1, which should not be affected by the current perturbation F according to the current effective transport sub-topology, repeatedly forms a pressure response consistent with the timing of F changes in multiple perturbations, then this forms candidate evidence of model mismatch. If, after on-site verification, it is found that the V3 feedback signal is incorrectly bound to the valve entity in the digital twin topology, then the V3 state association is updated according to the actual source of the feedback signal, the effective transport sub-topology and affected propagation edges are redefined, and it is checked whether the original propagation edge characteristics are still applicable to the update relationship, thereby avoiding misjudging the digital twin relationship as pipeline blockage.

[0086] Through the above implementation methods, this invention maintains the correlation between real-time equipment status and pressure data and the current actual conduction topology. Starting from the source of state change, it uses source-end propagation edges and measurement-point propagation edges to describe the continuous propagation process of pressure changes in the pipeline network. It enhances diagnostic evidence step by step through initial propagation mismatch, repeated verification of independent diagnostic events, independent entity status evidence, restricted diagnostic perturbation, and verification of actual on-site relationships. This enables the location of actual pipeline fault sections and specific fault entities, as well as the identification and correction of model mismatches in digital twin topology connections, measurement point bindings, valve status associations, and equipment locations. After updating the model relationships, it can also re-verify the applicability of propagation edge features to ensure that subsequent fault location continues to be based on a digital twin model consistent with actual on-site relationships.

[0087] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for digital twin state mapping and fault location of a waste transport pipeline network, characterized in that, include: Establish a digital twin topology for the waste transport network to record the connections and binding relationships between pipe sections, waste disposal ends, valves, suction power equipment, and pressure measuring points; Based on the working status of the waste disposal end, the conveying condition of the suction power equipment, and the valve feedback conduction status, an effective conveying sub-topology is determined. Real-time pressure and equipment status are mapped to the topology, and diagnostic events whose status change sources can be determined and whose changes have been confirmed by feedback are selected. The unique conductive section from the state change source entity to the first pressure measuring point in each direction, and the unique conductive section between adjacent pressure measuring points are determined as propagation edges. Propagation edge features that match the sub-topology, operating conditions, state change source, propagation edge topology relationship and change conditions are called to form diagnostic propagation paths from the state change source along each direction. The first mismatched edge coverage section of each path is determined as a fault candidate area. By repeatedly verifying independent diagnostic events, a stable propagation anomaly segment is formed when the first mismatch repeatedly covers the same segment; the fault segment is identified when the evidence of the state of an independent entity is consistent with the topological location and direction of influence of the anomaly, and the fault entity is identified when the evidence points to a specific entity; when the evidence is insufficient and there is a diagnostic execution device that meets the conditions, a restricted diagnostic perturbation that keeps the sub-topology unchanged is implemented; when the perturbation mismatch occurs, evidence of entity fault reinforcement is formed; when the location that the model predicts should not respond repeatedly forms an associated response in independent diagnostic events, candidate evidence of model mismatch is formed. For model mismatch candidate evidence, stable propagation anomaly segments that do not meet perturbation conditions and still lack independent entity state evidence after perturbation, obtain actual relationship verification data; When the actual relationship is inconsistent with the topology, the relationship is updated; when the actual relationship is consistent and the perturbation forms evidence of strengthening the entity fault, the faulty segment is determined. After verification, check the applicability of the propagation edge features, and continue positioning based on the verified topology.

2. The method for digital twin state mapping and fault location of waste transport pipeline network according to claim 1, characterized in that, Establishing a digital twin topology for the waste transport network includes: Acquire pipeline network structure data, pipeline geometric parameters, and equipment spatial locations, and assign a unique number to each entity; Pipe segments record geometric parameters and connection relationships; valves record connected pipe segments, control status and actual feedback status; pressure measuring points record the bound entities. The topology is formed based on the physical connections of the entities for path search and state propagation analysis, and the topological positions of valves, suction power equipment and pressure measuring points are recorded.

3. The method for digital twin state mapping and fault location of waste transport pipeline network according to claim 1, characterized in that, Determine the valid transport subtopology, including: The working status of the waste disposal end, the conveying condition of the suction power equipment, and the valve feedback conduction status are detected, and the period during which the combined status is maintained is determined as a waste conveying event. The current event ends when any of the following occurs: the working status of the waste disposal end changes, the valve conduction relationship changes, or the suction power equipment exits the conveying mode. Changes in the operating parameters of the suction power equipment during the conveying mode do not change the event boundary. Searching along the guiding path from the waste disposal end to the suction power equipment, entities that meet the guiding conditions and pressure measuring points are combined to form an effective transport sub-topology; when the measuring points outside the sub-topology form an effective response that is temporally correlated with the diagnostic event, evidence of abnormal relationship in the model prediction is formed.

4. The method for digital twin state mapping and fault location of waste transport pipeline network according to claim 1, characterized in that, The establishment of propagation edges and their characteristics includes: When there are no other configured pressure measurement points between the source entity of the state change and the first pressure measurement point in a certain propagation direction, and there is a unique continuous physical path, the corresponding entity and segment are determined as the source propagation edge; when there are no other configured pressure measurement points between adjacent configured pressure measurement points, and there is a unique continuous physical path, the corresponding segment is determined as the measurement point propagation edge. When there are two or more simultaneously conducting paths between the aforementioned endpoints, the area formed by the corresponding paths is defined as the composite propagation segment; Under the condition that the entity connection, measurement point binding and state association are verified to be consistent with the actual situation, the relevant entities are fault-free and the measurement point data is valid, the baseline diagnostic event that can be determined by the state change source is established according to the effective transmission sub-topology, operating conditions, the relationship between the state change source and the propagation edge topology, as well as the change type, direction and amplitude range. The normal propagation edge characteristics characterize the direction, time and amplitude relationship between the effective change on the front side and the pressure response on the back side. The measurement point propagation edge also characterizes the response order of adjacent measurement points and the steady-state effective pressure difference range. Under the aforementioned conditions, a single diagnostic execution device is used to implement a preset perturbation, and diagnostic propagation edge characteristics are established according to the topological relationship between the diagnostic execution device and the propagation edge, as well as the perturbation type, direction, and amplitude range. When the validity check of the data involved in the determination fails or the current conditions lack matching propagation edge features, the corresponding propagation edge will be determined as an undeterminable propagation edge.

5. The method for digital twin state mapping and fault location of waste transport pipeline network according to claim 4, characterized in that, The identification of diagnostic events and independent diagnostic events includes: During the effective delivery sub-topology conduction relationship maintenance, either the actual operating parameter change of the suction power equipment confirmed by feedback or the actual opening degree change of the valve maintaining conduction will be regarded as a diagnostic event; Record the source of state change, time of occurrence, type of change, direction of change, and magnitude of change; determine the conduction propagation direction of pressure measurement points from the source of state change along the effective transport sub-topology. When multiple state change sources exist within the same response time window and the response attribution cannot be distinguished, the event is not used to perform the first propagation mismatch localization; when the valve action changes the conduction relationship, the current waste conveying event is terminated and the effective conveying sub-topology is re-determined; When the state change sources of adjacent diagnostic events can be separately attributed, the response time windows do not overlap, and the dynamic response of the previous event ends and enters a stable state for a preset duration, the adjacent diagnostic events are determined to be independent diagnostic events, and the continuous responses generated by the same state change are not split and counted.

6. The method for digital twin state mapping and fault location of waste transport pipeline network according to claim 5, characterized in that, The determination of candidate fault regions includes: The diagnostic propagation path is formed by sequentially connecting the source end propagation edge and the measuring point propagation edge along each conduction propagation direction from the source of state change; For the source-end propagation edge, the actual change confirmed by the feedback of the state change source is taken as the effective change of the front side; for the measurement point propagation edge, the effective response of the measurement point before the reference response order is taken as the effective change of the front side. If a valid change is established on the front side and the data on the back side is valid, and the response on the back side meets the characteristics of a matching propagation edge, then the edge is determined as a propagation consistent edge; if no valid response is formed within the response time range or at least one of the response characteristics that should be judged does not meet the corresponding characteristics, then the edge is determined as a propagation mismatch edge. The segment covered by the mismatched edge during the first propagation of each path is identified as a candidate fault area; when an undeterminable propagation edge is encountered, the localization of the corresponding path is stopped; when a composite propagation segment is encountered, the current round of edge-by-edge localization is stopped and marked as a segment to be distinguished.

7. The method for digital twin state mapping and fault location of waste transmission pipeline network according to claim 6, characterized in that, The formation of evidence of the state of independent entities and evidence of anomalies in measurement point data includes: The physical status is determined by at least one of the following: the pipe section physical verification results obtained independently of the pressure propagation mismatch judgment, valve feedback, and equipment operating parameters; When the corresponding data indicates an entity anomaly, the entity is located in the fault candidate region, and the direction of its influence along the diagnostic propagation path is consistent with the actual propagation mismatch direction, independent entity status evidence is formed; when this evidence can independently point to a specific entity, the faulty entity is identified. When a pressure measurement point repeatedly exhibits at least one of the following in multiple independent diagnostic events: exceeding the measurement range or abnormal timestamps, it constitutes evidence of abnormal measurement point data. When a pressure measurement point maintains a fixed value across multiple independent diagnostic events, and adjacent valid pressure measurement points generate a pressure response consistent with the corresponding diagnostic event, this constitutes evidence of abnormal measurement point data.

8. The method for digital twin state mapping and fault location of waste transport pipeline network according to claim 7, characterized in that, Duplicate validation includes: Filter out multiple independent diagnostic events whose diagnostic propagation paths pass through the same fault candidate region, and call the propagation edge features that match their respective effective delivery sub-topology, operating conditions, state change sources, propagation edge topology relationships, and change conditions; If the data used for the determination is valid and the initial propagation mismatch edge repeatedly covers the same physical segment in multiple independent diagnostic events, the segment is identified as a stable propagation abnormal segment. If the segment to be distinguished forms a unique conductive path and has matching propagation edge characteristics in subsequent independent diagnostic events, it continues to be located according to the propagation edge; if the aforementioned conditions are not met, it remains in the state of being to be distinguished. When the location where the model predicts should not respond repeatedly forms a valid response that is temporally correlated with the source of state change in multiple independent diagnostic events, it forms candidate evidence of model mismatch.

9. The digital twin state mapping and fault location method for waste transport pipelines according to claim 8, characterized in that, Limited diagnostic perturbations include: When a stable propagation anomaly segment lacks independent entity state evidence to explain the propagation anomaly and can correspond to a propagation edge, determine whether there is a diagnostic execution device located outside the anomaly segment that is operating normally, provides consistent feedback, and whose diagnostic response can pass through the anomaly segment; When a diagnostic execution device that meets the conditions exists, a perturbation is implemented by selecting one of the suction power equipment or the continuous opening regulating valve; the suction power equipment changes its operating parameters within a range that does not trigger operating condition switching, protection, or shutdown, and the valve changes its opening within a range that maintains the conduction relationship unchanged; Invoke diagnostic propagation edge features that match the diagnostic execution device, propagation edge topology, and perturbation conditions; when a candidate propagation edge forms a valid change on the front side and the data on the back side is valid, but the back side does not form a valid response that satisfies the corresponding features, evidence of physical fault reinforcement is formed. After each perturbation, the diagnostic execution device is reset, and the next perturbation is implemented after the dynamic response ends and the device enters a stable state for a preset duration. When the external measurement point of the subtopology repeatedly forms an effective response that is temporally correlated with the perturbation in multiple diagnostic perturbations, it forms candidate evidence of model mismatch. When there is no diagnostic execution device that meets the conditions or when there is a lack of matching diagnostic propagation edge features, the corresponding perturbation is not implemented and the stable propagation abnormal segment is transferred to relation verification.

10. The method for digital twin state mapping and fault location of waste transport pipeline network according to claim 1, characterized in that, The handling of model mismatch includes: For model mismatch candidate evidence, stable propagation anomaly sections that do not meet perturbation execution conditions and still lack independent entity state evidence after perturbation, at least one of the following relationships to be verified is formed: pipe segment topology connection, pressure measurement point binding, valve state association, and equipment topology location. Acquire data on actual pipeline connections, measurement point installation locations, valve status, feedback signal sources, and actual equipment locations, and compare them with the digital twin topology; When the verification relationship is inconsistent, the corresponding relationship is updated according to the actual data, and the affected valid transport sub-topology and propagation edge are re-determined; when the existing propagation edge features cannot be confirmed to be applicable to the update relationship, the corresponding edge is determined as an undeterminable propagation edge, and the matching features are re-established. When a stable propagation anomaly segment with consistent verification relationships forms evidence of physical fault reinforcement in a restricted diagnostic perturbation, the corresponding fault segment is identified; when a stable propagation anomaly segment with consistent verification relationships does not form evidence of physical fault reinforcement, the stable propagation anomaly segment is retained and further verified using subsequent independent diagnostic events; when unexplained propagation conflicts still exist after model updates, the verified and confirmed update relationships are retained and the remaining conflicts are further verified.

Images