Underground pipe network full life cycle intelligent detection and visualization management system based on pnim

Abstract

The application discloses a kind of underground pipe network full life cycle intelligent detection and visual management system and method based on PNIM.The system includes: multi-source collaborative pipe network information acquisition module, static archives and real-time detection data are collected;Multi-modal data preprocessing module, time-space unification and filtering fusion are carried out;PNIM holographic data modeling module, data object containing static, dynamic and inference data domain is constructed for each pipe section;Dynamic risk early warning evaluation module, the history damage memory value is calculated using exponential-power hybrid attenuation function, and the comprehensive risk is calculated by combining benchmark risk and environmental comprehensive disturbance, and the grade is determined according to dynamic self-adaptive threshold;Visual modeling and early warning display module, data object and risk grade are converted into three-dimensional model and graded rendering early warning.The application realizes the full life cycle intelligent management of underground pipe network from static table to dynamic early warning, and improves the early warning accuracy and efficiency.

Description

Technical Field

[0001] This invention relates to the field of underground pipeline information detection technology, and more specifically to an intelligent detection and visualization management system for the entire life cycle of underground pipeline networks based on PNIM. Background Technology

[0002] Underground pipelines are considered the "lifeline" of cities and industrial areas, and their safe operation is directly related to production, daily life, and public safety. However, underground pipeline networks generally suffer from prominent problems such as a large span of construction periods, a wide variety of materials, chaotic spatial layouts, and missing original records. Existing detection technologies cannot simultaneously and quickly detect all pipelines of different depths and types, often requiring multiple changes of different equipment, resulting in low efficiency. Furthermore, the application of BIM technology in pipeline management is mostly limited to the static "digital sand table" level at the completion stage, only displaying geometric dimensions, materials, and as-built drawing locations. It cannot access and utilize dynamic data generated during pipeline operation (such as historical maintenance records, surrounding soil corrosivity monitoring data, pressure and flow data, etc.), causing the model to quickly become inaccurate after operation and failing to support full lifecycle management.

[0003] In risk assessment, existing methods largely rely on static scoring based on expert experience or simple linear weighting formulas, lacking the ability to model the time-varying characteristics of the impact of historical events and the coupling relationship between environmental disturbances and historical damage. For example, the contribution of a major repair two years ago to current risk should be significantly different from that of a routine repair last month, but current technology cannot quantitatively characterize this "time decay memory" effect. Furthermore, fixed risk thresholds can easily lead to misjudgments between plant areas with different environmental benchmarks—a normal pipeline in a highly corrosive environment might be misclassified as high-risk, while an older pipeline in a favorable environment might be underestimated.

[0004] Therefore, there is an urgent need for an intelligent pipeline management system that can integrate multi-source dynamic data and has time decay memory and adaptive threshold capabilities, so as to realize the leap from passive fault response to proactive risk warning in underground pipeline networks. Summary of the Invention

[0005] In view of this, the present invention provides a rapid detection and visualization modeling system for underground pipe networks.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A PNIM-based intelligent detection and visualization management system for the entire lifecycle of underground pipeline networks, comprising: The multi-source collaborative pipeline network information acquisition module is used to collect static archive information and real-time detection data of underground pipeline networks; The multimodal data preprocessing module is used to perform spatiotemporal unification and filtering fusion on the collected data; The PNIM holographic data modeling module is used to construct a PNIM data object for each pipe segment, which includes a static data domain, a dynamic data domain, and an inference data domain. The dynamic risk warning and assessment module based on the adaptive decay memory model is used to calculate the historical damage memory value based on the historical maintenance events in the static data domain using an exponential-power hybrid decay function, and to calculate the comprehensive environmental disturbance risk value based on the real-time environmental data in the dynamic data domain. The module combines the benchmark risk value, the historical damage memory value, and the comprehensive environmental disturbance risk value to generate a comprehensive risk value, and then determines the risk level based on the dynamic adaptive threshold. The PNIM visualization modeling and early warning display module is used to convert PNIM data objects and risk levels into three-dimensional visualization models and perform hierarchical rendering and early warning.

[0007] Preferably, the dynamic risk warning and assessment module based on the adaptive decay memory model specifically performs the following steps: Based on the historical maintenance events in the static data domain of the PNIM data object for each segment of the underground pipeline network, the historical damage memory value is calculated using an exponential-power hybrid decay function. The comprehensive risk value of environmental disturbances to the pipeline is obtained based on real-time environmental data in the dynamic data domain of the PNIM data object, which is affected by soil corrosion, medium flow, and meteorological parameters. A comprehensive risk value is generated based on the obtained historical damage memory value, comprehensive environmental disturbance risk value, and pipeline baseline risk value; Calculate and sort the comprehensive risk values ​​of all pipe segments in the current area, obtain the risk percentile of each pipe segment, and determine different risk levels based on the relationship between the risk percentile of each pipe segment and a set threshold.

[0008] Preferably, based on the historical maintenance events of each segment of the underground pipeline network, the historical damage memory value is calculated using an exponential-power hybrid decay function, specifically including the following calculation formula:

[0009] Where K is the total number of maintenance events that have occurred in pipe segment i since its completion, and the subscript j is the sequence number of the maintenance event, j=1 indicates the earliest maintenance event, and j=K indicates the most recent maintenance event; Let t be the time when the j-th maintenance event occurs, and t be the current assessment time. This represents the time interval between the j-th maintenance event and the current assessment time. The event type coefficient represents the severity of the j-th maintenance event. Let be the repair quality coefficient for the j-th repair event. Let λ be the time of the maintenance event, λ be the decay rate constant, and β be the decay shape parameter, where 0 < β ≤ 1.

[0010] Preferably, the comprehensive environmental disturbance risk value is calculated based on the real-time environmental data in the dynamic data domain using the following formula:

[0011] in, , , These are the weighting coefficients for each item. This represents the function that influences soil corrosivity. This represents the function that influences the flow of the medium in the pipeline. This represents the influence function of meteorological parameters, where pH is the soil acidity / alkalinity and ρ is the soil resistivity. stray current intensity, The pressure of the medium inside the pipe, Q , where T is the flow rate of the medium inside the pipe, T is the soil ambient temperature, and RH is the relative humidity of the soil ambient.

[0012] Preferably, the soil corrosivity influence function specifically includes the following functions:

[0013] In the formula, Here is the reference resistivity, and a and c are constants.

[0014] Preferably, the pipeline medium flow influence function specifically includes the following functions:

[0015]

[0016]

[0017] in, Design pressure for pipelines, Design flow rate for the pipeline, This represents the pressure risk growth coefficient. This represents the traffic risk growth coefficient. The pressure-flow synergistic influence coefficient is determined by the pipe section's radius of curvature and wall thickness.

[0018] Preferably, the environmental impact function specifically includes the following functions:

[0019]

[0020]

[0021] in, , , These are the preset lower temperature threshold, upper temperature threshold, and upper humidity threshold. This represents the low-temperature risk growth coefficient. This represents the high-temperature risk growth coefficient. This represents the humidity risk growth factor. The temperature and humidity synergistic effect coefficient is determined by the resistance test data of the pipeline external anti-corrosion coating material.

[0022] Preferably, the dynamic risk warning and assessment module based on the adaptive decay memory model determines the risk level according to the dynamic adaptive threshold, specifically including the following steps: Calculate and sort the comprehensive risk values ​​of all pipe segments within the current time area, and obtain the risk percentile P for each pipe segment. i (t); When P i When (t) ≥ 95%, it is classified as an extreme level of concern; When 85%≤P i When (t) < 95%, it is classified as a level of high concern; When P i When (t) < 85%, it is determined to be at the normal surveillance level.

[0023] Preferably, the dynamic risk early warning and assessment module further includes an adjacent pipe segment impact transmission unit: when the risk level of pipe segment i reaches the key concern level or above, the system retrieves the topological connection relationship in the PNIM static data domain to identify adjacent pipe segments, estimates the additional dynamic load increment ΔF{i→k} on adjacent pipe segment k when pipe segment i fails, and adjusts the risk value of adjacent pipe segments according to the following formula:

[0024] in, The transfer factor is determined by the pipe material and connection stiffness, 0 < ≤1.

[0025] Preferably, the PNIM visualization modeling and early warning display module includes: The input layer is used to load the pipeline space coordinates and geometric dimensions in the PNIM static data domain into the 3D GIS scene. The analysis layer is used to monitor updates to the PNIM dynamic data domain and inference data domain in real time through the change detection engine, driving updates to the 3D model rendering style. The output layer is used to render pipelines in three levels of colors—green, yellow, and red—based on risk percentiles, and to display flow arrows to indicate the risk transmission path when the impact of adjacent pipe segments is transmitted.

[0026] Preferably, the multi-source collaborative pipeline network information acquisition module simultaneously deploys electromagnetic induction equipment and ground penetrating radar equipment to collect real-time detection data of the underground pipeline network, achieves trigger synchronization through the control unit, and adopts an adaptive step size adjustment algorithm: a large step size is used for acquisition in straight pipeline sections, and the step size is automatically reduced in pipeline intersection or change-of-direction areas.

[0027] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a PNIM-based intelligent detection and visualization management system for the entire life cycle of underground pipelines, which has the following beneficial effects.

[0028] This invention uses a multi-source collaborative pipeline network information acquisition module to simultaneously collect static archive information and multiple types of real-time detection data, and employs an adaptive step size adjustment algorithm to improve acquisition efficiency while ensuring data quality.

[0029] This invention achieves spatiotemporal unification and filtering fusion of multi-source heterogeneous data through a multimodal data preprocessing module, effectively solving the data silo problem and providing a high-quality data foundation for subsequent modeling and evaluation.

[0030] This invention utilizes the PNIM holographic data modeling module to construct a three-domain integrated data object for each pipe segment, comprising a static data domain, a dynamic data domain, and an inference data domain, thereby realizing full lifecycle information management of underground pipeline networks from completion to decommissioning.

[0031] This invention calculates historical damage memory values ​​using an exponential-power hybrid attenuation function, enabling the risk assessment model to "memorize" historical maintenance events and adaptively adjust the attenuation shape parameters according to the pipe type, making the assessment results more consistent with engineering practice.

[0032] This invention systematically constructs soil corrosivity functions, pipeline medium flow influence functions, and environmental influence functions, enabling a refined quantitative assessment of the coupled effects of multiple factors such as electrochemical corrosion, hydraulic scouring, and temperature and humidity changes.

[0033] This invention employs a risk percentile dynamic adaptive threshold determination method, which dynamically adjusts the early warning standard based on the overall risk distribution of the regional pipeline network, effectively overcoming the inherent defects of the fixed threshold method.

[0034] This invention achieves risk transmission and tracking in the pipeline network topology through the mechanism of influence transmission between adjacent pipe segments, providing technical support for preventing the domino effect.

[0035] This invention utilizes the PNIM visualization modeling and early warning display module to achieve real-time dynamic 3D rendering of risk status and hierarchical early warning linkage, greatly improving the situational awareness and decision-making efficiency of operation and maintenance personnel. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0037] Figure 1 This is a schematic diagram of the overall structure provided in an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the execution steps of the dynamic risk warning and assessment module of the adaptive decay memory model provided in this embodiment of the invention.

[0038] Figure 3 This is a schematic diagram of the overall visualization modeling of a smart factory provided in an embodiment of the present invention.

[0039] Figure 4 This is a schematic diagram of the visualization detection results of an underground pipeline network in a smart factory, provided as an embodiment of the present invention. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] like Figure 1 As shown, this invention discloses a PNIM-based intelligent detection and visualization management system for the entire lifecycle of underground pipeline networks, comprising: The multi-source collaborative pipeline network information acquisition module is used to collect static archive information and real-time detection data of underground pipeline networks; The multimodal data preprocessing module is used to perform spatiotemporal unification and filtering fusion on the collected data; The PNIM holographic data modeling module is used to construct a PNIM data object for each pipe segment, which includes a static data domain, a dynamic data domain, and an inference data domain. The dynamic risk warning and assessment module based on the adaptive decay memory model is used to calculate the historical damage memory value based on the historical maintenance events in the static data domain using an exponential-power hybrid decay function, and to calculate the comprehensive environmental disturbance risk value based on the real-time environmental data in the dynamic data domain. The module combines the baseline risk value, the historical damage memory value, and the comprehensive environmental disturbance risk value to generate a comprehensive risk value, and then determines the risk level based on the dynamic adaptive threshold. The PNIM visualization modeling and early warning display module is used to convert PNIM data objects and risk levels into three-dimensional visualization models and perform hierarchical rendering and early warning.

[0042] The multi-source collaborative pipeline network information acquisition module is responsible for collecting static archive information and real-time monitoring data of underground pipeline networks. Specifically, the static archive information includes, but is not limited to: pipe segment number, pipe material type, pipe diameter, wall thickness, design pressure, design flow rate, completion year, burial depth, spatial coordinates, topological connections, and historical maintenance event records (occurrence time, event type coefficient, repair quality coefficient, etc. for each maintenance event). Real-time monitoring data includes, but is not limited to: internal medium pressure, flow rate, and temperature; external soil pH value, soil resistivity, stray current intensity; and ambient temperature and relative humidity. In practical applications, this module can simultaneously deploy electromagnetic induction devices and ground-penetrating radar devices, achieving trigger synchronization through a control unit and employing an adaptive step-size adjustment algorithm for real-time data detection: large step-size acquisition is used on straight pipeline sections, while the step-size is automatically reduced in areas where pipelines intersect or change direction.

[0043] The multimodal data preprocessing module is used to perform spatiotemporal unification and filtering fusion on the collected data. Specifically, this module performs spatiotemporal registration on data from different sources, in different formats, and with different timestamps, unifying them under the same spatiotemporal reference. It uses filtering algorithms (such as Kalman filtering, wavelet filtering, etc.) to denoise and remove outliers from the original detection data, and interpolates and completes missing data to form high-quality preprocessed data.

[0044] The PNIM holographic data modeling module is used to construct PNIM data objects for each pipe segment, comprising static, dynamic, and inference data domains. Specifically: The static data domain, derived from static archive information, primarily stores the inherent static attributes of the pipe segment, including pipe material, diameter, wall thickness, design parameters, as-built information, spatial coordinates, topological connections, and historical maintenance event records (time of each maintenance, event type coefficient, repair quality coefficient, etc.). These records can be added as maintenance activities occur but do not automatically change over time. The dynamic data domain is based on real-time monitoring data of the pipe segment, including pressure, flow rate, temperature, soil parameters, and environmental parameters, and is dynamically updated over time. The inference data domain stores derived data calculated based on the static and dynamic data domains, including historical damage memory values, environmental disturbance risk values, comprehensive risk values, and risk levels.

[0045] The dynamic risk warning and assessment module based on the adaptive decay memory model is used to perform dynamic risk assessment based on historical maintenance events in the static data domain. Specifically, it includes the following calculation steps: Based on the historical maintenance events of each section of the underground pipeline network, the historical damage memory value is calculated using an exponential-power hybrid decay function. The comprehensive environmental disturbance risk value of the pipeline is obtained based on the effects of soil corrosivity, medium flow, and meteorological parameters. The comprehensive risk value is obtained by summing the historical damage memory value, the comprehensive risk value of environmental disturbance, and the baseline risk value of the pipeline. Alternatively, a weighted summation can be performed as a preferred implementation method to obtain the comprehensive risk value. Calculate and sort the comprehensive risk values ​​of all pipe segments within the current time area to obtain the risk percentile for each segment. Based on the relationship between the risk percentile of each segment and a set threshold, determine different risk levels. Each step is explained in detail below: Historical damage memory values ​​were calculated using a hybrid exponential-power decay function:

[0046] Where K is the total number of maintenance events that have occurred in pipe segment i since its completion, and the subscript j is the sequence number of the maintenance event, j=1 indicates the earliest maintenance event, and j=K indicates the most recent maintenance event; Let t be the time when the j-th maintenance event occurs, and t be the current assessment time. This represents the time interval between the j-th maintenance event and the current assessment time. Let be the event type coefficient for the j-th maintenance event. Let be the repair quality coefficient for the j-th repair event. Let λ be the time of the maintenance event, λ be the decay rate constant, and β be the decay shape parameter, where 0 < β ≤ 1. The impact of historical maintenance events on the current risk of the pipeline segment is not constant but gradually decays over time.

[0047] The exponential-power hybrid decay function controls the shape of the decay curve through the parameter β. When β < 1, the decay rate gradually slows down over time (i.e., the "long memory" effect), which is consistent with the objective law in actual engineering that the impact of pipeline maintenance events weakens over time but does not completely disappear.

[0048] The event type coefficient for maintenance events is determined according to the following rules: When the maintenance event is a routine maintenance, it is defined as follows: =0.3; When the maintenance event is a critical maintenance, it is defined as follows: =0.7; When a maintenance event is an emergency repair, it is defined as follows: =1.0.

[0049] The more urgent and serious the repair, the greater its contribution to subsequent risks.

[0050] Repair quality coefficient of maintenance incidents The range of values ​​for is 0≤ ≤1 indicates that the higher the repair quality (closer to 1), the smaller the contribution of this repair event to subsequent risks.

[0051] The attenuation shape parameter β is determined by backfitting the historical maintenance data of the pipe section with subsequent fault records: corrosion and damage of metal pipes usually have a long "memory" effect, so the value of β is small; the aging behavior of plastic pipes is relatively simple and the memory effect is short, so the value of β is large. In the embodiments of the present invention, when the pipe section material is metal, the value of β is in the range of 0.6~0.8; when the pipe section material is plastic, the value of β is in the range of 0.8~1.0.

[0052] The environmental disturbance risk value is calculated using the following formula:

[0053] in, , , Weight coefficients for each item Soil corrosivity function

[0054] In the formula, Here, ρ is a soil corrosivity function, pH is the soil acidity / alkalinity, and ρ is the soil resistivity. stray current intensity, Using the reference resistivity, a and c are constants. This function is non-linear: the further the soil deviates from neutral, the lower the resistivity, and the greater the stray current, the higher the output value, reflecting the driving effect of the electrochemical corrosion environment on pipeline risks; the first term in this function reflects the combined effect of soil pH and resistivity on electrochemical corrosion. Characterizing the degree to which soil deviates from neutrality, The influence of resistivity on soil conductivity (lower resistivity means higher conductivity and more severe corrosion); the second item Characterizing the contribution of stray currents to the electrochemical corrosion of pipelines; a and c are constants, determined through fitting of laboratory soil corrosion simulation experiments, and the dimension of a is such that... Dimensionless, the dimension of c is m 2 / A; For reference resistivity, a fixed value of 50Ω is used. m; Function of influence of pipeline medium flow

[0055]

[0056]

[0057] The pressure of the medium inside the pipe, Q The flow rate of the medium inside the pipe, Design pressure for pipelines, Design flow rate for the pipeline, This represents the pressure risk growth coefficient. This represents the traffic risk growth coefficient. The pressure-flow synergistic effect coefficient, ≥0, its value is determined by the pipe section's radius of curvature and wall thickness; when the pressure P or flow rate Q exceeds 70% of the design value, the risk increases exponentially, reflecting the accelerating effect of pipeline overpressure and overflow operation on pipe wall fatigue and erosion corrosion; cross-terms This reflects the synergistic deterioration effect when high pressure and high flow rate coexist.

[0058] Environmental impact function:

[0059]

[0060]

[0061] T represents ambient temperature, and RH represents ambient relative humidity. , , These are the preset lower temperature threshold, upper temperature threshold, and upper humidity threshold. This represents the low-temperature risk growth coefficient. This represents the high-temperature risk growth coefficient. This represents the humidity risk growth factor. The coefficient representing the combined influence of temperature and humidity. ≥0, its value is determined by the resistance test data of the pipeline external anti-corrosion coating material.

[0062] The overall risk value of pipe segment i at the current time t is:

[0063] Among them, among them, As the benchmark risk value, This represents the historical damage memory value. This represents the environmental disturbance risk value.

[0064] The dynamic risk warning and assessment module also includes an adjacent pipe segment impact transmission unit: when the risk level of pipe segment i reaches the key concern level or above, the system retrieves the topological connection relationship in the PNIM static data domain to identify adjacent pipe segments, estimates the additional dynamic load increment ΔF{i→k} on adjacent pipe segment k when pipe segment i fails, and adjusts the risk value of adjacent pipe segments according to the following formula:

[0065] in, For the transmission coefficient, 0 < ≤1, determined by the pipe material and connection stiffness.

[0066] In the dynamic risk early warning and assessment module, the method for determining the dynamic adaptive threshold is as follows: calculate and sort the comprehensive risk value of all pipe segments in the current area, and obtain the risk percentile P of each pipe segment. i (t); when P i When (t) ≥ 95%, it is judged as an extreme level of concern; when 85% ≤ P i When (t) < 95%, it is judged as a key level of concern; when P i When (t) < 85%, it is determined to be at the normal monitoring level. The dynamic adaptive threshold mechanism in the dynamic risk early warning assessment module can dynamically adjust the early warning level of each pipe section according to the overall risk distribution of the regional pipeline network, avoiding the problem of missed reports when the overall risk is high and false reports when the overall risk is low when the fixed threshold is used.

[0067] The PNIM visualization modeling and early warning display module is used to transform PNIM data objects and risk levels into 3D visualization models and perform hierarchical rendering and early warning. Specifically, it includes: Input layer: Loads the pipeline spatial coordinates and geometric dimensions from the PNIM static data domain into a 3D GIS scene. Analysis layer: Monitors updates to the PNIM dynamic and inference data domains in real time using a change detection engine, driving updates to the 3D model rendering style. Output layer: Renders pipelines using three color levels (green, yellow, and red) based on risk percentile (green corresponds to normal monitoring level, yellow to key monitoring level, and red to extreme monitoring level), and displays flow arrows to indicate the risk transmission path when the impact of adjacent pipe segments propagates.

[0068] like Figure 1 As shown, the system of this invention comprises five core modules: a multi-source collaborative pipeline network information acquisition module, a multi-modal data fusion and enhancement processing module, a PNIM holographic data modeling module, a dynamic risk early warning assessment module based on an adaptive decay memory model, and a PNIM visualization modeling and early warning display module. The modules communicate with each other via a data bus, forming a complete data link of "acquisition-fusion-modeling-assessment-visualization".

[0069] In actual deployment, the multi-source collaborative pipeline network information acquisition module acquires raw data through various sensors (pressure sensors, flow sensors, temperature sensors, pH sensors, resistivity probes, etc.) and mobile detection devices (electromagnetic induction devices, ground penetrating radar, etc.) deployed in the pipeline network; the data is transmitted to the central server via wired and / or wireless networks; the multimodal data fusion and enhancement processing module in the central server performs spatiotemporal registration, filtering and denoising, and interpolation completion on the raw data; the PNIM holographic data modeling module writes the fused data into the dynamic data domain of the corresponding pipe segment and triggers the update calculation of the inference data domain; the dynamic risk warning assessment module performs risk assessment periodically (e.g., hourly) or event-driven (e.g., when abnormal data is detected); the PNIM visualization modeling and warning display module renders the assessment results in the 3D GIS scene in real time.

[0070] Taking a section of water supply pipe made of metal as an example, this section has experienced 3 maintenance events (K=3) since its completion. The parameters of each maintenance event are as follows:

[0071] The pipe section is made of metal. Historical data was backfitted to determine β=0.7 and λ=0.15 / month. The current assessment time t=0 (i.e., the current year and month) is used as the historical damage memory value formula to obtain the current historical damage memory value. :

[0072] The current monitoring data for the aforementioned metal water supply pipe section is as follows:

[0073] The soil function constant (acid-base resistance term) a is taken as 0.02, and the soil function constant (stray current term) c is taken as 0.5.

[0074] Soil corrosion risk value is obtained based on soil corrosion function. :

[0075] The risk value of the pipeline medium is obtained based on the pipeline medium flow influence function:

[0076]

[0077]

[0078] In this embodiment, the pressure risk growth coefficient kp is 5, the flow risk growth coefficient kq is 4, and the pressure-flow coordination coefficient γ is 0.3.

[0079] The environmental impact risk value is obtained from the environmental impact function:

[0080]

[0081]

[0082] In this embodiment, the high temperature risk growth coefficient is... The humidity risk growth coefficient is 0.1. The coefficient is 0.02, and the temperature and humidity synergy coefficient τ is 0.2.

[0083] The final comprehensive risk value of environmental disturbance is obtained by the following formula:

[0084] In this embodiment, the weighting coefficient , , The values ​​are 0.4, 0.35, and 0.25, respectively.

[0085] Determining the combined risk value and dynamic adaptive threshold: Let the baseline risk value of this pipe segment be Bi = 0.5. Using the historical damage memory value and the comprehensive risk value of environmental disturbance obtained previously, the comprehensive risk value of this pipe segment at the current time t is calculated. Assuming there are 200 pipe segments in the current area, and this pipe segment ranks 8th in comprehensive risk value after sorting, its risk percentile Pi = (200-8) / 200 × 100% = 96%. Since 96% ≥ 95%, where 95% is a set threshold, this pipe segment is determined to be of "extreme concern level." The system immediately renders this pipe segment in red in the 3D scene through the PNIM visualization modeling and early warning display module and pushes early warning information to maintenance personnel. Simultaneously, the system retrieves the topological connection relationship of this pipe segment and identifies the adjacent pipe segment k. The estimated... =0.3, take the transmission coefficient =0.6, then the risk value of the adjacent pipe segment k is adjusted as follows:

[0086] The flow arrows pointing from pipe segment i to pipe segment k are displayed simultaneously in the 3D scene, indicating the risk transmission path.

[0087] The system of this invention was deployed in a smart factory pipeline demonstration area, covering a total length of approximately 120 kilometers of underground pipelines, involving approximately 1500 segments. The overall visual modeling diagram of the smart factory and the visual detection results of the underground pipeline network at a certain moment are shown below. Figure 3 and Figure 4As shown, after the system has been running for 6 months, a comparison is made with traditional manual inspection and fixed threshold early warning methods: The system of this invention identified a total of 23 high-risk pipe sections, and after on-site verification, 20 sections were confirmed to have varying degrees of safety hazards (such as corrosion thinning, interface leakage, etc.), with an accuracy rate of about 87%; traditional manual methods only found 11 potential hazards during the same period.

[0088] The system of this invention issued early warnings 2-4 hours in advance for all three sudden pressure surge events, giving maintenance personnel sufficient time to respond; the traditional fixed threshold method failed to issue timely warnings in one of the events.

[0089] The false alarm rate of the system of this invention is about 12%, which is significantly lower than the 28% of the traditional method.

[0090] Through 3D visualization and hierarchical rendering, the efficiency of maintenance personnel in inspecting key pipeline sections has been improved by approximately 60%.

[0091] The above application examples fully demonstrate the effectiveness, accuracy, and practicality of the system of the present invention.

[0092] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0093] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A PNIM-based intelligent detection and visualization management system for the entire lifecycle of underground pipeline networks, characterized in that, include: The multi-source collaborative pipeline network information acquisition module is used to collect static archive information and real-time detection data of underground pipeline networks; The multimodal data preprocessing module is used to perform spatiotemporal unification and filtering fusion on the collected data; The PNIM holographic data modeling module is used to construct a PNIM data object for each pipe segment, which includes a static data domain, a dynamic data domain, and an inference data domain. The dynamic risk warning and assessment module based on the adaptive decay memory model is used to calculate the historical damage memory value based on the historical maintenance events in the static data domain using an exponential-power hybrid decay function, and to calculate the comprehensive environmental disturbance risk value based on the real-time environmental data in the dynamic data domain. The module combines the benchmark risk value, the historical damage memory value, and the comprehensive environmental disturbance risk value to generate a comprehensive risk value, and then determines the risk level based on the dynamic adaptive threshold. The PNIM visualization modeling and early warning display module is used to convert PNIM data objects and risk levels into three-dimensional visualization models and perform hierarchical rendering and early warning.

2. The system according to claim 1, characterized in that, Based on the historical maintenance events of each segment of the underground pipeline network, the historical damage memory value is calculated using an exponential-power hybrid decay function, specifically including the following calculation formula: Where K is the total number of maintenance events that have occurred in pipe segment i since its completion, and the subscript j is the sequence number of the maintenance event, j=1 indicates the earliest maintenance event, and j=K indicates the most recent maintenance event; Let t be the time when the j-th maintenance event occurs, and t be the current assessment time. This represents the time interval between the j-th maintenance event and the current assessment time. The event type coefficient represents the severity of the j-th maintenance event. Let be the repair quality coefficient for the j-th repair event. Let λ be the time of the maintenance event, λ be the decay rate constant, and β be the decay shape parameter, where 0 < β ≤ 1.

3. The system according to claim 1, characterized in that, The comprehensive environmental disturbance risk value is calculated based on the real-time environmental data in the dynamic data domain using the following formula: in, , , These are the weighting coefficients for each item. This represents the function that influences soil corrosivity. This represents the function that influences the flow of the medium in the pipeline. This represents the influence function of meteorological parameters, where pH is the soil acidity / alkalinity and ρ is the soil resistivity. stray current intensity, Where is the pressure of the medium inside the pipe, Q is the flow rate of the medium inside the pipe, T is the soil ambient temperature, and RH is the relative humidity of the soil ambient.

4. The system according to claim 3, characterized in that, The soil corrosivity influence function specifically includes the following functions: In the formula, Here is the reference resistivity, and a and c are constants.

5. The system according to claim 3, characterized in that, The pipeline medium flow influence function specifically includes the following functions: in, Design pressure for pipelines, Design flow rate for the pipeline, This represents the pressure risk growth coefficient. This represents the traffic risk growth coefficient. The pressure-flow synergistic influence coefficient is determined by the pipe section's radius of curvature and wall thickness.

6. The system according to claim 3, characterized in that, The environmental impact function specifically includes the following functions: in, , , These are the preset lower temperature threshold, upper temperature threshold, and upper humidity threshold. This represents the low-temperature risk growth coefficient. This represents the high-temperature risk growth coefficient. This represents the humidity risk growth factor. The coefficient representing the combined effect of temperature and humidity is determined from the resistance test data of the external anti-corrosion coating material of the pipeline.

7. The system according to claim 1, characterized in that, The dynamic risk warning and assessment module based on the adaptive decay memory model determines the risk level according to the dynamic adaptive threshold, specifically including the following steps: Calculate and sort the comprehensive risk values ​​of all pipe segments in the current time area to obtain the risk percentile Pi(t) of each pipe segment; When Pi(t) ≥ 95%, it is judged as an extreme level of concern; When 85%≤Pi(t)<95%, it is judged as a key level of concern; When Pi(t) < 85%, it is determined to be at the normal monitoring level.

8. The system according to claim 7, characterized in that, The dynamic risk early warning and assessment module also includes an adjacent pipe segment impact transmission unit: When the risk level of pipe segment i reaches the key concern level or above, the system retrieves the topological connection relationship in the PNIM static data domain to identify adjacent pipe segments, estimates the additional dynamic load increment ΔF{i→k} on adjacent pipe segment k when pipe segment i fails, and adjusts the risk value of adjacent pipe segments according to the following formula: in, The transfer factor is determined by the pipe material and connection stiffness, 0 < ≤1.

9. The system according to claim 1, characterized in that, The PNIM visualization modeling and early warning display module includes: The input layer is used to load the pipeline space coordinates and geometric dimensions in the PNIM static data domain into the 3D GIS scene. The analysis layer is used to monitor updates to the PNIM dynamic data domain and inference data domain in real time through the change detection engine, driving updates to the 3D model rendering style. The output layer is used to render pipelines in three levels of colors—green, yellow, and red—based on risk percentiles, and to display flow arrows to indicate the risk transmission path when the impact of adjacent pipe segments is transmitted.

10. The system according to claim 1, characterized in that, The multi-source collaborative pipeline network information acquisition module simultaneously deploys electromagnetic induction equipment and ground penetrating radar equipment to collect real-time detection data of underground pipeline networks. It achieves trigger synchronization through the control unit and adopts an adaptive step size adjustment algorithm: a large step size is used for acquisition in straight pipeline sections, and the step size is automatically reduced in pipeline intersection or change-of-direction areas.

Images