A Big Data-Based Method and System for Lifecycle Safety Monitoring of Urban Underground Pipelines

Abstract

This application discloses a method and system for lifecycle safety monitoring of urban underground pipelines based on big data. The method includes: acquiring multi-source data on the entire lifecycle of urban underground pipelines; performing fusion preprocessing on the multi-source data to construct a digital twin model of the pipelines; using a spatiotemporal correlation analysis algorithm to monitor the pipeline's operating status in real time and extract abnormal features; performing a graded assessment of the pipeline's safety status based on a pre-constructed deep learning risk assessment model to generate a risk level; combining historical fault data and real-time monitoring data to predict the remaining service life of the pipelines using a life prediction algorithm; and generating maintenance decision suggestions based on the risk level and remaining service life, thereby achieving full lifecycle safety monitoring of underground pipelines. By constructing a digital twin model through multi-source data fusion, dynamic and visual management of the entire lifecycle of underground pipelines is achieved. The combination of spatiotemporal correlation analysis and deep learning algorithms accurately identifies abnormal pipeline features, improving the accuracy of risk assessment.

Description

Technical Field

[0001] This application relates to the field of urban public facilities management technology, and in particular to a method and system for lifecycle safety monitoring of urban underground pipelines based on big data. Background Technology

[0002] In the process of urban modernization, underground pipelines serve as the lifeline of cities, playing a crucial supporting role in their normal operation. However, current urban underground pipeline management faces numerous challenges. Traditional methods of underground pipeline safety monitoring are mostly single-point and decentralized, making it difficult to comprehensively grasp the status changes of pipelines throughout their entire life cycle. During the planning and design phase, the lack of forward-looking analysis of future urban development needs and potential risks leads to irrational pipeline layouts, mutual interference between different pipelines, and insufficient reserved space. During construction, incomplete and non-standard data recording, and a disconnect between construction information and subsequent operation and maintenance, create hidden dangers for future safety management.

[0003] In terms of operational monitoring, monitoring methods are limited and lack precision, making it difficult to capture subtle changes in pipelines in real time. Most cities rely solely on a small number of sensors or manual inspections, failing to detect early signs of malfunctions promptly. Furthermore, data from different departments and pipeline types is stored independently, with inconsistent data formats and standards, creating data silos that severely hinder data fusion and analysis, making it impossible to assess the overall safety status of pipelines. Maintenance and repair also present significant problems. Due to a lack of accurate prediction of the remaining lifespan of pipelines, maintenance plans are often based on experience, leading to either over-maintenance resulting in resource waste or under-maintenance causing pipelines to operate with defects, triggering safety accidents. Simultaneously, given the complexity of underground pipeline systems, once a fault occurs, it is difficult to quickly locate the root cause, resulting in low repair efficiency and severely impacting urban operations. These problems urgently require an innovative, big data-based, full lifecycle safety monitoring method and system to address. Summary of the Invention

[0004] This application provides a method and system for lifecycle safety monitoring of urban underground pipelines based on big data, which solves the technical problems of isolated data and insufficient lifecycle safety assessment in traditional underground pipeline monitoring.

[0005] On the one hand, this application provides a method for lifecycle safety monitoring of urban underground pipelines based on big data, the method comprising the following steps:

[0006] Step S1: Obtain multi-source data on the entire lifecycle of urban underground pipelines;

[0007] Step S2: Perform fusion preprocessing on the multi-source data to construct a pipeline digital twin model; use a spatiotemporal correlation analysis algorithm to monitor the pipeline's operating status in real time and extract abnormal features;

[0008] Step S3: Based on the pre-built deep learning risk assessment model, the safety status of the pipeline is assessed in a graded manner to generate a risk level;

[0009] Step S4: Combine historical fault data with real-time monitoring data to predict the remaining service life of the pipeline using a life prediction algorithm; generate maintenance decision recommendations based on risk level and remaining service life to achieve full life cycle safety monitoring of underground pipelines.

[0010] In one implementation of this application, the multi-source data includes planning and design data, construction data, operation monitoring data, and maintenance and repair data.

[0011] In one implementation of this application, in step S2, the spatiotemporal correlation analysis algorithm specifically performs the following process:

[0012] The operational monitoring data is decomposed into time series components to extract trend, periodic, and abnormal fluctuation components. An association matrix is ​​constructed based on the pipeline spatial topology, and the spatiotemporal influence factors for each monitoring point are calculated. The spatial propagation range of abnormal characteristics is predicted using these spatiotemporal influence factors. The formula for calculating the spatiotemporal influence factors is as follows:

[0013] Among them, SIF ij W represents the spatiotemporal influence factor of monitoring point i on monitoring point j, used to quantify the degree of influence of monitoring point i on monitoring point j in the spatiotemporal dimension; ij Spatial distance weighting reflects the correlation weight between monitoring point i and monitoring point j caused by spatial distance. Different distances typically result in different weights, used to characterize the influence of spatial relationships; T ij W is the time delay coefficient, reflecting the time delay characteristics of the influence of monitoring point i on j. The time delay between different monitoring points can be reflected by this coefficient. ik The spatial distance weight between monitoring point i and its neighboring monitoring point k is T. ik It is the time delay coefficient between monitoring point i and its neighboring monitoring point k, where k is the index of the neighboring monitoring point, k = 1, 2, ..., n is the number of neighboring monitoring points, that is, the total number of monitoring points adjacent to monitoring point i, used to determine the range of the summation operation.

[0014] In one implementation of this application, the deep learning risk assessment model specifically performs the following process: preprocessed multi-source data is input into a convolutional layer for spatial feature extraction; time series features are processed through a long short-term memory network layer; key features are weighted using an attention mechanism to output a risk level probability distribution, and the risk level assessment formula is:

[0015]

[0016] In the formula, R is the risk level vector, X is the input data matrix, and Conv is the convolution operation. The function is a hyperbolic tangent; LSTM is a Long Short-Time Memory operation; W1, W2, and W3 are weight matrices; and b2 and b3 are bias vectors. This is the activation function.

[0017] In one implementation of this application, step S4 specifically performs the following process: fitting the pipeline performance degradation trajectory based on the Weibull degradation model and calculating the degradation rate; fusing real-time monitoring data using a Bayesian update method to update the remaining lifetime probability distribution; generating a lifetime prediction interval based on the updated distribution. The Weibull degradation model expression is:

[0018]

[0019] In the formula, L(t) represents the remaining lifetime at time t, and L0 represents the initial lifetime parameter. The degradation rate coefficient, For shape parameters.

[0020] In one implementation of this application, step S2, the multi-source data fusion preprocessing includes spatial coordinate unification and data quality assessment, as follows:

[0021] Transform planning and design data and construction data into a unified geographic coordinate system;

[0022] Establish a data quality assessment indicator system, including indicators for completeness, consistency, and timeliness; eliminate invalid data through quality scoring to generate a standardized dataset; the data quality scoring formula is:

[0023]

[0024] In the formula, Q is the data quality score, I is the completeness index, C is the consistency index, and T is the timeliness index. , , These are the weighting coefficients.

[0025] In one implementation of this application, the process of generating maintenance decision suggestions in step S4 specifically includes:

[0026] A multi-objective function is constructed with maintenance cost, safety benefits, and social impact as optimization objectives;

[0027] The Pareto optimal solution set is generated using a non-dominated sorting genetic algorithm;

[0028] An executable and maintainable solution is generated by adaptively modifying the optimal solution set using an expert knowledge base; the multi-objective function expression is as follows:

[0029] F(x) = {f1(x), f2(x), f3(x)}

[0030] In the formula, F(x) is the target vector, f1(x) is the maintenance cost function, f2(x) is the safety benefit function, f3(x) is the social impact function, and x is the maintenance scheme parameter vector.

[0031] In one implementation of this application, the method further includes:

[0032] When an anomaly is detected, the anomaly feature vector is extracted;

[0033] The possible regions of anomaly sources are inferred by using a spatiotemporal correlation algorithm;

[0034] Anomaly features are matched using a historical fault mode database to generate a source tracing probability matrix; the most likely location of the anomaly source is determined based on the source tracing probability matrix; the formula for calculating the source tracing probability is as follows:

[0035]

[0036] In the formula, P(s|a) is the probability that anomaly a originates from s, P(a|s) is the conditional probability that s generates a, P(s) is the prior probability that s occurs, and n is the number of anomaly sources.

[0037] In one implementation of this application, the method further includes:

[0038] A monitoring point coverage model is constructed based on pipeline risk level and importance index;

[0039] Utilize simulated annealing algorithm to optimize monitoring point locations and maximize risk coverage;

[0040] The coverage model is updated based on real-time monitoring data, and the layout of monitoring points is dynamically adjusted; the monitoring point coverage model is as follows:

[0041]

[0042] In the formula, C represents the total coverage, and w i Let cou be the weight of the i-th pipeline segment. i Let represent the coverage of the i-th pipeline segment, and n represent the number of pipeline segments.

[0043] On the other hand, this application also provides a big data-based urban underground pipeline lifecycle safety monitoring system, applied to the aforementioned big data-based urban underground pipeline lifecycle safety monitoring method. The system includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to complete the aforementioned big data-based urban underground pipeline lifecycle safety monitoring method.

[0044] The big data-based method and system for monitoring the lifecycle safety of urban underground pipelines provided in this application have the following beneficial effects:

[0045] (1) A digital twin model was constructed by fusion of multi-source data, which improved data consistency and realized dynamic visualization management of the entire life cycle of underground pipelines.

[0046] (2) By combining spatiotemporal correlation analysis with deep learning algorithms, the abnormal characteristics of pipelines can be accurately identified, thereby improving the real-time performance and accuracy of risk assessment.

[0047] (3) By using life prediction algorithms and multi-objective optimization to generate maintenance decisions, we can avoid over- or under-maintenance and reduce the accident rate and operation and maintenance costs. Attached Figure Description

[0048] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0049] Figure 1 Flowchart of a big data-based urban underground pipeline lifecycle safety monitoring method provided in this application embodiment;

[0050] Figure 2 A schematic diagram of a big data-based urban underground pipeline lifecycle safety monitoring system provided in this application embodiment. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0052] This application provides a method and system for lifecycle safety monitoring of urban underground pipelines based on big data. The technical solutions proposed in this application will be described in detail below with reference to the accompanying drawings.

[0053] Figure 1 A flowchart illustrating the big data-based urban underground pipeline lifecycle safety monitoring method provided in this application embodiment. Figure 1 As shown, the method mainly includes the following steps:

[0054] Step S1: Obtain multi-source data on the entire lifecycle of urban underground pipelines.

[0055] In this embodiment of the application, the multi-source data includes planning and design data, construction data, operation monitoring data, and maintenance and repair data.

[0056] Step S2: Perform fusion preprocessing on the multi-source data to construct a pipeline digital twin model; use a spatiotemporal correlation analysis algorithm to monitor the pipeline operation status in real time and extract abnormal features.

[0057] In this embodiment of the application, the spatiotemporal correlation analysis algorithm specifically performs the following process:

[0058] The operational monitoring data is decomposed into time series components to extract trend, periodic, and abnormal fluctuation components. An association matrix is ​​constructed based on the pipeline spatial topology, and the spatiotemporal influence factors for each monitoring point are calculated. The spatial propagation range of abnormal characteristics is predicted using these spatiotemporal influence factors. The formula for calculating the spatiotemporal influence factors is as follows:

[0059] Among them, SIF ij W represents the spatiotemporal influence factor of monitoring point i on monitoring point j, used to quantify the degree of influence of monitoring point i on monitoring point j in the spatiotemporal dimension; ij Spatial distance weighting reflects the correlation weight between monitoring point i and monitoring point j caused by spatial distance. Different distances typically result in different weights, used to characterize the influence of spatial relationships; T ij W is the time delay coefficient, reflecting the time delay characteristics of the influence of monitoring point i on j. The time delay between different monitoring points can be reflected by this coefficient. ik The spatial distance weight between monitoring point i and its neighboring monitoring point k is T. ik It is the time delay coefficient between monitoring point i and its neighboring monitoring point k, where k is the index of the neighboring monitoring point, k = 1, 2, ..., n is the number of neighboring monitoring points, that is, the total number of monitoring points adjacent to monitoring point i, used to determine the range of the summation operation.

[0060] In this embodiment of the application, the multi-source data fusion preprocessing includes spatial coordinate unification and data quality assessment, and the steps are as follows:

[0061] Transform planning and design data and construction data into a unified geographic coordinate system;

[0062] Establish a data quality assessment indicator system, including indicators for completeness, consistency, and timeliness; eliminate invalid data through quality scoring to generate a standardized dataset; the data quality scoring formula is:

[0063]

[0064] In the formula, Q is the data quality score, I is the completeness index, C is the consistency index, and T is the timeliness index. , , These are the weighting coefficients.

[0065] Step S3: Based on the pre-built deep learning risk assessment model, the pipeline safety status is graded and assessed to generate a risk level.

[0066] In this embodiment, the deep learning risk assessment model specifically performs the following process: preprocessed multi-source data is input into a convolutional layer for spatial feature extraction; time-series features are processed through a long short-term memory network layer; key features are weighted using an attention mechanism to output a risk level probability distribution, and the risk level assessment formula is:

[0067]

[0068] In the formula, R is the risk level vector, X is the input data matrix, and Conv is the convolution operation. The function is a hyperbolic tangent; LSTM is a Long Short-Time Memory operation; W1, W2, and W3 are weight matrices; and b2 and b3 are bias vectors. This is the activation function.

[0069] Step S4: Combine historical fault data with real-time monitoring data to predict the remaining service life of the pipeline using a life prediction algorithm; generate maintenance decision recommendations based on risk level and remaining service life to achieve full life cycle safety monitoring of underground pipelines.

[0070] In this embodiment, the pipeline performance degradation trajectory is fitted based on the Weibull degradation model to calculate the degradation rate; real-time monitoring data is fused using a Bayesian update method to update the remaining lifetime probability distribution; and a lifetime prediction interval is generated based on the updated distribution. The Weibull degradation model expression is as follows:

[0071]

[0072] In the formula, L(t) represents the remaining lifetime at time t, and L0 represents the initial lifetime parameter. The degradation rate coefficient, For shape parameters.

[0073] In this embodiment, the process of generating maintenance decision suggestions specifically includes: constructing a multi-objective function with maintenance cost, safety benefits, and social impact as optimization objectives; generating a Pareto optimal solution set using a non-dominated sorting genetic algorithm; and adaptively modifying the optimal solution set using an expert knowledge base to generate an executable maintenance plan. The multi-objective function expression is as follows:

[0074] F(x) = {f1(x), f2(x), f3(x)}

[0075] In the formula, F(x) is the target vector, f1(x) is the maintenance cost function, f2(x) is the safety benefit function, f3(x) is the social impact function, and x is the maintenance scheme parameter vector.

[0076] In this embodiment, when an anomaly is detected, an anomaly feature vector is extracted; the possible region of the anomaly source is inferred using a spatiotemporal correlation algorithm; the anomaly features are matched using a historical fault mode library to generate a source tracing probability matrix; and the most likely location of the anomaly source is determined based on the source tracing probability matrix. The formula for calculating the anomaly source tracing probability is as follows:

[0077]

[0078] In the formula, P(s|a) is the probability that anomaly a originates from s, P(a|s) is the conditional probability that s generates a, P(s) is the prior probability that s occurs, and n is the number of anomaly sources.

[0079] In this embodiment, a monitoring point coverage model is constructed based on pipeline risk level and importance index; simulated annealing algorithm is used to optimize monitoring point locations to maximize risk coverage; the coverage model is updated according to real-time monitoring data to dynamically adjust the monitoring point layout; wherein, the monitoring point coverage model is:

[0080]

[0081] In the formula, C represents the total coverage, and w i Let cou be the weight of the i-th pipeline segment. i Let represent the coverage of the i-th pipeline segment, and n represent the number of pipeline segments.

[0082] The above describes a method for monitoring the lifecycle safety of urban underground pipelines based on big data, as provided in the embodiments of this application. Based on the same inventive concept, embodiments of this application also provide a system for monitoring the lifecycle safety of urban underground pipelines based on big data. Figure 2 A schematic diagram of the composition of the urban underground pipeline lifecycle safety monitoring system based on big data provided in this application embodiment is shown below. Figure 2As shown, the system mainly includes: at least one processor 201; and a memory 202 communicatively connected to at least one processor 201; wherein, the memory 202 stores instructions that can be executed by at least one processor 201, and the instructions are executed by at least one processor 201 to enable at least one processor 201 to complete the aforementioned big data-based urban underground pipeline life cycle safety monitoring method.

[0083] The following section provides examples of the solutions provided in this application in specific application scenarios.

[0084] A 500-meter-long ductile iron main water supply pipe (buried 1.8 meters deep, DN400 diameter, 15 years of service) in a certain city is equipped with 10 pressure / vibration sensors (numbered P1-P10). The system monitors the operating status in real time and predicts maintenance strategies. The specific process is as follows:

[0085] 1. Multi-source data fusion preprocessing.

[0086]

[0087] Data quality score calculation: Completeness (I): Missing anti-corrosion layer thickness in planning and design data (I=0.8). Consistency (C): Construction coordinate system deviation 200 meters (C=0.7). Timeliness (T): Sensor data delay ≤2 seconds (T=0.95). Q=0.4×0.8+0.3×0.7+0.3×0.95=0.815

[0088] Result: Q>0.8, data is valid; coordinate system 1 is converted to CGCS2000.

[0089] 2. Spatiotemporal correlation analysis and anomaly detection.

[0090] Pressure mutation event: The pressure at monitoring point P5 suddenly drops by 0.3 MPa (time t0). The pressure at P4 drops by 0.2 MPa at t0+8 seconds, and the pressure at P6 drops by 0.1 MPa at t0+12 seconds.

[0091] Calculation of spatiotemporal influence factors:

[0092]

[0093] Conclusion: P5 contributes 31.8% to the propagation of P4. Based on topological inference, the anomaly source is located 30 meters upstream of P5.

[0094] 3. Deep learning risk assessment.

[0095] Input features: Spatial features (convolutional layer output): Corrosion patch area ratio of pipe segment is 12%; Temporal features (LSTM output): Pressure fluctuation frequency increases by 40%; Attention weights: Corrosion feature weight is 0.7, vibration feature weight is 0.3.

[0096] Risk level output:

[0097] 𝑅=Softmax([0.1,1.2,0.3])=[0.15,0.67,0.18] R =Softmax([0.1,1.2,0.3])=[0.15,0.67,0.18]

[0098] Risk level: Low = 0, Medium = 1, High = 2.

[0099] Result: Risk level "medium" (probability 67%), requires priority investigation.

[0100] 4. Remaining life prediction and maintenance decision-making.

[0101] Weibull degradation model parameters:

[0102] Initial life L0 = 40 years, λ = 0.008, β = 1.5 (typical values ​​for ductile iron).

[0103] Current service life t=15 years.

[0104] L (15) = 40 × exp(−0.008 × 151.5) = 11.3 years.

[0105] Lifetime after Bayesian update:

[0106] The real-time corrosion rate exceeded expectations by 20% → Corrected λ=0.0096.

[0107] L Update (15) = 40 × exp(−0.0096 × 151.5) = 9.1 years.

[0108] Forecast range: [7.2, 11.0] years (95% confidence level).

[0109] 5. Multi-objective maintenance decision optimization.

[0110] Optimize the objective function:

[0111]

[0112] Pareto optimal solution set (NSGA-II algorithm output):

[0113]

[0114] Experts revised: Although Option 2 is costly, it has low social impact and outstanding safety benefits → Recommended for implementation.

[0115] 6. Optimization of dynamic monitoring points.

[0116] Coverage model parameters:

[0117]

[0118] C=0.9×1.0+0.7×0.6+1.0×0.8=2.12.

[0119] After simulated annealing optimization: P11 is added at the midpoint of segment X2 → cou2 is increased to 0.9. The total coverage C=2.35 (↑10.8%), with 100% coverage of high-risk segments.

[0120] Results Summary: Anomaly Location Efficiency: From alarm to source location, the time taken is ≤3 minutes (traditional methods require 2 hours); Maintenance Cost Savings: Accurate lifespan prediction avoids over-maintenance and reduces annual costs.

[0121] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0122] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0123] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for lifecycle safety monitoring of urban underground pipelines based on big data, characterized in that, The method includes the following steps: Step S1: Obtain multi-source data on the entire lifecycle of urban underground pipelines; Step S2: Perform fusion preprocessing on the multi-source data to construct a pipeline digital twin model; use a spatiotemporal correlation analysis algorithm to monitor the pipeline's operating status in real time and extract abnormal features; wherein, the spatiotemporal correlation analysis algorithm specifically performs the following process: The operational monitoring data is decomposed into time series components to extract trend, periodic, and abnormal fluctuation components. An association matrix is ​​constructed based on the pipeline spatial topology, and the spatiotemporal influence factors for each monitoring point are calculated. The spatial propagation range of abnormal characteristics is predicted using these spatiotemporal influence factors. The formula for calculating the spatiotemporal influence factors is as follows: Among them, SIF ij W represents the spatiotemporal influence factor of monitoring point i on monitoring point j, used to quantify the degree of influence of monitoring point i on monitoring point j in the spatiotemporal dimension; ij Spatial distance weighting reflects the correlation weight between monitoring point i and monitoring point j caused by spatial distance. Different distances result in different weights, which are used to characterize the influence of spatial relationships; T ij W is the time delay coefficient, reflecting the time delay characteristics of the influence of monitoring point i on j. The time delay between different monitoring points can be reflected by this coefficient. ik The spatial distance weight between monitoring point i and its neighboring monitoring point k is T. ik It is the time delay coefficient between monitoring point i and its neighboring monitoring point k, where k is the index of the neighboring monitoring point, k = 1, 2, ..., n is the number of neighboring monitoring points, that is, the total number of monitoring points adjacent to monitoring point i, used to determine the range of the summation operation; Step S3: Based on the pre-built deep learning risk assessment model, the pipeline safety status is graded and assessed to generate a risk level; Step S4: Combine historical fault data with real-time monitoring data to predict the remaining service life of the pipeline using a life prediction algorithm; generate maintenance decision recommendations based on risk level and remaining service life to achieve full life cycle safety monitoring of underground pipelines.

2. The method for lifecycle safety monitoring of urban underground pipelines based on big data according to claim 1, characterized in that, The multi-source data includes planning and design data, construction data, operation monitoring data, and maintenance and repair data.

3. The method for lifecycle safety monitoring of urban underground pipelines based on big data according to claim 1, characterized in that, The deep learning risk assessment model specifically performs the following process: Preprocessed multi-source data is input into a convolutional layer for spatial feature extraction; time-series features are processed through a long short-term memory network layer; key features are weighted using an attention mechanism to output a risk level probability distribution. The risk level assessment formula is as follows: In the formula, R is the risk level vector, X is the input data matrix, and Conv is the convolution operation. The function is a hyperbolic tangent; LSTM is a Long Short-Time Memory operation; W1, W2, and W3 are weight matrices; and b2 and b3 are bias vectors. This is the activation function.

4. The method for lifecycle safety monitoring of urban underground pipelines based on big data according to claim 1, characterized in that, Step S4 specifically involves the following processes: fitting the pipeline performance degradation trajectory based on the Weibull degradation model and calculating the degradation rate; fusing real-time monitoring data using a Bayesian update method to update the remaining lifetime probability distribution; and generating a lifetime prediction interval based on the updated distribution. The Weibull degradation model expression is as follows: In the formula, L(t) represents the remaining lifetime at time t, and L0 represents the initial lifetime parameter. The degradation rate coefficient, For shape parameters.

5. The method for lifecycle safety monitoring of urban underground pipelines based on big data according to claim 1, characterized in that, In step S2, the multi-source data fusion preprocessing includes spatial coordinate unification and data quality assessment, and the steps are as follows: Transform planning and design data and construction data into a unified geographic coordinate system; Establish a data quality assessment index system, including indicators for completeness, consistency, and timeliness; eliminate invalid data through quality scoring to generate a standardized dataset; the data quality scoring formula is: In the formula, Q is the data quality score, I is the completeness index, C is the consistency index, and T is the timeliness index. , , These are the weighting coefficients.

6. The method for lifecycle safety monitoring of urban underground pipelines based on big data according to claim 1, characterized in that, In step S4, the process of generating maintenance decision suggestions is as follows: A multi-objective function is constructed with maintenance cost, safety benefits, and social impact as optimization objectives; The Pareto optimal solution set is generated using a non-dominated sorting genetic algorithm; An executable and maintainable solution is generated by adaptively modifying the optimal solution set using an expert knowledge base; the multi-objective function expression is as follows: F(x) = {f1(x), f2(x), f3(x)} In the formula, F(x) is the target vector, f1(x) is the maintenance cost function, f2(x) is the safety benefit function, f3(x) is the social impact function, and x is the maintenance scheme parameter vector.

7. The method for lifecycle safety monitoring of urban underground pipelines based on big data according to claim 1, characterized in that, The method further includes: When an anomaly is detected, the anomaly feature vector is extracted; The possible regions of anomaly sources are inferred by using a spatiotemporal correlation algorithm; By matching abnormal features using a historical fault mode library, a source tracing probability matrix is ​​generated; the most likely location of the abnormal source is determined based on the source tracing probability matrix.

8. The method for lifecycle safety monitoring of urban underground pipelines based on big data according to claim 1, characterized in that, The method further includes: A monitoring point coverage model is constructed based on pipeline risk level and importance index; Utilize simulated annealing algorithm to optimize monitoring point locations and maximize risk coverage; The coverage model is updated based on real-time monitoring data, and the layout of monitoring points is dynamically adjusted; the monitoring point coverage model is as follows: In the formula, C represents the total coverage, and w i Let cou be the weight of the i-th pipeline segment. i Let represent the coverage of the i-th pipeline segment, and n represent the number of pipeline segments.

9. A big data-based urban underground pipeline lifecycle safety monitoring system, applied to the big data-based urban underground pipeline lifecycle safety monitoring method described in any one of claims 1-8, characterized in that, The system includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the big data-based urban underground pipeline lifecycle safety monitoring method as described in any one of claims 1-8.

Images