A city flood control and waterlogging treatment engineering system failure risk diagnosis method and system

Abstract

The application relates to the field of data analysis, and discloses a city flood control and waterlogging treatment engineering system failure risk diagnosis method and system, which couples a city flood and waterlogging mechanism model with an XGBoost numerical simulation technology, constructs a city flood and waterlogging key information simulator, uses a Monte Carlo simulation technology to expand an envelope type scene sample library, obtains a city flood and waterlogging key information simulation sample set based on the mechanism model and the XGBoost model, comprehensively considers system failure reasons to construct a fault tree and carries out failure probability calculation based on a Bayesian network, calculates a prior probability of engineering failure, uses reverse reasoning and sensitivity analysis of the Bayesian network, and comprehensively considers flood and waterlogging influence to complete failure risk evaluation of key subsystems, so as to identify weak links of the city flood control and waterlogging treatment engineering system. The application can reasonably evaluate the overall performance of the city flood control and waterlogging treatment engineering system and the flood and waterlogging risk level of the subsystem, avoids interference of complex factors in actual application, and improves the accuracy and comprehensiveness of diagnosis.

Description

Technical Field

[0001] This invention relates to the field of data analysis, and in particular to a method and system for diagnosing the failure risk of urban flood control and drainage engineering systems. Background Technology

[0002] With the intensification of global climate change, extreme weather events are becoming more frequent and severe. Coupled with rapid urbanization, urban flooding is becoming increasingly serious. Urban flooding caused by extreme rainstorms has already, and will continue to, severely threaten people's normal lives and socio-economic development. Therefore, scientifically assessing urban flood risk levels, improving the Urban Flood Control and Drainage Engineering System (UFCES), and enhancing urban flood emergency response capabilities are fundamental strategies for addressing urban flooding disasters in various regions. However, urban flood control and drainage is a complex issue. The UFCES, composed of subsystems such as watershed storage systems (reservoirs), water conservancy and drainage systems (river and canal embankments, pumping stations), and urban drainage systems (municipal pipe networks, stormwater pumping stations), often faces the risk of partial or even complete failure. Therefore, identifying the failure mechanisms of UFCES and rationally assessing the failure risks of each subsystem, thereby providing targeted guidance for the upgrading, operation scheduling, and emergency response of UFCES, has significant theoretical and practical value.

[0003] There are two main reasons for the failure of UFCES: one is related to rainstorms and floods that exceed the engineering design capabilities of these subsystems; the other is related to the uncertainty of the physical and mechanical properties of the engineering structure itself. Current technologies mainly focus on failure risk analysis of individual subsystems in UFCES or on a comprehensive evaluation of the flood risks faced by UFCES. However, UFCES is an open mega-system involving numerous factors, and existing solutions are difficult to meet the complex actual situation and cannot provide more comprehensive diagnostic results that are more in line with real application scenarios.

[0004] Therefore, it is necessary to design a method for diagnosing the failure risks of urban flood control and drainage engineering systems in order to avoid interference from complex factors in practical applications and improve the accuracy and comprehensiveness of the diagnosis. Summary of the Invention

[0005] Based on this, the present invention proposes a method and system for diagnosing the failure risk of urban flood control and drainage engineering systems. By coupling an urban flood mechanism model with XGBoost numerical simulation technology, a simulator for key urban flood information is constructed. Monte Carlo simulation technology is used to expand the envelope-type scenario sample library. Based on the mechanism model and XGBoost model, a simulated sample set of key urban flood information is obtained. Then, considering the causes of system failure, a fault tree is constructed, and failure probability is calculated based on a Bayesian network. The prior probability of engineering failure is calculated. Using the backward reasoning and sensitivity analysis of Bayesian networks, and comprehensively considering the impact of flooding, the failure risk assessment of key subsystems is completed, thereby identifying the weak links in the urban flood control and drainage engineering system. This invention can reasonably assess the overall performance of the urban flood control and drainage engineering system and the flood risk level of its subsystems, avoiding interference from complex factors in practical applications and improving the accuracy and comprehensiveness of the diagnosis.

[0006] This invention proposes a method for diagnosing the failure risk of urban flood control and drainage engineering systems, comprising:

[0007] Obtain actual monitoring datasets and input them into the urban flood key information simulation model to obtain a simulation sample set of urban flood key information. The urban flood key information simulation model is based on Monte Carlo simulation and XGBoost model.

[0008] Construct a system fault tree and a system failure Bayesian network, wherein the system fault tree is based on risk factors of top event, intermediate event, and bottom event;

[0009] The failure probability is calculated based on the simulated sample set of key urban flood information and the Bayesian network of system failure to obtain the failure probability result of the urban flood control and drainage engineering system. The failure probability calculation is based on the failure probability calculation of sub-units and Bayesian network inference analysis.

[0010] In summary, based on the aforementioned method for diagnosing the failure risk of urban flood control and drainage engineering systems, this invention constructs a simulator for key urban flood information by coupling an urban flood mechanism model with XGBoost numerical simulation technology. Monte Carlo simulation technology is used to expand the envelope-based scenario sample library, and a simulated sample set of key urban flood information is obtained based on the mechanism model and XGBoost model. Then, considering the causes of system failure, a fault tree is constructed, and failure probability is calculated based on a Bayesian network. The prior probability of engineering failure is calculated, and the inverse reasoning and sensitivity analysis of the Bayesian network are used, along with comprehensive consideration of flood impact, to complete the failure risk assessment of key subsystems. This identifies the weak links in the urban flood control and drainage engineering system. This invention can reasonably assess the overall performance of the urban flood control and drainage engineering system and the flood risk level of its subsystems, avoiding interference from complex factors in practical applications and improving the accuracy and comprehensiveness of the diagnosis. Specifically, the invention involves acquiring actual monitoring datasets and inputting them into a simulation model of key urban flood information to obtain a simulation sample set of key urban flood information. This simulation model is based on Monte Carlo simulation and XGBoost. A system fault tree and a system failure Bayesian network are constructed. The system fault tree is based on risk factors of top, intermediate, and bottom events. Failure probability calculations are performed based on the simulation sample set of key urban flood information and the system failure Bayesian network to obtain the failure probability results of the urban flood control and drainage engineering system. This failure probability calculation is based on sub-unit failure probability calculation and Bayesian network inference analysis. This invention can reasonably assess the overall performance of the urban flood control and drainage engineering system and the flood risk level of its subsystems, avoiding interference from complex factors in practical applications and improving the accuracy and comprehensiveness of diagnosis.

[0011] Furthermore, the step of obtaining the actual monitoring dataset and inputting it into the urban flood key information simulation model to obtain the urban flood key information simulation sample set specifically includes:

[0012] Real-time data collection of relevant data from subsystems of the watershed flood storage system, water conservancy drainage system, and urban drainage system, and extraction of feature information to construct a real-world monitoring dataset;

[0013] The actual monitoring dataset is input into the urban flood key information simulation model to establish an envelope-scenario library measured sample set, which is established based on rainfall events-river network initial field-engineering scheduling;

[0014] The urban flood key information simulation model is based on Monte Carlo simulation and XGBoost model;

[0015] Based on the Monte Carlo simulation, an envelope-based scene library simulation sample set is generated;

[0016] Simulations were performed using the XGBoost model and an envelope-based scenario library simulation sample set to obtain a simulation sample set of key information on urban flooding. The specific algorithm of the XGBoost model is as follows:

[0017] ,

[0018] in, K Indicates the number of trees. It is the first i The predicted value for each sample, It is the first i The first sample k The regression equations corresponding to each regression tree.

[0019] Furthermore, the steps of constructing the system fault tree and the system failure Bayesian network specifically include:

[0020] A system fault tree is constructed, which is based on top event, intermediate event and bottom event risk factors. The top event is the failure of the urban flood control and drainage engineering system. The intermediate events include the failure of subsystems such as watershed storage system, water conservancy drainage system and urban drainage system, as well as the failure of sub-units in the subsystems. The bottom event risk factors are the engineering factors and natural factors of sub-unit failure.

[0021] Construct a system failure Bayesian network based on the system fault tree.

[0022] Furthermore, the specific algorithm of the Bayesian network for system failure is as follows:

[0023] ,

[0024] ,

[0025] in, pa(X i ) express X i The parent node, P(X i ) Represents a node X i The probability of occurrence express P(X i ) The probability of opposing events. X i and These represent events from different nodes. and These represent opposing events from different node events. n Indicates the number of nodes. iand j Indicates the ordinal number of different nodes.

[0026] Furthermore, the step of calculating the failure probability based on the simulated sample set of key urban flood information and the system failure Bayesian network to obtain the failure probability result of the urban flood control and drainage engineering system specifically includes:

[0027] Failure probability calculation is based on sub-unit failure probability calculation and Bayesian network inference analysis;

[0028] The sub-unit failure probability calculation includes the sub-unit engineering factor failure probability calculation and the sub-unit natural factor failure probability calculation. The specific algorithm for the sub-unit failure probability calculation is as follows:

[0029] ,

[0030] ,

[0031] in, This represents the failure probability of a sub-unit engineering factor. Represents the normal distribution function. Indicates the reliability index of the engineering structure. This represents the probability of failure due to natural factors in the subunit. This indicates the number of times each subunit experiences overflow, dike breach, or water accumulation exceeding the limit; N represents the number of simulation sample sets in the envelope-type scene library.

[0032] Then, conduct a failure risk assessment of the urban flood control and drainage engineering system to obtain the failure risk of sub-units;

[0033] The ultimate goal is to obtain the failure probability results of the urban flood control and drainage engineering system.

[0034] Furthermore, the steps of the Bayesian network inference analysis specifically include:

[0035] Bayesian network inference analysis is based on Bayesian network backward inference and Bayesian network sensitivity analysis;

[0036] The specific algorithm for Bayesian network reverse inference is as follows:

[0037] ,

[0038] in, Represents a node When =1, the node =1 posterior probability, Represents a node The parent node, Represents a node The probability of the event, iIndicates the ordinal number of the node;

[0039] The specific algorithm for Bayesian network sensitivity analysis is as follows:

[0040] ,

[0041] in, Indicates the sensitivity value. Represents the target node X The prior variance, It represents the posterior variance of the target node X after the state of a specific node Y is known.

[0042] Furthermore, the step of conducting a failure risk assessment of the urban flood control and drainage engineering system to obtain the failure risk of sub-units specifically includes:

[0043] The failure risk of the sub-unit is quantified, and the specific algorithm for quantifying the failure risk of the sub-unit is as follows:

[0044] ,

[0045] in, Indicates the failure risk value of the sub-unit. Indicates the probability of failure. Indicating the impact of flooding, Indicates a subunit.

[0046] This invention proposes a failure risk diagnosis system for urban flood control and drainage engineering systems, comprising:

[0047] The key information simulation module is used to acquire actual monitoring datasets and input them into the urban flood key information simulation model to obtain a simulation sample set of urban flood key information. The urban flood key information simulation model is based on Monte Carlo simulation and XGBoost model.

[0048] The construction module is used to construct a system fault tree and a system failure Bayesian network, wherein the system fault tree is based on the risk factors of the top event, intermediate events, and bottom events;

[0049] The probability calculation module is used to calculate the failure probability based on the simulated sample set of key urban flood information and the system failure Bayesian network, so as to obtain the failure probability result of the urban flood control and drainage engineering system. The failure probability calculation is based on the failure probability calculation of sub-units and Bayesian network inference analysis.

[0050] The present invention also provides a storage medium that stores one or more programs, which, when executed by a processor, implement the above-described method for diagnosing the failure risk of urban flood control and drainage engineering systems.

[0051] The present invention also provides a computer device, the computer device including a memory and a processor, wherein:

[0052] The memory is used to store computer programs;

[0053] When the processor executes the computer program stored in the memory, it implements the above-described method for diagnosing the failure risk of urban flood control and drainage engineering systems. Attached Figure Description

[0054] Figure 1 This is a flowchart of the method for diagnosing the failure risk of urban flood control and drainage engineering systems proposed in the first embodiment of the present invention;

[0055] Figure 2 This is a schematic diagram of the failure risk diagnosis system for urban flood control and drainage engineering system proposed in the second embodiment of the present invention;

[0056] Figure 3 This is a logical step diagram of the failure risk diagnosis method for urban flood control and drainage engineering system proposed in the first embodiment of the present invention;

[0057] Figure 4 This is an example analysis area map of the failure risk diagnosis method for urban flood control and drainage engineering systems proposed in the first embodiment of the present invention.

[0058] Figure 5 This is a fault tree diagram of the failure risk diagnosis method for urban flood control and drainage engineering system proposed in the first embodiment of the present invention.

[0059] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0060] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0061] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly on the other element or there may be an intervening element. When an element is considered to be "connected to" another element, it can be directly connected to the other element or there may be an intervening element. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this embodiment are for illustrative purposes only.

[0062] Unless otherwise defined, all technical and scientific terms used in this embodiment have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this embodiment is for the purpose of describing particular embodiments only and is not intended to limit the invention. The term "and / or" as used in this embodiment includes any and all combinations of one or more of the associated listed items.

[0063] Please see Figure 1 The diagram shows a flowchart of the failure risk diagnosis method for urban flood control and drainage engineering systems proposed in the first embodiment of the present invention. This method includes steps S01 to S03, wherein:

[0064] Step S01: Obtain the actual monitoring dataset and input it into the urban flood key information simulation model to obtain the urban flood key information simulation sample set;

[0065] It should be noted that, in this embodiment, the urban flood key information simulation model is based on Monte Carlo simulation and XGBoost model, and collects relevant data of the subsystems of the watershed flood storage system, water conservancy drainage system and urban drainage system in real time and extracts feature information to construct an actual monitoring dataset;

[0066] The actual monitoring dataset is input into the urban flood key information simulation model to establish an envelope-scenario library measured sample set, which is established based on rainfall events-river network initial field-engineering scheduling;

[0067] The urban flood key information simulation model is based on Monte Carlo simulation and XGBoost model;

[0068] Based on the Monte Carlo simulation, an envelope-based scene library simulation sample set is generated;

[0069] Simulations were performed using the XGBoost model and an envelope-based scenario library simulation sample set to obtain a simulation sample set of key information on urban flooding. The specific algorithm of the XGBoost model is as follows:

[0070] ,

[0071] in, K Indicates the number of trees. It is the first i The predicted value for each sample, It is the first i The first sample k The regression equations corresponding to each regression tree.

[0072] Step S02: Construct the system fault tree and the system failure Bayesian network;

[0073] It should be noted that in this embodiment, the system fault tree is constructed based on the risk factors of the top event, intermediate events, and bottom events. The system fault tree is based on the risk factors of the top event, intermediate events, and bottom events. The top event is the failure of the urban flood control and drainage engineering system. The intermediate events include the failure of subsystems such as the watershed storage system, the water conservancy drainage system, and the urban drainage system, as well as the failure of sub-units within the subsystems. The risk factors of the bottom events are the engineering factors and natural factors of the sub-unit failure.

[0074] The specific failure risk factors and their definitions in the system fault tree are shown in Table 1 below:

[0075] Table 1. UFCES Failure Risk Factors and Their Explanations

[0076]

[0077] Construct a system failure Bayesian network based on the system fault tree.

[0078] The specific algorithm of the Bayesian network for system failure is as follows:

[0079] ,

[0080] ,

[0081] in, pa(X i ) express X i The parent node, P(X i ) Represents a node X i The probability of occurrence express P(X i ) The probability of opposing events. X i and These represent events from different nodes. and These represent opposing events from different node events. n Indicates the number of nodes. i and j Indicates the ordinal number of different nodes.

[0082] Step S03: Calculate the failure probability based on the simulated sample set of key urban flood information and the Bayesian network of system failure to obtain the failure probability results of the urban flood control and drainage engineering system;

[0083] It should be noted that in this embodiment, the failure probability calculation is based on sub-unit failure probability calculation and Bayesian network inference analysis. The sub-unit failure probability calculation includes sub-unit engineering factor failure probability calculation and sub-unit natural factor failure probability calculation. The specific algorithm for the sub-unit failure probability calculation is as follows:

[0084] ,

[0085] ,

[0086] in, This represents the failure probability of a sub-unit engineering factor. Represents the normal distribution function. Indicates the reliability index of the engineering structure. This represents the probability of failure due to natural factors in the subunit. This indicates the number of times each subunit experiences overflow, dike breach, or water accumulation exceeding the limit; N represents the number of simulation sample sets in the envelope-type scene library.

[0087] Then, conduct a failure risk assessment of the urban flood control and drainage engineering system to obtain the failure risk of sub-units;

[0088] The ultimate goal is to obtain the failure probability results of the urban flood control and drainage engineering system.

[0089] Bayesian network inference analysis is based on Bayesian network backward inference and Bayesian network sensitivity analysis;

[0090] The specific algorithm for Bayesian network reverse inference is as follows:

[0091] ,

[0092] in, Represents a node When =1, the node =1 posterior probability, Represents a node The parent node, Represents a node The probability of the event, i Indicates the ordinal number of the node;

[0093] The specific algorithm for Bayesian network sensitivity analysis is as follows:

[0094] ,

[0095] in, Indicates the sensitivity value. Represents the target node X The prior variance, It represents the posterior variance of the target node X after the state of a specific node Y is known.

[0096] The failure risk of the sub-unit is quantified, and the specific algorithm for quantifying the failure risk of the sub-unit is as follows:

[0097] ,

[0098] in, Indicates the failure risk value of the sub-unit. Indicates the probability of failure. This refers to the impact of flooding, which includes the area affected by flooding and the losses caused by flooding. Indicates a subunit.

[0099] It should be noted that this embodiment analyzes the failure risk of UFCES in the Wusha River Basin using a case study. For the specific logical steps, please refer to [link / reference needed]. Figure 3 Nanchang City is located in northern Jiangxi Province, with the Gan River flowing from southwest to northeast, dividing it into the southern and northern urban areas. The Wusha River is the main water system in the northern urban area, with a drainage area of ​​263.3 km², an average elevation of 135 m, and a main channel length of approximately 40 km. The water level in its downstream section is closely related to that of the Gan River. When the Gan River's water level is high, it has a significant backwater effect on the Wusha River. The Xingfu Reservoir is located on the upper reaches of the Wusha River, controlling a drainage area of ​​30.2 km² above the dam site. The reservoir's normal storage level is 55 m, with design and check flood standards of 1% and 0.2%, respectively, corresponding to water levels of 56.65 m and 57.12 m. The total reservoir capacity is 2063 × 10⁴ m³, and the flood control capacity is 508 × 10⁴ m³. The Xingfu Reservoir Dam (XR), six typical levee sections (QD, MW, MLWL, MLW, JLWG, LQ), and six waterlogging points (XPG, NOPO, CIFA, XHC, WCCZ, CFDA) were selected as the subsystems to be studied. For specific example analysis areas, please refer to [link to relevant documentation]. Figure 4 .

[0100] The data used to construct the XGBoost model comes from an urban flood mechanism model. In this embodiment, a coupled model of one-dimensional pipe network, one-dimensional river channel, and two-dimensional topography in the Wusha River Basin is constructed based on the MIKE+ model. Since the flood control standard for the Wusha River Basin in Nanchang City is a 20-year return period, this study selects six rainfall return periods Ri (i=1,2,……,6), corresponding to 20, 50, 100, 200, 500, and 1000-year return periods (hereinafter referred to as 20a, 50a, 100a, 200a, 500a, and 1000a). The 24-hour rainfall corresponding to each return period is the rainfall characteristic variable in the rain-flood envelope scenario database of the Wusha River Basin in Nanchang City, as shown in Table 2 below:

[0101] Table 2 24-hour rainfall at each return period

[0102]

[0103] Historical rainstorm monitoring data from Nanchang station were clustered using the DTW-hierarchical clustering algorithm, extracting eight rainfall patterns Pj (j=1,2,……,8) and their corresponding probabilities. This embodiment uses these eight rainfall patterns as feature variables in the scene database. Furthermore, due to the significant backwater effect of the Ganjiang River water level on the Wusha River, based on the flood encounter analysis of internal and external rivers in the "Nanchang Changbei Urban Area Wusha River Water System Planning (Revised)", the design water level Lk (k=1,2,3,4) at the Wusha River estuary corresponding to four Ganjiang River flood frequencies were selected as feature variables for the downstream river level in the scene database, as shown in Table 3 below.

[0104] Table 3 Water level at the mouth of the Wusha River

[0105]

[0106] Taking into account six rainfall return periods (Ri), eight rainfall patterns (Pj), and four Wusha River estuary water levels (Lk), 192 samples (Sm) were combined to form a scenario library, serving as the input and boundary conditions for the urban flooding mechanism model. This simulated the urban flooding information sample library required for constructing the XGBoost urban flooding machine learning model. The sample library was divided into a training set and a validation set in a 7:3 ratio, thereby constructing an XGBoost model for predicting key urban flooding information. For calculating the UFCES failure probability in the Wusha River basin, this embodiment first used the Monte Carlo method to simulate and expand the sample, obtaining an envelope-type scenario library simulation sample set of 100,000 urban rainstorms. A roulette wheel selection method was used to ensure that the frequency of rainfall with the six return periods in the simulation sample matched the frequency of the original sample, and the frequency of the eight rainfall patterns matched the frequency of the original sample. Based on widely used distribution types, this embodiment selected the normal distribution, gamma distribution, exponential distribution, Weibull distribution, and log-normal distribution, and compared them with the theoretical probability density functions that correspond to the daily average water level time series characteristics of Nanchang Waizhou Station (1950–2013). The KS test showed that the distribution of the daily average water level at Waizhou Station was significantly similar to the log-normal distribution. Therefore, the water level was simulated according to the log-normal distribution, and the water level at the mouth of the Wusha River was calculated. Comparing the statistical parameters of the original and simulated water level samples, it was found that the mean and standard deviation of the original and simulated samples were not significantly different, with an error of 1.7% for the mean and 2.7% for the standard deviation. The results indicate that the Monte Carlo model not only ensures the accuracy of the sample but also expands the data range to a certain extent, improving the characteristic diversity of urban flood simulation. Based on a combined sample of rainfall amount and rainfall pattern, characteristic variables such as the maximum cumulative rainfall over 24 hours, 12 hours, 6 hours, 3 hours, 2 hours, and 1 hour for each rainfall sample were calculated. These, along with the water level at the Wusha River estuary, constituted the input conditions for the XGBoost model, thereby predicting key information about urban flooding, such as the key indicator HWLE for WCDS, and the key indicators MWD, TNFP, and MAFP for UDS. Additionally, the key indicator HWLR for BFSS was directly obtained based on flood regulation calculations, thus generating the calculation samples required for the failure probability of UFCES. The impact range of the UFCES subsystem in the Wusha River basin was determined through in-depth analysis of historical flood data, geographical and topographical data, and field surveys. Using GIS, the impact area of ​​each flood control subsystem within the Wusha River basin was determined for quantifying the failure risk of subunits, as shown in Table 4 below.

[0107] Table 4. Area affected by the failure of the flood control subunit

[0108]

[0109] This paper analyzes the failure paths of UFCES in the Wusha River Basin, summarizes its failure risk factors and their interpretations, and constructs a fault tree model for UFCES in the Wusha River Basin based on this. For details of the fault tree model implemented in this paper, please refer to [link to relevant documentation]. Figure 5 Node U1 indicates the failure of Xingfu Reservoir; nodes U2, U3, ..., U7 indicate the failure of six dike sections; nodes U8, U9, ..., U13 indicate the failure of the pipeline network at six waterlogging points; nodes X1, X2, ..., X13 indicate the failure of each flood control subunit due to engineering factors; and nodes Y1, Y2, ..., Y13 indicate the failure of each flood control subunit due to natural factors.

[0110] According to the "Standard", the reliability index of the Xingfu Reservoir and Wusha River embankment project is 3.2, and the failure probability caused by engineering factors is 0.0007, i.e., P(Xa)=0.0007 (a=1,2,...,7); the target reliability index of the waterlogging point pipeline network is 2.7, and the failure probability caused by engineering factors is 0.0035, i.e., P(Xa)=0.0035 (a=8,9,...,13). By conducting flood regulation calculations on Xingfu Reservoir, the probability of failure due to natural factors in the event of a rainstorm with a return period of more than 20 years is calculated to be 0.0001, i.e., P(Y1) = 0.0001. Based on the envelope-based scenario simulation sample library and the XGBoost model, key information of WCDS and UDS is obtained, and the probability of failure of the dike section and waterlogging point pipeline network (node ​​Ya (a=2,3,...,13)) due to natural factors in the event of a rainstorm with a return period of more than 20 years is calculated as shown in Table 5 below:

[0111] Table 5. Probability of Pipeline Failure Due to Natural Factors at Typical Cross-Sections and Waterlogging Points During Rainstorms with a Return Period of 1 in 20 Years or More

[0112]

[0113] Based on the failure probability calculation results of UFCES using Bayesian networks, the failure probability of all root nodes in the UFCES Bayesian network model of the Wusha River Basin was derived using reliability indices. Values ​​were assigned to the root nodes, and the failure probabilities of leaf nodes and intermediate nodes were finally determined using Netica software based on the prior probabilities of the root nodes and the conditional probability tables of non-root nodes. The results are shown in Table 6 below.

[0114] Table 6 Failure Probability of Leaf Nodes and Intermediate Nodes

[0115]

[0116] Table 6 shows that when encountering a rainstorm with a return period of 20 years or more, the failure probability of leaf node T is 0.545; the failure probabilities of nodes S1, S2, and S3 are 0.0008, 0.042, and 0.525, respectively. This indicates that when encountering a rainstorm with a return period of 20 years or more, the failure probability of UDS is relatively high, and the failure probability of most of its root nodes is also relatively high. Therefore, when formulating the UFCES renovation plan for the Wusha River Basin, the local government should pay attention to the upgrading and renovation of the drainage network system, appropriately improve the design standards, and enhance the city's flood drainage capacity. The failure risk diagnosis of UFCES in the Wusha River Basin, and the Bayesian network inverse reasoning results are shown in Table 7 below. The posterior probabilities of failure of the three flood control subsystems (nodes S1, S2, and S3) from high to low are: UDS (0.963), WCDS (0.077), and BFSS (0.0014). Further analysis of the second-level parent nodes (nodes U1, U2, ..., U13) of leaf node T reveals that the three nodes with the highest posterior probabilities are U8 "XPG" (1.0), U9 "NOPO" (0.985), and U10 "CIFA" (0.947). The posterior probability of failure for these three waterlogging points is significantly higher than that of other sub-units, indicating a significantly higher probability of urban flooding in these areas. Therefore, when developing urban flood control and drainage emergency plans, the focus should be on analyzing the causes of flooding and formulating targeted prevention and control measures. When flooding is forecast in the Wusha River basin, these three waterlogging points should be prioritized, and real-time scheduling and emergency response of the UFCES (Urban Flood Control and Disaster Reduction System) should be activated in a timely manner, with appropriate allocation of various disaster prevention and mitigation resources. Similarly, in the UFCES improvement plan for the Wusha River basin, it is necessary to prioritize improving the drainage capacity of these three waterlogging points.

[0117] Table 7 Posterior Probabilities of Each Node in a Bayesian Network

[0118]

[0119] Based on the Bayesian network sensitivity analysis, with leaf node T as the target node, the sensitivity of each flood control sub-unit of the UFCES in the Wusha River Basin is shown in Table 8. "CIFA" and "NOPO" are the two sub-units with the highest sensitivity in the UFCES of the Wusha River Basin, with sensitivity far exceeding that of other sub-units. This indicates that the failure of these two waterlogging points has a significant impact on the target variable; therefore, they can be considered as weak links in the UFCES of the Wusha River Basin.

[0120] Table 8. Overview of UFCES Sub-cell Sensitivity in the Wusha River Basin

[0121]

[0122] Based on the calculated failure probability and impact range, the failure risk of each sub-unit is calculated according to the failure risk assessment results, as shown in Table 9 below:

[0123] Table 9 Failure Risk of UFCES Subunit in Wusha River Basin

[0124]

[0125] The results showed that the failure risk of reservoirs and most dike sections in the Wusha River basin was much higher than that of waterlogging points, with dike sections “MW”, “MLWL”, and “JLWG” all having a high failure risk. These should also be considered important weak links in UFCES (Urban Flood Control and Resistant Water Shields); when formulating a UFCES renovation plan for the Wusha River basin, the crest elevation of the corresponding dike sections should be appropriately increased.

[0126] In summary, based on the aforementioned method for diagnosing the failure risk of urban flood control and drainage engineering systems, this invention constructs a simulator for key urban flood information by coupling an urban flood mechanism model with XGBoost numerical simulation technology. Monte Carlo simulation technology is used to expand the envelope-based scenario sample library, and a simulated sample set of key urban flood information is obtained based on the mechanism model and XGBoost model. Then, considering the causes of system failure, a fault tree is constructed, and failure probability is calculated based on a Bayesian network. The prior probability of engineering failure is calculated, and the inverse reasoning and sensitivity analysis of the Bayesian network are used, along with comprehensive consideration of flood impact, to complete the failure risk assessment of key subsystems. This identifies the weak links in the urban flood control and drainage engineering system. This invention can reasonably assess the overall performance of the urban flood control and drainage engineering system and the flood risk level of its subsystems, avoiding interference from complex factors in practical applications and improving the accuracy and comprehensiveness of the diagnosis. Specifically, the invention involves acquiring actual monitoring datasets and inputting them into a simulation model of key urban flood information to obtain a simulation sample set of key urban flood information. This simulation model is based on Monte Carlo simulation and XGBoost. A system fault tree and a system failure Bayesian network are constructed. The system fault tree is based on risk factors of top, intermediate, and bottom events. Failure probability calculations are performed based on the simulation sample set of key urban flood information and the system failure Bayesian network to obtain the failure probability results of the urban flood control and drainage engineering system. This failure probability calculation is based on sub-unit failure probability calculation and Bayesian network inference analysis. This invention can reasonably assess the overall performance of the urban flood control and drainage engineering system and the flood risk level of its subsystems, avoiding interference from complex factors in practical applications and improving the accuracy and comprehensiveness of diagnosis.

[0127] Please see Figure 2 The diagram shows a structural schematic of the urban flood control and drainage engineering system failure risk diagnosis system proposed in the second embodiment of the present invention. The system includes:

[0128] The key information simulation module 10 is used to acquire the actual monitoring dataset and input it into the urban flood key information simulation model to obtain the urban flood key information simulation sample set. The urban flood key information simulation model is based on Monte Carlo simulation and XGBoost model.

[0129] Module 20 is used to construct a system fault tree and a system failure Bayesian network, wherein the system fault tree is based on top event, intermediate event and bottom event risk factors;

[0130] The probability calculation module 30 is used to calculate the failure probability based on the simulated sample set of key urban flood information and the system failure Bayesian network, so as to obtain the failure probability result of the urban flood control and drainage engineering system. The failure probability calculation is based on the failure probability calculation of sub-units and Bayesian network inference analysis.

[0131] The present invention also proposes a computer storage medium storing one or more programs, which, when executed by a processor, implement the above-mentioned method for diagnosing the failure risk of urban flood control and drainage engineering systems.

[0132] The present invention also proposes a computer device, including a memory and a processor, wherein the memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory, so as to realize the above-mentioned method for diagnosing the failure risk of urban flood control and drainage engineering systems.

[0133] Those skilled in the art will understand that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain stored, communicated, propagated, or transmitted programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0134] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0135] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0136] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0137] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for diagnosing the failure risk of an urban flood control and drainage engineering system, characterized in that, include: Obtain actual monitoring datasets and input them into the urban flood key information simulation model to obtain a simulation sample set of urban flood key information. The urban flood key information simulation model is based on Monte Carlo simulation and XGBoost model. The urban flood key information simulation model obtains an envelope-type scenario library simulation sample set by simulating expanded samples using the Monte Carlo method, then predicts urban flood key information of water conservancy drainage system and urban drainage system based on XGBoost model, and obtains urban flood key information of watershed storage system based on flood regulation calculation, so as to generate the data sample required for failure probability calculation. Construct a system fault tree and a system failure Bayesian network, wherein the system fault tree is based on risk factors of top event, intermediate event, and bottom event; The failure probability is calculated based on the simulated sample set of key urban flood information and the Bayesian network of system failure, so as to obtain the failure probability result of the urban flood control and drainage engineering system. The failure probability calculation is based on the failure probability calculation of sub-units and Bayesian network inference analysis. The step of calculating the failure probability based on the simulated sample set of key urban flood information and the system failure Bayesian network to obtain the failure probability result of the urban flood control and drainage engineering system specifically includes: Failure probability calculation is based on sub-unit failure probability calculation and Bayesian network inference analysis; The sub-unit failure probability calculation includes the sub-unit engineering factor failure probability calculation and the sub-unit natural factor failure probability calculation. The specific algorithm for the sub-unit failure probability calculation is as follows: ， ， in, This represents the failure probability of a sub-unit engineering factor. Represents the normal distribution function. Indicates the reliability index of the engineering structure. This represents the probability of failure due to natural factors in the subunit. This indicates a natural failure event of the subsystem. This indicates the number of times each subunit experiences overflow, dike breach, or water accumulation exceeding the limit; N represents the number of simulation sample sets in the envelope-type scene library. Then, conduct a failure risk assessment of the urban flood control and drainage engineering system to obtain the failure risk of sub-units; To ultimately obtain the failure probability results of the urban flood control and drainage engineering system; The steps of the Bayesian network inference analysis specifically include: Bayesian network inference analysis is based on Bayesian network backward inference and Bayesian network sensitivity analysis; The specific algorithm for Bayesian network reverse inference is as follows: ， in, Represents a node When =1, the node =1 posterior probability, Represents a node The parent node, Represents a node The probability of the event, i Indicates the ordinal number of the node; The specific algorithm for Bayesian network sensitivity analysis is as follows: ， in, Indicates the sensitivity value. Represents the target node X The prior variance, It represents the posterior variance of the target node X given the state of a specific node Y; The step of conducting a failure risk assessment of the urban flood control and drainage engineering system and obtaining the failure risk of sub-units specifically includes: The failure risk of the sub-unit is quantified, and the specific algorithm for quantifying the failure risk of the sub-unit is as follows: ， in, Indicates the failure risk value of the sub-unit. Indicates the probability of failure. Indicating the impact of flooding, Indicates a subunit.

2. The method for diagnosing the failure risk of urban flood control and drainage engineering systems according to claim 1, characterized in that, The steps of obtaining the actual monitoring dataset and inputting it into the urban flood key information simulation model to obtain the urban flood key information simulation sample set specifically include: Real-time data collection of relevant data from subsystems of the watershed flood storage system, water conservancy drainage system, and urban drainage system, and extraction of feature information to construct a real-world monitoring dataset; The actual monitoring dataset is input into the urban flood key information simulation model to establish an envelope-scenario library measured sample set, which is established based on rainfall events-river network initial field-engineering scheduling; The urban flood key information simulation model is based on Monte Carlo simulation and XGBoost model; Based on the Monte Carlo simulation, an envelope-based scene library simulation sample set is generated; Simulations were performed using the XGBoost model and an envelope-based scenario library simulation sample set to obtain a simulation sample set of key information on urban flooding. The specific algorithm of the XGBoost model is as follows: ， in, K Indicates the number of trees. It is the first i The predicted value for each sample, It is the first i The first sample k The regression equations corresponding to each regression tree.

3. The method for diagnosing the failure risk of urban flood control and drainage engineering systems according to claim 1, characterized in that, The steps for constructing the system fault tree and the system failure Bayesian network specifically include: A system fault tree is constructed, which is based on top event, intermediate event and bottom event risk factors. The top event is the failure of the urban flood control and drainage engineering system. The intermediate events include the failure of subsystems such as watershed storage system, water conservancy drainage system and urban drainage system, as well as the failure of sub-units in the subsystems. The bottom event risk factors are the engineering factors and natural factors of sub-unit failure. Construct a system failure Bayesian network based on the system fault tree.

4. The method for diagnosing the failure risk of urban flood control and drainage engineering systems according to claim 3, characterized in that, The specific algorithm of the Bayesian network for system failure is as follows: ， ， in, pa(X i ) express X i The parent node, P(X i ) Represents a node X i The probability of occurrence express The probability of opposing events. X i and These represent events from different nodes. Represents the opposite event of a node event. n Indicates the number of nodes. i and j Indicates the ordinal number of different nodes.

5. A failure risk diagnosis system for urban flood control and drainage engineering systems, characterized in that, include: The key information simulation module is used to acquire actual monitoring datasets and input them into the urban flood key information simulation model to obtain a simulation sample set of urban flood key information. The urban flood key information simulation model is based on Monte Carlo simulation and XGBoost model. The urban flood key information simulation model obtains an envelope-type scenario library simulation sample set by simulating expanded samples using the Monte Carlo method, then predicts urban flood key information of water conservancy drainage system and urban drainage system based on XGBoost model, and obtains urban flood key information of watershed storage system based on flood regulation calculation, so as to generate the data sample required for failure probability calculation. The construction module is used to construct a system fault tree and a system failure Bayesian network, wherein the system fault tree is based on the risk factors of the top event, intermediate events, and bottom events; The probability calculation module is used to calculate the failure probability based on the simulated sample set of key urban flood information and the system failure Bayesian network, so as to obtain the failure probability result of the urban flood control and drainage engineering system. The failure probability calculation is based on the failure probability calculation of sub-units and Bayesian network inference analysis. The step of calculating the failure probability based on the simulated sample set of key urban flood information and the system failure Bayesian network to obtain the failure probability result of the urban flood control and drainage engineering system specifically includes: Failure probability calculation is based on sub-unit failure probability calculation and Bayesian network inference analysis; The sub-unit failure probability calculation includes the sub-unit engineering factor failure probability calculation and the sub-unit natural factor failure probability calculation. The specific algorithm for the sub-unit failure probability calculation is as follows: ， ， in, This represents the failure probability of a sub-unit engineering factor. Represents the normal distribution function. Indicates the reliability index of the engineering structure. This represents the probability of failure due to natural factors in the subunit. This indicates a natural failure event of the subsystem. This indicates the number of times each subunit experiences overflow, dike breach, or water accumulation exceeding the limit; N represents the number of simulation sample sets in the envelope-type scene library. Then, conduct a failure risk assessment of the urban flood control and drainage engineering system to obtain the failure risk of sub-units; To ultimately obtain the failure probability results of the urban flood control and drainage engineering system; The steps of the Bayesian network inference analysis specifically include: Bayesian network inference analysis is based on Bayesian network backward inference and Bayesian network sensitivity analysis; The specific algorithm for Bayesian network reverse inference is as follows: ， in, Represents a node When =1, the node =1 posterior probability, Represents a node The parent node, Represents a node The probability of the event, i Indicates the ordinal number of the node; The specific algorithm for Bayesian network sensitivity analysis is as follows: ， in, Indicates the sensitivity value. Represents the target node X The prior variance, It represents the posterior variance of the target node X given the state of a specific node Y; The step of conducting a failure risk assessment of the urban flood control and drainage engineering system and obtaining the failure risk of sub-units specifically includes: The failure risk of the sub-unit is quantified, and the specific algorithm for quantifying the failure risk of the sub-unit is as follows: ， in, Indicates the failure risk value of the sub-unit. Indicates the probability of failure. Indicating the impact of flooding, Indicates a subunit.

6. A storage medium, characterized in that, The storage medium stores one or more programs, which, when executed by a processor, implement the method for diagnosing the failure risk of urban flood control and drainage engineering systems as described in any one of claims 1-4.

7. A computer device, characterized in that, The computer device includes a memory and a processor, wherein: The memory is used to store computer programs; When the processor executes the computer program stored in the memory, it implements the method for diagnosing the failure risk of the urban flood control and drainage engineering system as described in any one of claims 1-4.

Images