Municipal drainage pipe network health degree diagnosis system

Abstract

The invention discloses a municipal drainage pipe network health degree diagnosis system. The system comprises a data acquisition module used for acquiring a pipeline defect identification result; the database construction module is used for establishing a drainage pipeline defect database based on the pipeline defect recognition result; the index system construction module is used for selecting drainage pipeline health assessment indexes from the drainage pipeline defect database and constructing a drainage pipeline health diagnosis index system; the fuzzy evaluation model building module is used for building a single drainage pipeline health degree fuzzy evaluation model based on the drainage pipeline health diagnosis index system, and obtaining a single drainage pipeline health degree evaluation result based on the single drainage pipeline health degree fuzzy evaluation model; and the comprehensive evaluation model construction module is used for integrating the health degree evaluation result of the single drainage pipeline with different management scales such as streets and communities, and comprehensively evaluating and determining the health degrees of drainage pipe networks with different scales. According to the invention, a pipeline detection technology of AI image recognition is combined, and multi-scale health degree diagnosis from a single pipeline to a regional pipe network is realized.

Description

Technical Field

[0001] The invention belongs to the field of drainage network health diagnosis, and in particular relates to a municipal drainage network health diagnosis system. Background Art

[0002] With the acceleration of urbanization, the layout of urban drainage pipe networks, as an important part of urban infrastructure, is becoming increasingly large, and the management work is becoming increasingly trivial and complicated. The drainage pipe network plays a vital role in the ecological environment of the city and the lives of residents, especially in urban flood control and drainage, and in protecting the lives and property of citizens. However, as the service life of the pipe network increases, problems such as aging, rupture, and foreign body intrusion of pipes are becoming increasingly prominent, resulting in a decrease in the drainage capacity of the pipe network, which may lead to urban flood disasters, road collapse and other problems, seriously threatening the lives and property safety of citizens.

[0003] At present, the traditional urban drainage network health detection method mainly relies on fixed-point detection at key nodes, collecting data through manual or simple mechanical equipment, and then summarizing and analyzing these data to obtain the overall health status of the network. This method has obvious limitations: the selection of detection points is often limited by cost and operability, resulting in large errors in the detection results and failing to fully reflect the actual condition of the network. In order to improve the accuracy and efficiency of detection, some more technologically advanced methods have begun to use mobile detection robots. These robots can perform detailed inspections inside the pipeline, collect a large amount of data, and diagnose the health status of the pipeline. Although this method has made significant progress in data collection, it is difficult to accurately diagnose the comprehensive health of the network due to the variety of pipeline defects and damage types, and the presentation of detection results is often not intuitive enough. Summary of the invention

[0004] In view of the shortcomings of the existing technology, the present invention proposes a municipal drainage network health diagnosis system, which uses AI recognition technology to intelligently identify pipeline image video data, automatically identify and classify pipeline defects, and constructs a fuzzy evaluation model based on the FAHP-RS weight analysis method to achieve comprehensive health diagnosis from single pipe sections to streets, communities and even regional pipe networks. By improving detection accuracy, optimizing data presentation, and enhancing the level of automation, scientific decision-making support is provided for municipal managers. This system will effectively improve the management efficiency and accuracy of urban drainage networks, safeguard urban flood control and drainage capabilities, ensure the safety of citizens' lives and property, and promote the intelligent upgrade of urban infrastructure.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A municipal drainage network health diagnosis system comprises the following steps: a data acquisition module, a database construction module, an indicator system construction module, a fuzzy evaluation model construction module and a comprehensive evaluation model construction module;

[0007] The data acquisition module is used to obtain pipeline defect identification results;

[0008] The database construction module is used to establish a drainage pipe defect database based on the pipe defect identification result;

[0009] The indicator system construction module is used to select drainage pipeline health assessment indicators from the drainage pipeline defect database to construct a drainage pipeline health diagnosis indicator system;

[0010] The fuzzy evaluation model building module is used to establish a fuzzy evaluation model for the health of a single drainage pipeline based on the drainage pipeline health diagnosis index system, and obtain a single drainage pipeline health evaluation result based on the fuzzy evaluation model for the health of the single drainage pipeline;

[0011] The comprehensive evaluation model building module is used to integrate the health evaluation results of the single drainage pipe with data of different management scales such as streets and communities, so as to comprehensively evaluate the health of drainage pipe networks of different scales.

[0012] Preferably, the pipeline defect recognition result includes: defect category, confidence level and position coordinates in the image.

[0013] Preferably, the fuzzy evaluation model construction module includes: a stratification unit, a judgment unit, a first calculation unit, a second calculation unit, a third calculation unit, a combination unit and an evaluation unit;

[0014] The hierarchical unit is used to establish a hierarchical structure based on the drainage pipeline health diagnosis indicator system;

[0015] The judgment unit is used to construct a pairwise comparison judgment matrix, i.e., a fuzzy complementary judgment matrix, based on the hierarchical structure;

[0016] The first calculation unit is used to construct a fuzzy consistency judgment matrix based on the fuzzy complementary judgment matrix, and calculate the single-layer weight based on the fuzzy consistency judgment matrix;

[0017] The second calculation unit is used to calculate the subjective weight of each scheme for the target layer based on the single-layer weight;

[0018] The third calculation unit is used to determine the objective weight based on the rough set theory;

[0019] The combination unit is used to determine the pipeline health evaluation index FAHP-RS combination weight based on the subjective weight of each scheme for the target layer and the objective weight;

[0020] The evaluation unit is used to establish a fuzzy evaluation model for the health of a single drainage pipeline based on the combined weight of the pipeline health evaluation index FAHP-RS to obtain a single drainage pipeline health evaluation result.

[0021] Preferably, the fuzzy complementary judgment matrix is:

[0022] A=(a ij ) n×n ,

[0023] Among them, a ij +a ji =1,a ii =0.5,0 ij <1,i,j=1,2,...,n.

[0024] Preferably, based on the fuzzy complementary judgment matrix, the process of constructing a fuzzy consistency judgment matrix includes:

[0025]

[0026] Among them, i, j, k are 1, 2, ..., n; n is the order of the judgment matrix.

[0027] Preferably, based on the single-layer weights, the process of calculating the subjective weights of each scheme for the target layer includes:

[0028] Select a row of the fuzzy consistency judgment matrix as the benchmark, and preset the weight vector as w = (w 1 ,w 2 ,...,w n );

[0029] For the i-th factor, according to the i-th row of the B matrix, we establish the equation Among them, b ij is obtained from the fuzzy consistency judgment matrix, w j is the weight of the jth factor.

[0030] Preferably, based on the pipeline health evaluation index FAHP-RS combined weight, the process of establishing an optimization decision model for the evaluation index combined weight includes:

[0031]

[0032] Among them, n is the number of evaluation indicators, W Fi is the subjective weight of the evaluation index, W Ri is the objective weight of the evaluation index, W i is the combined weight of the evaluation index, θ i is the empirical factor, 0<θ i ​<1, reflecting the decision maker's preference for subjective experience and objective data.

[0033] Preferably, the comprehensive evaluation model construction module includes: a summary unit, an extraction unit, a construction unit and a determination unit;

[0034] The summarizing unit is used to obtain the street-scale drainage pipeline health evaluation results based on the single drainage pipeline health evaluation results according to each street;

[0035] The extraction unit is used to extract key street variables that affect street-scale health scores based on street-scale drainage pipeline health evaluation results by using correlation analysis and stepwise regression;

[0036] The construction unit is used to construct a comprehensive model of street-scale health assessment based on the key street variables using a metric error model;

[0037] The determination unit is used to determine the health of drainage networks of different scales based on the street-scale health assessment comprehensive model.

[0038] Preferably, the key street variables include: 6 street variables: number of pipelines P q , Pipeline operating pressure P pr , Pipeline network density P de , Pipeline network coverage rate P cr , average pipe diameter P d 、Pipeline connection method P con , drainage capacity P capa , and 4 topographic variables, namely, altitude E LE , Slope S LO , Slope P OS and slope A SP .

[0039] Preferably, the process of constructing a comprehensive model of street-scale health assessment using a metric error model includes:

[0040]

[0041] Where: f(·) is an m-dimensional vector function, Y i is the p-dimensional vector truth value, y i is the vector Y i The estimated value of x i is the q-dimensional error-free variable observation, c is the k-dimensional vector parameter, e i is the model error, E(e i ) is the expected value of the model error, C ov (e i )=σ 2 Ψ is the covariance matrix of the error, where Ψ is e iThe error structure matrix, σ 2 is the estimation error.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] This invention realizes high-precision automatic identification of urban drainage network defects through the integrated application of deep learning algorithms, and provides municipal managers with a new, comprehensive and intuitive network health diagnosis tool through systematic analysis technology. After the implementation of the technology, government management departments can more accurately evaluate and manage the health of drainage networks, optimize resource allocation, and improve the efficiency of maintenance and repair. At the same time, it promotes the transformation of urban infrastructure management towards intelligence and automation, ensures the long-term stable operation of urban drainage systems, protects the lives and property of citizens, and lays a solid foundation for the construction of smart cities. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0045] Figure 1 A diagram of a drainage network health status indicator system according to an embodiment of the present invention;

[0046] Figure 2 A schematic diagram of a municipal drainage network health diagnosis system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0049] Embodiment 1

[0050] like Figure 2 As shown, the present invention provides a municipal drainage network health diagnosis system, comprising the following steps: a data acquisition module, a database construction module, an indicator system construction module, a fuzzy evaluation model construction module and a comprehensive evaluation model construction module;

[0051] The data acquisition module is used to obtain pipeline defect identification results;

[0052] The database construction module is used to establish a drainage pipeline defect database based on the pipeline defect identification result;

[0053] The indicator system construction module is used to select drainage pipe health assessment indicators from the drainage pipe defect database and construct a drainage pipe health diagnosis indicator system;

[0054] The fuzzy evaluation model building module is used to establish a fuzzy evaluation model for the health of a single drainage pipeline based on the drainage pipeline health diagnosis index system, and obtain a single drainage pipeline health evaluation result based on the fuzzy evaluation model for the health of the single drainage pipeline;

[0055] The comprehensive evaluation model construction module is used to integrate the health evaluation results of a single drainage pipe with different management scales such as streets and communities, so as to comprehensively assess and determine the health of drainage networks at different scales.

[0056] In this embodiment, in data collection and preprocessing, the pipeline defect recognition method based on edge AI technology proposed in the invention patent technology (CN118397362) that has been applied for and published is adopted, and the common defect feature data in the drainage pipeline is used to train the model for the characteristic application scenarios in the drainage pipeline detection work, so as to improve the efficiency and accuracy of the technology in the drainage pipeline defect recognition work. Finally, the defect recognition results are obtained, including the defect category, confidence and location coordinates in the image.

[0057] In this embodiment, a drainage pipe test result database is established: the test data obtained by the above technology is stored in a structured and highly scalable database. The database not only has the basic functions of raw data storage and query, but also supports data security, privacy protection, real-time update and multi-dimensional analysis. The design of the entire system ensures the accuracy, reliability and accessibility of the data, providing a solid foundation for pipe network maintenance, health diagnosis and decision support.

[0058] In this embodiment, the drainage pipeline health diagnosis index system is constructed: the diagnosis of the health of the drainage network is a comprehensive and complex process, involving careful consideration of the functional factors, structural factors and external environmental factors of the network. The indicators should cover pipeline aging, wear, sediment conditions, maintenance records, groundwater impact, rainwater and sewage confluence, and external factors such as climate change and land use. The selection of these indicators is based on a comprehensive consideration of the long-term stability and functionality of the network, ensuring the scientificity and practicality of the diagnostic results. The following is a diagram of the drainage network health status index system structure in the present invention (such as Figure 1 shown).

[0059] In this embodiment, a fuzzy evaluation model for the health of a single drainage pipe: In the weight analysis of the drainage network health evaluation index, the present invention adopts a weight assignment method that integrates subjective and objective factors, and establishes an optimization model for the combined weights of evaluation indicators based on optimization theory. This method combines an improved analytic hierarchy process, the fuzzy analytic hierarchy process (FAHP), which uses fuzzy mathematics to handle the uncertainty in subjective judgments, and rough set theory (RS), a mathematical tool that objectively determines the importance of indicators based on the intrinsic connection of data. This subjective and objective combined weight assignment method not only takes into account the decision maker's subjective cognition of the importance of the indicator, but also integrates the objective information of the indicator itself, thereby improving the accuracy of diagnosis while making up for the limitations of a single weighting method, providing a more comprehensive and reliable method for diagnosing the health of the drainage network.

[0060] Based on the health diagnosis of single pipeline and street scale, the health diagnosis results are divided into 6 levels, namely very healthy, healthy, sub-healthy, damaged, severely damaged, and failed.

[0061] Table 1 Drainage pipe health level and description

[0062]

[0063] FAHP-RS single drainage pipe health fuzzy evaluation model specific implementation process: FAHP-RS evaluation method is a comprehensive evaluation method that combines fuzzy analytic hierarchy process (FAHP) and rough set theory (RS). This method first determines the weight of each diagnostic indicator through fuzzy analytic hierarchy process, and then uses rough set theory to comprehensively evaluate the health status of a single pipeline. This method can effectively deal with the uncertainty and ambiguity in the diagnosis process and improve the accuracy and reliability of the diagnosis results.

[0064] The fuzzy evaluation model construction module includes: a stratification unit, a judgment unit, a first calculation unit, a second calculation unit, a third calculation unit, a combination unit, a decision unit and an evaluation unit;

[0065] (1) The hierarchical unit is used to establish a hierarchical structure based on the drainage pipeline health diagnosis index system; specifically, clarify the problem and establish a hierarchical structure of the system. According to the nature of the problem and the goal to be solved, the problem is decomposed into different components and arranged in the form of the highest layer (target layer), the middle layer (accurate measurement layer, which can be several layers), and the lowest layer (solution layer). This hierarchical structure is as follows: Figure 1 shown.

[0066] (2) The judgment unit is used to construct a pairwise comparison judgment matrix, i.e., a fuzzy complementary judgment matrix, based on a hierarchical structure; specifically, the importance of each factor at the same level is compared with that of a factor at the previous level to construct a comparison judgment matrix. This judgment matrix is ​​a fuzzy complementary judgment matrix, and its fuzzy scale and meaning are shown in Table 2:

[0067] Table 2 Fuzzy scale and meaning table

[0068]

[0069] The interval between each two levels can be quantified by 0.2, 0.4, 0.6, and 0.8. As shown in Table 2, the fuzzy complementary judgment matrix constructed according to the above fuzzy scale is: A = (a ij ) n×n , with the following properties:

[0070] Among them, a ij +a ji =1(1),a ii =0.5(2),0 ij <1(3),

[0071] In the formula, i, j are 1, 2,…, n; n is the matrix order.

[0072] In summary, the fuzzy judgment matrix of influencing factors is constructed, see Tables 2 to 5.

[0073] Evaluation system decision-making layer fuzzy judgment matrix:

[0074]

[0075] Structural Factor Fuzzy Judgment Matrix

[0076]

[0077]

[0078]

[0079] (3) The first calculation unit is used to construct a fuzzy consistency judgment matrix based on the fuzzy complementary judgment matrix, and calculate the single-layer weight (hierarchical single ranking) based on the fuzzy consistency judgment matrix; specifically, the fuzzy complementary judgment is converted into a fuzzy consistency judgment matrix, and then according to the relationship between the weights of the fuzzy consistency judgment matrix, an equation can be established to obtain the importance weight of each factor at the same level relative to a factor at the previous level, that is, hierarchical single ranking. The fuzzy complementary judgment matrix is ​​converted into a fuzzy consistency judgment matrix through formula 4.

[0080] ​

[0081] Among them, i, j, k are 1, 2, ..., n; n is the order of the judgment matrix.

[0082]

[0083] In order to determine whether the weight value calculated by formula 5 is reasonable, a consistency test is required in the comparative judgment process. Now, the compatibility index of the judgment matrix and its characteristic matrix is ​​obtained through matrix calculation, and its expressions are as follows:

[0084]

[0085]

[0086] A=(a ij ) n*n ,W*=(b ij ) n*n

[0087] Among them, A and W* are both fuzzy complementary judgment matrices. If the compatibility index value I(A, W*) is less than a specific threshold α (usually α = 0.1), the judgment matrix can be considered to be a satisfactory consistency matrix. The smaller α is, the higher the decision maker's consistency requirements for the fuzzy judgment matrix are.

[0088] Table 3 Calculation results of single-level weights in the decision layer

[0089]

[0090] Table 4 Calculation results of single-level weights of structural factors

[0091]

[0092] Table 5 Single-level weight calculation results of functional factors

[0093]

[0094] Table 6 Calculation results of single-level weights of environmental and social factors

[0095]

[0096]

[0097] (4) The second calculation unit is used to calculate the relative importance weight (subjective weight) of each solution layer element to the target layer element based on the single-layer weight, that is, the comprehensive weight (hierarchical total ranking) of the target layer elements; the composite weight of the indicator elements of the solution layer relative to the target layer is:

[0098]

[0099] Table 7 Comprehensive weight calculation results based on FAHP method

[0100]

[0101] (5) The third computing unit is used to determine the objective weight based on the rough set theory; the rough set is a set of attribute data that is good at data mining and processing complex information. The rough set theory can be expressed as a decision system S (U, A, V, f), where U = {u 1 ,u 1 ,…,u |U|}, represents the domain, u |U| is called an object; A = {a 1 , a 1 , …, a |A|} is represented as an attribute; now define that for each a∈A, there exists a mapping a:U→a(U) which is the value range of attribute a, and A=C∪D, S(U, A) is the decision table, C is the condition attribute, D is the decision attribute; V = U c＝A V C , V c is the value range of attribute C; f is U×A→V, which is an information function that assigns a value to the attribute of each object, expressed as:

[0102] The information system S is a drainage pipeline system. The 19 evaluation indicators of the scheme layer in the above fuzzy hierarchical analysis method are conditional attributes (evaluation indicators) C, C = {c1, c2, c3, ..., c18, c19}; decision attribute D = {d}, which refers to the number of pipeline failures, and the decision attribute value range V = {0, 1, 2, ...}. The original data of the information system is the statistical data of the drainage network health detection. Since the original data conditional attributes are complicated and difficult to reflect the commonality, it is not conducive to the attribute simplification of the rough set theory. Therefore, the original attributes are preprocessed, and the decision table can be established by combining the decision attribute assignment rules.

[0103] Table 8 Decision table excerpt

[0104]

[0105] Classify the domain U, calculate the degree of dependence of D on C (conditional information entropy), and the degree of dependence of the decision attribute D (U / D={D1,D2,…,Dk}) on the conditional attribute set C (U / C={C1,C2,…Cm}) is:

[0106]

[0107] In the formula, k is the number of sets D in the decision attributes; m is the number of sets C in the conditional attribute set; the symbol |*| is the number of elements in the set *.

[0108] The decision attribute D for each evaluation index C i The degree of dependence is:

[0109]

[0110] In the decision table S(U, A, V, f), the evaluation index C in the condition attribute set C i The attribute importance of is defined as:

[0111] Sig(c i )=I((D|C)-{c i})-I(D|C) (10)

[0112] Evaluation Index C i The weight is:

[0113]

[0114] The dependence of D on C in the sensory criteria is calculated using formula (8), and the solution is I(D|C) = 0.0013.

[0115] Calculate the degree of dependence of decision attribute D on condition attribute Ci (D|C-{c i}), attribute importance Sig(c i ) Finally, the evaluation index C is calculated i Weight index:

[0116] Table 9 Objective weight calculation results based on RS method

[0117] Evaluation indicators Weight index Weight Evaluation indicators Weight Pipe material 0.061 Maintenance and repair records 0.061 Groundwater level and water quality 0.043 Pipe diameter 0.063 Obstructions inside the pipeline 0.062 Geological and soil conditions 0.035 Pipe length 0.057 Mixed connection of rainwater and sewage 0.064 Construction status 0.041 Pipe wall thickness 0.059 Pipeline sediment condition 0.063 Surrounding land use type 0.045 Pipeline burial depth 0.032 Piping Invasive Tree Roots 0.071 Road usage 0.045 Pipeline Age 0.059 Pipeline inspection and monitoring results 0.012 Temperature and rainfall changes 0.058 Pipeline ageing and wear 0.069

[0118] (6) The combination unit is used to determine the pipeline based on the subjective weight and objective weight of each scheme on the target layer.

[0119] FAHP-RS combined weight of health evaluation index. Specifically, the subjective weight and objective weight of the evaluation index obtained by the above calculation are respectively recorded as W Fi , W Ri , W i Set as the combined weight to achieve the optimal combination of subjective weighting method and objective weighting method, and now establish an optimization decision-making model.

[0120]

[0121] Among them, n is the number of evaluation indicators, W Fi is the subjective weight of the evaluation index, W Ri is the objective weight of the evaluation index, Wi is the combined weight of the evaluation index, θ i is the empirical factor, 0<θ i<1, which reflects the preference degree of the decision maker for subjective experience and objective data. It is obtained that the optimal decision-making model has a unique solution in the feasible region, and the FAHP-RS combined weight is:

[0122] W i = θ i W Fi +(1 - θ i W Ri )

[0123] Table 10 Calculation Results of FAHP-RS Combined Weight

[0124] Evaluation indicators FAHP RS FAHP-RS Evaluation indicators FAHP RS FAHP-RS Pipe material 0.066 0.061 0.0640 Pipeline sediment condition 0.062 0.063 0.0624 Pipe diameter 0.054 0.063 0.0576 Piping Invasive Tree Roots 0.058 0.071 0.0632 Pipe length 0.048 0.057 0.0516 Pipeline inspection and monitoring results 0.05 0.012 0.0348 Pipe wall thickness 0.063 0.059 0.0614 Groundwater level and water quality 0.048 0.043 0.0460 Pipeline burial depth 0.042 0.032 0.0380 Geological and soil conditions 0.043 0.035 0.0398 Pipeline Age 0.071 0.059 0.0662 Construction status 0.040 0.041 0.0404 Pipeline ageing and wear 0.056 0.069 0.0612 Surrounding land use type 0.042 0.045 0.0426 Maintenance and repair records 0.055 0.061 0.0574 Road usage 0.033 0.045 0.0378 Obstructions inside the pipeline 0.058 0.062 0.0596 Temperature and rainfall changes 0.046 0.058 0.0514 Mixed connection of rainwater and sewage 0.065 0.064 0.0646

[0125] (7) The evaluation unit is used to establish a fuzzy evaluation model for the health degree of a single drainage pipeline based on the above FAHP-RS combined weight, and obtain the evaluation result of the health degree of the single drainage pipeline.

[0126] First, according to the health degree grade of the drainage pipeline in Table 1, the drainage pipeline is divided into six health degree grades: very healthy, healthy, sub-healthy, damaged, severely damaged, and failed. Then, for the 19 pipeline health degree evaluation indicators, an accurate fuzzy set membership function is established. For factors such as pipeline diameter and pipe material, which have discrete properties and cannot establish a continuous membership function, the method of direct assignment is used to determine their membership function.

[0127] For data such as pipeline burial depth, age, and groundwater level, which have an obvious correlation with the pipeline health degree, the calculation method of their membership degree is as follows:

[0128] ① When the actual value xi of the i-th index Ci is greater than the grade standard of "very healthy", its membership degree to the "very healthy" grade is 1, and its membership degrees to other health grades are 0. That is, when xi > V1, ri1 = 1; ri2 = 0; …; ri6 = 0.

[0129] ② When the actual value of the i-th index Ci is between the health degree standards Vj and Vj+1 of the j-th grade and the (j + 1)-th grade, that is, when Vj < xi < Vj+1, its membership degree to the (j + 1)-th grade of health degree is:

[0130]

[0131] Its membership degree to the j-th grade of health degree is:

[0132] r j = 1 - r j+1 , j = 0, 1, 2, 3, 4

[0133] And its membership degrees to other health grades are 0.

[0134] ③When the actual value xi of the i-th indicator Ci is less than its corresponding "failure" grade standard, the membership calculation is the same as ①② above and will not be repeated here.

[0135] According to the weight analysis of the above FAHP-RS method, the factor set U = {u1, u2, u3, ..., u18, u19} is determined, which includes the above 19 types of influencing factors.

[0136] Determine the drainage pipe evaluation set V, V = {v1, v2, v3, v4, v5, v6}, which are represented by six health levels: very healthy, healthy, sub-healthy, damaged, severely damaged and failed.

[0137] According to the above membership calculation results, the single factor evaluation matrix R is established:

[0138]

[0139] Combined with the weights of each evaluation index, a comprehensive evaluation is conducted:

[0140]

[0141] b k Indicates that the pipeline being evaluated has a good overall performance on V k The membership degree of the grade fuzzy subset. Finally, combined with the calculated vector B, the evaluated pipeline is comprehensively evaluated according to the maximum membership principle to determine the pipeline health level.

[0142] In summary, the health evaluation results of single drainage pipelines include a total of 13,459 pipe sections, of which 3,446 pipe sections are very healthy, 5,493 are healthy, 3,512 are sub-healthy, 685 are damaged, and 323 are severely damaged.

[0143] In this embodiment, the comprehensive evaluation model construction module includes: a summary unit, an extraction unit, a construction unit and a determination unit;

[0144] The summary unit is used to summarize the health evaluation results of the drainage pipelines at the street level based on the health evaluation results of the single drainage pipelines, that is, the average value (H m ), standard deviation (H sd ), coefficient of variation (H cv ), skewness (H pd ), kurtosis (H fd ). Average value (H m ) reflects the average health score of all pipelines in the street. The larger the value, the healthier the pipeline. The standard deviation (H sd) reflects the variation of the health scores of all pipelines in the street, and can reflect the differences in factors such as the material of each street pipeline and sediment blockage. The larger the value, the more complex the differences between street pipelines; the coefficient of variation (H cv ) is a comprehensive reflection of the variation of health scores of all pipelines in the street; skewness (H pd ) and kurtosis (H fd ) reflects the degree of deviation of the health score of each pipeline in the street from the normal distribution. If H pd >0, indicating that the data is obviously right-skewed normally distributed; if H fd >0, indicating that the kurtosis of the data distribution is relatively steep; if H fd <0, indicating that the kurtosis of the data distribution is relatively flat.

[0145] The extraction unit is used to extract the key street variables that affect the street-scale health score based on the health evaluation results of the street-scale drainage pipelines by using correlation analysis and stepwise regression. Specifically, in the screening process, the variance inflation factor critical value is set to 5, and the reference significance level is α = 0.05, that is, only the parameters with significant estimation results and I VIF <5 variables to reduce the multicollinearity between variables. Based on previous studies and data availability, the candidate variables include 6 street variables: number of pipelines (P q ), pipeline use pressure (P pr ), pipe network density (P de ), pipe network coverage rate (P cr ), average pipe diameter (P d ), pipeline connection method (P con ), drainage capacity (P capa ), and four topographic variables, namely, elevation (E LE ), slope (S LO ), slope position (P OS ) and slope aspect (A SP ).

[0146] The construction unit is used to construct a comprehensive model of street-scale health assessment based on the key street variables and using a metric error model;

[0147] The determination unit is used to determine the health of drainage pipe networks of different scales based on the comprehensive model of street-scale health assessment, as shown in Table 11.

[0148] Since the above variables are the summary data of the health evaluation results of a single pipeline, there are certain measurement errors and cumulative errors. At the same time, there is a certain correlation between the indicators. Therefore, it is not practical to estimate each variable separately without considering the relationship between them. Therefore, the present invention adopts the measurement error model to construct a comprehensive model of street-scale health evaluation.

[0149] Table 11 Street drainage network health level and description

[0150]

[0151] Among them, the key street variable screening method is:

[0152] The correlation analysis method extracts the key variables affecting street-scale health from the above five pipeline health statistical variables, six street variables, and four terrain variables, a total of 15 variables, thereby improving the explanatory power and prediction accuracy of the model.

[0153] This correlation analysis uses Pearson correlation analysis. The Pearson correlation coefficient method is a statistical method that accurately measures the closeness of the relationship between two variables. Its expression formula is as follows:

[0154]

[0155] In the formula, are the means of n test values ​​respectively. The correlation coefficient r ranges from -1 to 1, that is, |r|≤1. The closer |r| is to 1, the higher the linear correlation between x and y.

[0156] Stepwise regression analysis is a method in multiple regression, which is often used to establish the optimal or most appropriate regression model, so as to deepen the study of the dependence between variables. Its basic idea is to introduce new variables one by one, and consider whether to eliminate the selected variables each time a new variable is introduced, until no new variables are introduced.

[0157] In the stepwise regression process, correlation analysis helps to identify and deal with multicollinearity problems among variables. By setting the VIF threshold, variables that are highly correlated with other variables can be excluded, thereby avoiding instability and uncertainty in model estimation.

[0158] The F test is used to test whether the linear effect of one or more related variables in the regression model is significant:

[0159]

[0160] Among them, RSS 0 is the residual sum of squares, RSS, of the model containing only the intercept 1 is the residual sum of squares of the model after including the additional variables, p is the number of parameters in the model, and p 0 is the number of additional variables in the new model and n is the sample size.

[0161] Among them, the principles and steps of the measurement error model are:

[0162] Based on the summary data of the single pipeline health evaluation results, because there are certain measurement errors and cumulative errors, and there is a certain correlation between the various indicators, the measurement error model is selected to construct the street-scale drainage network health evaluation model:

[0163]

[0164] Where: f(·) is an m-dimensional vector function, Y i is the p-dimensional vector truth value, y i is the vector Y i The estimated value of x i is the q-dimensional error-free variable observation, c is the k-dimensional vector parameter, e i is the model error, E(e i ) is the expected value of the model error, C ov (e i )=σ 2 Ψ is the covariance matrix of the error, where Ψ is e i The error structure matrix, σ 2 is the estimation error.

[0165] Furthermore, the health status of the area is displayed visually.

[0166] In order to more intuitively display the spatial distribution of the health of the pipe network, the present invention also combines the urban drainage pipe network waterlogging survey data set and the digital elevation model (DEM) data to develop a visualization system. The system can not only display the geographical distribution of the health of the pipe network, but also has search and query functions. Users can query and visualize the diagnosis results of the health of the drainage pipe network by selecting different scales such as pipe sections, streets, and districts, so that users can quickly locate and understand the pipe network status in a specific area. The initial interface of the system provides a statistical overview of the health of the pipe network at the district level or administrative area level. After the user selects a specific district, the system further refines and displays the distribution of the health of the pipe network in each street in the district. By selecting a street, the user can go deep into the pipe section level and obtain detailed pipe section health information, including location, disease type, grade and quantity. This process realizes the layer-by-layer drilling of data, allowing users to continuously transition from the entire urban district to the specific status of a single pipe section. The GIS geographic information system integrated within the system combines the pipe network health data with the geographical location and presents it intuitively on the map through dynamic visualization technology. Users can also quickly locate and filter the pipe network areas of interest through interactive query tools, achieving flexible data query and in-depth analysis. In addition, the statistical reports and charts automatically generated by the system provide users with quantitative analysis results of the health of the pipe network, further enhancing the data support for decision-making. The visualization display function of the present invention not only improves the readability and ease of use of data, but also ensures the personalization and practicality of information through user customization and export functions.

[0167] The present invention provides an innovative multi-scale drainage network health diagnosis method. The core of this method is that it can flexibly perform comprehensive health diagnosis on networks from single pipe scale to street scale. By obtaining the weights of evaluation indicators based on FAHP-RS, and then establishing a fuzzy evaluation model to obtain the single pipe health result, and then using stepwise regression to measure the error to obtain the street-scale drainage network, the pipeline health evaluation results of different scales are obtained. The present invention can provide drainage network health diagnosis results of different scales according to different application requirements and diagnostic purposes.

[0168] The multi-scale drainage network health diagnosis method of the present invention not only improves the scientificity and practicality of the diagnosis, but also provides managers with a more flexible and efficient tool to adapt to changing diagnostic needs and challenges.

[0169] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A municipal drainage network health diagnosis system, characterized in that: The following steps are involved: Data collection module, database construction module, indicator system construction module, fuzzy evaluation model construction module and comprehensive evaluation model construction module; The data acquisition module is used to obtain pipeline defect identification results; The database construction module is used to establish a drainage pipe defect database based on the pipe defect identification result; The indicator system construction module is used to select drainage pipeline health assessment indicators from the drainage pipeline defect database to construct a drainage pipeline health diagnosis indicator system; The fuzzy evaluation model building module is used to establish a fuzzy evaluation model for the health of a single drainage pipeline based on the drainage pipeline health diagnosis index system, and obtain a single drainage pipeline health evaluation result based on the fuzzy evaluation model for the health of the single drainage pipeline; The comprehensive evaluation model building module is used to integrate the health evaluation results of the single drainage pipe with data of different management scales such as streets and communities, so as to comprehensively evaluate the health of drainage pipe networks of different scales.

2. The municipal drainage network health diagnosis system according to claim 1, characterized in that: The pipeline defect recognition result includes: defect category, confidence level and position coordinates in the image.

3. The municipal drainage network health diagnosis system according to claim 1, characterized in that: The fuzzy evaluation model construction module includes: a stratification unit, a judgment unit, a first calculation unit, a second calculation unit, a third calculation unit, a combination unit and an evaluation unit; The hierarchical unit is used to establish a hierarchical structure based on the drainage pipeline health diagnosis indicator system; The judgment unit is used to construct a pairwise comparison judgment matrix, i.e., a fuzzy complementary judgment matrix, based on the hierarchical structure; The first calculation unit is used to construct a fuzzy consistency judgment matrix based on the fuzzy complementary judgment matrix, and calculate the single-layer weight based on the fuzzy consistency judgment matrix; The second calculation unit is used to calculate the subjective weight of each scheme for the target layer based on the single-layer weight; The third calculation unit is used to determine the objective weight based on the rough set theory; The combination unit is used to determine the pipeline health evaluation index FAHP-RS combination weight based on the subjective weight of each scheme for the target layer and the objective weight; The evaluation unit is used to establish a fuzzy evaluation model for the health of a single drainage pipeline based on the combined weight of the pipeline health evaluation index FAHP-RS to obtain a single drainage pipeline health evaluation result.

4. The municipal drainage network health diagnosis system according to claim 3, characterized in that: The fuzzy complementary judgment matrix is: A=(a ij ) n×n , Among them, a ij +a ji = 1, a ii = 0.5, 0 < a ij <1, i, j = 1, 2, ..., n.

5. The municipal drainage network health diagnosis system according to claim 3, characterized in that: Based on the fuzzy complementary judgment matrix, the process of constructing the fuzzy consistency judgment matrix includes: Among them, i, j, k are 1, 2, ..., n; n is the order of the judgment matrix.

6. The municipal drainage network health diagnosis system according to claim 3, characterized in that: Based on the single-layer weights, the process of calculating the subjective weights of each solution to the target layer includes: Select a row of the fuzzy consistency judgment matrix as the benchmark, and preset the weight vector as w = (w1, w2, ..., w n ); For the i-th factor, according to the i-th row of the B matrix, we establish the equation Among them, b ij is obtained from the fuzzy consistency judgment matrix, w j is the weight of the jth factor.

7. The municipal drainage network health diagnosis system according to claim 3, characterized in that: Based on the FAHP-RS combined weight of the pipeline health evaluation index, the process of establishing an optimization decision model for the combined weight of the evaluation index includes: Among them, n is the number of evaluation indicators, W Fi is the subjective weight of the evaluation index, W Ri is the objective weight of the evaluation index, W i is the combined weight of the evaluation index, θ i is the empirical factor, 0<θ i <1, reflecting the decision maker's preference for subjective experience and objective data.

8. The municipal drainage network health diagnosis system according to claim 1, characterized in that: The comprehensive evaluation model construction module includes: a summary unit, an extraction unit, a construction unit and a determination unit; The summarizing unit is used to obtain the street-scale drainage pipeline health evaluation results based on the single drainage pipeline health evaluation results according to each street; The extraction unit is used to extract key street variables that affect street-scale health scores based on street-scale drainage pipeline health evaluation results by using correlation analysis and stepwise regression; The construction unit is used to construct a comprehensive model of street-scale health assessment based on the key street variables using a metric error model; The determination unit is used to determine the health of drainage networks of different scales based on the street-scale health assessment comprehensive model.

9. The municipal drainage network health diagnosis system according to claim 8, characterized in that: The key street variables include: 6 street variables: Pipeline quantity P q , Pipeline operating pressure P pr , Pipeline network density P de , Pipeline network coverage rate P cr , average pipe diameter P d 、Pipeline connection method P con , drainage capacity P capa , and 4 topographic variables, namely, altitude E LE , Slope S LO , Slope P OS and slope A SP .

10. The municipal drainage network health diagnosis system according to claim 8, characterized in that: The process of constructing a comprehensive model of street-scale health assessment using a metric error model includes: Where: f(·) is an m-dimensional vector function, Y i is the p-dimensional vector truth value, y i is the vector Y i The estimated value of x i is the q-dimensional error-free variable observation, c is the k-dimensional vector parameter, e i is the model error, E(e i ) is the expected value of the model error, C ov (e i )=σ 2 Ψ is the covariance matrix of the error, where Ψ is e i The error structure matrix, σ 2 is the estimation error.

Images