An intelligent decision-making method and system for urban underground drainage network maintenance

Abstract

The present invention is applicable to the technical field of pipe network maintenance, and relates to an intelligent decision-making method and system for urban underground drainage pipe network maintenance, including the following steps: three-dimensional instantaneous hydraulic model analysis of drainage pipe network functional diseases; finite element fitting analysis and full-scale test are used to calibrate parameters and verify the accuracy of the three-dimensional instantaneous hydraulic model of drainage pipe network functional diseases; combining node water level iteration method, Preissmann slit method, Godunov finite volume method and unstructured grid, reconstructing the surface one-two-dimensional underground coupling connection model; using R language, dynamic library link technology and long short-term memory neural network method of multi-source data samples to realize the engineering secondary development of the one-two-dimensional coupling connection model, and obtain the functional disease situation of urban drainage pipe network with waterlogging result labels; establishing a multi-objective planning intelligent decision-making model for drainage pipe network maintenance and its solution method. The present invention can meet the intelligent, precise and scientific management and decision-making needs of urban underground drainage pipe networks.

Description

Technical Field

[0001] The present invention belongs to the technical field of pipe network maintenance, and in particular relates to an intelligent decision-making method and system for maintaining an urban underground drainage pipe network. Background Art

[0002] In the existing technology, the research on the relationship between functional diseases of drainage pipes and waterlogging in my country did not consider the multi-phase flow feedback relationship between pipe functional diseases, water flow, and upper air, and the influence of multi-field coupling such as gravity field, stress field, and flow field on the flow characteristics of pipes. The "solid-liquid-gas" multi-phase flow and multi-field coupling changed the instantaneous state of pipe confluence and the parameter values ​​of the one-dimensional model of surface and groundwater flow. In addition, the workload of analyzing and calculating the corresponding relationship between pipe functional diseases and waterlogging areas one by one by the control variable method is large and difficult to achieve. The above reasons lead to the low accuracy of the existing urban waterlogging early warning and disaster prevention decision-making system. The patent with the publication number CN110298076A provides an intelligent modeling and analysis method for urban waterlogging based on GIS and SWMM, including building a simulation system based on GIS and SWMM; automatically preprocessing the model data and automatically identifying topological errors; combining rainfall, calculating the hydrological and hydrodynamic coupling model based on SWMM, calculating the amount of rainwater that flows into the drainage network system, simulating the actual information in the drainage network, and obtaining the overflow data of the pipe point; analyzing the surface water inundation according to the overflow data of the pipe point, simulating the surface water accumulation to obtain the depth of the surface water accumulation, and distributing the surface water flow based on the window method; conducting early warning analysis on the surface waterlogging, comprehensively considering the surface water accumulation prediction, based on the calculation results of the hydrological and hydrodynamic coupling model, and automatically dividing the waterlogging risk level of different regions according to the results of the surface water inundation analysis, and providing an early warning plan for drainage and waterlogging emergency. This patent mainly focuses on the research of surface waterlogging and on this basis, performs calculation and prediction based on the hydrological and hydrodynamic coupling model. The factors considered are relatively single, and the accuracy of the results is difficult to guarantee.

[0003] Therefore, how to provide a method and system that can consider the impact of pipeline functional diseases on the hydrological and hydrodynamics of urban waterlogging under multiphase flow and multi-field coupling and can make intelligent decisions accurately and efficiently is an urgent problem to be solved by personnel in this technical field. Summary of the invention

[0004] In view of the shortcomings of the prior art, the purpose of the present invention is to provide an intelligent decision-making method for the maintenance of urban underground drainage pipe networks, so as to solve the problems in the prior art that the hydrological and hydrodynamic effects of pipeline functional diseases under multiphase flow and multi-field coupling on urban waterlogging are not considered, and the accuracy of the intelligent decisions made is not high; in addition, the present invention also provides an intelligent decision-making system for the maintenance of urban underground drainage pipe networks.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present invention provides an intelligent decision-making method for urban underground drainage network maintenance, comprising the following steps:

[0007] S10. Analysis of three-dimensional instantaneous hydraulic model of functional diseases of drainage network based on fluid dynamics and “mass-momentum-energy” conservation theory;

[0008] S20. Calibrate the parameters using finite element fitting analysis and full-scale tests, and verify the accuracy of the three-dimensional instantaneous hydraulic model of functional diseases of the drainage network;

[0009] S30, combining the node water level iteration method, Preissmann slit method, Godunov finite volume method and unstructured grid, reconstructs a two-dimensional coupled connection model of the surface and underground;

[0010] S40. Use R language, dynamic library linking technology and long short-term memory neural network method of multi-source data samples to realize the engineering secondary development of the surface and underground two-dimensional coupling connection model, and obtain the functional disease status of the urban drainage network with waterlogging result labels;

[0011] S50. Introduce deep small-world neural network, genetic algorithm and simulated annealing algorithm to establish a multi-objective planning intelligent decision-making model for drainage network maintenance and its solution method.

[0012] Furthermore, in step S10, the specific steps of three-dimensional instantaneous hydraulic model analysis of functional diseases of drainage pipe network are as follows:

[0013] S101. Based on the incompressibility of water flow and the principle of volume conservation, we can obtain the time-averaged motion equation and the pulsating motion equation.

[0014] S102. Combined with the "mass-momentum-energy" conservation equation The "solid-liquid-gas" spatial distribution model is constructed and the functional disease index of the drainage network is obtained. The u, v, and w are the components of the vertical average flow velocity of the river section on the x, y, and z axes of the three-dimensional coordinate system, respectively. is the real-time pulsating flow velocity of the turbulent flow in each coordinate axis direction, V is the volume of any given space in the pipeline, and Γ V is the boundary of the space domain. When α is 1, 2, and 3, it represents the solid, liquid, and gas phases respectively. α is the α phase velocity, the Q α is the source term of the α phase, the ρ α is the density of the α phase, and n is the boundary of the space domain Γ V The normal vector, is the source term of the other phase to the α phase, the σ αis the stress tensor, b α is the physical force, and S is the boundary of the space domain Γ V The cross-sectional area, c α is the specific heat capacity of the α phase, the T α is the temperature of the α phase, is the heat source of the α phase, The heat source for the α phase is generated by the phase transformation of other phases.

[0015] Furthermore, in step S20, the finite element fitting analysis refers to constructing a "solid-liquid-gas" multiphase flow simulation model of functional diseases of the drainage network using Abaqus software to verify the parameters calibrated by the full-scale test and the theoretical structural formula obtained by the three-dimensional instantaneous hydraulic model of the functional diseases of the drainage network; the full-scale test refers to building a landing well, a flushing gate and a functional disease pipe section, and calculating the accurate values ​​of the instantaneous state parameters of the pipeline confluence by the theoretical structural formula.

[0016] Furthermore, in step S40, the specific steps of engineering secondary development of the surface-underground two-dimensional coupled connection model are:

[0017] S401. Based on the stormwater analysis software InforWorks ICM, the surface-underground two-dimensional coupling connection model is re-implemented using dynamic library linking technology and R language secondary development technology and embedded into the InforWorks ICM to obtain an improved InforWorks ICM;

[0018] S402. The improved InforWorks ICM needs to be based on the existing InforWorks ICM, and simultaneously integrate the "solid-liquid-gas" multiphase flow movement law model of pipeline functional diseases, the section confluence state model, the section instantaneous flow velocity, the flow model and the surface and underground two-dimensional coupling connection model, and secondary development to achieve the improvement and upgrade of InforWorks ICM, so as to output the urban waterlogging loss situation under different pipeline functional disease conditions.

[0019] Furthermore, in step S50, the solution method of the multi-objective planning intelligent decision-making model for drainage network maintenance refers to using genetic algorithm and simulated annealing algorithm to perform intelligent optimization to obtain the functional disease conditions of the urban drainage network with maintenance decision labels; the initial input data of the deep small-world neural network is the functional disease conditions of the urban drainage network with maintenance decision labels, and after clustering, outlier detection and interpolation processing, the multi-layer restricted Boltzmann machine of the deep small-world neural network is repeatedly trained and adjusted through data set expansion to obtain the intelligent decision-making results of urban drainage network maintenance that are dynamically adjusted with time and rainstorms with different recurrence periods.

[0020] Furthermore, in step S30, the surface and underground two-dimensional coupled connection model is a one-dimensional pipe network full flow model and a two-dimensional surface water flow model coupled in the horizontal and vertical directions; the one-dimensional pipe network full flow model and the two-dimensional surface water flow model are analyzed by the weir flow formula method, the mutual boundary method and the fixed node water level method to obtain relevant structural formulas.

[0021] In a second aspect, the present invention also provides an intelligent decision-making system for urban underground drainage network maintenance, comprising:

[0022] Operate the subsystem for periodic detection of functional diseases of urban drainage networks;

[0023] Drainage network maintenance intelligent decision-making and integrated management database;

[0024] Drainage network maintenance plan management subsystem;

[0025] Intelligent decision-making and evaluation subsystem for drainage network maintenance.

[0026] Furthermore, the operational urban drainage network functional disease periodic detection subsystem includes an ultrasonic detection robot based on the Internet of Things / base station data transmission technology, a video surveillance detector, a pipeline detection robot and a magnetic flux detector, as well as a drainage network functional disease detection data collection workstation; the drainage network maintenance intelligent decision-making integrated management database provides upper application layer users with real-time data of heavy rains with different recurrence periods, network maintenance costs, real-time collection of maintenance information, and access and access control functions of geographic information data of functional diseases of the network; the drainage network maintenance plan management subsystem includes the functions of maintenance and operation execution plan, work order execution and return, and work order data update; the drainage network maintenance intelligent decision-making evaluation subsystem includes a three-dimensional instantaneous hydraulic analysis module for functional diseases of the network, a finite element full-scale test module, a two-dimensional coupled connection model of the surface and underground, an engineering implementation module of the rainwater model, and a multi-objective planning decision and solution module.

[0027] Furthermore, the system is divided into a data access control layer part, an access control service layer part and an application layer part in terms of system structure; the comprehensive management database server of the data access control layer part is respectively connected to the geographic information system server and the software standard data interface; the access control service layer part is carried by the geographic information system server and provides geographic information system services, business query and operation services, computing services and database access relay access control services for the application layer; the application layer part adopts a hybrid architecture model of C / S business system and B / S business system.

[0028] Furthermore, the hardware support platform of the system includes a comprehensive management database server, an urban pipe network geographic information system server, a drainage pipe network functional disease detection data collection workstation, a drainage pipe network maintenance plan execution agency workstation, a drainage pipe network maintenance plan real-time execution status collection workstation, an urban pipe network maintenance department monitoring and decision-making workstation, a regional pipe network maintenance department monitoring and decision-making workstation, and subordinate detection and maintenance instruments, meters, and sensors connected through the Internet of Things / base stations.

[0029] Compared with the prior art, the intelligent decision-making method and system for urban underground drainage network maintenance provided by the present invention have at least the following beneficial effects:

[0030] The present invention constructs a two-dimensional coupled connection model of the urban surface and underground that takes into account the evolution law of the instantaneous state of pipeline water flow under the action of multi-phase flow and multi-field coupling, and uses engineering technology to carry out secondary development and implementation of InforWorks ICM, so as to accurately reveal the hydrological and hydrodynamic coupling mechanism between the functional diseases of the drainage network and urban waterlogging; establishes an intelligent decision-making method for the maintenance of urban drainage network based on multi-objective decision-making and deep small-world neural network, and establishes an intelligent decision-making system for the maintenance of urban drainage network that can accurately and efficiently identify the corresponding relationship between the functional diseases of the network and the waterlogging losses and give reasonable decisions; constructs an intelligent decision-making system for the maintenance of urban drainage network that combines software and hardware, and realizes a maintenance decision-making platform that integrates functional disease data collection, parameter calibration, secondary development and intelligent optimization of the network. Based on a hybrid model combining C / S and B / S service modes, with geographic information system servers and integrated database servers as data processing platforms, and fluid dynamics, full-scale experiments, multi-objective decision-making and genetic / simulated annealing algorithms as analysis methods, an intelligent decision-making system for urban drainage network maintenance was designed and developed. It can not only meet the requirements of complex data processing rate of the system, but also ensure the accuracy of the output of maintenance decision results, and solve the problem of intelligent and precise preventive maintenance of urban flood-prone pipeline networks considering the coupling effects of pipeline functional diseases and waterlogging under multi-objective constraints. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the scheme of the present invention, a brief introduction is given below to the figures required for use in the description of the embodiments. Obviously, the figures described below are some embodiments of the present invention. For ordinary technicians in this field, other figures can be obtained based on these figures without paying any creative work.

[0032] Figure 1 A flow chart of an intelligent decision-making method for urban drainage network maintenance provided by an embodiment of the present invention;

[0033] Figure 2A general block diagram of an intelligent decision-making system for urban drainage network maintenance provided by an embodiment of the present invention;

[0034] Figure 3 A detailed composition diagram of the data access control layer part, access control service layer part and application layer part of an urban drainage network maintenance intelligent decision-making system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field of the present invention; the terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For example, the directions or positions indicated by the terms "length", "width", "up", "down", "left", "right", "front", "back", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc. are based on the directions or positions shown in the drawings, which are only for the convenience of description and should not be understood as limitations on the present technical solution.

[0036] The terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions; the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects rather than to describe a specific order. In the specification and claims of the present invention and the above-mentioned drawings, when an element is referred to as being "fixed to" or "mounted on" or "disposed on" or "connected to" another element, it may be directly or indirectly located on the other element. For example, when an element is referred to as being "connected to" another element, it may be directly or indirectly connected to the other element.

[0037] In addition, reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0038] The invention provides an intelligent decision-making method for the maintenance of an urban underground drainage network, which is applied to the research work on the relationship between functional diseases of drainage pipes and waterlogging. The intelligent decision-making method for the maintenance of an urban underground drainage network comprises the following steps: S10, a three-dimensional instantaneous hydraulic model analysis of functional diseases of the drainage network based on fluid dynamics and the "mass-momentum-energy" conservation theory; S20, calibrating parameters by using finite element fitting analysis and full-scale tests, and verifying the accuracy of the three-dimensional instantaneous hydraulic model of functional diseases of the drainage network; S30, reconstructing a two-dimensional coupled connection model of the surface and the underground by combining a node water level iteration method, a Preissmann slit method, a Godunov finite volume method and an unstructured grid; S40, implementing the engineering secondary development of a two-dimensional coupled connection model of the surface and the underground by using R language, dynamic library linking technology and a long short-term memory neural network method of multi-source data samples, and obtaining the functional disease situation of the urban drainage network with waterlogging result labels; S50, introducing a deep small-world neural network, a genetic algorithm and a simulated annealing algorithm, and establishing a multi-objective planning intelligent decision-making model for the maintenance of the drainage network and a solution method thereof.

[0039] The present invention can meet the needs of intelligent, precise and scientific management and decision-making of urban underground drainage pipe networks.

[0040] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings.

[0041] The present invention provides an intelligent decision-making method for urban underground drainage network maintenance, which is applied to the research on the relationship between functional diseases of drainage pipes and waterlogging. Figure 1 As shown, the intelligent decision-making method for urban underground drainage network maintenance includes the following steps:

[0042] S10. Analysis of three-dimensional instantaneous hydraulic model of functional diseases of drainage network based on fluid dynamics and “mass-momentum-energy” conservation theory;

[0043] S20. Use finite element fitting analysis and full-scale tests to calibrate parameters and verify the accuracy of the three-dimensional transient hydraulic model of functional diseases of the drainage network;

[0044] S30, combining the node water level iteration method, Preissmann slit method, Godunov finite volume method and unstructured grid, reconstructs a two-dimensional coupled connection model of the surface and underground;

[0045] S40. Use R language, dynamic library linking technology and long short-term memory neural network method of multi-source data samples to realize the engineering secondary development of the surface and underground two-dimensional coupling connection model, and obtain the functional disease status of the urban drainage network with waterlogging result labels;

[0046] S50. Introduce deep small-world neural network, genetic algorithm and simulated annealing algorithm to establish a multi-objective planning intelligent decision-making model for drainage network maintenance and its solution method.

[0047] Furthermore, in this embodiment, in step S10, the three-dimensional instantaneous hydraulic model analysis of functional diseases of the drainage network is theoretically derived from the perspective of multiphase flow and multi-field, and the liquid phase stress tensor and shear viscosity spatial distribution model, the solid phase stress tensor and shear viscosity spatial distribution model, and the gas phase Newtonian viscous stress spatial distribution model under the combined action of gravity field, pressure field, stress field, and flow field are constructed respectively.

[0048] Furthermore, in this embodiment, the steps of constructing the liquid phase stress tensor and shear viscosity spatial distribution model, the solid phase stress tensor and shear viscosity spatial distribution model, and the gas phase Newtonian viscous stress spatial distribution model include:

[0049] S101. Based on the incompressibility of water flow and the principle of volume conservation, the time-averaged motion equation and the pulsating motion equation are obtained.

[0050] S102. Combined with the "mass-momentum-energy" conservation equation A "solid-liquid-gas" spatial distribution model is constructed to obtain the functional disease index of the drainage network; u, v, and w are the components of the vertical average flow velocity of the river section on the x, y, and z axes of the three-dimensional coordinate system, respectively. is the real-time pulsating flow velocity of turbulence in each coordinate axis direction, V is the volume of any given space in the pipeline, Γ V is the boundary of the space domain. When α is 1, 2, and 3, it represents solid, liquid, and gas phases respectively. α is the α phase velocity, Q α is the source term of the α phase, ρ α is the density of the α phase, n is the boundary Γ V The normal vector of is the source term of the other relative α phase, σ α is the stress tensor, b α is the physical strength, S is the boundary Γ V The cross-sectional area, c α is the specific heat capacity of the α phase, T α is the temperature of the α phase, is the heat source of the α phase, It is the heat source for the α phase generated by the phase transformation of other phases.

[0051] Furthermore, in this embodiment, the three-dimensional instantaneous hydraulic model analysis of the functional diseases of the drainage network can obtain the pulsation period changes of the average flow velocity and instantaneous flow velocity between typical sections under the action of gravity flow and pressure flow, the turbulence intensity distribution, the energy dissipation law, the Reynolds stress distribution of the pipe body and other characteristics, and then accurately solve the critical starting conditions of the functional diseases of the drainage network, the theoretical structural formula of the displacement distance and flow capacity of the bed load, and extract and analyze the flow characteristics and turbulent flow field structure parameters related to the functional diseases of the pipeline.

[0052] Furthermore, in this embodiment, in step S20, the finite element fitting analysis refers to the use of Abaqus software to construct a "solid-liquid-gas" multiphase flow simulation model of functional diseases of the drainage network to verify the validity of the parameters calibrated by the full-scale test and the theoretical structural formula obtained by the three-dimensional transient hydraulic model. The full-scale test refers to the construction of a landing well, a scouring gate and a functional disease pipe section, and the accurate values ​​of the instantaneous state parameters of the pipeline confluence are calculated by the theoretical structural formula in the fluid dynamics, sensors and three-dimensional transient hydraulic model. The full-scale test consists of four parts: a pipeline functional disease scenario adjustment module, a confluence parameter spatiotemporal variation law analysis module, a bedload transport situation analysis module, and a pipeline velocity and flow time variation law analysis module.

[0053] Furthermore, in this embodiment, in step S30, the surface-underground coupling relationship model refers to the coupling connection between the horizontal direction (surface) and the vertical direction (surface-underground) of the one-dimensional pipe network full flow model and the two-dimensional surface water flow model; the one-dimensional pipe network full flow model and the two-dimensional surface water flow model can be analyzed through the weir flow formula method, the mutual boundary method and the fixed node water level method to obtain the relevant structural formula.

[0054] Furthermore, in the present embodiment, in step S40, the engineering secondary development of the surface-underground two-dimensional coupling connection model refers to re-implementing the surface-underground two-dimensional coupling connection model that considers the coupling effects of pipeline functional diseases and urban waterlogging based on the most mature existing stormwater analysis software InforWorks ICM by using dynamic library linking technology and R language secondary development technology, and embedding the module into InforWorks ICM to obtain an improved InforWorks ICM; the improved InforWorks ICM needs to integrate the "solid-liquid-gas" multiphase flow motion law model of the pipeline network functional diseases, the section confluence state model, the section instantaneous flow velocity, the flow model and the surface-underground two-dimensional coupling connection model on the basis of the existing InforWorks ICM, and the secondary development realizes the improvement and upgrade of InforWorks ICM, so as to output the urban waterlogging loss situation under different pipeline network functional diseases.

[0055] Furthermore, in this embodiment, in step S50, the solution method of the multi-objective planning intelligent decision-making model for drainage network maintenance refers to the use of genetic algorithm and simulated annealing algorithm for intelligent optimization to obtain the functional disease conditions of the urban drainage network with maintenance decision labels; the initial input data of the deep small-world neural network is the functional disease conditions of the urban drainage network with maintenance decision labels, and after clustering, outlier detection and interpolation processing, the multi-layer restricted Boltzmann machine of the deep small-world neural network is repeatedly trained and adjusted through data set expansion to obtain the intelligent decision-making results of urban drainage network maintenance that are dynamically adjusted with time and different recurrence periods of rainstorms.

[0056] Specifically, the deep small-world neural network (DSWNN) consists of two stages: offline learning and online decision-making. The specific steps include:

[0057] S501. With the help of the Internet of Things / base stations and the local area network of the online operation and maintenance system of the urban drainage network, obtain multi-dimensional indicator data such as the type, degree, and dredging / cleaning cost of functional diseases of each important node of the network, including normal collection data, abnormal or large-scale malicious tampering sample data, and use indicator screening, anomaly detection and interpolation to process the data;

[0058] S502, using a sliding window of a certain length to intercept all sample data, and obtain the mean subsequence under the sliding window; all subsequences are organized into a training sample set, and a status label is added to each training sample according to the InforWorks ICM numerical simulation results and the pipe network maintenance multi-objective planning results, mainly to classify the training sample nodes of the urban pipe network functional diseases that need maintenance;

[0059] S503, using an unlabeled data set to pre-train a multi-layer restricted Boltzmann machine (RBM) network of a deep small-world neural network (DSWNN), and after fully obtaining feature representation, recording network parameters, including weights and bias values ​​of neurons;

[0060] S504, randomly adding edges to the multi-layer restricted Boltzmann machine (RBM) network according to the random edge adding probability p of the deep small-world neural network (DSWNN), and adding a fully connected classification layer at the top layer to convert it into a deep small-world neural network (DSWNN) with multi-classification function; supervised learning is used to train the network again, and the back propagation (BP, Back Propagation) algorithm is used to fine-tune the parameters until the training is completed, so as to obtain a trained deep small-world neural network (DSWNN) classifier model;

[0061] S505, online collection of functional disease indicator sets of urban drainage pipe networks, and processing of online data using the same sliding window as that used in training to obtain samples to be tested;

[0062] S506. Input the samples to be tested into the trained deep small-world neural network (DSWNN) classifier model to obtain the intelligent decision-making results and optimal response strategies for drainage network maintenance, thereby completing the intelligent decision-making management optimization process for urban underground drainage network maintenance.

[0063] The intelligent decision-making method for maintenance of urban underground drainage pipe networks described in the above embodiments adopts advanced fluid dynamics, "mass-momentum-energy" conservation theory, finite element high-precision fitting and full-scale test method, node water level iteration method, Preissmann slit method, Godunov finite volume method and unstructured grid, R language, dynamic library linking technology and long short-term memory neural network method of multi-source data samples, deep small-world neural network, genetic algorithm and simulated annealing algorithm, optimizes and improves the relevant links and details of urban underground drainage pipe network maintenance, couples the hydrological and hydrodynamic relationship between pipeline functional diseases and urban waterlogging, and makes full use of the technical route of "theoretical analysis-experimental calibration and verification-model construction-optimal decision solution" to provide strong technical support for intelligent management and decision-making of urban drainage pipe network maintenance, and meet the intelligent, precise and scientific management and decision-making needs of most urban underground drainage pipe networks.

[0064] The embodiment of the present invention further provides a system using the above-mentioned intelligent decision-making method for urban underground drainage network maintenance, comprising:

[0065] Operate the subsystem for periodic detection of functional defects of urban drainage pipe networks, including: ultrasonic detection robots, video surveillance detectors, pipeline detection robots and magnetic flux detectors based on Internet of Things / base station data transmission technology, and supporting drainage pipe network functional defect detection data collection workstations;

[0066] Drainage network maintenance intelligent decision-making and integrated management database: The drainage network maintenance intelligent decision-making and integrated management database provides users of the upper application layer with storage and access control functions for real-time data of heavy rains with different recurrence periods, network maintenance costs, real-time collection of maintenance information, geographic information of functional diseases of the network, and other data;

[0067] The drainage network maintenance plan management subsystem includes the functions of maintenance and operation execution plan, work order execution and return order and work order data update; the maintenance and operation execution plan for functional defects of the network includes three strategies: overall expansion of the drainage network, siltation dredging and scale body cleaning; work order data update refers to the feedback of the current drainage network maintenance strategy execution status to the comprehensive management database, and use it as the initial input of the next intelligent decision-making evaluation subsystem to improve the accuracy of maintenance decisions;

[0068] Drainage network maintenance intelligent decision-making and evaluation subsystem, network functional disease three-dimensional instantaneous hydraulic analysis module, finite element full-scale test module, surface and underground one- and two-dimensional coupled connection model, rainwater model engineering implementation module and multi-objective planning decision and solution module.

[0069] Furthermore, in this embodiment, the urban underground drainage network maintenance intelligent decision-making system consists of two parts: software and hardware support platform:

[0070] The hardware support platform includes a comprehensive management database server, an urban pipe network GIS server, a drainage pipe network functional disease detection data collection workstation, a drainage pipe network maintenance plan execution agency workstation, a drainage pipe network maintenance plan real-time execution status collection workstation, an urban pipe network maintenance department monitoring and decision-making workstation, a regional pipe network maintenance department monitoring and decision-making workstation, and subordinate detection and maintenance instruments, meters, and sensors connected through the Internet of Things / base stations.

[0071] The software includes code implementation of functional modules such as drainage network maintenance cost statistics, deep small-world neural network algorithm, multi-objective decision-making for network maintenance, and urban network functional disease statistics platform. These software all run on the above-mentioned hardware support platform.

[0072] like Figure 3 As shown in the figure, the intelligent decision-making system for urban underground drainage network maintenance is divided into the data access control layer, the access control service layer and the application layer from the perspective of system structure:

[0073] The integrated management database server of the data access control layer is connected to the geographic information system (GIS) server and the software standard data interface respectively.

[0074] The functional items corresponding to the software standard data interface include pipeline network geographic information query module, maintenance information real-time collection module, maintenance cost update module, pipeline network functional disease diagnosis module and rainstorm fitting module with different recurrence periods.

[0075] The functional items corresponding to the software standard data interface are all obtained by wired / wireless monitoring technology support.

[0076] The access control service layer is hosted by the GIS server, which provides GIS services, business query and operation services, computing services and database access relay access control services for the application layer.

[0077] GIS service refers to the provision of geographic information interaction-related services for intelligent decision-making on maintenance of upper-level urban drainage pipe networks.

[0078] Business query and operation services include client / server (C / S) business system and browser / server (B / S) business system.

[0079] Computing service refers to providing the application layer with complex data analysis and processing of three-dimensional transient hydraulic analysis module, pipe network test parameter calibration module, one and two-dimensional surface and underground coupling connection model, rainwater model secondary development module, multi-objective optimization decision module, sample set expansion module, machine learning module and runoff test parameter collection module through the C / S business system; and providing the application layer with complex data analysis and processing of maintenance decision cost display module, maintenance decision benefit display module, maintenance point geographic information module, decision visualization real-time refresh module, different recurrence period rainstorm scene switching module and drainage network maintenance intelligent decision visualization interface through the B / S business system.

[0080] The basic test conditions for the runoff generation and runoff test parameter collection module are provided by the full-scale test scenario design of the drainage network.

[0081] Database access relay access control service refers to the application layer accessing the comprehensive management database server through the GIS server relay to achieve real-time access control of data and parameters related to intelligent decision-making for drainage network maintenance.

[0082] The application layer adopts a hybrid architecture mode of C / S business system and B / S business system, including: three-dimensional transient hydraulic analysis module, pipe network test parameter calibration module, surface and underground two-dimensional coupling connection model, rain and flood model secondary development module, multi-objective optimization decision module, sample set expansion module, machine learning module and runoff test parameter collection module for complex data analysis and processing; through the B / S business system, the application layer is provided with maintenance decision cost display module, maintenance decision benefit display module, maintenance point geographic information module, decision visualization real-time refresh module, different recurrence period rainstorm scene switching module and drainage network maintenance intelligent decision visualization interface.

[0083] The drainage network maintenance intelligent decision visualization interface realizes the real-time visualization of user-side results through the B / S business system, and can also issue intelligent decision commands to various network maintenance departments in the city through the visualization interface. The drainage network maintenance intelligent decision visualization interface periodically updates the network maintenance decision results based on multi-objective decision output.

[0084] The system using the intelligent decision-making method for urban underground drainage pipe network maintenance described in the above embodiment is a complex giant system involving fluid dynamics theory, "mass-momentum-energy" conservation theory, full-scale test method, secondary development technology, dynamic link technology, deep learning technology, database technology, geographic information system, data automatic acquisition technology, middleware technology, network technology and system integration technology, realizing a maintenance decision-making platform integrating pipe network functional disease data acquisition, parameter calibration, secondary development and intelligent optimization. Based on the hybrid mode based on the combination of C / S and B / S service modes, with the geographic information system server and the comprehensive database server as the data processing platform, and with fluid dynamics, full-scale test, multi-objective decision-making and genetic / simulated annealing algorithm as the analysis method, the design and development of the intelligent decision-making system for urban drainage pipe network maintenance is carried out, which can not only meet the system's complex data processing rate requirements, but also ensure the accuracy of the output of the maintenance decision results, and solve the problem of intelligent and precise preventive maintenance of urban waterlogged pipe networks considering the coupling effects of pipeline functional diseases and waterlogging under multi-objective constraints.

[0085] Obviously, the embodiments described above are only preferred embodiments of the present invention, rather than all embodiments. The preferred embodiments of the present invention are shown in the accompanying drawings, but they do not limit the patent scope of the present invention. The present invention can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive. Although the present invention has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific embodiments, or to perform equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of the present invention, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present invention.

Claims

1. An intelligent decision-making method for urban underground drainage network maintenance, characterized in that: The following steps are involved: S10. Analysis of three-dimensional instantaneous hydraulic model of functional diseases of drainage network based on fluid dynamics and "mass-momentum-energy" conservation theory; In step S10, the three-dimensional instantaneous hydraulic model analysis of functional diseases of the drainage network is theoretically derived from the perspective of multiphase flow and multi-field, and the liquid phase stress tensor and shear viscosity spatial distribution model, the solid phase stress tensor and shear viscosity spatial distribution model, and the gas phase Newtonian viscous stress spatial distribution model under the combined action of gravity field, pressure field, stress field, and flow field are respectively constructed; The steps of constructing the liquid phase stress tensor and shear viscosity spatial distribution model, the solid phase stress tensor and shear viscosity spatial distribution model, and the gas phase Newtonian viscous stress spatial distribution model include: S101. Based on the incompressibility of water flow and the principle of volume conservation, the time-averaged motion equation and the pulsating motion equation are obtained. ; S102. Combined with the "mass-momentum-energy" conservation equation A "solid-liquid-gas" spatial distribution model was constructed to obtain the functional disease index of the drainage network; u, v, and w are the components of the vertical average flow velocity of the river section on the x, y, and z axes of the three-dimensional coordinate system, respectively. , , is the real-time pulsating flow velocity of turbulence in each coordinate axis direction, V is the volume of any given space in the pipeline, is the boundary of the space domain, The values ​​1, 2, and 3 represent solid, liquid, and gas phases respectively. for Phase velocity, for The source term of the phase, for The density of the phase, For the border The normal vector of Other relative The source term of the phase, is the stress tensor, For physical strength, For the border The cross-sectional area, for The specific heat capacity of the phase, for The temperature of the phase, for The heat source of the phase, The phase transition of other phases Phase heat source; S20. Calibrate the parameters using finite element fitting analysis and full-scale tests, and verify the accuracy of the three-dimensional instantaneous hydraulic model of functional diseases of the drainage network; S30, combining the node water level iteration method, Preissmann slit method, Godunov finite volume method and unstructured grid, reconstructs a two-dimensional coupled connection model of the surface and underground; In step S30, the surface and underground two-dimensional coupled connection model is a one-dimensional pipe network full flow model and a two-dimensional surface water flow model coupled in the horizontal direction and the vertical direction; the one-dimensional pipe network full flow model and the two-dimensional surface water flow model are analyzed by weir flow formula method, mutual boundary method and fixed node water level method to obtain relevant structural formulas; S40. Use R language, dynamic library linking technology and long short-term memory neural network method of multi-source data samples to realize the engineering secondary development of the surface and underground two-dimensional coupling connection model, and obtain the functional disease status of the urban drainage network with waterlogging result labels; In step S40, the specific steps of engineering secondary development of the surface-underground two-dimensional coupled connection model are: S401. Based on the stormwater analysis software InforWorks ICM, the surface-underground two-dimensional coupling connection model is re-implemented using dynamic library linking technology and R language secondary development technology and embedded into the InforWorks ICM to obtain an improved InforWorks ICM; S402. The improved InforWorks ICM needs to integrate the "solid-liquid-gas" multiphase flow motion law model of the functional diseases of the pipeline network, the section confluence state model, the section instantaneous flow velocity, the flow model and the surface and underground one-dimensional coupling connection model on the basis of the existing InforWorks ICM, and secondary development is carried out to improve and upgrade the InforWorks ICM, so as to output the urban waterlogging loss situation under different pipeline network functional diseases. S50. Introduce deep small-world neural network, genetic algorithm and simulated annealing algorithm to establish a multi-objective planning intelligent decision-making model for drainage network maintenance and its solution method; In the step S50, the method for solving the multi-objective planning intelligent decision-making model for drainage network maintenance refers to using genetic algorithm and simulated annealing algorithm to perform intelligent optimization to obtain the functional disease conditions of the urban drainage network with maintenance decision labels; the initial input data of the deep small-world neural network is the functional disease conditions of the urban drainage network with maintenance decision labels, and after clustering, outlier detection and interpolation processing, the multi-layer restricted Boltzmann machine of the deep small-world neural network is repeatedly trained and adjusted through data set expansion to obtain the intelligent decision-making results of urban drainage network maintenance that are dynamically adjusted over time and rainstorms with different return periods; The deep small-world neural network consists of two stages: offline learning and online decision-making. The specific steps include: S501. With the help of the Internet of Things / base stations and the local area network of the online operation and maintenance system of the urban drainage network, obtain the multi-dimensional indicator data of the type, degree, and dredging / cleaning cost of the functional diseases of each important node of the network, including normal collection data, abnormal or large-scale malicious tampering sample data, and use indicator screening, anomaly detection and interpolation to process the data; S502, using a sliding window of a certain length to intercept all sample data, and obtain the mean subsequence under the sliding window; all subsequences are organized into a training sample set, and a status label is added to each training sample according to the InforWorks ICM numerical simulation results and the pipe network maintenance multi-objective planning results, mainly to classify the training sample nodes of the urban pipe network functional diseases that need maintenance; S503, using an unlabeled data set to pre-train a multi-layer restricted Boltzmann machine network of a deep small-world neural network, and after fully obtaining feature representation, recording network parameters, including weights and bias values ​​of neurons; S504, randomly adding edges to the multi-layer restricted Boltzmann machine network according to the random edge adding probability p of the deep small-world neural network, and adding a fully connected classification layer on the top layer to convert it into a deep small-world neural network with multi-classification function; supervised learning is used again to supervise the network training, and the back propagation algorithm is used to fine-tune the parameters until the training is completed, so as to obtain a trained deep small-world neural network classifier model; S505, online collection of functional disease indicator sets of urban drainage pipe networks, and processing of online data using the same sliding window as that used in training to obtain samples to be tested; S506. Input the samples to be tested into the trained deep small-world neural network classifier model to obtain the intelligent decision-making results and optimal response strategies for drainage network maintenance, thereby completing the intelligent decision-making management optimization process for urban underground drainage network maintenance.

2. According to claim 1, an intelligent decision-making method for urban underground drainage network maintenance is characterized in that: In the step S20, the finite element fitting analysis refers to constructing a "solid-liquid-gas" multiphase flow simulation model of functional diseases of the drainage network using Abaqus software to verify the parameters calibrated by the full-scale test and the theoretical structural formula obtained by the three-dimensional instantaneous hydraulic model of functional diseases of the drainage network; the full-scale test refers to building a landing well, a scouring gate and a functional disease pipe section, and calculating the accurate values ​​of the instantaneous state parameters of the pipeline confluence by the theoretical structural formula.

3. A system using the intelligent decision-making method for urban underground drainage network maintenance as described in any one of claims 1 to 2, characterized in that: include: Operate the subsystem for periodic detection of functional diseases of urban drainage networks; Drainage network maintenance intelligent decision-making and integrated management database; Drainage network maintenance plan management subsystem; Intelligent decision-making and evaluation subsystem for drainage network maintenance.

4. The system of the intelligent decision-making method for urban underground drainage network maintenance according to claim 3 is characterized in that: The operational urban drainage network functional disease periodic detection subsystem includes an ultrasonic detection robot based on the Internet of Things / base station data transmission technology, a video monitoring detector, a pipeline detection robot and a magnetic flux detector, as well as a drainage network functional disease detection data collection workstation; the drainage network maintenance intelligent decision-making integrated management database provides upper application layer users with real-time data of heavy rains with different recurrence periods, network maintenance costs, real-time collection of maintenance information, and access and access control functions of geographic information data of functional diseases of the network; the drainage network maintenance plan management subsystem includes the functions of maintenance and operation execution plan, work order execution and return, and work order data update; the drainage network maintenance intelligent decision-making evaluation subsystem includes a three-dimensional instantaneous hydraulic analysis module for functional diseases of the network, a finite element full-scale test module, a two-dimensional surface and underground coupled connection model, a rain and flood model engineering implementation module, and a multi-objective planning decision and solution module.

5. The system of the intelligent decision-making method for urban underground drainage network maintenance according to claim 3 is characterized in that: The system is divided into a data access control layer part, an access control service layer part and an application layer part in terms of system structure; the comprehensive management database server of the data access control layer part is connected to the geographic information system server and the software standard data interface respectively; the access control service layer part is carried by the geographic information system server and provides geographic information system services, business query and operation services, computing services and database access relay access control services for the application layer; the application layer part adopts a hybrid architecture model of C / S business system and B / S business system.

6. The system of the intelligent decision-making method for urban underground drainage network maintenance according to claim 3 is characterized in that: The hardware support platform of the system includes a comprehensive management database server, an urban pipe network geographic information system server, a drainage pipe network functional disease detection data collection workstation, a drainage pipe network maintenance plan execution agency workstation, a drainage pipe network maintenance plan real-time execution status collection workstation, an urban pipe network maintenance department monitoring and decision-making workstation, a regional pipe network maintenance department monitoring and decision-making workstation, and subordinate detection and maintenance instruments, meters, and sensors connected through the Internet of Things / base stations.

Images