CN115127036B - Municipal gas pipe network leakage positioning method and system - Google Patents
Municipal gas pipe network leakage positioning method and system Download PDFInfo
- Publication number
- CN115127036B CN115127036B CN202211060851.6A CN202211060851A CN115127036B CN 115127036 B CN115127036 B CN 115127036B CN 202211060851 A CN202211060851 A CN 202211060851A CN 115127036 B CN115127036 B CN 115127036B
- Authority
- CN
- China
- Prior art keywords
- pipe network
- leakage
- gas pipe
- municipal gas
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F17—STORING OR DISTRIBUTING GASES OR LIQUIDS
- F17D—PIPE-LINE SYSTEMS; PIPE-LINES
- F17D5/00—Protection or supervision of installations
- F17D5/02—Preventing, monitoring, or locating loss
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F17—STORING OR DISTRIBUTING GASES OR LIQUIDS
- F17D—PIPE-LINE SYSTEMS; PIPE-LINES
- F17D5/00—Protection or supervision of installations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Mechanical Engineering (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Fluid Mechanics (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Mathematical Analysis (AREA)
- Algebra (AREA)
- Examining Or Testing Airtightness (AREA)
Abstract
A municipal gas pipe network leakage positioning method and system belong to the technical field of municipal gas pipe network leakage monitoring, and comprise the steps of establishing a hydraulic and thermodynamic model of a municipal gas pipe network system; obtaining the flow, pressure and temperature of each node of the municipal gas pipe network by using the model through a numerical analysis method; simulating a leakage position by using the model; determining a municipal gas pipe network leakage early warning value, and identifying newly-added leakage points of the pipe network; training an artificial intelligent municipal gas pipe network leakage position positioning model, and positioning the municipal gas pipe network leakage position by adopting the artificial intelligent municipal gas pipe network leakage position positioning model; and outputting the leakage position of the municipal gas pipe network obtained through calculation. According to the method and the system, the limited sensors are used for monitoring data on line, the hydraulic and thermodynamic equation sets of the pipe network are solved, and the artificial intelligence model is combined, so that the leakage state of the complex pipe network can be efficiently identified in a short time, the training is faster, and the required sample amount is greatly reduced.
Description
Technical Field
The invention belongs to the technical field of municipal gas pipe network leakage monitoring, and particularly relates to a municipal gas pipe network leakage positioning method and system.
Background
Municipal gas pipelines in cities cause accidents, so that gas leakage frequently occurs, normal delivery of gas is influenced, environmental pollution is caused, fire, explosion and the like are caused, and great threat is brought to life and property safety of people. In 2016-2019, 3395 gas explosion accidents occur together, 374 people die and 3754 people are injured; in spite of all gas explosion accidents, more than 80 percent of the accidents are caused by perforation and air leakage due to serious corrosion of pipelines. Therefore, the research on the gas pipeline leakage detection and positioning technology has extremely important practical significance and has attracted attention from countries all over the world.
Although research on the pipeline leakage detection method has been carried out for decades, no simple, reliable and universal method for detecting and positioning the leakage of the gas pipeline exists at present due to the complexity of gas leakage detection, the diversity of the environment where the pipeline is located and the diversity of the leakage form.
The manual inspection method is a common leakage detection method for various domestic urban gas companies at present. The gas leakage detector or the leak detection vehicle is held by a patrol worker to patrol along a pipeline laying path at regular intervals, and whether gas leakage exists is judged in multiple modes of seeing, smelling, listening and the like. After the gas leaks from the underground pipeline, the gas can leak in different directions due to different types, different specific gravities and different surrounding environments, so the accuracy of leak detection and positioning of the method has great relation with the experience and subjective judgment of inspection personnel.
The mass balance method is the most basic detection method. It judges whether the pipeline leaks according to the flow difference of the inlet and outlet pipeline fluids. The method is simple and intuitive and easy to realize, but the flow measurement is greatly influenced by the change of parameters such as fluid components, temperature, pressure and the like, so the measurement accuracy is lower, and meanwhile, the method needs to install high-precision flow meters at two ends of a pipeline, has higher cost and can not realize the positioning of a leakage position.
The negative pressure wave method is a method which is researched more in China, when a certain position of a pipeline leaks suddenly, transient pressure suddenly drops at the leaking position to form a negative pressure wave, the negative pressure wave is transmitted to two ends of the pipeline at a certain speed, and the specific position of a leakage point can be determined according to the time difference of the negative pressure wave transmitted to a pressure sensor at the upstream end and the downstream end and the transmission speed of the negative pressure wave. The current research focuses on how to accurately determine the time difference and correct the propagation speed formula, and the main research methods include a correlation analysis method, a wavelet transform method, a time series method, an image processing-based method and the like. For example, chinese patent CN200910070101.5 discloses a method for automatically detecting and positioning gas network leakage based on Geographic Information System (GIS) and data acquisition and monitoring (SCADA) technology: establishing a gas pipe network GIS system, and reading and storing a pipe network diagram, pipe network attribute data and the like; collecting the pressure, temperature and flow parameters of each section of gas pipeline in real time by an SCADA system; comparing and calculating the acquisition parameters with the stored data; calculating the deviation between the measured value and the calculated parameter value at the head end and the tail end of the pipeline, and displaying the abnormal pipeline information; searching an extreme point to determine a pressure catastrophe point, and calculating the position of the leakage point through a leakage positioning formula according to the gas flow, the temperature and the pressure parameters of the leakage pipe section. The method adopts a wavelet transform method to perform correlation analysis. However, the negative pressure wave method has the disadvantage that only large, sudden leaks can be detected, which is not suitable for small leaks that occur slowly.
Disclosure of Invention
The invention aims to solve the problems that in the prior art, a municipal gas pipe network with a city-level complex topological structure cannot accurately position leakage points, large-scale manual leakage detection is low in efficiency and the like, and provides a method and a system for positioning leakage of the municipal gas pipe network; the method and the system simulate the leakage by adopting numerical calculation, train an artificial intelligent model by adopting data obtained by numerical simulation, and identify and position the leakage of the municipal gas pipe network; because the numerical simulation calculation result is accurate, the precision of the artificial intelligence model can be ensured. In addition, another municipal gas pipe network leakage positioning method and system are provided, numerical calculation is adopted to identify and position leakage under the condition of less municipal gas pipe network leakage, after a certain amount of leakage data are accumulated, an artificial intelligence mode is adopted to identify and position leakage, and in addition, when an artificial intelligence model is trained, two parts of municipal gas pipe network leakage data and numerical simulation data which are truly recorded are adopted as training sets, so that sufficient sample data can be ensured when the real leakage data are less, and the calculation accuracy of the artificial intelligence model is ensured; the invention combines numerical calculation and artificial intelligence, and can quickly and accurately position leakage in each stage of operation of the municipal gas pipe network.
The invention provides a municipal gas pipe network leakage positioning method which is characterized by comprising the following steps:
s1, establishing a hydraulics and thermodynamics model of a municipal gas pipe network system; obtaining the flow, pressure and temperature of each node of the municipal gas pipe network by a numerical analysis method;
s2, simulating a leakage position by adopting the hydraulics and thermodynamics model of the municipal gas pipe network system in the step S1;
s3, determining a municipal gas pipe network leakage early warning value, and identifying newly-added leakage points of the pipe network;
s4, training an artificial intelligent municipal gas pipe network leakage position positioning model, and positioning the municipal gas pipe network leakage position by adopting the artificial intelligent municipal gas pipe network leakage position positioning model;
and S5, outputting the leakage position of the municipal gas pipe network obtained through calculation.
The invention provides another municipal gas pipe network leakage positioning method which is characterized by comprising the following steps of:
s1, establishing a hydraulics and thermodynamics model of a municipal gas pipe network system; obtaining and calculating the flow, pressure and temperature of each node of the municipal gas pipe network by a numerical analysis method;
s2, simulating a leakage position by adopting the hydraulics and thermodynamics model of the municipal gas pipe network system in the step S1;
s3, determining a municipal gas pipe network leakage early warning value, and identifying newly-added leakage points of the pipe network;
s4, selecting a proper pipe network leakage positioning mode; if the recorded real leakage events are less than K pieces, adopting the hydraulics and thermodynamics calculation mode of the step S5 to position the leakage position; if the recorded real leakage event is more than or equal to K pieces, positioning the leakage position by adopting the artificial intelligence calculation mode of the step S6;
s5, positioning the municipal gas pipe network leakage position by adopting the hydraulics and thermodynamics model of the municipal gas pipe network system in the step S1; the method comprises the following steps:
s501, sequentially assuming newly increased leakage on nodes at two ends of a computing unit divided by all pipe sections in a corrected municipal gas pipe network system hydraulics and thermodynamics model, traversing all the computing units in the hydraulics and thermodynamics model, and solving the flow, pressure and temperature of each node of the municipal gas pipe network;
and S502, comparing the calculated values of the pressure and the temperature of each node with the measured values of the pressure and the temperature measured by the sensor in the municipal gas pipeline network, and taking the node position assumed by the current calculation with the highest goodness of fit as an automatic positioning leakage position.
S6, training an artificial intelligent municipal gas pipe network leakage position positioning model, and positioning the municipal gas pipe network leakage position by adopting the artificial intelligent municipal gas pipe network leakage position positioning model;
and S7, outputting the leakage position of the municipal gas pipe network obtained through calculation.
Further, the step S1 specifically includes the steps of:
s101, establishing a hydraulics and thermodynamics model of the municipal gas pipe network system by reading physical layer information;
and S102, solving a pipe network hydraulics and thermodynamics simultaneous equation set in the municipal gas pipe network system model to obtain the flow, pressure and temperature of each node.
Further, the step S1 further includes the steps of: and monitoring flow, pressure and temperature data by using a sensor in the municipal gas pipe network system, and correcting a hydraulics and thermodynamics model of the municipal gas pipe network system.
Further, in step S3, identifying newly-added leakage of the municipal gas pipe network specifically includes:
s301, reasonably determining a leakage early warning value according to the design condition and the actual operation condition of the municipal gas pipe network;
and S302, collecting flow monitoring data of sensors arranged at an entrance and an exit and a user node in the municipal gas pipeline network system in a certain period, calculating the difference between the current collection value and the previous collection value, and judging that a newly increased leakage point occurs if the difference is greater than a leakage early warning value.
Further, the training of the artificial intelligence municipal gas pipe network leakage position positioning model comprises the following steps:
in a time period T, collecting flow, pressure and temperature data monitored by a municipal gas pipe network sensor at a certain sampling frequency to obtain actually measured flow, pressure and temperature data monitored by the pipe network sensor under the condition of leakage, and recording the leakage position;
randomly selecting the actually measured leakage position and part of data in the simulated leakage position obtained in the step S2 as an original training set, and using the rest part of data as a test set;
and obtaining an artificial intelligent municipal gas pipe network leakage position positioning model by utilizing the training set and the testing set and adopting a deep learning algorithm.
Further, the deep learning algorithm selects a random forest algorithm, and specifically comprises the following steps:
step 1, respectively training n decision tree models with the number of layers being p for n training sets;
step 2, for a single decision tree model, assuming that the number of training sample features is m, selecting the best feature to split according to the information gain ratio during each splitting;
step 3, splitting each tree according to the steps until all training samples of the node belong to the same class;
step 4, forming a random forest by the generated decision trees, and determining a prediction leakage position according to a mean value of prediction of the decision trees;
step 5, testing the model by adopting the data in the test set, and comparing the predicted leakage position with the known leakage position:
(1) The deviation degree is less than or equal to L' (meter), namely the test is qualified;
(2) If the deviation degree is larger than L' (meter), increasing the layer number p, and repeating the steps 2-4;
if the test is qualified, the test is qualified;
if the test is not goodThe number of layers p of the decision tree model and the number of layers p of the decision tree model are compared y Making a comparison if p is less than or equal to p y Increasing the number p of layers, and repeating the steps 2-4 until the test is qualified; if p > p y And prolonging the acquisition time period T, and repeating the steps 1-4 until the test is qualified.
Further, the position of the newly found leakage point is fed back to a training set or a testing set, so that the precision of the artificial intelligent model is automatically improved along with the increase of the sample data volume and the artificial intelligent model is adaptive to new data.
The invention provides a municipal gas pipe network leakage positioning system, which comprises: the system comprises a sensing system, a data transmission system and a leakage positioning system;
the sensing system is a pressure sensor, a flow sensor and a temperature sensor which are arranged in the municipal gas pipe network system and is used for monitoring the pressure, the flow and the temperature of the inlet and the outlet of the municipal gas pipe network, the user nodes, the key non-user nodes and the like in real time;
the data transmission system is used for transmitting the pressure, flow and temperature data measured by the sensing system to the leakage positioning system;
the leakage localization system comprises:
the data storage module is used for storing pressure, flow and temperature data measured in the sensing system and leakage position data;
the numerical simulation module is used for constructing a hydraulic and thermodynamic model of the municipal gas pipe network and solving the flow, pressure and temperature values of each node of the municipal gas pipe network through a numerical analysis method;
the leakage identification module is used for determining a leakage early warning value of the municipal gas pipe network by using the municipal gas pipe network flow, pressure and temperature data obtained from the sensing system, and realizing the identification of a newly added leakage point according to the leakage early warning value and real-time flow monitoring data of an inlet, an outlet and a user node in the municipal gas pipe network system;
the leakage position artificial intelligence positioning module is used for training an artificial intelligence model by utilizing pressure, flow and temperature data measured by the sensing system, historical leakage position data of the municipal gas pipe network and simulated leakage position data obtained by the numerical simulation module, wherein part of data is randomly selected as a training set, and the rest of data is used as a test set, so that an artificial intelligence municipal gas pipe network leakage position positioning model is obtained, and the municipal gas pipe network leakage position is determined by adopting the artificial intelligence municipal gas pipe network leakage position positioning model;
and the leakage position output module outputs the leakage position of the municipal gas pipe network.
And the model correction module is used for correcting the hydraulics and thermodynamics model of the municipal gas pipe network system in real time by using the municipal gas pipe network flow, pressure and temperature data obtained from the sensing system.
Further, still include: the leakage point positioning method selection module is used for selecting and adopting a leakage position numerical value positioning module or a leakage position artificial intelligence positioning module to position the leakage point according to the recorded real leakage event number;
and the leakage position numerical value positioning module is used for positioning the leakage position of the municipal gas pipe network by adopting a model in the numerical simulation module and a numerical analysis method.
The municipal gas pipe network leakage positioning method and the municipal gas pipe network leakage positioning system realize automatic positioning of municipal gas pipe network leakage and automatic pressure adjustment. The method and the system have the following beneficial effects:
1. the leakage position can be automatically positioned on line, the rush repair of the leakage of the municipal gas pipe network is accelerated, and the safety risk and the economic loss caused by the leakage are reduced;
2. the method has the advantages that the method adopts a limited number of sensors to monitor data on line, solves a hydraulic and thermodynamic equation set of the pipe network and combines random forest model training, can efficiently identify the leakage state of the complex pipe network in a short time, and is quicker and greatly reduces the required sample amount compared with other artificial intelligent training on leakage sample data;
3. the method is suitable for various topological structures (ring networks and dendritic networks) and pipe networks of various scales, and can adapt to the development current situation that the pipe network is continuously expanded, the gas distribution and the node flow are continuously changed;
4. with continuous application and continuous accumulation of data, the system can perform self-learning, thereby improving the accuracy of leakage identification and the positioning precision of leakage, and enhancing the anti-interference capability;
5. the workload of daily operation and maintenance inspection personnel is greatly saved, the operation and maintenance cost is reduced, and the working efficiency of daily inspection is improved;
6. the system can cooperate with operation and maintenance departments to find potential threats in time, effectively prevent safety accidents such as pipe explosion and the like, and provide a quick and reliable scientific basis for urban municipal gas emergency plans.
Drawings
For a more complete understanding of the present invention, reference is now made to the following descriptions taken in conjunction with the accompanying drawing, in which:
FIG. 1 is a flow chart of a municipal gas pipe network leakage positioning method according to an embodiment of the invention;
FIG. 2 is a flow chart of the present invention for locating the position of a leakage point of a municipal gas pipeline network by using an artificial intelligence method;
FIG. 3 is a flow chart of a municipal gas pipe network leakage positioning method according to an embodiment of the invention;
FIG. 4 is a schematic structural view of the municipal gas pipe network leakage positioning system of the present invention;
FIG. 5 is a sensor layout diagram of the municipal gas pipeline network leakage positioning system of the invention.
Detailed Description
For the purpose of illustrating the invention, its technical details and its practical application to thereby enable one of ordinary skill in the art to understand and practice the invention, reference will now be made in detail to the embodiments of the present invention with reference to the accompanying drawings. It is to be understood that the embodiments described herein are merely illustrative and explanatory of the invention and are not restrictive thereof.
The invention provides a municipal gas pipe network leakage positioning method, which correlates each system of a municipal gas pipe network physical layer, a perception layer, a control layer, a data layer, a model layer, an application layer and the like, and has the core of a pipe network hydraulics model, a thermodynamics model and an artificial intelligence model: the method comprises the following steps that a pipe network is divided through a hydraulic and thermodynamic model of the pipe network, the leakage point position is assumed by adopting an exhaustion method, the flow, pressure and temperature of each node are calculated, the calculated value of the pressure and temperature of each node is compared with the measured value of the pressure and temperature, and the assumed node position calculated at the current time with the highest goodness of fit is the leakage point position automatically positioned by the system; the artificial intelligence model is used for training by acquiring leakage sample data for a period of time and simulating leakage data and actual leakage data by using the model, and automatically positioning a leakage point. When the leakage events are less, calculating the positions of the leakage points by taking a hydraulic and thermodynamic model of a pipe network as a main model; when the leakage event is enough, calculating the position of the leakage point by taking the artificial intelligence model as a main part. According to the method, the leakage state of the complex pipe network can be efficiently identified within a short time after the complex pipe network is put into application by utilizing the on-line monitoring data of a limited number of sensors, compared with other methods for carrying out artificial intelligence training on leakage sample data, the method is quicker, the required sample amount is greatly reduced, the leakage position can be automatically positioned, the rush repair of the leakage of the municipal gas pipe network is accelerated, and the safety risk and the property loss are reduced.
Referring to the attached figure 1 of the specification, the municipal gas pipe network leakage positioning method mainly comprises the following steps:
s1, establishing a hydraulic and thermodynamic model of the municipal gas pipe network.
S101, establishing a hydraulics and thermodynamics model of a municipal gas pipe network system; a municipal gas pipe network system model is established by reading physical layer information including geographic information, pipe network basic information (including pipeline plane and vertical information, pipe diameter, pipes and valves), gas sources, pressure regulating stations, users and the like.
S102, solving a hydraulic and thermodynamic equation set of a pipe network in a municipal gas pipe network system model by a numerical analysis method to obtain the flow, pressure and temperature of each node; the method specifically comprises the following steps:
establishing a topological structure chart of a pipe network system for simulation analysis, dividing all pipe sections into computing unit pipe sections with the length less than or equal to L (such as 1-10 meters) meters according to the distribution condition of user nodes and the positioning precision requirement of leakage points,
whether the network is a branched network or a circular network, the solution of the nonlinear equation system is carried out based on the conservation law, and the conservation of energy and the conservation of mass are followed.
Assuming that no flow flows out in the calculation unit pipe section and only flow flows out at nodes at two ends of the pipe section, respectively calculating a flow value, a pressure value and a temperature value of nodes (including a user and a non-user) at two ends of the calculation unit pipe section according to the following formula:
wherein the content of the first and second substances,Mis the gas mass flow rate;Tis the gas temperature;Pis the gas pressure;xfor different positions (m) of the pipe section;Ais the cross-sectional area of the pipeline;gis the acceleration of gravity;θis the pipe inclination angle;Dis the inner diameter of the gas pipeline;λis the coefficient of frictional resistance;T 0 the temperature of soil in the deep buried position of the pipeline is measured;
wherein the content of the first and second substances,R 0 is the gas constant; a. The 0 、B 0 、C 0 、D 0 、E 0 、γ、a、b、c、d、
Are the parameters of the state equation.
hThe enthalpy value of the fuel gas can be calculated by the following formula:
wherein the content of the first and second substances,m i in a mixed gasiThe mass components of the components; a. The i 、B i 、C i 、D i 、E i 、F i Is composed ofiA component constant; m 0 Is the average molecular weight of the gas.
KThe total heat transfer coefficient from the gas to the soil can be calculated by:
wherein the content of the first and second substances,
the heat release coefficient from the gas to the inner wall of the tube; d in Is the inner diameter of the pipeline; lambda [ alpha ] i Is as followsiThermal conductivity of the layers (pipe wall, anti-corrosion layer); d i Is a pipelineiInner diameter of the layer; d i+1 Is a pipelinei+1 layer inner diameter;
the heat release coefficient from the outer wall of the tube to the surrounding medium; d ex Is the outside diameter of the pipe.
When the pipe network is not damaged: the pressure value, the flow value and the temperature value of the total inlet of the pipe network are known, the flow of the user nodes of the pipe network is known, the flow of the non-user nodes of the pipe network is 0, and the pressure value, the flow value and the temperature value of each node can be calculated according to the relational expression.
And S103, monitoring flow, pressure and temperature data by using a sensor in the municipal gas pipe network system, and correcting a finite element model of the municipal gas pipe network system.
After the pipe network operates for a period of time, due to the factors such as dirt accumulation inside the pipeline, corrosion of the pipe network and the like, the friction coefficient lambda of the pipeline and the total heat transfer coefficient K from the fuel gas to the soil can be changed, and therefore the theoretical friction resistance coefficient lambda and the total heat transfer coefficient K from the fuel gas to the soil need to be corrected, so that the pipe network model is closer to the actual operating state, and the model precision is guaranteed.
The model is modified in the following manner:
after the model is initially established, under the condition that a pipe network is not lost (when the leakage early warning value is lower than the leakage early warning value), collecting flow, pressure and temperature monitoring data of a sensor arranged in the municipal gas pipe network system, comparing the pressure monitoring value and the temperature detection value of each user node at a time point with the pressure calculation value and the temperature calculation value of the time point, and correcting pipe sections among the user nodes according to deviation. And the corrected lambda and K are used as the lambda and K values for model calculation in the subsequent step.
And S2, simulating a leakage position by adopting the hydraulic and thermodynamic model of the municipal gas pipe network in the step S1.
And (3) assuming a certain node of the calculation unit pipe section as a leakage node, assigning different leakage quantities, solving a hydraulic and thermodynamic equation set of the pipe network to obtain the flow, pressure and temperature simulation values of the monitoring point node in the leakage state, and recording the simulated leakage position.
The solving method for simulating the leakage position comprises the following steps: and (4) sequentially assuming the assumed leakage amount on nodes at two ends of the calculation units of all the pipe section divisions, performing the calculation of the step (S102), comparing the calculated values of the pressure and the temperature of each node with the measured values of the pressure and the temperature, and taking the position of the node which is assumed by the current calculation and has the highest goodness of fit as a simulated leakage position.
And S3, determining a municipal gas pipe network leakage early warning value, and identifying newly added leakage points of the pipe network.
According to the design conditions (pipe materials, pipe age and the like) and the actual operation conditions (pipeline inspection and maintenance level) of the municipal gas pipe network, the leakage early warning value is reasonably determined, and generally 5% -6% of the total gas supply can be taken.
And collecting flow monitoring data of sensors arranged at an entrance and an exit and a user node in the municipal gas pipe network system at a certain period (such as once in 5 minutes), calculating the difference between the current collected flow and the last collected flow, and judging that a newly-added leakage point occurs if the difference is greater than a leakage early warning value.
And S4, training an artificial intelligent municipal gas pipe network leakage position positioning model, and positioning the municipal gas pipe network leakage position by adopting the artificial intelligent municipal gas pipe network leakage position positioning model. The artificial intelligence method may adopt a common artificial intelligence calculation method, such as a random forest algorithm, a neural network, and the like, and the following description takes the random forest algorithm as an example, and the specific flow refers to fig. 2 in the description.
Step S401, collecting flow, pressure and temperature data monitored by the municipal gas pipe network sensor at a certain sampling frequency in a time period T to obtain actually measured flow, pressure and temperature data monitored by the pipe network sensor under the condition of leakage, and recording the leakage position.
And S402, randomly selecting 80% of the data collection in the steps S401 and S2 as an original training set by using a Bootstrap method, and using the rest 20% of data as a test set.
And S403, obtaining an artificial intelligent municipal gas pipe network leakage position positioning model by using the training set and the testing set in the step S402 and adopting a deep learning algorithm.
Taking a random forest algorithm as an example, the specific method is as follows:
step 4031, for n training sets, respectively training n decision tree models, wherein the number of layers is p.
Step 4032, for a single decision tree model, assuming that the number of training sample features is m, selecting the best feature to split according to the information gain ratio during each splitting.
4033, each tree is split according to the steps until all training examples of the node belong to the same class. And 4034, forming a random forest by the generated multiple decision trees, and determining the predicted leakage position according to the predicted mean value of the multiple trees.
Step 4035, the model is tested by adopting the data in the test set, and the predicted leakage position is compared with the known leakage position:
(1) The deviation degree is less than or equal to L' (meter), namely the test is qualified;
(2) If the deviation degree is larger than L' (meter), the number of layers p is increased, and the steps 4032-4034 are repeated.
If the test is qualified, the test is qualified;
if the test is not qualified, the number p of layers of the decision tree model and the number p of layers of the decision tree model are compared y (the highest number of layers defined) if p ≦ p y Increasing the number p of layers, and repeating the steps 4032-4034 until the test is qualified; if p > p y The acquisition time period T in step S401 is extended, and steps S401 to S403 are repeated until the test is qualified.
And S404, determining a leakage position by adopting the artificial intelligent municipal gas pipe network leakage position positioning model which is qualified in test.
And S5, outputting the calculated leakage position. For example, the leakage specific positioning information can be displayed on a municipal gas pipe network digital twin model, and is sent to an operation and maintenance management center, so that automatic identification and positioning of leakage are realized.
In order to improve the leakage positioning precision of the municipal gas pipe network, after the step S5 is finished, the position where leakage is newly found can be further fed back to the step S401 to increase the number of samples, so that the precision of the artificial intelligent model is automatically improved along with the increase of the sample data volume and the artificial intelligent model is adapted to new data.
Can be according to the least unfavorable user point minimum air feed pressure value H in municipal gas pipe network system topological structure min And the minimum pressure values of the inlet of the gas supply area, the important node and the user are calculated, and the regulation suggestions of the valve of the associated pipe section and the pressure regulating station are output according to the relation between the opening degree of the valve and the pressure loss. For example, more than one week, it is monitored that the minimum air supply pressure value of the worst user point is continuously higher than the standard value, that is, it is determined that the system pressure is too high, and a pressure regulating space exists.
In addition, considering that in the initial stage of operation of the municipal gas pipe network, the leakage occurs less, the number of samples during artificial intelligence model training is less, and the precision of the artificial intelligence model may be insufficient, the invention provides a municipal gas pipe network leakage positioning method based on mechanical calculation and artificial intelligence, which adds a selection step of the leakage positioning method and a step of calculating the leakage position by adopting a numerical method on the basis of the positioning method of the first embodiment of the invention, and the method specifically comprises the following steps with reference to the attached drawing 3 of the specification:
s1, establishing a hydraulics and thermodynamics model of the municipal gas pipe network system.
And S2, simulating a leakage position by adopting the hydraulics and thermodynamics model of the municipal gas pipe network system in the step S1.
And S3, determining a municipal gas pipe network leakage early warning value, and identifying newly added leakage points of the pipe network.
The steps S1 to S3 are the same as the corresponding steps in the first embodiment of the present invention, and are not described herein again.
And S4, selecting a proper pipe network leakage positioning mode.
Collecting data of node flow, pressure and temperature sensors transmitted in real time when the pipe network runs, and after identifying that a new leakage point is formed in the pipe network in the step S3, if the recorded real leakage event is less than K pieces, adopting a hydraulic and thermodynamic calculation mode in the step S5 to position a leakage position; and if the recorded real leakage event is more than or equal to K pieces, positioning the leakage position by adopting the artificial intelligence calculation mode of the step S6.
And S5, positioning the leakage position of the municipal gas pipe network by adopting the hydraulics and thermodynamics model of the municipal gas pipe network system in the step S1. The method comprises the following specific steps:
s501, sequentially assuming the newly increased leakage on nodes at two ends of calculation units of all pipe sections in the corrected municipal gas pipe network system hydraulics and thermodynamic model, traversing all the calculation units in the hydraulics and thermodynamic model, solving an equation set, and solving the flow, pressure and temperature of each node of the municipal gas pipe network;
and S502, comparing the calculated value of the pressure and the temperature of each node with the measured values of the pressure and the temperature measured by the sensor in the municipal gas pipeline network, and taking the node position (such as X, Y coordinates, the pipeline section where the node is located, nearby geographic information and the like) assumed by the current calculation with the highest matching degree as an automatic positioning leakage position.
And S6, training an artificial intelligent municipal gas pipe network leakage position positioning model, and positioning the municipal gas pipe network leakage position by adopting the artificial intelligent municipal gas pipe network leakage position positioning model.
This step is the same as step S4 in the first embodiment of the present invention, and is not described herein again.
And S7, outputting the leakage position of the municipal gas pipe network obtained through calculation.
The invention provides a municipal gas pipe network leakage positioning system, which is shown in the attached figure 4 of the specification and comprises the following components: a sensing system 100, a data transmission system 200, and a leak location system 300.
The sensing system 100 is a pressure sensor, a flow sensor and a temperature sensor which are arranged in a municipal gas pipe network system, and referring to fig. 5 of the specification, flow, pressure and temperature sensors are arranged at all inlets, outlets and important nodes of an area and are used for monitoring the pressure, the flow and the temperature of the municipal gas pipe network inlet, outlet, user nodes, important non-user nodes and the like in real time.
The data transmission system 200 is used to transmit the pressure, flow and temperature data measured by the sensing system 100 to the leakage localization system 300. The data transmission of the data transmission system 200 may adopt a wireless or wired transmission mode commonly used in the art; the data transmission system 200 may include network cables, optical fibers, routers, and the like.
The leak location system 300 includes:
a data storage module 301, configured to store pressure, flow, and temperature data measured in the sensing system 100, and leakage position data;
the numerical simulation module 302 is used for constructing a hydraulics and thermodynamics model of the municipal gas pipe network and solving the flow, pressure and temperature values of each node of the municipal gas pipe network by a numerical analysis method;
the leakage identification module 303 is configured to determine a leakage early warning value of the municipal gas pipe network by using the municipal gas pipe network flow, pressure, and temperature data obtained in the sensing system 100, and implement identification of a newly added leakage point according to the leakage early warning value and real-time flow monitoring data at an inlet, an outlet, and a user node in the municipal gas pipe network system;
the leakage position artificial intelligence positioning module 304 is used for training an artificial intelligence model by utilizing pressure, flow and temperature data measured by the sensing system 100, historical leakage position data of the municipal gas pipe network and simulated leakage position data obtained by the numerical simulation module 302, wherein 80% of data are randomly selected as a training set, 20% of data are selected as a test set, the artificial intelligence model is obtained, and the municipal gas pipe network leakage position is determined by adopting the artificial intelligence municipal gas pipe network leakage position positioning model;
and a leakage position output module 305 for outputting the leakage position of the municipal gas pipe network.
Optionally, the leakage localization system 300 may further include:
the model correction module 306 is used for correcting the hydraulics and thermodynamics model of the municipal gas pipe network system in real time by using the municipal gas pipe network flow, pressure and temperature data obtained from the sensing system 100;
a leakage point positioning method selecting module 307, configured to select to perform leakage point positioning by using the leakage position numerical value positioning module 308 or the leakage position artificial intelligence positioning module 305 according to the recorded actual number of leakage events;
and the leakage position numerical value positioning module 308 is used for positioning the leakage position of the municipal gas pipe network by adopting the model in the numerical simulation module 302 and adopting a numerical analysis method.
The municipal gas pipe network leakage positioning system can be connected with an application module, original data, leakage identification, leakage positioning, pressure/temperature prediction and the like obtained from the positioning system are displayed in a cloud picture mode and the like, and leakage positioning, flow simulation cloud pictures, pressure simulation cloud pictures, temperature simulation cloud pictures, real-time monitoring, information integration, operation and maintenance feedback, updating decision and the like are achieved.
The municipal gas pipe network leakage positioning system is used for realizing the municipal gas pipe network leakage positioning method.
Claims (10)
1. A municipal gas pipe network leakage positioning method is characterized by comprising the following steps:
s1, establishing a hydraulics and thermodynamics model of a municipal gas pipe network system; obtaining and calculating the flow, pressure and temperature of each node of the municipal gas pipe network by a numerical analysis method; the method specifically comprises the following steps:
s101, establishing a hydraulics and thermodynamics model of the municipal gas pipe network system by reading physical layer information;
s102, solving a pipe network hydraulics and thermodynamics simultaneous equation set in the municipal gas pipe network system hydraulics and thermodynamics model to obtain the flow, pressure and temperature of each node;
s2, simulating a leakage position by adopting the hydraulics and thermodynamics model of the municipal gas pipe network system in the step S1;
s3, determining a municipal gas pipe network leakage early warning value, and identifying newly-added leakage points of the pipe network;
s4, selecting a proper pipe network leakage positioning mode; if the recorded real leakage events are less than K pieces, adopting the hydraulics and thermodynamics calculation mode of the step S5 to position the leakage position; if the recorded real leakage event is more than or equal to K pieces, positioning the leakage position by adopting the artificial intelligence calculation mode of the step S6;
s5, positioning the municipal gas pipe network leakage position by adopting the hydraulics and thermodynamics model of the municipal gas pipe network system in the step S1;
s6, training an artificial intelligent municipal gas pipe network leakage position positioning model, and positioning the municipal gas pipe network leakage position by adopting the artificial intelligent municipal gas pipe network leakage position positioning model;
training of artificial intelligence municipal gas pipe network leakage position location model includes the following steps:
in a time period T, collecting flow, pressure and temperature data monitored by a municipal gas pipe network sensor at a certain sampling frequency to obtain actually measured flow, pressure and temperature data monitored by the pipe network sensor under the condition of leakage, and recording the leakage position;
randomly selecting the actually measured leakage position and part of data in the simulated leakage position obtained in the step S2 as an original training set, and using the rest data as a test set;
obtaining an artificial intelligent municipal gas pipe network leakage position positioning model by utilizing the training set and the testing set and adopting a deep learning algorithm;
and S7, outputting the leakage position of the municipal gas pipe network obtained through calculation.
2. The method according to claim 1, wherein the step S5 comprises the steps of:
s501, sequentially assuming newly increased leakage on nodes at two ends of a computing unit of all pipe sections in the municipal gas pipe network system hydraulics and thermodynamics model, traversing all the computing units in the hydraulics and thermodynamics model, and obtaining flow, pressure and temperature of each node of the municipal gas pipe network;
and S502, comparing the calculated values of the pressure and the temperature of each node with the measured values of the pressure and the temperature measured by the sensor in the municipal gas pipeline network, and taking the node position assumed by the current calculation with the highest goodness of fit as an automatic positioning leakage position.
3. The method according to claim 1, wherein the step S1 further comprises the steps of: and monitoring flow, pressure and temperature data by using a sensor in the municipal gas pipe network system, and correcting a hydraulics model and a thermodynamics model of the municipal gas pipe network system.
4. The method according to claim 1, wherein in the step S3, the identification of newly added leakage of the municipal gas pipe network is specifically as follows:
s301, reasonably determining a leakage early warning value according to the design condition and the actual operation condition of the municipal gas pipe network;
and S302, collecting flow monitoring data of sensors arranged at an entrance and an exit and a user node in the municipal gas pipeline network system in a certain period, calculating the difference between the current collection value and the previous collection value, and judging that a newly-increased leakage point occurs if the difference is greater than a leakage early warning value.
5. The method as claimed in claim 1, wherein the deep learning algorithm selects a random forest algorithm, comprising the specific steps of:
step 1, respectively training n decision tree models with the number of layers being p for n training sets;
step 2, for a single decision tree model, assuming that the number of training sample features is m, selecting the best feature to split according to the information gain ratio during each splitting;
step 3, splitting each tree according to the steps until all training samples of the node belong to the same class;
step 4, forming a random forest by the generated decision trees, and determining a prediction leakage position according to a mean value of prediction of the decision trees;
step 5, testing the model by adopting the data in the test set, and comparing the predicted leakage position with the known leakage position:
(1) The deviation degree is less than or equal to L' (meter), namely the test is qualified;
(2) If the deviation degree is larger than L' (meter), increasing the layer number p, and repeating the steps 2-4;
if the test is qualified, the test is considered to be qualified;
if the test is not qualified, the number p of layers of the decision tree model and the number p of layers of the decision tree model are compared y Making a comparison if p is less than or equal to p y Increasing the number p of layers, and repeating the steps 2-4 until the test is qualified; if p > p y And prolonging the acquisition time period T, and repeating the steps 1-4 until the test is qualified.
6. The method of claim 1, wherein the position of the newly found leakage point is fed back to the training set or the test set, so that the artificial intelligence model automatically improves accuracy and adapts to new data as the sample data volume increases.
7. A municipal gas pipe network leakage positioning system, comprising: the system comprises a sensing system, a data transmission system and a leakage positioning system;
the sensing system is a pressure sensor, a flow sensor and a temperature sensor which are arranged in the municipal gas pipe network system and is used for monitoring the pressure, the flow and the temperature of an inlet and an outlet of the municipal gas pipe network, a user node and a key non-user node in real time;
the data transmission system is used for transmitting the pressure, flow and temperature data measured by the sensing system to the leakage positioning system;
the leakage localization system comprises:
the data storage module is used for storing pressure, flow and temperature data measured in the sensing system and leakage position data;
the numerical simulation module is used for constructing a hydraulics and thermodynamics model of the municipal gas pipe network system and obtaining the flow, pressure and temperature values of each node of the municipal gas pipe network by a numerical analysis method;
the leakage identification module is used for determining a leakage early warning value of the municipal gas pipe network by using the municipal gas pipe network flow, pressure and temperature data obtained from the sensing system, and realizing the identification of a newly added leakage point according to the leakage early warning value and real-time flow monitoring data of an inlet, an outlet and a user node in the municipal gas pipe network system;
the leakage position artificial intelligence positioning module is used for training an artificial intelligence model by utilizing pressure, flow and temperature data measured by the sensing system, historical leakage position data of the municipal gas pipe network and simulated leakage position data obtained by the numerical simulation module, wherein part of data is randomly selected as a training set, and the rest of data is used as a test set, so that an artificial intelligence municipal gas pipe network leakage position positioning model is obtained, and the municipal gas pipe network leakage position is determined by adopting the artificial intelligence municipal gas pipe network leakage position positioning model;
and the leakage position output module outputs the leakage position of the municipal gas pipe network.
8. The system of claim 7, further comprising a model correction module for correcting the hydraulic and thermodynamic models of the municipal gas pipe network system in real time by using the municipal gas pipe network flow, pressure and temperature data obtained from the sensing system.
9. The system of claim 7, further comprising:
the leakage point positioning method selection module is used for selecting a leakage position numerical value positioning module or a leakage position artificial intelligence positioning module to position the leakage point according to the recorded real leakage event number;
and the leakage position numerical value positioning module is used for positioning the leakage position of the municipal gas pipe network by adopting a model in the numerical simulation module and a numerical analysis method.
10. A system according to any of claims 7-9, characterized in that the system is arranged to implement the method as claimed in any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211060851.6A CN115127036B (en) | 2022-09-01 | 2022-09-01 | Municipal gas pipe network leakage positioning method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211060851.6A CN115127036B (en) | 2022-09-01 | 2022-09-01 | Municipal gas pipe network leakage positioning method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115127036A CN115127036A (en) | 2022-09-30 |
| CN115127036B true CN115127036B (en) | 2022-11-08 |
Family
ID=83387289
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211060851.6A Active CN115127036B (en) | 2022-09-01 | 2022-09-01 | Municipal gas pipe network leakage positioning method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115127036B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116906839B (en) * | 2023-09-14 | 2023-12-01 | 浙江英集动力科技有限公司 | Safety intelligent monitoring and early warning method for thermodynamic pipeline integrating physical measurement and soft measurement |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101625071B (en) * | 2009-08-07 | 2012-11-28 | 天津大学 | Method for measuring and locating leakage of gas pipelines |
| WO2018097746A1 (en) * | 2016-11-28 | 2018-05-31 | Siemens Aktiengesellschaft | Method for detecting abnormal state of a fluid supply network based on pressure pattern analysis |
| CN107420743B (en) * | 2017-06-09 | 2023-06-13 | 中国计量大学 | Intelligent urban gas PE pipe network measurement and control system and measurement and control method |
| WO2020136475A1 (en) * | 2018-12-27 | 2020-07-02 | Atlas Copco Airpower, Naamloze Vennootschap | Method for detecting leaks in a gas network under pressure or under vacuum and gas network |
| CN111520615B (en) * | 2020-04-28 | 2021-03-16 | 清华大学 | Pipe network leakage identification and positioning method based on line spectrum pair and cubic interpolation search |
| CN112610903A (en) * | 2020-12-10 | 2021-04-06 | 合肥学院 | Water supply pipe network leakage positioning method based on deep neural network model |
| CN113139353B (en) * | 2021-05-11 | 2023-09-29 | 东北大学 | Steam pipe network dynamic calculation and online monitoring early warning analysis method |
| CN114048694B (en) * | 2021-11-09 | 2024-03-29 | 中船第九设计研究院工程有限公司 | Leakage positioning method based on Internet of things and hydraulic model |
| CN114352947B (en) * | 2021-12-08 | 2024-03-12 | 天翼物联科技有限公司 | Gas pipeline leakage detection method, system, device and storage medium |
-
2022
- 2022-09-01 CN CN202211060851.6A patent/CN115127036B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115127036A (en) | 2022-09-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107420743B (en) | Intelligent urban gas PE pipe network measurement and control system and measurement and control method | |
| CN100492233C (en) | Gas pipe network accidents early warning method | |
| CN115685787B (en) | Modeling simulation method, modeling simulation system and medium for urban gas pipe network | |
| CN115127037B (en) | Water supply pipe network leakage positioning method and system | |
| CN102235575A (en) | Data processing method and system for checking pipeline leakage | |
| CN110332465A (en) | A kind of long distance gas pipeline leakage monitoring decision-making technique and system | |
| CN110245411A (en) | A kind of cities and towns central heating system and leakage point detecting method | |
| CN113139353A (en) | Dynamic calculation and online monitoring early warning analysis method for steam pipe network | |
| CN111486930B (en) | Natural gas energy metering assignment checking method and system | |
| CN105757459B (en) | A kind of gas drainage pipe network parameter monitoring system and leak source accurate positioning method | |
| CN107013812B (en) | A kind of THM coupling line leakage method | |
| CN115127036B (en) | Municipal gas pipe network leakage positioning method and system | |
| CN105868444A (en) | Feature image based gas pipeline network leakage detection and positioning method | |
| CN112985713A (en) | Pipe network leakage monitoring method and system based on edge calculation | |
| WO2023284060A1 (en) | Geographic three-dimensional information-based method for analyzing uncertainty of flow rate of sewage pipe network | |
| CN112884250B (en) | Intelligent gas management system and method | |
| CN111706785A (en) | Natural gas dendritic pipe network leakage pipe section identification method | |
| CN100480662C (en) | Gas leakage automatic detecting method | |
| CN115493093A (en) | Steam heating pipe network leakage positioning method and system based on mechanical simulation | |
| CN115879324B (en) | Simulation method, simulation platform and medium for urban gas multi-stage pipe network | |
| CN110543719B (en) | Water supply pipeline leakage prediction method based on space metering model | |
| Zhou et al. | Risk index assessment for urban natural gas pipeline leakage based on artificial neural network | |
| CN116796473A (en) | Gas pipe network monitoring point position layout method based on improved greedy algorithm | |
| CN110657352B (en) | Gas pipeline monitoring and stationing optimization method, system, equipment and storage medium | |
| CN112664839A (en) | Method and system for predicting and tracing combustible gas diffusion of communication pipeline |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |