Virtual sensor-based monitoring method and system for leakage of thermal recovery medium delivery pipeline for heavy oil

Abstract

The present application provides a virtual sensor-based heavy oil thermal recovery medium pipeline leakage monitoring method and system. By acquiring and transmitting data of operating characteristic parameters such as pressure, temperature, and flow rate at the inlet and outlet of the pipeline, and processing the data, a software-based leakage detection algorithm model is used to solve the problem. By comparing the simulation results with the field data, the operating state and leakage parameters of the pipe manifold are determined. The present application has the beneficial effect of being suitable for real-time monitoring and routine maintenance of the pipeline leakage state of offshore thermal recovery platforms, overcoming the limitations of space and extreme environments. By using system simulation, testing, and other means, common normal conditions such as steam injection, valve opening and closing, and leakage are simulated. By setting relevant operating conditions in the test section and conducting test tests and data verification, dynamic simulation of valve opening and closing, injection, and leakage with saturated steam and superheated steam as the working medium is achieved. By studying the operating condition changes such as valve differential pressure opening and closing, the overall performance of the heavy oil thermal recovery medium leakage detection system and the accuracy of the virtual sensor model are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of heavy oil thermal recovery medium pipeline leakage monitoring, more particularly to a heavy oil thermal recovery medium pipeline leakage monitoring method and system based on virtual sensing. BACKGROUND

[0002] The content of asphalt gum in heavy oil is high, and the content of wax is low, which causes high viscosity, difficult flow, and great difficulty in exploitation. As the main means of heavy oil development, heavy oil thermal recovery has been widely used in domestic and foreign heavy oil reservoir development. Steam stimulation is the main method of heavy oil recovery, which injects a certain amount of steam into the oil well, closes the well for a period of time, and then opens the well for production to improve heavy oil production after the heat of steam diffuses to the oil layer. Compared with land oilfield steam stimulation thermal recovery, offshore thermal recovery has the characteristics of small space, high injection temperature and high injection pressure. Once high-temperature and high-pressure steam leaks, it will cause serious consequences and threaten the inspectors. Therefore, by studying the high-temperature and high-pressure steam pipeline leakage monitoring technology, developing the leakage monitoring method and system, and improving the safety guarantee level of unconventional oil and gas development.

[0003] The heavy oil thermal recovery medium has high temperature and high pressure characteristics. For high-temperature and high-pressure steam pipeline leakage detection, the detection methods are generally divided into hardware-based and software-based detection methods.

[0004] The existing steam leakage detection methods have many problems: the hardware-based detection methods such as infrared thermal imaging and optical fiber detection have high cost, and cannot quantify the leakage amount. At the same time, due to the environmental and spatial limitations of the thermal recovery platform, the equipment is not applicable or the precision is easily affected; the software-based detection method is mainly used for liquids, natural gas and other fluids, and is generally suitable for long-distance pipelines. It is less used for high-temperature and high-pressure media, and due to the unstable nature of steam, it is greatly affected by temperature, resulting in measurement deviation and uncertainty.

[0005] The existing pipeline leakage monitoring method and system mainly rely on related hardware monitoring equipment or sensor systems to obtain real-time pipeline operation data, and simulate and solve the characteristic parameters of the pipeline internal operation. However, the heavy oil thermal recovery steam pipeline is affected by various factors such as working condition changes, environmental temperature, and operating pressure during operation, and has a large error in leakage monitoring. Most detection equipment has a relatively slow calculation speed for steam pipeline related calculations, and has poor real-time performance. It is not completely applicable to complex pipeline systems, variable working conditions, and pipeline operation. SUMMARY

[0006] The present application overcomes the deficiencies in the prior art and provides a heavy oil thermal recovery medium pipeline leakage monitoring method and system based on virtual sensing.

[0007] The purpose of the present application is achieved by the following technical solutions.

[0008] The virtual sensor-based heavy oil thermal recovery medium conveying pipeline leakage monitoring method is a software-based leakage method, fully considers the changes of fluid flow parameters, has high model simulation output precision, fast response of equipment state identification, and low equipment cost, and specific steps include:

[0009] Step 1, deploy sensors in the collection pipeline, transmit the data collected by the sensors to the remote host computer through remote communication means, and form the operation and verification data set of the equipment physical properties and related characteristic parameters;

[0010] Step 2, combine the fluid transient and thermal models, and establish a virtual pipeline model according to the changes of the velocity, pressure, density and viscosity parameters of the fluid in the pipeline;

[0011] Step 3, solve the related data of the flow field in the pipeline under the boundary conditions of the field using the virtual pipeline model established in step 2, and construct the field pipeline operation data set and the virtual pipeline simulation data set using the collected data and the simulation data,

[0012] The field pipeline operation data set is composed of steam pipeline operation data, and the operation data is collected by the deployed sensors and realized data transmission, and the sensors are deployed at the inlet and outlet of the pipeline to monitor the inlet and outlet pressure, temperature, velocity parameter data;

[0013] The virtual pipeline simulation data set includes a steam pipeline transportation data set and a steam pipeline leakage data set, wherein the steam pipeline transportation data set is constructed according to the parameter data calculated by the pipeline steam medium flow model, and the steam pipeline leakage data set is constructed according to the parameter data calculated by the pipeline steam leakage model;

[0014] Compare the data of the field pipeline operation data set and the virtual pipeline simulation data set, compare the calculated value with the verification data precision, and complete the leakage identification, leakage estimation and leakage early warning;

[0015] The leakage identification is performed by correlating the field data of the known equipment working condition with the model simulation calculation results of the corresponding working condition to determine the equipment normal operation threshold, and the leakage identification is determined by comparing the real-time simulation model calculation results with the leakage threshold;

[0016] The leakage estimation is determined by the difference between the mass flow rate difference of the pipeline inlet and outlet and the leakage threshold;

[0017] The leakage warning is performed by judging whether the absolute value of the mass flow rate difference of the real-time simulation results and the difference value of the leakage threshold are greater than zero to output the leakage alarm, and the leakage estimation value is calculated and the subsequent result is displayed;

[0018] The model mainly comprises: a pipeline virtual model is a pipeline steam medium flow model, a continuity equation of steam flow of the pipeline steam medium flow model is,

[0019]

[0020] wherein,

[0021] ρ - gas density, kg / m 3 ;

[0022] x - distance of the gas along the pipeline flow direction, m;

[0023] w - flow velocity of the gas in the pipeline, m / s;

[0024] A - flow area of the pipeline cross section, m 2 ;

[0025] A momentum equation of steam flow of the pipeline steam medium flow model is,

[0026]

[0027] wherein, p - gas pressure, pa; D - pipeline inner diameter, m; λ - friction resistance coefficient;

[0028] An energy equation of steam flow of the pipeline steam medium flow model is,

[0029]

[0030] wherein, h - enthalpy value of the gas, J / kg; - heat exchange rate of unit mass flow of the gas on a unit pipeline length, is W / m 2 .

[0031] A solving method of the pipeline steam medium flow model is based on finite volume and finite difference method, partial differential terms in the model are discretized, through numerical discrete formula of high-precision format combined with corresponding restrictor, based on the coefficient nonlinear characteristics of the momentum equation, the model equation is solved by using group implicit iteration, according to the inner iteration of node coefficient equation and the outer iteration of equation group, simulation data in the pipeline and time dimension are obtained according to the residual error requirements before and after iteration, the data is mapped with real running data, the relationship between the virtual pipeline and the real pipeline under normal operation is obtained, which provides support for leakage discrimination and early warning.

[0032] The steam pipeline leakage data set in the step 3 is calculated and judged by using a pipeline steam leakage model, and an equation of the pipeline steam leakage model is,

[0033]

[0034] wherein,

[0035] u - internal energy of the gas, J / kg;

[0036] h - enthalpy of the gas, J / kg;

[0037] p - density of the gas, kg / m 3 ;

[0038] p - pressure of the gas, Pa;

[0039] w - flow velocity of the gas in the pipe, m / s;

[0040] t - time, s;

[0041] x - distance of the gas along the flow direction of the pipe, m

[0042] A - flow area of the pipe cross-section, m 2 ;

[0043] g - - acceleration due to gravity, m / s 2 ;

[0044] l - friction resistance coefficient;

[0045] D - inner diameter of the pipe, m;

[0046] - heat exchange rate of the gas per unit mass flow per unit pipe length, with the unit of W / m 2 ;

[0047] M 泄露 - mass leakage rate at the leakage position, with the unit of kg / (m·s);

[0048] Mv 泄露 - momentum leakage rate at the leakage position, with the unit of N / m 3 ;

[0049] E 泄露 - energy leakage rate at the leakage position, with the unit of W / m.

[0050] The calculation process of the pipe steam leakage model specifically comprises:

[0051] S1, input initial conditions;

[0052] S2, set fault conditions;

[0053] S3, read fault parameters;

[0054] S4, bring the collected pressure, temperature and flow data at the inlet and outlet into the pipe steam leakage model for calculation;

[0055] S5, residual error judgment is performed on the calculation result, and when the condition is met, the calculation is ended, and when the condition is not met, the step S3 is returned to and recalculation is performed.

[0056] The step 3 is compared with the leakage early warning algorithm, the leakage early warning algorithm is determined by comparing the inlet and outlet data of the field through the calculation of the inlet and outlet flow of the transient result under different working conditions, the preliminary judgment is performed on the simulation data of the inlet and outlet combined with the pressure point analysis method and the mass balance method, when the leakage occurs, the threshold of the leakage early warning is accurately processed according to the specific inlet and outlet parameters and the analysis and comparison of the field and simulation data, the model is established through the data statistical method in the leakage identification and leakage estimation of various methods, and the establishment of the leakage early warning algorithm is realized.

[0057] The system for monitoring leakage of a heavy oil thermal recovery medium conveying pipeline based on virtual sensing includes:

[0058] A sensor system is installed on the pipeline system, is arranged at the positions of the inlet and outlet of the pipeline and the positions of the change of the pipe diameter according to the relevant model of the fluid mechanics theory to collect the pipeline data, and includes a pressure sensor, a temperature sensor and a flow sensor for collecting corresponding data information;

[0059] A data acquisition device includes a data processor and an infinite communication module, data transmission is performed between the sensor and the node controller through serial communication, the node controller is composed of a control chip and a peripheral circuit of the data processor, is responsible for signal acquisition and sending, realizes data acquisition and communication transmission of each node, and the data of the sensor is sent to the wireless communication module through the node controller, and the data is packaged by the wireless communication module and transmitted to the cloud server.

[0060] A leakage early warning system includes a pipeline steam medium flow model, a pipeline steam leakage model and a leakage monitoring and early warning algorithm, data simulation is performed by using the pipeline steam medium flow model, relevant data analysis is performed according to the field and simulation results, and leakage judgment, leakage early warning and related leakage estimation are performed through the pipeline steam leakage model and the leakage monitoring and early warning algorithm.

[0061] A man-machine interaction system is used for relevant operating personnel to acquire pipeline normal operation and leakage accident related information and perform leakage alarm processing, guide field personnel to take measures, and perform feature processing, noise reduction analysis and signal filtering on the data collected by the sensor, and realize effective processing of the collected data.

[0062] A database is used for storing and managing operation data, pipeline physical property data and simulation result data.

[0063] The beneficial effects of the present application are as follows:

[0064] 1、The virtual sensing leakage monitoring system of the present application has lower cost than the conventional physical sensing detection equipment, is more suitable for development and daily maintenance, can overcome the limitations of space and extreme environment, and is especially suitable for the insufficient space, high temperature and corrosive environment of offshore thermal recovery platforms. The virtual sensing scheme can be redesigned according to requirements and is convenient to use, while the physical sensing means such as infrared detection needs to be repositioned through mechanical intervention.

[0065] 2、In the construction of the virtual sensing model, considering the variable operation condition of the steam injection process, the present application uses system simulation, test and other means to simulate the common normal conditions (steam injection, valve opening and closing, well blanking), and through the setting of relevant working condition components in the test section, the valve opening and closing, injection and well blanking process system dynamic simulation of the saturated steam and superheated steam working medium can be realized. Through the study of the system parameter changes in the valve differential pressure opening and closing process, the overall performance design of the heavy oil thermal recovery medium leakage detection system is provided with reference, and the accuracy of the virtual sensing model is improved. BRIEF DESCRIPTION OF DRAWINGS

[0066] Figure 1 Fig. 1 is a structural schematic diagram of the heavy oil thermal recovery medium pipeline leakage monitoring system based on virtual sensing;

[0067] Figure 2 Fig. 3 is a calculation flowchart of the pipeline steam leakage model;

[0068] Figure 3 Fig. 5 is a flowchart of the leakage identification and leakage amount estimation process. DETAILED DESCRIPTION

[0069] The technical scheme of the present application will be further described below through specific examples.

[0070] Example

[0071] The present application realizes field data acquisition and transmission by installing a sensor system at the inlet and outlet of the monitored pipeline and connecting a data acquisition and transmission system module, and realizes data management in combination with a host computer database. The sensor system module is composed of electromagnetic flowmeters, pressure sensors and temperature sensors, and the sensors are arranged on both sides of the inlet and outlet of the pipeline. The specific sensor selection should meet the requirements of high temperature and high pressure of the thermal recovery medium, and the sampling response time of the sensor should be short, and the signal acquisition frequency and precision should be high to meet the high-precision model input of the subsequent model solving. The data acquisition and transmission system realizes the storage of data from the field sensor equipment to the database platform end of the host computer under the condition of meeting the maximum sampling frequency requirement of the sensor data, and realizes the collection of the running data of the medium conveying pipeline equipment in combination with the communication protocols such as MQTT and TCP / IP. The simulation and field equipment state analysis results are completed by the man-machine interaction front end deployed on the platform end, and the leakage early warning system is completed by the back-end programming framework and the database deployed on the host computer. The man-machine interaction front end mainly uses the Vue front-end framework to build a webpage technology to display the leakage monitoring situation, so that the relevant operating personnel can obtain and process the leakage alarm related information and take measures to guide the on-site personnel. The back-end data processing part mainly performs feature processing, noise reduction analysis and signal filtering on the sensor collected data to realize the effectiveness processing of the collected data. The model simulation module takes the data processing of the sensor collected data as the model input, determines the virtual pipeline model construction and boundary conditions and initial conditions, adopts a specific model solving method, calculates the simulation data results, compares them with the field data and verification data, determines the leakage situation and specific results, deduces the relationship between the monitoring parameters and performance parameters and the leakage under different leakage amounts and different working conditions, deduces the mathematical relationship between the monitoring parameters and the leakage, and considers the mutual influence of pressure and temperature of steam flowing in the pipeline in the model construction aspect, so that the constructed pipeline model can as accurately as possible describe the real flow state of steam in the pipeline and improve the performance of the monitoring system. The leakage early warning module analyzes the relevant data according to the field and simulation results, judges the leakage, gives a leakage warning and estimates the relevant leakage amount in combination with other leakage analysis methods. The database mainly stores and manages the relevant data such as running data, pipeline physical property data and simulation result data.

[0072] By deploying the pressure, temperature and flow sensors selected to meet the requirements at the inlet and outlet of the pipeline to be measured, realizing sensor data acquisition through the RS-485 conversion interface, realizing data exchange in combination with the Modbus-RTU protocol, realizing real-time data storage through the database system after data processing of the collected signals, and realizing real-time calculation through the normal pipeline simulation model and the leakage simulation model by the monitoring algorithm through the simulation calculation of the collected data in the back end, the state analysis results of the equipment and the relevant data are realized by analyzing and comparing the threshold values after the test calibration and the algorithm test.

[0073] The method for monitoring leakage in heavy oil thermal recovery media pipelines based on virtual sensing acquires, transmits, and processes operational characteristic parameters such as pressure, temperature, and flow rate at the pipeline inlet and outlet. Then, it solves the problem by inputting an algorithm model and comparing simulation results, verification data, and field data to determine leakage-related parameters.

[0074] like Figure 1 As shown, a leakage monitoring system for heavy oil thermal recovery medium transportation pipelines based on virtual sensing is presented. The monitoring system includes a sensor system, a data acquisition system, a database management system, a human-machine interface front-end, and a leakage early warning system. The specific functions of the virtual sensing-based monitoring method and system are implemented through data acquisition and transmission, and the construction and computational analysis of a virtual pipeline model.

[0075] Data acquisition and transmission are accomplished by a sensor system and a data acquisition and transmission system. The sensor system is installed on the pipeline system and positioned at the pipeline inlet and outlet locations and at points where the pipeline diameter changes, based on relevant models of fluid mechanics theory. The sensor system includes pressure sensors, temperature sensors, and flow sensors.

[0076] Based on the high temperature and high pressure requirements of the steam medium, a high-temperature pressure sensor with a wide pressure range, wide temperature range, high stability, and strong wear resistance, impact resistance, and corrosion resistance was selected. The sensor is equipped with a communication protocol interface, enabling long-distance transmission, and has strong anti-interference capabilities and a high signal transmission rate during operation.

[0077] The temperature sensor selected is an explosion-proof digital display thermal temperature sensor with an operating range of -200 to 500℃, which meets the requirements for measuring the temperature of thermal mining media. The built-in communication module ensures data communication.

[0078] The flow sensor serves as the initial basis for leakage detection in subsequent algorithm models. It employs a high-temperature resistant, high-precision, high-sampling-frequency digital mass flow controller with preliminary flow reading capabilities. It features short warm-up time, low zero drift, high reliability, communication interface, and data storage functions. All three sensors utilize a unified communication protocol.

[0079] The installation positions of the aforementioned sensors are designed based on specific operating requirements. Generally, piping systems must meet pressure design and operational requirements at the pipe inlet. Pressure sensors are installed slightly forward to facilitate operating condition verification and confirmation of changes in operating conditions. The remaining sensors are located behind the pressure sensors for real-time data acquisition and transmission.

[0080] While meeting the maximum sampling frequency requirement of sensor data, the system combines communication protocols such as MQTT and TCP / IP to complete the data transfer from the field sensor devices to the database storage on the host computer platform, thereby realizing the collection of operating data from the media transportation pipeline equipment.

[0081] The data acquisition and transmission system is composed of a control chip, a basic circuit, a communication interface module circuit for receiving sensor signals, and a wireless communication module circuit for transmitting data. The control chip has abundant interfaces and can meet the industrial demand under low power consumption. The sensor transmits data between the serial port communication and the node controller. The node controller is composed of a control chip and peripheral circuits, responsible for signal acquisition and transmission, used to realize data acquisition and communication transmission of pressure and other data of each node.

[0082] The bus network is realized by the RS-485 conversion interface between each sensor, and the pressure, temperature, and flow data are collected by the sensor system. The node controller exchanges data through the Modbus-RTU protocol, and then sends the data to the wireless communication module through the RS-485 communication. Then the wireless communication module packs the collected data in JSON format, and transmits it to the cloud server through the MQTT transmission protocol. The storage and management of data are completed by the database system.

[0083] The running condition of the system is displayed by the man-machine interaction front end. The leakage monitoring condition on site is completed by the man-machine interaction front end deployed on the platform end, the leakage early warning system back end, and the database for data management. The man-machine interaction front end mainly uses the Vue front-end framework to build web technology to display the leakage monitoring condition, so that relevant operating personnel can obtain normal operation and leakage accident related information of the pipeline and handle the leakage alarm, guiding the on-site personnel to take measures. The back-end data processing part mainly processes the collected data of the sensor, analyzes the noise reduction, and filters the signal, etc., to realize the effectiveness of the collected data.

[0084] As shown in Figures 2-3 The leakage early warning system performs data simulation on the pipeline steam medium flow model through the pipeline steam medium flow model, and the calculation data is used as the operation basis of the pipeline steam leakage model. The pipeline steam leakage model operates according to the data collected by the sensor, and judges whether there is leakage through the residual error of the data. The simulation pipeline simulation data set is formed by the operation data of the pipeline steam medium flow model and the pipeline steam leakage model, and the simulation pipeline simulation data set and the on-site pipeline operation data set are brought into the leakage early warning algorithm to complete the work of leakage identification, leakage calculation, and leakage early warning.

[0085] The embodiments of the present application are described in detail above, but the content described is only the preferred embodiments of the present application, and cannot be considered as limiting the scope of the implementation of the present application. Any equivalent changes and improvements made within the scope of the present application should still belong to the patent coverage of the present application.

Claims

1. A method for monitoring leakage in heavy oil thermal recovery medium transportation pipelines based on virtual sensing, characterized in that, The software-based leakage method fully considers changes in fluid flow parameters, and the specific steps include: Step 1: Deploy sensors in the acquisition pipeline and transmit the data collected by the sensors to a remote host computer through remote communication to form an operational and verification dataset of equipment physical properties and related characteristic parameters; Step 2: Combining fluid transient and thermodynamic models, establish a virtual pipeline model based on the changes in the velocity, pressure, density, and viscosity parameters of the fluid within the pipeline; Step 3: Using the virtual pipeline model established in Step 2, solve for the relevant data of the flow field inside the pipeline under the boundary conditions of the site. Construct a field pipeline operation dataset and a virtual pipeline simulation dataset using the collected data and simulation data. The on-site pipeline operation dataset consists of steam pipeline operation data. The operation data is collected and transmitted through deployed sensors. The sensors are deployed at the pipeline inlet and outlet to focus on monitoring the inlet and outlet pressure, temperature, and velocity parameters. The virtual pipeline simulation dataset includes a steam pipeline transportation dataset and a steam pipeline leakage dataset. The steam pipeline transportation dataset is constructed based on the parameter data calculated by the pipeline steam medium flow model, and the steam pipeline leakage dataset is constructed based on the parameter data calculated by the pipeline steam leakage model. By comparing the data from the on-site pipeline operation dataset and the virtual pipeline simulation dataset, and comparing the accuracy of the calculated values ​​with the verification data, leakage identification, leakage estimation, and leakage early warning can be completed. Leakage identification involves correlating field data of known equipment operating conditions with model simulation calculation results for the corresponding operating conditions to determine the normal operating threshold of the equipment. Leakage identification is determined by comparing the calculation results of the real-time simulation model with the leakage threshold. Leakage is estimated by the difference between the mass flow rate difference at the pipeline inlet and outlet and the leakage threshold. The leakage early warning system outputs a leakage alarm by judging whether the absolute value of the difference between the inlet and outlet mass flow rates in the real-time simulation results and the leakage threshold is greater than zero. It also calculates the estimated leakage amount and displays the subsequent results. The model mainly includes: a virtual pipeline model for steam medium flow, and the continuity equation for steam flow in the pipeline steam medium flow model is as follows: In the formula, ρ – gas density, kg / m³ 3 ; x - the distance of gas flow along the pipe direction, in meters; w – the velocity of the gas in the pipe, m / s; A – Flow area of ​​the pipe cross-section, m² 2 ; The momentum equation for steam flow in the pipeline steam medium flow model is as follows: In the formula, p – gas pressure, pa; D – pipe inner diameter, m; λ – friction resistance coefficient; The energy equation for steam flow in the pipeline steam medium flow model is as follows: In the formula, h – enthalpy of the gas, J / kg; – The heat exchange rate of gas per unit mass flow rate per unit pipe length, in W / m 2 .

2. The method for monitoring leakage in heavy oil thermal recovery medium transportation pipelines based on virtual sensing according to claim 1, characterized in that, include: The solution method for the pipeline steam medium flow model is based on the finite volume and finite difference methods. The partial differential terms in the model are discretized, and the model equations are solved by using a high-precision numerical discretization formula combined with appropriate limiters. Based on the nonlinear characteristics of the momentum equation coefficients, the model equations are solved by a set of implicit iterations. According to the internal iteration of the nodal coefficient equations and the external iteration of the equation set, the simulation data of the pipeline and the time dimension are obtained according to the residual requirements before and after iteration. This data is mapped with the actual operation data to obtain the relationship between the virtual pipeline and the actual pipeline under normal operation, providing support for leakage detection and early warning.

3. The method for monitoring leakage in heavy oil thermal recovery medium transportation pipelines based on virtual sensing according to claim 1, characterized in that, include: In step 3, the steam pipeline leakage dataset is calculated and judged using a pipeline steam leakage model. The equation for the pipeline steam leakage model is as follows: In the formula: u - Internal energy of the gas, J / kg; h - Enthalpy of the gas, J / kg; ρ - gas density, kg / m³ 3 ; p - gas pressure, Pa; w - the velocity of the gas in the pipe, m / s; t represents time, in seconds; x - the distance of gas flow along the pipe direction, in meters; A - Flow area of ​​the pipe cross-section, m² 2 ; g - acceleration due to gravity, m / s 2 ; λ - coefficient of frictional resistance; D - Pipe inner diameter, in meters; – The heat exchange rate of gas per unit mass flow rate per unit pipe length, expressed in W / m. 2 ; M 泄露 - The mass leakage rate at the leakage location, in kg / (m·s); Mv 泄露 - Momentum leakage rate at the leak location, in N / m 3 ; E 泄露 - Energy leakage rate at the leak location, in W / m.

4. The method for monitoring leakage in heavy oil thermal recovery medium transportation pipelines based on virtual sensing according to claim 3, characterized in that, The calculation process of the pipeline steam leakage model specifically includes: S1. Input initial conditions; S2. Set fault conditions; S3, Read in fault parameters; S4. Input the pressure, temperature and flow data collected at the inlet and outlet into the pipeline steam leakage model for calculation; S5. Perform residual judgment on the calculation results. If the conditions are met, the calculation ends. If the conditions are not met, return to step S3 and recalculate.

5. The method for monitoring leakage in heavy oil thermal recovery medium transportation pipelines based on virtual sensing according to claim 1, characterized in that: In step 3, the on-site pipeline operation dataset and the virtual pipeline simulation dataset are compared using a leakage early warning algorithm. The leakage early warning algorithm calculates the inlet and outlet flow rates based on the transient results under different operating conditions, compares the on-site inlet and outlet data to make a preliminary early warning, and makes a preliminary judgment on the inlet and outlet simulation data by combining the pressure point analysis method and the mass balance method. When a leak is determined to have occurred, the leakage early warning threshold is refined based on the specific inlet and outlet parameters and the analysis and comparison of on-site and simulation data. The leakage identification and leakage estimation methods of various methods are modeled using data statistical methods to realize the establishment of the leakage early warning algorithm.

6. A system for monitoring leaks in heavy oil thermal recovery medium transportation pipelines based on virtual sensing, used to execute the method for monitoring leaks in heavy oil thermal recovery medium transportation pipelines based on virtual sensing as described in any one of claims 1 to 5, characterized in that, include: The sensor system, installed on the pipeline system, is set at the inlet and outlet positions of the pipeline and at the locations where the pipeline diameter changes, based on relevant models of fluid mechanics theory, to collect pipeline data. The sensor system includes pressure sensors, temperature sensors and flow sensors to collect corresponding data information. The data acquisition device includes a data processor and a wireless communication module. The sensor transmits data to the node controller via serial communication. The node controller consists of the control chip of the data processor and peripheral circuits. It is responsible for signal acquisition and transmission, realizing data acquisition and communication transmission of each node. The sensor data is sent to the wireless communication module via the node controller. The wireless communication module packages the data and transmits it to the cloud server. The leakage early warning system includes a pipeline steam medium flow model, a pipeline steam leakage model, and a leakage monitoring and early warning algorithm. It uses the pipeline steam medium flow model for data simulation, performs relevant data analysis based on the field and simulation results, and uses the pipeline steam leakage model and leakage monitoring and early warning algorithm to determine leakage, provide early warning of leakage, and estimate the relevant leakage amount. The human-machine interaction system is used to enable relevant operators to obtain information related to normal pipeline operation and leakage accidents, handle leakage alarms, guide on-site personnel to take measures, and perform feature processing, noise reduction analysis, and signal filtering on sensor-collected data to achieve effective processing of the collected data. The database is used for storing and managing operational data, pipeline property data, simulation results data, and related data.

Images