Gas-liquid two-phase discharge amount intelligent monitoring method based on multi-plane three-dimensional reconstruction

Abstract

The invention relates to a gas-liquid two-phase discharge technology in the technical field of petrochemical engineering safety engineering, in particular to a gas-liquid two-phase discharge amount intelligent monitoring method based on multi-plane three-dimensional reconstruction. The method comprises the following steps: a gas-liquid two-phase discharge fluid enters an observation area; intercepting a plurality of two-dimensional fluid plane areas in the observation area; acquiring two-dimensional information fields of gas and liquid phases in different two-dimensional fluid plane areas; and according to different two-dimensional information fields, reconstructing a gas-liquid two-phase three-dimensional information field in the whole observation area, and calculating to obtain the total discharge amount of the whole fluid domain. Dynamic distribution images of two-phase fluid are captured by means of a high-speed camera, deep analysis of image data is carried out by combining intelligent detection software, meanwhile, a three-dimensional coupling reconstruction technology and system integration of the data are applied, and finally the volume quantities of a gas phase and a liquid phase discharged from a tank are obtained respectively. And a key data support is provided for material balance analysis, process optimization and safety evaluation in the discharge process.

Description

Technical Field

[0001] The present invention relates to the gas-liquid two-phase relief technology in the technical field of petrochemical safety engineering. Specifically, it relates to an intelligent monitoring method for gas-liquid two-phase relief quantity based on multi-plane three-dimensional reconstruction. Background Art

[0002] At present, the safety situation of hazardous chemical production presents a severe and complex trend. Due to the flammable, explosive, toxic, and harmful characteristics of hazardous chemicals themselves, once accidents such as leakage occur during their production process, it is very likely to trigger catastrophic consequences such as fires, explosions, and poisonings. This will not only threaten the lives of enterprise employees but also pose a serious threat to the lives and health of surrounding community residents. At the same time, it may cause irreversible damage to the ecological environment such as the atmosphere, soil, and water bodies, bringing huge property losses and social negative impacts.

[0003] In this context, the analysis of the relief quantity is very necessary. The current existing technologies include:

[0004] Chinese invention patent CN118881954B discloses a gas safety relief device for chemical equipment, which can achieve the safe relief of chemical gases.

[0005] Chinese invention patent CN110812871B discloses a safety relief device, which is suitable for the emergency relief of dangerous materials in the accident state of an oxidation reactor during the production of phenol and acetone.

[0006] Chinese invention patent CN117455223A discloses a safety relief device for tower equipment, which can prevent safety accidents caused by insufficient relief area of the safety valve of the tower equipment under overpressure conditions.

[0007] In industrial production processes such as petrochemical industry, non-ambient temperature and pressure conditions are often involved. Especially when materials are discharged under high pressure, flashing is very likely to occur. Flashing refers to the phenomenon that when a liquid in a high-pressure state suddenly reduces in pressure, part of the liquid quickly vaporizes, thus forming a complex flow state with coexistence of gas and liquid phases. In this case, the traditional method of using a mass flowmeter to monitor the gaseous relief quantity faces severe challenges and is even no longer applicable. This is because the design and measurement principle of the mass flowmeter are often based on the stable flow characteristics of single-phase gas. When gas and liquid phases coexist, the internal flow state is complex and changeable, and the proportion, distribution, flow velocity, and interaction of gas and liquid phases are constantly changing. The presence of gas and liquid phases will interfere with the measuring elements of the mass flowmeter, making it unable to accurately sense the true flow rate and mass of the gas.

[0008] For example, the presence of liquid may cause a change in the resistance within the measurement pipeline, resulting in pressure fluctuations, which in turn affect the measurement accuracy of the flowmeter. The interfacial slip and turbulence phenomena between gas and liquid phases will also make the measurement signals unstable and unreliable.

[0009] In addition, the physical properties such as density and viscosity of gas-liquid two-phase are significantly different from those of single-phase gas, further increasing the difficulty of accurately measuring the gaseous discharge volume using traditional mass flowmeters. Therefore, under the working conditions of flash evaporation caused by high-pressure discharge and coexistence of gas-liquid two-phase, it is urgent to explore and apply new monitoring technologies and methods to achieve accurate monitoring of the gas-liquid two-phase discharge volume and ensure the safe and stable operation of industrial production. Summary of the Invention

[0010] The purpose of the present invention is to provide an intelligent monitoring method for gas-liquid two-phase discharge volume based on multi-plane three-dimensional reconstruction, so as to solve the problem of difficult monitoring of two-phase discharge volume caused by the coexistence of gas and liquid due to material flash evaporation under non-normal pressure conditions. The idea of three-dimensional reconstruction is introduced into the spatial domain, combined with a high-speed camera system and intelligent monitoring and recognition software, expanding from two-dimensional single-plane monitoring to multi-plane monitoring of three-dimensional spatial domain, and overall control from single-phase gas discharge volume to gas-liquid two-phase discharge volume, ensuring the high performance and wide application of the system.

[0011] To achieve the above object, the method includes the following steps:

[0012] S1. Select a monitoring point on the reactor, and the PLC controller controls the safety relief valve to discharge according to parameters such as pressure and temperature at the monitoring point;

[0013] S2. Add an observation area behind the safety relief valve. After the safety relief valve discharges, the discharged fluid containing gas-liquid two-phase enters the observation area;

[0014] S3. Use a high-speed camera to capture images of the discharged fluid in the observation area, adjust the light source to an appropriate brightness and angle, intercept multiple two-dimensional fluid plane areas in a variable focal length manner, and obtain real-time images of the droplet distribution on different position planes. The images will contain clearly visible droplets;

[0015] S4. Identify the droplets in the images through intelligent detection software, process the captured pictures using a computer, calculate information such as the size, position distribution, and movement speed of droplets at different positions on different planes, and reverse infer the gas flow rate based on the droplet movement speed to obtain the two-dimensional information field of gas-liquid two-phase on different monitoring planes;

[0016] S5. Reconstruct the three-dimensional information field of gas-liquid two-phase within the entire observed fluid domain according to the two-dimensional information fields on different position planes, and calculate the total discharge volume of the entire fluid domain;

[0017] S6. The two-phase fluid discharged is collected by a collection tank. According to the above method, the volumes of the gas phase and liquid phase discharged from the tank can be obtained.

[0018] Preferably, in S1, appropriate monitoring points should be selected from the key parts of the reactor, for example, in the areas where the reactor is prone to pressure and temperature fluctuations, such as the inlet and outlet, corners, and the core part with intense reactions. The PLC controller continuously collects key parameters such as pressure and temperature at each monitoring point and dynamically compares them with the preset safety thresholds and process parameters. Once the monitored parameters deviate from the normal range, the PLC controller immediately controls the opening of the safety relief valve for discharging according to the built-in control logic and algorithms, quickly adjusts the pressure and temperature in the reactor, and makes it return to the safe and stable operating range, thereby effectively ensuring the safe and efficient operation of the reactor.

[0019] Preferably, in S2, the space domain occupied by the gas-phase and liquid-phase fluids discharged by the safety relief valve is set as the observation area. It is necessary to accurately define and optimize the volume range of this observation area to ensure that it can completely cover the key flow areas of the discharged two-phase flow and meet the actual requirements of experimental measurement and data analysis. At the same time, in order to obtain more representative and valuable research data, it is necessary to ensure that the velocity distributions of the gas-liquid two phases in the observation area show significant gradient change characteristics on different monitored two-dimensional planes. By reasonably arranging the monitoring planes and optimizing the design of the observation area, the data error is minimized to provide an accurate data source for the subsequent shooting process and help improve the accuracy and precision of the entire monitoring method.

[0020] Preferably, in S3, the coordinated use of the high-speed camera and the light source is the key link to obtain high-quality observation data. To achieve the accurate capture of the gas-liquid doped fluid during the discharge process of the safety relief valve, it is necessary to systematically debug and optimize the equipment parameters before the fluid is discharged. Among them, the lighting angle of the light source needs to be accurately adjusted according to the spatial layout of the observation area and the fluid movement trajectory. Through multi-angle simulation tests, the best irradiation angle that can minimize shadow interference and enhance the reflection effect on the fluid surface is selected; the brightness of the light source needs to be finely adjusted according to the fluid characteristics and ambient light conditions to clearly present the interface characteristics of the gas-liquid two phases inside the fluid on the premise of avoiding overexposure or underexposure.

[0021] In terms of the parameter setting of the high-speed camera, it is necessary to reasonably set the frame rate of the high-speed camera according to the expected fluid motion speed and change details to be captured, ensuring that the dynamic evolution of the key moments during the fluid discharge process can be completely recorded. At the same time, by adjusting the focal length and focusing mode, the lens focus is accurately locked on the core area of the gas-liquid two-phase flow, guaranteeing the clarity and sharpness of the captured images. Through the optimized adjustment of the above parameters, finally, in the fluid images obtained by shooting, the droplets distributed at different positions are clearly distinguishable, and the spatial distribution and motion state of the gas-liquid two-phase can be completely presented, thus ensuring that the captured data has high effectiveness and practicality. These high-quality image data will lay a solid foundation for the subsequent image recognition and data analysis work based on intelligent software, effectively improving the accuracy and efficiency of fluid dynamics parameter measurement.

[0022] Preferably, in the step S4, the intelligent detection software relies on advanced digital image processing and analysis algorithms to deeply analyze and optimize the massive amount of original image data collected by the high-speed camera. The original images are converted into a standard image format that is clearer, more beautiful and conforms to the computer vision recognition specification. On this basis, with the help of the object detection algorithm based on deep learning, the droplets in the gas-liquid two-phase flow images are accurately identified and segmented to achieve the complete extraction of the droplet contours. Further, through the spatio-temporal correlation analysis of multiple frame image sequences, the software calculates the geometric sizes of the droplets at different positions in different two-dimensional planes under monitoring, establishes the two-dimensional information field of the droplets, and uses technologies such as particle image velocimetry (PIV) to accurately track the motion trajectories of the droplets, and then obtains the key dynamic parameters such as the motion speed and acceleration of the droplets at each moment, and infers the relevant dynamic parameters of the gas phase;

[0023] Preferably, in the step S5, through the in-depth excavation and integration of the gas-liquid two-phase flow information field data on multiple different monitored two-dimensional planes collected and processed in the early stage, the innovative use of advanced three-dimensional coupling reconstruction technology realizes the high-precision modeling and visual presentation of the discharge fluid spatial domain. This technology first performs coordinate mapping and spatial registration on the discrete information such as the gas-liquid flow velocity distribution, phase ratio, and droplet diameter on each two-dimensional plane, and through the construction of a multi-scale grid model and interpolation algorithm, the two-dimensional data points are orderly associated and supplemented in the three-dimensional space. Subsequently, based on the fluid mechanics continuity equation and conservation law, combined with the physical characteristics of the gas-liquid two-phase flow, the data in the three-dimensional space is dynamically coupled and analyzed to generate the three-dimensional information field distribution covering the entire discharge fluid spatial domain. This three-dimensional information field not only intuitively presents the dynamic distribution characteristics of the gas-liquid two-phase in space, but also accurately realizes the real-time monitoring and quantitative calculation of the total discharge amount of the gas-liquid two-phase through the spatial integral operation of parameters such as fluid density and velocity vector.

[0024] Preferably, in the step S6, the gas-liquid two-phase fluid generated during the discharge process is orderly introduced into the collection tank through a pipeline for centralized storage. Based on the three-dimensional reconstruction intelligent monitoring and analysis method described above, the system captures the dynamic distribution images of the two-phase fluid with a high-speed camera, combines the in-depth analysis of the image data by intelligent detection software, and at the same time uses the three-dimensional coupling reconstruction technology for systematic integration of the data. Finally, the volumes of the gas and liquid discharged from the tank are obtained respectively, providing key data support for the material balance analysis, process optimization, and safety assessment of the discharge process.

[0025] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0026] This intelligent monitoring method for the gas-liquid two-phase discharge volume based on multi-plane three-dimensional reconstruction, through the collaborative application of three-dimensional reconstruction technology, high-speed cameras, and intelligent detection and recognition software, reconstructs the two-dimensional information fields at different angles, different times, and different positions in three dimensions. By identifying and matching the same gas-liquid two-phase flow characteristic points in different photos, a three-dimensional spatial structure of the flow field is constructed, and the dynamic change process of the gas-liquid two-phase flow in the entire spatial basin is restored by combining time-series data. Through the analysis and calculation of the reconstructed model, the velocity vector of the liquid at each point in space, including the magnitude and direction of the velocity, can be accurately obtained. According to the principles of fluid mechanics, combined with parameters such as the density and flow rate of the gas-liquid two-phase, the velocity distribution of the gas phase is deduced, and then the gas-liquid two-phase discharge volume in the entire spatial basin is derived. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the gas-liquid two-phase monitoring process of the present invention;

[0028] Figure 2 It is a schematic diagram of the principle of the three-dimensional reconstruction technology of the present invention;

[0029] Figure 3 It is a schematic diagram of the multi-plane intelligent coupling monitoring image and analysis result of the present invention.

[0030] The meanings of the reference numerals in the figure are as follows:

[0031] 1, reactor; 2, monitoring point; 3, PLC controller; 4, safety relief valve; 5, observation area; 6, light source; 7, high-speed camera; 8, intelligent detection software; 9, collection tank. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0033] Since the area where the safety relief valve 4 sprays outwards is a three-dimensional space domain, during the relief process, the velocities of the fluid near and far from the safety relief valve 4 are unevenly distributed. Therefore, the ordinary method of using a high-speed camera 7 to capture the size, flow velocity, position distribution, etc. of the liquid phase on a two-dimensional plane is no longer applicable. The idea of performing three-dimensional reconstruction on the space domain, monitoring the distribution of gas-liquid two-phase at different positions, and then obtaining a more accurate relief volume value emerged.

[0034] The collaborative application of three-dimensional reconstruction technology, high-speed camera 7 and intelligent detection software 8 has opened up a new path for the dynamic monitoring of gas-liquid two-phase relief volume in the petrochemical field. With its extremely high shooting frame rate, the high-speed camera 7 can capture the subtle changes and complex forms of the gas-liquid two-phase jet flow in an extremely short time, quickly obtain a large number of high-definition photos containing the motion characteristics of the gas-liquid two-phase flow, record the dynamic information such as the morphological evolution and interface fluctuation of the jet flow from the start to the diffusion process, and provide a rich data basis for subsequent analysis. Based on deep learning algorithms and advanced image processing techniques, the intelligent detection software 8 can quickly analyze and process the massive image data collected by the high-speed camera 7. By identifying the droplets in the captured pictures, calculating their size, position distribution, motion speed and other information, and inversely deducing the speed of the gas phase based on the speed distribution of the liquid phase obtained from subsequent data processing, the two-dimensional information fields of the two phases are obtained respectively.

[0035] On this basis, three-dimensional reconstruction is performed on the two-dimensional information fields at different angles, different times and different positions. By identifying and matching the same gas-liquid two-phase flow feature points in different photos, the three-dimensional structure of the flow field in the three-dimensional space is constructed, and the dynamic change process of the gas-liquid two-phase flow in the entire space domain is restored by combining time series data. Through the analysis and calculation of the reconstructed model, the velocity vector of the liquid at each point in space, including the magnitude and direction of the velocity, can be accurately obtained. According to the principles of fluid mechanics, combined with parameters such as the density and flow rate of the gas-liquid two-phase, the velocity distribution of the gas phase is inversely deduced, and then the relief volume of the gas-liquid two-phase in the entire space domain is deduced.

[0036] Specifically, an intelligent monitoring method for gas-liquid two-phase relief volume based on multi-plane three-dimensional reconstruction, the technical process of this method includes:

[0037] Select the monitoring point 2 on the reactor 1. The PLC controller 3 controls the safety relief valve 4 to discharge according to parameters such as the pressure and temperature at the monitoring point 2. An observation area 5 is added behind the safety relief valve 4. After the safety relief valve 4 discharges, the discharged fluid containing gas-liquid two-phase enters the observation area 5. Use a high-speed camera 7 to capture the image of the discharged fluid in the observation area 5, and obtain images of different position planes in a variable focal length manner. The images will contain clearly visible liquid droplets. Identify the liquid droplets in the images through the intelligent detection software 8, and calculate information such as their size, position distribution, and movement speed. Then, infer the gas-phase flow rate based on the liquid droplet movement speed to obtain the two-dimensional information field of the gas-liquid two-phase. Reconstruct the three-dimensional information field of the gas-liquid two-phase in the entire observation area 5 according to the two-dimensional information fields of different position planes, and calculate the total discharge amount of the entire fluid domain. The discharged two-phase fluid is collected by the collection tank 9. According to the above method, the volume of gas and liquid discharged from the tank can be obtained respectively.

[0038] Figure 1 In the figure, the solid line represents the material flow, and the dashed line represents the data flow. During operation, the PLC controller 3 serves as the core hub. Based on the key parameters such as the pressure and temperature at the monitoring point 2 collected in real time, and according to the preset safety thresholds and control strategies, it accurately controls the safety relief valve 4 to perform the discharge operation. When the safety relief valve 4 is opened and the discharged fluid containing gas-liquid two-phase enters the observation area 5, the high-speed camera 7 is immediately started, and the movement form of the fluid is continuously and dynamically captured with the pre-optimized shooting parameters to ensure clear and detailed observation images are obtained.

[0039] It should be noted that the high-speed camera 7 and the light source 6 work together. Among them, the lighting angle of the light source 6 needs to be accurately adjusted according to the spatial layout of the observation area 5 and the fluid movement trajectory. Through multi-angle simulation tests, select the best irradiation angle that can minimize shadow interference and enhance the reflection effect on the fluid surface. The brightness of the light source 6 needs to be finely adjusted according to the fluid characteristics and ambient light conditions to ensure that the interface characteristics of the gas-liquid two-phase inside the fluid are clearly presented on the premise of avoiding overexposure or underexposure.

[0040] The intelligent detection software 8 deeply analyzes and integrates the massive image data collected by the camera by integrating advanced image processing algorithms and data analysis models. By using target recognition and particle tracking technologies, it accurately extracts information such as the distribution characteristics of the liquid phase and the movement trajectories of liquid droplets in the gas-liquid two-phase flow. By selecting multiple two-dimensional planes at different positions for monitoring and applying three-dimensional reconstruction, information on the two phases in the entire fluid domain is obtained, and finally the total discharge amount is obtained. The calculated discharge amount data is fed back to the PLC controller 3 in real time. Based on the feedback information and the preset discharge target, the PLC controller 3 dynamically corrects and adjusts the opening of the safety relief valve 4 using an adaptive control algorithm. Through a closed-loop control mechanism, precise regulation of the discharge amount is achieved, ensuring that the system operates under safe and efficient conditions and effectively improving the automation level and control accuracy of the entire discharge process.

[0041] According to Figure 3 The results shown, a distributed high-precision data processing flow is adopted to achieve accurate detection of the gas-liquid two-phase discharge amount. First, the high-speed camera 7 continuously shoots the discharged fluid at a preset high frame rate to capture the original image sequence containing the movement characteristics of the gas-liquid two-phase, providing basic data for subsequent analysis. Subsequently, the intelligent detection software 8 starts the image preprocessing program, using algorithms such as noise reduction filtering and contrast enhancement to eliminate image noise and interference and improve image clarity and data quality. In the two-dimensional plane analysis stage, the software selects a specific two-dimensional monitoring plane, performs time-series superposition on the processed images at different times on this plane, and uses the change in gray value to intuitively present the spatial position difference of the tracer particles. Through particle image velocimetry (PIV) technology, quantitative analysis of the gray distribution is carried out to accurately track the movement trajectories of the tracer particles, and then the velocity vector distribution of the liquid flow on this plane is obtained. To obtain the overall picture of the fluid movement in the three-dimensional space, the above processing method is extended to the entire spatial basin, and multiple two-dimensional planes are selected. Through the collaborative analysis of the data on each plane and using three-dimensional coupling reconstruction technology, the discrete two-dimensional information is spatially integrated to construct a three-dimensional distribution model containing parameters such as the velocity and concentration of the gas-liquid two-phase. Based on the continuity equation and momentum conservation principle of the gas-liquid two-phase flow, the intelligent detection software 8 uses the liquid velocity distribution data and combines the interaction relationship between the two phases to inversely calculate the gas velocity distribution. On this basis, through the spatial integration operation of the gas-liquid two-phase velocity field, the discharge amounts of the gas phase and the liquid phase are respectively obtained, and finally the high-precision detection result of the total gas-liquid two-phase discharge amount is obtained by superposition calculation, providing scientific and reliable data support for the dynamic monitoring and safety assessment of the discharge process.

[0042] Example 1

[0043] This example is applied to the emergency discharge system of the reactor 1 in a 200,000-ton / year synthetic ethylene oxide (EO) project.

[0044] Release of materials: The released materials are the raw materials and products for synthesizing ethylene oxide. The release pressure is 5 Mpa, the release temperature is 150 °C, the maximum release rate is 6300 kg / h, and the longest release time is 2 h;

[0045] Description of the release process of gas-liquid two-phase doping:

[0046] The inlet pipe diameter for material release is DN100, and the inlet for material release is set at the top of reactor 1. Monitoring sensors (constituting monitoring point 2) are evenly distributed around reactor 1. The size of reactor 1 is ψ1500 mm × 6000 mm. The monitoring sensors dynamically collect the status information of reactor 1 in real time and transmit it to PLC controller 3. PLC controller 3 analyzes and compares the information to identify abnormal states. If overpressure or abnormal temperature is detected, safety relief valve 4 is triggered. According to the results of identifying abnormal states and the calculated release rate, safety relief valve 4 is actively opened to release materials to achieve accurate control of the release rate. At the same time, high-speed camera 7 takes pictures of the monitored fluid domain during the release process, and intelligent detection software 8 performs secondary processing on the images. The images of particle flows at different times on the same plane are superimposed and sorted, and the images of different monitored two-dimensional planes are three-dimensionally reconstructed to obtain the data set information of gas-liquid two-phase in the entire spatial dimension. Finally, the released materials flow into collection tank 9 through a transparent pipe for collection and treatment. After treatment, the volume of gas and liquid phases in collection tank 9 can be known.

[0047] Example 2

[0048] As Figure 1 shown, this example is applied to the emergency release system of reactor 1 in a 400,000-ton / year ethylene glycol (EG) synthesis project.

[0049] Release of materials: The released materials are the raw materials and products for synthesizing ethylene glycol. The release pressure is 10 Mpa, the release temperature is 200 °C, the maximum release rate is 9600 kg / h, and the longest release time is 5 h;

[0050] Description of the process of intelligent monitoring of the release rate of gas-liquid two-phase doping:

[0051] The diameter of the material discharge inlet is DN230. The size of the reactor 1 is ψ2900mm×7500mm. The material discharge inlet is arranged at the top of the reactor 1. The monitoring sensors (constituting the monitoring point 2) are evenly distributed around and on the top of the reactor 1. The monitoring sensors dynamically collect the status information of the reactor 1 in real time and transmit it to the PLC controller 3. The PLC controller 3 analyzes and compares the information to identify abnormal states. If abnormal temperature and pressure are found, the safety relief valve 4 is triggered. According to the calculation result of the discharge amount, the safety relief valve 4 is actively opened and the opening degree of the safety relief valve 4 is adjusted to discharge the material. At the same time, the high-speed camera 7 shoots the discharge situation, and the intelligent detection software 8 performs secondary processing on the images, superimposes and arranges the images of the particle flow at different times on the same plane, and performs three-dimensional reconstruction on the images of different monitoring two-dimensional planes to obtain the data group information of the gas-liquid two-phase in the entire spatial dimension. Finally, the discharged material flows into the collection tank 9 through a transparent pipeline for collection and treatment. After treatment, the volume of the gas-liquid two-phase in the collection tank 9 can be known.

[0052] In the above embodiment, through technology integration, non-contact and high-precision dynamic monitoring of the gas-liquid two-phase discharge process is achieved. Compared with traditional monitoring means, it can not only provide more comprehensive and detailed flow field information, but also avoid the interference caused by the installation of sensors to the gas-liquid two-phase flow field, significantly improving the accuracy and reliability of the monitoring results. At the same time, through the visualization presentation and quantitative analysis of the entire dynamic process, it provides strong support for the in-depth study of the gas-liquid two-phase discharge phenomenon in petrochemical safety engineering, helps engineers better understand the discharge mechanism, optimize the design of the discharge system, improve the safety monitoring and risk prevention and control level, and is of great significance for ensuring the safety of petrochemical production.

[0053] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and the descriptions in the specification are only preferred examples of the present invention and are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements fall within the scope of the present invention claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent monitoring method for gas-liquid two-phase discharge volume based on multi-plane three-dimensional reconstruction, characterized in that, It includes the following steps: The discharged fluid containing gas-liquid two-phase enters the observation area; Intercept multiple two-dimensional fluid plane areas in the observation area to obtain the size, position distribution, and movement speed of the droplets; Back-calculate the gas-phase flow rate based on the droplet movement speed to obtain the two-dimensional information field of the gas-liquid two-phase in different two-dimensional fluid plane areas; According to different two-dimensional information fields, reconstruct the three-dimensional information field of the gas-liquid two-phase in the entire observation area, and calculate the total discharge amount of the entire fluid domain.

2. The intelligent monitoring method for gas-liquid two-phase discharge amount based on multi-plane three-dimensional reconstruction according to claim 1, wherein A reactor is arranged upstream of the observation area, and a safety relief valve is arranged at the top of the reactor; Among them, the opening steps of the safety relief valve are as follows: Select monitoring points; The PLC controller collects in real time the process parameters affecting the safety threshold at each monitoring point; Dynamically compare the collected process parameters with the preset safety threshold. Once deviating from the safety threshold, the PLC controller immediately controls the safety relief valve to open and discharge.

3. The intelligent monitoring method for gas-liquid two-phase discharge volume based on multi-plane three-dimensional reconstruction according to claim 2, wherein The monitoring points are arranged in the areas where process parameter fluctuations are likely to occur in the reactor.

4. The intelligent monitoring method for gas-liquid two-phase discharge amount based on multi-plane three-dimensional reconstruction according to claim 1, characterized in that The observation area is the space basin occupied by the gas-liquid two-phase fluid discharged by the safety relief valve.

5. The intelligent monitoring method for gas-liquid two-phase discharge amount based on multi-plane three-dimensional reconstruction according to claim 1, wherein The velocity distribution of the gas-liquid two-phase fluid in the observation area shows a gradient change on different two-dimensional fluid plane areas.

6. The intelligent monitoring method for gas-liquid two-phase discharge amount based on multi-plane three-dimensional reconstruction according to claim 1, wherein The steps for obtaining the droplet movement speed are as follows: Use a high-speed camera to capture the image of the discharged fluid in the observation area; Adjust the light source to an appropriate brightness and angle, and intercept multiple two-dimensional fluid plane areas in a variable focal length manner; Obtain the real-time images of the droplet distribution in different two-dimensional fluid plane areas; Identify the droplets in the image through intelligent detection software, and calculate the size, position distribution, and movement speed of the droplets at different positions in different two-dimensional fluid plane areas.

7. The intelligent monitoring method for gas-liquid two-phase discharge volume based on multi-plane three-dimensional reconstruction according to claim 1, wherein The steps for obtaining the two-dimensional information field are as follows: Perform in-depth analysis and optimization processing on the original image data collected by the high-speed camera; Convert the original image into a standard image format that conforms to the computer vision recognition specification; With the help of a target detection algorithm based on deep learning, identify and segment the droplets in the gas-liquid two-phase fluid image to achieve the complete extraction of the droplet contour; Through the spatio-temporal correlation analysis of multiple frame image sequences, calculate the size of the droplets at different positions in different two-dimensional fluid plane areas monitored, and establish a two-dimensional information field of the droplets. Among them, the two-dimensional information field includes gas-liquid flow rate distribution, phase ratio, and droplet particle size.

8. The intelligent monitoring method for gas-liquid two-phase discharge amount based on multi-plane three-dimensional reconstruction according to claim 7, characterized in that The steps for reconstructing the three-dimensional information field are as follows: Perform coordinate mapping and spatial registration on the gas-liquid flow rate distribution, phase ratio, and droplet particle size on each two-dimensional fluid plane area; Through constructing a multi-scale grid model and interpolation algorithm, orderly associate and supplement the two-dimensional data points in three-dimensional space; Based on the fluid mechanics continuity equation and conservation law, combined with the physical characteristics of the gas-liquid two-phase fluid, perform dynamic coupling analysis on the data in three-dimensional space to generate a three-dimensional information field distribution covering the entire discharged fluid space domain.

Images