Fluid flow monitoring method, device, equipment, storage medium and program product

Abstract

The invention relates to the technical field of fluid image processing, and discloses a fluid flow monitoring method, device and equipment, a storage medium and a program product, and the fluid flow monitoring method comprises the steps: collecting a plurality of motion images of a target fluid according to a preset frame rate; feature extraction is carried out on the multiple moving images, feature information of the target fluid is obtained, and the feature information comprises fluid main feature information and fluid particle feature information; on the basis of the feature information, a machine learning model is utilized to detect fluid abnormal behaviors, and abnormal fluid particle information is removed from the feature information; and calculating the fluid main flow velocity based on the updated feature information, and calculating the fluid flow in combination with the area of the fluid section of the target fluid. The influence of fluid abnormal behaviors can be solved, and the flow monitoring accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fluid image processing, and in particular to a fluid flow monitoring method, device, equipment, storage medium and program product. Background Art

[0002] Fluid is a physical form corresponding to solid, and is a general term for liquids and gases. It is composed of a large number of molecules that are constantly in thermal motion and have no fixed equilibrium position. Its basic characteristics are that it has no fixed shape and is fluid. Fluid flow is an important data that characterizes fluid movement.

[0003] Traditional fluid flow monitoring systems are often affected by insufficient time resolution or measurement errors when dealing with complex fluid dynamic behavior and instantaneous changes. High-speed imaging technology combined with image processing algorithms has been used for flow measurement, but there are still challenges when facing abnormal fluid behavior. It is unable to cope with the impact of abnormal conditions in the fluid on fluid flow calculations. Summary of the invention

[0004] In view of this, the present invention provides a fluid flow monitoring method, device, equipment, storage medium and program product to solve the influence of abnormal fluid behavior and improve the accuracy of flow monitoring.

[0005] In a first aspect, the present invention provides a fluid flow monitoring method, which includes: acquiring multiple motion images of a target fluid at a preset frame rate; performing feature extraction on the multiple motion images to obtain feature information of the target fluid, the feature information including main feature information of the fluid and feature information of fluid particles; based on the feature information, using a machine learning model to detect abnormal fluid behavior, and eliminating abnormal fluid particle information from the feature information; calculating the main flow velocity of the fluid based on the updated feature information, and calculating the fluid flow rate in combination with the area of ​​the fluid cross-section of the target fluid.

[0006] In this implementation, the application improves the spatial understanding of fluid motion by collecting motion images from multiple angles, and can capture more subtle flow changes. Compared with the traditional single-angle measurement method, it has higher accuracy. At the same time, abnormal behavior detection of fluid based on machine learning models can ensure the accuracy of data measurement results and reduce measurement errors caused by abnormal fluid behavior.

[0007] In an optional embodiment, feature extraction is performed on multiple motion images to obtain characteristic information of the target fluid, including: edge detection is performed on multiple motion images to identify the boundary contour of the target fluid to obtain the main direction information of the target fluid; fluid particle detection is performed on multiple motion images to obtain fluid particle morphology characteristic information of the target fluid.

[0008] In an alternative embodiment, feature extraction is performed on multiple motion images, and the characteristic information of the target fluid obtained includes: based on the morphological characteristic information of the fluid particles of the target fluid, the optical flow method is used to detect the velocity of the multiple motion images, and the vector velocity information of the fluid particles of the target fluid is obtained.

[0009] In an alternative embodiment, based on the characteristic information, a machine learning model is used to detect abnormal fluid behavior, including: inputting the fluid main direction information and the vector velocity information of the fluid particles into the machine learning model to detect the angular difference between the direction information of the fluid particles and the fluid main direction information; when the angular difference is greater than a preset angle, it is determined that there is an abnormality in the fluid particles.

[0010] In this implementation, a machine learning model is introduced to analyze and learn multi-angle image data, which not only improves the accuracy of flow measurement, but also can continuously optimize the model through data, enhancing the long-term adaptability and prediction ability of the system.

[0011] In an alternative embodiment, calculating the main flow velocity of the fluid based on the updated characteristic information includes: setting a plurality of flow velocity points in the fluid cross-section; calculating the flow velocity point velocity information of the flow velocity points based on the vector velocity information of the three fluid particles closest to the flow velocity points; calculating the main flow velocity of the fluid based on the flow velocity information of the plurality of flow velocity points.

[0012] In an alternative embodiment, the fluid flow monitoring method includes: pre-constructing a flow velocity frame rate relationship table, where the faster the flow velocity, the higher the frame rate; based on the main flow velocity of the fluid, using the flow velocity frame rate relationship table to determine the corresponding camera frame rate; adjusting the preset frame rate based on the camera frame rate.

[0013] In this implementation, the imaging frame rate can be dynamically adjusted according to the real-time fluid flow velocity, ensuring that data redundancy is reduced at low flow velocities and that fast-changing details can be captured at high flow velocities, improving the monitoring flexibility and efficiency of the system.

[0014] In a second aspect, the present invention provides a fluid flow monitoring device, which includes: an acquisition module for acquiring a plurality of motion images of the target fluid at a preset frame rate; an extraction module for performing feature extraction on the plurality of motion images to obtain the characteristic information of the target fluid, where the characteristic information includes fluid main characteristic information and fluid particle characteristic information; a detection module for detecting abnormal fluid behavior based on the characteristic information using a machine learning model, and removing abnormal fluid particle information from the characteristic information; a calculation module for calculating the main flow velocity of the fluid based on the updated characteristic information and calculating the fluid flow rate in combination with the area of the fluid cross-section of the target fluid.

[0015] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the fluid flow rate monitoring method according to the first aspect or any corresponding embodiment thereof.

[0016] In a fourth aspect, the present invention provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to cause a computer to perform the fluid flow rate monitoring method according to the first aspect or any corresponding embodiment thereof.

[0017] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions, and the computer instructions are used to cause a computer to perform the fluid flow rate monitoring method according to the first aspect or any corresponding embodiment thereof. Description of the Drawings

[0018] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 is a schematic diagram of a fluid flow rate monitoring system according to an embodiment of the present invention;

[0020] Figure 2 is a flowchart of a fluid flow rate monitoring method according to an embodiment of the present invention;

[0021] Figure 3 is a flowchart of another fluid flow rate monitoring method according to an embodiment of the present invention;

[0022] Figure 4 is a schematic diagram of a target fluid according to an embodiment of the present invention;

[0023] Figure 5 is a structural block diagram of a fluid flow rate monitoring device according to an embodiment of the present invention;

[0024] Figure 6 is a schematic hardware structure diagram of a computer device according to an embodiment of the present invention. Detailed Embodiments

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0026] According to an embodiment of the present invention, a fluid flow monitoring system is provided. Figure 1 It is a schematic diagram of a fluid flow monitoring system according to an embodiment of the present invention. The fluid flow monitoring system includes an image acquisition module, a data transmission module, and a data processing module. The data transmission module is respectively connected to the image acquisition module and the data processing module.

[0027] Among them, the image acquisition module is used to acquire image data of the target fluid, the data transmission module is used to transmit the image data to the data processing module, and the data processing module is used to process the image data to obtain the flow rate of the target fluid.

[0028] The image processing module is further used to generate a control signal according to the flow rate of the target fluid, the data transmission module is used to transmit the control signal to the image acquisition module, and the image acquisition module is used to adjust the acquisition method according to the control information.

[0029] Among them, the image acquisition module includes one or more image acquisition devices, and the acquisition method includes the acquisition frame rate.

[0030] Before using the fluid flow monitoring system to monitor the fluid flow, one or more image acquisition devices in the image acquisition module are initialized in advance.

[0031] In one implementation, the image acquisition device is a high-speed camera.

[0032] Specifically, the initialization includes calibration and setting of initial parameters.

[0033] In one implementation, a standard fluid is used to calibrate the acquisition system and sensors of one or more image acquisition devices to ensure the accuracy of image acquisition and flow data calculation.

[0034] In one implementation, the parameters include frame rate, resolution, shooting angle, etc. The initial frame rate, initial resolution, and initial shooting angle of one or more image acquisition devices are set during initialization.

[0035] This fluid flow monitoring system is not only applicable to simple and stable fluid monitoring environments, but also can work precisely in complex and non-linear flow environments (such as turbulence and eddy currents), with wide adaptability.

[0036] On this basis, according to an embodiment of the present invention, an embodiment of a fluid flow monitoring method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0037] In this embodiment, a fluid flow monitoring method is provided. Figure 2 It is a flowchart of a fluid flow monitoring method according to an embodiment of the present invention. It should be noted that if there are substantially the same results, this embodiment is not limited to Figure 2 the shown process sequence. As Figure 2 shown, the process includes the following steps:

[0038] Step S201, collect multiple motion images of the target fluid at a preset frame rate.

[0039] Among them, the preset frame rate is the initial frame rate initially set by the image acquisition device.

[0040] After the fluid flow monitoring system is started, the dynamic behavior of the target fluid is monitored in real time through the image acquisition device of the image acquisition module, and multiple motion images are collected. Among them, the multiple motion images include motion images at multiple moments and are used to monitor the motion behavior of the target fluid.

[0041] In one implementation, there are multiple image acquisition devices, and the multiple image acquisition devices collect multiple motion images of the target fluid from different angles.

[0042] Send the multiple motion images to the data processing module through the data transmission module.

[0043] Step S202, perform feature extraction on the multiple motion images to obtain the feature information of the target fluid.

[0044] Among them, feature extraction is to extract fluid feature data from multiple motion images, which is used to calculate the flow velocity and flow rate of the target fluid and monitor the fluid flow.

[0045] Among them, the feature information includes fluid main feature information and fluid particle feature information.

[0046] Among them, the fluid main feature information is the feature information about the whole obtained with the entire target fluid as the reference object. The fluid main feature information includes fluid main direction information. It can be understood that the fluid main direction information is the main motion direction of the target fluid.

[0047] The fluid particle characteristic information is characteristic information about the fluid particles obtained by taking each fluid particle in the target fluid as a reference object, and the fluid particle characteristic information includes fluid particle morphological characteristic information and fluid particle vector velocity information.

[0048] In one implementation, two independent feature extractions are performed on a plurality of motion images to obtain fluid main direction information, fluid particle morphology feature information, and fluid particle vector velocity information of the target fluid, respectively.

[0049] In another implementation, a first feature extraction is performed on multiple motion images to obtain fluid main direction information and fluid particle morphology feature information of the target fluid, and based on this, a second feature extraction is performed on the multiple motion images to obtain fluid particle vector velocity information of the target fluid.

[0050] Step S203, based on the feature information, using a machine learning model, detects abnormal fluid behavior and removes abnormal fluid particle information from the feature information.

[0051] Among them, abnormal behaviors include turbulence, eddies, etc.

[0052] Among them, when the flow velocity is very low, the fluid flows in layers without mixing, which is called laminar flow, also known as steady flow or sheet flow; gradually increase the flow velocity, the streamlines of the fluid begin to swing in a wave-like manner, and the frequency and amplitude of the swing increase with the increase in flow velocity. This flow condition is called transitional flow; when the flow velocity increases to a very large value, the streamlines are no longer clearly discernible, there are many small vortices in the flow field, the laminar flow is destroyed, and there is not only sliding but also mixing between adjacent flow layers. At this time, the fluid moves irregularly, and a component velocity perpendicular to the axis of the flow tube is generated. This movement is called turbulence. Vortex is a rotating fluid area formed in the fluid.

[0053] Based on the characteristics of turbulence and eddy currents and the characteristic information of the current target fluid, a machine learning model is used to determine whether turbulence and / or eddy currents exist in the current target fluid. When turbulence and / or eddy currents exist in the target fluid, the characteristic information of the corresponding area will affect the subsequent calculation of the main flow velocity of the fluid and the calculation of the fluid flow rate. In this case, the abnormal fluid particle information is eliminated from the characteristic information obtained in step S202.

[0054] Step S204, calculating the fluid main flow velocity based on the updated characteristic information, and calculating the fluid flow rate in combination with the area of ​​the fluid cross section of the target fluid.

[0055] The fluid flow monitoring method provided in this embodiment captures motion images from multiple angles, enhancing the spatial understanding of fluid motion and enabling the detection of more subtle flow changes. Compared with traditional single-angle measurement methods, it has higher accuracy. Meanwhile, based on a machine learning model, it detects abnormal fluid behaviors, ensuring the accuracy of data measurement results and reducing measurement errors caused by abnormal fluid behaviors.

[0056] In this embodiment, a fluid flow monitoring method is provided. Figure 3 It is a flowchart of another fluid flow monitoring method according to an embodiment of the present invention. It should be noted that if there are substantially the same results, this embodiment is not limited to Figure 3 the shown process sequence. As Figure 3 shown, the process includes the following steps:

[0057] Step S301, collect multiple motion images of the target fluid at a preset frame rate.

[0058] The image acquisition device is a high-speed camera. Before using the fluid flow monitoring system to monitor fluid flow, one or more image acquisition devices in the image acquisition module are initialized in advance.

[0059] Specifically, the initialization includes calibration and setting of initial parameters. The acquisition system and sensors of one or more image acquisition devices are calibrated using a standard fluid to ensure the accuracy of image acquisition and flow data calculation. The parameters include frame rate, resolution, shooting angle, etc. The initial frame rate, initial resolution, and initial shooting angle of one or more image acquisition devices are set during initialization.

[0060] After the fluid flow monitoring system is started, the dynamic behavior of the target fluid is monitored in real time through a high-speed camera. Multiple motion images of the target fluid are collected from different angles using multiple cameras to ensure comprehensive coverage of complex fluid motion, especially in the case of turbulent or non-linear flow.

[0061] Furthermore, the motion images from multiple angles are integrated. Specifically, the images from multiple camera angles are registered and stitched to form a three-dimensional view of the overall fluid flow, so as to obtain more comprehensive information on the flow of the target fluid.

[0062] Multiple motion images captured by the high-speed camera are transmitted to the data processing module through the data transmission module.

[0063] Among them, the imaging data from different angles are fused to avoid blind spots and errors that may exist in single-angle imaging, thereby providing more complete and accurate measurement results for complex fluid environments.

[0064] Step S302: Extract features from multiple motion images to obtain the feature information of the target fluid.

[0065] Before feature extraction, preprocess the multiple motion images in advance.

[0066] Among them, the preprocessing includes image denoising and enhancement. Specifically, perform denoising, contrast enhancement, sharpening, etc. on the multiple motion images.

[0067] Among them, the feature information includes the main fluid feature information and the fluid particle feature information. The main fluid direction information is the main motion direction of the target fluid. The fluid particle feature information includes the fluid particle morphological feature information and the fluid particle vector velocity information.

[0068] In one implementation, perform the first feature extraction on the multiple motion images to obtain the main fluid direction information and the fluid particle morphological feature information of the target fluid, and based on this, perform the second feature extraction on the multiple motion images to obtain the fluid particle vector velocity information of the target fluid.

[0069] Specifically, the above step S302 includes:

[0070] Step S3021: Perform edge detection on the multiple motion images, identify the boundary contour of the target fluid, and obtain the main fluid direction information of the target fluid.

[0071] Among them, the edge detection algorithm can be a differential operator, Roberts operator, Sobel operator, Prewitt operator, Log operator, Canny operator, etc.

[0072] In one implementation, use the Canny operator to perform edge detection on the multiple motion images. The Canny algorithm is a standard algorithm widely used in edge detection. Its goal is to find an optimal edge detection solution or find the position with the strongest change in gray intensity in an image. The optimal edge detection is mainly evaluated by three criteria: low error rate, high localization, and minimum response. Specifically, apply Gaussian filtering to smooth the image to remove noise, find the gradient of the image, apply non-maximum suppression technology to filter out non-edge pixels, and make the blurred boundary clear. This process retains the maximum value of the gradient intensity at each pixel point and filters out other values. Apply the double-threshold method to determine possible (potential) boundaries. If a weak boundary connected to a strong boundary at a certain pixel position is considered a boundary, other weak boundaries are deleted. Identify the boundary contour of the target fluid and obtain the main fluid direction information of the target fluid.

[0073] Step S3022: Perform fluid particle detection on the multiple motion images to obtain the fluid particle morphological feature information of the target fluid.

[0074] Among them, the fluid main direction information in step S3021 is information characterizing the overall flow direction of the target fluid and cannot characterize the states of individual fluid particles in the target fluid. Continue to perform individual fluid particle detection on multiple motion images, determine the position characteristics and state characteristics of each target fluid, and obtain the fluid particle morphological characteristic information of the target fluid.

[0075] Step S3023, based on the fluid particle morphological characteristic information of the target fluid, use the optical flow method to perform velocity detection on multiple motion images to obtain the fluid particle vector velocity information of the target fluid.

[0076] Among them, the fluid particle morphological characteristic information in step S3022 characterizes the states of individual fluid particles in the target fluid in each motion image and cannot characterize the motion states of individual fluid particles in the target fluid. Continue to use the optical flow method to perform velocity detection on multiple motion images based on the states of individual fluid particles in each motion image to obtain the fluid particle vector velocity information of the target fluid.

[0077] Among them, the optical flow method uses the change of pixels in the time domain in the image sequence and the correlation between adjacent frames to find the corresponding relationship between the previous frame and the current frame, so as to calculate the motion information of the object between adjacent frames.

[0078] Specifically, by considering the differences between each motion image, through fusing multi-angle image data, calculating the morphological changes of the flow velocity particles in each frame, and calculating the flow velocity vector field of the flow velocity particles, the fluid particle vector velocity information of the target fluid is obtained.

[0079] Step S303, based on the characteristic information, use a machine learning model to detect abnormal fluid behaviors and eliminate abnormal fluid particle information from the characteristic information.

[0080] Among them, abnormal behaviors include turbulence, eddy current, etc. Through learning historical data, train a deep neural network model to predict the change pattern of fluid flow velocity.

[0081] Specifically, the above step S303 includes:

[0082] Step S3031, input the fluid main direction information and the fluid particle vector velocity information into the machine learning model to detect the angle difference between the direction information of the fluid particle and the fluid main direction information.

[0083] Use the machine learning model to detect whether there is an angle difference between the direction information of the fluid particle and the fluid main direction information based on the fluid main direction information about the target fluid and the fluid particle vector velocity information about the fluid particle in the characteristic information. When there is an angle difference, it indicates that the fluid particle does not move in accordance with the main motion direction of the target fluid, and there may be fluid anomalies.

[0084] In step S3032, when the angular difference is greater than a preset angle, it is determined that there is an abnormality in the fluid particles, and the abnormal fluid particle information is excluded from the characteristic information.

[0085] Detect whether the angular difference between the direction information of the fluid particles and the main direction information of the fluid is greater than the preset angle. When the angular difference between the direction information of the fluid particles and the main direction information of the fluid is greater than the preset angle, it indicates that there is an abnormality in the area of the fluid particles, and there may be situations such as turbulence and eddy currents. Exclude the abnormal fluid particle information from the characteristic information obtained in step S302.

[0086] In one implementation, the preset angle is 150 degrees.

[0087] In one implementation, the fluid flow monitoring system has an abnormal detection and automatic correction function. Once an abnormal flow behavior occurs, the system can issue an alarm in a timely manner and make automatic adjustments, reducing the need for human intervention.

[0088] Among them, an adaptive algorithm is used to detect the flow velocity and correct abnormal data in real time to ensure more reliable flow measurement of the system in a complex fluid environment.

[0089] This application introduces a machine learning model to analyze and learn multi-angle image data, which not only improves the accuracy of flow measurement, but also can continuously optimize the model through data, enhancing the long-term adaptability and prediction ability of the system. Analyze the image data through the machine learning model to achieve flow prediction and correct the real-time flow velocity data to ensure the stability and accuracy of flow measurement. It has an abnormal detection and automatic correction function. Once an abnormal flow behavior occurs, the system can issue an alarm in a timely manner and make automatic adjustments, reducing the need for human intervention.

[0090] Step S304, calculate the main flow velocity of the fluid based on the updated characteristic information, and calculate the fluid flow rate in combination with the area of the fluid cross-section of the target fluid.

[0091] Set multiple flow velocity points in the fluid cross-section, calculate the flow velocity point velocity information of the flow velocity points based on the fluid particle vector velocity information of the three fluid particles closest to the flow velocity points; calculate the main flow velocity of the fluid based on the flow velocity information of multiple flow velocity points.

[0092] Specifically, please refer to Figure 4 , Figure 4 is a schematic diagram of the target fluid according to an embodiment of the present invention.

[0093] Please refer to Figure 4 , divide the target fluid into multiple triangular regions according to the grid division method, and obtain the fluid particle information of each intersection point by using the above method.

[0094] Divide the fluid cross-section of the target fluid into three segments to obtain two flow velocity points. Use the fluid particle vector velocity information of the three fluid particles closest to the flow velocity points for interpolation calculation to obtain the flow velocity point velocity information of the flow velocity points. Take a part of the total flow velocity point velocity information of the flow velocity points as the main fluid flow velocity. Specifically, the main fluid flow velocity is:

[0095]

[0096] Among them, is the main fluid flow velocity, and v1 and v2 are the flow velocity point velocity information of the flow velocity points.

[0097] Calculate the fluid flow rate based on the area of the fluid cross-section of the target fluid as:

[0098]

[0099] Among them, Q is the fluid flow rate, and A is the area of the fluid cross-section of the target fluid.

[0100] Furthermore, after calculating the fluid flow rate, it also includes: adjusting the frame rate of the image acquisition device.

[0101] Among them, a flow velocity frame rate relationship table is pre-constructed. The faster the flow velocity, the higher the frame rate. Based on the main fluid flow velocity, use the flow velocity frame rate relationship table to determine the corresponding camera frame rate. Adjust the preset frame rate based on the camera frame rate.

[0102] Specifically, monitor the flow velocity change in real time. When the flow velocity is low, the system will reduce the frame rate of the camera to reduce data redundancy. When the flow velocity is high, the system automatically increases the frame rate to ensure that fast-changing fluid details are captured. The fluid flow rate monitoring system has a built-in frame rate optimization algorithm. By learning the frame rate requirements at different flow velocities, it automatically adapts to different fluid environments and balances data processing efficiency and accuracy.

[0103] Among them, the system can dynamically adjust the imaging frame rate according to the real-time fluid flow velocity, ensuring reduced data redundancy at low flow velocities and being able to capture fast-changing details at high flow velocities, improving the monitoring flexibility and efficiency of the system.

[0104] Furthermore, the fluid flow rate monitoring system of the present application sets up a monitoring interface. Through the monitoring interface, data such as the flow rate, flow velocity, and fluid dynamic changes are displayed in real time. It has the function of real-time monitoring of the flow rate, and can provide instant feedback during the monitoring process to help the operator make quick judgments and adjustments.

[0105] Specifically, the fluid flow rate monitoring system displays information such as the processed flow rate, velocity, and abnormal detection results through the interface in real time for the user to monitor and analyze. When abnormal fluid behavior is detected, the system reminds the operator through the interface or sound alarm and records the abnormal event.

[0106] This application can output monitoring results in real time through a flow monitoring interface, enabling users to always grasp the flow conditions of the fluid and make timely adjustments.

[0107] Furthermore, the fluid flow monitoring system of this application is provided with a storage module. The storage module stores the monitoring data in real time, facilitating subsequent data analysis and system optimization.

[0108] Specifically, the fluid flow monitoring system stores all the monitored data in a database for subsequent analysis. Long-term flow trend analysis is carried out using historical data, or new machine learning models are trained to further optimize the accuracy and efficiency of flow measurement. Support for the storage of monitoring data is convenient for subsequent data analysis, model optimization, and trend prediction, contributing to the research and monitoring of long-term fluid behavior.

[0109] The fluid flow monitoring method provided in this embodiment is applicable to multiple fields such as scientific research, industrial monitoring, and environmental monitoring, and performs particularly well in occasions where precise monitoring of complex fluid behavior is required (such as chemical production and river monitoring). Compared with traditional flow monitoring technologies, the system can handle complex flow environments (such as unstable multiphase flows and environments with sudden fluctuations), and provide continuous and reliable data support.

[0110] At the same time, the hardware and software module designs of this application are flexible and can expand more devices such as cameras and sensors according to specific needs, so as to adapt to the requirements of different scenarios and fluid environments. The machine learning and image processing algorithms adopted can be further optimized through subsequent updates to adapt to new environments or introduce new technologies, improving the long-term stability and performance of the system.

[0111] At the same time, through advanced image processing technology and the extraction of flow velocity vector fields, the system can quickly and accurately calculate the flow rate of the fluid. Combining with an adaptive algorithm in a dynamic environment improves the real-time performance and accuracy of flow monitoring.

[0112] The method of this application combines high-speed imaging, intelligent algorithms, and machine learning technologies, not only improving the accuracy and flexibility of flow measurement, but also enhancing the adaptability of the system in complex environments. It is an efficient, intelligent, and widely applicable flow monitoring system with potential.

[0113] In this embodiment, a fluid flow monitoring device is also provided. This device is used to implement the above embodiments and preferred implementation manners, and those that have been described will not be repeated. As used hereinafter, the term "module" can be a combination of software and / or hardware that can achieve a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0114] This embodiment provides a fluid flow monitoring device. Figure 5 It is a schematic diagram of a fluid flow monitoring device according to an embodiment of the present invention. As Figure 5 shown, it includes:

[0115] An acquisition module 501, configured to acquire a plurality of motion images of a target fluid at a preset frame rate.

[0116] An extraction module 502, configured to perform feature extraction on the plurality of motion images to obtain feature information of the target fluid, where the feature information includes fluid main feature information and fluid particle feature information.

[0117] A detection module 503, configured to detect abnormal fluid behaviors based on the feature information by using a machine learning model, and eliminate abnormal fluid particle information from the feature information.

[0118] A calculation module 504, configured to calculate the main flow velocity of the fluid based on the updated feature information, and calculate the fluid flow rate in combination with the area of the fluid cross-section of the target fluid.

[0119] The further function descriptions of the above-mentioned various modules and units are the same as those in the corresponding embodiments above, and will not be repeated here.

[0120] The fluid flow monitoring device in this embodiment is presented in the form of functional units. Here, the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0121] The embodiment of the present invention further provides a computer device having the above-mentioned Figure 5 fluid flow monitoring device.

[0122] Please refer to Figure 6 , Figure 6 which is a schematic structural diagram of a computer device provided by an optional embodiment of the present invention. As Figure 6As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including a high-speed interface and a low-speed interface. Each component communicates with each other using different buses and can be installed on a common motherboard or in other ways as needed. The processor can process instructions executed within the computer device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some alternative embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a set of blade servers, or a multi-processor system). Figure 6 In the figure, a processor 10 is taken as an example.

[0123] The processor 10 can be a central processing unit, a network processor, or a combination thereof. Among them, the processor 10 can further include a hardware chip. The above hardware chip can be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The above programmable logic device can be a complex programmable logic device, a field programmable gate array, a generic array logic, or any combination thereof.

[0124] Among them, the memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiments.

[0125] The memory 20 can include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the computer device, etc. In addition, the memory 20 can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 can optionally include a memory remotely set relative to the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the above network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0126] The memory 20 can include a volatile memory, such as a random access memory; the memory can also include a non-volatile memory, such as a flash memory, a hard disk, or a solid-state drive; the memory 20 can also include a combination of the above types of memories.

[0127] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30, and the output device 40 may be connected by a bus or other means. Figure 6 Taking the connection by bus as an example.

[0128] The input device 30 can receive input digital or character information and generate key signal inputs related to the user settings and function controls of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED), and a haptic feedback device (e.g., a vibration motor), etc. The above display device includes but is not limited to a liquid crystal display, a light-emitting diode, a display, and a plasma display. In some alternative embodiments, the display device may be a touch screen.

[0129] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code that is originally stored in a remote storage medium or a non-transitory machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be processed by such software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiment is implemented.

[0130] A part of the present invention can be applied as a computer program product, such as computer program instructions. When executed by a computer, through the operation of the computer, the methods and / or technical solutions according to the present invention can be invoked or provided. Those skilled in the art should be able to understand that the forms in which computer program instructions exist in a computer-readable medium include but are not limited to source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include but are not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0131] Although embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations fall within the scope defined by the appended claims.

Claims

1. A method for monitoring fluid flow, characterized in that: The method comprises: Acquire multiple motion images of the target fluid at a preset frame rate; Extracting features from the plurality of motion images to obtain feature information of the target fluid, wherein the feature information includes main feature information of the fluid and feature information of fluid particles; Based on the characteristic information, using a machine learning model, detecting abnormal behavior of the fluid, and removing abnormal fluid particle information from the characteristic information; The main flow velocity of the fluid is calculated based on the updated characteristic information, and the fluid flow rate is calculated in combination with the area of ​​the fluid cross section of the target fluid.

2. The fluid flow monitoring method according to claim 1, characterized in that: The extracting features of the plurality of motion images to obtain feature information of the target fluid comprises: Performing edge detection on the multiple motion images, identifying the boundary contour of the target fluid, and obtaining fluid main direction information of the target fluid; Fluid particle detection is performed on the multiple motion images to obtain fluid particle morphological feature information of the target fluid.

3. The fluid flow monitoring method according to claim 2, characterized in that: The extracting features of the plurality of motion images to obtain feature information of the target fluid comprises: Based on the fluid particle morphological feature information of the target fluid, the optical flow method is used to perform velocity detection on the multiple motion images to obtain the fluid particle vector velocity information of the target fluid.

4. The fluid flow monitoring method according to claim 3, characterized in that: The detecting of abnormal fluid behavior based on the feature information and using a machine learning model includes: Inputting the fluid main direction information and the fluid particle vector velocity information into the machine learning model, and detecting the angle difference between the direction information of the fluid particles and the fluid main direction information; When the angle difference is greater than a preset angle, it is determined that the fluid particles are abnormal.

5. The fluid flow monitoring method according to claim 1, characterized in that: The calculating the main flow velocity of the fluid based on the updated characteristic information comprises: Setting a plurality of flow velocity points in the fluid cross section; Calculating the velocity information of the flow point based on the fluid particle vector velocity information of three fluid particles closest to the flow point; The main fluid velocity is calculated based on the velocity information of the plurality of flow velocity points.

6. The fluid flow monitoring method according to claim 1, characterized in that: The method further comprises: Pre-constructing a flow rate frame rate relationship table, the faster the flow rate, the higher the frame rate; Based on the main flow velocity of the fluid, determining the corresponding camera frame rate using the flow velocity frame rate relationship table; The preset frame rate is adjusted based on the camera frame rate.

7. A fluid flow monitoring device, characterized in that: The device comprises: An acquisition module, used for acquiring multiple motion images of the target fluid at a preset frame rate; An extraction module, used for performing feature extraction on the plurality of motion images to obtain feature information of the target fluid, wherein the feature information includes main feature information of the fluid and feature information of fluid particles; A detection module, configured to detect abnormal fluid behavior based on the feature information and to remove abnormal fluid particle information from the feature information by using a machine learning model; The calculation module is used to calculate the main flow velocity of the fluid based on the updated characteristic information, and calculate the fluid flow rate in combination with the area of ​​the fluid cross section of the target fluid.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the fluid flow monitoring method according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the fluid flow monitoring method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises computer instructions for causing a computer to execute the fluid flow monitoring method according to any one of claims 1 to 6.

Images