A method and device for detecting pipeline leakage in a flow battery system

Abstract

A method and device for detecting leakage in a pipeline in a flow battery system, comprising: collecting a pipe wall appearance image of a pipe in the flow battery system through an image acquisition device at a first preset acquisition frequency, and whenever a pipe wall appearance image of an acquisition cycle is obtained, it is used as a first target pipe wall appearance image and the following operations are performed: by inputting all the obtained first target pipe wall appearance images into a pre-trained image processing model to obtain leakage judgment information of whether the pipe corresponding to each first target pipe wall appearance image is leaking, and leakage color information of the leaking pipe corresponding to the first target pipe wall appearance images with leakage judgment information of leakage in all the first target pipe wall appearance images; determining the leaking pipe according to the leakage color information of the leaking pipe in the obtained first target pipe wall appearance image. The embodiment of the present application can realize automatic detection of leaking pipes, save human resources, and improve pipeline detection efficiency.

Description

Technical Field

[0001] The present invention relates to pipeline detection technology, and in particular to a method and device for detecting pipeline leakage in a liquid flow battery system. Background Art

[0002] Liquid flow battery systems have the functions of storing and efficiently converting electric energy, so they are widely used. At present, liquid flow batteries often have occasional leaks in pipes and components during operation, which greatly reduces the efficiency of charging and discharging, the loss of costs and the waste of resources.

[0003] In the related art, liquid leakage is often detected manually.

[0004] However, this method is very labor-intensive, and leaking pipes are often not discovered in a timely manner, resulting in low detection efficiency. Summary of the invention

[0005] The present application provides a method and device for detecting pipeline leakage in a liquid flow battery system, which can realize automatic detection, save human resources, and improve detection efficiency.

[0006] On the one hand, the present application provides a method for detecting pipeline leakage in a flow battery system, comprising:

[0007] The image acquisition device is used to acquire the appearance image of the pipe wall of the pipe in the liquid flow battery system at a first preset acquisition frequency, and whenever an appearance image of the pipe wall in an acquisition cycle is acquired, the appearance image of the pipe wall in the acquisition cycle is used as the first target appearance image of the pipe wall and the following operations are performed:

[0008] By inputting all the first target tube wall appearance images obtained into a pre-trained image processing model, leakage determination information of whether the pipeline corresponding to each first target tube wall appearance image is leaking, and leakage color information of the leaking pipeline corresponding to the first target tube wall appearance images with leakage determination information of leakage among all the first target tube wall appearance images;

[0009] The leaking pipeline is determined according to the leakage color information of the leaking pipeline in the obtained first target pipe wall appearance image.

[0010] On the other hand, the present application provides a device for detecting pipeline leakage in a liquid flow battery system, a memory and a processor, wherein the memory is used to store an executable program;

[0011] The processor is used to read and execute the executable program, and is applied to any of the above-mentioned methods for detecting pipeline leakage in a liquid flow battery system.

[0012] Compared with the related art, the embodiment of the present application can use an image acquisition device to collect the appearance image of the pipe wall of the pipeline in the liquid flow battery system, and use a pre-trained image processing model to judge whether the pipe is leaking and the color of the leakage of the leaking pipe on the collected pipe wall appearance image, thereby realizing automatic detection of leaking pipes, saving human resources to a great extent, and improving the efficiency of pipeline detection.

[0013] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or be understood by implementing the present application. Other advantages of the present application can be realized and obtained by the schemes described in the description and the drawings.

[0014] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or be understood by implementing the present application. Other advantages of the present application can be realized and obtained by the schemes described in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings are used to provide an understanding of the technical solution of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the technical solution of the present application and do not constitute a limitation on the technical solution of the present application.

[0016] Figure 1 This is a flow chart of a method for detecting pipeline leakage in a flow battery system according to an embodiment of the present application;

[0017] Figure 2 This is a schematic diagram of the structure of a flow battery according to an embodiment of the present application;

[0018] Figure 3 This is a schematic diagram of the structure of a device for detecting pipeline leakage in a liquid flow battery system according to an embodiment of the present application. DETAILED DESCRIPTION

[0019] The present application describes multiple embodiments, but the description is exemplary rather than restrictive, and it is obvious to those skilled in the art that there may be more embodiments and implementations within the scope of the embodiments described in the present application. Although many possible feature combinations are shown in the drawings and discussed in the specific embodiments, many other combinations of the disclosed features are also possible. Unless specifically limited, any feature or element of any embodiment may be used in combination with any other feature or element in any other embodiment, or may replace any other feature or element in any other embodiment.

[0020] The present application includes and contemplates combinations of features and elements known to those of ordinary skill in the art. The embodiments, features and elements disclosed in the present application may also be combined with any conventional features or elements to form a unique invention scheme defined by the claims. Any features or elements of any embodiment may also be combined with features or elements from other invention schemes to form another unique invention scheme defined by the claims. Therefore, it should be understood that any feature shown and / or discussed in the present application may be implemented individually or in any appropriate combination. Therefore, except for the limitations made according to the attached claims and their equivalents, the embodiments are not subject to other restrictions. In addition, various modifications and changes may be made within the scope of protection of the attached claims.

[0021] In addition, when describing representative embodiments, the specification may have presented the method and / or process as a specific sequence of steps. However, to the extent that the method or process does not rely on the specific order of the steps described herein, the method or process should not be limited to the steps of the specific order described. As will be understood by those of ordinary skill in the art, other sequences of steps are also possible. Therefore, the specific sequence of the steps set forth in the specification should not be interpreted as a limitation to the claims. In addition, the claims for the method and / or process should not be limited to the steps of performing them in the order written, and those skilled in the art can easily understand that these sequences can be changed and still remain within the spirit and scope of the embodiments of the present application.

[0022] As the demand for regulation capacity of the power system increases and the scale of new energy development and consumption continues to increase, new energy storage has a short construction period, simple and flexible site selection, strong regulation capacity, and better matching with new energy development and consumption. Its advantages are gradually becoming more prominent, and it is imperative to accelerate the large-scale application of advanced energy storage technology. New energy storage is one of the breakthroughs for spawning new forms of energy industry and creating new economic engines. In the context of building a new development pattern of mutual promotion of domestic and international dual circulation, there are major opportunities to accelerate the layout of new energy storage industry. Liquid flow batteries are widely used in the energy storage industry due to their good stability and safety, and have become the mainstream technical solution in the field of energy storage. Liquid flow batteries are based on the redox reaction generated by the flow of electrolyte, which enables vanadium ions to continuously convert their valence states, thereby achieving the purpose of charging and discharging. Since the flow of electrolyte is the core concept of the operation of liquid flow batteries, the design of the pipelines on which the electrolyte flow relies is crucial.

[0023] At present, liquid flow batteries always have occasional leaks in pipes and components during operation, which greatly reduces the efficiency of charging and discharging, resulting in cost losses and waste of resources. At the same time, since the leakage point is difficult to find and detect, it increases the difficulty of investigation. Therefore, when a leak occurs, remediation is often not timely. After the leakage occurs, the pressure of the instrument system will be unstable, the peak retention time will be prolonged, and the peak width will become wider. If the leakage is not discovered in time for a long time, it may cause a short circuit in the circuit and damage important parts of the instrument, which will lead to major safety accidents. Most of the existing leakage monitoring methods are discovered through manual inspection. For unmanned equipment and 24-hour online equipment, leakage cannot be discovered in time, resulting in instrument failure or safety accidents.

[0024] To this end, the present application provides a method for detecting leakage in a pipeline of a flow battery system, such as Figure 1 As shown, the method includes:

[0025] The image acquisition device is used to acquire the appearance image of the pipe wall of the pipe in the liquid flow battery system at a first preset acquisition frequency, and whenever an appearance image of the pipe wall in an acquisition cycle is acquired, the appearance image of the pipe wall in the acquisition cycle is used as the first target appearance image of the pipe wall and the following operations are performed:

[0026] Step 101: input all the first target tube wall appearance images obtained into a pre-trained image processing model to obtain leakage determination information of whether the pipeline corresponding to each first target tube wall appearance image is leaking, and leakage color information of the leaking pipeline corresponding to the first target tube wall appearance images whose leakage determination information is leakage among all the first target tube wall appearance images;

[0027] Step 102: determine the leaking pipeline according to the leakage color information of the leaking pipeline in the obtained first target pipe wall appearance image.

[0028] Among them, the first preset acquisition frequency can be set according to actual detection requirements. If the real-time requirement for pipeline leakage detection is high, the first preset acquisition frequency can be set higher. If the real-time requirement for pipeline leakage detection is not high, the first preset acquisition frequency can be set lower.

[0029] The image acquisition device can be set according to the pipeline position that needs to be monitored in the liquid flow battery system. The pipeline position that needs to be monitored can be a position in the liquid flow battery system that is prone to leakage, such as the interface of the pipeline in the liquid flow battery system. The interface can be the interface where different segmented pipelines are connected, or it can be the interface where other measuring devices (such as pressure sensors) are installed in the pipeline.

[0030] The image acquisition device can be set in a fixed manner or in a mobile manner. When the mobile manner is adopted, the image acquisition device can be moved to each pipeline position that needs to be monitored in turn to acquire the image of the appearance of the pipe wall. When the fixed manner is adopted, one or more image acquisition devices can be set for each pipeline position that needs to be monitored, and one or more image acquisition devices can also be set for at least two adjacent positions that are prone to leakage, that is, at least two adjacent positions that are prone to leakage "share" one or more image acquisition devices.

[0031] The image acquisition device may be a camera or a video camera. The video camera may be a charge coupled device (CCD) camera. When the appearance image of the pipe wall of the pipeline in the flow battery system is acquired by the CCD camera, the CCD camera will acquire the appearance image of the pipe wall of the pipeline at intervals of t under the mapping of the light source.

[0032] The image processing model may specifically be an Elman neural network model, which includes an input layer, a hidden layer, a correlation layer, and an output layer.

[0033] Among them, the input layer is expressed as:

[0034] The hidden layer is expressed as:

[0035]

[0036] The association layer is represented as:

[0037]

[0038] The output layer is represented as:

[0039]

[0040] The method for detecting leakage of pipelines in a flow battery system provided in an embodiment of the present application can use an image acquisition device to collect the appearance image of the pipe wall of the pipe in the flow battery system, and use a pre-trained image processing model to determine whether the pipe is leaking and the color of the leakage of the leaking pipe on the collected appearance image of the pipe wall, thereby realizing automatic detection of the leaking pipe, greatly saving human resources and improving the efficiency of pipe detection. In addition, the method for detecting leakage of pipelines in a flow battery system provided in an embodiment of the present application can quickly lock the leaking pipe, thereby promptly discovering the leakage point, greatly reducing the troubleshooting time, and reducing the leakage of the electrolyte.

[0041] In an exemplary embodiment, the liquid flow battery system includes: a positive electrode liquid storage tank, a negative electrode liquid storage tank and a pipeline, wherein the pipeline includes: a first liquid supply pipeline and a first liquid return pipeline connected to the positive electrode liquid storage tank, and a second liquid supply pipeline and a second liquid return pipeline connected to the negative electrode liquid storage tank.

[0042] For example, the flow battery system can be Figure 2 As shown, 1 and 9 are the positive electrode storage tank and the negative electrode storage tank of the system, respectively, which serve as the storage electrolyte of the circulation system, 2 represents the positive electrode supply pipeline (i.e., the first supply pipeline), the supply pipeline 2 is transmitted to the battery stack 3, and then enters the positive electrode storage tank 1 through the positive electrode return pipeline 4 (i.e., the first return pipeline), completing a positive cycle, and 6 and 7 in the negative electrode pipeline represent the negative electrode supply pipeline (i.e., the second supply pipeline) and the negative electrode return pipeline (i.e., the second return pipeline), respectively, after passing through the battery stack 3, the electrolyte enters the negative electrode storage tank 9 to complete a negative cycle. The reaction formula involved in the circulation process, in which the potential difference generated is transmitted through 5 and 8. Among them, the vanadium ions will undergo a change in valence during the circulation process, and the chemical equation is as follows, in which pentavalent vanadium ions will display yellow, tetravalent vanadium ions will display blue, trivalent vanadium ions will display green, and divalent vanadium ions will display purple. Therefore, during the circulation process of the system, the positive electrode's liquid supply pipeline will display yellow (pentavalent), the return liquid pipeline will display blue (quadrivalent), the negative electrode's liquid supply pipeline will display purple (divalent), and the return liquid pipeline will display green (trivalent).

[0043] Positive electrode: VO 2 + +2H + +e - <—>VO 2 2+ +H 2 O

[0044] Negative electrode: V 2+ <—>V 3+ +e -

[0045] Battery reaction: VO 2 + +2H + +V 2+ <—>VO 2 2+ +H 2 O+V 3+

[0046] In an exemplary embodiment, the image processing model includes: an image processing model configured to determine leakage judgment information indicating whether the corresponding pipeline is leaking based on the pipeline outer wall image, and an image processing model in which the leakage judgment information is leakage color information of the leaking pipeline corresponding to the pipeline outer wall image of the leaking pipeline.

[0047] In an exemplary embodiment, the image processing model is established in the following manner:

[0048] The following operations are performed for each pipeline: first, a plurality of pipeline outer wall images of the pipeline are collected by an image acquisition device as corresponding sample pipeline outer wall images; wherein the plurality of sample pipeline outer wall images corresponding to the pipeline include: a sample pipeline outer wall image of the pipeline leaking liquid and a sample pipeline outer wall image of the pipeline not leaking liquid;

[0049] Secondly, obtaining leakage determination information of whether the pipeline is leaking in all sample pipeline outer wall images as corresponding leakage determination information, and obtaining leakage color information of the leaking pipeline in the sample pipeline outer wall images with leakage determination information of leakage as corresponding leakage color information;

[0050] Finally, all the obtained sample pipeline outer wall images are taken as input, the leakage judgment information corresponding to all pipeline outer wall images, and the leakage color information corresponding to the sample pipeline outer wall images with the leakage judgment information being leakage are taken as labels of the corresponding sample pipeline outer wall images, and a pre-set neural network model is trained to obtain the image processing model.

[0051] When the image processing model is an Elman neural network model, the process of training the Elman neural network model to obtain the image processing model is as follows: the pipeline outer wall image of the pipeline is obtained and used as the data of the input layer of Elman, the corresponding leakage judgment information and the leakage color information corresponding to the sample pipeline outer wall image with the leakage judgment information as leakage are used as the label of the corresponding sample pipeline outer wall image. After importing the data, it is necessary to initialize the weight and threshold of Elamn, determine the number and hierarchy of neuron nodes, and use back propagation to transfer parameters.

[0052] In an exemplary instance, the image processing model includes: an image processing model configured to determine leakage judgment information indicating whether the corresponding pipeline is leaking based on the pipeline outer wall image, and an image processing model in which the leakage judgment information is leakage color information and leakage status information of the leaking pipeline corresponding to the pipeline outer wall image of the leaking pipeline.

[0053] In an exemplary embodiment, the image processing model is established in the following manner:

[0054] The following operations are performed for each pipeline: first, a plurality of pipeline outer wall images of the pipeline are collected by an image acquisition device as corresponding sample pipeline outer wall images; wherein the plurality of sample pipeline outer wall images corresponding to the pipeline include: a sample pipeline outer wall image of the pipeline leaking and a sample pipeline outer wall image of the pipeline not leaking;

[0055] Secondly, obtaining leakage determination information of whether the pipeline is leaking in all sample pipeline outer wall images as corresponding leakage determination information, obtaining leakage color information of the leaking pipeline in the sample pipeline outer wall images with leakage determination information being leakage as corresponding leakage color information, and obtaining leakage state information of the leaking pipeline in the sample pipeline outer wall images with leakage determination information being leakage as corresponding leakage state information;

[0056] Finally, all the obtained sample appearance images are taken as input, the leakage judgment information corresponding to all sample pipeline outer wall images, as well as the leakage color information and leakage state information corresponding to the sample pipeline outer wall images with the leakage judgment information being leakage, are taken as labels of the corresponding sample pipeline outer wall images, and a pre-set neural network model is trained to obtain the image processing model.

[0057] When the image processing model is an Elman neural network model, the process of training the Elman neural network model to obtain the image processing model is as follows: the pipeline outer wall image of the pipeline is obtained and used as the data of the input layer of Elman, and the corresponding leakage judgment information, as well as the leakage color information and leakage state information corresponding to the sample pipeline outer wall image with leakage judgment information as leakage are used as labels of the corresponding sample pipeline outer wall image. After importing the data, it is necessary to initialize the weights and thresholds of Elamn, determine the number and hierarchy of neuron nodes, and use back propagation to transfer parameters.

[0058] The method for detecting pipeline leakage in a liquid flow battery system provided in the embodiment of the present application can not only identify the leakage color information of the leaking pipeline in the sample pipeline outer wall image with leakage judgment information as leakage, but also identify the leakage status information of the leaking pipeline in the sample pipeline outer wall image with leakage judgment information as leakage. The leakage status information can be used for subsequent evaluation of the leakage amount.

[0059] In an exemplary embodiment, the method further includes: inputting all the first target tube wall appearance images obtained into a pre-trained image processing model to obtain leakage state information of the leaking tube in each first target tube wall appearance image as first leakage state information of each leaking tube;

[0060] Perform the following operations for each leaking pipeline:

[0061] First, the pipeline pressure of the leaking pipeline when there is no leakage is obtained by a pressure sensor as a first pipeline pressure, and the acquisition time of the first target pipe wall appearance image corresponding to the leaking pipeline is recorded as a first acquisition time;

[0062] Secondly, the information that the leaking pipeline has leaked is notified to the manager in a preset manner, and the repair time of the leaking pipeline by the manager is recorded, and the pipeline pressure after the leaking pipeline is repaired is obtained through the pressure sensor as the second pipeline pressure;

[0063] Finally, the leakage amount of the leaking pipeline is calculated according to the first leakage state information of the leaking pipeline, the first collection time of the leaking pipeline, the repair time of the leaking pipeline, the first pipeline pressure of the leaking pipeline, and the second pipeline pressure of the leaking pipeline.

[0064] In an exemplary embodiment, the leakage status information includes: the leakage point area and the leakage point aspect ratio.

[0065] Exemplarily, the leakage amount of the leaking pipeline can be calculated by the following formula:

[0066] Q leak =Q 漏液点流量 *T

[0067]

[0068] Among them, Q leak represents the leakage amount of the leaking pipeline, T represents the leakage time of the leaking pipeline, Q 漏液点流量 Indicates the flow rate of the leaking point of the leaking pipeline, in m 3 / s, A represents the leakage point area of ​​the leaking pipeline, the unit is m 2 , γ represents the aspect ratio of the leakage point of the leaking pipeline, D represents the pipeline diameter of the leaking pipeline, the unit is m, ΔP represents the pipeline pressure difference of the leaking pipeline, a, b, c, d, e, f are all constant factors, as shown in Figure 1.

[0069] Table 1

[0070] Constant Factor a b c d e f value 0.2364 -0.02989 0.08211 0.9241 1.327 0.0001543

[0071] In the specific calculation process, ΔP is the pressure difference between the first pipeline pressure of the leaking pipeline and the second pipeline pressure of the leaking pipeline, and T is obtained according to the first collection time of the leaking pipeline and the repair time of the leaking pipeline.

[0072] In an exemplary embodiment, the leakage state may further include: leakage type, and the leakage type includes: leakage, dripping, heavy leakage, and flow.

[0073] The method for detecting pipeline leakage in a flow battery system provided in an embodiment of the present application obtains leakage status information of the leaking pipeline in the image of the appearance of the pipe wall when leakage occurs, and then determines the leakage duration according to the first acquisition time and repair time of the leaking pipeline, determines the pressure difference caused by the leakage according to the first pipeline pressure and the second pipeline pressure of the leaking pipeline, and finally estimates the leakage amount according to the leakage status information, the leakage duration and the pressure difference caused by the leakage. In an exemplary embodiment, the method further includes:

[0074] Perform the following operations for each leaking pipeline:

[0075] The image of the outer wall of the leaking pipeline is collected by the image acquisition device at a second preset frequency until the manager repairs the leaking pipeline, and the image of the outer wall of the leaking pipeline collected at the second preset frequency is used as the second target pipeline outer wall image, and each time a second target pipeline outer wall image is obtained, the following operations are performed on the second target pipeline outer wall image:

[0076] Inputting the second target pipeline outer wall image into a pre-trained image processing model, obtaining leakage state information of the leaking pipeline in the second target pipeline outer wall image and using the information as the second leakage state information corresponding to the second target pipeline outer wall image of the leaking pipeline;

[0077] If the second target pipeline outer wall image of the leaking pipeline is the first second target pipeline outer wall image, determine whether the second leakage state information corresponding to the second target pipeline outer wall image of the leaking pipeline and the first leakage state information of the leaking pipeline have changed; if changed, use the second leakage state information corresponding to the second target pipeline outer wall image of the leaking pipeline as the third leakage state information of the leaking pipeline, obtain the acquisition time of the second target pipeline outer wall image of the leaking pipeline as the second acquisition time, and obtain the pipeline pressure when the second target pipeline outer wall image of the leaking pipeline is acquired through the pressure sensor corresponding to the leaking pipeline as the third pipeline pressure;

[0078] If the second target pipeline outer wall image of the leaking pipeline is not the first second target pipeline outer wall image, determine whether the second leakage state information corresponding to the second target pipeline outer wall image of the leaking pipeline has changed compared with the second leakage state information corresponding to the second target pipeline outer wall image before the second target pipeline outer wall image of the leaking pipeline, and if changed, use the second leakage state information corresponding to the second target pipeline outer wall image of the leaking pipeline as the third leakage state information, obtain the acquisition time of the second target pipeline outer wall image of the leaking pipeline as the second acquisition time, and obtain the pipeline pressure when the second target pipeline outer wall image of the leaking pipeline is acquired through the pressure sensor corresponding to the leaking pipeline as the third pipeline pressure;

[0079] The step of calculating the leakage amount of the leaking pipeline according to the first leakage state information of the leaking pipeline, the first collection time of the leaking pipeline, the repair time of the leaking pipeline, the first pipeline pressure of the leaking pipeline, and the second pipeline pressure of the leaking pipeline includes:

[0080] The leakage amount of the leaking pipeline is calculated according to the first leakage status information of the leaking pipeline, all the second leakage status information of the leaking pipeline, the first collection time of the leaking pipeline, all the second collection times of the leaking pipeline, the repair time of the leaking pipeline, the first pipeline pressure of the leaking pipeline, all the third pipeline pressures corresponding to the leaking pipeline, and the second pipeline pressure of the leaking pipeline.

[0081] For example, assuming that the leakage point of the leaking pipeline has changed N times from the time the leakage is discovered to the time the manager repairs the leakage, then N+1 leakage states will be generated accordingly. The above formula is used for every two adjacent leakage states, that is, Q leak =Q 漏液点流量 *T, you can get N Q 漏液点流量 , assuming Q 漏液点流量1 , Q 漏液点流量2 …Q 漏液点流量N Therefore, the total leakage amount of the leaking pipeline is Q leak =Q 漏液点流量1 *T 1 +Q 漏液点流量2 *T 2 +…Q 漏液点流量N *T N , where T 1 T is the time difference between the time corresponding to the initial leakage state and the time corresponding to the first leakage state change, 2 is the time difference between the time corresponding to the first leakage state change and the time corresponding to the second leakage state change, ..., T NIt is the time difference between the time corresponding to the N-1th leakage state change and the time corresponding to the Nth leakage state change.

[0082] The method for detecting pipeline leakage in a liquid flow battery system provided in an embodiment of the present application collects the appearance of the pipe wall of the leaking pipeline at a higher frequency after the leaking pipeline is determined, so that the changes in the leakage state of the routing pipeline can be better monitored when the management personnel do not repair it, thereby achieving a more accurate assessment of the leakage amount.

[0083] The embodiment of the present application also provides a device for detecting leakage in a pipeline of a liquid flow battery system, comprising: a memory and a processor, wherein the memory is used to store an executable program;

[0084] The processor is used to read and execute the executable program to implement the method for detecting pipeline leakage in a liquid flow battery system as described in any of the above embodiments.

[0085] The device for detecting leakage in a pipeline in a liquid flow battery system provided in an embodiment of the present application can use an image acquisition device to acquire an image of the appearance of the pipe wall of the pipe in the liquid flow battery system, and use a pre-trained image processing model to determine whether the pipe is leaking and the color of the leakage in the leaking pipe from the acquired image of the appearance of the pipe wall, thereby realizing automatic detection of leaking pipes, saving human resources to a great extent, and improving the efficiency of pipe detection.

[0086] The present application also provides a device for detecting leakage in a pipeline of a flow battery system, such as Figure 3 As shown, the detection device includes: an image sensor, a pressure sensor, an ARM processor, an A / D converter, a host computer and an alarm light.

[0087] The image sensor is used as the front-end input to collect the data of the system pipeline, and then after data processing, it is transmitted to the A / D converter. The A / D converter converts the received signal and then transmits the signal to the ARM processor. The A / D converter calculates the received signal using the Elman neural network algorithm, and then transmits the calculation result to the host computer, which prompts through the alarm light. During the pipeline leakage process, the pressure on the corresponding pipeline will change, so the pressure signal will be periodically obtained through the pressure sensor. The pressure difference between non-leaking and leaking will also pass through the ARM processor to calculate the pipeline leakage amount and transmit the calculation result to the host computer.

[0088] Correspondingly, the detection method of pipeline leakage in the flow battery system includes: collecting the outer wall image of the pipeline in the flow battery system, and then training the neural network model as a training sample to obtain an image processing model. Then set the acquisition time to t1 to obtain the outer wall image of the pipeline in the flow battery system, and input it into the image processing model. When a pipeline leak is found, adjust the acquisition time of the leaking pipeline and define it as t2, where t1>t2, the purpose is to monitor the change of the leakage amount to achieve accuracy.

[0089] The process of processing the pipeline outer wall image by the image processing model is carried out in the ARM processor. The ARM processor obtains the leakage color information of the leaking pipeline corresponding to the pipeline outer wall image with leakage judgment information. If the leakage color information is yellow, the leakage position is located in the positive electrode supply pipeline, if the leakage color information is blue, the leakage position is located in the positive electrode return pipeline, if it is purple, the leakage position is located in the negative electrode supply pipeline, if the leakage color information is green, the leakage position is located in the negative electrode return pipeline. After confirming the position, the next instruction is executed. Here, the positive electrode pipeline is taken as an example. Assuming that the judged image shows yellow, the instruction will be transmitted to the host computer for alarm prompts. There will be a time difference for personnel linkage. At this time, the collection system of the leakage position will continue to work, and then collect what the leakage status is. There are only four states of leakage. The first is occasional leakage, the second is dripping, the third is heavy leakage, and the fourth is flowing. Through the monitoring of the above four states, the leakage amount will be obtained, and then after maintenance, a notification prompt will be issued referring to the estimated amount of replenishment.

[0090] When there is no leakage, the test results can be uploaded to the host computer at intervals of t3 to report the results. Among them, t3 should be much longer than t1. The purpose is to avoid frequent communication and transmission of report results. When leakage occurs, the test results need to be uploaded to the host computer in time to achieve the effect of timely notification of management personnel.

[0091] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all components may be implemented as software executed by a processor, such as a digital signal processor or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

Claims

1. A method for detecting pipeline leakage in a flow battery system, characterized in that: The method comprises: The image acquisition device is used to acquire the appearance image of the pipe wall of the pipe in the liquid flow battery system at a first preset acquisition frequency, and whenever an appearance image of the pipe wall in an acquisition cycle is acquired, the appearance image of the pipe wall in the acquisition cycle is used as the first target appearance image of the pipe wall and the following operations are performed: By inputting all the first target tube wall appearance images obtained into a pre-trained image processing model, leakage determination information of whether the pipeline corresponding to each first target tube wall appearance image is leaking, and leakage color information of the leaking pipeline corresponding to the first target tube wall appearance images with leakage determination information of leakage among all the first target tube wall appearance images; Determine the leaking pipeline according to the leakage color information of the leaking pipeline in the obtained first target pipe wall appearance image; The liquid flow battery system comprises: a positive electrode liquid storage tank, a negative electrode liquid storage tank and pipelines, wherein the pipelines comprise: a first liquid supply pipeline and a first liquid return pipeline connected to the positive electrode liquid storage tank, and a second liquid supply pipeline and a second liquid return pipeline connected to the negative electrode liquid storage tank; The step of determining the leaking pipeline according to the leakage color information of the leaking pipeline in the first target pipe wall appearance image obtained includes: When the obtained leakage color information is yellow, it is determined that the leakage pipeline is the first liquid supply pipeline; When the obtained leakage color information is blue, it is determined that the leakage pipeline is the first liquid return pipeline; When the obtained leakage color information is purple, it is determined that the leakage pipeline is the second liquid supply pipeline; When the obtained liquid leakage color information is green, it is determined that the liquid leakage pipeline is the second liquid return pipeline.

2. The method according to claim 1, characterized in that The image processing model includes: an image processing model configured to determine leakage determination information indicating whether the corresponding pipeline is leaking according to the pipeline outer wall image, and an image processing model wherein the leakage determination information is leakage color information of the leaking pipeline corresponding to the pipeline outer wall image of the leaking pipeline.

3. The method according to claim 2, characterized in that The image processing model is established in the following way: The following operation is performed on each pipeline: a plurality of pipeline outer wall images of the pipeline are collected by an image acquisition device as corresponding sample pipeline outer wall images; wherein the plurality of sample pipeline outer wall images corresponding to the pipeline include: a sample pipeline outer wall image of the pipeline leaking and a sample pipeline outer wall image of the pipeline not leaking; Obtaining leakage determination information of whether the pipeline is leaking in all sample pipeline outer wall images as corresponding leakage determination information, and obtaining leakage color information of the leaking pipeline in the sample pipeline outer wall images with leakage determination information being leakage as corresponding leakage color information; All the obtained sample pipeline outer wall images are taken as input, the leakage judgment information corresponding to all pipeline outer wall images, and the leakage color information corresponding to the sample pipeline outer wall images with the leakage judgment information being leakage are taken as labels of the corresponding sample pipeline outer wall images, and a pre-set neural network model is trained to obtain the image processing model.

4. The method according to claim 1, characterized in that The image processing model includes: an image processing model configured to determine leakage determination information indicating whether the corresponding pipeline is leaking based on the pipeline outer wall image, and an image processing model in which the leakage determination information is leakage color information and leakage state information of the leaking pipeline corresponding to the pipeline outer wall image of the leaking pipeline.

5. The method according to claim 4, characterized in that The image processing model is established in the following way: The following operations are performed on each pipeline: a plurality of pipeline outer wall images of the pipeline are collected by an image acquisition device as corresponding sample pipeline outer wall images; wherein the plurality of sample pipeline outer wall images corresponding to the pipeline include: a sample pipeline outer wall image of the pipeline leaking and a sample pipeline outer wall image of the pipeline not leaking; Obtaining leakage determination information of whether the pipeline is leaking in all sample pipeline outer wall images as corresponding leakage determination information, and obtaining leakage color information of the leaking pipeline in the sample pipeline outer wall images whose leakage determination information is leakage as corresponding leakage color information, and obtaining leakage state information of the leaking pipeline in the sample pipeline outer wall images whose leakage determination information is leakage as corresponding leakage state information; All the obtained sample appearance images are taken as input, the leakage judgment information corresponding to all the sample pipeline outer wall images, and the leakage color information and leakage state information corresponding to the sample pipeline outer wall images with the leakage judgment information being leakage are taken as labels of the corresponding sample pipeline outer wall images, and a pre-set neural network model is trained to obtain the image processing model.

6. The method according to claim 5, characterized in that The method further includes: inputting all the first target tube wall appearance images obtained into a pre-trained image processing model to obtain leakage state information of the leaking pipeline in each first target tube wall appearance image as first leakage state information of each leaking pipeline; Perform the following operations for each leaking pipeline: Acquire the pipeline pressure of the leaking pipeline when there is no leakage through the pressure sensor as the first pipeline pressure, and record the acquisition time of the first target pipe wall appearance image corresponding to the leaking pipeline as the first acquisition time; Notify the manager of the leakage of the leaking pipeline in a preset manner, record the time when the manager repairs the leaking pipeline, and obtain the pipeline pressure of the leaking pipeline after repair through a pressure sensor as the second pipeline pressure; The leakage amount of the leaking pipeline is calculated according to the first leakage state information of the leaking pipeline, the first collection time of the leaking pipeline, the repair time of the leaking pipeline, the first pipeline pressure of the leaking pipeline, and the second pipeline pressure of the leaking pipeline.

7. The method according to claim 6, characterized in that The method further comprises: Perform the following operations for each leaking pipeline: The image of the outer wall of the leaking pipeline is collected by the image acquisition device at a second preset frequency until the manager repairs the leaking pipeline, and the image of the outer wall of the leaking pipeline collected at the second preset frequency is used as the second target pipeline outer wall image, and each time a second target pipeline outer wall image is obtained, the following operations are performed on the second target pipeline outer wall image: Inputting the second target pipeline outer wall image into a pre-trained image processing model, obtaining leakage state information of the leaking pipeline in the second target pipeline outer wall image and using the information as the second leakage state information corresponding to the second target pipeline outer wall image of the leaking pipeline; If the second target pipeline outer wall image of the leaking pipeline is the first second target pipeline outer wall image, determine whether the second leakage state information corresponding to the second target pipeline outer wall image of the leaking pipeline and the first leakage state information of the leaking pipeline have changed; if changed, use the second leakage state information corresponding to the second target pipeline outer wall image of the leaking pipeline as the third leakage state information of the leaking pipeline, obtain the acquisition time of the second target pipeline outer wall image of the leaking pipeline as the second acquisition time, and obtain the pipeline pressure when the second target pipeline outer wall image of the leaking pipeline is acquired through the pressure sensor corresponding to the leaking pipeline as the third pipeline pressure; If the second target pipeline outer wall image of the leaking pipeline is not the first second target pipeline outer wall image, determine whether the second leakage state information corresponding to the second target pipeline outer wall image of the leaking pipeline has changed compared with the second leakage state information corresponding to the second target pipeline outer wall image before the second target pipeline outer wall image of the leaking pipeline, and if changed, use the second leakage state information corresponding to the second target pipeline outer wall image of the leaking pipeline as the third leakage state information, obtain the acquisition time of the second target pipeline outer wall image of the leaking pipeline as the second acquisition time, and obtain the pipeline pressure when the second target pipeline outer wall image of the leaking pipeline is acquired through the pressure sensor corresponding to the leaking pipeline as the third pipeline pressure; The step of calculating the leakage amount of the leaking pipeline according to the first leakage state information of the leaking pipeline, the first collection time of the leaking pipeline, the repair time of the leaking pipeline, the first pipeline pressure of the leaking pipeline, and the second pipeline pressure of the leaking pipeline includes: The leakage amount of the leaking pipeline is calculated according to the first leakage status information of the leaking pipeline, all the second leakage status information of the leaking pipeline, the first collection time of the leaking pipeline, all the second collection times of the leaking pipeline, the repair time of the leaking pipeline, the first pipeline pressure of the leaking pipeline, all the third pipeline pressures corresponding to the leaking pipeline, and the second pipeline pressure of the leaking pipeline.

8. The method according to claim 4, 5 or 6, characterized in that: The leakage state information includes: the leakage point area and the leakage point aspect ratio.

9. A device for detecting pipeline leakage in a flow battery system, characterized in that: include: A memory and a processor, wherein the memory is used to store an executable program; The processor is used to read and execute the executable program to implement the method for detecting pipeline leakage in a liquid flow battery system as described in any one of claims 1 to 8.

Images