A method and system for detecting abnormal alkali solution flow rate in the production of basic nickel carbonate

Abstract

The present invention discloses a method and system for detecting abnormal alkali liquid flow rate for producing basic nickel carbonate. The method comprises: according to the initial alkali liquid video frame at the initial moment, a preset division strategy is used to divide and process the first alkali liquid video frame of each production line at the first moment, and at least one target first alkali liquid video frame set is obtained; according to a preset selection rule, a certain target first alkali liquid video frame is selected from a certain target first alkali liquid video frame set, and the certain target first alkali liquid video frame is used as the first alkali liquid video frame to be tested. After determining that the first alkali liquid flow rate data corresponding to the first alkali liquid video frame to be tested is normal through analysis, it is not necessary to analyze other target first alkali liquid video frames in a certain target first alkali liquid video frame set where the first alkali liquid video frame to be tested is located, thereby reducing the workload of re-analyzing other similar first alkali liquid flow rate data as much as possible.

Description

Technical Field

[0001] The present invention belongs to the technical field of flow detection, and in particular relates to a method and system for detecting abnormal flow of alkali solution used for producing basic nickel carbonate. Background Art

[0002] Basic nickel carbonate products are widely used in precision electroplating, printed circuit board electroplating, general alloy electroplating, nickel-cobalt alloy electroforming, ceramic industry, industrial catalysts and other industries. It has a wide range of applications and a large demand.

[0003] In the process of synthesizing basic nickel carbonate, the pH value condition is very demanding, so it is necessary to accurately grasp the flow rate of each alkali solution, so that the staff can accurately control the pH value condition in time through the alkali solution flow rate. In the prior art, the alkali solution image at each moment is input into a pre-built flow prediction model, and the flow prediction model continuously outputs the flow results corresponding to the alkali solution image, and compares the predicted flow results with the threshold value to determine whether the alkali solution flow at that moment is abnormal. However, for the mass production of basic nickel carbonate, when there are multiple production lines, such a method will undoubtedly increase the recognition amount of the flow prediction model. Therefore, there is an urgent need for a method and system for detecting abnormal alkali solution flow for the production of basic nickel carbonate to monitor the changes in alkali solution flow. Summary of the invention

[0004] The present invention provides a method and system for detecting abnormal flow of alkali solution for producing basic nickel carbonate, which is used to solve the technical problem that the recognition amount of a flow prediction model increases when there are multiple production lines for mass production of basic nickel carbonate.

[0005] In a first aspect, the present invention provides a method for detecting abnormal flow of alkali solution for producing basic nickel carbonate, comprising:

[0006] Acquire initial alkali liquid flow data of each production line at an initial moment, and first alkali liquid flow data at a first moment, wherein the initial alkali liquid flow data includes an initial alkali liquid video frame taken at an alkali liquid outlet at the initial moment, and the first alkali liquid flow data includes a first alkali liquid video frame taken at an alkali liquid outlet at the first moment;

[0007] According to the initial alkali liquid video frame at the initial moment, a preset division strategy is used to divide the first alkali liquid video frame of each production line at the first moment to obtain at least one target first alkali liquid video frame set, wherein a target first alkali liquid video frame set includes at least one target first alkali liquid video frame;

[0008] According to a preset selection rule, a certain target first alkali liquid video frame is selected from a certain target first alkali liquid video frame set, and the certain target first alkali liquid video frame is used as a first alkali liquid video frame to be tested;

[0009] Performing feature extraction on the first alkali solution video frame to be tested to obtain at least one target feature region corresponding to the first alkali solution video frame to be tested, and fusing the at least one target feature region to obtain a target feature region image;

[0010] The target feature area image is input into a pre-built flow recognition model, the flow recognition model outputs an alkali liquid flow value corresponding to the target feature area image, and determines whether the difference between the alkali liquid flow value and a preset dynamic alkali liquid flow threshold is greater than a first preset threshold, wherein the dynamic alkali liquid flow threshold is an alkali liquid flow threshold obtained by dynamic adjustment at different times;

[0011] If it is greater than a first preset threshold, it is determined that there is an abnormality in the first alkali liquid flow data corresponding to the first alkali liquid video frame to be tested, and other target feature area images are input into a pre-built flow recognition model for identification, wherein the other target feature area images are target feature area images obtained by feature extraction and fusion of other target first alkali liquid video frames in the set of the first alkali liquid video frames of a certain target.

[0012] In a second aspect, the present invention provides an abnormal alkali solution flow detection system for producing basic nickel carbonate, comprising:

[0013] an acquisition module, configured to acquire initial alkali liquid flow data of each production line at an initial moment, and first alkali liquid flow data at a first moment, wherein the initial alkali liquid flow data includes an initial alkali liquid video frame taken at an alkali liquid outlet at the initial moment, and the first alkali liquid flow data includes a first alkali liquid video frame taken at an alkali liquid outlet at the first moment;

[0014] a partitioning module configured to partition the first alkali liquid video frames of each production line at the first moment according to the initial alkali liquid video frames at the initial moment using a preset partitioning strategy to obtain at least one target first alkali liquid video frame set, wherein a target first alkali liquid video frame set includes at least one target first alkali liquid video frame;

[0015] A selection module is configured to select a certain target first alkali liquid video frame from a certain target first alkali liquid video frame set according to a preset selection rule, and use the certain target first alkali liquid video frame as a first alkali liquid video frame to be tested;

[0016] a fusion module configured to perform feature extraction on the first alkali solution video frame to be tested, obtain at least one target feature region corresponding to the first alkali solution video frame to be tested, and fuse the at least one target feature region to obtain a target feature region image;

[0017] A judgment module is configured to input the target feature area image into a pre-built flow recognition model, the flow recognition model outputs an alkali liquid flow value corresponding to the target feature area image, and judges whether the difference between the alkali liquid flow value and a preset dynamic alkali liquid flow threshold is greater than a first preset threshold, wherein the dynamic alkali liquid flow threshold is an alkali liquid flow threshold obtained by dynamic adjustment at different times;

[0018] The determination module is configured to determine that the first alkali liquid flow data corresponding to the first alkali liquid video frame to be tested is abnormal if it is greater than a first preset threshold, and input other target feature area images into a pre-built flow recognition model for identification, wherein the other target feature area images are target feature area images obtained by feature extraction and fusion of other target first alkali liquid video frames in the set of the first alkali liquid video frames of a certain target.

[0019] According to a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the steps of the method for detecting abnormal alkali solution flow for producing basic nickel carbonate according to any embodiment of the present invention.

[0020] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the steps of the method for detecting abnormal alkali solution flow for producing basic nickel carbonate according to any embodiment of the present invention.

[0021] The alkali liquid flow anomaly detection method and system for producing basic nickel carbonate of the present application, based on the initial alkali liquid video frame at the initial moment, adopt a preset division strategy to divide and process the first alkali liquid video frame of each production line at the first moment, and obtain at least one target first alkali liquid video frame set, according to the preset selection rule, select a target first alkali liquid video frame in a certain target first alkali liquid video frame set, and use the certain target first alkali liquid video frame as the first alkali liquid video frame to be tested, after determining through analysis that the first alkali liquid flow data corresponding to the first alkali liquid video frame to be tested is normal, there is no need to analyze other target first alkali liquid video frames in a certain target first alkali liquid video frame set where the first alkali liquid video frame to be tested is located, thereby reducing the workload of re-analyzing other similar first alkali liquid flow data as much as possible. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 A flow chart of a method for detecting abnormal alkali solution flow rate for producing basic nickel carbonate provided by one embodiment of the present invention;

[0024] Figure 2 A structural block diagram of an abnormal alkali solution flow detection system for producing basic nickel carbonate provided by one embodiment of the present invention;

[0025] Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] See also Figure 1 , which shows a flow chart of a method for detecting abnormal alkali solution flow for producing basic nickel carbonate of the present application.

[0028] like Figure 1 As shown, the method for detecting abnormal alkali solution flow rate for producing basic nickel carbonate specifically comprises the following steps:

[0029] Step S101, obtaining the initial alkali liquid flow data of each production line at the initial moment, and the first alkali liquid flow data at the first moment, wherein the initial alkali liquid flow data includes an initial alkali liquid video frame at the alkali liquid outlet taken at the initial moment, and the first alkali liquid flow data includes a first alkali liquid video frame at the alkali liquid outlet taken at the first moment.

[0030] In the process of producing basic nickel carbonate, multiple reactors are often used for production at the same time. In this process, the staff uses a controller to connect the control valves in the multiple reactors. By opening the control valves at the same time, the alkali liquid on each production line can flow out at the same time when no abnormal failure occurs. Therefore, the initial moment in this step can be the moment when the control instruction is generated to open the control valve and the alkali liquid flows out, or it can be the moment when the control valve is opened and the alkali liquid flows out after a very short interval. In this way, the initial alkali liquid flow data of each production line at the initial moment can be made the same as much as possible, so as to facilitate the subsequent monitoring of whether the first alkali liquid flow data at the first moment is abnormal.

[0031] Step S102, according to the initial alkali liquid video frame at the initial moment, a preset division strategy is used to divide the first alkali liquid video frame of each production line at the first moment to obtain at least one target first alkali liquid video frame set, wherein a target first alkali liquid video frame set contains at least one target first alkali liquid video frame.

[0032] In this step, an initial alkali liquid video frame and a first alkali liquid video frame of a certain production line are obtained, and pixel difference is performed between a first alkali liquid video frame and an initial alkali liquid video frame to obtain an alkali liquid change video frame; the alkali liquid change video frames corresponding to each production line are input into a preset image recognition model, and the image recognition model outputs feature quantities corresponding to each alkali liquid change video frame; at least two alkali liquid change video frames whose feature quantity change values ​​are not greater than a second preset threshold are clustered, wherein the feature quantity change value is the difference between the feature quantity of one alkali liquid change video frame and the feature quantity of another alkali liquid change video frame; and at least two first alkali liquid video frames corresponding to at least two alkali liquid change video frames are divided into the same target first alkali liquid video frame set according to the clustering result.

[0033] It should be noted that performing pixel difference calculation on a certain first alkali liquid video frame and a certain initial alkali liquid video frame specifically refers to: performing pixel difference calculation on a certain first alkali liquid video frame and a certain initial alkali liquid video frame. By clustering at least two alkali liquid change video frames whose feature quantity change values ​​are not greater than the second preset threshold, and dividing at least two first alkali liquid video frames corresponding to at least two alkali liquid change video frames into the same target first alkali liquid video frame set according to the clustering result, the first alkali liquid flow rate data on the production line with similar alkali liquid flow rate change degrees from the initial moment to the first moment can be clustered together, so that only the first alkali liquid flow rate data that is most likely to be abnormal needs to be analyzed later, reducing the workload of re-analyzing other similar first alkali liquid flow rate data.

[0034] Specifically, the image recognition model may be a VGGNet network model, which is obtained through machine learning using multiple sets of training images and detection images.

[0035] Step S103 : selecting a certain target first alkali liquid video frame from a certain target first alkali liquid video frame set according to a preset selection rule, and taking the certain target first alkali liquid video frame as a first alkali liquid video frame to be tested.

[0036] In this step, a target first alkali liquid video frame in a target first alkali liquid video frame set and an initial alkali liquid video frame corresponding to a target first alkali liquid video frame are respectively input into a preset image recognition model, and the image recognition model outputs a first feature quantity corresponding to the initial alkali liquid video frame and a second feature quantity corresponding to the target first alkali liquid video frame respectively; the feature change rate is calculated according to a feature quantity difference and a time interval, and a target first alkali liquid video frame with the smallest feature change rate is selected from a target first alkali liquid video frame set, wherein the feature quantity difference is the difference between the second feature quantity and the first feature quantity, and the time interval is the time interval between a first moment corresponding to a target first alkali liquid video frame and an initial moment corresponding to an initial alkali liquid video frame.

[0037] It should be noted that, in a set of first alkali liquid video frames of a certain target, the first alkali liquid video frame with the smallest feature change rate is selected, which indicates that the first alkali liquid flow data corresponding to the first alkali liquid video frame of a certain target has the smallest change degree compared with the initial alkali liquid flow data. The reason for this situation may be that the pipeline is blocked or there is a leakage, so that the alkali liquid flow at the alkali liquid outlet is not large. It may also be that the controller controls the control valve in the reactor on the production line to be closed and then opened, resulting in a small alkali liquid flow. Therefore, it is necessary to take the first alkali liquid video frame of a certain target as the first alkali liquid video frame to be tested that may be abnormal for subsequent analysis.

[0038] Step S104: extracting features from the first alkali solution video frame to obtain at least one target feature region corresponding to the first alkali solution video frame to be tested, and fusing the at least one target feature region to obtain a target feature region image.

[0039] In this step, the first alkali liquid video frame to be tested is input into a preset image segmentation model, and the image segmentation model outputs a feature area image corresponding to the first alkali liquid video frame to be tested, wherein the feature area image contains at least one feature area; a sliding window of a preset size is slid on the edge of at least one feature area, and a transition area covered by the sliding window when sliding is intercepted with at least one feature area to obtain at least one target feature area, wherein the transition area is an area connected to and does not overlap with at least one feature area.

[0040] It should be noted that the image segmentation model can be obtained by training the U-Net network using training samples and verification samples.

[0041] Further, a first initial alkali liquid video frame corresponding to the first alkali liquid video frame to be tested is obtained, and the first initial alkali liquid video frame is an initial alkali liquid video frame obtained on the same production line as the first alkali liquid video frame to be tested; at least one target feature area is embedded into the initial alkali liquid video frame according to at least two pixels in the transition area to obtain a target initial alkali liquid video frame; and the target initial alkali liquid video frame is segmented based on the image segmentation model to obtain a target feature area image. Since the alkali liquid feature points in the first alkali liquid video frame to be tested must cover the alkali liquid feature points in the initial alkali liquid video frame, positioning is performed according to at least two pixels in the transition area, and at least one positioned target feature area is embedded in the corresponding position of the initial alkali liquid video frame, so that while no additional alkali liquid feature points are generated, the alkali liquid feature points that may be missed in the feature area image can be better captured, thereby improving the accuracy of the alkali liquid flow value output by the subsequent flow recognition model.

[0042] Step S105, input the target feature area image into a pre-constructed flow recognition model, the flow recognition model outputs an alkali liquid flow value corresponding to the target feature area image, and determines whether the difference between the alkali liquid flow value and a preset dynamic alkali liquid flow threshold is greater than a first preset threshold, wherein the dynamic alkali liquid flow threshold is an alkali liquid flow threshold obtained by dynamic adjustment at different times.

[0043] In this step, the traffic recognition model can be obtained by training a convolutional neural network based on training samples and annotation results.

[0044] It should be noted that the setting rule of the dynamic alkali liquid flow threshold is: set the initial alkali liquid flow threshold at the initial moment, and dynamically increase the initial alkali liquid flow threshold according to the difference between the current moment and the initial moment. For example, the dynamic alkali liquid flow threshold at the initial moment A is a, and the dynamic alkali liquid flow threshold at the first moment B is , k is the threshold adjustment coefficient, and the dynamic alkali solution flow threshold of C at the second moment is Furthermore, during the production process, if the addition of alkali solution is suspended midway, the initial time is also the time when the alkali solution is added after the control valve is opened again.

[0045] In a specific embodiment, if it is not greater than the first preset threshold, it is determined that there is no abnormality in the first alkali liquid flow data corresponding to the first alkali liquid video frame to be tested, and there is no abnormality in the first alkali liquid flow data corresponding to other target first alkali liquid video frames in a target first alkali liquid video frame set where the first alkali liquid video frame to be tested is located.

[0046] Step S106, if it is greater than the first preset threshold, it is determined that the first alkali liquid flow data corresponding to the first alkali liquid video frame to be tested is abnormal, and other target feature area images are input into a pre-constructed flow recognition model for identification, wherein the other target feature area images are target feature area images obtained by feature extraction and fusion of other target first alkali liquid video frames in the set of the first alkali liquid video frames of a certain target.

[0047] In summary, the method of the present application, based on the initial alkali liquid video frame at the initial moment, adopts a preset division strategy to divide and process the first alkali liquid video frame of each production line at the first moment, and obtains at least one target first alkali liquid video frame set; according to a preset selection rule, a target first alkali liquid video frame is selected from a certain target first alkali liquid video frame set, and the certain target first alkali liquid video frame is used as the first alkali liquid video frame to be tested; after determining through analysis that the first alkali liquid flow data corresponding to the first alkali liquid video frame to be tested is normal, there is no need to analyze other target first alkali liquid video frames in a certain target first alkali liquid video frame set where the first alkali liquid video frame to be tested is located, thereby reducing the workload of re-analyzing other similar first alkali liquid flow data as much as possible.

[0048] See also Figure 2 , which shows a structural block diagram of an alkali solution flow anomaly detection system for producing basic nickel carbonate of the present application.

[0049] like Figure 2 As shown, the alkali solution flow abnormality detection system 200 includes an acquisition module 210, a division module 220, a selection module 230, a fusion module 240, a judgment module 250 and a determination module 260.

[0050] Among them, the acquisition module 210 is configured to obtain the initial alkali liquid flow data of each production line at the initial moment, and the first alkali liquid flow data at the first moment, wherein the initial alkali liquid flow data includes the initial alkali liquid video frame taken at the alkali liquid outlet at the initial moment, and the first alkali liquid flow data includes the first alkali liquid video frame taken at the alkali liquid outlet at the first moment; the division module 220 is configured to divide the first alkali liquid video frame of each production line at the first moment according to the initial alkali liquid video frame at the initial moment, using a preset division strategy to divide and process the first alkali liquid video frame of each production line at the first moment, and obtain at least one target first alkali liquid video frame set, wherein a target first alkali liquid video frame set contains at least one target first alkali liquid video frame; the selection module 230 is configured to select a certain target first alkali liquid video frame from a certain target first alkali liquid video frame set according to a preset selection rule, and use the certain target first alkali liquid video frame as the first alkali liquid video frame to be tested; the fusion module 240 is configured to perform special processing on the first alkali liquid video frame to be tested A judging module 250 is configured to input the target feature area image into a pre-constructed flow recognition model, the flow recognition model outputs the alkali liquid flow value corresponding to the target feature area image, and judges whether the difference between the alkali liquid flow value and a preset dynamic alkali liquid flow threshold is greater than a first preset threshold, wherein the dynamic alkali liquid flow threshold is an alkali liquid flow threshold obtained by dynamic adjustment at different times; a determining module 260 is configured to determine that if it is greater than the first preset threshold, the first alkali liquid flow data corresponding to the first alkali liquid video frame to be tested is abnormal, and input other target feature area images into the pre-constructed flow recognition model for identification, wherein the other target feature area images are target feature area images obtained by feature extraction and fusion of other target first alkali liquid video frames in the set of the first alkali liquid video frames of a certain target.

[0051] It should be understood that Figure 2 Modules and references documented in Figure 1 Therefore, the operations and features described above for the method and the corresponding technical effects are also applicable to Figure 2 The modules in it will not be described in detail here.

[0052] In other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor is caused to execute the method for detecting abnormal alkali solution flow rate for producing basic nickel carbonate in any of the above method embodiments;

[0053] As an implementation mode, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows:

[0054] Acquire initial alkali liquid flow data of each production line at an initial moment, and first alkali liquid flow data at a first moment, wherein the initial alkali liquid flow data includes an initial alkali liquid video frame taken at an alkali liquid outlet at the initial moment, and the first alkali liquid flow data includes a first alkali liquid video frame taken at an alkali liquid outlet at the first moment;

[0055] According to the initial alkali liquid video frame at the initial moment, a preset division strategy is used to divide the first alkali liquid video frame of each production line at the first moment to obtain at least one target first alkali liquid video frame set, wherein a target first alkali liquid video frame set includes at least one target first alkali liquid video frame;

[0056] According to a preset selection rule, a certain target first alkali liquid video frame is selected from a certain target first alkali liquid video frame set, and the certain target first alkali liquid video frame is used as a first alkali liquid video frame to be tested;

[0057] Performing feature extraction on the first alkali solution video frame to be tested to obtain at least one target feature region corresponding to the first alkali solution video frame to be tested, and fusing the at least one target feature region to obtain a target feature region image;

[0058] The target feature area image is input into a pre-built flow recognition model, the flow recognition model outputs an alkali liquid flow value corresponding to the target feature area image, and determines whether the difference between the alkali liquid flow value and a preset dynamic alkali liquid flow threshold is greater than a first preset threshold, wherein the dynamic alkali liquid flow threshold is an alkali liquid flow threshold obtained by dynamic adjustment at different times;

[0059] If it is greater than a first preset threshold, it is determined that there is an abnormality in the first alkali liquid flow data corresponding to the first alkali liquid video frame to be tested, and other target feature area images are input into a pre-built flow recognition model for identification, wherein the other target feature area images are target feature area images obtained by feature extraction and fusion of other target first alkali liquid video frames in the set of the first alkali liquid video frames of a certain target.

[0060] The computer-readable storage medium may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function; the storage data area may store data created according to the use of the alkali solution flow anomaly detection system for producing basic nickel carbonate, etc. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the alkali solution flow anomaly detection system for producing basic nickel carbonate via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0061] Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, the device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 3 In the example of connecting through a bus. The memory 320 is the above-mentioned computer-readable storage medium. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, the above-mentioned method embodiment is used to detect the abnormal flow rate of alkali solution in the production of basic nickel carbonate. The input device 330 can receive input digital or character information, and generate key signal input related to user settings and function control of the alkali flow rate abnormality detection system for producing basic nickel carbonate. The output device 340 may include a display device such as a display screen.

[0062] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0063] As an embodiment, the electronic device is applied to an abnormal alkali solution flow detection system for producing basic nickel carbonate, and is used for a client, comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can:

[0064] Acquire initial alkali liquid flow data of each production line at an initial moment, and first alkali liquid flow data at a first moment, wherein the initial alkali liquid flow data includes an initial alkali liquid video frame taken at an alkali liquid outlet at the initial moment, and the first alkali liquid flow data includes a first alkali liquid video frame taken at an alkali liquid outlet at the first moment;

[0065] According to the initial alkali liquid video frame at the initial moment, a preset division strategy is used to divide the first alkali liquid video frame of each production line at the first moment to obtain at least one target first alkali liquid video frame set, wherein a target first alkali liquid video frame set includes at least one target first alkali liquid video frame;

[0066] According to a preset selection rule, a certain target first alkali liquid video frame is selected from a certain target first alkali liquid video frame set, and the certain target first alkali liquid video frame is used as a first alkali liquid video frame to be tested;

[0067] Performing feature extraction on the first alkali solution video frame to be tested to obtain at least one target feature region corresponding to the first alkali solution video frame to be tested, and fusing the at least one target feature region to obtain a target feature region image;

[0068] The target feature area image is input into a pre-built flow recognition model, the flow recognition model outputs an alkali liquid flow value corresponding to the target feature area image, and determines whether the difference between the alkali liquid flow value and a preset dynamic alkali liquid flow threshold is greater than a first preset threshold, wherein the dynamic alkali liquid flow threshold is an alkali liquid flow threshold obtained by dynamic adjustment at different times;

[0069] If it is greater than a first preset threshold, it is determined that there is an abnormality in the first alkali liquid flow data corresponding to the first alkali liquid video frame to be tested, and other target feature area images are input into a pre-built flow recognition model for identification, wherein the other target feature area images are target feature area images obtained by feature extraction and fusion of other target first alkali liquid video frames in the set of the first alkali liquid video frames of a certain target.

[0070] Through the description of the above implementation modes, those skilled in the art can clearly understand that each implementation mode can be implemented by means of software plus a necessary general hardware platform, or of course by hardware. Based on such an understanding, the above technical solution can essentially or in other words be embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting abnormal flow of alkali solution for producing basic nickel carbonate, characterized in that: include: Acquire initial alkali liquid flow data of each production line at an initial moment, and first alkali liquid flow data at a first moment, wherein the initial alkali liquid flow data includes an initial alkali liquid video frame taken at an alkali liquid outlet at the initial moment, and the first alkali liquid flow data includes a first alkali liquid video frame taken at an alkali liquid outlet at the first moment; According to the initial alkali liquid video frame at the initial moment, a preset division strategy is used to divide the first alkali liquid video frame of each production line at the first moment to obtain at least one target first alkali liquid video frame set, wherein a target first alkali liquid video frame set includes at least one target first alkali liquid video frame; According to a preset selection rule, a certain target first alkali liquid video frame is selected from a certain target first alkali liquid video frame set, and the certain target first alkali liquid video frame is used as a first alkali liquid video frame to be tested; Performing feature extraction on the first alkali solution video frame to be tested to obtain at least one target feature region corresponding to the first alkali solution video frame to be tested, and fusing the at least one target feature region to obtain a target feature region image; The target feature area image is input into a pre-built flow recognition model, the flow recognition model outputs an alkali liquid flow value corresponding to the target feature area image, and determines whether the difference between the alkali liquid flow value and a preset dynamic alkali liquid flow threshold is greater than a first preset threshold, wherein the dynamic alkali liquid flow threshold is an alkali liquid flow threshold obtained by dynamic adjustment at different times; If it is greater than a first preset threshold, it is determined that there is an abnormality in the first alkali liquid flow data corresponding to the first alkali liquid video frame to be tested, and other target feature area images are input into a pre-built flow recognition model for identification, wherein the other target feature area images are target feature area images obtained by feature extraction and fusion of other target first alkali liquid video frames in the set of the first alkali liquid video frames of a certain target.

2. The method for detecting abnormal alkali solution flow rate for producing basic nickel carbonate according to claim 1, characterized in that: The first alkali liquid video frames of each production line at the first moment are divided and processed by using a preset division strategy according to the initial alkali liquid video frames at the initial moment to obtain at least one target first alkali liquid video frame set, including: Acquire an initial alkali solution video frame and a first alkali solution video frame of a production line, and perform pixel difference between the first alkali solution video frame and the initial alkali solution video frame to obtain an alkali solution change video frame; Inputting the alkali solution change video frames corresponding to each production line into a preset image recognition model, and the image recognition model outputs the feature values ​​corresponding to each alkali solution change video frame; Clustering at least two alkali solution change video frames whose feature quantity change values ​​are not greater than a second preset threshold, wherein the feature quantity change value is a difference between a feature quantity of one alkali solution change video frame and a feature quantity of another alkali solution change video frame; According to the clustering result, at least two first alkali liquid video frames corresponding to the at least two alkali liquid change video frames are divided into the same target first alkali liquid video frame set.

3. The method for detecting abnormal alkali solution flow rate for producing basic nickel carbonate according to claim 1, characterized in that: The step of selecting a target first alkali solution video frame from a target first alkali solution video frame set according to a preset selection rule includes: Inputting a certain target first alkali liquid video frame in the certain target first alkali liquid video frame set and a certain initial alkali liquid video frame corresponding to the certain target first alkali liquid video frame into a preset image recognition model respectively, and the image recognition model outputs a first feature value corresponding to the certain initial alkali liquid video frame and a second feature value corresponding to the certain target first alkali liquid video frame respectively; The feature change rate is calculated according to a certain feature quantity difference and a certain time interval, and a certain target first alkali liquid video frame with the smallest feature change rate is selected from the certain target first alkali liquid video frame set, wherein the certain feature quantity difference is the difference between the second feature quantity and the first feature quantity, and the certain time interval is the time interval between the first moment corresponding to the certain target first alkali liquid video frame and the initial moment corresponding to the certain initial alkali liquid video frame.

4. The method for detecting abnormal alkali solution flow rate for producing basic nickel carbonate according to claim 1, characterized in that: The extracting features of the first alkali solution video frame to be tested to obtain at least one target feature area corresponding to the first alkali solution video frame to be tested comprises: Inputting the first alkali solution video frame to be tested into a preset image segmentation model, the image segmentation model outputting a feature region image corresponding to the first alkali solution video frame to be tested, the feature region image including at least one feature region; A sliding window of a preset size is slid on the edge of the at least one feature area, and a transition area covered by the sliding window when sliding is intercepted with the at least one feature area to obtain at least one target feature area, wherein the transition area is an area connected to and does not overlap with the at least one feature area.

5. The method for detecting abnormal alkali solution flow rate for producing basic nickel carbonate according to claim 4, characterized in that: The fusing the at least one target feature region to obtain a target feature region image comprises: Acquire a first initial alkali liquid video frame corresponding to the first alkali liquid video frame to be tested, wherein the first initial alkali liquid video frame is an initial alkali liquid video frame acquired on the same production line as the first alkali liquid video frame to be tested; Embedding the at least one target feature region into the initial alkali solution video frame according to at least two pixel points in the transition region to obtain a target initial alkali solution video frame; The target initial alkali solution video frame is segmented based on the image segmentation model to obtain a target feature area image.

6. The method for detecting abnormal alkali solution flow rate for producing basic nickel carbonate according to claim 1, characterized in that: After determining whether the difference between the alkali solution flow rate value and the preset dynamic alkali solution flow rate threshold is greater than a first preset threshold, the method further includes: If it is not greater than the first preset threshold, it is determined that there is no abnormality in the first alkali liquid flow data corresponding to the first alkali liquid video frame to be tested, and there is no abnormality in the first alkali liquid flow data corresponding to other target first alkali liquid video frames in a target first alkali liquid video frame set where the first alkali liquid video frame to be tested is located.

7. A system for detecting abnormal flow of alkali solution for producing basic nickel carbonate, characterized in that: include: an acquisition module, configured to acquire initial alkali liquid flow data of each production line at an initial moment, and first alkali liquid flow data at a first moment, wherein the initial alkali liquid flow data includes an initial alkali liquid video frame taken at an alkali liquid outlet at the initial moment, and the first alkali liquid flow data includes a first alkali liquid video frame taken at an alkali liquid outlet at the first moment; a partitioning module configured to partition the first alkali liquid video frames of each production line at the first moment according to the initial alkali liquid video frames at the initial moment using a preset partitioning strategy to obtain at least one target first alkali liquid video frame set, wherein a target first alkali liquid video frame set includes at least one target first alkali liquid video frame; A selection module is configured to select a certain target first alkali liquid video frame from a certain target first alkali liquid video frame set according to a preset selection rule, and use the certain target first alkali liquid video frame as a first alkali liquid video frame to be tested; a fusion module configured to perform feature extraction on the first alkali solution video frame to be tested, obtain at least one target feature region corresponding to the first alkali solution video frame to be tested, and fuse the at least one target feature region to obtain a target feature region image; A judgment module is configured to input the target feature area image into a pre-built flow recognition model, the flow recognition model outputs an alkali liquid flow value corresponding to the target feature area image, and judges whether the difference between the alkali liquid flow value and a preset dynamic alkali liquid flow threshold is greater than a first preset threshold, wherein the dynamic alkali liquid flow threshold is an alkali liquid flow threshold obtained by dynamic adjustment at different times; The determination module is configured to determine that the first alkali liquid flow data corresponding to the first alkali liquid video frame to be tested is abnormal if it is greater than a first preset threshold, and input other target feature area images into a pre-built flow recognition model for identification, wherein the other target feature area images are target feature area images obtained by feature extraction and fusion of other target first alkali liquid video frames in the set of the first alkali liquid video frames of a certain target.

8. An electronic device, characterized in that: include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images