Filter tank cleaning monitoring method and device
The filter video data is captured by the camera and input the filter cleaning monitoring model for monitoring, which solves the problem of long filter cleaning time and relying on manual monitoring, realizes automatic monitoring, improves monitoring accuracy and saves human resources.
Patent Information
- Application Number
- CN202510174369.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-18
- Publication Date
- 2025-05-23
AI Technical Summary
In the prior art, filter cleaning requires a long time and relies on manual monitoring, which is prone to discontinuous cleaning due to human errors, and cannot ensure the effectiveness of filter cleaning.
By driving multiple cameras to capture video data on the filter and inputting these data into the trained filter cleaning monitoring model for monitoring, obtaining drug-dose monitoring results and backflush monitoring results, thereby realizing automated monitoring of the filter cleaning process.
It reduces the manpower required to monitor filter cleaning, saves human resources, and improves the accuracy of cleaning monitoring, avoids mistakes in manual monitoring.
Smart Images
Figure CN120034624A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of filter clarity monitoring, and in particular to a filter cleaning monitoring method and device. Background Art
[0002] Filters are used for filtering purposes. Some are used to remove suspended matter in water to obtain water with lower turbidity; some are used to remove water from sludge to obtain sludge with lower water content. There are many types of filters, which can be divided into fast filters and slow filters according to the filtration rate. In practical applications, most of them are fast filters. Fast filters have a large processing capacity and good effluent quality. Cleaning the filter is an important way to maintain the filtering capacity of the filter. However, in the prior art, cleaning the filter takes a long time, and manual observation and monitoring of the cleaning of the filter wastes a lot of human resources. Once the observer is distracted or goes to the bathroom, the cleaning of the filter will lack monitoring, and it is impossible to ensure that the cleaning of the filter is continuous. Summary of the invention
[0003] The purpose of this application is to provide a filter tank cleaning monitoring method and device that can overcome the shortcomings and deficiencies in the prior art.
[0004] A first aspect of an embodiment of the present application provides a filter tank cleaning monitoring method, comprising:
[0005] Driving multiple cameras to shoot multiple filter tanks respectively to obtain filter tank video data corresponding to each filter tank;
[0006] The filter tank video data are respectively input into the trained filter tank cleaning monitoring model for monitoring, and a plurality of dosing monitoring results and backwashing monitoring results are obtained;
[0007] The cleaning monitoring result of the filter pool is obtained based on the dosing monitoring result and the backwashing monitoring result of the same filter pool.
[0008] Further, the filter tank video data is respectively input into the trained filter tank cleaning monitoring model for monitoring, and the steps of obtaining multiple dosing monitoring results and backwashing monitoring results include:
[0009] The filter tank video data are respectively input into the corresponding trained filter tank cleaning monitoring model for water flow monitoring. If the filter tank cleaning monitoring model detects that the water flow in the filter tank video data is flowing normally, the drug addition monitoring result is determined to be normal. If the filter tank cleaning monitoring model detects that the water flow in the filter tank video data stops flowing, the drug addition monitoring result is determined to be abnormal.
[0010] Further, the filter tank video data is respectively input into the trained filter tank cleaning monitoring model for monitoring, and the steps of obtaining multiple dosing monitoring results and backwashing monitoring results include:
[0011] The filter tank video data is respectively input into the corresponding trained filter tank cleaning monitoring model for liquid level height monitoring and bubble monitoring. If the liquid level of the filter tank is higher than the preset height advance and bubbles continue to be generated on the liquid surface of the filter tank, the backwash monitoring result is determined to be normal; otherwise, the backwash monitoring result is determined to be abnormal.
[0012] Further, the filter tank cleaning monitoring model is trained by the following steps:
[0013] Acquire multiple training samples; the training samples are provided with water flow marks, liquid level height marks and bubble marks;
[0014] The initial monitoring model of the laboratory server cluster is trained according to the multiple sample groups to obtain the filter tank cleaning monitoring model.
[0015] Furthermore, the initial monitoring model includes a backbone network, a neck network and a head network connected in sequence; the backbone network includes a convolutional layer, a plurality of convolutional attention layers and a spatial pooling layer;
[0016] The step of training the initial monitoring model of the laboratory server cluster according to the multiple sample groups to obtain the filter tank cleaning monitoring model includes:
[0017] Inputting the training sample into the convolution layer for feature extraction to obtain a first sample feature;
[0018] Inputting the first sample feature into the plurality of cascaded convolutional attention layers for feature extraction and feature attention processing to obtain second sample features output by each convolutional attention layer;
[0019] Inputting the second sample feature output by the last convolutional attention layer into the spatial pooling layer for dimensionality reduction processing to obtain a pooling feature;
[0020] Inputting the pooled features and a plurality of preset second sample features into the neck network for fusion processing and spatial transformation attention processing to obtain a plurality of fused attention features;
[0021] The multiple fused attention features are input into the head network for predictive learning processing to obtain a trained filter tank cleaning monitoring model.
[0022] Further, the convolutional attention layer includes a convolution module and a dual attention module;
[0023] The convolution module performs feature extraction processing on the input to obtain a third sample feature;
[0024] The dual attention module performs feature attention processing on the third sample feature of the same convolutional attention layer to obtain the second sample feature, and the dual attention module outputs the second sample feature to the convolution module of the next-level convolutional attention layer.
[0025] Further, the dual attention module includes an enhanced convolution submodule, a dual adaptive attention submodule and a first superposition fusion submodule;
[0026] The enhanced convolution submodule performs convolution enhancement processing on the third sample feature to obtain a fourth sample feature;
[0027] The dual-adaptive attention submodule performs feature attention fusion processing on the third sample feature to obtain a fifth sample feature;
[0028] The first superposition and fusion submodule performs superposition and fusion processing on the third sample feature, the fourth sample feature and the fifth sample feature to obtain the second sample feature.
[0029] Further, the dual-adaptive attention submodule includes: a first convolution activation unit, a second convolution activation unit, a decomposition attention fusion unit and a feature fusion unit;
[0030] The first convolution activation unit performs convolution activation processing on the third sample feature to obtain a first activation feature;
[0031] The second convolution activation unit performs convolution activation processing on the first activation feature to obtain a second activation feature;
[0032] The decomposition and attention fusion unit performs decomposition processing and attention fusion convolution processing on the first activation feature and the second activation feature to obtain an attention fusion feature;
[0033] The feature fusion unit fuses the first activation feature, the second activation feature and the attention fusion feature to obtain the fifth sample feature.
[0034] Further, the decomposition attention fusion unit includes a decomposition subunit, a fusion subunit, an attention normalization processing subunit and an attention fusion convolution subunit;
[0035] The decomposition subunit is used to decompose the first activation feature and the second activation feature to obtain a first decomposition feature and a second decomposition feature;
[0036] The fusion subunit fuses the first decomposition feature and the second decomposition feature to obtain a decomposition fusion feature;
[0037] The attention normalization processing subunit performs attention processing and normalization processing on the decomposed fusion features to obtain attention normalized features;
[0038] The attention fusion convolution subunit performs fusion convolution processing on the first decomposition feature, the second decomposition feature and the attention normalization feature to obtain the attention fusion feature.
[0039] A second aspect of an embodiment of the present application provides a filter tank cleaning monitoring device, comprising:
[0040] The filter pool video data acquisition module is used to drive multiple cameras to respectively shoot multiple filter pools to obtain filter pool video data corresponding to each filter pool;
[0041] A monitoring module, used for inputting the filter tank video data into the trained filter tank cleaning monitoring model for monitoring, and obtaining a plurality of dosing monitoring results and backwashing monitoring results;
[0042] The cleaning monitoring result acquisition module is used to obtain the cleaning monitoring result of the filter pool according to the dosing monitoring result and the backwashing monitoring result of the same filter pool.
[0043] Compared with the prior art, the present application can input the filter pool video data captured by the camera into the trained filter pool cleaning monitoring model for monitoring, obtain multiple dosing monitoring results and backwashing monitoring results, and then obtain the cleaning monitoring results of the filter pool based on the dosing monitoring results and the backwashing monitoring results of the same filter pool. It is a technical solution for monitoring filter pool cleaning through filter pool video data and filter pool cleaning monitoring models, which can reduce the manpower required to monitor filter pool cleaning, save human resources, reduce errors in manual monitoring, and improve the monitoring accuracy of filter pool cleaning.
[0044] In order to provide a clearer understanding of the present application, the specific implementation of the present application will be described below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] Figure 1 This is a flow chart of a filter tank cleaning monitoring method according to an embodiment of the present application.
[0046] Figure 2 A schematic diagram of a filter tank cleaning monitoring model according to an embodiment of the present application.
[0047] Figure 3 This is a schematic diagram of the module connections of a filter tank cleaning monitoring device according to one embodiment of the present application.
[0048] 100. Filter tank video data acquisition module; 200. Monitoring module; 300. Cleaning monitoring result acquisition module. DETAILED DESCRIPTION
[0049] In order to make the objectives, technical solutions and advantages of the present application more clear, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.
[0050] It should be clear that the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the embodiments of the present application.
[0051] When the following description relates to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. In the description of the present application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances. The singular forms of "a", "said" and "the" used in the present application and the appended claims are also intended to include the majority form, unless the context clearly indicates other meanings. The words "if" / "if" used herein can be interpreted as "at the time of" or "when" or "in response to determination".
[0052] In addition, in the description of this application, unless otherwise specified, "plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects are in an "or" relationship.
[0053] See also Figure 1 , which is a flow chart of a filter tank cleaning monitoring method according to an embodiment of the present application, comprising:
[0054] S1: driving multiple cameras to respectively shoot multiple filter tanks to obtain filter tank video data corresponding to each filter tank.
[0055] Among them, one filter pool corresponds to at least one camera.
[0056] S2: Inputting the filter tank video data into the trained filter tank cleaning monitoring model for monitoring, and obtaining a plurality of dosing monitoring results and backwashing monitoring results.
[0057] The filter tank cleaning monitoring model is trained by the following steps:
[0058] A plurality of training samples are obtained; the training samples are provided with water flow marks, liquid level height marks and bubble marks.
[0059] The initial monitoring model of the laboratory server cluster is trained according to the multiple sample groups to obtain the filter tank cleaning monitoring model.
[0060] S3: Obtaining the cleaning monitoring result of the filter pool according to the dosing monitoring result and the backwashing monitoring result of the same filter pool.
[0061] In a feasible embodiment, the filter tank video data is input into the trained filter tank cleaning monitoring model for monitoring, and the steps of obtaining multiple dosing monitoring results and backwashing monitoring results include:
[0062] The filter tank video data are respectively input into the corresponding trained filter tank cleaning monitoring model for water flow monitoring. If the filter tank cleaning monitoring model detects that the water flow in the filter tank video data is flowing normally, the drug addition monitoring result is determined to be normal. If the filter tank cleaning monitoring model detects that the water flow in the filter tank video data stops flowing, the drug addition monitoring result is determined to be abnormal.
[0063] Among them, the filter tank cleaning monitoring model can determine whether the water flows normally based on the liquid surface ripples in the filter tank video data.
[0064] In a feasible embodiment, the filter tank video data is input into the trained filter tank cleaning monitoring model for monitoring, and the steps of obtaining multiple dosing monitoring results and backwashing monitoring results include:
[0065] The filter tank video data is respectively input into the corresponding trained filter tank cleaning monitoring model for liquid level height monitoring and bubble monitoring. If the liquid level of the filter tank is higher than the preset height advance and bubbles continue to be generated on the liquid surface of the filter tank, the backwash monitoring result is determined to be normal; otherwise, the backwash monitoring result is determined to be abnormal.
[0066] Among them, the filter tank cleaning monitoring model can determine whether bubbles continue to be generated on the liquid surface of the filter tank based on the liquid level bubble monitoring results of multiple consecutive video frames in the filter tank video data.
[0067] Compared with the prior art, the present application can input the filter pool video data captured by the camera into the trained filter pool cleaning monitoring model for monitoring, obtain multiple dosing monitoring results and backwashing monitoring results, and then obtain the cleaning monitoring results of the filter pool based on the dosing monitoring results and the backwashing monitoring results of the same filter pool. It is a technical solution for monitoring filter pool cleaning through filter pool video data and filter pool cleaning monitoring models, which can reduce the manpower required to monitor filter pool cleaning, save human resources, reduce errors in manual monitoring, and improve the monitoring accuracy of filter pool cleaning.
[0068] See also Figure 2 In a feasible embodiment, the initial monitoring model includes a backbone network (Backbone), a neck network (Neck) and a head network (Head) connected in sequence; the backbone network includes a convolutional layer, multiple convolutional attention layers and a spatial pooling layer;
[0069] The step of training the initial monitoring model of the laboratory server cluster according to the multiple sample groups to obtain the filter tank cleaning monitoring model includes:
[0070] Inputting the training sample into the convolution layer for feature extraction to obtain a first sample feature;
[0071] Inputting the first sample feature into the plurality of cascaded convolutional attention layers for feature extraction and feature attention processing to obtain second sample features output by each convolutional attention layer;
[0072] Inputting the second sample feature output by the last convolutional attention layer into the spatial pooling layer for dimensionality reduction processing to obtain a pooling feature;
[0073] Inputting the pooled features and a plurality of preset second sample features into the neck network for fusion processing and spatial transformation attention processing to obtain a plurality of fused attention features;
[0074] The multiple fused attention features are input into the head network for predictive learning processing to obtain a trained filter tank cleaning monitoring model.
[0075] In a feasible embodiment, the convolutional attention layer includes a convolutional module and a dual attention module, wherein there are multiple convolutional modules, such as P1, P2, P3, etc.
[0076] The convolution module performs feature extraction processing on the input to obtain a third sample feature;
[0077] The dual attention module performs feature attention processing on the third sample feature of the same convolutional attention layer to obtain the second sample feature, and the dual attention module outputs the second sample feature to the convolution module of the next-level convolutional attention layer.
[0078] In a feasible embodiment, the dual attention module includes an enhanced convolution submodule, a dual adaptive attention submodule and a first superposition fusion submodule;
[0079] The enhanced convolution submodule performs convolution enhancement processing on the third sample feature to obtain a fourth sample feature; wherein the process is shown in the following formula:
[0080] ecb_output = γ ECB ·ECB(BN(X A ))
[0081] ecb_output is the fourth sample feature; X A is the third sample feature; BN(·) is the normalization operation; ECB(·) is the convolution enhancement operation; γ ECB Learned parameters for convolutional augmentation.
[0082] The dual-adaptive attention submodule performs feature attention fusion processing on the third sample feature to obtain a fifth sample feature; wherein the process is shown in the following formula:
[0083] attention_output = γ DAA ·DAA(BN(X A ))
[0084] Among them, attention_output is the fifth sample feature; DAA(·) is the feature attention fusion operation; γ DAA are the learning parameters for feature attention fusion.
[0085] The first superposition and fusion submodule performs superposition and fusion processing on the third sample feature, the fourth sample feature and the fifth sample feature to obtain the second sample feature.
[0086] In a feasible embodiment, the dual adaptive attention submodule includes: a first convolution activation unit, a second convolution activation unit, a decomposition attention fusion unit and a feature fusion unit;
[0087] The first convolution activation unit performs convolution activation processing on the third sample feature to obtain a first activation feature; wherein the process is shown in the following formula:
[0088]
[0089] Among them, X1 is the first activated feature; X A is the third sample feature; BN(·) is the normalization operation; It is a convolution operation, a normalization operation, and an activation operation.
[0090] The second convolution activation unit performs convolution activation processing on the first activation feature to obtain a second activation feature; wherein the process is shown in the following formula:
[0091]
[0092] Among them, X 2 is the second activation feature; It is a convolution operation, a normalization operation, and an activation operation.
[0093] The decomposition and attention fusion unit performs decomposition processing and attention fusion convolution processing on the first activation feature and the second activation feature to obtain an attention fusion feature;
[0094] The feature fusion unit fuses the first activation feature, the second activation feature and the attention fusion feature to obtain the fifth sample feature; wherein the process is shown in the following formula:
[0095]
[0096] Among them, attention_output is the fifth sample feature; γ DAA is the learning parameter of feature attention fusion; SC is the attention fusion feature; X 1 is the first activated feature; X 2 is the second activated feature.
[0097] In a feasible embodiment, the decomposition attention fusion unit includes a decomposition subunit, a fusion subunit, an attention normalization processing subunit and an attention fusion convolution subunit;
[0098] The decomposition subunit is used to decompose the first activation feature and the second activation feature to obtain a first decomposition feature and a second decomposition feature;
[0099] The fusion subunit fuses the first decomposition feature and the second decomposition feature to obtain a decomposition fusion feature;
[0100] The attention normalization processing subunit performs attention processing and normalization processing on the decomposed fusion features to obtain attention normalized features;
[0101] The attention fusion convolution subunit performs fusion convolution processing on the first decomposition feature, the second decomposition feature and the attention normalization feature to obtain the attention fusion feature.
[0102] Among them, the training sample is input into the convolution layer for feature extraction to obtain the first sample feature; the first sample feature is input into the cascaded multiple convolution attention layers for feature extraction and feature attention processing to obtain the second sample feature output by each convolution attention layer; the second sample feature output by the last convolution attention layer is input into the spatial pooling layer for dimensionality reduction processing to obtain the pooling feature; the pooling feature and several preset second sample features are input into the neck network for fusion processing and spatial transformation attention processing to obtain multiple fusion attention features; the multiple fusion attention features are input into the head network for predictive learning processing to obtain the trained filter cleaning monitoring model, which has the ability to perform fusion processing and spatial change attention processing according to the second sample features and pooling features of different spatial dimensions. In addition, through the above training process, the feature capture ability of the model has been significantly improved. This not only enables the model to more accurately identify subtle changes in the filter, but also greatly improves the accuracy and efficiency of monitoring. Therefore, a filter cleaning monitoring model with excellent performance can be trained, which can accurately monitor the cleaning of the filter in a complex industrial environment, thereby ensuring that the filter is fully cleaned.
[0103] See also Figure 3 The second embodiment of the present application provides a filter tank cleaning monitoring device, comprising:
[0104] The filter pool video data acquisition module 100 is used to drive multiple cameras to respectively shoot multiple filter pools to obtain filter pool video data corresponding to each filter pool;
[0105] The monitoring module 200 is used to input the filter tank video data into the trained filter tank cleaning monitoring model for monitoring, and obtain multiple dosing monitoring results and backwashing monitoring results;
[0106] The cleaning monitoring result acquisition module 300 is used to obtain the cleaning monitoring result of the filter pool according to the dosing monitoring result and the backwashing monitoring result of the same filter pool.
[0107] It should be noted that the filter cleaning monitoring device provided in the second embodiment of the present application only uses the division of the above-mentioned functional modules as an example when executing the filter cleaning monitoring method. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the equipment is divided into different functional modules to complete all or part of the functions described above. In addition, the filter cleaning monitoring device provided in the second embodiment of the present application and the filter cleaning monitoring method of the first embodiment of the present application belong to the same concept. The implementation process thereof is detailed in the method embodiment and will not be repeated here.
[0108] The device embodiments described above are merely illustrative, wherein the components described as separate parts may or may not be physically separated, and the parts displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present application. Ordinary technicians in this field can understand and implement it without creative work.
[0109] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.
[0110] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the function selected in the process. Figure 1 A process or multiple processes and / or boxes Figure 1function selected in a box or multiple boxes.
[0111] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 steps for the function selected in a box or multiple boxes.
[0112] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0113] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0114] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0115] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0116] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.
Claims
1. A filter tank cleaning monitoring method, characterized in that: include: Driving multiple cameras to shoot multiple filter tanks respectively to obtain filter tank video data corresponding to each filter tank; The filter tank video data are respectively input into the trained filter tank cleaning monitoring model for monitoring, and a plurality of dosing monitoring results and backwashing monitoring results are obtained; The cleaning monitoring result of the filter pool is obtained based on the dosing monitoring result and the backwashing monitoring result of the same filter pool.
2. The filter tank cleaning monitoring method according to claim 1, characterized in that: The filter tank video data is input into the trained filter tank cleaning monitoring model for monitoring, and the steps of obtaining multiple dosing monitoring results and backwashing monitoring results include: The filter tank video data are respectively input into the corresponding trained filter tank cleaning monitoring model for water flow monitoring. If the filter tank cleaning monitoring model detects that the water flow in the filter tank video data is flowing normally, the drug addition monitoring result is determined to be normal. If the filter tank cleaning monitoring model detects that the water flow in the filter tank video data stops flowing, the drug addition monitoring result is determined to be abnormal.
3. The filter tank cleaning monitoring method according to claim 1, characterized in that: The filter tank video data is input into the trained filter tank cleaning monitoring model for monitoring, and the steps of obtaining multiple dosing monitoring results and backwashing monitoring results include: The filter tank video data is respectively input into the corresponding trained filter tank cleaning monitoring model for liquid level height monitoring and bubble monitoring. If the liquid level of the filter tank is higher than the preset height advance and bubbles continue to be generated on the liquid surface of the filter tank, the backwash monitoring result is determined to be normal; otherwise, the backwash monitoring result is determined to be abnormal.
4. The filter tank cleaning monitoring method according to claim 1, characterized in that: The filter tank cleaning monitoring model is trained by the following steps: Acquire multiple training samples; the training samples are provided with water flow marks, liquid level height marks and bubble marks; The initial monitoring model of the laboratory server cluster is trained according to the multiple sample groups to obtain the filter tank cleaning monitoring model.
5. The filter tank cleaning monitoring method according to claim 4, characterized in that: The initial monitoring model includes a backbone network, a neck network and a head network connected in sequence; the backbone network includes a convolutional layer, multiple convolutional attention layers and a spatial pooling layer; The step of training the initial monitoring model of the laboratory server cluster according to the multiple sample groups to obtain the filter tank cleaning monitoring model includes: Inputting the training sample into the convolution layer for feature extraction to obtain a first sample feature; Inputting the first sample feature into the plurality of cascaded convolutional attention layers for feature extraction and feature attention processing to obtain second sample features output by each convolutional attention layer; Inputting the second sample feature output by the last convolutional attention layer into the spatial pooling layer for dimensionality reduction processing to obtain a pooling feature; Inputting the pooled features and a plurality of preset second sample features into the neck network for fusion processing and spatial transformation attention processing to obtain a plurality of fused attention features; The multiple fused attention features are input into the head network for predictive learning processing to obtain a trained filter tank cleaning monitoring model.
6. The filter tank cleaning monitoring method according to claim 5, characterized in that: The convolutional attention layer includes a convolution module and a dual attention module; The convolution module performs feature extraction processing on the input to obtain a third sample feature; The dual attention module performs feature attention processing on the third sample feature of the same convolutional attention layer to obtain the second sample feature, and the dual attention module outputs the second sample feature to the convolution module of the next-level convolutional attention layer.
7. The filter tank cleaning monitoring method according to claim 6, characterized in that: The dual attention module includes an enhanced convolution submodule, a dual adaptive attention submodule and a first superposition fusion submodule; The enhanced convolution submodule performs convolution enhancement processing on the third sample feature to obtain a fourth sample feature; The dual-adaptive attention submodule performs feature attention fusion processing on the third sample feature to obtain a fifth sample feature; The first superposition and fusion submodule performs superposition and fusion processing on the third sample feature, the fourth sample feature and the fifth sample feature to obtain the second sample feature.
8. The filter tank cleaning monitoring method according to claim 7, characterized in that: The dual-adaptive attention submodule includes: a first convolution activation unit, a second convolution activation unit, a decomposition attention fusion unit and a feature fusion unit; The first convolution activation unit performs convolution activation processing on the third sample feature to obtain a first activation feature; The second convolution activation unit performs convolution activation processing on the first activation feature to obtain a second activation feature; The decomposition and attention fusion unit performs decomposition processing and attention fusion convolution processing on the first activation feature and the second activation feature to obtain an attention fusion feature; The feature fusion unit fuses the first activation feature, the second activation feature and the attention fusion feature to obtain the fifth sample feature.
9. The filter tank cleaning monitoring method according to claim 8, characterized in that: The decomposition attention fusion unit includes a decomposition subunit, a fusion subunit, an attention normalization processing subunit and an attention fusion convolution subunit; The decomposition subunit is used to decompose the first activation feature and the second activation feature to obtain a first decomposition feature and a second decomposition feature; The fusion subunit fuses the first decomposition feature and the second decomposition feature to obtain a decomposition fusion feature; The attention normalization processing subunit performs attention processing and normalization processing on the decomposed fusion features to obtain attention normalized features; The attention fusion convolution subunit performs fusion convolution processing on the first decomposition feature, the second decomposition feature and the attention normalization feature to obtain the attention fusion feature.
10. A filter tank cleaning monitoring device, characterized in that: include: The filter pool video data acquisition module is used to drive multiple cameras to respectively shoot multiple filter pools to obtain filter pool video data corresponding to each filter pool; A monitoring module, used for inputting the filter tank video data into the trained filter tank cleaning monitoring model for monitoring, and obtaining a plurality of dosing monitoring results and backwashing monitoring results; The cleaning monitoring result acquisition module is used to obtain the cleaning monitoring result of the filter pool according to the dosing monitoring result and the backwashing monitoring result of the same filter pool.