Cleaning control method and device for high-voltage frequency converter

Abstract

The invention provides a cleaning control method and device for a high-voltage frequency converter, and relates to the technical field of cleaning control, and the method comprises the steps: obtaining a target frequency converter assembly; obtaining a bilateral monitoring result of the target component; generating a target component cleaning practice instruction; when the target component cleaning practice instruction is a dust removal cleaning practice instruction, generating a target component dust removal decision, and performing cleaning control on the target frequency converter component; and when the target component cleaning practice instruction is a decontamination cleaning practice instruction, generating a target component decontamination decision, and performing cleaning control on the target frequency converter component. The technical problem that the cleaning effect is poor due to the fact that dust and dirt are difficult to recognize and clean at the same time and the cleaning scheme is inaccurate in setting in the prior art can be solved, timely dust removal and dirt removal control is achieved, and the purposes of improving the dust removal and dirt removal control accuracy of the high-voltage frequency converter and improving the cleaning efficiency of the high-voltage frequency converter are achieved. And the cleaning degree of the high-voltage frequency converter is ensured.

Description

Technical Field

[0001] The present application relates to the field of cleaning control technology, and in particular to a cleaning control method and device for a high-voltage frequency converter. Background Art

[0002] High-voltage inverters play an important role in the power system, responsible for converting electrical energy from one frequency to another to meet the operating requirements of different equipment or systems. However, due to the complex and changeable working environment of high-voltage inverters, their surfaces often accumulate dirt such as oil and dust, which not only affects the heat dissipation performance of the equipment, but also poses a potential threat to the normal operation of the equipment.

[0003] High-voltage inverters are prone to dust particles, as well as contaminants such as oil stains and chemicals. In the prior art, dust or dirt is often identified, and it is difficult to identify and clean both dust and dirt at the same time. In addition, the cleaning plan is often set based on manual experience, and the cleaning plan setting is inaccurate, resulting in poor cleaning effect. Summary of the invention

[0004] The purpose of the present application is to provide a cleaning control method and device for a high-voltage inverter, so as to solve the technical problems in the prior art that it is difficult to identify and clean dust and dirt at the same time, and the cleaning scheme is not set accurately, resulting in poor cleaning effect.

[0005] In view of the above problems, the present application provides a cleaning control method and device for a high-voltage inverter.

[0006] In a first aspect, the present application provides a cleaning control method for a high-voltage inverter, the method being implemented by a cleaning control device for a high-voltage inverter, wherein the method comprises: obtaining a target inverter component according to a high-voltage inverter component distribution, wherein the high-voltage inverter component distribution comprises a plurality of components of a high-voltage inverter, and the target inverter component is any one of the high-voltage inverter component distribution; monitoring the target inverter component in real time according to a multi-source monitoring terminal to obtain a bilateral monitoring result of the target component, wherein the multi-source monitoring terminal comprises a dust monitoring array and an image monitoring array, and the bilateral monitoring result of the target component comprises target dust feature data and target image feature data; according to the bilateral cleaning The practice analysis channel performs cleaning practice feature analysis on the bilateral monitoring results of the target component to generate a cleaning practice instruction for the target component; when the cleaning practice instruction for the target component is a dust removal cleaning practice instruction, the dust removal analysis of the target inverter component is performed according to the bilateral inverter cleaning decision channel to generate a target component dust removal decision, and the target inverter component is cleaned and controlled according to the target component dust removal decision; when the cleaning practice instruction for the target component is a contamination removal cleaning practice instruction, the contamination removal analysis of the target inverter component is performed according to the bilateral inverter cleaning decision channel to generate a target component contamination removal decision, and the target inverter component is cleaned and controlled according to the target component contamination removal decision.

[0007] In a second aspect, the present application further provides a cleaning control device for a high-voltage inverter, which is used to execute a cleaning control method for a high-voltage inverter as described in the first aspect, wherein the device includes: a target inverter component acquisition module, which is used to obtain a target inverter component according to a high-voltage inverter component distribution, wherein the high-voltage inverter component distribution includes multiple components of a high-voltage inverter, and the target inverter component is any one of the high-voltage inverter component distribution; a bilateral monitoring result acquisition module, which is used to monitor the target inverter component in real time according to a multi-source monitoring terminal to obtain a bilateral monitoring result of the target component, wherein the multi-source monitoring terminal includes a dust monitoring array and an image monitoring array, and the bilateral monitoring result of the target component includes target dust feature data and target image feature data; cleaning practice feature An analysis module is used to perform cleaning practice feature analysis on the bilateral monitoring results of the target component according to the bilateral cleaning practice analysis channel to generate a cleaning practice instruction for the target component; a dust removal analysis module is used to perform dust removal analysis of the target inverter component according to the bilateral inverter cleaning decision channel when the cleaning practice instruction of the target component is a dust removal cleaning practice instruction, generate a dust removal decision for the target component, and perform cleaning control on the target inverter component according to the dust removal decision for the target component; a pollution removal analysis module is used to perform pollution removal analysis of the target inverter component according to the bilateral inverter cleaning decision channel when the cleaning practice instruction of the target component is a pollution removal cleaning practice instruction, generate a pollution removal decision for the target component, and perform cleaning control on the target inverter component according to the pollution removal decision for the target component.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: According to the distribution of high-voltage inverter components, a target inverter component is obtained, wherein the distribution of high-voltage inverter components includes multiple components of the high-voltage inverter, and the target inverter component is any one of the distribution of high-voltage inverter components; according to the multi-source monitoring end, the target inverter component is monitored in real time to obtain a bilateral monitoring result of the target component, wherein the multi-source monitoring end includes a dust monitoring array and an image monitoring array, and the bilateral monitoring result of the target component includes target dust feature data and target image feature data; according to the bilateral cleaning practice analysis channel, the bilateral monitoring result of the target component is subjected to cleaning practice feature analysis, Generate a cleaning execution instruction for the target component; when the cleaning execution instruction for the target component is a dust removal cleaning execution instruction, perform dust removal analysis of the target inverter component according to the bilateral inverter cleaning decision channel, generate a target component dust removal decision, and perform cleaning control on the target inverter component according to the target component dust removal decision; when the cleaning execution instruction for the target component is a dirt removal cleaning execution instruction, perform dirt removal analysis of the target inverter component according to the bilateral inverter cleaning decision channel, generate a target component dirt removal decision, and perform cleaning control on the target inverter component according to the target component dirt removal decision. Thus, by respectively identifying dust and dirt, generating dust removal cleaning execution instructions and dirt removal cleaning execution instructions, respectively optimizing dust removal schemes and dirt removal schemes, and realizing timely dust removal and dirt removal control, achieving the technical effect of improving the accuracy of dust removal and dirt removal control of high-voltage inverters and ensuring the cleanliness of high-voltage inverters.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0011] Figure 1 A schematic flow chart of a cleaning control method for a high-voltage inverter according to the present application; Figure 2 This is a structural schematic diagram of a cleaning control device for a high-voltage inverter in the present application.

[0012] Explanation of the reference numerals: target inverter component acquisition module 11, bilateral monitoring result acquisition module 12, cleaning practice feature analysis module 13, dust removal analysis module 14, pollution removal analysis module 15. DETAILED DESCRIPTION

[0013] The present application provides a cleaning control method and device for a high-voltage frequency converter, which solves the technical problem in the prior art that it is difficult to identify and clean dust and dirt at the same time, and the cleaning scheme is not set accurately, resulting in poor cleaning effect. By identifying dust and dirt separately, generating dust removal cleaning execution instructions and dirt removal cleaning execution instructions, optimizing dust removal schemes and dirt removal schemes respectively, timely dust removal and dirt removal control is achieved, so as to achieve the technical effect of improving the accuracy of dust removal and dirt removal control of the high-voltage frequency converter and ensuring the cleanliness of the high-voltage frequency converter.

[0014] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0015] Embodiment 1 Please see attached Figure 1 The present application provides a cleaning control method for a high-voltage frequency converter, wherein the method is applied to a cleaning control device for a high-voltage frequency converter, and the method specifically comprises the following steps: Step 1: obtaining a target inverter component according to a high-voltage inverter component distribution, wherein the high-voltage inverter component distribution includes a plurality of high-voltage inverter components, and the target inverter component is any one of the high-voltage inverter component distribution; Specifically, in modern industrial automation and power systems, high-voltage inverter is an important device that can convert high-voltage AC power in the power grid into AC power with adjustable frequency and amplitude. The high-voltage inverter is composed of multiple components, including but not limited to rectifiers, intermediate DC links, inverters, control units, protection devices, etc. The specific components need to be determined in combination with actual conditions, so that the multiple components that constitute the high-voltage inverter are distributed as high-voltage inverter components, and any one of the components is extracted as the target inverter component.

[0016] Step 2: monitoring the target inverter component in real time according to the multi-source monitoring terminal to obtain a bilateral monitoring result of the target component, wherein the multi-source monitoring terminal includes a dust monitoring array and an image monitoring array, and the bilateral monitoring result of the target component includes target dust feature data and target image feature data; Specifically, the dust monitoring array includes a plurality of dust sensors arranged on the target inverter component, which are used to monitor dust concentration, particle size distribution and other parameters in real time as target dust feature data. The image monitoring array includes a plurality of image acquisition devices for acquiring images, such as smart cameras, which capture the appearance image data of the target inverter component in real time as target image feature data. The target image feature data can intuitively reflect the surface stains of the target component. The target dust feature data and the target image feature data constitute the bilateral monitoring results of the target component.

[0017] Step 3: Perform cleaning practice feature analysis on the bilateral monitoring results of the target component according to the bilateral cleaning practice analysis channel, and generate cleaning practice instructions for the target component; Specifically, the bilateral cleaning practice analysis channel includes a dust removal cleaning practice analysis channel and a contamination removal cleaning practice analysis channel, both of which are built based on existing machine learning models. Among them, the dust removal cleaning practice analysis channel determines whether dust removal is required after analyzing the target dust feature data, and obtains dust removal cleaning practice instructions; the contamination removal cleaning practice analysis channel determines whether contamination removal is required after analyzing the target image feature data, and obtains contamination removal cleaning practice instructions. The target component cleaning practice instructions include dust removal cleaning practice instructions or contamination removal cleaning practice instructions.

[0018] Step 4: When the target component cleaning execution instruction is a dust removal cleaning execution instruction, dust removal analysis of the target inverter component is performed according to the bilateral inverter cleaning decision channel, a target component dust removal decision is generated, and cleaning control of the target inverter component is performed according to the target component dust removal decision; Specifically, the bilateral inverter cleaning decision channel includes an inverter dust removal cleaning decision channel and an inverter contamination removal cleaning decision channel. Simply put, dust removal is to remove dust, particulate matter, etc. on the surface of the inverter, and contamination removal is to remove oil stains, chemical residues, etc. on the surface of the inverter. The cleaning devices corresponding to dust removal and contamination removal are also different. When the target component cleaning execution instruction is a dust removal cleaning execution instruction, the dust removal analysis of the target inverter component is performed according to the inverter dust removal cleaning decision channel in the bilateral inverter cleaning decision channel to generate a target component dust removal decision. The target component dust removal decision refers to a dust removal plan, such as dust removal control parameters. Finally, the target inverter component is cleaned and controlled according to the target component dust removal decision to achieve cleaning control of the target inverter component.

[0019] Step 5: When the target component cleaning execution instruction is a decontamination cleaning execution instruction, a decontamination analysis of the target inverter component is performed according to the bilateral inverter cleaning decision channel, a target component decontamination decision is generated, and the target inverter component is cleaned and controlled according to the target component decontamination decision.

[0020] Similarly, when the target component cleaning execution instruction is a decontamination cleaning execution instruction, the decontamination analysis of the target inverter component is performed according to the inverter decontamination cleaning decision channel in the bilateral inverter cleaning decision channel to generate a target component decontamination decision, which is a decontamination plan, such as a decontamination control parameter. Finally, the target inverter component is cleaned and controlled according to the target component decontamination decision. Furthermore, step three of this application also includes: Activate the dust removal cleaning practice analysis channel within the bilateral cleaning practice analysis channel, wherein the dust removal cleaning practice analysis channel includes a dust removal cleaning necessity identification subchannel and a dust removal cleaning necessity judgment subchannel; perform a dust removal demand depth evaluation on the target dust feature data according to the dust removal cleaning necessity identification subchannel, and generate a target dust removal cleaning necessity coefficient; input the target dust removal cleaning necessity coefficient into the dust removal cleaning necessity judgment subchannel, wherein the dust removal cleaning necessity judgment subchannel includes a dust removal cleaning necessity threshold; if the target dust removal cleaning necessity coefficient is greater than / equal to the dust removal cleaning necessity threshold, generate the dust removal cleaning practice instruction, and add the dust removal cleaning practice instruction to the target component cleaning practice instruction; if the target dust removal cleaning necessity coefficient is less than the dust removal cleaning necessity threshold, generate a tracking dust monitoring instruction.

[0021] Specifically, the process of performing cleaning practice feature analysis on the bilateral monitoring results of the target component according to the bilateral cleaning practice analysis channel and generating cleaning practice instructions for the target component is as follows: The dust removal and cleaning practice analysis channel includes a dust removal and cleaning necessity identification subchannel and a dust removal and cleaning necessity judgment subchannel, wherein the dust removal and cleaning necessity identification subchannel is used to perform dust removal demand analysis, and outputs a dust removal and cleaning necessity coefficient that can represent the degree of dust removal demand, and the dust removal and cleaning necessity judgment subchannel stores a dust removal and cleaning necessity threshold value, which is used to compare whether the dust removal and cleaning necessity coefficient is greater than or equal to the dust removal and cleaning necessity threshold value, so as to judge whether dust removal is required, and the dust removal and cleaning necessity threshold value is set by professionals and technicians in this field based on historical experience. The dust removal and cleaning necessity identification subchannel can be constructed based on existing machine learning models, such as neural networks, decision trees, etc. For example, professionals and technicians in this field obtain a large number of dust feature data samples, and obtain dust removal and cleaning necessity coefficient samples based on existing technologies, and use the neural network model to construct the dust removal and cleaning necessity identification subchannel, use the dust feature data samples as training input, and use the dust removal and cleaning necessity coefficient samples for output supervision and adjustment, and train the dust removal and cleaning necessity identification subchannel until convergence.

[0022] The target dust feature data is input into the dust removal and cleaning necessity identification sub-channel to perform a dust removal demand depth evaluation, and a target dust removal and cleaning necessity coefficient is output. The target dust removal and cleaning necessity coefficient is then input into the dust removal and cleaning necessity judgment sub-channel, and the dust removal and cleaning necessity judgment sub-channel compares the target dust removal and cleaning necessity coefficient with the dust removal and cleaning necessity threshold. If the target dust removal and cleaning necessity coefficient is greater than / equal to the dust removal and cleaning necessity threshold, it means that there is a lot of dust, particulate matter, etc. on the target inverter component, and it needs to be cleaned. The dust removal and cleaning execution instruction is generated, and the dust removal and cleaning execution instruction is added to the target component cleaning execution instruction. Subsequently, dust removal analysis and control are performed according to the dust removal and cleaning execution instruction.

[0023] If the target dust removal and cleaning necessary coefficient is less than the dust removal and cleaning necessary threshold, a tracking dust monitoring instruction is generated, which is used to control the dust monitoring array to continue to monitor the dust of the target inverter component, and to continuously identify the necessity of dust removal through the dust removal and cleaning practice analysis channel. When the target dust removal and cleaning necessary coefficient is greater than / equal to the dust removal and cleaning necessary threshold, a corresponding dust removal and cleaning practice instruction is generated. In this way, dust removal necessity identification is realized, providing support for subsequent dust removal analysis and control.

[0024] Furthermore, the present application also includes the following steps: Activate the contamination removal cleaning practice analysis channel in the bilateral cleaning practice analysis channel, wherein the contamination removal cleaning practice analysis channel includes an adaptive enhancement subchannel, a contamination removal cleaning necessity identification subchannel and a contamination removal cleaning necessity judgment subchannel; enhance the target image feature data according to the adaptive enhancement subchannel to generate enhanced image feature data; perform contamination feature recognition and evaluation on the enhanced image feature data according to the contamination removal cleaning necessity identification subchannel to generate a target contamination feature recognition result and a target contamination removal cleaning necessity coefficient; input the target contamination removal cleaning necessity coefficient into the contamination removal cleaning necessity judgment subchannel, wherein the contamination removal cleaning necessity judgment subchannel includes a contamination removal cleaning necessity threshold; if the target contamination removal cleaning necessity coefficient is greater than / equal to the contamination removal cleaning necessity threshold, generate the contamination removal cleaning practice instruction, and add the contamination removal cleaning practice instruction to the target component cleaning practice instruction; if the target contamination removal cleaning necessity coefficient is less than the contamination removal cleaning necessity threshold, generate a tracking image monitoring instruction.

[0025] Furthermore, the present application also includes: The adaptive enhancement subchannel includes an adaptive enhancement function, and the adaptive enhancement function is: ; Among them, EIFD(x, y) represents the enhanced image feature data, DLG represents the brightness enhancement coefficient, IFD(x, y) represents the brightness feature layer of the target image feature data, ALG represents the detail enhancement coefficient, and IFA(x, y) represents the detail feature layer of the target image feature data.

[0026] Specifically, the cleaning practice feature analysis is performed on the bilateral monitoring results of the target component according to the bilateral cleaning practice analysis channel to generate the cleaning practice instruction of the target component, and also includes: The decontamination cleaning practice analysis channel in the bilateral cleaning practice analysis channel is activated to perform decontamination necessity identification on the target image feature data. The decontamination cleaning practice analysis channel includes an adaptive enhancement subchannel, a decontamination cleaning necessity identification subchannel and a decontamination cleaning necessity judgment subchannel. The adaptive enhancement subchannel is used to perform image enhancement processing on the target image feature data to improve image quality and highlight potential dirty areas.

[0027] The adaptive enhancement subchannel includes an adaptive enhancement function, and the adaptive enhancement function is: ; Among them, EIFD (x, y) represents the enhanced image feature data, that is, the image feature data obtained after processing the target image feature data through the adaptive enhancement function, DLG represents the brightness enhancement coefficient, which can be set by a technician in this field based on actual experience and is used to adjust the brightness level of the image. IFD (x, y) represents the brightness feature layer of the target image feature data, which represents the brightness information of the pixel point located at the (x, y) coordinate in the original target image feature data, and ALG represents the detail enhancement coefficient, which is used to enhance the detail features of the image, making the texture and edges of the image clearer, and is set by a technician in this field based on actual experience. IFA (x, y) represents the detail feature layer of the target image feature data, which represents the detail information of the pixel point located at the (x, y) coordinate in the original target image feature data, such as edges and textures. By constructing an adaptive enhancement subchannel, a model support is provided for image enhancement processing, the signal-to-noise ratio of the target image feature data is improved, and then the accuracy of dirt recognition is improved, and then the accuracy of cleaning control is improved.

[0028] The target image feature data is enhanced according to the adaptive enhancement function in the adaptive enhancement subchannel, and the enhanced image feature data is output. Further, the enhanced image feature data is evaluated for dirt feature recognition through the dirt removal and cleaning necessity identification subchannel, that is, the dirt type and area and other parameters are identified as the target dirt feature recognition result, the current dirtiness level is analyzed, and the target dirt removal and cleaning necessity coefficient is output. The target dirt removal and cleaning necessity coefficient represents the dirtiness level. The greater the dirtiness level, the greater the target dirt removal and cleaning necessity coefficient. Specifically, the dirt removal and cleaning necessity identification subchannel can be constructed based on an existing machine learning model. For example, the dirt removal and cleaning necessity identification subchannel can be constructed based on a convolutional neural network model. By obtaining image feature data samples and corresponding dirt feature recognition result samples and dirt removal and cleaning necessity coefficient samples, the dirt removal and cleaning necessity identification subchannel is trained. The training of the machine learning model is a common technical means used by those skilled in the art and will not be expanded here.

[0029] The target necessary coefficient for decontamination and cleaning is input into the necessary judgment sub-channel for decontamination and cleaning. The necessary judgment sub-channel for decontamination and cleaning contains a preset necessary threshold for decontamination and cleaning, which is used to judge whether cleaning is required. The necessary threshold for decontamination and cleaning is set by professional and technical personnel in this field based on actual experience, and refers to the critical value of the necessary coefficient for decontamination and cleaning when cleaning is required. If the target necessary coefficient for decontamination and cleaning is greater than / equal to the necessary threshold for decontamination and cleaning, it means that decontamination and cleaning are required, thereby generating the decontamination and cleaning practice instruction, and adding the decontamination and cleaning practice instruction to the target component cleaning practice instruction. If the target necessary coefficient for decontamination and cleaning is less than the necessary threshold for decontamination and cleaning, a tracking image monitoring instruction is generated, and the image monitoring array is controlled by the tracking image monitoring instruction to continuously monitor the target inverter component to ensure that it can be cleaned in time when needed in the future. Thus, by performing necessary decontamination identification, support is provided for subsequent decontamination analysis and control.

[0030] Furthermore, step 4 of this application also includes: When the target component cleaning execution instruction is a dust removal cleaning execution instruction, the inverter dust removal cleaning decision channel in the bilateral inverter cleaning decision channel is activated, wherein the inverter dust removal cleaning decision channel includes a dust removal decision analysis sub-channel, a dust removal decision deduction sub-channel, a dust removal deduction judgment sub-channel and a dust removal decision optimization sub-channel; according to the dust removal decision analysis sub-channel, the dust removal analysis is performed in combination with the target dust feature data to generate a component dust removal plan; according to the dust removal decision deduction sub-channel, the component dust removal plan is deduced to generate a deduction plan deduced dust removal index; inputting the deduced dust removal index into the dust removal deduction and judgment subchannel, wherein the dust removal deduction and judgment subchannel includes dust removal feature constraints; if the deduced dust removal index satisfies the dust removal feature constraints, adding the component dust removal scheme to the target component dust removal decision; if the deduced dust removal index does not satisfy the dust removal feature constraints, iteratively optimizing and adjusting the component dust removal scheme according to the dust removal decision optimization subchannel, the dust removal decision deduction subchannel and the dust removal deduction and judgment subchannel to obtain the target component dust removal decision.

[0031] Specifically, when the target component cleaning execution instruction is a dust removal cleaning execution instruction, the dust removal analysis of the target inverter component is performed according to the bilateral inverter cleaning decision channel, and the process of generating the target component dust removal decision is as follows: First, activate the inverter dust removal cleaning decision channel in the bilateral inverter cleaning decision channel, wherein the inverter dust removal cleaning decision channel includes a dust removal decision analysis sub-channel, a dust removal decision deduction sub-channel, a dust removal deduction judgment sub-channel and a dust removal decision optimization sub-channel. The dust removal decision analysis sub-channel is used to perform dust removal analysis and generate component dust removal plans. The component dust removal plan includes dust removal parameters. Exemplarily, inverter dust removal is generally performed through dust removal equipment. The dust removal equipment includes a dust removal box, a dust removal hood, a fan, a filter, a dust collector, etc. The dust removal parameters are control parameters such as air volume, wind speed, and wind pressure that control the operation of the dust removal equipment. Specifically, by obtaining historical dust removal plans and corresponding historical dust feature data, a dust removal decision analysis sub-channel can be constructed based on the existing machine learning model training, so that the dust removal decision analysis sub-channel can generate dust removal plans based on dust feature data to obtain component dust removal plans.

[0032] The dust removal decision deduction subchannel will deduce the generated component dust removal scheme. The deduction process may include simulating the implementation process of the dust removal scheme, predicting the dust removal effect, and generating a deduced dust removal index. The deduced dust removal index is a quantitative indicator used to measure the expected effect of the deduced component dust removal scheme. Specifically, a large number of dust removal scheme samples and deduced dust removal index samples can be obtained based on the existing technology, wherein the deduced dust removal index samples can be obtained based on the degree of difference between the dust feature data after and before the execution of the dust removal scheme samples, that is, 1 minus the degree of difference, and the result can be used as the deduced dust removal index sample. Then, the dust removal scheme samples and the deduced dust removal index samples are used as training data, and a dust removal decision deduction subchannel is constructed and trained based on existing machine learning models, such as neural network models, decision trees, etc., to obtain a dust removal decision deduction subchannel. Then, the component dust removal scheme is deduced through the dust removal decision deduction subchannel, and the deduced dust removal index is output.

[0033] Further, the deduced dust removal index is input into the dust removal deduction judgment subchannel, wherein the dust removal deduction judgment subchannel includes a dust removal feature constraint, and the dust removal feature constraint refers to the deduced dust removal index range in which the dust removal effect of the target inverter component reaches the expected value, which is obtained by professional and technical personnel in this field in combination with practical experience. The dust removal deduction judgment subchannel compares the deduced dust removal index with the dust removal feature constraint. If the deduced dust removal index satisfies the dust removal feature constraint, the component dust removal scheme is added to the target component dust removal decision for dust removal control, that is, the dust removal equipment is controlled through the target component dust removal decision.

[0034] If the deduced dust removal index does not meet the dust removal feature constraint, it means that the component dust removal scheme cannot achieve the expected dust removal effect. At this time, the component dust removal scheme is iteratively optimized and adjusted according to the dust removal decision optimization sub-channel, the dust removal decision deduction sub-channel and the dust removal deduction judgment sub-channel. Specifically, the dust removal decision optimization sub-channel is based on the existing intelligent optimization algorithm, such as particle swarm optimization algorithm, ant colony algorithm, simulated annealing algorithm, etc. Through the dust removal decision optimization sub-channel, the component dust removal scheme is adjusted according to a preset step size, including adjusting control parameters according to a preset step size, etc., to obtain an updated component dust removal scheme, and the updated component dust removal scheme is input into the dust removal decision deduction sub-channel for re-deduction, and an updated deduced dust removal index is generated and input into the dust removal deduction judgment sub-channel for judgment. If the updated deduced dust removal index still does not meet the dust removal feature constraint, repeat the analysis and perform iterative optimization adjustment until a dust removal scheme that meets the dust removal feature constraint is found and added to the target component dust removal decision. This enables the optimal decision-making of the dust removal plan and achieves the technical effect of improving the accuracy of dust removal control.

[0035] Furthermore, step five of this application also includes: When the target component cleaning execution instruction is a decontamination cleaning execution instruction, the inverter decontamination cleaning decision channel in the bilateral inverter cleaning decision channel is activated, wherein the inverter decontamination cleaning decision channel includes a decontamination decision analysis subchannel, a decontamination decision deduction subchannel, a decontamination deduction judgment subchannel and a decontamination decision optimization subchannel; according to the decontamination decision analysis subchannel, a decontamination analysis is performed in combination with the target contamination feature recognition result to generate a component decontamination plan; according to the decontamination decision deduction subchannel, the component decontamination plan is deduced to generate a decontamination plan deduced pollution removal index; inputting the deduced pollution removal index into the pollution removal deduction and judgment subchannel, wherein the pollution removal deduction and judgment subchannel includes pollution removal feature constraints; if the deduced pollution removal index satisfies the pollution removal feature constraints, adding the component pollution removal scheme to the target component pollution removal decision; if the deduced pollution removal index does not satisfy the pollution removal feature constraints, iteratively optimizing and adjusting the component pollution removal scheme according to the pollution removal decision optimization subchannel, the pollution removal decision deduction subchannel and the pollution removal deduction and judgment subchannel to obtain the target component pollution removal decision.

[0036] Specifically, when the target component cleaning execution instruction is a decontamination cleaning execution instruction, the decontamination analysis of the target inverter component is performed according to the bilateral inverter cleaning decision channel, and the process of generating the target component decontamination decision is as follows: First, activate the inverter decontamination cleaning decision channel in the bilateral inverter cleaning decision channel, wherein the inverter decontamination cleaning decision channel includes a decontamination decision analysis subchannel, a decontamination decision deduction subchannel, a decontamination deduction judgment subchannel and a decontamination decision optimization subchannel. The decontamination decision analysis subchannel is used to perform decontamination analysis and generate a component decontamination plan. The component decontamination plan includes decontamination parameters. Exemplarily, inverter decontamination is generally performed by decontamination equipment, such as high-pressure cleaning equipment. The decontamination parameters are control parameters such as water pressure and water volume that control the operation of the decontamination equipment. Specifically, by obtaining historical decontamination plans and corresponding historical contamination feature recognition results, a decontamination decision analysis subchannel can be constructed based on the existing machine learning model training, so that the decontamination decision analysis subchannel can generate a decontamination plan based on the target contamination feature recognition results to obtain a component decontamination plan. The training of the machine learning model is a common technical means used by those skilled in the art and will not be expanded here.

[0037] The decontamination decision deduction subchannel will deduce the generated component decontamination plan. The deduction process may include simulating the implementation process of the decontamination plan, predicting the decontamination effect, and generating a deduced decontamination index. The deduced decontamination index is a quantitative indicator used to measure the expected effect of the deduced component decontamination plan. Specifically, a large number of decontamination plan samples and deduced decontamination index samples can be obtained based on the existing technology, wherein the deduced decontamination index samples can be obtained based on the degree of difference between the dirt feature recognition results after and before the execution of the decontamination plan samples, that is, 1 minus the degree of difference, and the result can be used as the deduced decontamination index sample. Then, the decontamination plan samples and the deduced decontamination index samples are used as training data, and the decontamination decision deduction subchannel is constructed and trained based on the existing machine learning models, such as neural network models, decision trees, etc., to obtain the decontamination decision deduction subchannel. Then, the component decontamination plan is deduced through the decontamination decision deduction subchannel, and the deduced decontamination index is output.

[0038] Further, the deduced pollution removal index is input into the pollution removal deduction judgment subchannel, wherein the pollution removal deduction judgment subchannel includes pollution removal characteristic constraints, and the pollution removal characteristic constraints refer to the deduced pollution removal index range in which the pollution removal effect of the target inverter component reaches the expected value, which is obtained by professional and technical personnel in this field in combination with practical experience. The pollution removal deduction judgment subchannel compares the deduced pollution removal index with the pollution removal characteristic constraint. If the deduced pollution removal index satisfies the pollution removal characteristic constraint, the component pollution removal scheme is added to the target component pollution removal decision for pollution removal control, that is, the pollution removal equipment is controlled through the target component pollution removal decision.

[0039] If the deduced pollution removal index does not meet the pollution removal feature constraint, it means that the component pollution removal scheme cannot achieve the expected pollution removal effect. At this time, the component pollution removal scheme is iteratively optimized and adjusted according to the pollution removal decision optimization sub-channel, the pollution removal decision deduction sub-channel and the pollution removal deduction judgment sub-channel. Specifically, the pollution removal decision optimization sub-channel is based on the existing intelligent optimization algorithm, such as particle swarm optimization algorithm, ant colony algorithm, simulated annealing algorithm, etc., through the pollution removal decision optimization sub-channel, the component pollution removal scheme is adjusted according to a preset step size, including adjusting control parameters according to a preset step size, etc., to obtain an updated component pollution removal scheme, and the updated component pollution removal scheme is input into the pollution removal decision deduction sub-channel for re-deduction, and an updated deduced pollution removal index is generated and input into the pollution removal deduction judgment sub-channel for judgment. If the updated deduced pollution removal index still does not meet the pollution removal feature constraint, the analysis is repeated, and iterative optimization adjustment is performed until a pollution removal scheme that meets the pollution removal feature constraint is found and added to the target component pollution removal decision. This enables optimal decision-making for pollution removal solutions and achieves the technical effect of improving the accuracy of pollution removal control.

[0040] Furthermore, the embodiment of the present application also includes the following steps: When the target component cleaning execution instruction is the dust removal cleaning execution instruction and the contamination removal cleaning execution instruction, dust removal analysis and contamination removal analysis are performed on the target inverter component according to the bilateral inverter cleaning decision channel to obtain the target component dust removal decision and the target component contamination removal decision; the target component dust removal decision and the target component contamination removal decision are integrated to generate a target component cleaning plan; and the target inverter component is cleaned and controlled according to the target component cleaning plan.

[0041] Specifically, the aforementioned steps 4 and 5 respectively analyze the situation when the target component cleaning practice instruction is any one of the dust removal cleaning practice instruction or the contamination removal cleaning practice instruction, that is, when the target component cleaning practice instruction is any one of the dust removal cleaning practice instruction or the contamination removal cleaning practice instruction, only dust removal control or contamination removal control needs to be performed. When the target component cleaning practice instruction is the dust removal cleaning practice instruction and the contamination removal cleaning practice instruction, that is, it includes both the dust removal cleaning practice instruction and the contamination removal cleaning practice instruction, at this time, the inverter dust removal cleaning decision channel and the inverter contamination removal cleaning decision channel in the bilateral inverter cleaning decision channel are used to perform dust removal analysis and contamination removal analysis on the target inverter component respectively, and obtain the target component dust removal decision and the target component contamination removal decision. The target component dust removal decision and the target component contamination removal decision are integrated, that is, the target component dust removal decision and the target component contamination removal decision are combined as the target component cleaning plan. Finally, the target inverter component is subjected to dust removal and dirt removal control according to the target component cleaning plan, thereby achieving cleaning control of the target inverter component, achieving joint control of dust removal and dirt removal, improving the accuracy of cleaning control, and further improving the cleaning effect.

[0042] In summary, the cleaning control method of a high-voltage inverter provided by the present application has the following technical effects: According to the distribution of high-voltage inverter components, a target inverter component is obtained, wherein the distribution of high-voltage inverter components includes multiple components of the high-voltage inverter, and the target inverter component is any one of the distribution of high-voltage inverter components; according to the multi-source monitoring end, the target inverter component is monitored in real time to obtain a bilateral monitoring result of the target component, wherein the multi-source monitoring end includes a dust monitoring array and an image monitoring array, and the bilateral monitoring result of the target component includes target dust feature data and target image feature data; according to the bilateral cleaning practice analysis channel, the bilateral monitoring result of the target component is subjected to cleaning practice feature analysis, Generate a cleaning execution instruction for the target component; when the cleaning execution instruction for the target component is a dust removal cleaning execution instruction, perform dust removal analysis of the target inverter component according to the bilateral inverter cleaning decision channel, generate a target component dust removal decision, and perform cleaning control on the target inverter component according to the target component dust removal decision; when the cleaning execution instruction for the target component is a dirt removal cleaning execution instruction, perform dirt removal analysis of the target inverter component according to the bilateral inverter cleaning decision channel, generate a target component dirt removal decision, and perform cleaning control on the target inverter component according to the target component dirt removal decision. Thus, by respectively identifying dust and dirt, generating dust removal cleaning execution instructions and dirt removal cleaning execution instructions, respectively optimizing dust removal schemes and dirt removal schemes, and realizing timely dust removal and dirt removal control, achieving the technical effect of improving the accuracy of dust removal and dirt removal control of high-voltage inverters and ensuring the cleanliness of high-voltage inverters.

[0043] Embodiment 2 Based on the same inventive concept as the cleaning control method of a high-voltage inverter in the above embodiment, the present application also provides a cleaning control device for a high-voltage inverter, see the attached Figure 2 , the device comprises: A target inverter component acquisition module 11 is used to obtain a target inverter component according to a high-voltage inverter component distribution, wherein the high-voltage inverter component distribution includes a plurality of components of a high-voltage inverter, and the target inverter component is any one of the high-voltage inverter component distribution; A bilateral monitoring result acquisition module 12 is used to monitor the target inverter component in real time according to a multi-source monitoring terminal to obtain a bilateral monitoring result of the target component, wherein the multi-source monitoring terminal includes a dust monitoring array and an image monitoring array, and the bilateral monitoring result of the target component includes target dust feature data and target image feature data; A cleaning practice feature analysis module 13 is used to perform cleaning practice feature analysis on the bilateral monitoring result of the target component according to the bilateral cleaning practice analysis channel, and generate a cleaning practice instruction for the target component; The dust removal analysis module 14 is used for performing dust removal analysis of the target inverter component according to the bilateral inverter cleaning decision channel when the target component cleaning execution instruction is a dust removal cleaning execution instruction, generating a target component dust removal decision, and performing cleaning control on the target inverter component according to the target component dust removal decision; The decontamination analysis module 15 is used to perform decontamination analysis of the target inverter component according to the bilateral inverter cleaning decision channel when the target component cleaning implementation instruction is a decontamination cleaning implementation instruction, generate a target component decontamination decision, and perform cleaning control on the target inverter component according to the target component decontamination decision.

[0044] Furthermore, the cleaning practice characteristic analysis module 13 in the device is also used for: Activating a dust removal cleaning practice analysis channel in the bilateral cleaning practice analysis channel, wherein the dust removal cleaning practice analysis channel includes a dust removal cleaning necessity identification subchannel and a dust removal cleaning necessity judgment subchannel; Performing a dust removal requirement depth evaluation on the target dust feature data according to the dust removal and cleaning necessity identification sub-channel to generate a target dust removal and cleaning necessity coefficient; inputting the target dust removal cleaning necessity coefficient into the dust removal cleaning necessity determination subchannel, wherein the dust removal cleaning necessity determination subchannel includes a dust removal cleaning necessity threshold; If the target dust removal cleaning necessity coefficient is greater than / equal to the dust removal cleaning necessity threshold, generating the dust removal cleaning execution instruction, and adding the dust removal cleaning execution instruction to the target component cleaning execution instruction; If the target dust removal cleaning necessity coefficient is less than the dust removal cleaning necessity threshold, a tracking dust monitoring instruction is generated.

[0045] Furthermore, the cleaning practice characteristic analysis module 13 in the device is also used for: Activating a decontamination cleaning practice analysis channel in the bilateral cleaning practice analysis channel, wherein the decontamination cleaning practice analysis channel includes an adaptive enhancement subchannel, a decontamination cleaning necessity identification subchannel, and a decontamination cleaning necessity judgment subchannel; Performing enhancement processing on the target image feature data according to the adaptive enhancement subchannel to generate enhanced image feature data; Performing dirt feature recognition evaluation on the enhanced image feature data according to the dirt removal and cleaning necessity recognition subchannel to generate a target dirt feature recognition result and a target dirt removal and cleaning necessity coefficient; inputting the target decontamination cleaning necessity coefficient into the decontamination cleaning necessity judgment subchannel, wherein the decontamination cleaning necessity judgment subchannel includes a decontamination cleaning necessity threshold; If the target decontamination cleaning necessity coefficient is greater than / equal to the decontamination cleaning necessity threshold, generating the decontamination cleaning practice instruction, and adding the decontamination cleaning practice instruction to the target component cleaning practice instruction; If the target decontamination cleaning necessity coefficient is less than the decontamination cleaning necessity threshold, a tracking image monitoring instruction is generated.

[0046] Furthermore, the cleaning practice feature analysis module 13 in the device further includes: The adaptive enhancement subchannel includes an adaptive enhancement function, and the adaptive enhancement function is: ; Among them, EIFD(x, y) represents the enhanced image feature data, DLG represents the brightness enhancement coefficient, IFD(x, y) represents the brightness feature layer of the target image feature data, ALG represents the detail enhancement coefficient, and IFA(x, y) represents the detail feature layer of the target image feature data.

[0047] Furthermore, the dust removal analysis module 14 in the device is also used for: When the target component cleaning execution instruction is a dust removal cleaning execution instruction, the inverter dust removal cleaning decision channel in the bilateral inverter cleaning decision channel is activated, wherein the inverter dust removal cleaning decision channel includes a dust removal decision analysis subchannel, a dust removal decision deduction subchannel, a dust removal deduction judgment subchannel and a dust removal decision optimization subchannel; According to the dust removal decision analysis subchannel, dust removal analysis is performed in combination with the target dust feature data to generate a component dust removal plan; Deducing the component dust removal scheme according to the dust removal decision deduction subchannel to generate a deduced dust removal index; Inputting the deduced dust removal index into the dust removal deduction judgment subchannel, wherein the dust removal deduction judgment subchannel includes dust removal feature constraints; If the deduced dust removal index satisfies the dust removal characteristic constraint, adding the component dust removal scheme to the target component dust removal decision; If the deduced dust removal index does not satisfy the dust removal feature constraint, the component dust removal scheme is iteratively optimized and adjusted according to the dust removal decision optimization sub-channel, the dust removal decision deduction sub-channel and the dust removal deduction judgment sub-channel to obtain the target component dust removal decision.

[0048] Furthermore, the decontamination analysis module 15 in the device is also used for: When the target component cleaning execution instruction is a decontamination cleaning execution instruction, the inverter decontamination cleaning decision channel in the bilateral inverter cleaning decision channel is activated, wherein the inverter decontamination cleaning decision channel includes a decontamination decision analysis subchannel, a decontamination decision deduction subchannel, a decontamination deduction judgment subchannel and a decontamination decision optimization subchannel; According to the decontamination decision analysis subchannel, a decontamination analysis is performed in combination with the target dirt feature recognition result to generate a component decontamination plan; Deducing the component decontamination plan according to the decontamination decision deduction subchannel to generate a deduced decontamination index; Inputting the deduced pollution removal index into the pollution removal deduction judgment subchannel, wherein the pollution removal deduction judgment subchannel includes pollution removal feature constraints; If the deduced decontamination index satisfies the decontamination characteristic constraint, adding the component decontamination solution to the target component decontamination decision; If the deduced pollution removal index does not meet the pollution removal characteristic constraint, the component pollution removal plan is iteratively optimized and adjusted according to the pollution removal decision optimization sub-channel, the pollution removal decision deduction sub-channel and the pollution removal deduction judgment sub-channel to obtain the target component pollution removal decision.

[0049] Furthermore, the device further comprises a fusion analysis module, and the fusion analysis module is used to: When the target component cleaning execution instruction is the dust removal cleaning execution instruction and the contamination removal cleaning execution instruction, performing dust removal analysis and contamination removal analysis on the target inverter component according to the bilateral inverter cleaning decision channel to obtain the target component dust removal decision and the target component contamination removal decision; The dust removal decision of the target component and the contamination removal decision of the target component are integrated to generate a cleaning plan for the target component; Cleaning control is performed on the target inverter component according to the target component cleaning plan.

[0050] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The cleaning control method and specific examples of a high-voltage inverter in the first embodiment are also applicable to a cleaning control device of a high-voltage inverter in the present embodiment. Through the above detailed description of the cleaning control method of a high-voltage inverter, those skilled in the art can clearly know the cleaning control device of a high-voltage inverter in the present embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0051] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0052] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A cleaning control method for a high voltage inverter, characterized in that: The method comprises: Obtaining a target inverter component according to a high-voltage inverter component distribution, wherein the high-voltage inverter component distribution includes a plurality of high-voltage inverter components, and the target inverter component is any one of the high-voltage inverter component distribution; According to the multi-source monitoring end, the target inverter component is monitored in real time to obtain a bilateral monitoring result of the target component, wherein the multi-source monitoring end includes a dust monitoring array and an image monitoring array, and the bilateral monitoring result of the target component includes target dust feature data and target image feature data; Performing cleaning practice feature analysis on the bilateral monitoring results of the target component according to the bilateral cleaning practice analysis channel, and generating cleaning practice instructions for the target component; When the target component cleaning execution instruction is a dust removal cleaning execution instruction, performing dust removal analysis of the target inverter component according to the bilateral inverter cleaning decision channel, generating a target component dust removal decision, and performing cleaning control on the target inverter component according to the target component dust removal decision; When the target component cleaning execution instruction is a decontamination cleaning execution instruction, a decontamination analysis of the target inverter component is performed according to the bilateral inverter cleaning decision channel, a target component decontamination decision is generated, and the target inverter component is cleaned and controlled according to the target component decontamination decision.

2. The method according to claim 1, characterized in that The cleaning practice characteristics of the bilateral monitoring results of the target component are analyzed according to the bilateral cleaning practice analysis channel, and the cleaning practice instructions of the target component are generated, including: Activating a dust removal cleaning practice analysis channel in the bilateral cleaning practice analysis channel, wherein the dust removal cleaning practice analysis channel includes a dust removal cleaning necessity identification subchannel and a dust removal cleaning necessity judgment subchannel; Performing a dust removal requirement depth evaluation on the target dust feature data according to the dust removal and cleaning necessity identification sub-channel to generate a target dust removal and cleaning necessity coefficient; inputting the target dust removal cleaning necessity coefficient into the dust removal cleaning necessity determination subchannel, wherein the dust removal cleaning necessity determination subchannel includes a dust removal cleaning necessity threshold; If the target dust removal cleaning necessary coefficient is greater than / equal to the dust removal cleaning necessary threshold, generating the dust removal cleaning execution instruction, and adding the dust removal cleaning execution instruction to the target component cleaning execution instruction; If the target dust removal cleaning necessity coefficient is less than the dust removal cleaning necessity threshold, a tracking dust monitoring instruction is generated.

3. The method according to claim 1, characterized in that The cleaning practice characteristics of the bilateral monitoring results of the target component are analyzed according to the bilateral cleaning practice analysis channel, and the cleaning practice instructions of the target component are generated, including: Activating a decontamination cleaning practice analysis channel in the bilateral cleaning practice analysis channel, wherein the decontamination cleaning practice analysis channel includes an adaptive enhancement subchannel, a decontamination cleaning necessity identification subchannel, and a decontamination cleaning necessity judgment subchannel; Performing enhancement processing on the target image feature data according to the adaptive enhancement subchannel to generate enhanced image feature data; Performing dirt feature recognition evaluation on the enhanced image feature data according to the dirt removal and cleaning necessity recognition subchannel to generate a target dirt feature recognition result and a target dirt removal and cleaning necessity coefficient; inputting the target decontamination cleaning necessity coefficient into the decontamination cleaning necessity judgment subchannel, wherein the decontamination cleaning necessity judgment subchannel includes a decontamination cleaning necessity threshold; If the target decontamination cleaning necessity coefficient is greater than / equal to the decontamination cleaning necessity threshold, generating the decontamination cleaning practice instruction, and adding the decontamination cleaning practice instruction to the target component cleaning practice instruction; If the target decontamination cleaning necessity coefficient is less than the decontamination cleaning necessity threshold, a tracking image monitoring instruction is generated.

4. The method according to claim 3, characterized in that The adaptive enhancement subchannel includes an adaptive enhancement function, and the adaptive enhancement function is: ； Among them, EIFD(x, y) represents the enhanced image feature data, DLG represents the brightness enhancement coefficient, IFD(x, y) represents the brightness feature layer of the target image feature data, ALG represents the detail enhancement coefficient, and IFA(x, y) represents the detail feature layer of the target image feature data.

5. The method according to claim 1, characterized in that When the target component cleaning execution instruction is a dust removal cleaning execution instruction, performing dust removal analysis of the target inverter component according to the bilateral inverter cleaning decision channel to generate a target component dust removal decision includes: When the target component cleaning execution instruction is a dust removal cleaning execution instruction, the inverter dust removal cleaning decision channel in the bilateral inverter cleaning decision channel is activated, wherein the inverter dust removal cleaning decision channel includes a dust removal decision analysis subchannel, a dust removal decision deduction subchannel, a dust removal deduction judgment subchannel and a dust removal decision optimization subchannel; According to the dust removal decision analysis subchannel, dust removal analysis is performed in combination with the target dust feature data to generate a component dust removal plan; Deducing the component dust removal scheme according to the dust removal decision deduction subchannel to generate a deduced dust removal index; Inputting the deduced dust removal index into the dust removal deduction judgment subchannel, wherein the dust removal deduction judgment subchannel includes dust removal feature constraints; If the deduced dust removal index satisfies the dust removal characteristic constraint, adding the component dust removal scheme to the target component dust removal decision; If the deduced dust removal index does not satisfy the dust removal feature constraint, the component dust removal scheme is iteratively optimized and adjusted according to the dust removal decision optimization sub-channel, the dust removal decision deduction sub-channel and the dust removal deduction judgment sub-channel to obtain the target component dust removal decision.

6. The method according to claim 1, characterized in that When the target component cleaning execution instruction is a decontamination cleaning execution instruction, performing decontamination analysis of the target inverter component according to the bilateral inverter cleaning decision channel to generate a target component decontamination decision includes: When the target component cleaning execution instruction is a decontamination cleaning execution instruction, the inverter decontamination cleaning decision channel in the bilateral inverter cleaning decision channel is activated, wherein the inverter decontamination cleaning decision channel includes a decontamination decision analysis subchannel, a decontamination decision deduction subchannel, a decontamination deduction judgment subchannel and a decontamination decision optimization subchannel; According to the decontamination decision analysis subchannel, a decontamination analysis is performed in combination with the target dirt feature recognition result to generate a component decontamination plan; Deducing the component decontamination plan according to the decontamination decision deduction subchannel to generate a deduced decontamination index; Inputting the deduced pollution removal index into the pollution removal deduction judgment subchannel, wherein the pollution removal deduction judgment subchannel includes pollution removal feature constraints; If the deduced decontamination index satisfies the decontamination characteristic constraint, adding the component decontamination solution to the target component decontamination decision; If the deduced pollution removal index does not meet the pollution removal characteristic constraint, the component pollution removal plan is iteratively optimized and adjusted according to the pollution removal decision optimization sub-channel, the pollution removal decision deduction sub-channel and the pollution removal deduction judgment sub-channel to obtain the target component pollution removal decision.

7. The method according to claim 1, characterized in that The method comprises: When the target component cleaning execution instruction is the dust removal cleaning execution instruction and the contamination removal cleaning execution instruction, performing dust removal analysis and contamination removal analysis on the target inverter component according to the bilateral inverter cleaning decision channel to obtain the target component dust removal decision and the target component contamination removal decision; The dust removal decision of the target component and the contamination removal decision of the target component are integrated to generate a cleaning plan for the target component; The target inverter component is cleaned and controlled according to the target component cleaning plan.

8. A cleaning control device for a high voltage frequency converter, characterized in that: For implementing the steps of the method according to any one of claims 1 to 7, the device comprises: A target inverter component acquisition module, configured to obtain a target inverter component according to a high-voltage inverter component distribution, wherein the high-voltage inverter component distribution includes a plurality of components of a high-voltage inverter, and the target inverter component is any one of the high-voltage inverter component distribution; A bilateral monitoring result acquisition module, used for monitoring the target inverter component in real time according to a multi-source monitoring terminal to obtain a bilateral monitoring result of the target component, wherein the multi-source monitoring terminal includes a dust monitoring array and an image monitoring array, and the bilateral monitoring result of the target component includes target dust feature data and target image feature data; A cleaning practice feature analysis module, used to perform cleaning practice feature analysis on the bilateral monitoring results of the target component according to the bilateral cleaning practice analysis channel, and generate cleaning practice instructions for the target component; a dust removal analysis module, for performing dust removal analysis of the target inverter component according to the bilateral inverter cleaning decision channel when the target component cleaning execution instruction is a dust removal cleaning execution instruction, generating a target component dust removal decision, and performing cleaning control on the target inverter component according to the target component dust removal decision; A decontamination analysis module is used to perform decontamination analysis of the target inverter component according to the bilateral inverter cleaning decision channel when the target component cleaning implementation instruction is a decontamination cleaning implementation instruction, generate a target component decontamination decision, and perform cleaning control on the target inverter component according to the target component decontamination decision.

Images