Conductor cleaning agent and cleaning device

Abstract

The invention discloses an electric conductor cleaning agent and a cleaning device, the electric conductor cleaning agent comprises a rust remover and a thickening agent which are mixed, and the rust remover and the thickening agent which are mixed are in a gel state. According to the electric conductor cleaning agent provided by the invention, the liquid cleaning agent is converted into a gel form through the thickening agent. The electric conductor cleaning agent has excellent adhesiveness and can stay on the surface of an electric conductor for a long time, liquid dripping and volatilization are avoided while the cleaning effect is improved, and harm to the environment and operators is reduced. The invention further provides a cleaning device, the cleaning device is used for cleaning, an operator is prevented from making direct contact with a cleaning agent, the body safety of the operator is guaranteed, the cleaning efficiency is improved, the cleaning quality is optimized, the service life of the electrical equipment is prolonged, and the maintenance cost of the electrical equipment is reduced. Abnormity and faults of the power equipment caused by unclean contact surfaces of the conductors are reduced, and stability and safety of the power equipment are enhanced.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of cleaning agents, and in particular to a conductive cleaning agent and a cleaning device. Background Art

[0002] In modern power systems, substation power equipment is the key to ensuring stable power transmission. However, as the operating time of power equipment increases, dust, dirt and oxides are easily accumulated on the surface of the conductor of power equipment, which not only affects the conductivity and heat dissipation performance, but also may cause abnormalities in the power equipment and lead to power equipment failure. Current cleaning methods mostly rely on liquid cleaning agents, which are cleaned by spraying and wiping. This method has certain disadvantages: liquid cleaning agents evaporate quickly, have poor adhesion, and are easy to cause skin damage or respiratory discomfort to workers during operation, and the cleaning efficiency is low. Especially in high-voltage environments, the risk of manual operation is high, and existing power equipment cannot provide effective antioxidant protection. Summary of the invention

[0003] The present invention provides a conductor cleaning agent and a cleaning device to achieve the purpose of solving at least one defect existing in the prior art.

[0004] In a first aspect, an embodiment of the present invention provides a conductor cleaning agent, comprising a mixed rust remover and a thickener, wherein the mixed rust remover and thickener are in a gel state.

[0005] Optionally, the thickener is an oily thickener.

[0006] Optionally, the conductor cleaner is used to clean the surface of a conductor.

[0007] In a second aspect, an embodiment of the present invention further provides a cleaning device, comprising a cleaning agent storage bottle, wherein the cleaning agent storage bottle stores any one of the conductor cleaning agents described in the embodiments of the present invention.

[0008] Optionally, it also includes a controller and a camera, wherein the controller is used to execute a preset conductor cleaning control method;

[0009] The camera is used to obtain image data required when the conductor cleaning control method is executed.

[0010] Optionally, it also includes a power supply, a motor, and a brush head;

[0011] The power supply is used to supply power to the motor and the detergent storage bottle. The motor is used to drive the brush head to rotate. The detergent storage bottle is also used to pump the conductive body detergent into the brush head.

[0012] Optionally, an insulating rod is further included, and the insulating rod serves as a handheld portion of the cleaning device.

[0013] Optionally, the insulating rod is also configured with a telescopic structure.

[0014] Optionally, the brush head is equipped with a cotton wheel.

[0015] Optionally, the controller is further configured to adjust the flow rate of the conductive cleaning agent pumped into the brush head.

[0016] Compared with the prior art, the present invention has the following beneficial effects: the present invention provides a conductor cleaning agent, which converts a liquid cleaning agent into a gel form through a thickener. The conductor cleaning agent has excellent adhesion and can stay on the surface of the conductor for a long time, thereby improving the cleaning effect while avoiding liquid dripping and volatilization, and reducing harm to the environment and operators. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 Schematic diagram of the cleaning device structure in the embodiment.

[0018] Figure 2 4 is a flow chart of the conductor cleaning control method in the embodiment. DETAILED DESCRIPTION

[0019] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0020] Embodiment 1

[0021] The embodiment of the present invention provides a conductor cleaning agent, comprising a mixed rust remover and a thickener, wherein the mixed rust remover and thickener are in a gel state.

[0022] For example, in this solution, phosphoric acid can be used as the rust remover, with a concentration of 20% to 30% (mass fraction), and the amount used accounts for 40% to 50% of the total formula. Phosphoric acid has good rust removal ability, can effectively remove rust and oxides on the metal surface, and is relatively less corrosive to metals. It is suitable for use in conductor cleaners to ensure that the conductivity of the metal surface after cleaning is not greatly affected;

[0023] The thickener can be sodium carboxymethyl cellulose (CMC), with the dosage accounting for 3% to 5% of the total formula. CMC can make the cleaner gel-like, increase the adhesion and residence time of the cleaner on the metal surface, thereby improving the rust removal effect. It has good water solubility and thickening properties, and can maintain a stable gel structure in a wide temperature range. ;

[0024] The solvent may be deionized water, with the amount accounting for 45% to 57% of the total formula. Deionized water can dissolve the rust remover and the thickener, making them uniformly mixed, and will not introduce impurity ions to avoid interference with the conductive properties.

[0025] Exemplarily, in this solution, to make a conductive cleaning agent, deionized water is first added to a stirring container, a stirring device is turned on, and a stirring speed is set to 300 to 500 rpm;

[0026] Slowly add phosphoric acid to the deionized water in the stirring container, stirring while adding to ensure that the phosphoric acid is evenly dispersed in the water to prevent excessive local concentration from causing violent reactions;

[0027] After the phosphoric acid is completely dissolved, slowly sprinkle sodium carboxymethyl cellulose (CMC) into the solution, increase the stirring speed to 800-1000 rpm, and continue stirring for 30-45 minutes to allow the CMC to fully swell and disperse, and mix evenly with the phosphoric acid solution to form a gel-like conductor cleaner.

[0028] This proposal proposes a conductor cleaning agent, which converts a liquid cleaning agent into a gel form through a thickener. The conductor cleaning agent has excellent adhesion and can stay on the surface of the conductor for a long time, improving the cleaning effect while avoiding liquid dripping and volatilization, reducing harm to the environment and operators.

[0029] Exemplarily, in this solution, the following cleaning agents may also be used as the conductor cleaning agent:

[0030] WD-40 liquid has a strong solvent odor and is easy to splash, causing the risk of inhalation or splashing into the eyes. Kontakt60 paste and Kontakt61 paste are relatively safe, odorless, more comfortable to use, and have less residue during cleaning and are not easy to splash;

[0031] WD-40, Kontakt60 paste, and Kontakt61 paste are good for cleaning the sample conductors and removing dust and oil on the surface. The effect of cleaning the oxide layer is average, and it needs to be wiped repeatedly with a soft cloth to achieve a good cleaning effect. Compared with soft cloth, scouring pad is more effective in cleaning the oxide layer;

[0032] WD-40 liquid was mixed with Kontakt60 paste and Kontakt61 paste, and the mixture was gel-like. After a while, separation occurred, indicating that the two pastes had poor fusion and could not be stably mixed. Mixing did not significantly improve the removal effect of the oxide layer, and the oxide layer still needed to be cleaned by physical means.

[0033] WD-40 liquid was mixed with J0602 and J0613 oil thickeners to form a gel. After a period of time, no separation occurred, indicating that the two had good fusion and stable mixing. The new cleaning gel maintained a good cleaning effect without obvious residue. It was stable in terms of user experience and was suitable for different cleaning needs.

[0034] Based on any of the above schemes, in one possible implementation manner, the thickener is an oily thickener.

[0035] For example, in this solution, the oily thickener can be polyisobutylene, with the amount accounting for 5% to 8% of the total formula. Polyisobutylene has an excellent thickening effect, which can make the detergent gel-like, enhance its adhesion and retention time on the metal surface, improve the rust removal efficiency, and has high chemical stability and is not easily affected by environmental factors.

[0036] For example, in this solution, white oil can be used as a solvent, and the amount used accounts for 42% to 55% of the total formula. White oil can effectively dissolve the rust remover and the oily thickener, and promote the uniform mixing of the two. At the same time, white oil itself has certain lubricity and stability, which helps to improve the comprehensive performance of the cleaning agent.

[0037] Based on any of the above schemes, in one possible implementation manner, the conductor cleaner is used to clean the surface of the conductor.

[0038] Embodiment 2

[0039] This embodiment provides a cleaning device, including a cleaning agent storage bottle, wherein the cleaning agent storage bottle stores any one of the conductive body cleaning agents described in the first embodiment.

[0040] Based on any of the above solutions, in one possible implementation scheme, the cleaning device further includes a controller and a camera, and the controller is used to execute a preset conductor cleaning control method;

[0041] The camera is used to obtain image data required when the conductor cleaning control method is executed.

[0042] Based on any of the foregoing solutions, in one possible implementation, the cleaning device further includes a power source, a motor, and a brush head;

[0043] The power supply is used to supply power to the motor and the detergent storage bottle. The motor is used to drive the brush head to rotate. The detergent storage bottle is also used to pump the conductive body detergent into the brush head.

[0044] Exemplarily, in this solution, the power source is used to power the motor and the detergent storage bottle, the motor is used to drive the brush head to rotate, and the detergent storage bottle is used to store the detergent and pump the detergent into the brush head.

[0045] Exemplarily, in this solution, the cleaning device uses a sturdy and lightweight engineering plastic shell, and inside the shell, components such as a power supply, a motor, a brush head, and a detergent storage bottle are reasonably arranged;

[0046] The power source is located at the bottom or rear of the device to ensure the overall center of gravity is stable, and it is convenient to lead out the power cord or replace the battery (if it is a rechargeable battery). The motor is installed near the brush head and connected to the brush head through a suitable transmission mechanism (such as gear drive or belt drive) to ensure efficient and stable power transmission. The detergent storage bottle is set above or on the side of the device to facilitate observation of the detergent level and is connected to the brush head through a pipeline, which is equipped with a precise pump body to control the delivery of the detergent.

[0047] For example, in this solution, the cleaning device may use a rechargeable lithium-ion battery pack, which has the advantages of high energy density, convenient charging, and long service life. The capacity of the battery pack may be designed according to the endurance requirements of the cleaning device.

[0048] For example, in this embodiment, the cleaning device may use a brushless DC motor, which has the advantages of high speed, high torque, high efficiency, low noise, etc. The power of the motor may be determined according to factors such as the cleaning width of the cleaning device, the brush head pressure, and the desired cleaning speed. For example, the motor power may be selected between 100 and 300 watts.

[0049] For example, in this solution, the brush head adopts a variety of materials and structural designs to adapt to different cleaning objects and dirt types. For example, a brush head with harder and denser bristles, such as a nylon bristle brush head, can be used, and a brush head with softer and more elastic bristles, such as a polyester fiber bristle brush head, can be used to avoid damaging carpet fibers. The brush head can be driven by direct drive or indirect drive. Direct drive means that the motor shaft is directly connected to the brush head shaft. This method has high transmission efficiency and simple structure.

[0050] For example, in this solution, the detergent storage bottle is connected to the brush head through a pipeline, and a precise pump body, such as a peristaltic pump or a gear pump, is installed on the pipeline. The pump body is controlled by the control system of the cleaning device to accurately control the supply amount of the detergent according to the cleaning effect evaluation result or the cleaning mode set by the user.

[0051] Exemplarily, in this embodiment, the brush head may also be provided with a cotton wheel.

[0052] Exemplarily, in this solution, after the cleaning device is started, the power supply supplies power to the motor and the detergent storage bottle, the motor drives the appropriate brush head to rotate, and the detergent storage bottle pumps an appropriate amount of detergent into the brush head according to the control instruction of the cleaning device. The cleaning device can evaluate the cleaning effect in real time according to a pre-set cleaning program or through image analysis, dynamically adjust parameters such as motor speed, brush head pressure and detergent flow, and efficiently complete the cleaning work. While ensuring the cleaning quality, it realizes the rational use of detergents and reduces waste and environmental pollution.

[0053] Based on any of the foregoing solutions, in one possible implementation manner, it further includes an insulating rod, which serves as a handheld portion of the cleaning device.

[0054] For example, in this solution, the insulating rod is made of a high-strength, high-insulation composite material, such as glass fiber reinforced epoxy resin. Its length can be customized according to the actual height of the high-voltage wire and operational requirements, generally ranging from 3 to 10 meters.

[0055] In this solution, in actual high-voltage wire cleaning operations, such as line maintenance in a city power grid, the cleaning device and operating method of this embodiment are used, and operators can clean high-voltage wires safely on the ground;

[0056] The insulating rod effectively isolates the operator from the high voltage electricity, reducing the risk of operation. The intelligent control function of the cleaning device enables the operator to flexibly adjust the cleaning parameters according to the actual dirt situation of the wire, improving the cleaning efficiency and quality, ensuring that the high voltage wire can operate in a good state, and ensuring the safe and stable power supply of the urban power grid.

[0057] Based on any of the foregoing solutions, in one possible implementation manner, the controller is further configured to adjust the flow rate of the conductive cleaning agent pumped into the brush head.

[0058] Based on any of the above solutions, in one possible implementation scheme, the insulating rod is further configured with a telescopic structure.

[0059] Figure 1 is a schematic diagram of the cleaning device structure in the embodiment, refer to Figure 1 Based on any of the foregoing solutions, in one possible implementation scheme, the cleaning device includes:

[0060] Controller 100, camera 200, power supply 300, motor 400, brush head 500, detergent storage bottle 600, insulating rod 700, wherein the insulating rod 700 is configured with a telescopic structure, the detergent storage bottle 600 has a built-in micro water pump, and the motor 400 is configured with a telescopic module;

[0061] In this solution, the power supply provides stable power support and drives the entire device (including motors, controllers, and other modules) by converting voltage and current. The power supply design needs to ensure stable voltage and sufficient output power to meet the working requirements of the entire device.

[0062] The controller adjusts the motor speed, the movement of the telescopic module, and the amount of detergent delivered according to the preset program and sensor feedback to ensure the accuracy and efficiency of the cleaning process;

[0063] The insulating rod can ensure that the device can contact the contact surface of the conductor safely and efficiently, preventing operators from directly contacting high-voltage equipment and providing additional safety protection;

[0064] The detergent storage bottle is used to store detergent and is designed to be connected to a micro water pump so as to deliver the detergent to the brush head. The capacity of the detergent storage bottle needs to be designed according to actual work needs to ensure that the detergent can be continuously supplied and that the detergent does not leak.

[0065] The telescopic module is responsible for the forward and backward movement of the cleaning device, adjusting the working position of the cleaning device so that the brush head can accurately align with the surface of the conductive body to be cleaned. The precise control of the telescopic module ensures that the cleaning device can cover every corner to avoid omissions.

[0066] The motor's gripper grabs and holds the brush head, ensuring it remains stable during operation. The motor drives the gripper and telescopic module, allowing the device to be flexibly adjusted and precisely positioned on the surface to be cleaned.

[0067] In this solution, the insulating rod is made of high-strength composite insulating material with an insulation resistance of 1012Ω*cm, ensuring the safety of operation under high-voltage environment. Its insulation performance is stable, and it can maintain a high insulation effect even in bad weather or humid and high salt fog conditions. The leakage current is less than 1μA / cm2 to ensure the safety of personnel operation.

[0068] The insulating rod is retractable and can be flexibly adjusted in length from 1.5m to 6m according to the height of the cleaning object. The maximum load-bearing capacity of the rod is 100N. This design not only adapts to insulating equipment of different heights, but also effectively reduces the operating burden and ensures the safety of the rod.

[0069] In this solution, the brush head can accurately control the flow rate within the range of 10mL / min to 100mL / min (error ±5%), and can adjust the spray angle between 0° and 60° to achieve the best cleaning effect. The flow rate and angle parameters under different surface and dirt conditions can be automatically set according to needs.

[0070] The brush head is designed with a self-cleaning and anti-clogging structure, which can automatically clear away tiny impurities and ensure smooth spraying, making it suitable for long-term continuous use.

[0071] In this solution, the camera is connected to the controller, and the controller establishes a communication connection with the monitoring device or mobile terminal. The camera can also be used to achieve real-time visual feedback and can transmit the cleaning process image to the monitoring device or mobile terminal App, making it convenient for the staff to check the cleaning effect at any time.

[0072] In this solution, a cleaning device is used for cleaning, which avoids direct contact between operators and cleaning agents, ensures the physical safety of operators, improves cleaning efficiency, optimizes cleaning quality, extends the service life of power equipment, reduces the maintenance cost of power equipment, reduces power equipment abnormalities and failures caused by unclean conductor contact surfaces, and enhances the stability and safety of power equipment.

[0073] Figure 2 is a flow chart of the conductor cleaning control method in the embodiment, refer to Figure 2 , illustratively, in this embodiment, the conductor cleaning control method includes:

[0074] S101. Determine the area to be cleaned, judge the depth parameter of the dirt in the area to be cleaned, and generate a cleaning device position adjustment instruction according to the depth parameter.

[0075] Exemplarily, in this solution, the area to be cleaned is determined based on the image (taken by the camera), and the method used may be:

[0076] Use the kernel matrix of the Sobel operator to perform convolution operations with the image respectively. For each pixel in the image, calculate the horizontal gradient and the vertical gradient. The formula used is:

[0077]

[0078] Where Gx(i,j) represents the horizontal gradient, Gy(i,j) represents the vertical gradient, I(i,j) represents the pixel value of the (i,j) image coordinate, Sx represents the kernel matrix in the horizontal direction, and Sy represents the kernel matrix in the vertical direction;

[0079] The gradient amplitude is calculated based on the horizontal gradient and the vertical gradient. If the gradient amplitude of the image coordinate point is greater than the set threshold, the image coordinate point is marked as an edge point. By connecting the edge points, the boundary of the area to be cleaned can be determined.

[0080] The formula used to calculate the gradient amplitude G(i,j) is:

[0081]

[0082] Alternatively, the area to be cleaned may be determined by:

[0083] Determine the number k of cluster centers, set k to 2, divide the image into a clean area and an unclean area, and randomly select k pixel points as the initial cluster centers μ1 and μ2;

[0084] Calculate the distance d from each pixel point in the image to each cluster center l (i,j), and the formula used is:

[0085]

[0086] In the formula, I(i,j) represents the eigenvalue (color, brightness, etc.) of the pixel point with image coordinates (i,j);

[0087] Assign the pixel point to the type Ci of the nearest cluster center, and update each cluster center μl so that it is equal to the average value of the eigenvalues of all pixel points belonging to this cluster. The formula used is:

[0088]

[0089] In the formula, Nl is the number of pixel points in the cluster Ci. Repeat the process of assigning the pixel point to the type Ci of the nearest cluster center and updating each cluster center μl until the cluster center no longer changes significantly or reaches the predetermined number of iterations;

[0090] According to the clustering result, determine the category representing the unclean area as the area to be cleaned. The area to be cleaned can be displayed on the image by means of marking or color differentiation, etc.;

[0091] Alternatively, the method for determining the area to be cleaned can be:

[0092] And convert the image into a grayscale image, and determine the grayscale threshold T;

[0093] For each pixel point (i,j) in the image, if its grayscale value I(i,j) < T (assuming the dirt area is darker), then mark this pixel point as the dirt area (area to be cleaned). If I(i,j) ≥ T, then mark it as the clean area. In this way, the image is segmented into a clean area and an area to be cleaned.

[0094] Exemplarily, in this solution, set the depth parameter to represent the minimum vertical distance between the dirt and the camera (imaging plane) in the area to be cleaned;

[0095] In this solution, the depth parameter can be determined based on a deep learning-based depth estimation method;

[0096] Specifically, use a deep neural network model and train it with a large amount of image data with depth labels, so that the model can automatically predict the depth information of the object according to the input image, including:

[0097] Collect image datasets containing various surfaces (including those with dirt) and annotate these images with depth information (through manual annotation);

[0098] Choose a suitable deep learning model, such as a variant of a convolutional neural network (CNN) (e.g., ResNet, DenseNet, etc.) or a network specifically designed for depth estimation (e.g., Monodepth, DepthNet, etc.);

[0099] The labeled image data is divided into training set, validation set and test set. The model is trained using the training set, and the model parameters are continuously adjusted through the back propagation algorithm to make the model's prediction results as close as possible to the labeled depth information. During the training process, the validation set is used to monitor the performance of the model to prevent overfitting;

[0100] The image of the object to be cleaned is input into the trained model, and the model outputs the depth estimation value of each pixel in the image. Among these depth estimation values, the minimum value of the depth estimation values ​​corresponding to the dirt area is found as the minimum value of the vertical distance between the dirt and the imaging plane (depth parameter).

[0101] Illustratively, in this solution, the cleaning device position adjustment instruction is used to adjust the vertical distance between the cleaning device and the dirt in the area to be cleaned, so that the cleaning device can clean the dirt.

[0102] S102. Acquire first image data of the area to be cleaned before cleaning and second image data of the area to be cleaned after cleaning.

[0103] Exemplarily, in this solution, the first image data is a first image acquired (taken by a camera), and the second image data is a second image acquired (taken by a camera).

[0104] S103. Determine image similarity, image contrast, and cleaning coverage based on the first image data and the second image data.

[0105] Exemplarily, in this solution, determining the image similarity may be:

[0106] Calculate the sum of the absolute values ​​of the corresponding pixel differences between the first image and the second image, MAD. The smaller the MAD value, the more similar the images are. The formula used is:

[0107]

[0108] Where I1 represents the first image, I2 represents the second image, and N represents the total number of pixels;

[0109] Alternatively, determining image similarity can be:

[0110] Calculate the average MSE of the sum of squares of the corresponding pixel differences of the first image and the second image. The smaller the MSE value, the more similar the images are. The formula used is:

[0111]

[0112] Exemplarily, in this solution, the image contrast may be determined by a histogram, including:

[0113] Generate histograms HA and HB of the first image and the second image respectively, where the abscissa of the histogram is the grayscale level (e.g., 0 to 255) and the ordinate is the number of pixels at the corresponding grayscale level;

[0114] For the histogram HA of the first image, its variance VA is:

[0115]

[0116] For the histogram HB of the second image, its variance VB is:

[0117]

[0118] Where L represents the total number of gray levels. The contrast is determined by comparing VA and VB. The larger the variance, the more dispersed the histogram distribution, and the higher the contrast may be.

[0119] Exemplarily, in this solution, a deep learning model (such as a convolutional neural network, CNN) can be used to automatically learn and identify clean areas and unclean areas, and then calculate the cleaning coverage, including:

[0120] Collect a large amount of image data before and after cleaning, annotate these images, and mark the clean and unclean areas;

[0121] Build a suitable CNN model (such as U-Net, MaskR-CNN, etc.) and divide the labeled data into training set, validation set and test set;

[0122] Use the training set to train the model, and adjust the model parameters through the back propagation algorithm so that the model can accurately segment the clean area and the unclean area. During the training process, use the validation set to monitor the performance of the model to prevent overfitting;

[0123] The cleaned image to be tested is input into the trained model, and the model outputs the segmentation result of the clean area. The area ratio of the clean area is calculated to determine the cleaning coverage.

[0124] S104. The cleaning effect is determined according to the image similarity, image contrast and cleaning coverage. If the cleaning effect reaches the expected level, the cleaning is stopped; otherwise, the cleaning of the area to be cleaned is repeated.

[0125] Exemplarily, in this solution, if the image similarity is less than the first threshold, the image contrast is greater than the second threshold, and the cleaning coverage is greater than the third threshold, it can be determined that the cleaning effect meets expectations; otherwise, it is determined that the cleaning effect does not meet expectations.

[0126] Illustratively, in this solution, repeatedly cleaning the area to be cleaned may be repeatedly performing steps S101 to S104.

[0127] This embodiment proposes a cleaning control method, in which, by accurately determining the area to be cleaned and judging the depth parameters of the dirt therein, an accurate position adjustment instruction can be generated for the cleaning device, which not only enables the cleaning device to operate at the most appropriate distance and angle to the cleaning area, greatly improving the pertinence and effectiveness of cleaning, but also effectively avoids problems such as incomplete cleaning or unnecessary damage to the cleaning object caused by an improper distance between the cleaning device and the dirt;

[0128] During the cleaning process, the first image data before cleaning and the second image data after cleaning are analyzed to obtain key indicators such as image similarity, image contrast and cleaning coverage. Image similarity can reflect the degree of change in the area before and after cleaning in terms of overall structure and detail features, and accurately quantify the impact of cleaning operations on the integrity of the area's appearance; image contrast highlights the changes in the difference between bright and dark areas in the area before and after cleaning, which can intuitively reflect the improvement in image clarity and layering after dirt removal, and effectively evaluate the improvement of image visual quality by cleaning; cleaning coverage accurately measures the scope and extent of cleaning from the dimension of regional area, providing a reliable basis for a comprehensive understanding of the progress of the cleaning task.

[0129] exist Figure 1 Based on the scheme shown, in one possible implementation scheme, determining the depth parameter of the dirt in the area to be cleaned includes determining the depth parameter using the following formula:

[0130]

[0131] Where Z represents the depth parameter, f represents the focal length of the camera, B represents the baseline distance between the two cameras, and d represents the parallax between the images taken by the two cameras.

[0132] Exemplarily, in this solution, the cleaning device can be equipped with a binocular camera, wherein the optical axes of the two cameras are parallel and on the same horizontal plane, and the baseline distance between the cameras can be designed and fixed according to the working range and accuracy requirements of the cleaning equipment;

[0133] For example, for cleaning scenarios targeting electrical conductors, the baseline distance can be set between 50 mm and 150 mm;

[0134] Before the cleaning operation starts, the binocular camera simultaneously captures images of the target area to be cleaned and pre-processes the captured left and right images to improve the image clarity and feature recognition.

[0135] The preprocessing steps include grayscale processing, which converts color images into grayscale images, reducing the amount of data while highlighting the edge and texture information of the image; then Gaussian filtering is performed to remove noise interference in the image; finally, image enhancement processing is performed, such as histogram equalization, to expand the grayscale dynamic range of the image, enhance the contrast of the image, and make the details in the image clearer and more visible;

[0136] Use advanced feature extraction algorithms, such as deep learning-based feature extraction networks or the classic scale-invariant feature transform (SIFT) algorithm, to extract feature points from the preprocessed left and right images;

[0137] Using a feature matching algorithm, such as the nearest neighbor matching algorithm based on Euclidean distance, the feature points in the left and right images are matched. For each feature point in the left image, the feature point in the right image that is closest to its feature descriptor is found as the matching point.

[0138] In order to improve the accuracy and reliability of matching, a two-way matching strategy can be adopted, that is, matching is performed not only from left to right, but also from right to left. At the same time, a reasonable matching threshold is set, such as the feature point description sub-Euclidean distance threshold of 0.7 to 0.9 times the ratio of the nearest neighbor distance to the next nearest neighbor distance, to remove the feature point pairs that were mismatched, and retain high-quality matching feature point pairs for subsequent disparity calculation;

[0139] For the feature point pairs that have been matched and verified, calculate their disparity in the left and right images. The disparity calculation is based on the pixel coordinate difference of the feature point in the image coordinate system, and the disparity value of the feature point is obtained through the precise coordinate difference calculation.

[0140] Substitute the pre-calibrated camera focal length f, the fixed baseline distance B, and the calculated parallax d into the above formula, and quickly and accurately calculate the depth parameter through the calculation module in the cleaning equipment control system;

[0141] For example, after calibration, the focal length of the camera is 80 mm, the baseline distance is 120 mm, and the parallax calculated for a certain matching feature point pair is 20 mm, then the dirt depth parameter corresponding to the feature point is 480 mm;

[0142] The calculated depth parameter Z will be transmitted to the control system of the cleaning equipment to generate position adjustment instructions for the cleaning device (such as nozzles, cleaning brushes, etc.), so that the cleaning device can perform precise spatial positioning and motion control according to the depth of the dirt.

[0143] In this solution, binocular vision technology is used to achieve a fully automated operation process from depth perception to automatic control of the cleaning device. The entire process can be completed accurately without human intervention, greatly reducing labor costs and human operation errors, and improving the production efficiency and standardization of cleaning operations.

[0144] Based on any of the above solutions, in one possible implementation, determining the image similarity includes determining a similarity index using the following formula:

[0145]

[0146] Where SSIM(x,y) represents the similarity index, μx represents the pixel average of the first image, μy represents the pixel average of the second image, σxy represents the pixel covariance between the first image and the second image, and σx 2 represents the pixel variance of the first image, σy 2 represents the pixel variance of the second image, C1 represents the first constant term, and C2 represents the second constant term.

[0147] Exemplarily, in this solution, before the cleaning operation is started, the image acquisition device is triggered to capture a first image of the area to be cleaned, and after the cleaning operation is completed, the image acquisition device is used again to capture a second image of the same area to be cleaned, and the acquisition conditions are consistent with the image before cleaning to ensure the comparability of the images;

[0148] The collected first image and the second image are preprocessed respectively. First, the image is cropped to remove irrelevant areas that may exist at the edge of the image, and only the core part related to the area to be cleaned is retained;

[0149] Then grayscale processing is performed to convert the color image into a grayscale image, which simplifies the subsequent calculation process and reduces the interference of color information on similarity evaluation;

[0150] Next, image filtering is performed, using methods such as Gaussian filtering to remove noise points in the image and improve the clarity and quality of the image. The size of the Gaussian filter kernel can be adjusted according to the degree of image noise, and is generally between 3x3 and 5x5 pixels.

[0151] Calculate the pixel mean μx of the first image and the pixel mean μy of the second image, the pixel covariance σxy of the first image and the second image, and the pixel variance σx of the first image 2 , the pixel variance σy of the second image 2 ;

[0152] According to the above formula, the similarity index SSIM(x,y) is calculated, where C1 and C2 are constant terms, and their values ​​are determined according to the pixel value range of the image and the calculation accuracy requirements. For example, for an 8-bit grayscale image (pixel value range 0 to 255), C1 and C2 can be 0.01 and 0.03 respectively;

[0153] The cleaning effect is judged according to the calculated similarity index SSIM(x,y). If the value of SSIM(x,y) is low, for example, less than 0.7, it means that the image before and after cleaning is quite different, the cleaning effect is good, and the expected goal is achieved. The cleaning equipment can stop the cleaning operation or enter the next cleaning area;

[0154] If SSIM(x,y) is close to 1, for example, greater than 0.9, it means that the image after cleaning is very similar to the image before cleaning, and the cleaning effect is not ideal. The cleaning equipment needs to repeat the cleaning process or adjust the cleaning parameters and clean again until the value of SSIM(x,y) reaches an acceptable range.

[0155] In this solution, by calculating the image similarity index SSIM(x,y) based on a specific formula, the cleaning effect can be converted from a subjective visual perception to an objective numerical indicator, greatly improving the accuracy and reliability of the evaluation. Whether it is a tiny stain residue or a change in the overall cleaning level, it can be accurately reflected by this index, providing scientific and rigorous data support for the quality control of cleaning operations;

[0156] When calculating SSIM(x,y), multiple statistical parameters such as the pixel mean, variance and covariance of the image are taken into comprehensive consideration. This can capture the changes in the structure, texture and details of the image before and after cleaning, thereby more comprehensively and carefully evaluating the actual impact of the cleaning operation on the clean area represented by the image, avoiding the one-sidedness and misjudgment that may be caused by a single indicator evaluation.

[0157] Based on any of the foregoing schemes, in one possible implementation manner, determining the image contrast includes determining a contrast index according to the following formula:

[0158]

[0159] Wherein, Contrast represents the contrast index, max(I) represents the maximum pixel value of the second image, and min(I) represents the minimum pixel value of the second image.

[0160] Exemplarily, in this solution, each pixel point in the preprocessed second image is traversed to find the maximum value max(I) and the minimum value min(I) of the pixel value. In the traversal process, an efficient data search algorithm is used to reduce computing time and resource consumption. For example, the image can be divided into blocks first, and the maximum value is found in each sub-block, and then the maximum and minimum pixel values ​​of the entire image are obtained by comprehensive comparison;

[0161] Substitute the obtained max(I) and min(I) into the formula and perform accurate calculations through the calculation unit of the cleaning device. For example, if the maximum pixel value in the image is 220 and the minimum pixel value is 30, then Contrast = (220-30) / (220+30) = 0.76.

[0162] For example, in this solution, the cleaning effect is judged according to the calculated contrast index Contrast. A contrast threshold range is preset, and the threshold range can be determined by a large amount of experimental data and actual cleaning requirements;

[0163] If the value of Contrast falls within the preset good cleaning threshold range (for example, for general cleaning tasks, the threshold range can be set to 0.6 to 0.9), it indicates that the image after cleaning has a high contrast and the cleaning effect is good, and the cleaning device can stop the cleaning operation or enter the preparation stage for the next cleaning area;

[0164] If the Contrast value is lower than the lower threshold, it means that the contrast of the image after cleaning is low, there may be residual dirt or incomplete cleaning. The cleaning device needs to start a repeated cleaning program or adjust the cleaning parameters (such as increasing the cleaning intensity, changing the cleaning fluid, etc.) and then clean again.

[0165] For example, in this solution, the control system of the cleaning device automatically generates corresponding control instructions based on the judgment result of the contrast index. When repeated cleaning is required, the control system adjusts the working parameters of the cleaning device and triggers the image acquisition and contrast calculation process again to form a closed feedback control loop. Through continuous monitoring and adjustment, it is ensured that the cleaning effect finally reaches the expected standard, realizing the intelligent and automated cleaning operation.

[0166] Based on any of the foregoing schemes, in one possible implementation scheme, determining the cleaning coverage includes determining the cleaning area using the following formula:

[0167]

[0168] In the formula, A clean represents the cleaning area, I(x i ,y i) represents the pixel value of the pixel in the i-th row and j-th column after the second image is binarized.

[0169] After the cleaning operation is completed, the image acquisition device is triggered in time to capture an image of the area to be cleaned, which is defined as a second image, and a suitable threshold is selected to perform binarization processing on the grayscale image;

[0170] According to the determined threshold, each pixel in the grayscale image is binarized. Let the grayscale value of the pixel be g(x i ,y i ), the threshold is T, then the pixel value after binarization is I(x i ,y i ) can be expressed as:

[0171]

[0172] Among them, a pixel value of 1 represents a clean area, and a pixel value of 0 represents an unclean area or a stained area. Through binarization, the image is converted into a binary image containing only two pixel values, which greatly simplifies the calculation process of the clean area;

[0173] According to the following formula:

[0174]

[0175] Calculate the clean area, traverse each pixel in the binary image, and accumulate the pixel values ​​​​of 1;

[0176] Since each pixel in the image represents a certain actual area, for example, if the resolution of the image is M×N, and the length and width of the actual cleaning area corresponding to the image are known to be L and W respectively, the actual area represented by each pixel is:

[0177]

[0178] By accumulating the number of pixels with a pixel value of 1 and multiplying it by the actual area represented by each pixel, the precise cleaning area A can be obtained. clean If the number of pixels with a pixel value of 1 in the binary image is p, and the actual area represented by each pixel is a, then A clean Can be pa.

[0179] Assume the total area of ​​the clean area is A total (can be obtained by the geometric dimensions of the cleaning object or by pre-measurement), the cleaning coverage C is:

[0180]

[0181] Exemplarily, in this solution, by calculating the cleaning coverage rate, the completion degree and effect of the cleaning work can be intuitively understood;

[0182] According to the preset cleaning standards, it can be judged whether the cleaning operation meets the requirements. If the cleaning coverage rate does not meet the expected standards, the cleaning equipment can adjust the cleaning strategy according to the specific situation, such as re-cleaning the uncovered area, increasing the cleaning time or adjusting the working parameters of the cleaning tool, and then perform image acquisition and coverage calculation again until a satisfactory cleaning effect is achieved.

[0183] Based on any of the foregoing solutions, in one possible implementation scheme, it also includes generating a cleaning agent flow control instruction according to the cleaning effect.

[0184] Exemplarily, in this solution, a control strategy for the flow rate of a cleaning agent is formulated based on the cleaning effect evaluation index, wherein a cleaning agent flow rate adjustment rule corresponding to different cleaning effect index ranges can be set;

[0185] For example, when the image similarity index is higher than a set threshold (e.g., 0.8) and the cleaning coverage is lower than a set value (e.g., 70%), it indicates that the cleaning effect is poor and there are many areas that are not cleaned, and the flow rate of the cleaning agent should be increased;

[0186] When the image contrast index is high (eg, greater than 0.6) and the cleaning coverage reaches a high level (eg, above 90%), it indicates that the cleaning effect is good and the flow rate of the cleaning agent can be appropriately reduced to avoid waste.

[0187] Exemplarily, in this solution, the control system of the cleaning equipment generates a detergent flow control instruction according to the index value calculated by the cleaning effect evaluation module and a pre-set control strategy. The instruction is transmitted to a detergent supply device, such as a detergent pump or a valve controller, through a communication interface to achieve precise adjustment of the detergent flow.

[0188] In this solution, during the entire cleaning process, the image acquisition and cleaning effect evaluation module works continuously, continuously collecting images after cleaning and calculating cleaning effect indicators, and the detergent flow control module adjusts the detergent flow in real time according to these indicators. This real-time feedback mechanism enables the cleaning equipment to quickly adapt to changes in the dirt on the surface of the cleaning object, ensuring that the appropriate detergent flow can be used at different cleaning stages, thereby improving cleaning efficiency and quality.

[0189] Illustratively, in this embodiment, any of the aforementioned schemes can be freely arranged and combined, and the specific implementation methods of each step are the same as the corresponding recorded contents above, and the specific contents will not be described in detail again.

[0190] exist Figure 1Based on the scheme shown, in one possible implementation scheme, determining the depth parameter of the dirt in the area to be cleaned includes determining the depth parameter using the following formula:

[0191]

[0192] In the formula, Z represents the depth parameter, f represents the focal length of the camera, and d represents the parallax between the images taken by the two cameras;

[0193] Determining the image similarity includes determining a similarity index using the following formula:

[0194]

[0195] Where SSIM(x,y) represents the similarity index, μx represents the pixel average of the first image, μy represents the pixel average of the second image, σxy represents the pixel covariance between the first image and the second image, and σx 2 represents the pixel variance of the first image, σy 2 represents the pixel variance of the second image, C1 represents the first constant term, and C2 represents the second constant term;

[0196] Determining the image contrast includes determining a contrast index according to the following formula:

[0197]

[0198] Wherein, Contrast represents the contrast index, max(I) represents the maximum pixel value of the second image, and min(I) represents the minimum pixel value of the second image;

[0199] Determining the cleaning coverage includes determining the cleaning area using the following formula:

[0200]

[0201] In the formula, A clean represents the cleaning area, I(x i ,y i ) represents the pixel value of the pixel in the i-th row and j-th column after the second image is binarized.

[0202] In this solution, if the image similarity is less than the first threshold, the image contrast is greater than the second threshold, and the cleaning coverage is greater than the third threshold, it can be determined that the cleaning effect meets expectations, otherwise it is determined that the cleaning effect does not meet expectations;

[0203] The specific values ​​of the first threshold, the second threshold, and the third threshold can be determined based on experience or simulation experiments.

[0204] In this scheme, the detergent spraying flow rate of the cleaning device is adjusted by the detergent flow control instruction according to the image similarity detergent flow control instruction, wherein the detergent flow control instruction is set to accurately control the flow rate within the range of 10mL / min to 100mL / min (error ±5%), and the spraying angle can be adjusted between 0° and 60° to achieve the best cleaning effect.

[0205] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A conductor cleaning agent, characterized in that: The invention comprises a mixed rust remover and a thickener, wherein the mixed rust remover and the thickener are in a gel state.

2. The conductor cleaning agent according to claim 1, characterized in that The thickener is an oily thickener.

3. The conductor cleaning agent according to any one of claims 1 and 2, characterized in that: The conductor cleaning agent is used for cleaning the surface of a conductor.

4. A cleaning device, characterized in that: A cleaning agent storage bottle is included, wherein the cleaning agent storage bottle stores the conductor cleaning agent according to claim 1 .

5. The cleaning device according to claim 1, characterized in that Also included is a controller and a camera, wherein the controller is used to execute a preset conductor cleaning control method; The camera is used to obtain image data required when the conductor cleaning control method is executed.

6. The cleaning device according to claim 5, characterized in that It also includes a power supply, a motor, and a brush head; The power supply is used to supply power to the motor and the detergent storage bottle. The motor is used to drive the brush head to rotate. The detergent storage bottle is also used to pump the conductive body detergent into the brush head.

7. The cleaning device according to claim 6, characterized in that Also included is an insulating rod, which serves as a handheld portion of the cleaning device.

8. The cleaning device according to claim 7, characterized in that The insulating rod is also provided with a telescopic structure.

9. The cleaning device according to claim 6, characterized in that: The brush head is equipped with a cotton wheel.

10. The cleaning device of claim 6, wherein the controller is further configured to adjust a flow rate of the conductor cleaning agent pumped into the brush head.

Images