A nozzle intelligent control optimization method for a carrier cleaning machine

Abstract

The present application provides a nozzle intelligent control optimization method for a carrier cleaning machine, which relates to the field of carrier cleaning technology. The method involves obtaining pollution distribution information of the carrier to be cleaned; dividing the carrier to be cleaned according to the pollution distribution information, and outputting an identified pollution area; detecting the real-time cleaning area of ​​the nozzle element in real time; activating the nozzle protection module if the real-time cleaning area meets the identified pollution area, and the nozzle protection module receives the pollution information of the identified pollution area and the real-time position information of the nozzle element; performing nozzle protection control analysis on the real-time position information and pollution information, outputting protection control parameters, and controlling the position of the nozzle element to move according to the protection control parameters. The present application solves the technical problem in the prior art that the nozzle is easily corroded by the contaminated return fluid due to the lack of effective control of the distance between the nozzle and the pollution area, and achieves the technical effect of extending the service life of the nozzle.

Description

Technical Field

[0001] The present application relates to the field of carrier cleaning technology, and in particular to a nozzle intelligent control optimization method for a carrier cleaning machine. Background Art

[0002] Carrier cleaning machines are widely used in industries such as semiconductor manufacturing and optoelectronic equipment to clean contaminants generated during the production process. Carrier cleaning machines are equipped with multiple nozzles that remove contaminants from the surface of the carrier by spraying cleaning fluid. Most existing carrier cleaning machines use a fixed nozzle position or a uniform spraying method. The movement of the nozzle is usually controlled according to a set program to perform a uniform cleaning operation on the carrier. This method does not take into account the differences in contamination distribution on the carrier. When cleaning severely contaminated areas, the cleaning fluid of the nozzle may mix with the contaminants on the carrier to form contaminated reflux liquid, which can easily cause corrosion and damage to the nozzle. This not only shortens the service life of the nozzle, but may also increase the maintenance cost of the equipment and affect the continuity of the cleaning process. Summary of the Invention

[0003] The present application provides an intelligent control optimization method for a nozzle of a vehicle cleaning machine, which solves the technical problem in the prior art of lacking effective control over the distance between the nozzle and the contaminated area, resulting in the nozzle being easily corroded by contaminated return fluid, thereby achieving the technical effect of extending the service life of the nozzle.

[0004] In view of the above problems, the present application provides a nozzle intelligent control optimization method for a carrier cleaning machine, the method comprising:

[0005] Acquire pollution distribution information of the carrier to be cleaned; divide the carrier to be cleaned according to the pollution distribution information, and output an identified pollution area, wherein the identified pollution area is an area where the pollution index is greater than a first preset pollution index; connect the nozzle element of the carrier cleaning machine to detect the real-time cleaning area of ​​the nozzle element in real time; if the real-time cleaning area meets the identified pollution area, activate the nozzle protection module, the nozzle protection module receives the pollution information of the identified pollution area and the real-time position information of the nozzle element; perform nozzle protection control analysis on the real-time position information and the pollution information, output protection control parameters, and control the movement of the nozzle element according to the protection control parameters.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] By capturing contamination distribution information on the carrier being cleaned, this system provides foundational data for subsequent precision cleaning, ensuring that each cleaning step is tailored to the specific contamination situation. By categorizing the carrier into distinct contamination zones, cleaning efforts can be focused on those areas requiring critical cleaning, avoiding ineffective cleaning of uncontaminated areas. This provides guidance for subsequent nozzle protection, conserving cleaning fluid and energy, and improving cleaning efficiency. By connecting the nozzle components of the carrier cleaning machine and monitoring their cleaning zones in real time, the alignment of the nozzles with the contaminated areas can be dynamically monitored, making the cleaning process more precise and providing accurate data for adjusting nozzle positioning and cleaning strategies. When the nozzle approaches a contaminated area, the nozzle protection module activates to prevent excessive contact with the contaminated fluid. The nozzle protection module acquires nozzle position and contamination information, performs nozzle protection control analysis, and outputs protection control parameters to prevent excessive contact with the contaminated fluid. This prevents corrosion from backflow during the cleaning process, extending nozzle life and improving cleaning stability.

[0008] In summary, this application achieves precise control of vehicle cleaning and effective protection of the nozzles through a series of steps such as obtaining vehicle contamination distribution information, dividing contaminated areas, detecting nozzle cleaning areas in real time, triggering the nozzle protection module and performing position control. In terms of vehicle cleaning, it can improve cleaning efficiency and avoid excessive or incomplete cleaning; in terms of nozzle protection, it effectively prevents the nozzles from being corroded by contaminated return fluid, thereby extending the service life of the nozzles, reducing equipment maintenance costs, and ensuring the stable progress of cleaning work. In addition, precise control of the nozzle position and cleaning area reduces the waste of water resources and cleaning fluid, thereby saving production costs.

[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 in accordance with the contents of the specification. 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 listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A flow chart of a method for intelligent nozzle control optimization of a carrier cleaning machine provided in an embodiment of the present application.

[0011] Figure 2 A schematic diagram of a flow chart for determining cleaning control parameters in a nozzle intelligent control optimization method for a carrier cleaning machine provided in an embodiment of the present application.

[0012] Figure 3 A flow chart of outputting protection control parameters in a nozzle intelligent control optimization method for a vehicle cleaning machine provided in an embodiment of the present application. DETAILED DESCRIPTION

[0013] The embodiments of the present application provide a nozzle intelligent control optimization method for a carrier cleaning machine, obtain contamination distribution information of the carrier to be cleaned, divide the carrier to be cleaned according to the contamination distribution information, and identify the contamination area; detect the real-time cleaning area of ​​the nozzle element in real time, and when the real-time cleaning area meets the identified contamination area, activate the nozzle protection module to perform nozzle protection control analysis on the real-time position information and contamination information, output protection control parameters, and control the position of the nozzle element according to the protection control parameters. This solves the technical problem in the prior art that the nozzle is easily corroded by contaminated return fluid due to the lack of effective control of the distance between the nozzle and the contaminated area, and achieves the technical effect of extending the service life of the nozzle.

[0014] like Figure 1 As shown, an embodiment of the present application provides a nozzle intelligent control optimization method for a carrier cleaning machine, the method comprising:

[0015] Step S1: Obtaining pollution distribution information of the carrier to be cleaned.

[0016] Specifically, a carrier to be cleaned is a tool used to hold semiconductor materials or devices during the semiconductor manufacturing process, such as a silicon wafer carrier. Contamination distribution information refers to the contamination at different locations on the surface of the carrier to be cleaned, including information such as the type, density, and location of the contaminants.

[0017] Before cleaning begins, the carrier to be cleaned is scanned and inspected to obtain the distribution of contaminants on its surface. The inspection process can be achieved through high-precision sensors, scanning equipment or imaging technology. For example, a high-resolution camera is used to photograph the surface of the carrier, or a sensor is used to detect the chemical composition of the contaminants, to obtain detailed images or data of the contamination on the surface of the carrier and generate a contamination distribution map. These data will show the degree of contamination in each area on the carrier, thus providing a basis for the subsequent cleaning process. For example, for a carrier used to carry semiconductor wafers, a contamination distribution map is obtained by scanning before cleaning, and it is found that the contaminants are more concentrated in the corners or holes of the carrier. By obtaining the contamination distribution information of the carrier to be cleaned, high-contamination areas can be identified, providing data support for the next cleaning work.

[0018] Step S2: Divide the carrier to be cleaned according to the pollution distribution information, and output an identified pollution area, wherein the identified pollution area is an area with a pollution index greater than a first preset pollution index.

[0019] Specifically, the marked contaminated areas refer to the areas on the vehicle that are more seriously polluted, determined based on the pollution distribution information and the first preset pollution index. These areas require special attention and special treatment. The first preset pollution index is a pre-set standard value used to determine whether a certain area on the vehicle is a seriously polluted area. This index can be a specific value of the pollutant concentration or the pollutant area. The pollution distribution information obtained in step S1 is compared with the preset pollution index. By analyzing the degree of pollution in different areas of the vehicle surface, the areas where the pollutant concentration is greater than the first preset pollution index are marked, i.e., the marked contaminated areas. By dividing the contaminated areas, cleaning control can be carried out in a more targeted manner, improving the accuracy of cleaning, and providing key area information for nozzle protection to prevent the nozzle from being too close to seriously polluted areas.

[0020] Step S3: connecting to a nozzle element of a carrier cleaning machine and detecting a real-time cleaning area of ​​the nozzle element in real time.

[0021] Specifically, a nozzle element is the component in a vehicle washer that sprays cleaning fluid. Typically consisting of multiple nozzles, the nozzle element is used to evenly or directionally apply the cleaning fluid to the vehicle. The live cleaning area refers to the area of ​​the vehicle surface being cleaned by the nozzle element at any given moment.

[0022] The movement trajectory and working range of the nozzle are monitored in real time using position sensors and cleaning area monitoring equipment (such as micro cameras or laser positioning devices) installed on the nozzle element. During the cleaning process, the nozzle element will continuously move on the surface of the vehicle. The position sensor can determine the real-time position of the nozzle element, while the camera or laser positioning device can locate the area of ​​the vehicle surface covered by the nozzle cleaning fluid in real time, that is, the real-time cleaning area. Through these monitoring devices, the cleaning status of the nozzle can be monitored in real time, providing accurate data for subsequent judgment of whether the nozzle is approaching a dangerous contaminated area, so that protective measures can be taken in a timely manner.

[0023] Step S4: If the real-time cleaning area meets the identified contaminated area, the nozzle protection module is activated, and the nozzle protection module receives the contamination information of the identified contaminated area and the real-time position information of the nozzle element.

[0024] Specifically, the printhead protection module is a functional module dedicated to protecting the printhead. It can make decisions based on the information it receives to prevent the printhead from being corroded by contaminated return fluid. Real-time position information refers to the current position of the printhead during the cleaning process.

[0025] When the real-time cleaning area of ​​the printhead matches the identified contaminated area, the printhead protection module is activated. The printhead protection module receives contamination information (such as the type and area of ​​the contaminant) and the real-time position information of the printhead (such as the height and horizontal position of the printhead from the carrier surface) through the data interface to determine whether to adjust the printhead position or take other protective measures.

[0026] By timely activating the nozzle protection module and reacting when the nozzle approaches the contaminated area, the nozzle can be prevented from being corroded by the contaminated return fluid and the normal operation of the nozzle can be guaranteed.

[0027] Step S5: performing nozzle protection control analysis on the real-time position information and the pollution information, outputting protection control parameters, and controlling the movement of the nozzle element according to the protection control parameters.

[0028] Specifically, the protection control parameters are parameters used to control the movement of the nozzle, such as the direction and distance the nozzle should move, obtained by analyzing the real-time position information of the nozzle element and the pollution information identifying the contaminated area.

[0029] The nozzle protection module uses an internal algorithm to comprehensively analyze real-time position and pollution information to generate protection control parameters. For example, if a nozzle is too close to a severely polluted area, the module calculates that the nozzle should move back 5 cm and up 3 cm based on factors such as the diffusion characteristics of the contaminated return fluid and the nozzle's corrosion resistance. This parameter is then transmitted to the nozzle's drive device (such as a motor) to control the position and movement of the nozzle components.

[0030] By controlling the movement of the nozzle elements through protection control parameters, the position of the nozzle can be precisely controlled, effectively preventing the nozzle from being corroded by contaminated return fluid, extending the service life of the nozzle, and ensuring the continuous cleaning work.

[0031] Further, such as Figure 2 As shown, after controlling the movement of the nozzle element according to the protection control parameter, the method further includes:

[0032] Step S61: Acquire cleaning control parameters of the nozzle element based on the protection control parameters, wherein the cleaning control parameters include a cleaning angle, a cleaning pressure, and a cleaning flow rate.

[0033] Step S62: Predicting the cleaning effect of the cleaning control parameters and outputting the predicted cleaning effect.

[0034] Step S63: If the predicted cleaning effect is less than a preset threshold, adaptively optimize the cleaning control parameters with the preset threshold as a feedback target, and output cleaning accompanying control parameters corresponding to the protection control parameters.

[0035] Specifically, cleaning control parameters are key parameters used to control the cleaning process, mainly including cleaning angle, cleaning pressure and cleaning flow. These parameters will affect the spray effect and cleaning quality of the cleaning liquid. After the nozzle element is controlled to move according to the protection control parameters, the sensor on the interactive nozzle element or the control system connected to the nozzle obtains the current cleaning control parameters. For example, the cleaning pressure is monitored by installing a pressure sensor inside the nozzle, the cleaning flow is monitored by a flow sensor, and the cleaning angle information is obtained through the mechanical structure and angle adjustment device of the nozzle. A comprehensive understanding of the parameters related to the cleaning ability of the nozzle in the current position provides a data basis for subsequent prediction and optimization of the cleaning effect, which helps to improve the accuracy and effectiveness of cleaning.

[0036] The cleaning effect after cleaning is predicted based on the cleaning control parameters that have been obtained. The cleaning effect can be expressed by indicators such as cleanliness and dryness. For example, a mathematical model is used to predict the cleaning effect. A theoretical model is established based on the empirical formula between the cleaning angle, pressure, flow rate and pollutant removal rate or based on physical principles. For example, for a certain type of pollutant, the pollutant removal rate is proportional to the square of the cleaning pressure, proportional to the cleaning flow rate, and proportional to the cosine value of the cleaning angle. The obtained cleaning control parameters are substituted into this model to calculate the predicted cleaning effect. A machine learning algorithm can also be used to train with a large amount of cleaning experimental data, input the cleaning control parameters, and output the predicted cleaning effect. Predicting the cleaning effect before actual cleaning can timely discover possible problems such as incomplete cleaning, provide a basis for further optimization, avoid ineffective cleaning operations, and improve cleaning quality.

[0037] The preset threshold is a pre-set standard value used to measure whether the cleaning effect meets the requirements. The cleaning accompanying control parameter is a parameter obtained by optimizing the cleaning control parameter with the preset threshold as the target. This parameter is a parameter that can meet the cleaning effect requirements when the nozzle moves according to the protection control parameter. When the predicted cleaning effect is less than the preset threshold, the cleaning control parameter is adjusted using an optimization algorithm. For example, if the predicted cleaning effect is not ideal because the cleaning pressure is insufficient, the pressure value that needs to be increased can be calculated based on a certain proportional relationship according to the preset threshold and the current prediction result, and parameters such as the cleaning flow rate and cleaning angle can be adjusted accordingly to ensure that the cleaning effect requirements are achieved while meeting the nozzle protection. An algorithm based on the feedback control principle can also be used to continuously compare the gap between the predicted cleaning effect and the preset threshold, gradually adjust the cleaning control parameters until the requirements are met, and finally output the cleaning accompanying control parameters.

[0038] By optimizing the cleaning control parameters of the nozzle elements under the protection control parameters, it is possible to ensure that the cleaning effect can meet the requirements. While protecting the nozzle, the carrier can be effectively cleaned, the overall quality of the carrier cleaning can be improved, and subsequent production problems caused by incomplete cleaning can be reduced.

[0039] Further, such as Figure 3 As shown, step S5 includes:

[0040] Step S51: performing pollution backflow risk identification based on the real-time location information and the pollution information to obtain a pollution backflow risk index.

[0041] Step S52: When the contaminated backflow risk index is greater than the preset backflow risk index, an objective function is constructed to minimize the backflow risk index difference to perform nozzle protection control analysis, wherein the response data of the objective function includes a position movement vector.

[0042] Step S53: Decomposing the position movement vector according to the movement control unit of the nozzle element, and outputting protection control parameters.

[0043] Specifically, contamination backflow refers to the possibility that during the cleaning process, cleaning fluid and contaminants may flow back from the contaminated area to the components of the nozzle, causing contamination, corrosion or damage to the nozzle. This backflow phenomenon will cause the cleaning fluid to mix with the contaminants, reducing the cleaning effect and exacerbating nozzle wear. The risk of contamination backflow is identified using real-time position information (the position of the nozzle) and contamination information (the location and concentration distribution of the contaminated area). By calculating factors such as the relative position of the nozzle and the contaminated area, the flow direction of the cleaning fluid, and the nozzle injection pressure, the risk of contaminants possibly flowing back into the nozzle is inferred, and the risk of contamination backflow is quantified to generate a contamination backflow risk index. This index is usually a numerical value, and the larger the value, the higher the backflow risk. Through contamination backflow risk identification, potential contamination backflow problems can be predicted and identified in a timely manner, so that corresponding protective measures can be taken to reduce damage to the nozzle and waste of cleaning fluid, ensuring the stability and efficiency of the cleaning process.

[0044] The preset backflow risk index is a pre-set risk threshold that indicates the maximum acceptable backflow risk. When the backflow risk index exceeds this value, the backflow risk is considered excessive and protective measures are required. When the backflow risk index exceeds the preset backflow risk index, a sprinkler protection control analysis is triggered. With the goal of minimizing the difference between the backflow risk index and the preset backflow risk index, an objective function is constructed. This objective function can be constructed based on physical principles and mathematical models, taking into account factors such as the location of the sprinkler and the characteristics of the contaminated area. For example, a quadratic objective function can be established based on the distance between the sprinkler and the contaminated area and the diffusion patterns of the contaminants. This objective function is then solved using an optimization algorithm (such as gradient descent or genetic algorithm) to obtain a position movement vector, thereby reducing the backflow risk. This position movement vector is the response data of the objective function, indicating the direction and distance the sprinkler needs to move. By constructing and analyzing the objective function, the direction and distance required for sprinkler movement can be scientifically determined to minimize the risk of contaminated backflow, effectively protecting the sprinkler from corrosion caused by contaminated backflow fluid.

[0045] The motion control unit is a functional unit in the nozzle element used to control the movement of the nozzle. It usually includes a motor, servo controller, or stepper controller, and is responsible for adjusting the position of the nozzle element according to control instructions. The motion control unit of the nozzle element receives the position movement vector from the objective function and decomposes this movement vector into components in different directions. For example, if the motion control unit is a three-dimensional motion control based on a Cartesian coordinate system and the position movement vector is (3, 4, 5), then according to the working principle of the motion control unit, this vector is decomposed into protection control parameters such as the movement distance in the x-direction is 3 units, the movement distance in the y-direction is 4 units, and the movement distance in the z-direction is 5 units.

[0046] By analyzing the risk of contamination backflow in real time and generating protection control parameters, precise printhead position adjustment can be achieved to minimize the risk of backflow. This not only protects the printhead from damage by contaminants, but also ensures the continued effectiveness of the cleaning process.

[0047] Furthermore, step S51 includes:

[0048] Step S511: obtaining flow samples of a historical cleaning process, performing trajectory simulation on the flow samples of the historical cleaning process using fluid dynamics equations, and constructing a contamination backflow model.

[0049] Step S512: Acquire a position sample and a contamination information sample of the nozzle element, wherein the contamination information sample includes a carrier geometry sample, a contaminant type sample, and a cleaning parameter sample.

[0050] Step S513: loading the contaminated backflow model, training it based on the position samples, contamination information samples, cleaning parameter samples, and labels identifying the degree of backflow damage to the nozzle components, and generating a contaminated backflow risk identification model.

[0051] Step S514: performing pollution backflow risk identification on the real-time location information and the pollution information according to the pollution backflow risk identification model to obtain a pollution backflow risk index.

[0052] Specifically, historical cleaning process flow samples refer to the flow behavior data of the cleaning fluid on the vehicle surface during past cleaning operations, including parameters such as the cleaning fluid's flow trajectory, velocity, and direction. The contamination backflow model is constructed based on these historical cleaning process flow samples and fluid dynamics equations. It describes the contamination backflow phenomenon and reflects the flow and backflow of cleaning fluid and contaminants on the vehicle surface.

[0053] The data recording module of the interactive cleaning equipment collects flow sample data of the historical cleaning process from sensor records or historical cleaning experiment data. These data are then substituted into the fluid dynamics equation, and trajectory simulation is performed using computer simulation software (such as ANSYS Fluent) to construct a contamination reflux model. This model simulates the flow trajectory of the cleaning fluid to infer whether the cleaning fluid will flow back to the nozzle area, and provides data support for the prediction of reflux risk. For example, in the simulation, parameters such as the initial flow rate of the cleaning fluid and the roughness of the carrier surface are set, and the flow trajectory of the cleaning fluid on the carrier surface at different times is calculated according to the fluid dynamics equation, thereby constructing a contamination reflux model.

[0054] Position samples are data sets that record the position of the printhead components during different cleaning processes. For example, the coordinate position of the printhead at different points above the carrier. Contamination information samples include various contamination-related data that impact cleaning effectiveness and backflow risk, such as carrier geometry samples, contaminant type samples, and cleaning parameter samples. Carrier geometry samples are data records of different carrier shapes (such as square, round, and curved). Carrier geometry affects the flow of cleaning fluid and contaminants. Contaminant type samples include data records of different types of contaminants, such as chemical residues, dust, and metal particles. Different types of contaminants have different physical and chemical properties, which can affect contaminant backflow. Cleaning parameter samples include data records of cleaning operation-related parameters such as cleaning fluid flow rate, pressure, and cleaning angle. Printhead component position samples, carrier geometry samples, contaminant type samples, and cleaning parameter samples are obtained from historical operating data, experimental data, and sensor monitoring data of the cleaning equipment. For example, position samples are obtained using position sensors installed around the printhead and carrier, carrier geometry samples are obtained through surface inspection and analysis of the carrier, contaminant type samples are obtained through compositional analysis of contaminants, and cleaning parameter samples are obtained through the cleaning equipment's control system. These sample data provide rich input information for building a pollution backflow risk identification model, which can comprehensively consider various factors affecting the pollution backflow risk and improve the reliability of the model.

[0055] The backflow damage label indicates the degree of damage to printhead components caused by contamination backflow. For example, it can be represented by a value between 0 and 1, with 0 indicating no damage and 1 indicating severe damage. The contamination backflow risk identification model accurately identifies the contamination backflow risk based on input data such as printhead location, contamination information, and cleaning parameters.

[0056] The constructed contaminated backflow model is loaded into a machine learning algorithm framework (such as TensorFlow, PyTorch, etc.). Then, position samples, contamination information samples, cleaning parameter samples, and labels of the degree of backflow damage are input into the model as training data to train the contaminated backflow risk identification model. For example, a neural network algorithm in a supervised learning algorithm is used to continuously adjust the weight parameters of the model according to the input data and labels. After multiple iterative trainings, until the prediction accuracy of the model reaches a certain requirement, a contaminated backflow risk identification model is generated. The generated contaminated backflow risk identification model can accurately identify the contaminated backflow risk of the nozzle at different positions, under different contamination and cleaning conditions, and provides an effective tool for the subsequent acquisition of contaminated backflow risk indicators.

[0057] Real-time location and contamination information are input into the generated contamination backflow risk identification model. The model calculates the contamination backflow risk index based on pre-learned patterns and algorithms. By obtaining the contamination backflow risk index for the current state of the printhead, it provides a basis for determining whether to perform printhead protection control analysis, thereby effectively protecting the printhead from corrosion caused by contaminated backflow fluid.

[0058] Furthermore, step S514 includes:

[0059] Step S514 - 1 : Load the contaminated reflux model, perform reflux prediction based on the position samples, contamination information samples, and cleaning parameter samples, and output a first predicted reflux radius.

[0060] Step S514 - 2 : Taking the carrier to be cleaned as the center, divide the area according to the first predicted reflow radius and output a safe area.

[0061] Step S514 - 3 : The contamination backflow risk identification model performs contamination backflow risk identification on the real-time location information and the contamination information in the safe area.

[0062] Specifically, the first predicted reflux radius is the radius of the area where contamination reflux may occur, predicted by the contamination reflux model based on location samples, contamination information samples, and cleaning parameter samples. The constructed contamination reflux model is loaded and the location samples, contamination information samples, and cleaning parameter samples are input into the contamination reflux model. The model outputs the first predicted reflux radius based on fluid dynamics principles, taking into account factors such as the flow rate of the cleaning fluid and the diffusion characteristics of contaminants.

[0063] The safe zone is the area outside the first predicted return flow radius, centered on the carrier being cleaned. This area is relatively less affected by contaminated return flow. The zone is divided based on the value of the first predicted return flow radius, centered on a specific point on the carrier being cleaned (such as its geometric center). For example, if the first predicted return flow radius is 5 cm, with the carrier center as the origin, then the area within a 5 cm radius from the nozzle element to the carrier center is the safe zone.

[0064] Real-time location and contamination information are input into the contamination backflow risk identification model, which then performs risk identification within a defined safe zone. Based on the learned relationships between location and contamination within the safe zone, the contamination backflow risk identification model identifies the contamination backflow risk of the nozzle at its current location and under its current contamination status. Performing contamination backflow risk identification within a specific safe zone allows for more accurate assessment of the nozzle's contamination backflow risk, avoiding unnecessary complex calculations over a large area. This improves the speed and accuracy of risk identification, thereby supporting the subsequent accurate acquisition of contamination backflow risk indicators.

[0065] Furthermore, the outputting of the marked contaminated area in step S2 further includes:

[0066] Acquire the cleaning area of ​​the nozzle element; and constrain the area of ​​the marked contaminated region according to the cleaning area.

[0067] Specifically, the cleaning area refers to the area that the nozzle element can cover and clean in one operation. The cleaning area can be determined by the design parameters of the nozzle element and experimental tests. When identifying the contaminated area, the size or number of the contaminated area is adjusted taking into account the limitation of the nozzle cleaning area. Compare the cleaning area of ​​the obtained nozzle element with the area of ​​the identified contaminated area. If the area of ​​the identified contaminated area is larger than the cleaning area of ​​the nozzle element, the identified contaminated area needs to be divided or redefined to ensure that the size of each identified contaminated area does not exceed the cleaning capacity of the nozzle. For example, the identified contaminated area can be divided into several sub-areas according to the cleaning area, and cleaned one by one in a certain order.

[0068] This constraint can ensure that the cleaning work is carried out within the capabilities of the nozzle, avoiding problems such as incomplete cleaning or excessive cleaning time due to the large area of ​​the marked contaminated area, and improving the overall efficiency and quality of the cleaning work.

[0069] Furthermore, the connected carrier cleaning machine in step S3 also includes a self-cleaning component, which is connected to the nozzle element; when the nozzle element is cleaned, the real-time pollution index of the nozzle element is detected; if the real-time pollution index is greater than a second preset pollution index, the self-cleaning component is started to self-clean the nozzle element.

[0070] Specifically, the self-cleaning component is an additional device connected to the nozzle element, which can automatically perform a cleaning operation after the nozzle cleaning is completed. The real-time contamination index is a numerical value of the concentration or degree of contaminants on the nozzle element monitored during the cleaning process. The contamination index may be detected based on optical sensors, laser detection, pressure changes, etc., indicating the amount or concentration of contaminants remaining on the nozzle. The second preset contamination index is a pre-set contamination threshold used to determine whether the nozzle needs to perform a self-cleaning operation. If the detected real-time contamination index exceeds the threshold, it means that there are too many contaminants accumulated on the nozzle and the self-cleaning program needs to be started.

[0071] In addition to the nozzle element, the design of the carrier cleaning machine also incorporates a self-cleaning component. This component automatically cleans the nozzle after it has finished cleaning the carrier to remove any contaminants that may have adhered to it. Self-cleaning components typically include a backwash mechanism, a cleaning fluid circulation system, or other cleaning technologies. While establishing a connection with the carrier cleaning agent, a communication connection with the self-cleaning component is also established. After the nozzle element completes the cleaning operation, real-time contamination indicators are measured. For example, an optical sensor can be used to detect the intensity of reflected light from the nozzle element surface, inferring contaminant coverage based on changes in reflected light intensity. Alternatively, an electrical sensor can be used to detect changes in the nozzle element's electrical conductivity to determine the presence and extent of contamination. The data collected by these sensors is then input into a pre-established computational model. This model, based on experimental data and theoretical analysis, determines the relationship between different sensor data and contamination indicators, thereby calculating real-time contamination indicators.

[0072] The detected real-time pollution index is compared with a second preset pollution index. If the real-time pollution index is greater than the second preset pollution index, a control command is initiated to activate the self-cleaning component, which controls the self-cleaning component to self-clean the nozzle element. Exemplarily, the self-cleaning process includes controlling the automatic movement of the nozzle element to a cleaning position, activating a cleaning liquid supply system, and executing a cleaning program. The cleaning liquid supply device supplies cleaning liquid at a set pressure and flow rate, the flushing nozzle sprays the cleaning liquid evenly on the nozzle element, and the cleaning liquid recovery device promptly recovers the used cleaning liquid.

[0073] Through the above steps, automatic cleaning control of the nozzle components can be achieved, pollutants on the nozzle components can be removed in a timely and effective manner, the cleanliness of the nozzle components can be ensured, the working efficiency and service life of the nozzle components can be improved, and the working performance of the entire carrier cleaning machine can be improved.

[0074] Furthermore, step S1 includes:

[0075] Step S11: installing a pollution sensing device to collect surface pollution data of the vehicle to be cleaned.

[0076] Step S12: collecting the surface contamination image of the carrier to be cleaned using a visual recognition device.

[0077] Step S13: Analyze the surface contamination data and the surface contamination image to obtain contamination distribution information of the carrier to be cleaned.

[0078] Specifically, the pollution sensing device is a device used to detect the contamination of the surface of the carrier to be cleaned. It senses the presence of pollutants through physical or chemical principles and converts them into measurable data. Such as optical sensors, chemical sensors, etc. The pollution sensing device is installed in a suitable position, and this position must be able to effectively contact the surface of the carrier to be cleaned. For example, for a flat carrier, the pollution sensing device can be installed on a movable bracket above the carrier to fully collect data. When collecting data, the pollution sensing device operates according to its own working principle to obtain surface pollution data of the carrier to be cleaned, such as pollutant concentration, type, etc. For example, a sensor using optical principles will emit light of a specific wavelength to the surface of the carrier, and then receive the reflected light. By analyzing information such as the intensity and wavelength change of the reflected light, the surface pollution data is calculated.

[0079] A visual recognition device refers to a device used to capture images of surface contamination on the carrier to be cleaned, such as a high-resolution camera or infrared camera. The visual recognition device should be installed in a suitable location to ensure that the surface of the carrier to be cleaned is fully captured. For example, for irregularly shaped carriers, multiple visual recognition devices can be installed to capture images from different angles. After the visual recognition device is activated, it captures images of the carrier surface according to the set resolution, frame rate, and other parameters to obtain an image of the surface contamination of the carrier to be cleaned. This image intuitively demonstrates the contamination status of the carrier surface.

[0080] The acquired surface contamination data and surface contamination images are comprehensively analyzed. For example, image analysis software can be used to process the surface contamination images to identify the shape and location of the contaminated areas within the images. This information, combined with information such as contaminant concentration from the surface contamination data, can be used to determine the distribution of contamination within the contaminated areas. Data fusion technology can be used to integrate information from these two sources to obtain contamination distribution information for the vehicle being cleaned.

[0081] Through the above steps, the pollution distribution information of the carrier to be cleaned can be accurately obtained, thereby accurately identifying the contaminated area and providing data support for subsequent nozzle control and cleaning parameter adjustment.

[0082] In summary, the embodiment of the present application provides a nozzle intelligent control optimization method for a carrier cleaning machine, which has the following technical effects:

[0083] First, by installing pollution sensors and visual recognition devices, the pollution distribution information of the carrier to be cleaned is collected and analyzed in real time, and the location, shape and concentration of the contaminated area are accurately identified, thereby providing detailed data support for subsequent cleaning decisions. Next, the contaminated area is divided according to the pollution distribution information, and the contaminated area is constrained in combination with the cleaning area of ​​the nozzle to ensure that the cleaning range matches the nozzle capacity and avoid unnecessary waste of resources. Based on real-time location information and pollution data, the risk of contamination backflow is identified, and the nozzle movement path is optimized through the objective function to minimize the risk of contamination backflow, avoid corrosive damage to the nozzle by pollutants, further protect the nozzle and optimize cleaning efficiency. Finally, the pollution backflow risk identification model and self-cleaning control are combined to ensure that the nozzle not only cleans the contaminated area efficiently, but also automatically performs maintenance after cleaning, extending the life of the nozzle. In addition, before the cleaning operation, the cleaning control parameters are adaptively optimized through cleaning effect prediction to ensure cleaning quality while avoiding contact between the nozzle and pollutants.

[0084] Overall, the embodiments of this application address the prior art issue of printhead corrosion due to backflow of contaminated fluid through intelligent control and real-time monitoring, effectively protecting the printhead. This method effectively extends the life of the printhead by accurately identifying contaminated areas and dynamically adjusting the printhead's position and cleaning strategy, ensuring long-term stable operation of the equipment. This method also avoids ineffective cleaning of uncontaminated areas, conserving cleaning fluid and energy, and achieving a more efficient, safe, and economical vehicle cleaning process.

[0085] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one 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 is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A nozzle intelligent control optimization method for a carrier cleaning machine, characterized in that: The method comprises: Obtaining contamination distribution information of the vehicle to be cleaned; Dividing the carrier to be cleaned according to the pollution distribution information and outputting a marked pollution area, wherein the marked pollution area is an area where the pollution index is greater than a first preset pollution index; Connecting to a nozzle element of a carrier cleaning machine to detect a real-time cleaning area of ​​the nozzle element in real time; If the real-time cleaning area meets the identified contaminated area, activating a nozzle protection module, the nozzle protection module receives contamination information of the identified contaminated area and real-time position information of the nozzle element; Performing a nozzle protection control analysis on the real-time position information and the pollution information, outputting a protection control parameter, and controlling the movement of the nozzle element according to the protection control parameter; Performing nozzle protection control analysis on the real-time position information and the pollution information and outputting protection control parameters, the method includes: Identify pollution backflow risks based on the real-time location information and the pollution information, and obtain a pollution backflow risk indicator; When the contaminated backflow risk index is greater than a preset backflow risk index, an objective function is constructed to minimize the backflow risk index difference to perform nozzle protection control analysis, wherein the response data of the objective function includes a position movement vector; The position movement vector is decomposed according to the movement control unit of the nozzle element, and a protection control parameter is output.

2. The method according to claim 1, wherein After controlling the movement of the nozzle element according to the protection control parameter, the method includes: Acquiring cleaning control parameters of the nozzle element based on the protection control parameters, wherein the cleaning control parameters include a cleaning angle, a cleaning pressure, and a cleaning flow rate; Predicting the cleaning effect of the cleaning control parameters and outputting the predicted cleaning effect; If the predicted cleaning effect is less than a preset threshold, the cleaning control parameter is adaptively optimized with the preset threshold as a feedback target, and a cleaning accompanying control parameter corresponding to the protection control parameter is output.

3. The method according to claim 1, wherein Identifying pollution backflow risk based on the real-time location information and the pollution information to obtain a pollution backflow risk index includes: Obtaining flow samples from a historical cleaning process, simulating trajectories of the flow samples from the historical cleaning process using fluid dynamics equations, and constructing a contamination backflow model; Acquire a position sample of the nozzle element and a contamination information sample, wherein the contamination information sample includes a carrier geometry sample, a contaminant type sample, and a cleaning parameter sample; Loading the contaminated backflow model, training it based on the position samples, contamination information samples, cleaning parameter samples, and labels identifying the degree of backflow damage to the nozzle components, to generate a contaminated backflow risk identification model; The pollution backflow risk is identified by performing pollution backflow risk identification on the real-time location information and the pollution information according to the pollution backflow risk identification model to obtain a pollution backflow risk index.

4. The method according to claim 3, wherein Performing pollution backflow risk identification on the real-time location information and the pollution information according to the pollution backflow risk identification model further includes: Loading the contaminated reflux model, performing reflux prediction based on the position samples, contamination information samples, and cleaning parameter samples, and outputting a first predicted reflux radius; Taking the carrier to be cleaned as the center, dividing the area according to the first predicted reflux radius and outputting a safe area; The contamination backflow risk identification model performs contamination backflow risk identification on the real-time location information and the contamination information in the safe area.

5. The method according to claim 1, wherein The output identifies the contaminated area, and the method further includes: Obtaining a cleaning area of ​​the nozzle element; The area of ​​the marked contaminated region is constrained according to the cleaning area.

6. The method according to claim 1, wherein The connection carrier cleaning machine further includes a self-cleaning component connected to the nozzle element; After the cleaning of the nozzle element is completed, detecting the real-time pollution index of the nozzle element; If the real-time pollution index is greater than a second preset pollution index, the self-cleaning component is activated to perform self-cleaning on the nozzle element.

7. The method according to claim 1, wherein Obtaining contamination distribution information of the carrier to be cleaned, the method includes: Installing a pollution sensing device to collect surface pollution data of the vehicle to be cleaned; According to the visual recognition device, a surface contamination image of the carrier to be cleaned is collected; An analysis is performed based on the surface contamination data and the surface contamination image to obtain contamination distribution information of the carrier to be cleaned.

Images