Writing brush cleaning machine nozzle moving track control system

Abstract

The invention relates to the technical field of writing brush cleaning, and particularly discloses a writing brush cleaning machine nozzle moving track control system which comprises the steps that in each cleaning task, writing brush data and cleaning data thereof are obtained, feature extraction is conducted on the writing brush data and the cleaning data thereof, and a cleaning reference value is obtained; synchronously obtaining the cleaning duration; performing statistics on cleaning reference values and cleaning durations of the cleaning tasks of preset times, and training a neural network model from the cleaning reference values to the cleaning durations; and generating feedback data based on the neural network model. According to the method, the writing brush data and the cleaning data in the multiple cleaning tasks are obtained, the feature values are extracted, the cleaning duration is synchronously obtained, the neural network model from the feature values to the cleaning duration is constructed, in actual use, the constructed neural network model can be used for feeding back tentative adjustment of a user, the flexibility is extremely high, and the user experience is improved. And a user can be assisted to make an optimal cleaning scheme according to actual conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of writing brush cleaning, and particularly to a control system for the running track of a nozzle of a writing brush cleaning machine. Background Art

[0002] The art of Chinese calligraphy in our country has a long history and has developed for about 4,000 years. It is one of the most elegant arts of mankind and an important manifestation of traditional Chinese culture. Currently, calligraphy is increasingly popular among the people. According to incomplete statistics, there are more than 50 million calligraphy lovers globally. The follow-up work after calligraphy practice and creation, including cleaning writing brushes, cleaning ink trays, and treating sewage, etc., is hereinafter referred to as "calligraphy follow-up work" for short. Usually, after calligraphy activities, calligraphy lovers must clean the writing brushes and ink trays. Otherwise, the glue in the ink will cause irreversible damage to the bristles of the writing brush, and the remaining ink in the ink tray will be unusable after drying. Even if a small amount of dried ink clumps is used for writing, it will also cause damage to the bristles. Therefore, calligraphy follow-up work is an inevitable link. However, the reality is that calligraphy follow-up work often brings negative emotions and impacts: It takes 3 - 5 minutes to clean a writing brush properly. Incorrect brush cleaning methods may even damage the writing brush. Due to the characteristics of the ink, the cleaning process will dirty utensils such as sinks and basins, requiring secondary cleaning. If the ink gets on the hands, the hands need to be cleaned. If the ink is accidentally splashed on clothes and other items, it will be even more difficult to clean.

[0003] The patent with the application publication number CN111596610B discloses an industrial control computer control system with a running track measurement and control function. By means of a highly abnormal track level, an excellent running track level, a deep analysis of the abnormal bending level of the track, and a deep analysis of the constant change level of the track, the running track curvature abnormal text is respectively edited and sent to the display screen. Then, the running track is classified into each region and each interval, and through the track length, boundary area, and projection area conditions therein, the measurement and control situations of each section of the running track are specifically divided. Furthermore, a track curvature analysis combining the peak-to-valley amplitude, the number of peak valleys, and the characteristic point spacing of the measurement and control situation is carried out, and based on this, precise track feedback and recording operations are made, and a targeted evaluation and in-depth processing of the overall running track are carried out to achieve a gradually advancing detailed track measurement and control effect.

[0004] Existing cleaning equipment does not have a prediction function. That is, it is difficult for users to know in advance the cleaning effects corresponding to different cleaning methods. This means that the cleaning process completely depends on a preset fixed scheme, and it is difficult for users to make tentative adjustments. Although the fixed scheme can bring a relatively good cleaning effect, since the writing brushes themselves are different, for example, the ink content of brushes with different usage durations is different, this fixed scheme reduces the possibility of achieving a better cleaning effect. How to provide a feedback architecture to provide adjustment references for users is the technical problem that the technical solution of the present invention wants to solve. Summary of the Invention

[0005] The purpose of the present invention is to provide a control system for the running trajectory of the nozzle of a writing brush cleaning machine to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] A control system for the running trajectory of the nozzle of a writing brush cleaning machine, the system includes:

[0008] A feature extraction module, which is used to obtain writing brush data and its cleaning data in each cleaning task, extract features from the writing brush data and its cleaning data to obtain a cleaning reference value; wherein, the writing brush data includes weight data, length data and size data, and the cleaning data includes intensity data, frequency data and angle data;

[0009] A cleaning duration calculation module, which is used to obtain the cleaning process of each cleaning task and calculate the cleaning duration according to the cleaning process;

[0010] A model training module, which is used to count the cleaning reference values and their cleaning durations of a preset number of cleaning tasks, and train a neural network model from the cleaning reference value to the cleaning duration;

[0011] A feedback data generation module, which is used to open an information acquisition port when receiving a cleaning request, obtain the current writing brush data and its cleaning data, extract the cleaning reference value of the current writing brush data and its cleaning data, and generate feedback data based on the neural network model.

[0012] As a further solution of the present invention: the feature extraction module includes:

[0013] A weight data acquisition unit, which is used to monitor the device running signal in real time. When the device running signal is monitored, it collects the writing brush weight value through a weight sensing module built in the device as the weight data;

[0014] A size data acquisition unit, which is used to obtain the length value and the pen barrel circumference value of the writing brush based on a visual sensor built in the device to obtain the size data and the length data;

[0015] The device parameter acquisition unit is configured to acquire the spray intensity value, spray frequency value, and spray angle value based on the sensor connected to the nozzle, and obtain the intensity data, frequency data, and angle data;

[0016] The data statistics unit is configured to use the weight data, size data, and length data as the writing brush data, and use the intensity data, frequency data, and angle data as the cleaning data;

[0017] The execution unit is configured to perform feature extraction on the writing brush data and its cleaning data to obtain the cleaning reference value.

[0018] As a further solution of the present invention: the content of performing feature extraction on the writing brush data and its cleaning data to obtain the cleaning reference value includes:

[0019] By substituting into the calculation formula: Obtain the pen tip reference value, where Ab is the weight data, Ac is the size data, Ad is the length data, and A1 is the length coefficient;

[0020] By substituting into the calculation formula: Obtain the cleaning reference value, where Bb is the intensity data, Bc is the frequency data, and Bd is the angle data.

[0021] As a further solution of the present invention: the cleaning duration calculation module includes:

[0022] The image acquisition unit is configured to acquire the image of the cleaning water flow in real time based on the visual sensor built in the device;

[0023] The color value recognition unit is configured to perform color value recognition on the image and record the cleaning duration according to the color value recognition result.

[0024] As a further solution of the present invention: the content of performing color value recognition on the image includes:

[0025] Perform gray-scale conversion on the image, read the gray-scale values of each pixel point in the converted image, and calculate the gray-scale average value of the image;

[0026] Calculate the change rate of the gray-scale average value according to the time stamp of the image;

[0027] When the change rate reaches the preset threshold, mark the corresponding time point;

[0028] Read the image at the time point, and determine the start time and end time according to the gray-scale average value;

[0029] Among them, when the gray-scale average value is less than the preset first threshold, mark the earliest time point as the start time, and when the gray-scale average value is greater than the preset second threshold, mark the latest time point as the end time.

[0030] As a further solution of the present invention: The model training module includes:

[0031] A sample set construction unit, configured to count the cleaning reference values and their cleaning durations of the cleaning tasks for a preset number of times, and construct a sample set;

[0032] A sample set splitting unit, configured to divide the sample set into a training set containing 70% of the data, a test set containing 15% of the data, and a validation set containing 15% of the data;

[0033] A first cluster center selection unit, configured to obtain historical feature vectors in the training set, preset inefficient data clusters L1, normal data clusters L2, and efficient data clusters L3, and select three data points as the first cluster centers in the training set through manual setting, respectively representing the inefficient data cluster L1, the normal data cluster L2, and the efficient data cluster L3;

[0034] By substituting into the calculation formula to obtain the distances between data items, calculate the distances between the data items in the training set and the inefficient data cluster L1, the normal data cluster L2, and the efficient data cluster L3 respectively, and allocate the data items to the data cluster with the closest distance to obtain three new data cluster sets as the second cluster centers, where x and y are the coordinate values of the data points, x i and y i are the values of two data points on i sub-data items respectively, and n is the number of sub-data items;

[0035] Calculate the means of the three new data cluster sets in the second cluster centers respectively, and use them as the new first cluster centers for re-calculation;

[0036] Loop until a preset number of iterations is reached to obtain a neural network model from the cleaning reference value to the cleaning duration.

[0037] As a further solution of the present invention: The feedback data generation module includes:

[0038] A port opening unit, configured to open an information acquisition port when a cleaning request is received;

[0039] A port application unit, configured to obtain the current brush data and its cleaning data based on the information acquisition port;

[0040] An extraction execution unit, configured to perform feature extraction on the obtained current brush data and its cleaning data to obtain the current cleaning reference value;

[0041] A duration output unit, configured to input the current cleaning reference value into the trained neural network model and output the cleaning duration as feedback data.

[0042] As a further solution of the present invention: The system includes:

[0043] An image acquisition unit for real-time acquiring an image of the cleaning water flow based on a visual sensor built in the device;

[0044] A cleaning position acquisition unit for acquiring the cleaning position of the writing brush at each moment based on a pressure sensor built in the fixture;

[0045] A duration segmentation unit for clustering the images at each moment based on the cleaning position and calculating the cleaning duration at different cleaning positions;

[0046] An ink accumulation amount acquisition unit for determining the ink accumulation amount at different cleaning positions based on the cleaning duration and a preset cleaning speed; the cleaning speed is the amount of ink washed per unit time;

[0047] A 3D modeling unit for constructing a 3D model of the writing brush according to the ink accumulation amount at different cleaning positions and feeding it back to the user.

[0048] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention acquires the writing brush data and its cleaning data in multiple cleaning tasks, extracts feature values, synchronously acquires the cleaning duration, constructs a neural network model from the feature values to the cleaning duration, and in actual use, with the help of the constructed neural network model, it can give feedback on the tentative adjustments of the user, with extremely high flexibility, and can assist the user in making an optimal cleaning plan in combination with the actual situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0050] Figure 1 It is a block diagram of the composition structure of the nozzle running trajectory control system of the writing brush cleaning machine. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention clearer, the following further details the present invention with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0052] Figure 1 It is a block diagram of the composition structure of the nozzle running trajectory control system of the writing brush cleaning machine. In an embodiment of the present invention, a nozzle running trajectory control system of a writing brush cleaning machine, the system 10 includes:

[0053] A feature extraction module 11, which is used to obtain brush data and its cleaning data in each cleaning task, extract features from the brush data and its cleaning data to obtain a cleaning reference value; wherein, the brush data includes weight data, length data and dimension data, and the cleaning data includes intensity data, frequency data and angle data;

[0054] A cleaning duration calculation module 12, which is used to obtain the cleaning process of each cleaning task and calculate the cleaning duration according to the cleaning process;

[0055] A model training module 13, which is used to count the cleaning reference values and their cleaning durations of a preset number of cleaning tasks and train a neural network model from the cleaning reference values to the cleaning durations;

[0056] A feedback data generation module 14, which is used to open an information acquisition port when receiving a cleaning request, obtain the current brush data and its cleaning data, extract the cleaning reference values of the current brush data and its cleaning data, and generate feedback data based on the neural network model.

[0057] The cleaning process is continuous. Each time a cleaning request is received, the cleaning device will execute a cleaning task. In each cleaning task, the brush data and its cleaning data are obtained, features are extracted from the brush data and its cleaning data to obtain a cleaning reference value. The brush data is the physical information of the brush itself, including weight data, length data and dimension data, and the cleaning data is the parameters of the device during the cleaning process, including intensity data, frequency data and angle data.

[0058] Further, the cleaning process of each cleaning task is obtained, and the cleaning duration is calculated according to the cleaning process. The cleaning process is the cleaning process, including the nozzle flushing the brush along a preset trajectory. The cleaning duration is the duration consumed by one cleaning task; the cleaning reference values and their cleaning durations of multiple cleaning tasks are counted, and a neural network model from the cleaning reference values to the cleaning durations is trained, wherein the cleaning reference value is used as the input and the cleaning duration is used as the output.

[0059] In actual application, the requester inputs a cleaning request, opens the information acquisition port, and based on the information acquisition port, obtains the current brush data and its cleaning data. In this process, the brush data is fixed, and the cleaning data is input by the requester independently to indicate how to clean. After extracting the cleaning reference values of the current brush data and its cleaning data, the trained neural network model is input, and the cleaning duration is output. The cleaning duration itself reflects the cleaning efficiency. After being fed back to the requester, the requester can adjust the cleaning data independently. Correspondingly, different cleaning durations can also be obtained. After multiple attempts by the requester, a better cleaning duration can be obtained, and the corresponding cleaning data is also used as the final reference for the device, and then converted into a control instruction.

[0060] It is worth mentioning that the working process of the feedback data generation module 14 is actually equivalent to performing a cleaning task, and it will be used as a new sample to recursively update the training process of the neural network model.

[0061] As a preferred embodiment of the technical solution of the present invention, the feature extraction module 11 includes:

[0062] A weight data acquisition unit, configured to monitor the device operation signal in real time. When the device operation signal is monitored, it acquires the weight value of the writing brush through the weight sensing module built in the device as the weight data.

[0063] A dimension data acquisition unit, configured to obtain the length value and the pen barrel circumference value of the writing brush based on the visual sensor built in the device to obtain the dimension data and the length data.

[0064] The above is used to acquire writing brush data, including weight data, dimension data and length data.

[0065] A device parameter acquisition unit, configured to obtain the spray intensity value, the spray frequency value and the spray angle value based on the sensor connected to the nozzle to obtain the intensity data, the frequency data and the angle data; wherein, the signals acquired by the sensor connected to the nozzle all contain time stamps.

[0066] The above is used to acquire cleaning data, including intensity data, frequency data and angle data. It should be explained that the spray intensity is the water spray volume per unit time of the nozzle. For example, the nozzle sprays 0.2 liters of water per minute; the spray frequency is the number of water sprays per unit time of the nozzle. For example, the nozzle sprays 60 times per minute; the spray angle is the included angle between the water spray direction of the nozzle and the axis of the nozzle. For example, the spray angle of the nozzle is 30 degrees.

[0067] A data statistics unit, configured to use the weight data, the dimension data and the length data as the writing brush data, and use the intensity data, the frequency data and the angle data as the cleaning data;

[0068] An execution unit, configured to perform feature extraction on the writing brush data and its cleaning data to obtain a cleaning reference value.

[0069] Finally, the writing brush data and the cleaning data are statistically obtained, and feature extraction is performed on the writing brush data and its cleaning data to obtain a cleaning reference value.

[0070] Specifically, the content of performing feature extraction on the writing brush data and its cleaning data to obtain a cleaning reference value includes:

[0071] By substituting into the calculation formula: The pen tip reference value is obtained, where Ab is the weight data, Ac is the dimension data, Ad is the length data, and A1 is the length coefficient;

[0072] By substituting into the calculation formula: A cleaning reference value is obtained, where Bb is intensity data, Bc is frequency data, and Bd is angle data.

[0073] As a preferred embodiment of the technical solution of the present invention, the cleaning duration calculation module 12 includes:

[0074] An image acquisition unit for real-time acquiring an image of the cleaning water flow based on a visual sensor built in the device;

[0075] A color value recognition unit for performing color value recognition on the image and recording the cleaning duration according to the color value recognition result.

[0076] In an example of the technical solution of the present invention, an image of the cleaning water flow is real-time acquired based on a visual sensor built in the device, color value recognition is performed on the image, and the cleaning duration is recorded according to the color value recognition result; wherein, since the ink has a definite color value, such as black (a small part is other colors), the recognition process is not complicated. Based on the recognition process of the ink color, the precise cleaning start time and cleaning end time can be determined, and thus the accurate cleaning duration can be determined.

[0077] Further, the content of performing color value recognition on the image includes:

[0078] Performing grayscale conversion on the image, reading the grayscale values of each pixel point in the converted image, and calculating the grayscale average value of the image;

[0079] Calculating the change rate of the grayscale average value according to the time stamp of the image;

[0080] When the change rate reaches a preset threshold, marking the corresponding time point;

[0081] Reading the image at the time point and determining the start time and end time according to the grayscale average value;

[0082] Wherein, when the grayscale average value is less than a preset first threshold, the earliest time point is marked as the start time, and when the grayscale average value is greater than a preset second threshold, the latest time point is marked as the end time.

[0083] In an example of the technical solution of the present invention, a specific color value recognition process is provided. The image is subjected to grayscale conversion, the grayscale values of each pixel point in the converted image are read, and the grayscale mean value of the image is calculated. Since each image has a timestamp, the obtained grayscale mean value is also the grayscale mean value at each moment. The difference between the grayscale mean values at adjacent moments is calculated, which is called the change rate. Since this application only needs to locate the start moment and the end moment, the change rate at these moments will be very high. For example, transparent color changes to black, black changes to transparent color, etc. Therefore, only the images corresponding to the moments with a relatively large change rate need to be specifically analyzed to screen out the start moment and the end moment. Specifically, when the grayscale mean value is less than a preset first threshold, it means that the image is dark enough (black when the grayscale value is 0), and the earliest moment point is marked as the start moment. When the grayscale mean value is greater than a preset second threshold, it means that the image is white enough (white when the grayscale value is 255), and the latest moment point is marked as the end moment.

[0084] As a preferred embodiment of the technical solution of the present invention, the model training module 13 includes:

[0085] A sample set construction unit for counting the cleaning reference values and their cleaning durations of the cleaning tasks for a preset number of times and constructing a sample set;

[0086] A sample set splitting unit for splitting the sample set into a training set containing 70% of the data, a test set containing 15% of the data, and a validation set containing 15% of the data;

[0087] A first cluster center selection unit for obtaining the historical feature vectors in the training set, presetting an inefficient data cluster L1, a normal data cluster L2, and an efficient data cluster L3, and selecting three data points as the first cluster centers in the training set through manual setting, respectively representing the inefficient data cluster L1, the normal data cluster L2, and the efficient data cluster L3;

[0088] By substituting into the calculation formula the distances between data items are obtained, the distances between the data items in the training set and the inefficient data cluster L1, the normal data cluster L2, and the efficient data cluster L3 are calculated, and the data items are assigned to the data cluster with the closest distance to obtain three new data cluster sets as the second cluster centers, where x and y are the coordinate values of the data points, x i and y i are the values of two data points on i sub - data items respectively, and n is the number of sub - data items;

[0089] The means of the three new data cluster sets in the second cluster center are calculated respectively as the new first cluster centers for re - calculation;

[0090] The loop is executed until a preset number of iterations is reached to obtain a neural network model from the cleaning reference value to the cleaning duration.

[0091] As a preferred embodiment of the technical solution of the present invention, the feedback data generation module 14 includes:

[0092] A port opening unit, configured to open an information acquisition port when a cleaning request is received;

[0093] A port application unit, configured to acquire current brush data and its cleaning data based on the information acquisition port;

[0094] An extraction execution unit, configured to perform feature extraction on the acquired current brush data and its cleaning data to obtain a current cleaning reference value;

[0095] A duration output unit, configured to input the current cleaning reference value into a trained neural network model and output a cleaning duration as feedback data.

[0096] In an example of the technical solution of the present invention, when a cleaning request is received, an information acquisition port is opened, current brush data and its cleaning data are acquired based on the information acquisition port, feature extraction is performed on the acquired current brush data and its cleaning data to obtain a current cleaning reference value, and the feature extraction scheme adopts the feature extraction scheme already mentioned in the above content. The current cleaning reference value is input into a trained neural network model, and a cleaning duration is output as feedback data.

[0097] As a preferred embodiment of the technical solution of the present invention, the system includes:

[0098] An image acquisition unit, configured to acquire an image of the cleaning water flow in real time based on a visual sensor built in the device;

[0099] A cleaning position acquisition unit, configured to acquire the cleaning position of the brush at each moment based on a pressure sensor built in the fixture;

[0100] A duration segmentation unit, configured to cluster the images at each moment based on the cleaning position and calculate the cleaning duration at different cleaning positions;

[0101] An ink accumulation amount acquisition unit, configured to determine the ink accumulation amount at different cleaning positions based on the cleaning duration and a preset cleaning speed; the cleaning speed is the amount of ink washed per unit duration;

[0102] A three-dimensional modeling unit, configured to construct a three-dimensional model of the brush according to the ink accumulation amount at different cleaning positions and feedback it to the user.

[0103] In an example of the technical solution of the present invention, an additional solution is also provided. Based on the introduction of a visual sensor in this application, during the actual cleaning process, the cleaning position of the writing brush at each moment can be obtained based on a pressure sensor built into the fixture. On this basis, an image of the cleaning water flow is obtained in real time based on the visual sensor built into the device. At this time, the data obtained includes the cleaning position at each time point and the image at each time point. The cleaning position and the image are registered based on time, and then all the images at the same cleaning position are grouped into one category to obtain the images at the same cleaning position at different moments. Finally, the cleaning duration at different cleaning positions is obtained. This process is equivalent to a specific refinement of the original cleaning duration calculation process, and what is obtained is the cleaning duration at each cleaning position.

[0104] For a nozzle, the cleaning speed corresponding to its cleaning data is known and can be obtained, that is, the amount of ink washed per unit time. By calculating the product of the cleaning duration and the cleaning speed, the total amount of ink washed can be obtained as the amount of ink accumulated at the cleaning position; finally, a three-dimensional model of the writing brush is constructed based on the amount of ink accumulated at different cleaning positions and fed back to the user.

[0105] The practical significance of the above solution is that the person cleaning the writing brush is generally the owner of the writing brush. When using the writing brush, he has his own habits. When he knows how much ink is accumulated at different positions when using the writing brush, he can adjust his usage habits during the use process, thereby increasing the service life of the writing brush and making the use process of the writing brush more uniform.

[0106] As a preferred embodiment of the technical solution of the present invention, the cleaning duration reflects the efficiency problem, and there can be some other applications for the cleaning duration, specifically as follows:

[0107] A preset cleaning threshold interval (W1, W2) is set. When the cleaning duration is less than W1, a high-efficiency signal is generated. When the cleaning duration is greater than W1 and less than W2, a normal signal is generated. When the cleaning duration is greater than W2, a low-efficiency signal is generated;

[0108] The high-efficiency signal includes a set of fields representing a relatively high cleaning efficiency of the writing brush. The normal signal includes a set of fields representing a general cleaning efficiency of the writing brush. The low-efficiency signal includes a set of fields representing a relatively low cleaning efficiency;

[0109] The high-efficiency signal, the normal signal, and the low-efficiency signal are packaged to obtain the cleaning classification result.

[0110] On this basis, the cleaning trajectory of the nozzle can be adjusted according to the cleaning classification result. The ways to adjust the cleaning trajectory of the nozzle according to the cleaning classification result include:

[0111] Based on the quick positioning instruction, when the cleaning classification result is an efficient signal, a stop instruction is generated; when the cleaning classification result is a normal signal, a low-speed instruction is generated; when the cleaning classification result is an inefficient signal, a high-speed instruction is generated;

[0112] The stop instruction includes a set of fields representing that the position of the nozzle does not change; the low-speed instruction includes a set of fields representing that the nozzle moves at a slow speed; the high-speed instruction includes a set of fields representing that the nozzle moves at a high speed;

[0113] Package the stop instruction, the low-speed instruction and the high-speed instruction to obtain the cleaning adjustment result.

[0114] Furthermore, the cleaning classification result can also be analyzed. The specific analysis process is as follows:

[0115] The ways to analyze the cleaning classification result include:

[0116] When the cleaning classification result is an efficient signal, a reminder text message is sent to the user through the communication unit; when the cleaning classification result is a normal signal, a text message to be adjusted is sent to the user through the communication unit; when the cleaning classification result is an inefficient signal, a text message urgently needed to be adjusted is sent to the user through the communication unit;

[0117] The reminder text message includes an explanation that the current equipment cleaning efficiency is high, reminding the user to maintain the equipment regularly, replace the cleaning liquid regularly and observe the cleaning effect;

[0118] The text message to be adjusted includes an explanation that the current equipment cleaning efficiency is general, reminding the user to adjust the cleaning data set according to their own needs;

[0119] The text message urgently needed to be adjusted includes an explanation that the current equipment cleaning efficiency is low, requiring the user to adjust the cleaning data set, observe the operation status of the nozzle and select whether to replace the cleaning nozzle according to the operation status;

[0120] Package the reminder text message, the text message to be adjusted and the text message urgently needed to be adjusted to obtain the cleaning decision result.

[0121] Regarding the mechanical structure of the technical solution of the present invention, it generally includes four modules, namely the writing brush cleaning module, the ink tray cleaning module, the water circulation module and the electromechanical module. The specific description is as follows:

[0122] I. Writing brush cleaning module:

[0123] 1. Main components: pen fixing seat, servo motor, gear, rack, push rod, waterproof silicone sleeve, nozzle, water valve, cleaning chamber body and other supporting parts.

[0124] 2. Working principle: This application adopts a reciprocating mechanical structure that simulates the squeezing motion of fingers on the brush hair. During the cleaning process, the most stubborn part is the root of the brush hair because it is the tightest, and the ink that has penetrated is the most difficult to clean out efficiently and quickly. In the hand-washing method, pressing the belly of the brush is the most practical method. Therefore, in the experiment, we immersed the brush hair below the water surface and repeatedly squeezed the brush hair 100 times in 125 mL of clean water by simulating finger squeezing, and then changed the water, which was used as a cleaning sub-process. The sub-process was repeated 4 times, using a total of 500 mL of water. The cleanliness of the brush cleaned by this method is close to that of hand-washing. At the same time, the waterproof silicone sleeve can fully transmit the squeezing force while protecting the brush hair and the pen barrel from the damage of the mechanical mechanism's squeezing force. Such a design not only realizes the upgrade of the cleaning effect but also saves cleaning water.

[0125] 3. Technical solution:

[0126] A. The brush is fixed through a pen fixing seat, and after fixing, the brush hair is located between two waterproof silicone sleeves.

[0127] B. The nozzle injects 125 ml of water into the cleaning chamber, and the water level submerges the brush hair.

[0128] C. The servo motor drives the gear to rotate, driving two racks to move reciprocally towards each other. The push rod on the rack forms a cyclic reciprocating extrusion of the brush hair in the cleaning chamber. After the extrusion is repeated a certain number of times, the ink dissolved in the water in the cleaning chamber reaches the maximum concentration. The motor stops running.

[0129] D. The water valve is opened, and the sewage flows out of the cleaning chamber, and then the water valve is closed.

[0130] The above BCD links are a cleaning sub-process, which is controlled by the circuit system to run. According to the characteristics of the brush and the requirements of the user, the cleaning sub-process is repeated a certain number of times, and the cleaning work of the brush is regarded as completed.

[0131] II. Ink tray cleaning module:

[0132] 1. Main components: Ink tray cleaning chamber body, servo motor, gear, elastic telescopic rod, brush head, bottom suction cup, water injection port, water valve, etc.

[0133] 2. Working principle: This invention is mainly compatible with circular ink trays with a diameter less than 150 mm. The upper end of the elastic telescopic rod is connected to the drive, and the lower end is connected to the brush head. When the ink tray is placed in the cleaning chamber, it can be fixed on the bottom suction cup under the thrust of the brush head, and the brush head rotates driven by the motor, and the brush head rubs against the ink tray to achieve the cleaning of the ink tray.

[0134] 3. Technical solution:

[0135] A. After the ink tray is placed in the cleaning chamber, it falls to the bottom.

[0136] B. Close the lid of the cleaning chamber. The spring in the elastic telescopic rod is compressed, pushing the ink tray to adsorb on the suction cup at the bottom.

[0137] C. The water injection port injects a certain volume of water into the chamber body.

[0138] D. The motor drives the brush head to rotate at a speed of M revolutions per minute. The brush head rubs the ink tray to wash off the ink.

[0139] E. Open the water valve, and the sewage flows out of the cleaning chamber, then close the water valve.

[0140] The above CDE steps form a cleaning sub-process B, which is controlled by the circuit system to operate. The cleaning sub-process is repeated a certain number of times, and it is regarded as the end of the ink tray cleaning work.

[0141] III. Water circulation module:

[0142] 1. Main components: water pump, clean water tank, sewage tank, PVDF ultrafilter, several water pipes.

[0143] 2. Working principle: Ink is composed of two main components, carbon particles and colloid. The working principle of ink is that carbon particles adhere to the colloid and stick to paper or other objects along with the colloid; in this application, to solve the series of problems brought by ink, it is temporarily impossible to start from decomposing carbon particles and colloid. On the one hand, carbon particles are extremely stable and almost impossible to be decomposed and transformed; on the other hand, in the water circulation design of this application, any chemical reagent will eventually come into contact with the pen tip along with the water, causing damage to the pen tip and thus losing the meaning and value of the entire solution. Therefore, we choose a physical method: separate carbon particles and colloid from water. The size of carbon particles is about 50 - 300 nm, and the size of colloid particles is about 10 - 200 nm. Using this physical property, a 10-inch PVDF ultrafilter with a pore size of 10 nm is selected as the filtering tool in the experiment, successfully achieving "black and white separation" with stable effects, completing the most important part of the water circulation module of this application. The separated water can be reused to clean the writing brush and the ink tray.

[0144] 3. Technical solution: The water pump pumps clean water out of the clean water tank. The clean water enters the writing brush cleaning module and the ink tray cleaning module through the water pipe. The sewage generated after the cleaning work (regardless of the specific working sequence) enters the sewage tank through the water pipe. The water pump pumps the sewage into the PVDF ultrafilter. After being filtered by the ultrafilter, carbon particles and most of the colloid are separated in the PVDF ultrafilter. Water and a small amount of colloid enter the clean water tank through the water pipe, waiting for the next cleaning work or to be replaced; after the PVDF ultrafilter filters a certain amount of ink, its filtering ability will reach the limit, and the filter element needs to be replaced before further filtering work can be carried out.

[0145] IV. Electromechanical module:

[0146] 1. Main components: circuit board, motor, water pump, valve, several wires, etc.

[0147] 2. Working principle: Relying on the circuit program to control the brush cleaning module, ink tray cleaning module and water circulation module to work together to complete the post-calligraphy work collaboratively.

[0148] It should be noted that brushes have different materials and size specifications, and at the same time, calligraphy lovers have personalized usage preferences. Therefore, there is no universal cleaning process to complete the brush cleaning work. It is necessary to provide option combinations to adjust the cleaning time T or let the user directly set the time T to flexibly adjust the cleaning process and time in the scenario. The matching schemes of preset times for different specifications need to be written into the embedded program in advance.

[0149] Suppose we preset the cleaning time: the brush hair material determines t1. The cleaning time for a goat hair brush is 5 minutes, for a combined hair brush is 4 minutes, and for a wolf hair brush is 3 minutes; the brush hair size determines t2. The cleaning time for a brush hair with a length of more than 5 cm is 3 minutes, for 3 - 5 cm is 2 minutes, and for less than 3 cm is 1 minute (these data are only for illustrative purposes). In one use, the user selects a goat hair brush with a tip length of 3 - 5 cm. The system presets the cleaning time t1 for the goat hair brush to be 5 minutes, and the cleaning time t2 for a brush with a tip length of 3 - 5 cm to be 2 minutes. Then T = t1 + t2 = 7 minutes.

[0150] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A control system for the running trajectory of a nozzle of a brush cleaning machine, characterized in that: The system comprises: A feature extraction module is used to obtain the writing brush data and its cleaning data in each cleaning task, perform feature extraction on the writing brush data and its cleaning data, and obtain a cleaning reference value; wherein the writing brush data includes weight data, length data and size data, and the cleaning data includes intensity data, frequency data and angle data; A cleaning time calculation module is used to obtain the cleaning process of each cleaning task and calculate the cleaning time according to the cleaning process; A model training module is used to count the cleaning reference values ​​and cleaning durations of cleaning tasks of a preset number of times, and train a neural network model from the cleaning reference values ​​to the cleaning durations; The feedback data generation module is used to open the information acquisition port when receiving a cleaning request, obtain the current brush data and its cleaning data, extract the cleaning reference values ​​of the current brush data and its cleaning data, and then generate feedback data based on the neural network model.

2. The control system for the nozzle running trajectory of a brush cleaning machine according to claim 1 is characterized in that: The feature extraction module comprises: The weight data acquisition unit is used to monitor the device operation signal in real time. When the device operation signal is monitored, the weight value of the brush is collected as weight data through the weight sensing module built into the device; A size data acquisition unit, used to acquire a length value of the brush and a circumference value of the pen shaft based on a visual sensor built into the device, to obtain size data and length data; The equipment parameter acquisition unit is used to acquire the spray intensity value, the spray frequency value and the spray angle value based on the sensor connected to the sprinkler head, and obtain the intensity data, the frequency data and the angle data; A data statistics unit, used for taking weight data, size data and length data as brush data, and taking strength data, frequency data and angle data as cleaning data; The execution unit is used to extract features from the brush data and its cleaning data to obtain a cleaning reference value.

3. The control system for the nozzle running trajectory of the brush cleaning machine according to claim 2 is characterized in that: The feature extraction of the writing brush data and its cleaning data to obtain the cleaning reference value includes: By substituting into the calculation formula: Obtain a pen tip reference value, where Ab is weight data, Ac is size data, Ad is length data, and A1 is length coefficient; By substituting into the calculation formula: A cleaning reference value is obtained, wherein Bb is intensity data, Bc is frequency data, and Bd is angle data.

4. The control system for the nozzle running trajectory of a brush cleaning machine according to claim 1 is characterized in that: The cleaning time calculation module includes: An image acquisition unit, used for acquiring an image of the cleaning water flow in real time based on a visual sensor built into the device; The color value recognition unit is used to perform color value recognition on the image and record the cleaning time according to the color value recognition result.

5. The control system for the nozzle running trajectory of a brush cleaning machine according to claim 4 is characterized in that: The content of performing color value recognition on the image includes: Perform grayscale conversion on the image, read the grayscale value of each pixel in the converted image, and calculate the grayscale mean of the image; Calculate the rate of change of the grayscale mean according to the timestamp of the image; When the change rate reaches a preset threshold, mark the corresponding time point; Read the image at the time point, and determine the start time and end time according to the grayscale mean; When the grayscale mean is less than a preset first threshold, the earliest time point is marked as the start time, and when the grayscale mean is greater than a preset second threshold, the latest time point is marked as the end time.

6. The control system for the nozzle running trajectory of a brush cleaning machine according to claim 1, characterized in that: The model training module includes: A sample set construction unit, used to count the cleaning reference values ​​and cleaning durations of the cleaning tasks of a preset number of times, and to construct a sample set; A sample set splitting unit is used to divide the sample set into a training set containing 70% of the data, a test set containing 15% of the data, and a validation set containing 15% of the data; The first center selection unit is used to obtain the historical feature vector in the training set, preset the inefficient data cluster L1, the normal data cluster L2 and the efficient data cluster L3, and manually select three data points in the training set as the first cluster center, representing the inefficient data cluster L1, the normal data cluster L2 and the efficient data cluster L3 respectively; By substituting into the calculation formula Get the distance between data items, calculate the distance between the data items in the training set and the inefficient data cluster L1, normal data cluster L2 and efficient data cluster L3, and assign the data items to the data clusters with the closest distances, and get three new data clusters as the second cluster centers, where x and y are the coordinate values ​​of the data points, x i and i are the values ​​of two data points on i sub-data items respectively, and n is the number of sub-data items; Calculate the means of the three new data clusters in the second cluster center respectively, and use them as the new first cluster center for recalculation; The loop is executed until the preset number of iterations is reached to obtain a neural network model from the cleaning reference value to the cleaning time.

7. The control system for the nozzle running trajectory of a brush cleaning machine according to claim 1, characterized in that: The feedback data generating module comprises: A port opening unit, used for opening an information acquisition port when receiving a cleaning request; A port application unit, used for acquiring current brush data and cleaning data thereof based on an information acquisition port; An extraction execution unit is used to extract features from the acquired current brush data and its cleaning data to obtain a current cleaning reference value; The duration output unit is used to input the current cleaning reference value into the trained neural network model and output the cleaning duration as feedback data.

8. The control system for the nozzle running trajectory of a brush cleaning machine according to claim 4 is characterized in that: The system comprises: An image acquisition unit, used for acquiring an image of the cleaning water flow in real time based on a visual sensor built into the device; A cleaning position acquisition unit, used to acquire the cleaning position of the brush at each moment based on a pressure sensor built into the fixture; A duration segmentation unit, used to cluster the images at each moment based on the cleaning position, and calculate the cleaning duration of different cleaning positions; An ink accumulation amount acquisition unit, used to determine the ink accumulation amount at different cleaning positions based on the cleaning time and a preset cleaning speed; the cleaning speed is the ink amount per unit time; The three-dimensional modeling unit is used to construct a three-dimensional model of the brush according to the amount of ink collected at different cleaning positions and provide feedback to the user.

Images