An AI-driven fully automated animal-coupled cleaning unit and cleaning method
Through the fully automatic IoT cleaning unit driven by AI, the vessel type is identified and suitable cleaning methods are generated, which solves the problem of single cleaning methods and pure water waste in the prior art, and achieves efficient and water-saving glassware cleaning effects.
Patent Information
- Application Number
- CN202411473084.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-22
- Publication Date
- 2025-02-18
- Estimated Expiration
- 2044-10-22
AI Technical Summary
The existing glassware cleaning machine has a single cleaning method and cannot be properly cleaned according to the specific type of the vessel and the experimental subject, resulting in poor cleaning results and serious waste of pure water.
Using AI-driven fully automatic IoT cleaning unit, the vessel type and layout diagram are identified through the vessel AI recognition unit and visual tracking unit to generate suitable cleaning methods. Combined with cleaning technologies such as ultrasonic, spraying and hot air drying, we can achieve the purpose of different vessels. Efficient cleaning.
It realizes efficient cleaning of glassware, saves pure water, and ensures cleaning effect. It is suitable for a variety of experimental scenarios.
Smart Images

Figure CN118988873B_ABST
Abstract
Description
Technical Field
[0001] The present invention specifically relates to an AI-driven fully automatic animal-coupled cleaning unit and a cleaning method, and belongs to the technical field of utensil cleaning. Background Art
[0002] Glassware is often used as experimental equipment, such as test tubes and beakers. When cleaning glassware, it is generally cleaned inside and outside by a cleaning machine to reduce the contaminants on the outside of the glassware and facilitate the subsequent use of the glassware. The cleaning quality is directly related to the credibility of experimental data. When glassware stores residues, it will affect the experimental results, resulting in inconsistent experimental data deviations. Therefore, it is necessary to ensure the cleaning effect of glassware. The cleaning method of existing glassware cleaning machines is relatively simple and is only suitable for cleaning in a single experimental scenario. For example, China Patent Authorization Announcement No.: CN110340092B discloses a fully automatic glassware cleaning machine that combines an ultrasonic cleaning system and a spray cleaning system into one, and operates fully automatically. However, the cleaning method of the cleaning machine is still relatively simple, and the utensils cannot be cleaned according to the specific type of the utensils and the experimental objects, and the cleaning effect cannot be guaranteed. In addition, pure water is seriously wasted during the cleaning process. Summary of the invention
[0003] To solve the above problems, the present invention proposes an AI-driven fully automatic animal-coupled cleaning unit and cleaning method, which can save more pure water and ensure the cleaning effect of glassware.
[0004] The AI-driven fully automatic animal-coupled cleaning unit of the present invention comprises:
[0005] A cleaning execution unit, the cleaning execution unit comprises a cleaning body, a cleaning cabin is arranged at the bottom of the inner side of the cleaning body, a placement rack inlet and outlet is arranged above the inner side of the cleaning body, an electric lifting unit is arranged at the top of the inner side of the cleaning body, a sealing plate that is movably pressed with the top surface of the cleaning cabin is fixed at the lifting end of the electric lifting unit; a placement rack clamping guide rail is fixed to the bottom surface of the sealing plate; a sealing door is arranged at the placement rack inlet and outlet; after the glassware completes the experiment, the glassware is placed on the placement rack, and then the placement rack is clamped into the placement rack clamping guide rail, and then the sealing door is closed, and the electric lifting unit is driven to be sent into the cleaning cabin, and the top of the cleaning cabin is sealed by the sealing plate; waiting for cleaning; the inner side wall of the cleaning cabin is provided with a horizontal and vertical two-axis slide; the two-axis slide needs to be waterproofed, and the two-axis slide A variety of pressure nozzles are arranged on the slider, and the pressure nozzles are all connected to the switching valve, the switching valve is connected to the cleaning pump, and the cleaning pump is respectively connected to the washing liquid storage tank and the clean water storage tank through a three-way valve; an ultrasonic cleaning unit, a water inlet valve and a drain valve are arranged at the bottom of the cleaning cabin, the water inlet valve is connected to the clean water storage tank; the drain valve is connected to the waste water bucket; a temperature-controlled heater is arranged at the output end of the clean water storage tank; a hot air drying system is arranged at the center of the inner side wall of the cleaning cabin or the bottom surface of the sealing plate; a functional cleaning unit is embedded and fixed on the inner side wall of the cleaning cabin, and the functional cleaning unit includes one or more of a vacuum unit, a pressurizing unit, a gas sterilization unit and an ultraviolet sterilization unit; the vacuum unit, the pressurizing unit, the gas sterilization unit and the ultraviolet sterilization unit can be used for a specific cleaning process or participate in a general cleaning process;
[0006] A cleaning cabin environment collection unit, the cleaning cabin environment collection unit is fixed to the bottom surface of the sealing plate; the cleaning cabin environment collection unit monitors the internal environment of the cleaning cabin in real time, and feeds back the monitoring data to the main control unit in real time, and the main control unit displays the monitoring data on the screen in real time;
[0007] A visual tracking unit is arranged inside the cleaning machine body and directly facing the entrance and exit of the rack. When the sealed door is closed, the visual tracking unit is triggered to track the rack and the glassware on the rack. The visual tracking unit is arranged at multiple angles or equipped with a pan-tilt head to enable a visual probe to collect multiple angles of the rack.
[0008] A main control unit for controlling the entire unit; the inside of the main control unit is provided with a vessel AI recognition unit, an experimental object and experimental category selection unit and a cleaning method generation unit; when the vessel AI tracking unit is working, the placement rack is used as the bearing area, the position of the vessel on the bearing area is obtained, and the type and size of each vessel are identified; the layout diagram of the vessel on the bearing area is output; when the experimental category selection unit is not selected, the normal cleaning method is used by default, and the selection process of the experimental category selection unit includes selecting one, multiple or all vessels according to the layout diagram, and then selecting a certain experimental category through a preset experimental category library; when the main control unit obtains the experimental category in which the vessel participates, the cleaning method is obtained by retrieving the experimental category and cleaning method list; when the cleaning method generation unit generates the cleaning process, it first determines whether the experimental category selection unit is selected. If an object is selected, the obtained cleaning method is first executed, and then the normal cleaning process is entered. If no object is selected, the normal cleaning process is entered.
[0009] Furthermore, the working process of the ordinary cleaning process is as follows: according to the layout diagram on the load-bearing area, the cleaning area and the number of cleanings are determined. When the cleaning area or the number of cleanings exceeds the set value, the water inlet valve is controlled to let water in and water bath ultrasonic cleaning is performed; when the cleaning area and the number of cleanings do not reach the set value, the two-axis slide moves and moves to the corresponding vessel end. The main control unit obtains the cleaning method by retrieving the pre-stored vessel and cleaning method list according to the vessel type and size. The cleaning method includes selecting a certain pressure nozzle, the frequency modulation output power of the cleaning pump and the reciprocating swing spacing of the two-axis slide; the cleaning execution unit executes the cleaning process according to the cleaning method, and the pressure nozzle completes the spraying of detergent, standing and high-temperature hot water rinsing processes; after cleaning is completed, the drain valve executes the drainage process. After drainage is completed, hot air drying is directly performed through the hot air drying system, or secondary sterilization is performed through the gas sterilization unit or the ultraviolet sterilization unit.
[0010] Furthermore, when the vessel AI tracking unit identifies, a picture of the vessels on the display rack is obtained through the visual tracking unit. The vessel AI tracking unit compares the picture of the vessels with the picture of the empty display rack to obtain the number of vessels and the position of the vessels, and directly identifies the vessels, or performs ultrasonic bath washing and spray rinsing selection judgment in advance. When it is determined to be spray rinsing, the vessel identification is performed. When selecting and judging, the projection image of the display rack and the end face of the vessel is obtained through the visual tracking unit, and the total projection area of each vessel is obtained. When the number of vessels and the total projection area are both less than the set value, spray rinsing is determined. When identifying the vessels, the position where the vessels are placed on the vessel placement picture is used as a feature bit, the feature bit is identified through a pre-trained recognition model, and the layout diagram of the vessels on the load-bearing area is output.
[0011] Furthermore, after the feature bits of the vessel placement picture are determined, the layout diagram on the carrying area can also be obtained by feature similarity calculation; specifically: extract the current feature information of the grayscale image of the feature bits, the feature information including the current texture feature and the current spatial position feature; calculate the similarity between the current feature information and the reference feature information; determine the type and size of the glassware according to the similarity, the display rack serves as the carrying area, and the glassware on the display rack serves as the feature area on the carrying area, and calculates the current feature information of the feature area to obtain the similarity between the feature information of the feature area and the pre-stored reference feature information, and determine the type and size of the glassware according to the similarity.
[0012] Furthermore, an inclination adjustment unit is provided on the slider of the two-axis slide, and a spray seat is fixed to the adjustment end of the inclination adjustment unit. The pressure nozzle is fixed on the spray seat. The inclination of the spray seat can be adjusted by the inclination adjustment unit, thereby achieving comprehensive cleaning of the inner and outer walls of the glassware.
[0013] Furthermore, the cleaning cabin environment collection unit includes one or more of a pressure sensor, a temperature sensor, a sterilization gas concentration transmitter, a rack position sensor and a TOC online monitoring unit. The cleaning cabin environment collection unit can collect the temperature, pressure and TOC data inside the cleaning cabin, and judge whether the cleaning execution unit executes the cleaning process and the cleaning effect normally by monitoring the temperature, pressure, TOC data and sterilization gas concentration data.
[0014] An AI-driven fully automated inline cleaning method, using an AI-driven fully automated inline cleaning unit, the method is specifically as follows:
[0015] Put the utensils on the rack, put the utensils to be cleaned after the experiment on the rack, then insert the rack into the rack clamping guide rail and close the sealing door;
[0016] Vessel tracking: After the main control unit receives the sealing door closing signal, it triggers the visual tracking unit, which collects tracking data and sends it to the main control unit for data recognition to obtain the layout of the vessels on the carrying area;
[0017] Cleaning method generation: the main control unit generates a cleaning method through a cleaning method generation unit according to the vessel tracking data and determines whether a trigger and selection instruction of the experimental object and experimental category selection unit is received;
[0018] During the cleaning process, the electric lifting unit drives the display rack downward into the cleaning chamber. At the same time, the sealing plate compresses and seals the top of the cleaning chamber. Then, the cleaning execution unit performs direct ordinary cleaning or specific cleaning followed by ordinary cleaning according to the cleaning method. The specific cleaning method includes one or more of pre-cleaning, vacuum sterilization cleaning, pressurized sterilization cleaning, gas sterilization cleaning and ultraviolet sterilization cleaning of specific vessels. After the cleaning is completed, the electric lifting unit is reset and waits for the display rack to be taken out.
[0019] Compared with the prior art, the AI-driven fully automatic animal-coupled cleaning unit and cleaning method of the present invention are simple to operate. They only require the glassware to be put on the shelf and the experimental object to be selected for the experimental vessel to be cleaned, and the cleaning of the glassware can be automatically completed. During cleaning, visual tracking can be performed inside the cleaning execution unit, and AI recognition of the glassware can be performed to automatically formulate and generate a general cleaning plan. The cleaning plan saves more pure water, and the corresponding cleaning plan can be automatically generated according to the experimental object of the vessel. A specific cleaning plan and a general cleaning plan are combined to ensure the cleaning effect of the glassware. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 It is a schematic diagram of the overall structure of the cleaning execution unit of the present invention.
[0021] Figure 2 It is a schematic diagram of the overall internal structure of the cleaning cabin of the present invention.
[0022] Figure 3 A flow chart is generated for the cleaning method of the present invention.
[0023] Figure 4 It is a schematic diagram of a common cleaning workflow of the present invention.
[0024] Figure 5 It is a schematic diagram of the vessel AI tracking unit identification process of the present invention.
[0025] Figure 6 A schematic flow chart of another embodiment of glassware identification according to the present invention.
[0026] Figure 7 It is a schematic diagram of the working process of the cleaning execution unit of the present invention.
[0027] Figure numerals: 1. Cleaning body, 2. Cleaning cabin, 3. Display rack inlet and outlet, 4. Electric lifting unit, 5. Sealing plate, 6. Display rack clamping guide rail, 7. Sealing door, 8. Display rack, 9. Two-axis slide, 10. Pressure nozzle, 11. Switching valve, 12. Cleaning pump, 13. Three-way valve, 14. Detergent storage tank, 15. Clean water storage tank, 16. Ultrasonic cleaning unit, 17. Water inlet valve, 18. Drain valve, 19. Waste water bucket, 20. Temperature control heater, 21. Hot air drying system, 22. Functional cleaning unit, 23. Cleaning cabin environment collection unit, 24. Visual tracking unit. DETAILED DESCRIPTION
[0028] Example:
[0029] like Figures 1 to 7 The AI-driven fully automatic linked cleaning unit shown includes:
[0030] A cleaning execution unit, the cleaning execution unit comprises a cleaning body 1, a cleaning cabin 2 is arranged at the bottom of the inner side of the cleaning body 1, a rack inlet and outlet 3 is arranged on the upper inner side of the cleaning body 1, an electric lifting unit 4 is arranged at the top of the inner side of the cleaning body 1, and a sealing plate 5 movably pressed against the top surface of the cleaning cabin 2 is fixed at the lifting end of the electric lifting unit 4; a rack clamping guide rail 6 is fixed to the bottom surface of the sealing plate 5; a sealing door 7 is arranged at the rack inlet and outlet 3; after the glassware completes the experiment, the glassware is placed on the rack 8, and then the rack 8 is clamped into the rack clamping guide rail 6, and then the sealing door 7 is closed, and the electric lifting unit 4 is driven to be sent into the cleaning cabin 2, and the top of the cleaning cabin 2 is sealed by the sealing plate 5; waiting for cleaning; the inner side wall of the cleaning cabin 2 is provided with a horizontal and vertical two-axis slide 9; the two-axis slide 9 needs to be waterproofed, and the slider of the two-axis slide 9 is provided with a variety of pressure The pressure nozzles 10 are connected to a switching valve 11, the switching valve 11 is connected to a cleaning pump 12, and the cleaning pump 12 is connected to a washing liquid storage tank 14 and a clean water storage tank 15 respectively through a three-way valve 13; an ultrasonic cleaning unit 16, a water inlet valve 17 and a drain valve 18 are arranged at the bottom of the cleaning cabin 2, the water inlet valve 17 is connected to the clean water storage tank 15; the drain valve 18 is connected to a waste water bucket 19; a temperature control heater 20 is arranged at the output end of the clean water storage tank 15; a hot air drying system 21 is arranged at the center of the inner side wall of the cleaning cabin 2 or the bottom surface of the sealing plate 5; a functional cleaning unit 22 is embedded and fixed on the inner side wall of the cleaning cabin 2, and the functional cleaning unit 22 includes one or more of a vacuum unit, a pressurizing unit, a gas sterilization unit and an ultraviolet sterilization unit; the vacuum unit, the pressurizing unit, the gas sterilization unit and the ultraviolet sterilization unit can be used for a specific cleaning process or participate in a general cleaning process;
[0031] The cleaning cabin environment collection unit 23 is fixed to the bottom surface of the sealing plate 5; the cleaning cabin environment collection unit 23 monitors the internal environment of the cleaning cabin 2 in real time, and feeds back the monitoring data to the main control unit in real time, and the main control unit displays the monitoring data in real time;
[0032] The visual tracking unit 24 is arranged inside the cleaning machine body 1 and is arranged opposite to the inlet and outlet 3 of the display rack 8. When the sealing door 7 is closed, the visual tracking unit 24 is triggered to track the display rack 8 and the glassware on the display rack 8. The visual tracking unit 24 is arranged at multiple angles or equipped with a pan-tilt head to realize that one visual probe collects multiple angles of the display rack 8.
[0033] A main control unit for controlling the entire unit; a vessel AI recognition unit, an experimental object and experimental category selection unit and a cleaning method generation unit are arranged on the inner side of the main control unit, the vessel AI tracking unit includes taking the placement rack 8 as the bearing area, obtaining the position of the vessel on the bearing area, and identifying the type and size of each vessel; outputting the layout diagram of the vessel on the bearing area; when the experimental category selection unit is not selected, the default is the ordinary cleaning method, and the selection process of the experimental category selection unit includes selecting one, multiple or all vessels according to the layout diagram, and then selecting a certain experimental category through a preset experimental category library; when the main control unit obtains the experimental category in which the vessel participates, the cleaning method is obtained by retrieving the experimental category and cleaning method list; when the cleaning method generation unit generates the cleaning process, it first determines whether the experimental category selection unit is selected. If an object is selected, the obtained cleaning method is first executed, and then the ordinary cleaning process is entered. If no object is selected, the ordinary cleaning process is entered.
[0034] The working process of the ordinary cleaning process is as follows: according to the layout diagram on the load-bearing area, the cleaning area and the number of cleanings are determined. When the cleaning area or the number of cleanings exceeds the set value, the water inlet valve 17 is controlled to let water in and water bath ultrasonic cleaning is performed; when the cleaning area and the number of cleanings do not reach the set value, the two-axis slide 9 moves and moves to the corresponding vessel end. The main control unit retrieves the pre-stored vessel and cleaning method list according to the vessel type and size to obtain the cleaning method. The cleaning method includes selecting a certain pressure nozzle 10, the frequency modulation output power of the cleaning pump 12 and the reciprocating swing spacing of the two-axis slide 9; the cleaning execution unit executes the cleaning process according to the cleaning method, and the pressure nozzle 10 completes the spraying of detergent, standing and high-temperature hot water rinsing processes; after cleaning is completed, the drain valve 18 executes the drainage process. After drainage is completed, hot air drying is directly performed through the hot air drying system 21, or secondary sterilization is performed through the gas sterilization unit or the ultraviolet sterilization unit.
[0035] When the vessel AI tracking unit identifies, the vessel placement picture on the display rack 8 is obtained through the visual tracking unit 24. The vessel AI tracking unit compares the vessel placement picture with the empty picture of the display rack 8 to obtain the number of vessels and the placement position of the vessels, and directly identifies the vessels, or performs ultrasonic bath washing and spray rinsing selection judgment in advance. When it is determined to be spray rinsing, the vessel identification is performed. When selecting and judging, the projection image of the display rack 8 and the end face of the vessel is obtained through the visual tracking unit 24, and the total projection area of each vessel is obtained. When the number of vessels and the total projection area are both less than the set value, spray rinsing is determined. When identifying the vessels, the position where the vessels are placed on the vessel placement picture is used as a feature bit, the feature bit is identified by a pre-trained recognition model, and the layout map of the vessels on the load-bearing area is output.
[0036] After the feature bits of the vessel placement picture are determined, the layout diagram on the carrying area can also be obtained by feature similarity calculation; specifically: extract the current feature information of the grayscale image of the feature bits, the feature information including the current texture feature and the current spatial position feature; calculate the similarity between the current feature information and the reference feature information; determine the type and size of the glassware according to the similarity, the placement rack 8 is used as the carrying area, and the glassware on the placement rack 8 is used as the feature area on the carrying area, and the current feature information of the feature area is calculated to obtain the similarity between the feature information of the feature area and the pre-stored reference feature information, and determine the type and size of the glassware according to the similarity.
[0037] The slider of the two-axis slide table 9 is also provided with an inclination adjustment unit, and a spray seat is fixed to the adjustment end of the inclination adjustment unit. The pressure nozzle 10 is fixed on the spray seat. The inclination of the spray seat can be adjusted by the inclination adjustment unit, so that the inner and outer walls of the glassware can be fully cleaned.
[0038] The cleaning cabin environment collection unit 23 includes one or more of a pressure sensor, a temperature sensor, a sterilization gas concentration transmitter, a rack position sensor and a TOC online monitoring unit. The cleaning cabin environment collection unit 23 can collect the temperature, pressure and TOC data inside the cleaning cabin 2, and by monitoring the temperature, pressure, TOC data and sterilization gas concentration data, it is determined whether the cleaning execution unit is executing the cleaning process and the cleaning effect normally.
[0039] An AI-driven fully automated inline cleaning method, using an AI-driven fully automated inline cleaning unit, the method is specifically as follows:
[0040] Put the utensils on the rack, put the utensils to be cleaned after the experiment on the rack 8, then insert the rack 8 into the rack clamping guide rail 6, and close the sealing door 7;
[0041] Vessel tracking: after the main control unit receives the closing signal of the sealing door 7, it triggers the visual tracking unit 24, which collects tracking data and sends it to the main control unit for data recognition, and obtains the layout diagram of the vessels on the carrying area;
[0042] Cleaning method generation: the main control unit generates a cleaning method through a cleaning method generation unit according to the vessel tracking data and determines whether a trigger and selection instruction of the experimental object and experimental category selection unit is received;
[0043] During the cleaning process, the electric lifting unit 4 drives the display rack 8 downward into the cleaning chamber 2. At the same time, the sealing plate 5 compresses and seals the top of the cleaning chamber 2. Then, the cleaning execution unit performs direct ordinary cleaning or specific cleaning followed by ordinary cleaning according to the cleaning method. The specific cleaning method includes one or more of pre-cleaning, vacuum sterilization cleaning, pressurized sterilization cleaning, gas sterilization cleaning and ultraviolet sterilization cleaning of specific vessels. After the cleaning is completed, the electric lifting unit 4 is reset to wait for the display rack 8 to be taken out.
[0044] The above embodiments are only preferred implementations of the present invention, so any equivalent changes or modifications made according to the structures, features and principles described in the scope of application of the present invention are included in the scope of application of the present invention.
Claims
1. An AI-driven fully automated animal-coupled cleaning unit, characterized in that: include A cleaning execution unit, the cleaning execution unit comprises a cleaning body, a cleaning cabin is arranged at the bottom of the inner side of the cleaning body, a placement rack inlet and outlet are arranged above the inner side of the cleaning body, an electric lifting unit is arranged at the top of the inner side of the cleaning body, a sealing plate that is movably pressed with the top surface of the cleaning cabin is fixed to the lifting end of the electric lifting unit; a placement rack clamping guide rail is fixed to the bottom surface of the sealing plate; a sealing door is arranged at the placement rack inlet and outlet; a horizontal and vertical two-axis slide table is arranged on the inner side wall of the cleaning cabin; a variety of pressure nozzles are arranged on the slider of the two-axis slide table, and the pressure nozzles are all connected to the cutting A switching valve is connected to a cleaning pump, and the cleaning pump is respectively connected to a washing liquid storage tank and a clean water storage tank through a three-way valve; an ultrasonic cleaning unit, a water inlet valve and a drain valve are arranged at the bottom of the cleaning cabin, and the water inlet valve is connected to the clean water storage tank; the drain valve is connected to a waste water bucket; a temperature control heater is arranged at the output end of the clean water storage tank; a hot air drying system is arranged at the center of the inner side wall of the cleaning cabin or the bottom surface of the sealing plate; a functional cleaning unit is embedded and fixed on the inner side wall of the cleaning cabin, and the functional cleaning unit includes one or more of a vacuum unit, a pressurizing unit, a gas sterilization unit and an ultraviolet sterilization unit; A cleaning cabin environment collection unit, wherein the cleaning cabin environment collection unit is fixed to the bottom surface of the sealing plate; A visual tracking unit, wherein the visual tracking unit is arranged inside the cleaning machine body and is arranged directly opposite to the entrance and exit of the display rack; A main control unit for controlling the entire unit; a vessel AI recognition unit, an experimental object and experimental category selection unit and a cleaning method generation unit are arranged inside the main control unit. When the vessel AI recognition unit is working, the placement rack is used as the bearing area, the position of the vessel on the bearing area is obtained, and the type and size of each vessel are identified; the layout diagram of the vessel on the bearing area is output; when the experimental object and experimental category selection unit are not selected, the common cleaning method is used by default, and the selection process of the experimental object and experimental category selection unit includes selecting one, multiple or all vessels according to the layout diagram, and then selecting a certain experimental category through a preset experimental category library; when the main control unit obtains the experimental category in which the vessel participates, the cleaning method is obtained by retrieving the experimental category and cleaning method list; when the cleaning method generation unit generates the cleaning process, it first determines whether the experimental object and experimental category selection unit is selected. If an object is selected, the obtained cleaning method is first executed, and then the common cleaning process is entered. If no object is selected, the common cleaning process is entered; The working process of the ordinary cleaning process is as follows: according to the layout diagram on the load-bearing area, the cleaning area and the number of cleanings are determined. When the cleaning area or the number of cleanings exceeds the set value, the water inlet valve is controlled to inlet water to perform water bath ultrasonic cleaning; when the cleaning area and the number of cleanings do not reach the set value, the two-axis slide moves and moves to the corresponding vessel end. The main control unit retrieves the pre-stored vessel and cleaning method list according to the vessel type and size to obtain the cleaning method. The cleaning method includes selecting a certain pressure nozzle, the frequency modulation output power of the cleaning pump and the reciprocating swing spacing of the two-axis slide; the cleaning execution unit executes the cleaning process according to the cleaning method, and the pressure nozzle completes the spraying of detergent, standing and high-temperature hot water flushing processes; after the cleaning is completed, the drain valve executes the drainage process. After the drainage is completed, the hot air drying system is directly used for hot air drying, or the gas sterilization unit or the ultraviolet sterilization unit is used for secondary sterilization; When the vessel AI recognition unit recognizes, the visual tracking unit obtains the vessel placement picture on the display rack, and the vessel AI recognition unit compares the vessel placement picture with the empty picture of the display rack to obtain the number of vessels and the placement position of the vessels, and directly performs vessel recognition, or performs ultrasonic bath washing and spray rinsing selection judgment in advance. When it is determined to be spray rinsing, the vessel recognition is performed. When selecting and judging, the visual tracking unit obtains the projection map of the display rack and the end face of the vessel, and obtains the total projection area of each vessel. When the number of vessels and the total projection area are both less than the set value, spray rinsing is determined. When recognizing the vessel, the position where the vessel is placed on the vessel placement picture is used as a feature bit, and the feature bit is recognized by the recognition model learned in advance, and the layout map of the vessel on the load-bearing area is output; After the feature bits of the container placement image are determined, the layout diagram on the carrying area can be obtained by calculating the feature similarity; specifically, the current feature information of the grayscale image of the feature bits is extracted, and the feature information Including current texture features and current spatial position features; Calculate the similarity between the current feature information and the reference feature information; and determine the type and size of the vessel based on the similarity.
2. The AI-driven fully automated animal-coupled cleaning unit according to claim 1, characterized in that: An inclination adjustment unit is also provided on the slider of the two-axis slide table, a spray seat is fixed to the adjustment end of the inclination adjustment unit, and the pressure nozzle is fixed on the spray seat.
3. The AI-driven fully automated animal-coupled cleaning unit according to claim 1, characterized in that: The cleaning cabin environment acquisition unit includes one or more of a pressure sensor, a temperature sensor, a sterilizing gas concentration transmitter, a rack position sensor and a TOC online monitoring unit.
4. An AI-driven fully automated inline cleaning method, using the AI-driven fully automated inline cleaning unit according to any one of claims 1 to 3, characterized in that: The method is specifically as follows: Put the utensils on the rack, put the utensils to be cleaned after the experiment on the rack, then insert the rack into the rack clamping guide rail and close the sealing door; Vessel tracking: After the main control unit receives the sealing door closing signal, it triggers the visual tracking unit, which collects tracking data and sends it to the main control unit for data recognition to obtain the layout of the vessels on the carrying area; Cleaning method generation: the main control unit generates a cleaning method through a cleaning method generation unit according to the vessel tracking data and determines whether a trigger and selection instruction of the experimental object and experimental category selection unit is received; During the cleaning process, the electric lifting unit drives the display rack downward into the cleaning chamber. At the same time, the sealing plate presses and seals the top of the cleaning chamber. Then, the cleaning execution unit performs direct ordinary cleaning or specific cleaning followed by ordinary cleaning according to the cleaning method. The specific cleaning method includes one or more of pre-cleaning, vacuum sterilization cleaning, pressurized sterilization cleaning, gas sterilization cleaning and ultraviolet sterilization cleaning of specific vessels. After the cleaning is completed, the electric lifting unit resets and waits for the display rack to be taken out.
Citation Information
Patent Citations
A fully automatic glassware washing machine
CN110340092B
Conveyor dishwasher and method for operating conveyor dishwasher
CN117770725A
Portable logging rock debris cleaning device
CN220049155U