Closed-jet cleaning automation control method and system

Abstract

The present application relates to the field of closed jet cleaning and automatic solvent circulation recovery, and discloses a closed jet cleaning automatic control method and system, which comprises dynamic mapping of multi-point pneumatic jetting on the surface of workpieces in a closed cleaning cavity, performing rotary plate roller, tunnel cleaning and web roller linkage operation on the cleaning priority sequence, identifying potential cleaning uneven areas according to the analysis of jetting intensity changes; the potential cleaning uneven areas and the corresponding workpiece surface are calibrated as candidate cleaning units, and the local cleaning efficiency optimization atlas is constructed by integrating the rotary plate roller speed, tunnel transmission node, jetting pressure and web roller parameters; based on the local cleaning efficiency optimization atlas and the global VOCs fluctuation trend, the pneumatic jetting pressure, liquid level control and jetting time weight are dynamically reconstructed, and the automatic control instruction is generated. The present application has the advantages of improving cleaning efficiency and reducing enterprise operating cost while ensuring the occupational health of workers.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of closed jet cleaning and automatic solvent recycling, in particular to a closed jet cleaning automatic control method and system. BACKGROUND

[0002] With the continuous development of printing and painting industries, part cleaning is widely used in the production process to remove ink, paint and other contaminants. However, in some enterprises, due to the complexity of production process, the variety of part shapes and the tediousness of cleaning process, it is very common for parts to be cleaned frequently in a short time. Such high-frequency cleaning not only affects production efficiency, but also increases the pressure of enterprises in terms of labor cost, solvent consumption and environmental management. In the prior art, enterprises usually rely on manual soaking or ultrasonic cleaning, and use semi-open distillation tanks to recover solvents, but these methods have obvious shortcomings: first, manual soaking cleaning is low in efficiency and time-consuming, and workers need to directly contact solvents, so the occupational health cannot be effectively guaranteed; second, ultrasonic cleaning equipment is mostly open box, which has serious solvent waste, unstable cleaning quality, and still needs manual operation, so the safety risk is high; finally, the existing solvent recovery distillation method has great safety hazards and it is difficult to meet environmental standards, and cannot realize efficient and automatic management. Therefore, it is necessary to design a closed jet cleaning automatic control method and system which can improve the cleaning efficiency while guaranteeing the occupational health of workers and reducing the operating cost of enterprises. SUMMARY

[0003] In view of the deficiencies of the prior art, the present application provides a closed jet cleaning automatic control method and system, which has the advantages of improving the cleaning efficiency while guaranteeing the occupational health of workers and reducing the operating cost of enterprises, and solves the problems in the background art.

[0004] To achieve the above purpose of improving the cleaning efficiency while guaranteeing the occupational health of workers and reducing the operating cost of enterprises, the present application provides the following technical scheme: a closed jet cleaning automatic control method and system, comprising the following steps:

[0005] Performing multi-point pneumatic jet dynamic mapping on the surface of the workpiece in the closed cleaning cavity, and generating a cleaning task sequence based on pollution identification and cleaning priority algorithm;

[0006] Performing rotary plate roll, tunnel cleaning and net roll linkage operation on the cleaning priority sequence, and combining pressure sensor, liquid level monitoring and VOCs concentration drift trajectory to construct a multi-dimensional time sequence feature tensor of the workpiece cleaning state;

[0007] Based on the multi-dimensional time sequence feature tensor of the workpiece cleaning state, identifying the potential uneven cleaning area according to the analysis of jet intensity change, liquid level fluctuation anomaly and solvent concentration entropy drift inflection point;

[0008] The potential cleaning uneven area and the corresponding workpiece surface are calibrated as a candidate cleaning unit, and the rotating plate roller speed, tunnel transmission node, spray pressure and mesh roller parameters are fused to construct a local cleaning efficiency optimization atlas;

[0009] Based on the local cleaning efficiency optimization atlas and the global VOCs fluctuation trend, the pneumatic spray pressure, liquid level control and spray time weight are dynamically reconstructed, and the automatic control instruction is generated.

[0010] Preferably, the generation of the cleaning task sequence process is:

[0011] Image recognition or sensor data analysis is used to obtain workpiece surface contamination distribution information;

[0012] According to the pollution type, pollution degree and workpiece material, the cleaning priority score is calculated;

[0013] A hierarchical cleaning task list is generated, and the cleaning task sequence is constructed in priority order;

[0014] The high-priority task is labeled and assigned a spray path and time window in real time.

[0015] Preferably, the rotating plate roller, tunnel cleaning and mesh roller linkage operation process for the cleaning priority sequence is:

[0016] Based on the cleaning task sequence, the rotating plate roller speed and direction are dynamically adjusted;

[0017] The tunnel transmission device is controlled to realize continuous movement and positioning of the workpiece;

[0018] The mesh roller linkage cooperates with the rotating plate roller and the tunnel transmission.

[0019] Preferably, the process of constructing a multi-dimensional time sequence feature tensor of the workpiece cleaning state is:

[0020] While the rotating plate roller, tunnel transmission and mesh roller linkage execute the cleaning task, real-time acquisition of cavity pressure, liquid level and solvent concentration data is performed;

[0021] The collected data is mapped to a multi-dimensional feature space according to the time sequence and workpiece spatial position;

[0022] Fusion of spray intensity, spray time, cleaning path and mechanical action state information forms a multi-dimensional time sequence feature tensor of the workpiece cleaning state.

[0023] Preferably, the process of identifying potential cleaning uneven areas is:

[0024] The multi-dimensional time sequence feature tensor of the cleaning state is mapped to a time sequence analysis model;

[0025] Monitoring the spray intensity fluctuation, liquid level anomaly and solvent concentration entropy drift inflection point;

[0026] Based on statistical and threshold analysis method, identify potential cleaning uneven area.

[0027] Preferably, the potential cleaning uneven area and the corresponding workpiece surface are marked as candidate cleaning unit process:

[0028] Based on the potential cleaning uneven area and the corresponding workpiece position;

[0029] The dynamic cleaning efficiency evolution path is compared with the standard cleaning reference trajectory in topology;

[0030] Identify the spray coverage sequence deviation and local cleaning abnormal area;

[0031] Comprehensive analysis combined with spray intensity, liquid level fluctuation and solvent concentration change;

[0032] The determined abnormal area forms a candidate cleaning unit.

[0033] Preferably, the process of constructing a local cleaning efficiency optimization map is:

[0034] Extract the corresponding rotary roll speed, tunnel transmission node, spray pressure, web roll parameter process data of the candidate cleaning unit;

[0035] Analyze the cleaning coverage dependency relationship between each candidate unit and the process parameter influence range, and divide the workpiece surface into local optimization area;

[0036] Based on the local process parameters and cleaning efficiency indicators, the cleaning efficiency optimization level is generated, and the local cleaning efficiency optimization map is formed.

[0037] Preferably, the process of dynamically reconstructing the pneumatic spray pressure, liquid level control and spray time weight is:

[0038] Continuously collect the spray pressure, liquid level, cleaning efficiency and VOCs concentration data of each candidate cleaning unit in the actual cleaning process;

[0039] Compare the collected data with the local cleaning efficiency optimization map to identify the area with local cleaning efficiency decline or coverage anomaly;

[0040] Based on the global VOCs concentration fluctuation and workpiece surface pollution distribution trend, analyze the deviation between local and overall cleaning uniformity;

[0041] Real-time dynamic adjustment of the spray pressure, liquid level control and spray time weight of the candidate cleaning unit.

[0042] Preferably, the process of generating an automatic control instruction is:

[0043] mapping the dynamically reconstructed jetting pressure, liquid level control, and jetting time weight to the instruction template of the automation control system;

[0044] According to the local cleaning efficiency optimization map and the global VOCs fluctuation trend, the priority execution order and jetting path of each candidate cleaning unit are distributed;

[0045] The instructions are issued to the rotary plate roller, tunnel transmission, and web roller control unit;

[0046] Access real-time monitoring data stream, close-loop feedback on control instruction execution effect, dynamically adjust subsequent jetting pressure, liquid level, and jetting time weight;

[0047] Output complete automation control execution record and workpiece cleaning state data.

[0048] The closed jet flow cleaning automation control system comprises:

[0049] The jetting mapping module: multi-point pneumatic jetting scanning of the workpiece surface and generation of a cleaning task sequence;

[0050] The state modeling module: fusion of multi-source sensor data to construct a multi-dimensional time series feature tensor of the workpiece cleaning state;

[0051] The area identification module: analysis of jetting intensity, liquid level, and solvent fluctuation to detect potential cleaning uneven areas;

[0052] The efficiency map module: fusion of candidate units and equipment parameters to generate a local cleaning efficiency optimization map;

[0053] The dynamic control module: dynamically adjusting jetting pressure, liquid level, and time based on the optimization map and VOCs trend and issuing control instructions.

[0054] Compared with the prior art, the present application provides a closed jet flow cleaning automation control method and system, which has the following beneficial effects:

[0055] The present application realizes accurate identification of workpiece surface contamination and task serialization distribution by introducing multi-point pneumatic jet dynamic mapping and cleaning priority algorithm in the closed cleaning cavity; combines rotary version roller, tunnel cleaning and net roller linkage operation, and uses real-time monitoring data of pressure, liquid level and VOCs concentration to construct multi-dimensional time sequence feature tensor, effectively capturing dynamic changes in the cleaning process; further identifies potential cleaning uneven areas through feature analysis, and generates local efficiency optimization atlas based on candidate cleaning units and key equipment parameters, so as to realize adaptive adjustment of jet pressure, liquid level control and jet time; the technology not only improves the intelligent level and accuracy of the cleaning process, ensures the stability and consistency of the cleaning quality, but also reduces resource waste and energy consumption, and meets environmental protection and safety requirements, and finally can improve the cleaning efficiency and reduce the operating cost of enterprises while ensuring the occupational health of workers. BRIEF DESCRIPTION OF DRAWINGS

[0056] Figure 1 A schematic diagram of the method of the present application is shown;

[0057] Figure 2 A schematic diagram of the system of the present application is shown. DETAILED DESCRIPTION

[0058] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0059] Embodiment 1: please refer to Figure 1 The closed jet cleaning automatic control method described in the embodiments of the present application includes the following steps:

[0060] S1: multi-point pneumatic jet dynamic mapping is performed on the workpiece surface in the closed cleaning cavity, and a cleaning task sequence is generated based on contamination identification and cleaning priority algorithm;

[0061] The process of generating the cleaning task sequence in S1 is:

[0062] Image recognition or sensor data analysis is used to obtain workpiece surface contamination distribution information;

[0063] The image data is collected by a workpiece surface imaging device or a multi-point sensor array installed in the cleaning cavity, and the image data is pre-processed, including light equalization, edge enhancement and noise removal; the sensor signals are normalized and abnormal point marked, and then the spatial distribution characteristics of the contaminated area are extracted by using image recognition algorithm or sensor data analysis method, and the contaminated distribution matrix is output, which is indexed by the workpiece surface coordinates and attached with the contamination density parameters.

[0064] According to the contamination type, contamination degree and workpiece material, the cleaning priority score is calculated;

[0065] The obtained contaminated distribution matrix is analyzed to extract information such as contamination type, contamination coverage area and contamination thickness, and combined with the surface characteristic parameters in the workpiece material database, a priority score model is constructed, which scores the contaminated area based on a weighted formula or a multi-factor decision tree, outputs a numerical cleaning priority index, and stores the result as a structured data table.

[0066] A hierarchical cleaning task list is generated, and a cleaning task sequence is constructed according to the priority order;

[0067] According to the cleaning priority index, the contaminated area is sorted according to the score, and a hierarchical cleaning task list is generated, the task list adopts a chain structure record, including task number, corresponding surface coordinate range, priority level and estimated cleaning time, the system combines the sorting results to form a cleaning task sequence, and retains the dependency relationship between tasks in the sequence, so that the tasks can be expanded in priority order.

[0068] Real-time labeling and allocation of spray path and time window for high-priority tasks;

[0069] In the task sequence, the cleaning tasks in high-priority level are labeled in real time, and the corresponding execution label is generated in the control system, the labeling information includes task coordinates, contamination type, cleaning parameter range and available spray device number, the system automatically allocates spray path and time window combined with the spatial position of spray unit and available time slot, so that high-priority tasks can be accurately scheduled and executed within the scheduling period.

[0070] S2: Perform rotary plate roller, tunnel cleaning and web roller linkage operation on the cleaning priority sequence, and construct a multi-dimensional time sequence feature tensor of workpiece cleaning state combined with pressure sensor, liquid level monitoring and VOCs concentration drift trajectory;

[0071] The process of performing rotary plate roller, tunnel cleaning and web roller linkage operation on the cleaning priority sequence in S2 is:

[0072] Based on the cleaning task sequence, the speed and direction of the rotary plate roller are dynamically adjusted;

[0073] According to the priority and position coordinates of each task in the cleaning task sequence, the motion parameters of the rotating plate roller, including the rotating speed, rotating direction and acceleration / deceleration curve, are calculated in real time, and the controller sends PWM signals or servo control signals through the motor drive module to realize dynamic adjustment of the plate roller, so that the workpiece surface relative to the spraying device maintains the optimal cleaning angle and coverage range, and the real-time position and speed data of the plate roller are recorded for subsequent task synchronization.

[0074] The control tunnel transmission device realizes the continuous movement and positioning of the workpiece;

[0075] The tunnel transmission system receives the position and time information in the cleaning task sequence, controls the transmission device to realize the continuous translation of the workpiece in the cleaning cavity, and drives the chain / rail through the stepping motor or servo motor to realize the movement and precise positioning of the workpiece. The residence time and movement speed of each workpiece are matched with the spraying time window of the cleaning task to ensure that the task sequence is executed in order according to the priority.

[0076] The net roller linkage cooperates with the rotating plate roller and the tunnel transmission;

[0077] The net roller system is linked and controlled with the rotating plate roller and the tunnel transmission to realize auxiliary cleaning of special parts or edge areas of the workpiece surface. The controller adjusts the rotating speed, rotating direction and synchronization timing of the net roller according to the region coordinates and cleaning requirements marked in the task sequence, so that the movement of the net roller is coordinated with that of the rotating plate roller and the tunnel transmission, and the angle and time of contact between the spraying device and the workpiece surface during cleaning meet the distributed task parameters.

[0078] The process of constructing a multi-dimensional time sequence feature tensor of the workpiece cleaning state in S2 is:

[0079] While the rotating plate roller, tunnel transmission and net roller are linked to execute the cleaning task, the pressure, liquid level and solvent concentration data in the cavity are collected in real time;

[0080] While the rotating plate roller, tunnel transmission and net roller are linked to execute the cleaning task, the system collects data in real time through the pressure sensor, liquid level sensor and solvent concentration sensor arranged in the closed cleaning cavity. The pressure sensor records the change of the spraying pressure with a sampling period of 100 ms, the liquid level sensor collects the liquid level with a sampling period of 1 s, and the solvent concentration sensor continuously collects the VOCs concentration with a precision of 0.1 ppm. All sensor data are time-stamped and corresponding to the workpiece position coordinates, which are used for subsequent feature mapping.

[0081] The collected data are mapped to a multi-dimensional feature space according to the time sequence and the spatial position of the workpiece;

[0082] The collected pressure, liquid level and solvent concentration data are mapped to a multi-dimensional feature space according to time series and workpiece surface spatial position. The system takes different regions of each workpiece surface as an index to construct a time-space coordinate matrix, associates the sensor signals with the corresponding workpiece position, and forms a preliminary multi-dimensional data structure, providing a basis for subsequent feature fusion and tensor construction.

[0083] Fusion of spray intensity, spray time, cleaning path and mechanical action state information forms a multi-dimensional time sequence feature tensor of workpiece cleaning state;

[0084] On the basis of the preliminary feature matrix, further fusion of spray intensity, spray time, cleaning path and mechanical action state information, including rotating plate roller speed, tunnel transmission speed and web roller motion state, is carried out. Through combination and coding of data at each time step, a structured multi-dimensional time sequence feature tensor is formed, the dimensions of which include time, spatial position, spray parameter and mechanical action state, for complete description of the dynamic state of the workpiece in the cleaning process.

[0085] S3: Based on the multi-dimensional time sequence feature tensor of workpiece cleaning state, potential cleaning uneven areas are identified according to analysis of spray intensity variation, liquid level fluctuation anomaly and solvent concentration entropy drift inflection point;

[0086] The process of identifying potential cleaning uneven areas in S3 is as follows:

[0087] Map the multi-dimensional time sequence feature tensor of cleaning state to a time series analysis model;

[0088] Input the constructed multi-dimensional time sequence feature tensor of workpiece cleaning state into the time series analysis model, and expand the data of each workpiece surface area according to time step sequence. Each time step contains features such as pressure, liquid level, solvent concentration, spray intensity and mechanical action state. By decomposing the multi-dimensional tensor into time series form, structured data input is provided for subsequent dynamic fluctuation monitoring and anomaly detection.

[0089] Monitor spray intensity fluctuation, liquid level anomaly and solvent concentration entropy drift inflection point;

[0090] In the time series model, for each workpiece surface area, the change amplitude of spray intensity, the fluctuation amplitude of liquid level and the change trend of solvent concentration entropy are calculated in real time. Pressure fluctuation and liquid level anomaly are calculated by sliding window algorithm to obtain mean, standard deviation and threshold boundary. The solvent concentration entropy drift is calculated by information entropy change rate to obtain the time inflection point. The monitoring results of each time step are recorded with the corresponding spatial position to associate the surface area.

[0091] Based on statistical and threshold analysis methods, potential cleaning uneven areas are identified;

[0092] According to the monitoring results, a statistical analysis and threshold judgment method is used to mark the surface area that exceeds the set fluctuation range or has an abnormal inflection point. The statistical method includes mean deviation, standard deviation exceeding the threshold, and local entropy change mutation detection. The system records the identified area as a potential cleaning uneven unit and stores its spatial coordinates and time period information, providing input for subsequent local efficiency optimization and dynamic control.

[0093] S4: The potential cleaning uneven area and the corresponding workpiece surface are marked as candidate cleaning units, and the rotating plate roller speed, tunnel transmission node, spray pressure, and mesh roller parameters are fused to construct a local cleaning efficiency optimization map.

[0094] The process of marking the potential cleaning uneven area and the corresponding workpiece surface as candidate cleaning units in S4 is as follows:

[0095] Based on the potential cleaning uneven area and the corresponding workpiece position;

[0096] According to the identification of the potential uneven area, the potential cleaning uneven area identified is matched with the spatial coordinates of the workpiece surface. The system establishes an index relationship between the time series characteristics of each area and the corresponding workpiece surface grid coordinates, providing accurate spatial positioning information for subsequent analysis.

[0097] Topological comparison between the dynamic cleaning efficiency evolution path and the standard cleaning reference trajectory;

[0098] Topological comparison between the dynamic cleaning efficiency evolution path recorded in the actual cleaning process and the pre-set standard cleaning reference trajectory. This process includes time synchronization matching of the spray path, liquid level change curve, and mechanical action sequence, and calculating the deviation matrix between the actual path and the reference trajectory to quantify the local cleaning sequence and coverage difference.

[0099] Identify the spray coverage sequence deviation and local cleaning abnormal area;

[0100] Through analysis of the deviation matrix, the system calculates the difference index between the actual spray sequence and the standard sequence for each area, including time delay, insufficient or repeated coverage, and marks the area that exceeds the set threshold as an abnormal area.

[0101] Comprehensive analysis combined with spray intensity, liquid level fluctuation, and solvent concentration change;

[0102] For the marked abnormal area, comprehensive analysis is conducted on multiple features such as spray intensity, liquid level fluctuation, and solvent concentration change. Through weighted calculation or multi-factor scoring, the cleaning state of each area is quantified to form a cleaning abnormality index, which is used to determine the final candidate cleaning unit.

[0103] Forming a candidate cleaning unit from the determined abnormal area;

[0104] Forming a candidate cleaning unit data structure from the determined abnormal area according to spatial position and time period, including workpiece surface coordinates, task number, priority index and related cleaning parameters, and storing the candidate cleaning unit into a task scheduling database to provide input for subsequent local efficiency optimization and dynamic control.

[0105] The process of constructing a local cleaning efficiency optimization map in S4 is as follows:

[0106] Extracting the rotary plate roller speed, tunnel transmission node, spraying pressure, and web roller parameter process data corresponding to the candidate cleaning unit;

[0107] The system indexes the candidate cleaning unit generated in the previous step and extracts its corresponding rotary plate roller speed, tunnel transmission node state, spraying pressure curve, and web roller parameter process data. The data includes real-time recorded mechanical action state, execution time sequence, and spraying device operation parameters. These information is organized as a structured data set associated with the spatial coordinates of each candidate unit, providing a basis for subsequent analysis.

[0108] Analyzing the cleaning coverage dependency relationship between each candidate unit and the process parameter influence range to divide the workpiece surface into local optimization areas;

[0109] Analyzing the cleaning coverage dependency relationship between each candidate cleaning unit, including spraying path overlap, mechanical motion coordination, and liquid coverage range, combining the influence radius of process parameters to divide the workpiece surface into several local optimization areas, each area containing mutually influencing candidate cleaning units, and recording the process parameter configuration and cleaning coverage information within the area for local optimization calculation.

[0110] Based on the local process parameters and cleaning efficiency index, a cleaning efficiency optimization level is generated, and a local cleaning efficiency optimization map is formed;

[0111] For each local optimization area, according to the local process parameters and cleaning efficiency index, a quantitative score is made, and the score results form a cleaning efficiency level for identifying the relative difference in cleaning effect. The system integrates all local optimization areas and their corresponding efficiency levels to generate a local cleaning efficiency optimization map, using the workpiece surface spatial coordinates as an index to record the process parameters and optimization level of each area, providing input for dynamic control.

[0112] S5: Based on the local cleaning efficiency optimization map and the global VOCs fluctuation trend, dynamically reconstructing the pneumatic spraying pressure, liquid level control and spraying time weight, and generating automatic control instructions.

[0113] The dynamic reconfiguration of the pneumatic injection pressure, liquid level control and injection time weight process in S5 is:

[0114] The injection pressure, liquid level, cleaning efficiency and VOCs concentration data of each candidate cleaning unit during the actual cleaning process are continuously collected;

[0115] During the cleaning process, the system continuously collects injection pressure, liquid level, cleaning efficiency and VOCs concentration data of each candidate cleaning unit. The pressure sensor records the pressure change of the injection device with 100ms sampling, the liquid level sensor monitors the liquid height change with 1s sampling, the cleaning efficiency is calculated by injection coverage and solvent distribution uniformity, and the VOCs concentration is continuously collected by the gas sensor. All data have time stamp and spatial coordinate information, which provides synchronous reference for subsequent analysis.

[0116] The collected data are compared with the local cleaning efficiency optimization map to identify the areas with decreased local cleaning efficiency or abnormal coverage;

[0117] The real-time collected data are compared with the local cleaning efficiency optimization map generated in the previous step. The system calculates the deviation of the actual injection coverage from the expected coverage of the map for each candidate cleaning unit, identifies the areas with decreased cleaning efficiency or abnormal local coverage, and records the abnormal information according to the spatial coordinates and time steps to form a local cleaning state monitoring matrix.

[0118] Based on the global VOCs concentration fluctuation and the workpiece surface contamination distribution trend, the deviation between local and overall cleaning uniformity is analyzed;

[0119] Based on the global VOCs concentration fluctuation and the workpiece surface contamination distribution trend, the deviation between local and overall cleaning uniformity is analyzed;

[0120] The injection pressure, liquid level control and injection time weight of the candidate cleaning unit are dynamically adjusted in real time;

[0121] According to the above analysis results, the injection pressure, liquid level control and injection time weight of the candidate cleaning unit are dynamically adjusted in real time. The controller adjusts the injection device output pressure range, liquid level pump start-stop and continuous injection time by issuing instructions, so that each candidate unit updates the operation according to the optimized parameters during the execution process, realizes real-time correction and uniformity adjustment of cleaning coverage.

[0122] The automatic control instruction generation process in S5 is:

[0123] The dynamically reconstructed jetting pressure, liquid level control and jetting time weight are mapped to the instruction template of the automation control system;

[0124] The dynamically reconstructed jetting pressure, liquid level control and jetting time weight data are mapped to the instruction template of the automation control system, and the template includes parameters such as workpiece surface coordinates, jetting device number, pressure range, liquid level height and jetting duration, each instruction is provided with a time stamp and a task number, so that the control system can parse and execute specific operations.

[0125] According to the local cleaning efficiency optimization map and the global VOCs fluctuation trend, the priority execution order and the jetting path of each candidate cleaning unit are distributed;

[0126] According to the local cleaning efficiency optimization map and the global VOCs concentration fluctuation trend, the system distributes the priority execution order and the jetting path of each candidate cleaning unit, and the order distribution considers the local cleaning efficiency grade, the adjacent unit coverage dependency and the workpiece surface space position, and the path planning includes the rotation plate roller angle, the tunnel transmission displacement and the web roller movement direction, to ensure that the task is completed in sequence within the execution cycle.

[0127] The instructions are issued to the rotation plate roller, tunnel transmission and web roller control units;

[0128] The control system issues the generated automation control instructions to the rotation plate roller, tunnel transmission and web roller control units, the instructions are transmitted through the industrial bus or control network, and the control units adjust the motor drive signal, pump start-stop and jetting valve opening and closing according to the instructions, to ensure that each candidate cleaning unit performs cleaning operation according to the preset parameters, while recording the response state of each instruction.

[0129] Access real-time monitoring data stream, close-loop feedback control instruction execution effect, dynamically adjust subsequent jetting pressure, liquid level and jetting time weight;

[0130] The system accesses real-time monitoring data stream, including pressure, liquid level, jetting coverage and VOCs concentration, etc., close-loop feedback control instruction execution effect, dynamically adjusts subsequent jetting pressure, liquid level and jetting time weight according to the deviation between monitoring data and preset target, to realize real-time optimization of continuous cleaning process.

[0131] Output complete automation control execution record and workpiece cleaning state data;

[0132] The system stores the complete execution record of automation control and the workpiece cleaning state data in the database. The record includes the issuing time, execution state, actual mechanical action parameters and cleaning feature tensor of the corresponding workpiece surface area of each instruction, which provides basic data support for subsequent task analysis, process optimization and data tracing.

[0133] Embodiment 2: As shown in Figure 2 The closed-jet cleaning automation control system comprises:

[0134] Spray mapping module: multi-point pneumatic spray scanning on workpiece surface and generating cleaning task sequence;

[0135] State modeling module: fusion of multi-source sensor data to build multi-dimensional time-series feature tensor of workpiece cleaning state;

[0136] Region identification module: analysis of spray intensity, liquid level and solvent fluctuation to detect potential cleaning non-uniform regions;

[0137] Efficiency atlas module: fusion of candidate cells and equipment parameters to generate local cleaning efficiency optimization atlas;

[0138] Dynamic control module: dynamic adjustment of spray pressure, liquid level and time based on optimization atlas and VOCs trend and issuing control instructions.

[0139] It should be noted that the relational terms herein such as first and second and the like are used solely to distinguish one entity or action from another, without necessarily requiring or implying any such actual relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0140] Although embodiments of the present application have been shown and described, it is to be understood that various modifications, substitutions, replacements and variations can be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A method for closed-jet washing automation control, characterized in that, The method comprises the following steps: Multi-point pneumatic jet dynamic mapping is performed on the surface of the workpiece in the closed cleaning cavity, and a cleaning task sequence is generated based on contamination identification and cleaning priority algorithm; Rotary plate roller, tunnel cleaning and web roller linkage operation are performed on the cleaning priority sequence, and a multi-dimensional time sequence feature tensor of the workpiece cleaning state is constructed in combination with pressure sensor, liquid level monitoring and VOCs concentration drift trajectory; Based on the multi-dimensional time sequence feature tensor of the workpiece cleaning state, potential uneven cleaning areas are identified according to analysis of jet intensity change, liquid level fluctuation anomaly and solvent concentration entropy drift inflection point; The process of identifying potential uneven cleaning areas is: Mapping the multi-dimensional time sequence feature tensor of the cleaning state to a time sequence analysis model; Monitoring jet intensity fluctuation, liquid level anomaly and solvent concentration entropy drift inflection point; Based on statistical and threshold analysis method, identifying potential uneven cleaning areas; The potential uneven cleaning areas and the corresponding workpiece surface are calibrated as candidate cleaning units, and the local cleaning efficiency optimization atlas is constructed by integrating the rotary plate roller speed, tunnel transmission node, jet pressure and web roller parameters; The process of calibrating the potential uneven cleaning areas and the corresponding workpiece surface as candidate cleaning units is: Based on the potential uneven cleaning areas and the corresponding workpiece position; Topologically comparing the dynamic cleaning efficiency evolution path with the standard cleaning reference trajectory; Identifying jet coverage sequence deviation and local cleaning abnormal area; Comprehensive analysis is conducted in combination with jet intensity, liquid level fluctuation and solvent concentration change; The determined abnormal area forms a candidate cleaning unit; Based on the local cleaning efficiency optimization atlas and the global VOCs fluctuation trend, the pneumatic jet pressure, liquid level control and jet time weight are dynamically reconstructed, and the automatic control instruction is generated.

2. The closed loop automated control method of a fluidic wash as claimed in claim 1, wherein, The process of generating the cleaning task sequence is: Obtaining the workpiece surface contamination distribution information by image recognition or sensor data analysis; According to the contamination type, contamination degree and workpiece material, the cleaning priority score is calculated; A hierarchical cleaning task list is generated, and the cleaning task sequence is constructed in priority order; Real-time labeling and assigning jet path and time window to high-priority tasks.

3. The closed loop automated control method of a fluidic wash as claimed in claim 2, wherein, The process of performing rotary plate roller, tunnel cleaning and web roller linkage operation on the cleaning priority sequence is: Based on the cleaning task sequence, dynamically adjusting the speed and direction of the rotary plate roller; Controlling the tunnel transmission device to realize continuous movement and positioning of the workpiece; Web roller linkage cooperates with rotary plate roller and tunnel transmission.

4. The closed loop automated control method of a fluidic wash as claimed in claim 3, wherein, The process of constructing the multi-dimensional time sequence feature tensor of the workpiece cleaning state is: While the rotary plate roller, tunnel transmission and web roller linkage are performing the cleaning task, real-time acquisition of pressure, liquid level and solvent concentration data in the cavity is performed; The collected data is mapped to a multi-dimensional feature space according to time sequence and workpiece spatial position; Fusion of jet intensity, jet time, cleaning path and mechanical action state information forms the multi-dimensional time sequence feature tensor of the workpiece cleaning state.

5. The closed loop automated control method of a fluidic wash as claimed in claim 1, wherein, The process of constructing the local cleaning efficiency optimization atlas is: Extracting the rotary plate roller speed, tunnel transmission node, jet pressure and web roller parameter process data corresponding to the candidate cleaning unit; Analyzing the cleaning coverage dependency relationship between each candidate unit and the process parameter influence range, and dividing the workpiece surface into local optimization areas; Based on the local process parameters and cleaning efficiency indicators, a quantitative score is generated to form a cleaning efficiency optimization grade and a local cleaning efficiency optimization map.

6. The closed-loop jet cleaning automated control method according to claim 5, characterized in that, The dynamic reconstruction of the pneumatic injection pressure, liquid level control, and injection time weight process is: Continuous acquisition of injection pressure, liquid level, cleaning efficiency, and VOCs concentration data during actual cleaning process of each candidate cleaning unit; Comparing the collected data with the local cleaning efficiency optimization map to identify areas of local cleaning efficiency decline or coverage anomalies; Based on the global VOCs concentration fluctuation and workpiece surface contamination distribution trend, analyze the deviation between local and overall cleaning uniformity; Real-time dynamic adjustment of injection pressure, liquid level control, and injection time weight of candidate cleaning units.

7. The closed loop automated control method of a fluidic wash as claimed in claim 6, wherein, The process of generating automatic control instructions is: Map the dynamically reconstructed injection pressure, liquid level control, and injection time weight to the instruction template of the automatic control system; According to the local cleaning efficiency optimization map and the global VOCs fluctuation trend, assign the priority execution order and injection path of each candidate cleaning unit; Issue instructions to the rotary plate roller, tunnel transmission, and web roller control unit; Access real-time monitoring data stream to perform closed-loop feedback on control instruction execution effect, dynamically adjust subsequent injection pressure, liquid level, and injection time weight; Output complete automatic control execution records and workpiece cleaning state data.

8. Closed-jet washing automation control system, applied to the method according to any one of claims 1 to 7, characterized in that, It includes: Injection mapping module: multi-point pneumatic injection scanning of workpiece surface and generation of cleaning task sequence; State modeling module: fusion of multi-source sensor data to construct a multi-dimensional time series feature tensor of workpiece cleaning state; Region identification module: analyze injection intensity, liquid level, and solvent fluctuation to detect potential cleaning uneven areas; Efficiency map module: fusion of candidate units and equipment parameters to generate a local cleaning efficiency optimization map; Dynamic control module: dynamically adjust injection pressure, liquid level, and time based on the optimization map and VOCs trend and issue control instructions.

Images