PH-temperature two-parameter linkage cleaning regulation and control system, device and method

Abstract

The invention relates to the technical field of paint cleaning control, and discloses a PH-temperature two-parameter linkage cleaning regulation and control system, device and method, and the system comprises a data collection module, a state fusion module, an optimization decision module, a control execution module, a learning feedback module and a cloud interaction module. The method comprises the steps that the pH value, the temperature value, the conductivity and the liquid level value in the cleaning process are collected; fusing the collected data to construct a state vector; solving control variables of a stock solution pump, a clean water pump and a heating valve based on a multi-objective optimization function; linkage control is executed, so that the PH and the temperature of the cleaning solution approach target values; recording a control process state and a result; and uploading the operation data to a cloud server. According to the method, the pH value and temperature double-parameter linkage cleaning regulation and control method is adopted, real-time sensor data and an optimization decision algorithm are combined, accurate regulation of the pH value and temperature of the cleaning liquid is achieved, and the high controllability of the cleaning effect is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of paint cleaning control, specifically to a cleaning control system, device and control method for pH-temperature dual-parameter linkage. Background Art

[0002] In modern industrial production and laboratory environments, the cleaning process is critical, especially when it comes to cleaning sensitive equipment and containers. The pH value and temperature of the cleaning fluid must be maintained within a specific range to ensure cleaning effectiveness and equipment safety.

[0003] Existing cleaning control technologies mostly focus on a single control method based on temperature or pH value. The temperature and pH value are monitored in real time by sensors, and the temperature or pH value of the liquid is adjusted by simple switches or valves.

[0004] However, existing technologies in cleaning and control systems generally rely on single parameter adjustment, making it difficult to achieve coordinated control of multiple key parameters, resulting in untimely system response and a lack of adaptive learning and optimization mechanisms. It is difficult to dynamically adjust control strategies based on historical data or real-time feedback. Most existing cleaning and control systems are limited to local control and have difficulty in effective data interaction and strategy optimization with cloud platforms. Therefore, the present invention provides a cleaning and control system, device, and control method with pH-temperature dual-parameter linkage to address the shortcomings of the existing technology. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the purpose of this application is to provide a cleaning control system, device and control method with pH-temperature dual parameter linkage, which solves the problems that the existing technology is difficult to accurately adjust multiple parameters at the same time, lacks adaptive capabilities and remote collaborative control.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a pH-temperature dual-parameter linkage cleaning control system, including the following modules: The data acquisition module is used to collect the pH value, temperature value and conductivity value of the soaking liquid during the cleaning process, and output the collected multiple sets of parameters as raw data; A state fusion module is used to receive the pH value, temperature value and conductivity value output by the data acquisition module, and fuse them into a state vector to represent the comprehensive state of the current cleaning liquid; an optimization decision module, configured to generate control variables including a heating adjustment amount and a liquid ratio adjustment amount by solving an optimization control function based on a deviation between the state vector and a preset target state vector; A control execution module, configured to control the start and stop or regulation of the raw liquid pump, the clean water pump, and the heating valve in a coordinated manner according to the control variables output by the optimization decision module, so as to drive the pH value and the temperature closer to the target state; A learning feedback module is used to receive the new state vector corresponding to the control execution module after executing the control variables, and record the new and old states and control variables to form training samples, so as to update and optimize the control function in the optimization decision module; The cloud interaction module is used to upload the control data generated in the learning feedback module to the cloud server, receive the control strategy parameters formed by the cloud server based on big data analysis, and dynamically adjust the control model in the optimization decision module.

[0007] Preferably, the data acquisition module includes: PH sensor, used to detect the real-time pH value in the cleaning fluid and generate a pH value signal; A temperature sensor is used to detect the current temperature of the soaking liquid and generate a temperature signal; Conductivity sensor, used to detect changes in the conductivity of the cleaning fluid and generate a conductivity signal.

[0008] Preferably, the state fusion module includes: a state vector construction unit, configured to combine the pH value, temperature value, and conductivity value into a state vector; a deviation calculation unit, used to calculate the deviation between the current state vector and the target state vector; The buffer filter unit is used to smooth the collected original parameters and filter outliers.

[0009] Preferably, the state vector construction unit combines the pH value, temperature value and conductivity value in the following form: ; in, Indicates time The state vector of is the pH value, is the temperature value, is the conductivity value; The deviation calculation unit drives the optimization decision module to execute control variable solution according to the deviation result.

[0010] Preferably, the optimization decision module includes: Optimization model unit, used to construct the control objective function and set constraints; A control variable solving unit is used to generate heating adjustment amount and liquid replenishment adjustment amount through optimization calculation; The weight adjustment unit is used to dynamically adjust the parameter weights in the control model based on historical operation data.

[0011] Preferably, the control objective function constructed by the optimization model unit is: ; in, is the predicted state vector, is the target state vector, is the adjustment amount of the control variable, is the regularization term weight.

[0012] Preferably, the control execution module includes: A raw liquid pump control unit is used to perform pH value deviation correction operations; Clean water pump control unit, used to dilute the cleaning fluid as needed and correct high pH conditions; The heating valve control unit is used to adjust the heating intensity so that the temperature approaches the target range.

[0013] Preferably, the learning feedback module includes: A data storage unit for recording status changes and control instructions for each round of cleaning; Strategy learning unit, used to modify the parameters of the optimization model based on the control effect; The model updating unit is used to dynamically update the control algorithm parameter structure according to the strategy learning results.

[0014] It also provides a pH-temperature dual parameter linkage cleaning control device, including: Soak the tank; A raw liquid pump, a clean water pump and a heating valve connected to the tank; A pH sensor, a temperature sensor and a conductivity sensor are provided on the tank; The main controller is connected to each sensor and pump valve respectively.

[0015] A cleaning control method for pH-temperature dual parameter linkage is also provided, comprising the following steps: Collect pH value, temperature value, conductivity and liquid level value during the cleaning process; The collected data is fused to construct a state vector and compared with the target state vector; Solve the control variables of the raw liquid pump, clean water pump and heating valve based on the multi-objective optimization function; Execute linkage control to make the pH and temperature of the cleaning fluid approach the target values; Record control process status and results for control model learning and strategy optimization; Upload the operating data to the cloud server and receive the updated control strategy from the cloud.

[0016] In summary, this application includes at least one of the following beneficial technical effects: 1. This invention utilizes a dual-parameter cleaning control method that integrates pH and temperature, combining real-time sensor data with an optimized decision-making algorithm to achieve precise regulation of the pH and temperature of the cleaning fluid. By simultaneously monitoring and controlling multiple key parameters, a high degree of controllability of the cleaning effect is achieved. Compared to traditional methods that rely on single-parameter regulation, this method addresses the issues of delayed adjustment and unstable results in the face of complex environmental changes.

[0017] 2. By incorporating a learning feedback module, this invention combines historical data with real-time feedback to adaptively adjust the control strategy within the decision-making module. This allows the system to continuously optimize control parameters based on varying cleaning environments and requirements, enhancing its intelligence. Compared to existing systems lacking self-learning mechanisms, this invention significantly reduces manual intervention and improves long-term operational stability and adaptability.

[0018] 3. This invention deeply integrates the cloud-based interaction module with the system, enabling data sharing and control strategy optimization between the cleaning control system and the cloud platform. The cloud platform can dynamically adjust control strategies based on big data analysis results and push them to the local system, effectively enhancing the flexibility and remote monitoring capabilities of the cleaning process. Compared to the local control methods limited to traditional systems, the cloud-based interaction module breaks through geographical and device limitations, enabling more efficient device management and optimization.

[0019] 4. This invention achieves real-time adjustment of the cleaning fluid state through precise regulation of the raw liquid pump, clean water pump, and heating valve. The control instructions calculated by the optimization decision module are efficiently executed by the control execution module, ensuring that the pH value and temperature remain within the target range. Unlike the manual adjustment or extensive control methods used in traditional cleaning systems, this invention effectively improves the system's response speed and regulation accuracy, reducing the risk of human error. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is the system architecture diagram of this application; Figure 2 This is the device architecture diagram of this application; Figure 3 It is a flow chart of the method of this application. DETAILED DESCRIPTION

[0021] The following is combined with Figure 1 -Attached Figure 3 , further details of this application are given.

[0022] Please see the attached Figure 1 , the PH-temperature dual parameter linkage cleaning control system includes the following modules: The data acquisition module is used to collect the pH value, temperature value and conductivity value of the soaking liquid during the cleaning process, and output the collected multiple sets of parameters as raw data; The state fusion module is used to receive the pH value, temperature value and conductivity value output by the data acquisition module and fuse them into a state vector to represent the comprehensive state of the current cleaning liquid; An optimization decision module is used to generate control variables including heating adjustment amount and liquid ratio adjustment amount by solving an optimization control function based on the deviation between the state vector and the preset target state vector; The control execution module is used to control the start and stop or adjust the raw liquid pump, clean water pump and heating valve according to the control variables output by the optimization decision module to drive the pH value and temperature to the target state; The learning feedback module is used to receive the new state vector corresponding to the control execution module after executing the control variables, and record the new and old states and control variables to form training samples, and update and optimize the control function in the optimization decision module; The cloud interaction module is used to upload the control data generated in the learning feedback module to the cloud server, receive the control strategy parameters formed by the cloud server based on big data analysis, and dynamically adjust and optimize the control model in the decision-making module.

[0023] In this embodiment, the data acquisition module is used to collect the pH, temperature, and conductivity values ​​of the soaking liquid in real time during the cleaning process, and provide this collected raw data to subsequent modules for processing. The design and functionality of the data acquisition module ensure that the system can obtain accurate cleaning liquid parameters in real time to support the efficient operation of the cleaning control system.

[0024] The data acquisition module is one of the foundational modules of the system. By continuously monitoring the pH, temperature, and conductivity of the cleaning fluid, it provides crucial input for subsequent decision-making and control. In one embodiment, the data acquisition module includes a pH sensor, a temperature sensor, and a conductivity sensor, which monitor the pH, temperature, and conductivity of the soaking fluid, respectively. The core task of the data acquisition module is to ensure that these sensors acquire data stably and accurately and convert it into signals that the system can recognize and process.

[0025] The pH sensor detects the pH level of the immersion liquid and generates a pH signal. pH measurement typically relies on electrochemical principles. Specifically, the pH sensor measures pH based on the potential difference between an electrode and the immersion liquid. This allows the pH sensor to accurately and in real time reflect the pH level of the immersion liquid. The pH signal output by the sensor is transmitted to the data acquisition module, forming part of the data transmission chain, ensuring real-time performance and data accuracy.

[0026] The temperature sensor is used to detect the temperature of the immersion liquid in real time and output a temperature signal. Temperature measurement is typically based on the thermoelectric effect or other principles of changes in heat-sensitive materials. Specifically, the temperature sensor in this embodiment uses a high-precision thermocouple or RTD sensor, which has high response speed and stability. The sensor detects temperature changes and converts the data into an electrical signal for output, ensuring that the system can obtain the current immersion liquid temperature in real time and provide this data to subsequent processing modules for analysis and adjustment.

[0027] The conductivity sensor measures the conductivity of the soaking liquid, indirectly reflecting the concentration of dissolved substances in the liquid. Conductivity is positively correlated with the total dissolved substances in the liquid (such as salt and acid). Therefore, conductivity data can help the system infer changes in the cleaning liquid's concentration and possible pH fluctuations. In this embodiment, the conductivity sensor applies an electric field and measures the change in current to obtain the liquid's conductivity value. This ultimately generates a conductivity signal, which is transmitted to subsequent modules via a data transmission chain.

[0028] In one possible implementation, to improve the real-time nature and accuracy of data acquisition, the data acquisition frequency of all sensors can be adjusted according to system needs. For example, during a high-precision cleaning process, the system can be configured to collect data frequently (e.g., once every second) to provide real-time feedback and adjust the pH and temperature of the cleaning fluid. Furthermore, to ensure data accuracy, the data acquisition module may include calibration and filtering functions. These functions effectively eliminate measurement errors caused by the environment or the sensors themselves.

[0029] The collected pH, temperature, and conductivity signals undergo signal conversion and are then transmitted to subsequent modules for processing. The data acquisition module converts the analog signals into digital signals using an analog-to-digital converter (ADC) for use by the state fusion module. This ensures that raw sensor data can be accurately transmitted to other modules in the control system, ensuring the accuracy and effectiveness of subsequent processing.

[0030] In specific implementations, to improve system stability and response speed, the data acquisition module may also employ anti-interference technologies to ensure sensor signal stability and accuracy under varying operating environments. Signal interference, noise, or electrical fluctuations between sensors can affect data accuracy. Therefore, employing techniques such as shielding, filtering, and signal amplification can effectively improve system signal quality.

[0031] In this embodiment, the state fusion module is responsible for integrating the pH, temperature, and conductivity values ​​output by the data acquisition module into a comprehensive state vector representing the overall state of the cleaning fluid. The state fusion module converts the multiple sets of sensor data acquired by the data acquisition module into a comprehensive state description and provides the necessary input for the subsequent optimization and decision-making module. This process ensures that the system can accurately perceive changes in the cleaning fluid state and provides a basis for precise control decisions.

[0032] In the aforementioned data acquisition module, pH, temperature, and conductivity signals are collected and output in real time. The state fusion module processes this raw data to construct a multidimensional state vector, providing comprehensive information for subsequent control. In this embodiment, the state fusion module not only performs data fusion but also includes data preprocessing and outlier filtering to ensure that the data used in subsequent steps is accurate and stable.

[0033] In this embodiment, one of the core tasks of the state fusion module is to fuse the raw data from different sensors into a unified state vector. Specifically, the state fusion module combines the pH value, temperature value, and conductivity value into a three-dimensional state vector according to the following formula: ; in, Indicates time The state vector of is the pH value, is the temperature value, is the conductivity value. This state vector intuitively describes the current state of the cleaning fluid, integrating key factors such as the acidity, alkalinity, temperature, and conductivity changes of the cleaning fluid.

[0034] In one possible implementation, the state vector construction process involves more than simple data combination; it may also include filtering and normalization. This filtering process removes noise from the raw data, ensuring that the fused data is more accurate and stable. Specifically, low-pass filtering or Kalman filtering algorithms may be used to smooth the data, eliminating transient measurement fluctuations and improving data reliability.

[0035] Once the state vector is constructed, the state fusion module then calculates the deviation between the current state vector and the preset target state vector. The purpose of the deviation calculation is to measure the difference between the current state of the cleaning fluid and the desired state, providing important information for the subsequent optimization decision module.

[0036] In this embodiment, the deviation calculation unit calculates the deviation through the following steps: Input: Real-time state vector obtained by sensors ; Calculation target: preset target state vector , indicating the ideal pH value, temperature and conductivity; Calculate the difference: get the deviation value by comparing the difference between the current state vector and the target state vector ,The difference value reflects the distance between the current state of the cleaning fluid and the target state.

[0037] Specifically, the deviation calculation process can be expressed as: ; in, is the state deviation at time t. This deviation value will serve as the input of the optimization decision module to help the system calculate the control quantity that needs to be adjusted.

[0038] Typically, raw data can be affected by environmental noise or sensor errors. Therefore, in addition to constructing the state vector and calculating deviations, data filtering and outlier handling are also crucial in the state fusion module. In some embodiments, the state fusion module applies filtering algorithms, such as low-pass filtering, Kalman filtering, or sliding average filtering, to smooth the data and reduce occasional noise interference.

[0039] Outlier processing involves detecting and eliminating abnormal data caused by equipment failure or external interference. For example, if a sensor's reading significantly deviates from the normal range (such as a pH value exceeding a predetermined range or an abnormal temperature change), the state fusion module will automatically flag and remove this data to ensure the reliability of subsequent system processing.

[0040] In this embodiment, the state fusion module serves as a bridge between the data acquisition module and the optimization and decision-making module, undertaking multiple tasks including data preprocessing, fusion, and deviation calculation. The data acquisition module provides raw sensor data, which the state fusion module effectively integrates to generate a state vector reflecting the current state of the cleaning fluid. This information is then passed to the optimization and decision-making module. The optimization and decision-making module calculates deviations and outputs control instructions, which drive the control execution module to adjust the pH and temperature, ensuring that the cleaning fluid reaches the preset target state as quickly as possible.

[0041] Specifically, in one implementation, the state fusion module ensures the accuracy and stability of data processing through various means. For example, the state fusion module may perform adaptive adjustments based on historical data, allowing the system to automatically adjust the state vector construction and deviation calculation methods based on complex situations that may arise during the cleaning process.

[0042] In this embodiment, the optimization decision module is responsible for calculating the control variable and generating control instructions based on the state vector output by the state fusion module and its deviation from the target state vector. These control instructions are then transmitted to the control execution module to guide the operation of the heating valve, raw liquid pump, and clean water pump. By solving the control objective function and combining historical data with real-time feedback, the optimization decision module ensures that the pH value and temperature of the cleaning fluid remain within the set range.

[0043] The optimization decision module is a core component of the cleaning control system, enabling precise control and optimized performance. After the state fusion module generates a state vector and calculates the deviation, the optimization decision module uses this deviation information to determine the amount of cleaning fluid to be adjusted using a specific optimization algorithm.

[0044] The optimization decision module solves the control variable by constructing an optimization objective function. This objective function is usually designed to minimize the deviation between the current state vector and the target state vector, and adds a penalty term for the control variable to ensure the balance of the control variable. Specifically, the objective function can be expressed as: ; in, is the predicted state vector, representing the moment expected state; is the target state vector, representing the ideal pH value, temperature and conductivity; is the adjustment amount of the controlled quantity, indicating the adjustment of the raw liquid pump, clean water pump and heating valve; is the regularization term weight, which is used to balance the control accuracy and the size of the control amount.

[0045] The first term of the objective function is used to minimize the deviation between the current state and the target state, ensuring that the cleaning fluid can be as close to the target value as possible; the second term is a penalty for changes in the control quantity to avoid over-adjustment that leads to unnecessary energy consumption or system fluctuations. The introduction of makes the objective function not only focus on minimizing the deviation, but also optimize the system performance by balancing the size of the adjustment quantity.

[0046] In this embodiment, the optimization decision module solves the objective function to generate control variables. These control variables cover the control variables of the heating valve, raw liquid pump, and clean water pump. Using optimization algorithms (such as gradient descent and Newton's method), the module can quickly determine the optimal control variables.

[0047] Heating valve regulation: The control variable for the heating valve is determined based on the temperature deviation. The optimization decision module calculates the temperature deviation and generates a control signal for the heating valve, adjusting the heating intensity to maintain the temperature within the target range.

[0048] Stock solution pump adjustment: The stock solution pump adjustment amount is based on the pH deviation. If the pH value is too low, the system will increase the stock solution amount to increase the pH value, and vice versa, reduce the stock solution amount.

[0049] Clean water pump adjustment: When the pH value is too high, the control volume of the clean water pump will increase to dilute the cleaning solution and lower the pH value to the target range.

[0050] These control quantities are obtained through calculations in the optimization decision module, and each control behavior is ensured to meet the system objectives. The final generated control signal is passed to the control execution module.

[0051] Alternatively, the optimization decision module can leverage historical operational data to dynamically adjust the optimization process. By learning the correspondence between state changes and control variables from historical data, the optimization decision module can continuously refine the parameters of its objective function, thereby improving control accuracy and system response speed.

[0052] For example, in one embodiment, the optimization decision module can record the state vector and control quantity during each cleaning process, and optimize the decision process through this learning feedback mechanism. As historical data accumulates, the module can adjust the weight factors in the objective function. or other algorithm parameters, thereby improving the adaptability and accuracy of the system.

[0053] Specifically, in some embodiments, the optimization decision module can employ adaptive control techniques to adjust the system's control strategy. In this case, the parameters of the objective function can be adjusted online through a feedback mechanism, enabling the system to dynamically respond to various changes during the cleaning process, such as dramatic fluctuations in liquid temperature or pH.

[0054] Under this adaptive control strategy, the optimization decision module not only performs calculations based on real-time data, but also continuously optimizes the control strategy through continuous learning of historical feedback to ensure the stability and accuracy of the system in different working environments.

[0055] In this embodiment, the control execution module receives control variables transmitted by the optimization decision module and adjusts the operating states of the raw liquid pump, clean water pump, and heating valve according to these control instructions to achieve precise control of pH and temperature. By performing specific adjustment operations, this module converts the adjustment variables calculated by the optimization decision module into actual operations, ensuring that the pH and temperature of the cleaning fluid are always maintained within the set target range.

[0056] The connection between the control execution module and the aforementioned modules is crucial. After the optimization decision module calculates the control variables, the control execution module adjusts the execution equipment according to these instructions, directly affecting the liquid state during the cleaning process. The precise operation of the control execution module ensures the efficient and stable operation of the entire system and is a crucial link in system regulation.

[0057] In this embodiment, the control amount of the stock solution pump is calculated by the optimization decision module and adjusted based on the pH deviation. Specifically, the adjustment amount of the stock solution pump is directly related to the magnitude of the pH deviation. When the pH value is low, the optimization decision module calculates a specific control signal to start the stock solution pump, increase the stock solution concentration in the cleaning solution, and raise the pH value. Conversely, when the pH value is high, the control signal of the stock solution pump is reduced, reducing the amount of stock solution added.

[0058] The control execution module receives control instructions from the optimization decision module to start or stop the stock solution pump and adjust the stock solution flow rate. This process ensures that the pH value can remain stable within the target range without drastic fluctuations in the liquid state due to excessive or insufficient addition of stock solution.

[0059] The clean water pump is also controlled by the optimization decision module based on pH deviation. In this embodiment, when the pH value is too high, the clean water pump control signal is increased, starting the clean water pump and adjusting the flow rate, thereby diluting the cleaning fluid and lowering the pH value until it reaches the target range. This control mechanism ensures that the cleaning effect is not affected by excessively high pH values.

[0060] The accuracy of clean water pump control is also crucial. With the support of the optimization decision module, the control execution module can make timely adjustments based on real-time pH deviations to avoid excessive or insufficient clean water pump output, ensuring that each adjustment meets optimal control requirements.

[0061] The heating valve is regulated by the optimization and decision-making module based on temperature deviation. In this embodiment, the control variable of the heating valve is calculated by the optimization and decision-making module to regulate the temperature of the cleaning fluid. Specifically, when the cleaning fluid temperature falls below the target range, the optimization and decision-making module calculates a control signal for the heating valve, activating the heating device to increase the fluid temperature. Conversely, when the temperature is too high, the optimization and decision-making module reduces the heating intensity or shuts down the heating device to prevent overheating.

[0062] Precise control of the heating valve not only impacts temperature stability but also indirectly influences pH regulation. Because temperature influences pH measurement and response, controlling the operating state of the heating valve is crucial. The control execution module, receiving instructions from the optimization decision module, ensures that the operation of the heating valve matches the actual needs of the cleaning fluid, thereby improving the overall performance and effectiveness of the system.

[0063] Typically, the control execution module implements the control instructions calculated by the optimization decision module by executing the aforementioned operations. The coordinated regulation of the raw liquid pump, clean water pump, and heating valve constitutes the core control mechanism of the entire cleaning control system. Specifically, the control execution module receives real-time instructions from the optimization decision module and operates the execution devices accordingly, ensuring that the system can make optimal adjustments based on the real-time status of the cleaning fluid.

[0064] Alternatively, the control execution module's operating status can interact with the learning feedback module through a feedback mechanism. Through this interaction, the control execution module can dynamically adjust the control strategy based on historical data and real-time feedback, making the system more adaptable to different working environments and cleaning requirements.

[0065] In one possible implementation, data transmission between the control execution module and the optimization decision module utilizes a high-speed communication interface to ensure real-time transmission of control signals. This efficient data transmission ensures rapid response to control signals, thereby improving the system's real-time performance and accuracy. Furthermore, the control execution module may utilize an integrated execution unit to achieve unified control of multiple pumps and valves, thereby reducing system complexity and improving operational convenience and stability.

[0066] In this embodiment, the learning feedback module is used to record the new state of the control execution module after executing the control variables. The correspondence between the old and new states and the control variables is recorded to form training samples, thereby updating and optimizing the control functions in the optimization decision module. The core task of the learning feedback module is to promote system self-learning and adjustment through feedback mechanisms, enabling continuous optimization of the control strategy through repeated application, thereby improving the long-term stability and control accuracy of the system.

[0067] The learning feedback module is a crucial component of the optimization decision-making mechanism in this invention. After the control execution module completes its equipment adjustments, the learning feedback module records the state changes and corresponding control instructions for each cleaning process in real time. Through continuous feedback and learning, the system's control capabilities are gradually improved. By accumulating historical data, the system can adaptively adjust its control strategy to accommodate varying cleaning environments and operating conditions.

[0068] In this embodiment, the learning feedback module first includes a data storage unit, which records state changes, control variables, and execution results during each cleaning cycle. The data storage unit stores all relevant data, including pH values, temperature, conductivity, and control instructions, in chronological order. By accumulating long-term historical data, the data storage unit provides sufficient input for subsequent strategy learning and optimization.

[0069] In some embodiments, the data storage unit can use an efficient data storage structure, such as a database or distributed storage system, to ensure efficient access and processing of large amounts of data. This stored historical data will be fed back into the learning feedback module as input, ensuring that the system can make appropriate adjustments based on actual operational history.

[0070] The policy learning unit is the core component of the learning feedback module. It analyzes historical data, particularly the relationship between control variables and state changes, to modify the control function in the optimization decision module. Generally, the policy learning unit uses machine learning algorithms (such as supervised learning and reinforcement learning) to extract the optimal direction of the control strategy based on the control effect and target deviation from historical data.

[0071] Specifically, the strategy learning unit can work through the following steps: Input historical data: obtain the control variables and status information of each round of cleaning process from the data storage unit; Calculate deviations and errors: Evaluate the effectiveness of the current control strategy based on the deviation between the actual performance of the control execution module and the target state; Optimize control strategy: Based on the deviation calculation results, the strategy learning unit uses a suitable optimization algorithm (such as gradient descent, Q-learning, etc.) to adjust the control strategy and update the parameters in the control model.

[0072] Through this process, the strategy learning unit continuously optimizes the control function of the decision module, enabling the system to continuously improve control accuracy and efficiency over time. Specifically, the strategy learning unit adjusts and optimizes the weight parameters in the decision module to enhance the system's responsiveness to changes in pH and temperature.

[0073] To enable adaptive adjustments in the optimization decision module, the learning feedback module also includes a model update unit. This unit dynamically updates the control algorithm parameters in the optimization decision module based on the optimization results of the strategy learning unit. In this embodiment, the model update unit analyzes new training samples and control strategies to adjust weights, coefficients, or other parameters in the control function to adapt to new operational requirements or environmental changes.

[0074] For example, assuming that the system experiences different working conditions or environmental changes over a period of time, the model update unit will make adaptive adjustments based on these changes to ensure that the optimization decision module can always perform optimal control based on the latest operating data.

[0075] In one possible implementation, the model update unit can be implemented through incremental learning. Incremental learning means that the system can make small updates to the control strategy based on each new cleaning process without retraining the entire model, thereby improving the efficiency and real-time performance of the learning process.

[0076] In some embodiments, the learning feedback module also features dynamic adjustment capabilities, adjusting the frequency of learning and policy updates based on real-time feedback. For example, if the system experiences significant fluctuations or abnormal conditions during the cleaning process, the learning feedback module can increase the learning frequency to quickly adapt to the current changes. Conversely, when the system is operating stably, the learning frequency can be appropriately reduced to avoid overtraining and unnecessary computational overhead.

[0077] Specifically, the learning feedback module can adjust the weight and priority of each round of data update based on the feedback of the control effect, ensuring that important control strategy adjustments can be processed first to improve the response speed and stability of the overall system.

[0078] In this embodiment, the cloud interaction module is primarily responsible for uploading system operation data to the cloud server and receiving control strategy adjustment instructions from the cloud. This module enables effective communication and data exchange between the cleaning control system and the cloud platform, leveraging the cloud's big data processing capabilities to dynamically adjust and optimize the system's control strategy. The introduction of this module not only ensures the system's data storage and analysis capabilities but also enhances its intelligence, making the cleaning process more flexible and efficient.

[0079] The cloud interaction module serves as a bridge between the system and the cloud, acquiring real-time device operating data and synchronizing it with the cloud server. This module uploads data generated in the aforementioned optimization decision module and learning feedback module to the cloud, providing the cloud platform with a basis for optimizing the control model. Furthermore, the cloud platform analyzes the uploaded data and generates new control strategies, which are then distributed to the local system via the cloud interaction module, enabling precise control and long-term optimization.

[0080] In this embodiment, the cloud interaction module uploads system-generated operational data to a cloud server via a communication interface. This data includes information such as pH, temperature, and conductivity collected during the cleaning process, as well as adjustment instructions from the control execution module. By centrally storing and managing this data on the cloud platform, the system enables cross-device and cross-scenario data sharing and analysis.

[0081] Typically, the cloud interaction module establishes a connection with the cloud platform via standard internet communication protocols (such as HTTP, WebSocket, or MQTT). During data upload, the cloud interaction module first processes, encodes, and encrypts the real-time data collected, and then transmits it to the cloud server. The cloud platform can store this data long-term and provide a foundation for subsequent data analysis.

[0082] As an option, the cloud interaction module can also implement data compression to reduce the data traffic generated during the upload process. This measure is particularly important for large-scale systems or when multiple devices work together, and can effectively improve the communication efficiency and stability of the system.

[0083] Once the cloud platform receives operational data from the local system, it will be deeply processed by the cloud analytics platform using big data analysis and machine learning algorithms. By analyzing historical data and current operational status, the cloud platform can generate more precise control strategies and push them to the local system for timely adjustments.

[0084] Specifically, the cloud platform performs the following analysis based on the uploaded data: Data aggregation and trend analysis: The cloud platform aggregates and analyzes data from multiple cleaning devices or multiple operating cycles to identify possible control bottlenecks or optimization opportunities. Control strategy adjustment: Based on data analysis results, the cloud platform adjusts the parameters in the control algorithm and generates a new control strategy; Feedback optimization: The cloud platform can also use feedback mechanisms to monitor the implementation effects of new control strategies and conduct further optimization.

[0085] The generated control strategies may involve optimizing the pH and temperature of the cleaning fluid, or adjusting the control quantity according to different environmental conditions. The cloud interaction module sends these control strategies to the optimization decision module of the local system via a secure communication link.

[0086] Generally, the cloud interaction module transmits data and receives instructions in real time, ensuring collaborative operation between the system and the cloud. Specifically, in some embodiments, the cloud interaction module may continuously synchronize data with the cloud in the background to ensure that the local system receives the latest control strategy adjustments at any time. Furthermore, the cloud interaction module can also provide fault diagnosis and early warning capabilities. If the system experiences anomalies or deviations, the cloud platform will issue emergency adjustment instructions through the cloud interaction module to ensure the stable operation of the cleaning process.

[0087] In some cases, the cloud interaction module can also support remote monitoring, allowing operators to monitor and control the system in real time through the cloud platform. This function is very important for the centralized management of large cleaning systems or multiple cleaning equipment, reducing manual intervention and improving equipment reliability.

[0088] In this embodiment, the cloud interaction module is designed with data security in mind. To prevent data tampering or leakage during transmission, the cloud interaction module utilizes data encryption and identity authentication technologies. Every data upload is encrypted, ensuring that it cannot be illegally accessed or modified during transmission. Furthermore, communication between the system and the cloud platform utilizes a two-way authentication mechanism to ensure that every command push originates from a legitimate cloud platform.

[0089] Specifically, the cloud interaction module may use the SSL / TLS protocol to encrypt the transmitted data to ensure the confidentiality and integrity of the information. In addition, in some embodiments, the cloud interaction module can also be integrated with the cloud platform's identity authentication system to ensure that only authorized systems can upload data or receive control instructions.

[0090] The PH-temperature dual-parameter linkage cleaning control device described below and the PH-temperature dual-parameter linkage cleaning control system described above can correspond to each other.

[0091] Please see the attached Figure 2 The present invention also provides a pH-temperature dual-parameter linkage cleaning control device, comprising: Soak the tank; The raw liquid pump, clean water pump and heating valve connected to the tank; PH sensor, temperature sensor and conductivity sensor installed on the tank; The main controller is connected to each sensor and pump valve respectively.

[0092] In order to further illustrate the collaborative working process of the technical solution described in the present invention, a specific working scenario example will be used below to illustrate.

[0093] The entire cleaning process can be divided into four main stages: initial liquid addition and preparation, soaking and paint stripping, waste discharge and recycling, and cleaning and rinsing. 1. Initial liquid addition and preparation stage: The main controller (the brain of the system, integrating modules such as optimization decision-making and control execution) sends instructions to control the solenoid valve and electric diaphragm pump (water supply pump, a type of clean water pump) corresponding to the water inlet to open.

[0094] Clean water is pumped from the clean water tank into the soaking tank until the level sensor in the soaking tank detects a preset low water level.

[0095] The data acquisition module starts to collect the pH value, temperature value and conductivity value in the immersion tank in real time.

[0096] The optimization decision module calculates the initial injection volume of the required high-concentration stock solution based on the preset target pH value of the cleaning solution.

[0097] The main controller controls the solenoid valve and electric diaphragm pump (stock liquid pump) corresponding to the supply tank to open, and the high-concentration stock liquid is injected into the immersion tank from the supply tank. The data acquisition module continuously monitors the real-time changes of the pH value in the immersion tank.

[0098] The state fusion module combines real-time pH, temperature, and conductivity values ​​into a state vector. It automatically establishes a relationship between pH and stock solution replenishment time, taking into account the current clean water volume and stock solution injection time.

[0099] The learning feedback module records the amount, time and final pH value of the stock solution injection to form a training sample for subsequent iterative optimization of the stock solution replenishment strategy.

[0100] Once the pH sensor detects that the pH value of the cleaning liquid in the immersion tank reaches the target range, the main controller immediately turns off the raw liquid pump.

[0101] Subsequently, the system will automatically complete the subsequent filling of clean water or original liquid according to the relationship until the liquid level sensor in the immersion tank detects that the immersion liquid level has reached full load. At this time, the main controller closes all liquid inlet pumps and corresponding solenoid valves to obtain the pH value and liquid level of the immersion liquid.

[0102] 2. Soaking and paint stripping stage: Place the parts in the immersion tank, and use the data acquisition module to continuously collect real-time data from the pH sensor, temperature sensor, and conductivity sensor in the immersion tank.

[0103] These real-time parameters are displayed on the operation screen through the main controller, and the cloud interaction module uploads this data to the cloud server at regular intervals.

[0104] The state fusion module constructs the state vector in real time and calculates the deviation with the preset target state vector.

[0105] Based on this deviation, the optimization decision module calculates the precise heating adjustment amount and the possible required liquid ratio adjustment amount by solving the internal optimization control function (if there is a large deviation in the pH value during the soaking process, it may trigger the replenishment of trace stock solution or clean water).

[0106] The control execution module receives instructions from the optimization decision module: When the temperature deviates from the target range, the start and stop and adjustment of the heating valve are controlled to keep the temperature within the optimal immersion paint stripping range.

[0107] When there is a slight deviation in the pH value, a micro-replenishment is made through an electric diaphragm pump (raw liquid pump or clean water pump) according to the optimization decision to ensure the stability of the pH value.

[0108] The learning feedback module records each parameter fluctuation, control instruction and its effect, and continuously optimizes the control function parameters in the decision-making module.

[0109] The cloud interaction module uploads these control data to the cloud, which further optimizes the control strategy parameters through big data analysis and sends them to the local system for dynamic adjustment.

[0110] The main controller has a built-in timer. When the preset soaking time is reached, the soaking process is completed.

[0111] 3. Waste discharge and recycling stage: The main controller first starts the chain conveyor type slag discharge (the conveyor belt starts running with the water filtering function). At the same time, the main controller controls the solenoid valve to open the soaking tank drain valve.

[0112] The paint residue and diluted soaking liquid in the soaking tank begin to be discharged. The paint residue is transported to the waste bucket by the conveyor belt on the chain conveyor type slag discharge, and the soaking liquid flows into the recovery water tank below.

[0113] When the liquid level sensor in the soaking tank detects that the water level has been drained, the main controller stops the chain conveyor discharge after a few seconds. The liquid level sensor also checks whether the water level in the turnover tank is low (recyclable). If the recovery conditions are met, the main controller controls the solenoid valve to open the recovery water pump, pumping the soaking liquid from the recovery water tank into the turnover tank. The filter tank performs preliminary filtration on the liquid entering the turnover tank, extending the service life of the recovered liquid.

[0114] 4. Cleaning and showering stage: The main controller starts the chain conveyor type slag discharge again, and at the same time controls the solenoid valve to open the immersion tank drain valve and the recovery tank drain valve to ensure that the flushing wastewater can be discharged directly to the sewage outlet.

[0115] When the liquid level sensor in the recovery water tank detects that there is basically no liquid, the main controller stops the recovery water pump and completes the recovery process.

[0116] The main controller controls the electric diaphragm pump (cleaning pump) to start, draws clean water from the clean water tank, and uses the spray device to perform high-pressure flushing on the parts in the immersion tank. The residual paint residue flushed down continues to be transferred to the waste residue bucket through the chain-type slag discharge, and the flushing waste water is discharged through the drain valve and the sink drain outlet.

[0117] When the preset designated cleaning time is reached, the main controller stops the cleaning pump and completes the cleaning process.

[0118] Please see the attached Figure 3The PH-temperature dual-parameter linkage cleaning control method described below and the PH-temperature dual-parameter linkage cleaning control system described above can refer to each other.

[0119] S1. Collect pH value, temperature value, conductivity and liquid level value during the cleaning process; S2, fuse the collected data to construct a state vector and compare it with the target state vector; S3. Solve the control variables of the raw liquid pump, clean water pump and heating valve based on the multi-objective optimization function; S4, execute linkage control to make the pH and temperature of the cleaning fluid approach the target values; S5. Record the control process status and results for control model learning and strategy optimization; S6. Upload the operating data to the cloud server and receive the control strategy updated by the cloud.

[0120] The method of this embodiment can be used to execute the above system embodiment. Its principles and technical effects are similar and will not be described in detail here.

[0121] The examples of this specific embodiment are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Identical components are represented by the same reference numerals. Therefore, any equivalent changes made based on the structure, shape, and principle of this application should be included in the scope of protection of this application.

Claims

1. The cleaning control system with pH-temperature dual parameter linkage is characterized by: Includes the following modules: The data acquisition module is used to collect the pH value, temperature value and conductivity value of the soaking liquid during the cleaning process, and output the collected multiple sets of parameters as raw data; A state fusion module is used to receive the pH value, temperature value and conductivity value output by the data acquisition module, and fuse them into a state vector to represent the comprehensive state of the current cleaning liquid; an optimization decision module, configured to generate control variables including a heating adjustment amount and a liquid ratio adjustment amount by solving an optimization control function based on a deviation between the state vector and a preset target state vector; A control execution module, configured to control the start and stop or regulation of the raw liquid pump, the clean water pump, and the heating valve in a coordinated manner according to the control variables output by the optimization decision module, so as to drive the pH value and the temperature closer to the target state; A learning feedback module is used to receive the new state vector corresponding to the control execution module after executing the control variables, and record the new and old states and control variables to form training samples, so as to update and optimize the control function in the optimization decision module; The cloud interaction module is used to upload the control data generated in the learning feedback module to the cloud server, receive the control strategy parameters formed by the cloud server based on big data analysis, and dynamically adjust the control model in the optimization decision module.

2. The pH-temperature dual-parameter linkage cleaning control system according to claim 1 is characterized in that: The data acquisition module includes: PH sensor, used to detect the real-time pH value in the cleaning fluid and generate a pH value signal; A temperature sensor is used to detect the current temperature of the soaking liquid and generate a temperature signal; Conductivity sensor, used to detect changes in the conductivity of the cleaning fluid and generate a conductivity signal.

3. The pH-temperature dual-parameter linkage cleaning control system according to claim 1 is characterized in that: The state fusion module includes: a state vector construction unit, configured to combine the pH value, temperature value, and conductivity value into a state vector; a deviation calculation unit, used to calculate the deviation between the current state vector and the target state vector; The buffer filter unit is used to smooth the collected original parameters and filter outliers.

4. The pH-temperature dual-parameter linkage cleaning control system according to claim 3 is characterized in that: The state vector construction unit combines the pH value, temperature value and conductivity value in the following form: ； in, Indicates time The state vector of is the pH value, is the temperature value, is the conductivity value; The deviation calculation unit drives the optimization decision module to execute control variable solution according to the deviation result.

5. The pH-temperature dual-parameter linkage cleaning control system according to claim 1 is characterized in that: The optimization decision module includes: Optimization model unit, used to construct the control objective function and set constraints; A control variable solving unit is used to generate heating adjustment amount and liquid replenishment adjustment amount through optimization calculation; The weight adjustment unit is used to dynamically adjust the parameter weights in the control model based on historical operation data.

6. The pH-temperature dual-parameter linkage cleaning control system according to claim 5 is characterized in that: The control objective function constructed by the optimization model unit is: ； in, is the predicted state vector, is the target state vector, is the adjustment amount of the control variable, is the regularization term weight.

7. The pH-temperature dual-parameter linkage cleaning control system according to claim 1 is characterized in that: The control execution module includes: A raw liquid pump control unit is used to perform pH value deviation correction operations; Clean water pump control unit, used to dilute the cleaning fluid as needed and correct high pH conditions; The heating valve control unit is used to adjust the heating intensity so that the temperature approaches the target range.

8. The pH-temperature dual-parameter linkage cleaning control system according to claim 1 is characterized in that: The learning feedback module includes: A data storage unit for recording status changes and control instructions for each round of cleaning; Strategy learning unit, used to modify the parameters of the optimization model based on the control effect; The model updating unit is used to dynamically update the control algorithm parameter structure according to the strategy learning results.

9. A cleaning control device with pH-temperature dual-parameter linkage, applied to the cleaning control system with pH-temperature dual-parameter linkage according to any one of claims 1 to 8, characterized in that: include: Soak the tank; A raw liquid pump, a clean water pump and a heating valve connected to the tank; A pH sensor, a temperature sensor and a conductivity sensor are provided on the tank; The main controller is connected to each sensor and pump valve respectively.

10. A cleaning control method with pH-temperature dual-parameter linkage, applied to a cleaning control system with pH-temperature dual-parameter linkage according to any one of claims 1 to 8, characterized in that: The following steps are involved: Collect pH value, temperature value, conductivity and liquid level value during the cleaning process; The collected data is fused to construct a state vector and compared with the target state vector; Solve the control variables of the raw liquid pump, clean water pump and heating valve based on the multi-objective optimization function; Execute linkage control to make the pH and temperature of the cleaning fluid approach the target values; Record control process status and results for control model learning and strategy optimization; Upload the operating data to the cloud server and receive the updated control strategy from the cloud.

Images