Intelligent control method and system for cleaning tank
By establishing a predictive model within the cleaning tank and dynamically adjusting the valve opening degree and timing, the problem of wasted pure water and steam during the cleaning process was solved, achieving efficient resource utilization and improved production efficiency.
Patent Information
- Application Number
- CN202511037565.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-25
- Publication Date
- 2025-11-07
Smart Images
Figure CN120901005A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of metal surface treatment, in particular to an intelligent control method and control system of a cleaning tank. BACKGROUND
[0002] Metal surface treatment is one of the important processes in the production and manufacturing of metal parts. A large amount of pure water is needed to clean the workpieces during production, and the pure water needs to be heated during cleaning. If the workpiece surface is not cleaned properly, the workpiece surface will have color difference, concave-convex points, uneven plating, and other defects, affecting the uniformity of subsequent dyeing, resulting in differences in local dyeing results of the workpiece, and reducing the yield of the product.
[0003] In related technologies, when cleaning the workpiece with pure water, it is necessary to ensure that there is a certain height and purity of pure water in the cleaning tank as cleaning liquid. Because the dyeing solution remaining on the surface of the workpiece accumulates in the cleaning tank during the cleaning process, if the concentration of the dyeing solution in the tank is too high, the surface of the workpiece will cause color difference and defects. Therefore, it is necessary to constantly supplement pure water to dilute the cleaning liquid. Excessive pure water can reduce the adhesion of the residual solution of the workpiece, but it will cause a lot of waste of pure water.
[0004] At the same time, in order to maintain the cleaning liquid in the cleaning tank at the required temperature, a heating pipe is needed to be connected to the bottom of the cleaning tank to supply steam to heat the cleaning liquid in the tank. Excessive heating will also cause waste of steam.
[0005] Therefore, the problems of waste of pure water and waste of steam have become urgent problems to be solved. SUMMARY
[0006] The present application relates to a cleaning control method of a cleaning tank to solve the problems of waste of pure water and waste of steam in the prior art when cleaning the workpiece.
[0007] In a first aspect, the embodiments of the present application provide an intelligent control method of a cleaning tank, and the method comprises: Based on the state of the cleaning liquid in the cleaning tank, collecting historical data of related production parameters and preprocessing, determining the influence relationship of the production parameters on the cleaning liquid state, and constructing a prediction model of the production parameters on the cleaning liquid state; Obtaining the current state of the cleaning liquid in the cleaning tank, if the current state of the cleaning liquid in the cleaning tank does not meet the preset requirements, based on the trained prediction model, determining the opening degree and / or opening time of the valve connected to the cleaning tank according to the influence relationship and the current state of the cleaning liquid.
[0008] Optionally, it comprises: The state of the cleaning liquid in the cleaning tank includes a water quality state and / or a water temperature state. The production parameters include at least one of a number of processed workpieces, a steam amount, and a pure water replenishment amount; the valve includes a replenishment valve for replenishing the cleaning tank with water and / or a steam valve for heating the cleaning liquid in the cleaning tank, and the opening degree and / or opening time of the valve include a first opening degree and a first opening time of the replenishment valve determined according to the first target influence relationship and the water quality state, and a second opening degree of the steam valve determined according to the second target influence relationship and the water temperature state. A first opening instruction is sent to the replenishment valve according to the first opening degree and the first opening time to control the replenishment valve to open to replenish the cleaning liquid with water, and a second opening instruction is sent to the steam valve according to the second opening degree to control the steam valve to open to heat the cleaning liquid.
[0009] Optionally, the conductivity of the cleaning liquid in the cleaning tank is monitored in real time to obtain the water quality state of the cleaning liquid. The first opening degree and the first opening time of the replenishment valve are the opening degree and the opening time of the minimum water consumption to ensure that the conductivity of the cleaning liquid is always lower than a preset conductivity threshold. The second opening degree of the steam valve is the opening degree of the minimum steam amount to ensure that the temperature of the cleaning liquid is always at a preset temperature threshold.
[0010] Optionally, the prediction model includes a first prediction model, and the first prediction model is a model for determining a first target influence relationship of the opening degree and the opening time of the replenishment valve on the conductivity of the cleaning liquid under the number of the current processed workpieces. The opening degree and / or opening time of the valve are determined according to the influence relationship and the current state of the cleaning liquid based on the trained prediction model, including: The number of the current processed workpieces is input into the first prediction model that is pre-trained, and the first opening degree and the first opening time of the replenishment valve are determined by the first prediction model. The first prediction model is used to determine the first opening degree and the first opening time of the replenishment valve according to the current conductivity of the cleaning liquid and the first target influence relationship of the opening degree and the opening time of the replenishment valve on the conductivity of the cleaning liquid.
[0011] Optionally, the obtaining the current state of the cleaning liquid in the cleaning tank comprises obtaining a current water temperature of the cleaning liquid, and the prediction model further comprises a second prediction model, the second prediction model being a model for determining a second target influence relationship between the opening degree of the steam valve and the water temperature of the cleaning liquid under the current pure water replenishment flow; The determining the opening degree and / or opening time of the valve based on the prediction model trained and the current state of the cleaning liquid comprises: inputting the pure water replenishment flow into the second prediction model trained in advance, and determining the second opening degree of the steam valve through the second prediction model; The second prediction model is used to determine the second opening degree of the steam valve according to the current water temperature of the cleaning liquid and the second target influence relationship between the opening degree of the steam valve and the water temperature of the cleaning liquid.
[0012] Optionally, the constructing the prediction model of the production parameter on the state of the cleaning liquid further comprises: training the first prediction model in advance; collecting first historical data of the related production parameters, the first historical data at least comprising one of the following: replenishment temperature, number of processed workpieces, actual conductivity before cleaning, actual conductivity after cleaning, target conductivity, time from cleaning to next workpiece feeding, and opening degree of the replenishment valve; deriving features and data from the first historical data respectively to determine first training data; training the first prediction model according to the first training data.
[0013] Optionally, the deriving features from the first historical data comprises: determining a conductivity difference before and after cleaning according to the actual conductivity before cleaning and the actual conductivity after cleaning; determining a distance target conductivity difference according to the target conductivity and the actual conductivity after cleaning; taking the conductivity difference before and after cleaning and the distance target conductivity difference as data features in the first historical data.
[0014] Optionally, the deriving data from the first historical data comprises: deriving a sample from the first historical data; determining a plurality of values of the opening degree of the replenishment valve based on an actual range of the opening degree of the replenishment valve; determining values of other data features in the first historical data corresponding to each value of the opening degree of the replenishment valve through the sample derivation model, and determining the values of the data features as the first training data.
[0015] Optionally, the model training of the first prediction model according to the first training data comprises: determining a first relationship between pure water replenishment flow and valve opening degree and valve opening time; determining a target function, the target function being used to represent a relationship between each data feature in the first training data and pure water replenishment flow; determining the valve opening degree and valve opening time under the minimum pure water replenishment flow, and corresponding pure water replenishment flow and conductivity based on the target function.
[0016] Optionally, the constructing of the prediction model of production parameters on cleaning liquid state further comprises: pre-completing the training of a second prediction model; collecting second historical data of related production parameters, and pre-processing the second historical data into second training data, the second historical data comprising a plurality of groups of corresponding pure water replenishment flow, steam amount, cleaning liquid water temperature, and steam valve opening degree; determining a second relationship between pure water replenishment flow and cleaning liquid water temperature, a third relationship between steam valve opening degree and steam amount, and a fourth relationship between steam amount and cleaning liquid water temperature according to the second training data; determining an influence relationship model of the steam valve opening degree on cleaning liquid water temperature at different pure water replenishment flow according to the second relationship, the third relationship and the fourth relationship.
[0017] In a second aspect, an embodiment of the present application provides an intelligent control system of a cleaning tank, comprising: a cleaning tank; a valve connected to the cleaning tank; a water temperature sensor for detecting a liquid temperature in the cleaning tank; a conductivity sensor for detecting a conductivity of the liquid in the cleaning tank to monitor a water quality state of the liquid in the cleaning tank; a control device, comprising: a processor configured to collect historical data of related production parameters and pre-process the historical data based on a state of cleaning liquid in the cleaning tank, determine an influence relationship of production parameters on the cleaning liquid state, and construct a prediction model of production parameters on the cleaning liquid state; and obtain a current state of the cleaning liquid in the cleaning tank, and if the current state of the cleaning liquid in the cleaning tank does not meet a preset requirement, determine an opening degree and / or opening time of the valve based on the prediction model completed by training, the influence relationship and the current state of the cleaning liquid.
[0018] Optionally, the state of the cleaning liquid in the cleaning tank comprises a water quality state and / or a water temperature state. The production parameters at least include at least one of the number of processed workpieces, the steam amount, and the pure water replenishment amount; the valve includes a replenishment valve for replenishing the cleaning tank with water and / or a steam valve for heating the cleaning liquid in the cleaning tank, and the opening degree and / or opening time of the valve include a first opening degree and a first opening time of the replenishment valve determined according to the influence relationship and the water quality state, and a second opening degree of the steam valve determined according to the influence relationship and the water temperature state.
[0019] Optionally, the application further comprises a plate heat exchanger, which is used to exchange the heat of the high-temperature water discharged from the cleaning tank to the incoming water, so as to improve the temperature of the incoming water during replenishment.
[0020] The embodiment of the application determines the influence relationship of the number of processed workpieces and the pure water replenishment on the water quality state of the cleaning liquid and the influence relationship of the steam amount and the pure water replenishment on the water temperature state of the cleaning liquid through the establishment of a prediction model and machine learning, so as to determine the optimal replenishment flow and steam amount under the current state of the cleaning liquid. The minimum replenishment flow and steam flow are controlled by adjusting the replenishment valve and the steam valve, the dynamic on-demand supply of pure water and steam is realized, the waste of pure water and steam is avoided, and the effect of energy saving is achieved. BRIEF DESCRIPTION OF DRAWINGS
[0021] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiments. Obviously, the drawings in the following description only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor.
[0022] Figure 1 The structure schematic diagram of the intelligent control system of the cleaning tank provided by the application is shown. Figure 2 The structure schematic diagram of the intelligent control system of a specific cleaning tank provided by the application is shown. Figure 3 The flow chart of the intelligent control method of the cleaning tank provided by the embodiment of the application is shown. DETAILED DESCRIPTION
[0023] In order to better understand the technical solutions of the application, the embodiments of the application will be described in detail below with reference to the drawings.
[0024] It should be clear that the described embodiments are only some of the embodiments of the application, not all. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.
[0025] As Figure 1 shown, a structure schematic diagram of an intelligent control system of a cleaning tank provided by an embodiment of the present application is shown. Referring to Figure 1 , the intelligent control system of the cleaning tank comprises a cleaning tank, a sensor 110, a control device 120 and a valve 130. The sensor 110 comprises an electric conductivity sensor 111 and a water temperature sensor 112, and the valve 130 comprises a water replenishing valve 131 and a steam valve 132, which are connected to the cleaning tank. The cleaning tank is not shown in the figure. Figure 1
[0026] The cleaning tank replenishes pure water through a water inlet pipeline and holds the pure water in the cleaning tank. After the pure water enters the cleaning tank, it is used as cleaning liquid to clean the workpieces in the cleaning tank. The cleaning tank discharges the cleaning liquid through an overflow pipeline.
[0027] The sensor 110 is arranged in the cleaning tank and is used to obtain the current state of the cleaning liquid in the cleaning tank. The state of the cleaning liquid in the cleaning tank includes the water quality state and the water temperature state.
[0028] In the embodiment of the present application, the sensor 110 is specifically implemented as the electric conductivity sensor 111 and the water temperature sensor 112. The electric conductivity sensor 111 is used to detect the current electric conductivity of the cleaning liquid to monitor the water quality state of the liquid in the cleaning tank, and the water temperature sensor 112 is used to detect the temperature of the liquid in the cleaning tank.
[0029] The control device 120 is used to collect historical data of relevant production parameters and perform preprocessing based on the state of the cleaning liquid in the cleaning tank. The influence relationship of the production parameters on the state of the cleaning liquid is determined through the historical data after preprocessing, and a prediction model of the production parameters on the state of the cleaning liquid is constructed.
[0030] The production parameters at least include the number of processed workpieces, and / or the steam amount, and / or the pure water replenishment.
[0031] The control device 120 is also used to obtain the current state of the cleaning liquid in the cleaning tank through the sensor 110. If the current state of the cleaning liquid in the cleaning tank does not meet the preset requirements, the opening degree and / or the opening time of the valve 130 are determined based on the trained prediction model according to the influence relationship and the current state of the cleaning liquid. A first opening instruction is sent to the water replenishing valve 131 according to the determined opening degree and the first opening time, and a second opening instruction is sent to the steam valve 132 according to the determined second opening degree.
[0032] The water replenishing valve 131 is arranged at the end of the water replenishing pipeline and is used to control the water replenishing flow and time, and the steam valve 132 is arranged at the end of the steam pipeline and is used to control the steam amount.
[0033] The opening degree of the valve 130, and / or the opening time includes the first opening degree and the first opening time of the water replenishment valve 131 determined by the control device 120 according to the influence relationship and the water quality state, and the second opening degree of the steam valve 132 determined by the control device 120 according to the influence relationship and the water temperature state.
[0034] The first opening degree and the first opening time of the water replenishment valve 131 are the minimum opening degree and the opening time of the water replenishment valve 131 to ensure that the conductivity of the cleaning liquid is always lower than the preset conductivity threshold; the second opening degree of the steam valve 132 is the minimum opening degree of the steam valve 132 to ensure that the temperature of the cleaning liquid is always within the preset temperature threshold.
[0035] The valve 130 is specifically used for opening according to the first opening instruction and the second opening instruction issued by the control device 120. The water replenishment valve 131 is opened according to the first opening degree and the first opening time in the first opening instruction to replenish water for the cleaning liquid in the cleaning tank; and the steam valve 132 is opened according to the second opening degree to heat the cleaning liquid in the cleaning tank.
[0036] Optionally, in the embodiment of the present application, the cleaning tank further includes a heat recovery device. In a specific implementation, the heat recovery device can be a plate heat exchanger.
[0037] The plate heat exchanger is connected with the water replenishment pipeline and the overflow water pipeline of the cleaning tank respectively, and is used for obtaining the recovered heat of the overflow water from the overflow water pipeline, and heating the pure water replenishment in the water replenishment pipeline through the recovered heat, so as to exchange the heat of the high-temperature water discharged from the cleaning tank to the incoming water, improve the temperature of the incoming water when replenishing water, and reduce the steam consumption of the cleaning tank when heating the pure water replenishment through the opening of the steam valve 132.
[0038] Based on the schematic diagram of the intelligent control system of the cleaning tank as shown in Figure 1 , a specific schematic diagram of the intelligent control system of the cleaning tank provided by the embodiment of the present application is shown in Figure 2 . Referring to Figure 2 , the water temperature sensor, the conductivity sensor and the control device are respectively arranged in the cleaning tank. Figure 2 The water temperature sensor, the conductivity sensor and the control device shown in Figure 1 are the water temperature sensor 112, the conductivity sensor 111 and the control device 120 shown in
[0039] The water replenishment valve 131 is arranged at the end of the water inlet pipeline of the cleaning tank, and the steam valve 132 is arranged at the end of the steam pipeline at the bottom of the cleaning tank. The control device 120 is in communication connection with the water temperature sensor 112, the conductivity sensor 111, the water replenishment valve 131 and the steam valve 132 respectively.
[0040] The conductivity sensor 111 and the water temperature sensor 112 are used to obtain the current state of the cleaning solution in the cleaning tank, i.e., to obtain the conductivity and the water temperature of the cleaning solution in the cleaning tank, respectively. The control device 120 determines the first opening degree and the first opening time of the water replenishing valve 131 and the second opening degree of the steam valve 132 based on the prediction model trained in advance according to the obtained conductivity and water temperature. The control device 120 controls the water replenishing valve 131 and the steam valve 132 to open correspondingly.
[0041] The plate heat exchanger arranged at the bottom of the cleaning tank is connected with the water replenishing pipeline and the overflow pipeline, respectively, to obtain the recovered heat of the overflow water from the overflow pipeline, so as to heat the pure water in the pure water pipeline by the recovered heat. In a specific embodiment, the pure water replenishment can be heated from 25℃ to 50℃ by recovering the heat of the 90℃ overflow water, thereby saving the amount of steam.
[0042] As shown in FIG. 1, a cleaning tank control method provided by an embodiment of the present application is provided. The method is applied to the cleaning tank shown in FIG. 1 and the control device 120 shown in FIG. 2. The specific steps of the method include: Figure 3 As shown in FIG. 1, a cleaning tank control method provided by an embodiment of the present application is provided. The method is applied to the cleaning tank shown in FIG. 1 and the control device 120 shown in FIG. 2. The specific steps of the method include: Figure 1 As shown in FIG. 1, a cleaning tank control method provided by an embodiment of the present application is provided. The method is applied to the cleaning tank shown in FIG. 1 and the control device 120 shown in FIG. 2. The specific steps of the method include: Figure 1 As shown in FIG. 1, a cleaning tank control method provided by an embodiment of the present application is provided. The method is applied to the cleaning tank shown in FIG. 1 and the control device 120 shown in FIG. 2. The specific steps of the method include: S301, based on the state of the cleaning solution in the cleaning tank, collecting historical data of relevant production parameters and preprocessing, determining the influence relationship of the production parameters on the state of the cleaning solution, and constructing a prediction model of the production parameters on the state of the cleaning solution.
[0043] Specifically, the production parameters at least include the number of processed workpieces, and / or the amount of steam, and / or the pure water replenishment. The state of the cleaning solution in the cleaning tank includes the water quality state and / or the water temperature state. The water quality state of the cleaning solution can be specifically represented by the conductivity of the cleaning solution.
[0044] After each cleaning of the workpieces by the cleaning tank, the data of the relevant production parameters are recorded as historical data for training and constructing the prediction model.
[0045] S302, obtaining the current state of the cleaning solution in the cleaning tank, and if the current state of the cleaning solution in the cleaning tank does not meet the preset requirement, determining the opening degree and / or the opening time of the valve 130 based on the trained prediction model according to the influence relationship and the current state of the cleaning solution.
[0046] Specifically, when the conductivity of the cleaning solution in the cleaning tank is too high, or the water temperature is too low, it is determined that the current state of the cleaning solution in the cleaning tank does not meet the preset requirement, and the water replenishment or heating is needed by opening the valve 130.
[0047] The valve 130 includes a water replenishment valve 131 for replenishing the cleaning liquid with pure water, and a steam valve 132 for heating the cleaning liquid. The opening degree and / or opening time of the valve includes a first opening degree and a first opening time of the water replenishment valve 131 determined by the prediction model based on the first target influence relationship and the water quality state, and a second opening degree of the steam valve 132 determined by the prediction model based on the second target influence relationship and the water temperature state.
[0048] The first opening degree and the first opening time of the water replenishment valve 131 are the minimum opening degree and the minimum opening time of the water replenishment valve 131 to ensure that the conductivity of the cleaning liquid is always lower than the preset conductivity threshold. The second opening degree of the steam valve 132 is the minimum opening degree of the steam valve 132 to ensure that the temperature of the cleaning liquid is always within the preset temperature threshold.
[0049] After the first opening degree, the second opening degree, and the first opening time are determined by the prediction model, a first opening instruction is sent to the water replenishment valve 131 according to the first opening degree and the first opening time to control the water replenishment valve 131 to open to replenish the cleaning liquid with pure water. A second opening instruction is sent to the steam valve 132 according to the second opening degree to control the steam valve 132 to open to heat the cleaning liquid.
[0050] The embodiment of the present application determines the influence relationship of the number of processed workpieces and pure water replenishment on the water quality state of the cleaning liquid, and the influence relationship of the steam amount and pure water replenishment on the water temperature state of the cleaning liquid by establishing a prediction model and performing machine learning in an offline state, thereby determining the optimal replenishment flow and steam amount of the cleaning liquid under the current state of the cleaning liquid. The replenishment flow and steam flow are controlled by adjusting the water replenishment valve 131 and the steam valve 132, which realizes dynamic on-demand supply of energy, avoids waste of pure water replenishment and steam, and also saves the heat caused by the temperature drop in the cleaning tank due to pure water replenishment.
[0051] Optionally, in the embodiment of the present application, when the opening degree and / or opening time of the valve 130 is determined based on the completed training prediction model according to the influence relationship and the current state of the cleaning liquid by performing S302, specifically, the first opening degree and the first opening time are determined by the first prediction model, and the second opening degree is determined by the second prediction model.
[0052] The first prediction model is a model for determining the first target influence relationship of the opening degree and the opening time of the water replenishment valve 131 on the conductivity of the cleaning liquid under the current number of processed workpieces. At this time, the current state of the cleaning liquid in the cleaning tank used for calculation is the current conductivity of the cleaning liquid.
[0053] The current number of processed workpieces is input into the first prediction model which has been trained in advance. The first prediction model determines the first opening degree and the first opening time of the water replenishment valve 131 according to the current conductivity of the cleaning solution and the first target influence relationship of the opening degree and the opening time of the water replenishment valve 131 on the conductivity of the cleaning solution.
[0054] The second prediction model is a model for determining the second target influence relationship of the opening degree of the steam valve 132 on the water temperature of the cleaning solution under the current pure water replenishment flow. At this time, the current state of the cleaning solution in the cleaning tank used for calculation is the current water temperature of the cleaning solution.
[0055] The current pure water replenishment flow is input into the second prediction model which has been trained in advance. The second prediction model determines the second opening degree of the steam valve 132 according to the current water temperature of the cleaning solution and the second target influence relationship of the opening degree of the steam valve 132 on the water temperature of the cleaning solution.
[0056] Optionally, in the embodiment of the present application, the training and construction of the prediction model are also needed to be completed by performing S301.
[0057] When training the first prediction model, the first historical data of related production parameters need to be collected, and the first training data is determined by performing feature derivation and data derivation on the first historical data, and the first prediction model is trained by using the first training data.
[0058] The data features in the first historical data at least include one of the water replenishment temperature, the number of processed workpieces, the actual conductivity before cleaning, the actual conductivity after cleaning, the target conductivity, the time from cleaning to the next workpiece feeding, and the opening degree of the water replenishment valve 131. The time from cleaning to the next workpiece feeding is the duration of the opening of the water replenishment valve 131 between two times of material cleaning.
[0059] The obtained first historical data is preprocessed, including missing value filling and abnormal value processing.
[0060] The feature derivation is a step of deriving other data features that can reflect the opening size of the water replenishment valve 131 according to the basic data features and combining the actual business requirements.
[0061] When performing feature derivation on the first historical data, the conductivity difference before and after cleaning is determined according to the actual conductivity before cleaning and the actual conductivity after cleaning. The distance target conductivity difference is determined according to the target conductivity and the actual conductivity after cleaning.
[0062] The conductivity difference before and after cleaning and the distance target conductivity difference are used as the derived data features in the first historical data.
[0063] The conductivity difference before and after cleaning is used to reflect the change rate between the reaction conductivity and the workpiece. The distance target conductivity difference is used to reflect the distance from the target conductivity. The smaller the distance target conductivity difference, the closer to the target value, the worse the water quality, and the more water supplement valve 131 needs to be opened to supplement more water.
[0064] The data derivation is a step of deriving more sample data through a derivation model, thereby enriching the data quantity, reducing the test cost, and improving the prediction model accuracy.
[0065] In the data derivation of the first historical data, the sample derivation model is trained by the first historical data. Based on the actual range of the water supplement valve 131 opening size, the values of the water supplement valve 131 opening size are determined. Through the sample derivation model, the values of each data feature in the first historical data corresponding to each water supplement valve 131 opening size value are determined, and the determined values of each data feature are determined as the first training data.
[0066] In the data derivation of the first historical data, the sample derivation model is trained by the first historical data. Based on the actual range of the water supplement valve 131 opening size, the values of the water supplement valve 131 opening size are determined. Through the sample derivation model, the values of each data feature in the first historical data corresponding to each water supplement valve 131 opening size value are determined, and the determined values of each data feature are determined as the first training data.
[0067] In the model training of the first prediction model according to the first training data, the first relationship between the pure water supplement flow and the valve 130 opening degree and the valve 130 opening time is determined. The target function is determined, which is used to represent the relationship between each data feature in the first training data and the pure water supplement flow. Based on the target function, the valve 130 opening degree and the valve 130 opening time under the minimum pure water supplement flow, and the corresponding pure water supplement flow and conductivity are determined. Generally, in the determination of the valve 130 opening degree, the valve 130 opening time, and the corresponding pure water flow and conductivity, the search space defined for the valve 130 opening degree and the valve 130 opening time is searched and determined by the optimization algorithm.
[0068] Specifically, in the determination of the pure water supplement flow, the relationship model q(v) of the flow and the valve 130 opening degree is established according to the physical characteristics or the past experimental data, thereby establishing the water consumption model y(v,t)=q(v)t, and determining the first relationship between the pure water supplement flow and the valve 130 opening degree and the valve 130 opening time. Wherein t is the valve 130 opening time.
[0069] The target function f(x,v,t)=y(v,t)=q(v) t, wherein x is each data feature in the first historical data, v is the valve 130 opening degree, and t is the valve 130 opening time.
[0070] The search space is defined, for example, the valve 130 opening degree v can be defined as [0, 100], the valve 130 opening time t is defined as [0, 100]. Based on the objective function, the particle swarm optimization algorithm (PSO) is repeatedly run to solve the boundary range of the search space. In the process of repeatedly running to solve, each function value of the objective function is evaluated, and the optimal solution is updated. The optimal solution of the valve 130 opening degree and the valve 130 opening time t is obtained, and the corresponding pure water replenishment flow and conductivity are obtained, and the training of the first model is completed.
[0071] When training the second prediction model, the second historical data of the related production parameters is collected, and the second historical data is preprocessed into second training data. The second training data includes several groups of corresponding pure water replenishment flow, steam quantity, cleaning liquid water temperature, and steam valve 132 opening degree.
[0072] According to the second training data, the second relationship between the pure water replenishment flow and the cleaning liquid water temperature, the third relationship between the steam valve 132 opening degree and the steam quantity, and the fourth relationship between the steam quantity and the cleaning liquid water temperature are determined.
[0073] According to the second relationship, the third relationship and the fourth relationship, the influence relationship model of the steam valve 132 opening degree on the cleaning liquid water temperature at different pure water replenishment flow is determined.
[0074] Optionally, in the embodiment of the present application, the heat is recovered from the overflow water of the cleaning liquid through the heat exchanger, and the recovered heat is used to heat the replenished pure water.
[0075] The embodiment of the present application recovers the heat in the overflow water which is originally discharged to heat the pure water replenishment by adding the heat exchanger as a heat recovery device, so as to reduce the steam quantity required when heating the cleaning liquid.
[0076] The above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order than the order in which they are recited and still achieve desirable results. In addition, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve the desired results. In certain implementations, multitasking and parallel processing can be advantageous.
[0077] In addition, the terms "first", "second", etc. are used herein only to describe different instances, and cannot be construed as indicating or implying relative importance or an indicated number of the technical features. Thus, the features defined with "first", "second", etc. can explicitly or implicitly include at least one of the features. In the description of the specification, the meaning of "a plurality of" is at least two, such as two, three, etc., unless otherwise specifically limited.
[0078] Any process or method descriptions or descriptions of the flow diagrams described herein or otherwise described herein can be understood as representing modules, segments, or portions of code that include one or more executable instructions for implementing specific logical functions or steps in the process, and the preferred embodiments of the specification include additional implementations in which the order of steps can be executed differently, including substantially simultaneously or in reverse order, depending on the functionality involved, as will be understood by those skilled in the art of the embodiments of the specification.
[0079] Depending on the context, the word "if" as used herein can be interpreted to mean "when" or "while" or "in response to determining" or "in response to detecting". Similarly, the phrase "if it is determined" or "if [a stated condition or event] is detected" can be interpreted to mean "upon determining" or "in response to determining" or "upon detecting [the stated condition or event]" or "in response to detecting [the stated condition or event]".
[0080] It should be noted that the terminal involved in the embodiments of the specification can include, but is not limited to, a personal computer (Personal Computer; hereinafter referred to as PC), a personal digital assistant (Personal Digital Assistant; hereinafter referred to as PDA), a wireless handheld device, a tablet computer, a mobile phone, an MP3 player, an MP4 player, etc.
[0081] In the embodiments provided in the specification, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division, and actual implementation can have another division manner. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or other forms.
[0082] In addition, the various functional units in the various embodiments of the present specification can be integrated into one processing unit, or each unit can exist physically, or two or more units can be integrated into one unit. The integrated unit can be implemented in the form of hardware, or in the form of hardware plus software function units.
[0083] The integrated unit implemented in the form of software function units can be stored in a computer readable storage medium. The software function unit stored in the storage medium includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute part of the steps of the method described in the various embodiments of the present specification.
[0084] The above only describes the preferred embodiments of the present specification and is not intended to limit the present specification. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present specification shall be included in the protection scope of the present specification.
Claims
1. A method for intelligent control of a cleaning tank, characterized in that, The method comprises the following steps: Based on the state of the cleaning liquid in the cleaning tank, collecting historical data of relevant production parameters and preprocessing, determining the influence relationship of production parameters on the state of the cleaning liquid, and constructing a prediction model of production parameters on the state of the cleaning liquid; Obtain the current state of the cleaning liquid in the cleaning tank, if the current state of the cleaning liquid in the cleaning tank does not meet the preset requirements, based on the trained prediction model, according to the influence relationship and the current state of the cleaning liquid, determine the opening degree and / or opening time of the valve connected with the cleaning tank.
2. The method of claim 1, wherein, Including: The state of the cleaning liquid in the cleaning tank includes water quality state and / or water temperature state; The production parameters at least include at least one of the number of processed workpieces, steam volume, and pure water replenishment volume; the valve includes a water replenishment valve for replenishing water to the cleaning tank and / or a steam valve for heating the cleaning liquid in the cleaning tank, and the opening degree and / or opening time of the valve include the first opening degree and the first opening time of the water replenishment valve determined according to the first target influence relationship and the water quality state, and the second opening degree of the steam valve determined according to the second target influence relationship and the water temperature state; According to the first opening degree and the first opening time, a first opening instruction is sent to the water replenishment valve to control the water replenishment valve to open and replenish water to the cleaning liquid, and according to the second opening degree, a second opening instruction is sent to the steam valve to control the steam valve to open and heat the cleaning liquid.
3. The method of claim 2, wherein, It also includes real-time monitoring the conductivity of the cleaning liquid in the cleaning tank to obtain the water quality state of the cleaning liquid; The first opening degree and the first opening time of the water replenishment valve are the opening degree and the opening time of the minimum water consumption to ensure that the conductivity of the cleaning liquid is always lower than the preset conductivity threshold; The second opening degree of the steam valve is the opening degree of the minimum steam volume to ensure that the temperature of the cleaning liquid is always within the preset temperature threshold.
4. The method of claim 3, wherein, The prediction model includes a first prediction model, which is a model for determining the first target influence relationship of the opening degree and the opening time of the water replenishment valve on the conductivity of the cleaning liquid under the current number of processed workpieces; Based on the trained prediction model, according to the influence relationship and the current state of the cleaning liquid, the opening degree and / or opening time of the valve are determined, which includes: Input the current number of processed workpieces into the first prediction model which has been trained in advance, and determine the first opening degree and the first opening time of the water replenishment valve through the first prediction model; Wherein, the first prediction model is used to determine the first opening degree and the first opening time of the water replenishment valve according to the current conductivity of the cleaning liquid and the first target influence relationship of the opening degree and the opening time of the water replenishment valve on the conductivity of the cleaning liquid.
5. The method of claim 3, wherein, The obtaining the current state of the cleaning liquid in the cleaning tank comprises obtaining a current water temperature of the cleaning liquid, and the prediction model further comprises a second prediction model, which is a model for determining a second target influence relationship between the opening degree of the steam valve and the water temperature of the cleaning liquid under a current pure water replenishment flow rate; The determining the opening degree and / or opening time of the valve based on the prediction model trained comprises: inputting the pure water replenishment flow rate into the second prediction model trained in advance, and determining the second opening degree of the steam valve through the second prediction model; The second prediction model is used to determine the second opening degree of the steam valve according to the current water temperature of the cleaning liquid and the second target influence relationship between the opening degree of the steam valve and the water temperature of the cleaning liquid.
6. The method of claim 4, wherein, The constructing the prediction model of the production parameters on the state of the cleaning liquid further comprises: pre-training the first prediction model; The first historical data of the related production parameters comprises at least one of the following: replenishment temperature, number of processed workpieces, actual conductivity before cleaning, actual conductivity after cleaning, target conductivity, time from cleaning to next workpiece feeding, and opening degree of the replenishment valve. The first training data is determined by performing feature derivation and data derivation on the first historical data, respectively. The first prediction model is trained according to the first training data.
7. The method of claim 6, wherein, The feature derivation on the first historical data comprises: determining the difference between the actual conductivity before cleaning and the actual conductivity after cleaning according to the actual conductivity before cleaning and the actual conductivity after cleaning; determining the difference from the target conductivity according to the target conductivity and the actual conductivity after cleaning; the difference between the actual conductivity before cleaning and the actual conductivity after cleaning and the difference from the target conductivity are used as data features in the first historical data.
8. The method of claim 7, wherein, The data derivation on the first historical data comprises: deriving a sample derivation model through the first historical data; determining a plurality of values of the opening degree of the replenishment valve based on the actual range of the opening degree of the replenishment valve; determining the values of the other data features in the first historical data corresponding to each value of the opening degree of the replenishment valve through the sample derivation model, and determining the values of the data features as the first training data.
9. The method of claim 6, wherein, The model training of the first prediction model according to the first training data comprises: determining a first relationship between the pure water replenishment flow rate and the opening degree and opening time of the valve; determining a target function, which is used to represent the relationship between each data feature in the first training data and the pure water replenishment flow rate; determining the opening degree and opening time of the valve under the minimum pure water replenishment flow rate, and the corresponding pure water replenishment flow rate and conductivity based on the target function.
10. The method of claim 5, wherein, The constructing the prediction model of the production parameters on the state of the cleaning liquid further comprises: pre-training the second prediction model; collecting second historical data of related production parameters, and preprocessing the second historical data into second training data, the second historical data including a plurality of sets of corresponding pure water replenishment flow, steam amount, cleaning liquid water temperature, and steam valve opening degree; determining a second relationship between pure water replenishment flow and cleaning liquid water temperature, a third relationship between steam valve opening degree and steam amount, and a fourth relationship between steam amount and cleaning liquid water temperature, respectively, according to the second training data; determining an influence relationship model of the steam valve opening degree on the cleaning liquid water temperature at different pure water replenishment flows, according to the second relationship, the third relationship, and the fourth relationship.
11. An intelligent control system for a cleaning tank, characterized in that, comprise: a cleaning tank; a valve connected to the cleaning tank; a water temperature sensor for detecting the temperature of liquid in the cleaning tank; a conductivity sensor for detecting the conductivity of liquid in the cleaning tank to monitor the water quality state of the liquid in the cleaning tank; a control device comprising: a processor configured to collect historical data of related production parameters and preprocess the historical data based on the state of cleaning liquid in the cleaning tank, determine an influence relationship of production parameters on the state of cleaning liquid, and construct a prediction model of production parameters on the state of cleaning liquid; and obtain the current state of cleaning liquid in the cleaning tank, and if the current state of cleaning liquid in the cleaning tank does not meet the preset requirements, determine the opening degree and / or opening time of the valve based on the trained prediction model, the influence relationship, and the current state of the cleaning liquid.
12. The system of claim 11, wherein, the state of cleaning liquid in the cleaning tank includes water quality state and / or water temperature state; the production parameters at least include at least one of the number of processed workpieces, steam amount, and pure water replenishment amount; the valve includes a water replenishment valve for replenishing water to the cleaning tank and / or a steam valve for heating cleaning liquid in the cleaning tank, and the opening degree and / or opening time of the valve includes a first opening degree and a first opening time of the water replenishment valve determined according to the influence relationship and the water quality state, and a second opening degree of the steam valve determined according to the influence relationship and the water temperature state.
13. The system of claim 11 or 12, wherein: Further comprising a plate heat exchanger for exchanging heat of high-temperature water discharged from the cleaning tank to incoming water to improve the temperature of incoming water when replenishing water.
Citation Information
Cited By
Adaptive control method and system for pure water preparation equipment
CN122501944A