Parameter tuning method and glue spraying apparatus

Abstract

The application discloses a parameter tuning method and a glue spraying device. The parameter tuning method can be applied to a device, and parameters of the device include input parameters, output values and target values. The parameter tuning method comprises the following steps: obtaining a mathematical model of the input parameters and the output values of the device; obtaining the target values; obtaining the output values; if an absolute value of a difference between the output values and the target values is greater than a first threshold value, adjusting the input parameters according to the mathematical model, so that an absolute value of a difference between the adjusted output values and the target values is reduced. The application can automatically adjust the input parameters according to the mathematical model of the input parameters and the output values of the device, not only improves the intelligent level of the device, but also reduces the frequency of manual operation and saves the cost.

Description

Technical Field

[0001] This invention relates to a parameter optimization method and a glue spraying device. Background Technology

[0002] For devices with Internet of Things (IoT) capabilities (such as, but not limited to, adhesive spraying equipment), after a period of operation, due to aging and system deviations, it is usually necessary to adjust the device's input parameters. These devices commonly use Statistical Process Control (SPC) rules, triggering parameter adjustments through time, counts, events, signals, alarms, and other triggering methods.

[0003] Current parameter tuning methods mostly rely on manual adjustments, which depend on expert experience and require system downtime. If the equipment's input parameters are complex, further testing is needed, leading to even longer downtime and greater productivity losses. Summary of the Invention

[0004] The purpose of this invention is to provide a parameter optimization method and a glue spraying device that can automatically adjust parameters and effectively solve at least one defect of the prior art.

[0005] To achieve the above objectives, the present invention provides a parameter tuning method applied to a device, wherein the parameters of the device include input parameters, output values, and target values. The parameter tuning method includes: obtaining a mathematical model of the input parameters and the output values ​​of the device; obtaining the target value; obtaining the output value; and if the absolute value of the difference between the output value and the target value is greater than a first threshold, adjusting the input parameters according to the mathematical model to reduce the absolute value of the difference between the adjusted output value and the target value.

[0006] The present invention also provides a glue spraying device, including a control unit, wherein the control unit is used to execute the parameter optimization method described above, wherein the input parameters include temperature, air pressure and glue spraying height, the output value includes glue spraying thickness, and the target value includes standard glue thickness.

[0007] According to one aspect of the present invention, the present invention can automatically adjust the input parameters of the device based on a mathematical model, which not only improves the intelligence level of the device, but also reduces the frequency of manual operation and saves costs.

[0008] According to another aspect of the present invention, the present invention can also issue a fault alert when parameters cannot be adjusted, notify personnel to wait in advance, and realize fault prediction and alert.

[0009] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be apparent from the description, or may be learned by practice of the invention. Attached Figure Description

[0010] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0011] Figure 1 This is a flowchart of the parameter tuning method of the present invention;

[0012] Figure 2 This is a schematic flowchart illustrating the specific operation of the parameter tuning method of the present invention;

[0013] Figure 3 yes Figure 2 An exemplary flowchart for obtaining the output value and calculating the average deviation value in operations 223 and 224, and determining whether the deviation value is too large;

[0014] Figure 4 yes Figure 2 An exemplary flowchart for obtaining new input parameters is shown in operation 225;

[0015] Figure 5 yes Figure 2 An exemplary flowchart for implementing the input of new input parameters to the device in operation 226;

[0016] Figure 6 This is an exemplary flowchart of the present invention implementing dynamic optimization on a single device based on a simple mathematical model (e.g., a polynomial);

[0017] Figure 7 This is a schematic diagram illustrating the changes in output values ​​after using parameter tuning methods;

[0018] Figure 8 This is a schematic diagram illustrating the change of the controllable input parameter x1 in the polynomial;

[0019] Figure 9 This is a schematic diagram illustrating the changes in output values ​​without using parameter tuning methods. Detailed Implementation

[0020] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that the invention will be thorough and complete, and the concept of the exemplary embodiments will be fully conveyed to those skilled in the art. The same reference numerals in the drawings denote the same or similar structures, and therefore their detailed description will be omitted.

[0021] In describing the elements / components / etc. described and / or illustrated herein, the terms “a,” “an,” “the,” “the,” and “at least one” are used to indicate the presence of one or more elements / components / etc. The terms “comprising,” “including,” and “having” are used to indicate an open-ended inclusion and to mean that additional elements / components / etc. may exist in addition to those listed. Furthermore, the terms “first,” “second,” etc., in the claims are used only as designations and are not intended to limit the number of objects to which they pertain.

[0022] like Figure 1 The diagram illustrates the flow of a parameter tuning method 100 provided by the present invention. The parameter tuning method 100 of the present invention can be applied to a device, whose parameters may include input parameters, output values, and target values. The parameter tuning method 100 may include the following steps:

[0023] Step S101: Obtain the mathematical model of the device's input parameters and output values.

[0024] Step S102: Obtain the target value.

[0025] Step S103: Obtain the output value of the device.

[0026] Step S104: If the absolute value of the difference between the output value and the target value is greater than the first threshold, then adjust the input parameters according to the mathematical model so that the absolute value of the difference between the adjusted output value and the target value is reduced.

[0027] In this invention, the device is preferably a device with Internet of Things (IoT) capabilities, such as, but not limited to, a glue spraying device. Furthermore, the device parameters can be read or set digitally. In this invention, the mathematical model can be established using methods such as linear regression analysis or the Analytic Hierarchy Process (AHP), but this invention is not limited thereto.

[0028] In this invention, the input parameters of the device may include controllable input parameters and uncontrollable input parameters. For controllable input parameters, the device of this invention can control and / or modify them through methods including but not limited to parameter modification commands. For uncontrollable input parameters, the device of this invention cannot control and / or modify them through methods such as parameter modification commands.

[0029] In this invention, the output value of the device can be obtained through methods including but not limited to, listening.

[0030] In this invention, the so-called "target value" is a result or standard value that is related to the output value of the device and is preset or expected to be achieved.

[0031] In some embodiments of the present invention, the step of obtaining the output value of the device (i.e., step S103) may include: obtaining the real-time output parameters of the device periodically; and the output value of the device is the average value of multiple real-time output parameters. The period can be set according to actual needs. Preferably, the output value of the device may be, for example, the average value of ten consecutive real-time output parameters. By using the average value calculation to obtain the output value, extreme values ​​can be effectively filtered out. However, it is understood that the method of obtaining the output value of the device in the present invention is not limited to the above method; for example, the output value can also be calculated and obtained based on other numbers of real-time output parameters, and these are not intended to limit the present invention.

[0032] In some embodiments of the present invention, the input parameters of the device may include a first input parameter. The first input parameter is a controllable input parameter, wherein the step of adjusting the input parameter (i.e., step S104) may include: obtaining the coefficients of the first input parameter in the mathematical model; obtaining an adjustment value based on the absolute value and the coefficients; and adjusting the first input parameter based on the adjustment value.

[0033] Furthermore, the first input parameter has an upper bound and a lower bound. The parameter tuning method further includes comparing the adjusted first input parameter with the upper bound or the lower bound.

[0034] If the adjusted first input parameter is greater than or equal to the upper bound value, then the first input parameter is adjusted according to the upper bound value;

[0035] If the adjusted first input parameter is less than or equal to the lower bound value, then the first input parameter is adjusted according to the lower bound value;

[0036] If the adjusted first input parameter is less than the upper bound and greater than the lower bound, then the first input parameter is adjusted according to the adjustment value.

[0037] In some embodiments of the present invention, such as Figure 1 As shown, the parameter tuning method 100 of the present invention may further include: step S105, determining the parameter adjustment result. Step S105 may include: recording the moment when the adjustment of the input parameter is first performed, and using this as the starting moment; if the absolute value is greater than the first threshold within a first time period from the starting moment, the parameter adjustment fails; if the absolute value is less than or equal to the first threshold within the first time period, the parameter adjustment succeeds.

[0038] In some embodiments of the present invention, such as Figure 1 As shown, the parameter optimization method 100 of the present invention may further include: step S106, adjusting the mathematical model according to the parameter adjustment results.

[0039] In some embodiments of the present invention, the parameter tuning method of the present invention may further include: if the absolute value of the difference between the output value and the target value is greater than a second threshold, then a warning message is issued.

[0040] The present invention also provides a glue spraying device, which may include a control unit, the control unit being used to execute the parameter optimization method of the present invention as described above. Preferably, the input parameters of the glue spraying device may include temperature, air pressure, and glue spraying height; the output value may include glue spraying thickness; and the target value may include standard glue thickness.

[0041] In some embodiments of the present invention, the adhesive spraying device of the present invention may store programs and / or instruction codes, and the control unit may execute the parameter tuning method of the present invention by reading the programs and / or instruction codes.

[0042] like Figure 2 As shown, it illustrates the specific operational procedure for parameter optimization of the device according to the present invention. Figure 2 In the diagram, operations 221-229 illustrate the specific process by which the present invention dynamically optimizes the input parameters of the device using a mathematical model.

[0043] refer to Figure 2 Operations 221-227, in conjunction with references Figures 3-5 The following details the specific process by which the present invention uses a mathematical model to dynamically optimize the input parameters of the device.

[0044] In operation 221, dynamic tuning begins, which means the parameter tuning method is executed.

[0045] In operation 222, the mathematical model of the equipment is obtained. The method of obtaining the mathematical model is not limited; for example, it can be obtained by combining modeling tools to analyze a large number of pre-acquired input and output parameters of the equipment, thereby obtaining the relevant mathematical model of the input and output parameters.

[0046] In operation 223, KPOV (Key Parameter Output Value) is collected in real time, and the average deviation value is calculated. KPOV is the real-time output parameter of the device. In some embodiments, the real-time output parameter is regarded as the output value of the device. The real-time acquired KPOV is compared with the target value to obtain the difference between the output value and the target value, that is, the deviation value. In other embodiments, the output value of the device is the average value of multiple output parameters, that is, the average value of multiple KPOVs is regarded as the output value of the device. In this case, the output value is compared with the target value to obtain the average deviation value between the output value and the target value.

[0047] In operation 224, determine if the deviation value is too large. If yes, proceed to operation 225; otherwise, return and continue with operation 223. (Refer to the reference.) Figure 3 An exemplary process for the present invention to realize the real-time collection of KPOV and the calculation of the average deviation value, as well as the determination of whether the deviation value is too large (i.e., operations 223 and 224), includes the following operations: In operation 2231, start the parameter monitoring process. In operation 2232, obtain the output parameters (e.g., KPOV) reported by the device. In operation 2233, determine whether more than the number of moving average reference points (i.e., N, N can be, for example, 10, but the present invention is not limited thereto) of data has been received. If yes, execute operation 2234 to calculate the moving average of the latest N points of KPOV; if no, return and continue to execute operation 2232. In operation 2235, determine whether the moving average of the latest N points of KPOV exceeds the allowable range of the target value. If yes, execute operation 2236 to start the process of obtaining the KPIV (Key Parameter Input Value) adjustment value; if no, return and continue to execute operation 2232. The present invention monitors whether the output parameters have undergone systematic deviation by observing the moving average, that is, whether the moving average exceeds a first threshold. For example, assuming the monitored output parameter is Y and the number of reference points is N, when the data of N real-time output parameter Y is obtained, the moving average can be calculated. If the moving average exceeds the allowable range of the target value (that is, the target value ± the first threshold), it is considered that a systematic deviation has occurred.

[0048] In operation 225, the deviation value is input into the mathematical model to obtain a new KPIV. In some embodiments, the input parameters include a first input parameter, which refers to a controllable input parameter, i.e., a parameter that can be actively adjusted by the user, and can be called a KPIV (Key Parameter Input Value). The new KPIV refers to the adjusted first input parameter. (Refer to the reference...) Figure 4An exemplary process for obtaining a new KPIV according to the present invention includes the following operations: In operation 2251, the process of obtaining the KPIV adjustment value begins. In operation 2252, the deviation between the KPIV moving average and the target value is calculated. In operation 2253, the tangent slope of the optimal target point of the mathematical model is obtained using an algorithm. Preferably, in some embodiments, an algorithm such as gradient descent can be used to search for the optimal target point of the mathematical model, thereby obtaining the tangent slope corresponding to the KPIV, which is the coefficient of the KPIV in the mathematical model. The method of obtaining the tangent slope will be further illustrated in detail below. In operation 2254, the adjustment value of each controllable input parameter can be obtained from the slope and the deviation value. That is, when multiple first input parameters (i.e., controllable input parameters) are included, according to the mathematical model, each first input parameter has its corresponding slope, thereby obtaining the corresponding adjustment value. In operation 2255, the current value of the controllable input parameter plus the adjustment value yields the new KPIV, i.e., the adjusted first input parameter. In operation 2256, determine whether the new KPIV exceeds its upper or lower bound. If yes, execute operations 2257-2259 sequentially; otherwise, directly execute operation 2259. Specifically, in operation 2257, the upper or lower bound is used as the new KPIV. In operation 2258, an alarm is issued indicating that the KPIV is at its upper or lower bound. In operation 2259, the new KPIV is obtained, and the device input parameter modification process begins (e.g., entering...). Figure 2 Operation 226 in the middle.

[0049] In operation 226, input the new KPIV. After obtaining the new KPIV through the aforementioned process, the corresponding first input parameter of the device can be corrected through the device's input parameter modification process. (Refer to the reference.) Figure 5This illustrates an exemplary process for implementing a new KPIV input according to the present invention. In operation 22601, the device input parameter modification process begins. In operation 22602, tuning process message listening begins. When the tuning process is triggered, in operation 22603, a tuning instruction is sent to the device's data monitoring module. In operation 22604, the device's data monitoring module continuously listens for tuning instructions. When a relevant tuning instruction is detected, in operation 22605, the device's data monitoring module modifies the device's input parameters accordingly. After modification, in operation 22606, the device's data monitoring module can retrieve the modified input parameters. Furthermore, in operation 22607, it is determined whether the modification was successful, that is, whether the absolute value of the difference between the tuned output value and the target value is less than or equal to a first threshold. If yes, then operation 22608 is executed, sending the tuning result (e.g., sending a result indicating "tuning successful") to the tuning system. If no, then operation 22609 is further executed to determine whether the modification timed out. In some embodiments, a method for determining whether optimization is successful can be implemented, for example, by setting a first time when a new KPIV is input, which is recorded as the start time. The optimization time is recorded from this start time. If the optimization time is less than the first time, and the absolute value of the difference between the device output value and the target value is still greater than a first threshold, operation 22605 is returned and executed again to modify the device's input parameters. If the optimization time is greater than or equal to the first time, and the absolute value of the difference between the device output value and the target value is still greater than the first threshold, the optimization fails, the modification is unsuccessful, and operation 22608 is executed, sending the optimization result (e.g., sending a result indicating "optimization failed") to the optimization system. If the optimization time is less than the first time, and the absolute value of the difference between the device output value and the target value decreases to less than the first threshold, the optimization is successful, and operation 22608 is executed, sending the optimization result (e.g., sending a result indicating "optimization successful") to the optimization system. In some embodiments, after sending the optimization result, the data monitoring module continues to execute operation 22604, that is, continues to continuously listen for optimization commands.

[0050] Operation 227 ends dynamic tuning.

[0051] Continue to refer to Figure 5 Preferably, in some embodiments, the tuning system further performs the following operations: In operation 22610, receiving the tuning result; In operation 22611, interpreting / verifying the tuning result; In operation 22612, reporting the tuning result; In operation 22613, ending the device input parameter modification process.

[0052] The following will combine Figures 6-9 An exemplary process for implementing the dynamic optimization of the present invention is given as an example of implementing the present invention on a single device based on a simple mathematical model (e.g., a polynomial).

[0053] like Figure 6 As shown, the present invention can be implemented, for example, by configuring a dynamic tuning module (e.g., as shown in the figure). Figure 5 The "data monitoring module" in the present invention executes the dynamic tuning method of the present invention, thereby enabling the output of new KPIV (i.e., New KPIV) to the device for device control, such as including but not limited to modifying the input parameters of the device.

[0054] The dynamic optimization module can be used to obtain a mathematical model, such as a polynomial that reflects the relationship between the input parameters and output values ​​of the device. The input parameters of the device may include controllable input parameters and uncontrollable input parameters, and the output values ​​of the device may be, for example, a moving average of multiple real-time output parameters.

[0055] The dynamic tuning module can also be used to configure mathematical models and tuning strategies, including but not limited to the configuration of target values, tolerance values, moving average reference points, and warning upper and lower limits for output parameters.

[0056] The dynamic tuning module can also be used to monitor device data, including but not limited to obtaining real-time output parameters of the device (such as KPOV), and performing operations including but not limited to calculating the moving average of real-time output parameters, judging the trigger conditions of the dynamic tuning process, and using the model to solve when dynamic tuning is triggered to obtain a new KPIV.

[0057] The dynamic optimization method of this invention will be explained in more detail below using a simple mathematical model as an example. For example, taking a mathematical model that reflects the relationship between the input parameters and output values ​​of a device as an example, the relationship is as follows:

[0058] y=k1*x1 + k2*x1*x2 + k3*x3 (Formula 1)

[0059] Where y represents the output parameter of the device; x1, x2, and x3 are all input parameters of the device, where x1 is, for example, a controllable input parameter (i.e., modifiable, such as the "first input parameter"), while x2 and x3 are, for example, uncontrollable input parameters (i.e., unmodifiable). For the first input parameter x1, according to mathematical model 1, its corresponding coefficient m = k1 + k2 * x2 can be obtained, where the output value (KPOV) is the moving average of N output parameters.

[0060] For example, the target value of y can be configured as "368", the first threshold is set to "±0.3" (i.e., the allowable error, the upper limit of the target error is "368.3", and the lower limit of the target error is "367.7"); the second threshold is set to 1, that is, the upper limit of the warning for y is configured as "369", and the lower limit of the warning is "367", that is, when the output value exceeds the warning range (i.e., "367~369"), an alarm can be issued; the number of moving average reference points N can be configured to 10, that is, the offset trend of KPOV can be determined by the average value of the real-time output parameters of 10 points (which can be combined with reference). Figure 7 The gray line represents the "moving average," and the average of the 10 real-time output parameters is considered the device's output value. Dynamic optimization is initiated when the gray line exceeds the target value's allowable range (e.g., greater than the upper target error bound or less than the lower target error bound), meaning the absolute value of the difference between the output value and the target value exceeds a first threshold. For example, the following preset values ​​can be used: k1 = 1.5; k2 = 0.008; k3 = -4; x1 = 160; x2 = 225; x3 = 40. Therefore, the coefficient m of the first input parameter x1 can be obtained as m = k1 + k2 * x2 = 3.3.

[0061] like Figure 7 As shown, it illustrates the change in the output value KPOV after dynamic tuning. Figure 7 It can be seen that at multiple times, such as t = 45, 81, ..., 160, ... (in seconds), the output value KPOV exceeded the allowable range of the target value. For example, it exceeded the upper limit of the target error (e.g., at t1 = 45s) or the lower limit of the target error (e.g., at t2 = 81s). That is, the absolute value of the difference between the output value and the target value was greater than the first threshold. These times correspond to the tuning trigger times. In other words, at these times, the action of adjusting the input parameters can be triggered to perform dynamic tuning, such as... Figure 2 Operations 225-227, etc., are examples of this. More specifically, for example, it can be achieved by executing... Figure 4 Operations 2251-2259 and Figure 5 The operation 22601 to 22613 and other operations in the process can modify the first input parameter x1 of the device.

[0062] like Figure 8 As shown, this illustrates the variation of the controllable input parameter x1. After dynamic optimization, the controllable input parameter x1 can be adjusted (or modified).

[0063] More specifically, with Figure 7Taking time t=160s as an example, the calculated moving average of 10 points is 367.68. The absolute value of the difference between this value and the target value (368) is 0.32 (greater than the first threshold "0.3"), which exceeds the allowable range of the target value, thus triggering dynamic optimization. The adjustment parameter for x1 can be calculated based on the absolute value of the difference between the output value and the target value (e.g., the absolute value of the difference between KPOV and the target value, i.e., the average deviation value), i.e., Δx = |367.68-368| / 3.3 = 0.096. Furthermore, the adjusted first input parameter (i.e., the new KPIV) can be obtained by adding the adjustment value to the current value of x1. For example, the new x1 (represented by x1_new) is: x1_new = 160.178 + 0.096 = 160.274, and x1 can be adjusted accordingly (i.e., x1 is modified). For example, as... Figure 8 As shown in the figure, x1 is adjusted from 160.178 to 160.274 at time t = 160s.

[0064] In some embodiments, the controllable input parameter x1 has an upper bound and a lower bound. When the adjusted first input parameter is greater than or equal to the upper bound, the controllable input parameter x1 is adjusted according to the upper bound (e.g., the upper bound is used as a new KPIV to modify the controllable input parameter x1); when the adjusted first input parameter is less than or equal to the lower bound, the controllable input parameter x1 is adjusted according to the lower bound (e.g., the lower bound is used as a new KPIV to modify x1). More specifically, with Figure 7 For example, when x1 undergoes dynamic optimization at t = 160s, the calculated x1 is 160.274. If this value is less than the upper bound of x1 but greater than the lower bound, then this calculated x1 (x1_new = 160.274) is used for parameter adjustment. Conversely, if the new x1 obtained after calculation is greater than the upper bound of x1, then the upper bound of x1 is used as the new KPIV to modify x1.

[0065] exist Figure 7 In this context, the first tuning trigger occurs at t1 = 45s, and this time can be considered the moment when the input parameter adjustment action is first executed. In some embodiments, in conjunction with reference... Figure 1 In the parameter tuning method 100 of the present invention, step S105 may further include: recording the moment when the first action of adjusting the output parameter is performed, and using that moment as the starting moment (e.g. Figure 7In the first time interval (t1 = 45s), if the absolute value is greater than the first threshold, the parameter adjustment fails; if the absolute value is less than or equal to the first threshold, the parameter adjustment succeeds. For example, if the first time interval is 10s, that is, if the absolute value is still greater than the first threshold at t = 55s, meaning the output value exceeds the upper or lower bound of the target error, the parameter adjustment fails; if the absolute value is less than or equal to the first threshold, meaning the output value is within the upper or lower bound of the target error, the parameter adjustment succeeds.

[0066] In some embodiments, if the absolute value of the difference between the output value and the target value is greater than a second threshold, the device may also issue a warning message. For example, in Figure 7 In the process, when the gray line representing the "moving average" exceeds the warning range (for example, if it is greater than the upper warning limit "369" or less than the lower warning limit "367", then the second threshold is, for example, "1"), it means that the absolute value of the difference between the output value and the target value is greater than the second threshold. At this time, the device can issue a warning message, such as a fault reminder indicating that "parameters cannot be adjusted", to notify personnel to wait in advance, thereby realizing fault prediction and reminder.

[0067] Depend on Figure 7 It can also be seen that after dynamic optimization (i.e., the controllable input parameter x1 is adjusted or modified), KPOV is adjusted to be close to the target value.

[0068] like Figure 9 As shown, it illustrates the change in KPOV (i.e., the y-value) without dynamic tuning. Figure 9 It can be seen that when dynamic optimization is not implemented, the KPOV fluctuates significantly, even exceeding the warning range. (Comparison) Figure 7 and Figure 9 It is evident that by implementing the dynamic optimization of this invention, the KPOV of the device can be effectively adjusted to near the target value.

[0069] As described above, in this embodiment, the controllable input parameter x1 can be adjusted through a dynamic tuning process. For example, the coefficient (e.g., slope m) of the controllable input parameter x1 (i.e., the first input parameter) can be obtained, and an adjustment value for the controllable input parameter x1 can be obtained based on the absolute value of the difference between the output value and the target value (e.g., the absolute value of the difference between the moving average of KPOV and the target value, i.e., the average deviation value) and the coefficient (e.g., slope m). The controllable input parameter x1 can then be adjusted according to this adjustment value. Of course, it is understood that for other embodiments of a mathematical model, such as those where the input parameters include two or more controllable input parameters, each controllable input parameter can be dynamically tuned using the dynamic tuning method described above, according to the priority of the multiple controllable input parameters.

[0070] The present invention also provides a glue spraying device, which includes a control unit. The input parameters of the glue spraying device include, for example, temperature, air pressure, and spraying height; the output values ​​of the glue spraying device include, for example, spraying thickness; and the target values ​​of the glue spraying device include, for example, standard glue thickness. The temperature, air pressure, and spraying height of the glue spraying device are represented by x1, x2, and x3, respectively, and the spraying thickness of the glue spraying device is represented by y.

[0071] Before adjusting the equipment parameters, a mathematical model was first obtained based on simulation to correlate the adhesive thickness y with temperature x1 and air pressure x2 with adhesive height x3:

[0072] y=c3×(c1×x1+c2×x2)+c4×(c1×x1+c2×x2)×x3.

[0073] After obtaining the mathematical model of the aforementioned glue spraying equipment, the control unit of the glue spraying equipment, through the parameter optimization method provided by this invention, can adjust controllable input parameters such as temperature, air pressure, and glue spraying height according to the glue spraying thickness y and the standard thickness of the glue, thereby ensuring that the glue spraying thickness y remains near the standard thickness of the glue. Of course, it is understood that in some other embodiments, for example, only one of temperature and air pressure may be considered as a controllable input parameter and adjusted or modified accordingly; these are not intended to limit the invention.

[0074] In addition, it is understood that in this invention, when the mathematical model includes multiple controllable input parameters (i.e., adjustable / modifiable), the corresponding local target point can be found using gradient descent in multiple dimensions. The coefficient of the corresponding controllable input parameter in the mathematical model can be obtained by calculating the tangent slope of the corresponding local target point, thereby adjusting the corresponding controllable input parameter.

[0075] Based on a mathematical model, this invention can automatically adjust the input parameters of the device, thereby not only improving the intelligence level of the device but also reducing the frequency of manual operation and saving costs.

[0076] In addition, the present invention can also issue a fault alert when parameters cannot be adjusted, notifying personnel to wait in advance, thus realizing fault prediction and alerting.

[0077] Exemplary embodiments of the present invention have been specifically illustrated and described above. It should be understood that the present invention is not limited to the disclosed embodiments; rather, the present invention is intended to cover various modifications and equivalent arrangements contained within the spirit and scope of the appended claims.

Claims

1. A parameter tuning method applied to a device, wherein the parameters of the device include input parameters, output values, and target values, characterized in that, include: A mathematical model for obtaining the input parameters and output values ​​of the device; Obtain the target value; Obtain the output value; If the absolute value of the difference between the output value and the target value is greater than the first threshold, the input parameters are adjusted according to the mathematical model so that the absolute value of the difference between the adjusted output value and the target value is reduced.

2. The parameter tuning method according to claim 1, characterized in that, The steps for obtaining the output value include: The real-time output parameters of the device are acquired periodically. The output value is the average of the multiple real-time output parameters.

3. The parameter tuning method according to claim 2, characterized in that, The output value is the average of ten consecutive real-time output parameters.

4. The parameter tuning method according to claim 1, characterized in that, The input parameters include a first input parameter, and the step of adjusting the input parameters includes: Obtain the coefficients of the first input parameter in the mathematical model; The adjustment value is obtained based on the absolute value and the coefficient; Adjust the first input parameter according to the adjustment value.

5. The parameter tuning method according to claim 4, characterized in that, The first input parameter has an upper bound and a lower bound, and the parameter tuning method further includes: Compare the adjusted first input parameter with the upper bound value or the lower bound value; If the adjusted first input parameter is greater than or equal to the upper bound value, then the first input parameter is adjusted according to the upper bound value; If the adjusted first input parameter is less than or equal to the lower bound value, then the first input parameter is adjusted according to the lower bound value; If the adjusted first input parameter is less than the upper bound and greater than the lower bound, then the first input parameter is adjusted according to the adjustment value.

6. The parameter tuning method according to claim 1, characterized in that, Also includes: The results of parameter adjustment are judged, including: Record the moment when the action of adjusting the input parameters is first performed, and use that moment as the starting time; If the absolute value is greater than the first threshold within the first time period starting from the start time, the parameter adjustment fails. If the absolute value is less than or equal to the first threshold within the first time period, the parameter adjustment is successful.

7. The parameter tuning method according to claim 1, characterized in that, Also includes: If the absolute value of the difference between the output value and the target value is greater than the second threshold, a warning message will be issued.

8. A glue spraying device, comprising a control unit, characterized in that, The control unit is used to execute the parameter tuning method as described in any one of claims 1 to 7, wherein the input parameters include temperature, air pressure and spray height, the output value includes spray thickness, and the target value includes standard adhesive thickness.

Images