Parameter optimization method, device and equipment based on industrial internet and medium

Abstract

The embodiment of the invention provides a parameter optimization method and device based on an industrial internet, equipment and a medium. The method comprises the following steps: acquiring an initial training sample, wherein the initial training sample comprises a production process parameter set and an influence parameter set of a mold on different injection molding machines; training a preset Gaussian model by using the initial training sample to obtain an agent model; acquiring a feature information value of the target injection molding machine; and optimizing the proxy model by using a preset sampling function and the feature information value to obtain a recommended influence parameter set of the mold on the target injection molding machine. The method is used for achieving the effect of reducing the debugging time of the target injection molding machine.

Description

Technical Field

[0001] The present application relates to a plastic product molding process, and in particular to an optimization parameter method, device, equipment and medium based on the Industrial Internet. Background Art

[0002] In the plastic injection molding process, due to the inherent performance differences between injection molding machines and the diversity of mold and injection molding machine combinations, when the same set of molds is produced on different injection molding machines, it is necessary to manually go through a tedious process of adjusting the influencing parameters based on experience. This not only prolongs the production preparation time and increases the debugging cost, but may also affect the stability of production efficiency and product quality. Summary of the invention

[0003] The embodiments of the present application provide an optimization parameter method, device, equipment and medium based on the Industrial Internet to achieve the effect of reducing debugging time.

[0004] In a first aspect, an embodiment of the present application provides a parameter optimization method based on the industrial Internet, including:

[0005] Acquire initial training samples, where the initial training samples include a set of production process parameters of the mold on different injection molding machines and a set of influencing parameters;

[0006] Using the initial training samples to train a preset Gaussian model to obtain a proxy model;

[0007] Obtain characteristic information values ​​of the target injection molding machine;

[0008] The proxy model is optimized using a preset sampling function and the characteristic information value to obtain a set of recommended influencing parameters of the mold on the target injection molding machine.

[0009] In a possible implementation, the proxy model is optimized using a preset sampling function and the characteristic information value to obtain a set of recommended influencing parameters of the mold on the target injection molding machine, including:

[0010] Determine, based on the characteristic information value, a parameter type and a corresponding value range of each influencing parameter in the influencing parameter set;

[0011] Obtaining a target parameter set that can maximize the acquisition function from the value ranges corresponding to each of the parameter types;

[0012] Inputting the target parameter set into the proxy model to obtain the target function value corresponding to the target parameter set;

[0013] Updating the initial training samples using the target parameter set and its corresponding target function value;

[0014] The Gaussian model is iteratively trained using the updated initial training samples until a preset stop condition is met, and the target parameter set is determined as the recommended influencing parameter set.

[0015] In a possible implementation, obtaining a target parameter set capable of maximizing the acquisition function from the value ranges corresponding to the parameter types includes:

[0016] Selecting a Pareto frontier from the initial training samples;

[0017] Randomly combining the values ​​of the influencing parameters within the value ranges corresponding to the parameter types to obtain a set of candidate influencing parameters;

[0018] Calculating the expected hypervolume improvement of each of the candidate influencing parameter sets relative to the Pareto frontier;

[0019] The candidate influencing parameter set corresponding to the maximum expected super volume improvement is determined as the target parameter set.

[0020] In a possible implementation, selecting the Pareto front from the initial training sample includes:

[0021] Inputting each of the influencing parameter sets into the proxy model to obtain the objective function values ​​corresponding to each of the influencing parameter sets;

[0022] Selecting a non-dominated solution from the objective function values ​​corresponding to each of the influencing parameter sets;

[0023] The influencing parameter set corresponding to the non-dominated solution is determined as the Pareto front.

[0024] In a possible implementation manner, determining the parameter type and the corresponding value range of each influencing parameter in the influencing parameter set based on the characteristic information value includes:

[0025] Determining a parameter type of each influencing parameter in the influencing parameter set;

[0026] If the parameter type includes an injection molding machine feature, the value range of the parameter type corresponding to the injection molding machine feature is determined as the feature information value.

[0027] In a possible implementation manner, after determining the parameter type of each influencing parameter in the influencing parameter set, the method further includes:

[0028] If the parameter type includes environmental information, then obtaining the environmental information value of the target injection molding machine;

[0029] The value range of the parameter type corresponding to the environmental information is determined as the environmental information value.

[0030] In a possible implementation, the parameter optimization method based on the industrial Internet further includes:

[0031] The recommended influencing parameter set is sent to the target injection molding machine, triggering the target injection molding machine to operate according to the recommended influencing parameter set.

[0032] In a second aspect, an embodiment of the present application provides an optimization parameter device based on the industrial Internet, including:

[0033] A first acquisition module is used to acquire an initial training sample, wherein the initial training sample includes a production process parameter set and an influencing parameter set of the mold on different injection molding machines;

[0034] A training module, used to train a preset Gaussian model using the initial training samples to obtain a proxy model;

[0035] A second acquisition module is used to acquire a characteristic information value of a target injection molding machine;

[0036] The optimization module is used to optimize the proxy model by using a preset sampling function and the characteristic information value to obtain a set of recommended influencing parameters of the mold on the target injection molding machine.

[0037] In a third aspect, an embodiment of the present application provides an electronic device, including: a memory, a processor;

[0038] The memory stores computer-executable instructions;

[0039] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementations of the first aspect.

[0040] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the first aspect above and / or various possible implementations of the first aspect.

[0041] The industrial Internet-based optimization parameter method, device, equipment and medium provided in the embodiments of the present application construct a proxy model by utilizing the production process parameter set and the influencing parameter set of the mold on different injection molding machines, and then optimize the proxy model using a sampling function, that is, the influencing parameter set is optimized by a Bayesian optimization algorithm to quickly find the recommended influencing parameter set for the target injection molding machine, thereby achieving the effect of reducing the parameter debugging time of the target injection molding machine. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0043] Figure 1 A flow chart of the optimization parameter method based on the industrial Internet provided in this application;

[0044] Figure 2 A schematic diagram of a process for optimizing a proxy model provided in this application;

[0045] Figure 3 A schematic diagram of the structure of an optimization parameter device based on the industrial Internet provided in this application;

[0046] Figure 4 A schematic diagram of the structure of the electronic device provided in this application.

[0047] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0048] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0049] Due to the development of technology or the loss of existing injection molding machines, it is inevitable to replace injection molding machines. Due to the differences between injection molding machines, in order to ensure that different injection molding machines achieve the same production effect, the process parameters of the injection molding machine need to be adjusted when replacing the injection molding machine. The existing method of adjusting the process parameters of the injection molding machine is basically to manually debug the injection molding machine based on their experience. The debugging results are proved by the products produced by the injection molding machine after debugging. If the products produced by the injection molding machine after debugging cannot meet the production requirements, the manual will continue to adjust the process parameters of the injection molding machine until the products produced by the injection molding machine meet the production requirements. In this way, not only a lot of time is needed, but also a lot of debugging costs are consumed.

[0050] In order to reduce the debugging time and debugging cost of the injection molding machine, this embodiment provides a method for optimizing parameters, by collecting all historical data generated by the mold on different injection molding machines, such as a production process parameter set and an influencing parameter set. Then a multi-objective Bayesian optimization algorithm is used to optimize the process parameters in the injection molding process, realize automatic adjustment and recommendation of process parameters, reduce manual intervention, improve the automation and intelligence level of production, and achieve the effect of reducing debugging time and debugging cost.

[0051] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0052] Figure 1 A flow chart of the optimization parameter method based on the industrial Internet provided for this application is as follows: Figure 1 As shown, the method includes:

[0053] Step S101, obtaining initial training samples, where the initial training samples include a production process parameter set and an influencing parameter set of a mold on different injection molding machines.

[0054] Production process parameters refer to the variables that need to be set and controlled during the injection molding process of the mold on the injection molding machine. Production process parameters are, for example, the starting point of holding pressure, the starting point of injection, the end point of injection, the injection tip pressure, the storage tip pressure, the holding time, the injection time, and the storage time. Among them, the starting point of holding pressure refers to the position where the injection is switched to the holding pressure stage. The starting point of injection refers to the position where the injection process starts, that is, the position where the screw starts to push the plastic melt forward into the mold cavity. The end point of injection refers to the position where the injection process ends, that is, the screw advances to the set maximum position to complete the injection action. The injection tip pressure refers to the maximum pressure value reached during the injection process. The storage tip pressure refers to the maximum pressure value reached during the pre-plasticization process, which is the process of heating and melting plastic particles. The holding time refers to the time maintained under the holding pressure, the purpose of which is to compensate for the shrinkage of the material during the cooling process and ensure the stability and consistency of the product size. The injection time refers to the total time from the start of injection to the end of injection. The storage time refers to the time it takes for the screw to retreat and melt the new plastic particles into a melt.

[0055] The influencing parameter set is, for example, an initial process parameter set, environmental information parameters, and injection molding machine characteristic parameters.

[0056] Initial process parameters are variables that directly affect the injection molding process and can be set and adjusted through the human-machine interface of the injection molding machine. Initial process parameters include temperature parameters, pressure parameters, time parameters, position parameters, and speed parameters. Temperature parameters include barrel temperature and mold temperature. Barrel temperature refers to the heating temperature of each section of the barrel, which can affect the molten state of the plastic. Mold temperature refers to the temperature at which the mold is heated or cooled, which can affect the cooling and solidification process of the plastic. Pressure parameters include injection pressure and holding pressure. Injection pressure refers to the pressure that pushes the plastic into the mold. Holding pressure refers to maintaining a certain pressure after the plastic fills the mold to eliminate the shrinkage of the plastic. Time parameters include holding time, cooling time, and injection time. Injection time refers to the time it takes for the plastic melt to be injected into the mold from the nozzle. Cooling time refers to the time required for the plastic to cool completely in the mold. Holding time refers to the duration of the holding pressure. Position parameters include injection position, transfer position, and holding position. Injection position refers to the distance the screw or plunger moves forward during the injection process, which determines the volume of the plastic melt injected into the mold. The pressure transfer position refers to the position from the injection stage to the holding stage. The holding position refers to the position of the screw or plunger when switching from the injection stage to the holding stage. Speed ​​parameters include injection speed and holding speed. Injection speed refers to the speed at which the plastic melt is pushed into the mold. Holding speed refers to the speed at which the screw or plunger moves during the holding stage.

[0057] Environmental information parameters include the temperature and humidity of the environment where the injection molding machine is located. Injection molding machine characteristic parameters include the clamping force and usage time of the injection molding machine.

[0058] Among them, the size of the influencing parameters directly affects the production process parameters. Therefore, to determine the production process parameters of the mold on the target injection molding machine, the prerequisite is to determine the influencing parameter set of the mold on the target injection molding machine.

[0059] Step S102: train the preset Gaussian model using the initial training samples to obtain a proxy model.

[0060] The Gaussian model representation uses a normal distribution to model the probability distribution of the initial training samples.

[0061] In the embodiment of the present application, the influencing parameter x and the unknown input parameter and production process parameters y, prediction results The relationship follows a multidimensional Gaussian distribution: where K is the covariance matrix whose elements are the Matern 5 / 2 kernel function to calculate x i ,x j deviation.

[0062] Define x i With x j The covariance between is:

[0063]

[0064] Among them, K ij = k(x i ,x j ) represents the covariance between the i-th influencing parameter and the j-th influencing parameter. σ f is the variance, l is the region length, σ ε is the observation noise, and the kernel hyperparameters are set by maximum likelihood estimation. For x i and The covariance vector between is defined as in, The kernel covariance function is determined by the distance between the two inputs, which means that the mean and variance of a given unknown input are predicted by the proximity between the collected data and the unknown input.

[0065] Step S103, obtaining characteristic information values ​​of the target injection molding machine.

[0066] In some embodiments, the characteristic information value of the target injection molding machine is, for example, the clamping force and usage time of the target injection molding machine.

[0067] The clamping force refers to the maximum clamping force applied by the clamping mechanism of the injection molding machine to the mold. Its function is to ensure that the mold will not be separated by the pressure of the plastic melt during the injection process, thereby preventing the generation of flash or overflow. This clamping force is to resist the force generated by the injection pressure. It can be seen that the clamping force is closely related to the injection pressure. For example, when the clamping force is insufficient, the injection pressure and speed need to be reduced to avoid flash.

[0068] Since the use time of the injection molding machine will affect the degree of wear of the injection molding machine, for example, the longer the use time, the greater the degree of wear of the injection molding machine may be. The wear of different parts on the injection molding machine will bring different consequences. For example, if the parts of the clamping system are worn, it will cause the stability of the clamping force to decrease. If the screw, nozzle and other parts of the injection system are worn, it may cause changes in the injection pressure and injection speed. Therefore, by obtaining the use time of the target injection molding machine, the degree of wear of the target injection molding machine can be inferred, which is convenient for adaptive adjustment of process parameters such as injection speed and injection pressure.

[0069] Step S104 , optimizing the proxy model using a preset sampling function and characteristic information value to obtain a set of recommended influencing parameters of the mold on the target injection molding machine.

[0070] The goal of the sampling function is to find the next point among the existing observation points that is most likely to improve the current optimal solution.

[0071] Specifically, the acquisition function can be calculated by improving the Pareto front and the expected hypervolume.

[0072] The optimization parameter method based on the industrial Internet provided in the embodiment of the present application constructs a proxy model by using the production process parameter set and the influencing parameter set of the mold on different injection molding machines, and then optimizes the proxy model using the sampling function, that is, the influencing parameter set is optimized by the Bayesian optimization algorithm, and the recommended influencing parameter set for the target injection molding machine is quickly found, thereby achieving the effect of reducing the parameter debugging time of the target injection molding machine. In addition, by optimizing the proxy model in combination with the characteristic information value of the target injection molding machine, the optimization result is closer to the actual production situation, and has strong adaptability and robustness.

[0073] Furthermore, after obtaining the recommended influencing parameter set, a reminder message can be sent to the user, and the reminder message is used to remind the user to review the recommended influencing parameter set. If the user receives the information that the review is passed, the mold model of the mold is obtained, and the mold model and its corresponding recommended influencing parameter set are sent to the target injection molding machine, triggering the target injection molding machine to inject plastic into the mold of the corresponding mold model according to the recommended influencing parameter set. If the user receives the information that the review failed, the parameters are optimized again according to the above-mentioned optimization parameter method based on the industrial Internet.

[0074] Furthermore, after triggering the target injection molding machine to inject plastic into the mold of the corresponding mold model according to the recommended influencing parameter set, the quality of the products actually produced by the target injection molding machine can also be inspected. If the products produced by the target injection molding machine do not meet the production requirements, the parameters are optimized again according to the above-mentioned optimization parameter method based on the industrial Internet.

[0075] Figure 2 A flow chart of the method for optimizing the proxy model provided in this application. Figure 1 Based on the embodiment, the optimization of the proxy model using the preset sampling function and characteristic information value to obtain the recommended influencing parameter set of the mold on the target injection molding machine is described in detail. Figure 2 As shown, the method for optimizing the proxy model may include the following steps:

[0076] Step S201: determining the parameter type and the corresponding value range of each influencing parameter in the influencing parameter set based on the characteristic information value.

[0077] The parameter type of each influencing parameter in the influencing parameter set may be determined first, and then the value range corresponding to each parameter type may be determined.

[0078] Specifically, if the parameter type is injection molding machine characteristics, the value range of the parameter type corresponding to the injection molding machine characteristics is determined as the characteristic information value of the target injection molding machine. If the parameter type is environmental information, the value range of the parameter type corresponding to the environmental information is determined as the environmental information value of the target injection molding machine.

[0079] If the parameter type is neither an injection molding machine feature nor environmental information, the parameter value corresponding to the parameter type is obtained from the influencing parameter set. The maximum value of the parameter type is determined as the maximum parameter value corresponding to the parameter type, and the minimum value corresponding to the parameter type is determined as the minimum parameter value corresponding to the parameter type. The value range of the parameter type is from the minimum parameter value corresponding to the parameter type to the maximum parameter value corresponding to the parameter type.

[0080] Step S202, selecting the Pareto front from the initial training samples.

[0081] Furthermore, each influencing parameter set can be input into the proxy model to obtain the objective function values ​​corresponding to each influencing parameter set; a non-dominated solution is selected from the objective function values ​​corresponding to each influencing parameter set; and the influencing parameter set corresponding to the non-dominated solution is determined as the Pareto frontier.

[0082] The Pareto frontier is defined as follows: P(Y) = {y′∈Y: {y″∈Y: y″>y′, y′≠y″} = φ}, that is, when the objective function value y″∈R m Better than the objective function value y′∈R corresponding to another set of influencing parameters m For example, in a minimization problem, if at least one component of y” is strictly smaller than the corresponding component of y’, and all components of y” are smaller than or equal to the corresponding components of y’, then it can be expressed as y”>y’.

[0083] Step S203: randomly combine the influencing parameter values ​​within the value ranges corresponding to the parameter types to obtain a candidate influencing parameter set.

[0084] Step S204, calculating the expected hypervolume improvement of each candidate influencing parameter set relative to the Pareto frontier.

[0085] For each candidate influencing parameter set, the following steps are performed to obtain the expected hypervolume improvement of each candidate influencing parameter set relative to the Pareto frontier.

[0086] First, the first hypervolume of the Pareto front relative to the reference point is calculated, then the Pareto front is updated using the set of candidate influencing parameters, then the second hypervolume of the updated Pareto front relative to the reference point is calculated, and then the increment of the second hypervolume relative to the first hypervolume is calculated. Finally, the expected hypervolume improvement is calculated based on the increment.

[0087] The increment of the second hypervolume relative to the first hypervolume satisfies the formula HVI(P,f(x))=HV(P∪f(x))-HV(P), where HV(P∪f(x)) is the second hypervolume and HV(P) is the first hypervolume. HVI(P,f(x)) is the increment of the second hypervolume relative to the first hypervolume. f(x) is the target parameter value obtained after the alternative parameter set X is input into the proxy model.

[0088] The expected hypervolume improvement satisfies the formula EHVI = E[HVI(P, f(x))] = ∫HVI(P, f(x))p(f(x))df(x), where the reference point is used to define the boundary of the hypervolume, and all non-dominated solutions are inside the reference point. The reference point can be set in advance, or the objective function values ​​corresponding to each influencing parameter set can be cheaply calculated to obtain the reference point. The hypervolume is an area defined by the Pareto front and the reference point. The expected hypervolume improvement is used to analyze the hypervolume increment brought about by the addition of non-dominated solutions to the alternative influencing parameter set.

[0089] Step S205: determining the candidate influencing parameter set corresponding to the maximum expected hypervolume improvement as the target parameter set.

[0090] Since the expected hypervolume improvement can analyze the expected contribution of the alternative influencing parameter set to the current non-dominated solution set hypervolume, the greater the expected hypervolume improvement, the greater the expected contribution of the alternative influencing parameter set to the current non-dominated solution set hypervolume, and the greater the expected contribution, the more likely the final parameter value will cover a wider performance space. Therefore, the alternative influencing parameter set corresponding to the maximum expected hypervolume improvement is determined as the target parameter set.

[0091] Step S206: input the target parameter set into the proxy model to obtain the target function value corresponding to the target parameter set.

[0092] Since the proxy model has learned the relationship between the influencing parameter set and the generating process parameter set, by inputting the target parameter set into the proxy model, it is possible to predict the generating process parameter set corresponding to the target parameter set, that is, to obtain the objective function value corresponding to the target parameter set.

[0093] Step S207: Update the initial training samples using the target parameter set and its corresponding target function value.

[0094] Specifically, the target parameter set and its corresponding target function value are added to the initial training sample to achieve the purpose of updating the initial training sample.

[0095] Step S208, iteratively training the Gaussian model using the updated initial training samples.

[0096] Step S209, determining whether a preset stop condition is met, if so, executing step S210, otherwise, executing step S202.

[0097] If step S202 is executed, the Pareto frontier is selected from the updated initial training sample. Since the initial training sample has been updated using the target parameter set and its corresponding target function value, the selection of the Pareto frontier from the initial training sample can be replaced by the selection of the Pareto frontier from the updated initial training sample.

[0098] The preset stop condition is, for example, a preset number of iterations, a preset iteration time, a preset objective function value, etc.

[0099] If the preset stop condition is a preset number of iterations, step S209 can be replaced by: determining whether the number of training times of the Gaussian model reaches the preset number of iterations, if so, executing step S210, otherwise, executing step S202.

[0100] If the preset stop condition is a preset iteration time, step S209 can be replaced by: determining whether the training time of the Gaussian model reaches the preset iteration time, if so, executing step S210, otherwise, executing step S202.

[0101] If the preset stop condition is a preset objective function value, step S209 can be replaced by: determining whether the objective function value corresponding to the target parameter set reaches the preset objective function value, if so, executing step S210, otherwise, executing step S202.

[0102] Step S210: sending the target parameter set as the recommended influencing parameter set to the target injection molding machine, triggering the target injection molding machine to operate according to the recommended influencing parameter set.

[0103] The method for optimizing the proxy model provided in the embodiment of the present application can analyze the expected contribution of the alternative influencing parameter set to the current non-dominated solution set hypervolume by calculating the expected hypervolume improvement of each alternative influencing parameter set relative to the Pareto frontier, thereby finding a variety of high-quality solutions within a limited number of experiments, and then quickly finding the optimal or near-optimal influencing parameter combination, reducing debugging time and trial and error costs, and improving production efficiency and product quality. At the same time, controlling the target injection molding machine to operate according to the optimized influencing parameters, that is, operating according to the recommended influencing parameter set, can reduce energy consumption and raw material waste, and reduce production costs.

[0104] Figure 3 The schematic diagram of the structure of the optimization parameter device based on the industrial Internet provided in this application is as follows: Figure 3 As shown, the parameter optimization device 300 based on the industrial Internet provided in this embodiment includes:

[0105] The first acquisition module 301 is used to acquire initial training samples, where the initial training samples include a production process parameter set and an influencing parameter set of the mold on different injection molding machines;

[0106] The training module 302 is used to train the preset Gaussian model using the initial training samples to obtain the proxy model;

[0107] The second acquisition module 303 is used to acquire the characteristic information value of the target injection molding machine;

[0108] The optimization module 304 is used to optimize the proxy model using a preset sampling function and characteristic information value to obtain a set of recommended influencing parameters of the mold on the target injection molding machine.

[0109] Optionally, an optimization module is used to determine the parameter type and the corresponding value range of each influencing parameter in the influencing parameter set based on the feature information value; obtain the target parameter set that can maximize the acquisition function from the value range corresponding to each parameter type; input the target parameter set into the proxy model to obtain the target function value corresponding to the target parameter set; use the target parameter set and its corresponding target function value to update the initial training sample; use the updated initial training sample to iteratively train the Gaussian model until the preset stop condition is met, and determine the target parameter set as the recommended influencing parameter set.

[0110] Optionally, the optimization module is used to select the Pareto frontier from the initial training samples; randomly combine the values ​​of each influencing parameter within the value range corresponding to each parameter type to obtain an alternative influencing parameter set; calculate the expected hypervolume improvement of each alternative influencing parameter set relative to the Pareto frontier; and determine the alternative influencing parameter set corresponding to the maximum expected hypervolume improvement as the target parameter set.

[0111] Optionally, the optimization module is used to input each influencing parameter set into the agent model to obtain the objective function values ​​corresponding to each influencing parameter set; select a non-dominated solution from the objective function values ​​corresponding to each influencing parameter set; and determine the influencing parameter set corresponding to the non-dominated solution as the Pareto frontier.

[0112] Optionally, if the parameter type includes an injection molding machine feature, the optimization module is used to determine a value range of the parameter type corresponding to the injection molding machine feature as the feature information value.

[0113] Optionally, if the parameter type includes environmental information, the optimization module is further used to obtain the environmental information value of the target injection molding machine; and determine the value range of the parameter type corresponding to the environmental information as the environmental information value.

[0114] Optionally, the parameter optimization device based on the industrial Internet also includes a sending module, which is used to send the recommended influencing parameter set to the target injection molding machine, triggering the target injection molding machine to operate according to the recommended influencing parameter set.

[0115] The industrial Internet-based optimization parameter device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effects are similar, and this embodiment will not be described in detail here.

[0116] Figure 4 This is a schematic diagram of the structure of the electronic device provided in this application. Figure 4 As shown, the electronic device 400 provided in this embodiment includes: at least one processor 401 and a memory 402. Optionally, the device 400 also includes a communication component 403. The processor 401, the memory 402 and the communication component 403 are connected via a bus.

[0117] In a specific implementation process, at least one processor 401 executes the computer-executable instructions stored in the memory 402, so that at least one processor 401 executes the above method.

[0118] The specific implementation process of the processor 401 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0119] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the invention may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0120] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (Non-volatile Memory, NVM), such as at least one disk storage.

[0121] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of the present application is not limited to only one bus or one type of bus.

[0122] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0123] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0124] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0125] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0126] The division of units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0127] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0128] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0129] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0130] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0131] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention, which follow the general principles of the present invention and include common knowledge or customary technical means in the art not disclosed by the present invention, are not limited to the precise structure described above and shown in the drawings, and may be modified and changed in various ways without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A parameter optimization method based on industrial Internet, characterized in that: include: Acquire initial training samples, where the initial training samples include a set of production process parameters of the mold on different injection molding machines and a set of influencing parameters; Using the initial training samples to train a preset Gaussian model to obtain a proxy model; Obtain characteristic information values ​​of the target injection molding machine; The proxy model is optimized using a preset sampling function and the characteristic information value to obtain a set of recommended influencing parameters of the mold on the target injection molding machine.

2. The method according to claim 1, characterized in that: The proxy model is optimized using a preset sampling function and the characteristic information value to obtain a set of recommended influencing parameters of the mold on the target injection molding machine, including: Determine, based on the characteristic information value, a parameter type and a corresponding value range of each influencing parameter in the influencing parameter set; Obtaining a target parameter set that can maximize the acquisition function from the value ranges corresponding to each of the parameter types; Inputting the target parameter set into the proxy model to obtain the target function value corresponding to the target parameter set; Updating the initial training samples using the target parameter set and its corresponding target function value; The Gaussian model is iteratively trained using the updated initial training samples until a preset stop condition is met, and the target parameter set is determined as the recommended influencing parameter set.

3. The method according to claim 2, characterized in that From the value ranges corresponding to the parameter types, a target parameter set capable of maximizing the acquisition function is obtained, including: Selecting a Pareto frontier from the initial training samples; Randomly combining the values ​​of the influencing parameters within the value ranges corresponding to the parameter types to obtain a set of candidate influencing parameters; Calculating the expected hypervolume improvement of each of the candidate influencing parameter sets relative to the Pareto frontier; The candidate influencing parameter set corresponding to the maximum expected hypervolume improvement is determined as the target parameter set.

4. The method according to claim 3, characterized in that Selecting a Pareto frontier from the initial training sample includes: Inputting each of the influencing parameter sets into the proxy model to obtain the objective function values ​​corresponding to each of the influencing parameter sets; Selecting a non-dominated solution from the objective function values ​​corresponding to each of the influencing parameter sets; The influencing parameter set corresponding to the non-dominated solution is determined as the Pareto front.

5. The method according to claim 2, characterized in that: Determining the parameter type and the corresponding value range of each influencing parameter in the influencing parameter set based on the characteristic information value includes: Determining a parameter type of each influencing parameter in the influencing parameter set; If the parameter type includes an injection molding machine feature, the value range of the parameter type corresponding to the injection molding machine feature is determined as the feature information value.

6. The method according to claim 5, characterized in that After determining the parameter type of each influencing parameter in the influencing parameter set, the method further includes: If the parameter type includes environmental information, then obtaining the environmental information value of the target injection molding machine; The value range of the parameter type corresponding to the environmental information is determined as the environmental information value.

7. The method according to any one of claims 1 to 6, characterized in that: Also includes: The recommended influencing parameters are sent to the target injection molding machine, triggering the target injection molding machine to inject plastic into the mold according to the recommended influencing parameter set.

8. An optimization parameter device based on industrial Internet, characterized in that: include: A first acquisition module is used to acquire an initial training sample, wherein the initial training sample includes a production process parameter set and an influencing parameter set of the mold on different injection molding machines; A training module, used to train a preset Gaussian model using the initial training samples to obtain a proxy model; A second acquisition module is used to acquire a characteristic information value of a target injection molding machine; The optimization module is used to optimize the proxy model by using a preset sampling function and the characteristic information value to obtain a set of recommended influencing parameters of the mold on the target injection molding machine.

9. An electronic device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

Images