Method and device for determining residual oil film thickness evaluation model

Abstract

The invention provides a method and a device for determining a residual oil film thickness evaluation model. First evaluation models corresponding to the influence types are constructed at least based on the multiple first sample parameter values of the multiple influence types; obtaining a first output parameter value of the first sub-model corresponding to the influence type and a first evaluation parameter value of the first evaluation model corresponding to the influence type; obtaining second output parameter values of the corresponding influence types based on the first evaluation parameter values of the various influence types; constructing a second evaluation model based on the second output parameter values of the multiple influence types and the second weight variables of the corresponding influence types; second weight variable values of multiple influence types are obtained; generating a predicted oil film thickness value based on the second weight variable values of the multiple influence types and the first output parameter values of the corresponding influence types; and when the predicted oil film thickness value is verified to be valid, determining that the first evaluation model and the second evaluation model are valid. And a residual oil film thickness evaluation model is accurately determined.

Description

Technical Field

[0001] The present application relates to the technical field of oil film thickness, and in particular to a method, device, medium and electronic device for determining a residual oil film thickness evaluation model. Background Art

[0002] The oil film thickness of stamped sheet metal plays a vital role in the stamping process. After the stamped sheet metal is squeezed out by a wet cleaning machine, the control of the oil film thickness is affected by many complex factors, including but not limited to the ambient temperature, sheet metal temperature, cleaning oil temperature, cleaning oil viscosity, impurity content, sheet metal size, squeeze roller hardness, pressure and force between the two squeeze rollers, residual oil on the sheet metal before flushing, and the impact force of the cleaning oil during flushing.

[0003] At present, the single factor verification method is used to study the influence of these factors on the oil film thickness. However, in the actual production environment, the various factors are interrelated and they work together to form the oil film thickness. When a certain factor changes within a certain range, it may be positively correlated or even linearly positively correlated with the oil film thickness, but once it exceeds this range, its influence may no longer be significant, or even produce a negative correlation effect.

[0004] Therefore, the present application provides a method for determining a residual oil film thickness evaluation model to solve the above technical problems. Summary of the invention

[0005] The purpose of the present application is to provide a method, device, medium and electronic device for determining a residual oil film thickness evaluation model, which can solve at least one of the technical problems mentioned above.

[0006] The specific plan is as follows:

[0007] According to a specific embodiment of the present application, in a first aspect, the present application provides a method for determining a residual oil film thickness evaluation model, comprising:

[0008] Acquire a plurality of first sample parameter values ​​of respective plurality of influence types affecting the residual oil film thickness;

[0009] constructing first evaluation models corresponding to the impact types respectively based on at least a plurality of first sample parameter values ​​of the respective impact types, wherein the first evaluation model includes a first sub-model;

[0010] Combining and verifying the preset first verification models of the respective multiple impact types with the respective first evaluation models to obtain a first output parameter value of the first sub-model corresponding to the impact type and a first evaluation parameter value of the first evaluation model corresponding to the impact type;

[0011] When the first evaluation parameter values ​​of the first evaluation models of the plurality of impact types are all greater than a preset first effective threshold, applying the first evaluation parameter values ​​of the plurality of impact types to the second sub-models of the corresponding impact types to obtain second output parameter values ​​of the corresponding impact types;

[0012] constructing a second evaluation model based on the second output parameter values ​​of the plurality of impact types and second weight variables corresponding to the impact types;

[0013] Combining and verifying the preset second verification models of the respective multiple impact types with the second evaluation model to obtain the second weight variable values ​​of the respective second weight variables of the corresponding impact types and the second evaluation parameter values ​​of the second evaluation model;

[0014] When the second evaluation parameter value is greater than a preset second effective threshold, generating a predicted oil film thickness value based on second weight variable values ​​of the second weight variables of the plurality of influence types and first output parameter values ​​of the corresponding influence types;

[0015] When the detected multiple measured oil film thickness values ​​verify that the predicted oil film thickness value is valid, it is determined that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid.

[0016] Optionally, constructing first assessment models corresponding to the impact types respectively based on at least a plurality of first sample parameter values ​​of the plurality of impact types includes:

[0017] Based on the respective multiple first sample parameter values ​​and respective multiple first weight variables of the multiple impact types, first evaluation models corresponding to the impact types are respectively constructed.

[0018] Optionally, the combining and verifying the preset first verification models of the respective multiple impact types with the respective first evaluation models to obtain the first output parameter value of the first sub-model corresponding to the impact type and the first evaluation parameter value of the first evaluation model corresponding to the impact type includes:

[0019] Applying the first sub-models of the plurality of impact types to the preset first verification model of the corresponding impact type respectively, and then combining and verifying with the first evaluation model of the corresponding impact type, to obtain the first weight variable value of each of the plurality of first weight variables of the corresponding impact type;

[0020] Applying first weight variable values ​​of the plurality of first weight variables of the plurality of influence types to first sub-models of corresponding influence types to obtain first output parameter values ​​of the first sub-models of corresponding influence types;

[0021] The first output parameter values ​​of each of the plurality of impact types are applied to the first evaluation model of the corresponding impact type to obtain the first evaluation parameter value of the first evaluation model of the corresponding impact type.

[0022] Optionally, the combining and verifying the preset second verification models of the respective multiple impact types with the second evaluation model to obtain the second weight variable values ​​of the respective second weight variables of the corresponding impact types and the second evaluation parameter values ​​of the second evaluation model includes:

[0023] Based on the second output parameter values ​​of the multiple impact types in the second evaluation model, the second output parameter values ​​are applied to the preset second verification models of the corresponding impact types respectively, and then combined with the second evaluation model for verification to obtain the second weight variable values ​​of the multiple second weight variables of the corresponding impact type;

[0024] The second weight variable values ​​of the second weight variables of the plurality of impact types are applied to the second evaluation model to obtain second evaluation parameter values.

[0025] Optionally, when the detected multiple measured oil film thickness values ​​verify that the predicted oil film thickness value is valid, determining that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid includes:

[0026] Calculating the error values ​​between the predicted oil film thickness value and the plurality of measured oil film thickness values;

[0027] Calculating the ratio of the absolute value of the error value of each of the plurality of measured oil film thickness values ​​to the respective measured oil film thickness value to obtain the error rate of each corresponding measured oil film thickness value;

[0028] Counting the target number of each of the plurality of measured oil film thickness values ​​whose error rate is less than a preset error rate threshold;

[0029] Calculating the error acceptance percentage between the target quantity and the total quantity of the plurality of actually measured oil film thickness values;

[0030] When the error qualification percentage is greater than or equal to a preset error qualification percentage threshold, it is determined that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid.

[0031] According to a specific embodiment of the present application, in a second aspect, the present application provides a device for determining a residual oil film thickness evaluation model, comprising:

[0032] An acquisition unit, used for acquiring a plurality of first sample parameter values ​​of respective types of influences affecting the residual oil film thickness;

[0033] A first construction unit, configured to respectively construct first evaluation models corresponding to the impact types based at least on the respective first sample parameter values ​​of the multiple impact types, wherein the first evaluation model includes a first sub-model;

[0034] A first obtaining unit is used to combine and verify the preset first verification models of the plurality of impact types with the respective first evaluation models to obtain a first output parameter value of the first sub-model corresponding to the impact type and a first evaluation parameter value of the first evaluation model corresponding to the impact type;

[0035] A second obtaining unit is configured to apply the first evaluation parameter values ​​of the respective multiple impact types to the second sub-model of the corresponding impact type to obtain a second output parameter value of the corresponding impact type when the first evaluation parameter values ​​of the respective first evaluation models of the multiple impact types are all greater than a preset first effective threshold value;

[0036] A second construction unit, configured to construct a second evaluation model based on second output parameter values ​​of the plurality of impact types and second weight variables corresponding to the impact types;

[0037] A third obtaining unit is used to combine and verify the preset second verification models of the plurality of impact types with the second evaluation model to obtain the second weight variable values ​​of the second weight variables of the corresponding impact types and the second evaluation parameter values ​​of the second evaluation model;

[0038] A generating unit, configured to generate a predicted oil film thickness value based on second weight variable values ​​of the second weight variables of the plurality of influence types and first output parameter values ​​of the corresponding influence types when the second evaluation parameter value is greater than a preset second effective threshold;

[0039] A determination unit is used to determine that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid when the detected multiple measured oil film thickness values ​​verify that the predicted oil film thickness value is valid.

[0040] Optionally, constructing first assessment models corresponding to the impact types respectively based on at least a plurality of first sample parameter values ​​of the plurality of impact types includes:

[0041] Based on the respective multiple first sample parameter values ​​and respective multiple first weight variables of the multiple impact types, first evaluation models corresponding to the impact types are respectively constructed.

[0042] Optionally, the combining and verifying the preset first verification models of the respective multiple impact types with the respective first evaluation models to obtain the first output parameter value of the first sub-model corresponding to the impact type and the first evaluation parameter value of the first evaluation model corresponding to the impact type includes:

[0043] Applying the first sub-models of the plurality of impact types to the preset first verification model of the corresponding impact type respectively, and then combining and verifying with the first evaluation model of the corresponding impact type, to obtain the first weight variable value of each of the plurality of first weight variables of the corresponding impact type;

[0044] Applying first weight variable values ​​of the plurality of first weight variables of the plurality of influence types to first sub-models of corresponding influence types to obtain first output parameter values ​​of the first sub-models of corresponding influence types;

[0045] The first output parameter values ​​of each of the plurality of impact types are applied to the first evaluation model of the corresponding impact type to obtain the first evaluation parameter value of the first evaluation model of the corresponding impact type.

[0046] Optionally, the combining and verifying the preset second verification models of the respective multiple impact types with the second evaluation model to obtain the second weight variable values ​​of the respective second weight variables of the corresponding impact types and the second evaluation parameter values ​​of the second evaluation model includes:

[0047] Based on the second output parameter values ​​of the multiple impact types in the second evaluation model, the second output parameter values ​​are applied to the preset second verification models of the corresponding impact types respectively, and then combined with the second evaluation model for verification to obtain the second weight variable values ​​of the multiple second weight variables of the corresponding impact type;

[0048] The second weight variable values ​​of the second weight variables of the plurality of impact types are applied to the second evaluation model to obtain second evaluation parameter values.

[0049] Optionally, when the detected multiple measured oil film thickness values ​​verify that the predicted oil film thickness value is valid, determining that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid includes:

[0050] Calculating the error values ​​between the predicted oil film thickness value and the plurality of measured oil film thickness values;

[0051] Calculating the ratio of the absolute value of the error value of each of the plurality of measured oil film thickness values ​​to the respective measured oil film thickness value to obtain the error rate of each corresponding measured oil film thickness value;

[0052] Counting the target number of each of the plurality of measured oil film thickness values ​​whose error rate is less than a preset error rate threshold;

[0053] Calculating the error acceptance percentage between the target quantity and the total quantity of the plurality of actually measured oil film thickness values;

[0054] When the error qualification percentage is greater than or equal to a preset error qualification percentage threshold, it is determined that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid.

[0055] According to a specific implementation of the present application, in a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the method for determining the residual oil film thickness assessment model as described in any of the above items is implemented.

[0056] According to the specific implementation of the present application, in a fourth aspect, the present application provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method for determining the residual oil film thickness assessment model as described in any of the above items.

[0057] Compared with the prior art, the above solution of the embodiment of the present application has at least the following beneficial effects:

[0058] The present application provides a method, device, medium and electronic device for determining a residual oil film thickness assessment model. The present application constructs a first assessment model of a corresponding influence type based on a plurality of first sample parameter values ​​of each of the plurality of influence types that affect the residual oil film thickness, wherein the first assessment model includes a first sub-model; the preset first verification models of the plurality of influence types are combined with the respective first assessment models for verification to obtain a first output parameter value of the first sub-model of the corresponding influence type and a first assessment parameter value of the first assessment model of the corresponding influence type; when the first assessment parameter values ​​of the first assessment models of the plurality of influence types are all greater than a preset first effective threshold, the first assessment parameter values ​​of the plurality of influence types are applied to the second sub-model of the corresponding influence type to obtain a second output parameter value of the corresponding influence type; based on The second output parameter values ​​of the multiple impact types and the second weight variables of the corresponding impact types are used to construct a second evaluation model; the preset second verification models of the multiple impact types are combined with the second evaluation model for verification to obtain the second weight variable values ​​of the second weight variables of the corresponding impact types and the second evaluation parameter values ​​of the second evaluation model; when the second evaluation parameter value is greater than the preset second effective threshold, the predicted oil film thickness value is generated based on the second weight variable values ​​of the second weight variables of the multiple impact types and the first output parameter value of the corresponding impact type; when the multiple measured oil film thickness values ​​detected verify that the predicted oil film thickness value is valid, the evaluation model of the combination of the first evaluation model and the second evaluation model is determined to be valid. Through the comprehensive analysis of multi-dimensional factors, the limitations of the traditional single factor verification method are abandoned, and the comprehensive analysis of multi-dimensional factors is adopted to comprehensively consider the influence of multiple key factors such as ambient temperature, plate temperature, cleaning oil temperature, viscosity, impurity rate, plate size, squeeze roller hardness, etc. on the oil film thickness. It can more accurately reveal the interaction relationship between the factors, accurately determine the residual oil film thickness evaluation model, and provide a scientific basis for formulating precise measures. A multiple evaluation model of oil film thickness has been constructed, which can automatically learn and capture the complex nonlinear relationship between various factors, and accurately predict the weight inflection point of the influence of various factors on oil film thickness. It not only improves the accuracy and reliability of the prediction, but also provides strong technical support for optimizing the oil film thickness control strategy of stamping plates. It realizes the precise control and optimization of the oil film thickness of stamping plates, improves production efficiency and improves product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A flow chart showing a method for determining a residual oil film thickness evaluation model according to an embodiment of the present application is shown;

[0060] Figure 2 A unit block diagram of a device for determining a residual oil film thickness evaluation model according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0062] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two.

[0063] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0064] It should be understood that although the terms first, second, third, etc. may be used to describe in the embodiments of the present application, these descriptions should not be limited to these terms. These terms are only used to distinguish the descriptions. For example, without departing from the scope of the embodiments of the present application, the first may also be referred to as the second, and similarly, the second may also be referred to as the first.

[0065] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0066] It should also be noted that the term "includes", "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, so that a commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprising a ..." do not exclude the existence of other identical elements in the commodity or device including the elements.

[0067] It should be particularly noted that any symbols and / or numbers in the specification that are not marked in the accompanying drawings are not drawing marks.

[0068] The optional embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0069] The embodiment provided in this application is an embodiment of a method for determining a residual oil film thickness evaluation model.

[0070] Combine the following Figure 1 The embodiments of the present application are described in detail.

[0071] Step S101 : obtaining a plurality of first sample parameter values ​​of respective types of influences on the residual oil film thickness.

[0072] For example, the various influence types include: temperature type, pressure width and hardness type, impurity rate type, residual oil type and cleaning type; the temperature type includes the ambient temperature type and the plate temperature type, denoted as x (F-11, F-21, ..., F-n1); the pressure width and hardness type includes the pressure of each squeezing roller of the wet cleaning machine, and the resulting pressure width type and hardness type, denoted as x (F-12, F-22, ..., F-n2); the cleaning oil type includes the viscosity type and the impurity rate type, denoted as x (F-13, F-23, ..., F-n3); the size, thickness, shape of the plate and the residual oil type of the plate before flushing, denoted as x (F-14, F-24, ..., F-n4); the cleaning type such as the flushing intensity and the temperature of the cleaning oil, denoted as x (F-1n, F-2n, ..., F-nn).

[0073] After the plate passes through the wet cleaning machine, an online monitoring mechanism for oil film thickness is installed, and the test results are uploaded through the communication module. According to a certain sampling period T, the online monitoring mechanism is used to obtain multiple samples A1, A2, ..., An. Considering the influence of the data collection process, the data preprocessing method of the Raida criterion method is used to eliminate samples greater than μ+3σ and less than μ-3σ. μ represents the mathematical expectation of the normal population, and σ represents the standard deviation of the normal population.

[0074] Step S102: constructing first evaluation models corresponding to the impact types respectively based on at least a plurality of first sample parameter values ​​of the respective impact types.

[0075] Wherein, the first evaluation model includes a first sub-model.

[0076] For example, the first evaluation model of the first impact type is g(z1)=g(x(F-11), x(F-21),..., x(F-n1), a1), the first evaluation model of the second impact type is g(z2)=g(x(F-12), x(F-22),..., x(F-n2), a2),..., the first evaluation model of the nth impact type is g(zn)=g(x(F-1n), x(F-2n),..., x(F-nn), an), where a1, a2,..., an are all bias values, that is, the minimum values ​​that produce an impact, and are empirical values ​​obtained through experiments.

[0077] In some specific embodiments, constructing first evaluation models corresponding to the impact types respectively based on at least a plurality of first sample parameter values ​​of the plurality of impact types includes:

[0078] Step S102a: constructing first evaluation models corresponding to the impact types based on the respective first sample parameter values ​​and the respective first weight variables of the respective impact types.

[0079] For example, according to the collected sample data, weights (ω1, ω2, ..., ωn)n are assigned respectively. Due to the nonlinear influence of the weights, the first evaluation model of the first influence type is g(z1)=g(ω11*x(F-11)+ω12*x(F-21)+......+ω1n*x(F-n1)+a1), the first sub-model of the first influence type is z1=ω11*x(F-11)+ω12*x(F-21)+......+ω1n*x(F-n1)+a1; the first evaluation model of the second influence type is g(z2)=g(ω21*x(F-12)+ω22*x(F-22 )+......+ω2n*x(F-n2)+a2), the first submodel of the second influence type z2=ω21*x(F-12)+ω22*x(F-22)+......+ω2n*x(F-n2)+a2;......;the first evaluation model of the nth influence type g(zn)=g(ωn1*x(F-1n)+ωn2*x(F-2n)+......+ωnn*x(F-nn)+an), the first submodel of the nth influence type zn=ωn1*x(F-1n)+ωn2*x(F-2n)+......+ωnn*x(F-nn)+an.

[0080] Step S103, combining and verifying the preset first verification models of the plurality of impact types with their respective first evaluation models to obtain the first output parameter value of the first sub-model corresponding to the impact type and the first evaluation parameter value of the first evaluation model corresponding to the impact type.

[0081] In some specific embodiments, the combining and verifying the preset first verification models of the respective multiple impact types with the respective first evaluation models to obtain the first output parameter value of the first sub-model corresponding to the impact type and the first evaluation parameter value of the first evaluation model corresponding to the impact type includes:

[0082] Step S103-1, respectively apply the first sub-models of the plurality of impact types to the preset first verification model of the corresponding impact type, and then combine and verify with the first evaluation model of the corresponding impact type to obtain the first weight variable value of each of the plurality of first weight variables of the corresponding impact type.

[0083] For example, continuing with the above example, the preset first verification model of the nth influence type g(zn)=1 / [1+e^(-zn)], where the first submodel of the nth influence type zn=ωn1*x(F-1n)+ωn2*x(F-2n)+......+ωnn*x(F-nn)+an, the preset first verification model of the nth influence type is combined with the first evaluation model of the nth influence type for verification, and the first weight variable values ​​(ωn1, ωn2,..., ωnn) of the nth influence type are obtained; based on this, the first weight variable values ​​of the n first weight variables of other influence types can also be obtained.

[0084] Step S103-2: Apply the first weight variable values ​​of the multiple first weight variables of the multiple impact types to the first sub-model of the corresponding impact type to obtain the first output parameter value of the first sub-model of the corresponding impact type.

[0085] For example, continuing with the above example, substituting the first weight variable value (ωn1, ωn2, ..., ωnn) of each of the n first weight variables of the nth influence type into the first sub-model of the nth influence type zn = ωn1*x(F-1n)+ωn2*x(F-2n)+......+ωnn*x(F-nn)+an, the first output parameter value of the nth influence type can be obtained; based on this, the first sub-models of other influence types can also be obtained.

[0086] Step S103 - 3 : applying the first output parameter values ​​of the plurality of impact types to the first evaluation models of the corresponding impact types respectively, to obtain the first evaluation parameter values ​​of the first evaluation models of the corresponding impact types.

[0087] For example, continuing with the above example, substituting the first output parameter value of the nth influence type into the first evaluation model g(zn) of the nth influence type of the nth influence type, the first evaluation parameter value of the nth influence type can be obtained; based on this, the first evaluation parameter values ​​of other influence types can also be obtained.

[0088] Step S104, when the first evaluation parameter values ​​of the first evaluation models of the plurality of impact types are all greater than the preset first effective threshold, the first evaluation parameter values ​​of the plurality of impact types are applied to the second sub-model of the corresponding impact type to obtain the second output parameter value of the corresponding impact type.

[0089] For example, the first effective threshold is preset to 98%, and 98% is an empirical value obtained through experiments.

[0090] The embodiment of the present application constrains the first evaluation parameter value of the first evaluation model of each of multiple impact types by presetting a first effective threshold, with the aim of ensuring the validity of the first evaluation parameter value, accelerating data convergence, and improving the efficiency of determining the residual oil film thickness evaluation model.

[0091] If the first evaluation parameter value of the first evaluation model of any impact type is less than or equal to the preset first effective threshold, return to step S101 to reacquire multiple first sample parameter values ​​of multiple impact types until the first evaluation parameter values ​​of the multiple impact types are all greater than the preset second effective threshold.

[0092] For example, continuing the above example, the second sub-model of the nth influence type is y(g(zn)), based on which the second sub-models of other influence types can also be obtained.

[0093] Step S105 , constructing a second evaluation model based on the second output parameter values ​​of the multiple impact types and second weight variables corresponding to the impact types.

[0094] For example, continuing the above example, the second estimation model f(y)=λ1*y(g(z1))+λ2*y(g(z2))+...+λn*y(g(zn)).

[0095] Step S106, combining and verifying the preset second verification models of the plurality of impact types with the second evaluation model to obtain the second weight variable values ​​of the second weight variables of the corresponding impact types and the second evaluation parameter values ​​of the second evaluation model.

[0096] In some specific embodiments, the combining and verifying the preset second verification models of the plurality of impact types with the second evaluation model to obtain the second weight variable values ​​of the second weight variables of the corresponding impact types and the second evaluation parameter values ​​of the second evaluation model includes:

[0097] Step S106-1, based on the second output parameter values ​​of the multiple impact types in the second evaluation model, the preset second verification model of the corresponding impact type is respectively applied, and then combined with the second evaluation model for verification to obtain the second weight variable value of each of the multiple second weight variables of the corresponding impact type.

[0098] For example, continuing with the above example, the preset second verification model of the first influence type f(y(g(z1))=1 / [1+e^(-y(g(z1))]; the preset second verification model of the second influence type f(y(g(z2))=1 / [1+e^(-y(g(z2))]; ...; the preset second verification model of the nth influence type f(y(g(zn))=1 / [1+e^(-y(g(zn))]; are combined with the second evaluation model for verification to obtain the second weight variable values ​​(λ1, λ2, ..., λn) of multiple second weight variables of multiple influence types.

[0099] Step S106 - 2 : applying the second weight variable values ​​of the second weight variables of the plurality of impact types to the second evaluation model to obtain a second evaluation parameter value.

[0100] For example, continuing with the above example, the second weight variable values ​​(λ1, λ2, ..., λn) of multiple second weight variables of multiple influence types are substituted into the second evaluation model f(y) = λ1*y(g(z1))+λ2*y(g(z2))+......+λn*y(g(zn)) to obtain the second evaluation parameter value.

[0101] Step S107 : when the second evaluation parameter value is greater than a preset second effective threshold, generating a predicted oil film thickness value based on second weight variable values ​​of the second weight variables of the plurality of influence types and first output parameter values ​​of the corresponding influence types.

[0102] For example, the second effective threshold is preset to 95%, and 95% is an empirical value obtained through experiments.

[0103] The embodiment of the present application constrains the second evaluation parameter value of the second evaluation model by presetting a second effective threshold, with the aim of ensuring the validity of the second evaluation parameter value, accelerating data convergence, and improving the efficiency of determining the residual oil film thickness evaluation model.

[0104] If the second evaluation parameter value is less than or equal to the preset second effective threshold, the process returns to step S101 to reacquire multiple first sample parameter values ​​of multiple impact types until the second evaluation parameter value is greater than the preset second effective threshold.

[0105] For example, continuing the above example, the second weight variable value of the second weight variable of the multiple impact types and the first output parameter value of the corresponding impact type generate the predicted oil film thickness value, including the following formula:

[0106] t=λ1*(z1)+λ2(z2)+...+λn(zn);

[0107] Where t represents the predicted oil film thickness value.

[0108] Step S108: when the detected multiple measured oil film thickness values ​​verify that the predicted oil film thickness value is valid, determining that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid.

[0109] In some specific embodiments, when the detected multiple measured oil film thickness values ​​verify that the predicted oil film thickness value is valid, determining that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid includes:

[0110] Step S108 - 1 , calculating the error values ​​between the predicted oil film thickness value and the plurality of actually measured oil film thickness values.

[0111] Step S108-2, calculating the ratio of the absolute value of the error value of each of the plurality of measured oil film thickness values ​​to the respective measured oil film thickness value, and obtaining the error rate of each corresponding measured oil film thickness value.

[0112] Step S108-3, counting the target number of the plurality of measured oil film thickness values ​​whose respective error rates are less than a preset error rate threshold.

[0113] For example, the preset error rate threshold is 5%.

[0114] Step S108-4, calculating the error acceptance percentage between the target number and the total number of the plurality of actually measured oil film thickness values.

[0115] For example, if the target quantity is 98 and the total quantity is 100, the error acceptance percentage = 98%.

[0116] Step S108-5: when the error pass percentage is greater than or equal to a preset error pass percentage threshold, it is determined that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid.

[0117] For example, continuing with the above example, the preset error pass percentage threshold is 95%, and the error pass percentage (i.e., 98%) is greater than the preset error pass percentage threshold (i.e., 95%), and it is determined that the evaluation model of the combination of the first evaluation model and the second evaluation model is valid.

[0118] If the error acceptance percentage is less than the preset error acceptance percentage threshold, return to step S101 to re-acquire multiple first sample parameter values ​​of each of the multiple impact types until it can be determined that the evaluation model of the combination of the first evaluation model and the second evaluation model is valid.

[0119] The embodiment of the present application constructs a first evaluation model of a corresponding influence type based on at least a plurality of first sample parameter values ​​of each of the plurality of influence types that affect the residual oil film thickness, wherein the first evaluation model includes a first sub-model; a preset first verification model of each of the plurality of influence types is combined with the respective first evaluation model for verification to obtain a first output parameter value of the first sub-model of the corresponding influence type and a first evaluation parameter value of the first evaluation model of the corresponding influence type; when the first evaluation parameter values ​​of the first evaluation models of each of the plurality of influence types are all greater than a preset first effective threshold, the first evaluation parameter values ​​of each of the plurality of influence types are applied to a second sub-model of the corresponding influence type to obtain a second output parameter value of the corresponding influence type; The second evaluation model is constructed based on the second output parameter values ​​of the multiple influence types and the second weight variables of the corresponding influence types; the preset second verification models of the multiple influence types are combined with the second evaluation model for verification to obtain the second weight variable values ​​of the second weight variables of the corresponding influence types and the second evaluation parameter values ​​of the second evaluation model; when the second evaluation parameter value is greater than the preset second effective threshold, the predicted oil film thickness value is generated based on the second weight variable values ​​of the second weight variables of the multiple influence types and the first output parameter value of the corresponding influence type; when the multiple measured oil film thickness values ​​detected verify that the predicted oil film thickness value is valid, the evaluation model of the combination of the first evaluation model and the second evaluation model is determined to be valid. Through the comprehensive analysis of multi-dimensional factors, the limitations of the traditional single factor verification method are abandoned, and the comprehensive analysis of multi-dimensional factors is adopted to comprehensively consider the influence of multiple key factors such as ambient temperature, plate temperature, cleaning oil temperature, viscosity, impurity rate, plate size, squeeze roller hardness, etc. on the oil film thickness. It can more accurately reveal the interaction relationship between the factors, accurately determine the residual oil film thickness evaluation model, and provide a scientific basis for formulating precise measures. A multiple evaluation model of oil film thickness has been constructed, which can automatically learn and capture the complex nonlinear relationship between various factors, and accurately predict the weight inflection point of the influence of various factors on oil film thickness. It not only improves the accuracy and reliability of the prediction, but also provides strong technical support for optimizing the oil film thickness control strategy of stamping plates. It realizes the precise control and optimization of the oil film thickness of stamping plates, improves production efficiency and improves product quality.

[0120] The present application also provides a device embodiment that is based on the above embodiment, which is used to implement the method steps described in the above embodiment. The explanation based on the same name meaning is the same as the above embodiment, and has the same technical effect as the above embodiment, which will not be repeated here.

[0121] like Figure 2 As shown, the present application provides a device 200 for determining a residual oil film thickness evaluation model, comprising:

[0122] An acquisition unit 201 is used to acquire a plurality of first sample parameter values ​​of respective types of influences affecting the residual oil film thickness;

[0123] A first construction unit 202, configured to respectively construct first evaluation models corresponding to the impact types based at least on a plurality of first sample parameter values ​​of the plurality of impact types, wherein the first evaluation model includes a first sub-model;

[0124] A first obtaining unit 203 is used to combine and verify the preset first verification models of the plurality of impact types with the respective first evaluation models to obtain a first output parameter value of the first sub-model corresponding to the impact type and a first evaluation parameter value of the first evaluation model corresponding to the impact type;

[0125] A second obtaining unit 204 is configured to apply the first evaluation parameter values ​​of the respective multiple impact types to the second sub-model of the corresponding impact type to obtain a second output parameter value of the corresponding impact type when the first evaluation parameter values ​​of the respective first evaluation models of the multiple impact types are all greater than a preset first effective threshold;

[0126] A second construction unit 205, configured to construct a second evaluation model based on the second output parameter values ​​of the plurality of impact types and second weight variables corresponding to the impact types;

[0127] A third obtaining unit 206 is used to combine and verify the preset second verification models of the plurality of impact types with the second evaluation model to obtain the second weight variable values ​​of the second weight variables of the corresponding impact types and the second evaluation parameter values ​​of the second evaluation model;

[0128] A generating unit 207, configured to generate a predicted oil film thickness value based on second weight variable values ​​of the second weight variables of the plurality of influence types and first output parameter values ​​of the corresponding influence types when the second evaluation parameter value is greater than a preset second effective threshold;

[0129] The determination unit 208 is configured to determine that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid when the detected multiple measured oil film thickness values ​​verify that the predicted oil film thickness value is valid.

[0130] Optionally, constructing first assessment models corresponding to the impact types respectively based on at least a plurality of first sample parameter values ​​of the plurality of impact types includes:

[0131] Based on the respective multiple first sample parameter values ​​and respective multiple first weight variables of the multiple impact types, first evaluation models corresponding to the impact types are respectively constructed.

[0132] Optionally, the combining and verifying the preset first verification models of the respective multiple impact types with the respective first evaluation models to obtain the first output parameter value of the first sub-model corresponding to the impact type and the first evaluation parameter value of the first evaluation model corresponding to the impact type includes:

[0133] Applying the first sub-models of the plurality of impact types to the preset first verification model of the corresponding impact type respectively, and then combining and verifying with the first evaluation model of the corresponding impact type, to obtain the first weight variable value of each of the plurality of first weight variables of the corresponding impact type;

[0134] Applying first weight variable values ​​of the plurality of first weight variables of the plurality of influence types to first sub-models of corresponding influence types to obtain first output parameter values ​​of the first sub-models of corresponding influence types;

[0135] The first output parameter values ​​of each of the plurality of impact types are applied to the first evaluation model of the corresponding impact type to obtain the first evaluation parameter value of the first evaluation model of the corresponding impact type.

[0136] Optionally, the combining and verifying the preset second verification models of the respective multiple impact types with the second evaluation model to obtain the second weight variable values ​​of the respective second weight variables of the corresponding impact types and the second evaluation parameter values ​​of the second evaluation model includes:

[0137] Based on the second output parameter values ​​of the multiple impact types in the second evaluation model, the second output parameter values ​​are applied to the preset second verification models of the corresponding impact types respectively, and then combined with the second evaluation model for verification to obtain the second weight variable values ​​of the multiple second weight variables of the corresponding impact type;

[0138] The second weight variable values ​​of the second weight variables of the plurality of impact types are applied to the second evaluation model to obtain second evaluation parameter values.

[0139] Optionally, when the detected multiple measured oil film thickness values ​​verify that the predicted oil film thickness value is valid, determining that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid includes:

[0140] Calculating the error values ​​between the predicted oil film thickness value and the plurality of measured oil film thickness values;

[0141] Calculating the ratio of the absolute value of the error value of each of the plurality of measured oil film thickness values ​​to the respective measured oil film thickness value to obtain the error rate of each corresponding measured oil film thickness value;

[0142] Counting the target number of each of the plurality of measured oil film thickness values ​​whose error rate is less than a preset error rate threshold;

[0143] Calculating the error acceptance percentage between the target quantity and the total quantity of the plurality of actually measured oil film thickness values;

[0144] When the error qualification percentage is greater than or equal to a preset error qualification percentage threshold, it is determined that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid.

[0145] The embodiment of the present application constructs a first evaluation model of a corresponding influence type based on at least a plurality of first sample parameter values ​​of each of the plurality of influence types that affect the residual oil film thickness, wherein the first evaluation model includes a first sub-model; a preset first verification model of each of the plurality of influence types is combined with the respective first evaluation model for verification to obtain a first output parameter value of the first sub-model of the corresponding influence type and a first evaluation parameter value of the first evaluation model of the corresponding influence type; when the first evaluation parameter values ​​of the first evaluation models of each of the plurality of influence types are all greater than a preset first effective threshold, the first evaluation parameter values ​​of each of the plurality of influence types are applied to a second sub-model of the corresponding influence type to obtain a second output parameter value of the corresponding influence type; The second evaluation model is constructed based on the second output parameter values ​​of the multiple influence types and the second weight variables of the corresponding influence types; the preset second verification models of the multiple influence types are combined with the second evaluation model for verification to obtain the second weight variable values ​​of the second weight variables of the corresponding influence types and the second evaluation parameter values ​​of the second evaluation model; when the second evaluation parameter value is greater than the preset second effective threshold, the predicted oil film thickness value is generated based on the second weight variable values ​​of the second weight variables of the multiple influence types and the first output parameter value of the corresponding influence type; when the multiple measured oil film thickness values ​​detected verify that the predicted oil film thickness value is valid, the evaluation model of the combination of the first evaluation model and the second evaluation model is determined to be valid. Through the comprehensive analysis of multi-dimensional factors, the limitations of the traditional single factor verification method are abandoned, and the comprehensive analysis of multi-dimensional factors is adopted to comprehensively consider the influence of multiple key factors such as ambient temperature, plate temperature, cleaning oil temperature, viscosity, impurity rate, plate size, squeeze roller hardness, etc. on the oil film thickness. It can more accurately reveal the interaction relationship between the factors, accurately determine the residual oil film thickness evaluation model, and provide a scientific basis for formulating precise measures. A multiple evaluation model of oil film thickness has been constructed, which can automatically learn and capture the complex nonlinear relationship between various factors, and accurately predict the weight inflection point of the influence of various factors on oil film thickness. It not only improves the accuracy and reliability of the prediction, but also provides strong technical support for optimizing the oil film thickness control strategy of stamping plates. It realizes the precise control and optimization of the oil film thickness of stamping plates, improves production efficiency and improves product quality.

[0146] Example 3

[0147] This embodiment provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method steps described in the above embodiment.

[0148] Example 4

[0149] An embodiment of the present application provides a non-volatile computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions can execute the method steps described in the above embodiment.

[0150] Finally, it should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other. For the system or device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0151] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for determining a residual oil film thickness evaluation model, characterized in that: include: Acquire a plurality of first sample parameter values ​​of respective plurality of influence types affecting the residual oil film thickness; constructing first evaluation models corresponding to the impact types respectively based on at least a plurality of first sample parameter values ​​of the respective impact types, wherein the first evaluation model includes a first sub-model; Combining and verifying the preset first verification models of the respective multiple impact types with the respective first evaluation models to obtain a first output parameter value of the first sub-model corresponding to the impact type and a first evaluation parameter value of the first evaluation model corresponding to the impact type; When the first evaluation parameter values ​​of the first evaluation models of the plurality of impact types are all greater than a preset first effective threshold, applying the first evaluation parameter values ​​of the plurality of impact types to the second sub-model of the corresponding impact type to obtain a second output parameter value of the corresponding impact type; constructing a second evaluation model based on the second output parameter values ​​of the plurality of impact types and second weight variables corresponding to the impact types; Combining and verifying the preset second verification models of the plurality of impact types with the second evaluation model to obtain the second weight variable values ​​of the second weight variables of the corresponding impact types and the second evaluation parameter values ​​of the second evaluation model; When the second evaluation parameter value is greater than a preset second effective threshold, generating a predicted oil film thickness value based on second weight variable values ​​of the second weight variables of the plurality of influence types and first output parameter values ​​of the corresponding influence types; When the detected multiple measured oil film thickness values ​​verify that the predicted oil film thickness value is valid, it is determined that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid.

2. The method according to claim 1, characterized in that The constructing first evaluation models corresponding to the impact types respectively based on at least a plurality of first sample parameter values ​​of the plurality of impact types comprises: Based on the respective multiple first sample parameter values ​​and respective multiple first weight variables of the multiple impact types, first evaluation models corresponding to the impact types are respectively constructed.

3. The method according to claim 2, characterized in that The combining and verifying the preset first verification models of the respective multiple impact types with the respective first evaluation models to obtain the first output parameter value of the first sub-model corresponding to the impact type and the first evaluation parameter value of the first evaluation model corresponding to the impact type includes: Applying the first sub-models of the plurality of impact types to the preset first verification model of the corresponding impact type respectively, and then combining and verifying with the first evaluation model of the corresponding impact type, to obtain the first weight variable value of each of the plurality of first weight variables of the corresponding impact type; Applying first weight variable values ​​of the plurality of first weight variables of the plurality of influence types to first sub-models of corresponding influence types to obtain first output parameter values ​​of the first sub-models of corresponding influence types; The first output parameter values ​​of each of the plurality of impact types are applied to the first evaluation model of the corresponding impact type to obtain the first evaluation parameter value of the first evaluation model of the corresponding impact type.

4. The method according to claim 1, characterized in that: The combining and verifying the preset second verification models of the respective multiple impact types with the second evaluation model to obtain the second weight variable values ​​of the respective second weight variables of the corresponding impact types and the second evaluation parameter values ​​of the second evaluation model includes: Based on the second output parameter values ​​of the multiple impact types in the second evaluation model, the second output parameter values ​​are applied to the preset second verification models of the corresponding impact types respectively, and then combined with the second evaluation model for verification to obtain the second weight variable values ​​of the multiple second weight variables of the corresponding impact type; The second weight variable values ​​of the second weight variables of the plurality of impact types are applied to the second evaluation model to obtain second evaluation parameter values.

5. The method according to claim 1, characterized in that When the detected multiple measured oil film thickness values ​​verify that the predicted oil film thickness value is valid, determining that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid includes: Calculating the error values ​​between the predicted oil film thickness value and the plurality of measured oil film thickness values; Calculating the ratio of the absolute value of the error value of each of the plurality of measured oil film thickness values ​​to the respective measured oil film thickness value to obtain the error rate of each corresponding measured oil film thickness value; Counting the target number of each of the plurality of measured oil film thickness values ​​whose error rate is less than a preset error rate threshold; Calculating the error acceptance percentage between the target quantity and the total quantity of the plurality of actually measured oil film thickness values; When the error qualification percentage is greater than or equal to a preset error qualification percentage threshold, it is determined that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid.

6. A device for determining a residual oil film thickness evaluation model, characterized in that: include: An acquisition unit, used for acquiring a plurality of first sample parameter values ​​of respective types of influences affecting the residual oil film thickness; A first construction unit, configured to respectively construct first evaluation models corresponding to the impact types based at least on the respective first sample parameter values ​​of the multiple impact types, wherein the first evaluation model includes a first sub-model; A first obtaining unit is used to combine and verify the preset first verification models of the respective multiple impact types with the respective first evaluation models to obtain a first output parameter value of the first sub-model corresponding to the impact type and a first evaluation parameter value of the first evaluation model corresponding to the impact type; A second obtaining unit is used to apply the first evaluation parameter values ​​of the respective multiple impact types to the second sub-model of the corresponding impact type to obtain a second output parameter value of the corresponding impact type when the first evaluation parameter values ​​of the respective first evaluation models of the multiple impact types are all greater than a preset first effective threshold value; A second construction unit, configured to construct a second evaluation model based on second output parameter values ​​of the plurality of impact types and second weight variables corresponding to the impact types; A third obtaining unit is used to combine and verify the preset second verification models of the plurality of impact types with the second evaluation model to obtain the second weight variable values ​​of the second weight variables of the corresponding impact types and the second evaluation parameter values ​​of the second evaluation model; A generating unit, configured to generate a predicted oil film thickness value based on second weight variable values ​​of the second weight variables of the plurality of influence types and first output parameter values ​​of the corresponding influence types when the second evaluation parameter value is greater than a preset second effective threshold; A determination unit is used to determine that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid when the detected multiple measured oil film thickness values ​​verify that the predicted oil film thickness value is valid.

7. The device according to claim 6, characterized in that The constructing first evaluation models corresponding to the impact types respectively based on at least a plurality of first sample parameter values ​​of the plurality of impact types comprises: Based on the respective multiple first sample parameter values ​​and respective multiple first weight variables of the multiple impact types, first evaluation models corresponding to the impact types are respectively constructed.

8. The device according to claim 7, characterized in that The combining and verifying the preset first verification models of the respective multiple impact types with the respective first evaluation models to obtain the first output parameter value of the first sub-model corresponding to the impact type and the first evaluation parameter value of the first evaluation model corresponding to the impact type includes: Applying the first sub-models of the plurality of impact types to the preset first verification model of the corresponding impact type respectively, and then combining and verifying with the first evaluation model of the corresponding impact type, to obtain the first weight variable value of each of the plurality of first weight variables of the corresponding impact type; Applying first weight variable values ​​of the plurality of first weight variables of the plurality of influence types to first sub-models of corresponding influence types to obtain first output parameter values ​​of the first sub-models of corresponding influence types; The first output parameter values ​​of each of the plurality of impact types are applied to the first evaluation model of the corresponding impact type to obtain the first evaluation parameter value of the first evaluation model of the corresponding impact type.

9. The device according to claim 6, characterized in that The combining and verifying the preset second verification models of the respective multiple impact types with the second evaluation model to obtain the second weight variable values ​​of the respective second weight variables of the corresponding impact types and the second evaluation parameter values ​​of the second evaluation model includes: Based on the second output parameter values ​​of the multiple impact types in the second evaluation model, the second output parameter values ​​are applied to the preset second verification models of the corresponding impact types respectively, and then combined with the second evaluation model for verification to obtain the second weight variable values ​​of the multiple second weight variables of the corresponding impact type; The second weight variable values ​​of the second weight variables of the plurality of impact types are applied to the second evaluation model to obtain second evaluation parameter values.

10. The device according to claim 6, characterized in that When the detected multiple measured oil film thickness values ​​verify that the predicted oil film thickness value is valid, determining that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid includes: Calculating the error values ​​between the predicted oil film thickness value and the plurality of measured oil film thickness values; Calculating the ratio of the absolute value of the error value of each of the plurality of measured oil film thickness values ​​to the respective measured oil film thickness value to obtain the error rate of each corresponding measured oil film thickness value; Counting the target number of each of the plurality of measured oil film thickness values ​​whose error rate is less than a preset error rate threshold; Calculating the error acceptance percentage between the target quantity and the total quantity of the plurality of actually measured oil film thickness values; When the error qualification percentage is greater than or equal to a preset error qualification percentage threshold, it is determined that the evaluation model formed by combining the first evaluation model and the second evaluation model is valid.

Images