CN106960100B - Technological parameter reasoning method and device - Google Patents
Technological parameter reasoning method and device Download PDFInfo
- Publication number
- CN106960100B CN106960100B CN201710192283.8A CN201710192283A CN106960100B CN 106960100 B CN106960100 B CN 106960100B CN 201710192283 A CN201710192283 A CN 201710192283A CN 106960100 B CN106960100 B CN 106960100B
- Authority
- CN
- China
- Prior art keywords
- value
- values
- input
- delta
- groups
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/18—Manufacturability analysis or optimisation for manufacturability
Abstract
The invention discloses a technological parameter reasoning method and a device thereof, comprising received input technological parameters; screening data in a preset parameter database according to the input process parameters to obtain a screening result of which the proximity degree of the preset parameter database to the input process parameters is in a preset range, wherein the preset parameter database comprises a plurality of groups of pre-obtained process parameters and corresponding parameters to be inferred; and analyzing the relationship between each process parameter and the parameter to be inferred according to the screening result, and obtaining the parameter to be inferred corresponding to the input process parameter according to the relationship for subsequent processing according to the parameter to be inferred. The method can reason the process parameters in various ranges, and has the advantages of wide application range, small calculated amount and short consumed time.
Description
Technical Field
The invention relates to the technical field of bent plate processing, in particular to a technological parameter reasoning method and a device thereof.
Background
The line heating plate is a hull complex outer plate processing and forming process commonly adopted by shipyards at home and abroad. The process has the advantages of large technical difficulty, multiple influencing factors, large difficulty of operation skills, low efficiency and unstable quality, and is completed by depending on experienced workers and manual operation according to experience.
The automatic reasoning problem of the line heating process parameters is to reason the required heating speed according to the process parameters during the plate bending process, so as to process the line heating plate (that is, to reason one parameter according to a plurality of parameters). The finite element simulation and prediction method is characterized in that a finite element model of the line heating plate is established, and the heating speed is calculated in a reverse-deduction mode according to a simulation result, but the analysis process of the method needs tens of thousands of iterative calculations, so that the time is wasted, the efficiency is low, and the calculation amount is large. The BP network process parameter forecasting method needs to establish a multilayer BP network model and train a network by using a large amount of experimental data, but the method lacks generalization capability and can only accurately forecast parameters aiming at a processing outer plate with specific parameters.
Therefore, how to provide a method and a device for reasoning process parameters with wide application and small calculation amount is a problem that needs to be solved by those skilled in the art.
Disclosure of Invention
The invention aims to provide a technological parameter reasoning method and a device thereof, which can reason technological parameters in various ranges, and have the advantages of wide application range, small calculated amount and short consumed time.
In order to solve the technical problem, the invention provides a process parameter reasoning method, which comprises the following steps:
receiving an input of process parameters;
screening data in a preset parameter database according to the input process parameters to obtain a screening result of which the proximity degree of the preset parameter database to the input process parameters is in a preset range, wherein the preset parameter database comprises a plurality of groups of pre-obtained process parameters and corresponding parameters to be inferred;
and analyzing the relationship between each process parameter and the parameter to be inferred according to the screening result, and obtaining the parameter to be inferred corresponding to the input process parameter according to the relationship for subsequent processing according to the parameter to be inferred.
Preferably, the process parameters comprise the plate thickness Hp of a processed outer plate of the line heating plate, the length L of a flame path and the target shrinkage Delta B; the parameter to be inferred is heating speed.
Preferably, the preset parameter database comprises a plurality of sets of data obtained in advance, each set of data comprises a set of Hp value, L value and Δ B value and V value corresponding to the Hp value, L value and Δ B value;
the process of screening data in a preset parameter database according to the input process parameters to obtain a screening result in the preset parameter database, wherein the proximity degree of the screening result to the input process parameters is within a preset range, analyzing the relationship between each process parameter and the heating speed according to the screening result, and obtaining the heating speed corresponding to the input process parameters according to the relationship specifically comprises the following steps:
comparing the input process parameters with the data in the preset parameter database to determine N groups of data which are the same as or closest to the input Hp value in the preset parameter database;
selecting N groups of data with L same as or closest to the input L value from the N groups of data to obtain N ^2 groups of data;
selecting a plurality of groups of data with the same delta B value as the input delta B from the N ^2 groups of data as the screening result, and if the data with the same delta B value as the input delta B does not exist, taking all the N ^2 groups of data as the screening result;
dividing the screening results into a plurality of large groups according to different Hp values, and dividing each large group into a plurality of small groups according to different L values;
drawing an empirical regression curve of the delta B and the V of each group according to the delta B and the V in each group, and obtaining a first output value VBi corresponding to the input delta B value on the empirical regression curve of the delta B and the V by a curve interpolation method; i is more than or equal to 1 and less than or equal to N1, and N1 is the total number of subgroups;
drawing an empirical regression curve of L and VBi in each large group according to the first output quantity VBi of each small group and the L value corresponding to the small group, and obtaining a second output quantity VLj corresponding to the input L value on the empirical regression curve of L and VBi by a curve interpolation method; j is more than or equal to 1 and less than or equal to N2, and N2 is the total number of the large groups;
and drawing an empirical regression curve of Hp and VLj according to the second output quantity VLj of each large group and the Hp value corresponding to the large group, and obtaining a third output quantity V3 corresponding to the input Hp value on the empirical regression curve of Hp and VLj by a curve interpolation method, wherein V3 is the heating speed corresponding to the input process parameter.
Preferably, the preset parameter database comprises 3 layers of relational data tables, and each layer of relational data table is a data table with two rows and multiple columns; the first row of the first layer relational data table stores the Hp values obtained in advance, and the second row stores the address pointers of the next layer relational data table corresponding to all the Hp values; a first row in the second-layer relational data table stores L values obtained in advance, and a second row stores address pointers of a next-layer relational data table corresponding to each L value; the first row in the third layer of relational data table stores a delta B value obtained in advance, and the second row of relational data table stores a corresponding V value;
the process of screening data in a preset parameter database according to the input process parameters to obtain a screening result in the preset parameter database, wherein the proximity degree of the screening result to the input process parameters is within a preset range, analyzing the relationship between each process parameter and the heating speed according to the screening result, and obtaining the heating speed corresponding to the input process parameters according to the relationship specifically comprises the following steps:
acquiring N Hp values which are the same as or closest to the input Hp value from a first-layer relational data table, and extracting address pointers of a second-layer relational data table corresponding to the N Hp values;
inquiring all L values pointed by the address pointers of the second layer corresponding to the N Hp values, acquiring N L values which are the same as or closest to the input L values, and extracting the address pointers of the third layer corresponding to the N L values;
querying all delta B values pointed by the address pointers of the third layer corresponding to the N L values to obtain a delta B value same as the input delta B and a corresponding V value, and if the delta B value same as the input delta B does not exist, obtaining all delta B values pointed by the address pointers corresponding to the N L values and a corresponding V value; the obtained V value and a group of delta B value, L value and Hp value corresponding to the V value are the screening result;
dividing the screening results into a plurality of large groups according to different Hp values, and dividing each large group into a plurality of small groups according to different L values;
drawing an empirical regression curve of the delta B and the V of each group according to the delta B and the V in each group, and obtaining a first output value VBi corresponding to the input delta B value on the empirical regression curve of the delta B and the V by a curve interpolation method; i is more than or equal to 1 and less than or equal to N1, and N1 is the total number of subgroups;
drawing an empirical regression curve of L and VBi in each large group according to the first output quantity VBi of each small group and the L value corresponding to the small group, and obtaining a second output quantity VLj corresponding to the input L value on the empirical regression curve of L and VBi by a curve interpolation method; j is more than or equal to 1 and less than or equal to N2, and N2 is the total number of the large groups;
and drawing an empirical regression curve of Hp and VLj according to the second output quantity VLj of each large group and the Hp value corresponding to the large group, and obtaining a third output quantity V3 corresponding to the input Hp value on the empirical regression curve of Hp and VLj by a curve interpolation method, wherein V3 is the heating speed corresponding to the input process parameter.
Preferably, N is 4.
In order to solve the above technical problem, the present invention further provides a process parameter inference device, including:
the receiving module is used for receiving the input process parameters;
the screening module is used for screening data in a preset parameter database according to the input process parameters to obtain a screening result of which the proximity degree of the preset parameter database to the input process parameters is in a preset range, wherein the preset parameter database comprises a plurality of groups of pre-obtained process parameters and corresponding parameters to be inferred;
and the reasoning module is used for analyzing the relationship between each process parameter and the parameter to be inferred according to the screening result, obtaining the parameter to be inferred corresponding to the input process parameter according to the relationship, and carrying out subsequent processing according to the parameter to be inferred.
Preferably, the preset parameter database comprises a plurality of groups of data obtained in advance, each group of data comprises a group of process parameters and corresponding parameters to be inferred, the process parameters comprise a plate thickness Hp value of a processed outer plate of a line heating plate, a length L value of a flame path and a target shrinkage delta B value, and the parameters to be inferred are heating speed V; the screening module specifically comprises:
the first comparison unit is used for performing parameter comparison on the input process parameters and data in the preset parameter database, and determining N groups of data which are the same as or closest to the input Hp value in the preset parameter database;
the second comparison unit is used for selecting N groups of data with L same as or closest to the input L value from the N groups of data to obtain N ^2 groups of data;
a third comparing unit, configured to select, from the N ^2 group of data, a plurality of groups of data having a Δ B value identical to an input Δ B as the screening result, and if there is no data identical to the input Δ B, take all of the N ^2 group of data as the screening result;
the reasoning module specifically comprises:
the grouping unit is used for dividing the screening results into a plurality of large groups according to different Hp values and dividing each large group into a plurality of small groups according to different L values;
the first curve analysis unit is used for drawing an empirical regression curve of the delta B and the V of each group according to the delta B and the V in each group, and obtaining a first output value VBi corresponding to the input delta B value on the empirical regression curve of the delta B and the V by a curve interpolation method; i is more than or equal to 1 and less than or equal to N1, and N1 is the total number of subgroups;
a second curve analysis unit, configured to draw an empirical regression curve of L and VBi in each large group according to the first output quantity VBi of each small group and the L value corresponding to the small group, and obtain a second output quantity VLj corresponding to the input L value on the empirical regression curve of L and VBi by using a curve interpolation method; j is more than or equal to 1 and less than or equal to N2, and N2 is the total number of the large groups;
and a third curve analysis unit, configured to draw an empirical regression curve of Hp and VLj according to each of the large groups of second output quantities VLj and Hp values corresponding to the large groups, and obtain a third output quantity V3 corresponding to the input Hp value on the empirical regression curve of Hp and VLj by using a curve interpolation method, where V3 is a heating speed corresponding to the input process parameter.
Preferably, the preset parameter database comprises 3 layers of relational data tables, and each layer of relational data table is a data table with two rows and multiple columns; the first row of the first layer relational data table stores the Hp values obtained in advance, and the second row stores the address pointers of the next layer relational data table corresponding to all the Hp values; a first row in the second-layer relational data table stores L values obtained in advance, and a second row stores address pointers of a next-layer relational data table corresponding to each L value; the first row in the third layer of relational data table stores a delta B value obtained in advance, and the second row of relational data table stores a corresponding V value;
the screening module specifically comprises:
the first comparison unit is used for acquiring N Hp values which are the same as or closest to the input Hp value from the first-layer relational data table, and extracting address pointers of the second-layer relational data table corresponding to the N Hp values;
a second comparing unit, configured to query all L values pointed by the address pointers of the second layer corresponding to the N Hp values, obtain N L values that are the same as or closest to the input L value, and extract address pointers of a third layer corresponding to the N L values;
a third comparing unit, configured to query all Δ B values pointed by address pointers of a third layer corresponding to the N L values, to obtain Δ B values identical to the input Δ B and V values corresponding to the Δ B values, and if there is no Δ B value identical to the input Δ B, to obtain all Δ B values pointed by address pointers corresponding to the N L values and V values corresponding to the Δ B values; the obtained V value and a group of delta B value, L value and Hp value corresponding to the V value are the screening result;
the reasoning module specifically comprises:
the grouping unit is used for dividing the screening results into a plurality of large groups according to different Hp values and dividing each large group into a plurality of small groups according to different L values;
the first curve analysis unit is used for drawing an empirical regression curve of the delta B and the V of each group according to the delta B and the V in each group, and obtaining a first output value VBi corresponding to the input delta B value on the empirical regression curve of the delta B and the V by a curve interpolation method; i is more than or equal to 1 and less than or equal to N1, and N1 is the total number of subgroups;
a second curve analysis unit, which draws an empirical regression curve of L and VBi in each large group according to the first output quantity VBi of each small group and the L value corresponding to the small group, and obtains a second output quantity VLj corresponding to the input L value on the empirical regression curve of L and VBi by a curve interpolation method; j is more than or equal to 1 and less than or equal to N2, and N2 is the total number of the large groups;
and a third curve analysis unit, which draws an empirical regression curve of Hp and VLj according to the second output quantity VLj of each large group and the Hp value corresponding to the large group, and obtains a third output quantity V3 corresponding to the input Hp value on the empirical regression curve of Hp and VLj by a curve interpolation method, wherein the V3 is the heating speed corresponding to the input process parameter.
The invention provides a process parameter reasoning method and a device thereof, which are characterized in that after input process parameters are received, partial data which are closer to the input process parameters are screened out from a plurality of groups of process parameters which are obtained in advance from a preset parameter database and the corresponding parameters to be inferred, then the relationship between each process parameter and the parameters to be inferred in the partial data is analyzed, and the parameters to be inferred which correspond to the input process parameters are obtained according to the relationship. Therefore, the invention does not limit the parameter range of the process parameters, and as long as the data in the preset parameter database is enough, the process parameters in any range can be inferred by the method to the corresponding parameters to be inferred, so that the application range is wide; the calculation amount of the method depends on the number of the screening results and the types of the process parameters, the iteration times are less, and the calculation amount is small, so that the calculation time is reduced.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed in the prior art and the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a flow chart of a process parameter inference method provided by the present invention;
FIG. 2 is a flow chart of the process of another process parameter inference method provided by the present invention;
FIG. 3 is a flow chart of the process of another process parameter inference method provided by the present invention;
FIG. 4 is a schematic diagram of the reasoning process for obtaining the final heating speed according to the screening result in the present invention;
fig. 5 is a schematic structural diagram of a process parameter inference device provided by the present invention.
Detailed Description
The core of the invention is to provide a process parameter reasoning method and a device thereof, which can reason process parameters in various ranges, and have the advantages of wide application range, small calculated amount and short consumed time.
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention provides a process parameter reasoning method, which is shown in figure 1, wherein figure 1 is a flow chart of the process parameter reasoning method provided by the invention; the method comprises the following steps:
step s 101: receiving an input of process parameters;
step s 102: screening data in a preset parameter database according to the input process parameters to obtain a screening result of which the proximity degree of the preset parameter database to the input process parameters is in a preset range, wherein the preset parameter database comprises a plurality of groups of pre-obtained process parameters and corresponding parameters to be inferred;
the data in the preset parameter database is data obtained in advance through experiments or practical experience.
Step s 103: and analyzing the relationship between each process parameter and the parameter to be inferred according to the screening result, and obtaining the parameter to be inferred corresponding to the input process parameter according to the relationship for subsequent plate bending processing according to the parameter to be inferred.
Preferably, the process parameters comprise the plate thickness Hp of a processed outer plate of the line heating plate, the length L of a flame path and the target shrinkage Delta B; the parameter to be inferred is the heating speed. Of course, the steps of the present invention may also be applied to other processing devices or scenarios that determine a parameter to be inferred according to a plurality of process parameters, and the present invention is not limited thereto.
Further, the preset parameter database comprises a plurality of groups of data obtained in advance, and each group of data comprises a group of Hp values, L values and Δ B values and V values corresponding to the Hp values, L values and Δ B values;
that is, in this embodiment, data is stored in the preset parameter database in the form of data groups, and each group of data includes four numbers of Hp, L, Δ B, and V.
It can be understood that different values of V are required for different sets of data, and different values of V are obtained if one of Hp, L, and Δ B is different.
Referring to fig. 2, fig. 2 is a flow chart illustrating a process of another process parameter inference method provided by the present invention;
step s102, the process of step s103 is specifically:
step s 201: comparing the input process parameters with data in a preset parameter database, and determining N groups of data which are the same as or closest to the input Hp value in the preset parameter database;
step s 202: selecting N groups of data with L same as or closest to the input L value from the N groups of data to obtain N ^2 groups of data;
step s 203: selecting a plurality of groups of data with the delta B value being the same as the input delta B from the N ^2 groups of data as a screening result, and if the data with the delta B value being the same as the input delta B does not exist, taking all the N ^2 groups of data as the screening result;
step s 301: dividing the screening results into a plurality of large groups according to different Hp values, and dividing each large group into a plurality of small groups according to different L values;
step s 302: drawing an empirical regression curve of the delta B and the V of each group according to the delta B and the V in each group, and obtaining a first output quantity VBi corresponding to an input delta B value on the empirical regression curve of the delta B and the V by a curve interpolation method; i is more than or equal to 1 and less than or equal to N1, and N1 is the total number of subgroups;
the empirical regression curve is drawn by using a nonlinear least squares method.
Step s 303: drawing an empirical regression curve of L and VBi in each large group according to the first output quantity VBi of each small group and the L value corresponding to the small group, and obtaining a second output quantity VLj corresponding to the L value input on the empirical regression curve of each L and VBi by a curve interpolation method; j is more than or equal to 1 and less than or equal to N2, and N2 is the total number of the large groups;
step s 304: and drawing an empirical regression curve of Hp and VLj according to the second output VLj of each large group and the Hp value corresponding to the large group, and obtaining a third output V3 corresponding to the Hp value input on the empirical regression curve of Hp and VLj by a curve interpolation method, wherein V3 is the heating speed corresponding to the input process parameter.
In addition, on the empirical regression curve of each Δ B and V obtained by the curve interpolation method, the process of the first output quantity VBi corresponding to the input Δ B value is essentially to determine the empirical regression curve, and when Δ B is equal to the input Δ B value, the corresponding V value is the first output quantity VBi. Other steps can be obtained by the same method.
In step s301, the groups may be divided into a plurality of major groups according to the difference of the L values, and each major group may be divided into a plurality of minor groups according to the difference of the Hp values; the groups are divided into a large group and a small group according to the value of which process parameter, and the present invention is not particularly limited. Correspondingly, in step s302, an empirical regression curve is drawn according to two variables in each group to be finally divided to obtain a first output quantity, then, in step s303, an empirical regression curve is drawn according to the first output quantity and parameters according to which the group is divided to obtain a second output quantity, and finally, in step s304, an empirical regression curve is drawn according to the second output quantity and parameters according to which the group is divided to obtain a final output value.
In addition, when the process parameters include 2 or 4 or other numbers of parameter types, when the screening results are grouped, the grouping times are correspondingly reduced or increased, and the final grouping result needs to ensure that the obtained variables in the minimum group only include V and one parameter. For example, when the process parameters include 4, assuming that the fourth parameter is X, each large group is divided into a plurality of small groups according to the difference of L values, and each small group is further divided into a plurality of minimum groups according to the difference of Δ B, where two variables included in each minimum group are X and V. Correspondingly, when curve drawing is performed according to each divided group, one more layer of iteration is needed. Of course, the present invention does not limit how many types of parameters the process parameters include.
In another embodiment, the preset parameter database comprises 3 layers of relational data tables, and each layer of relational data table is a data table with two rows and multiple columns; the first row of the first layer relational data table stores the Hp values obtained in advance, and the second row stores the address pointers of the next layer relational data table corresponding to all the Hp values; a first row in the second-layer relational data table stores L values obtained in advance, and a second row stores address pointers of a next-layer relational data table corresponding to each L value; the first row in the third layer of relational data table stores a delta B value obtained in advance, and the second row of relational data table stores a corresponding V value;
of course, the above arrangement of the numerical values is only a preferred scheme, and the L value may also be placed in the first layer of relationship data table, and what type of process parameter is specifically placed in each layer of relationship data table is not specifically limited in the present invention.
Referring to fig. 3 and 4, fig. 3 is a flow chart illustrating a process of another process parameter inference method provided by the present invention. FIG. 4 is a schematic diagram of the reasoning process for obtaining the final heating speed according to the screening result in the present invention;
step s102, the process of step s103 is specifically:
step s 401: acquiring N Hp values which are the same as or closest to the input Hp value from a first-layer relational data table, and extracting address pointers of a second-layer relational data table corresponding to the N Hp values;
at this point, each Hp value typically corresponds to multiple address pointers.
Step s 402: inquiring all L values pointed by the address pointers of the second layer corresponding to the N Hp values, acquiring N L values which are the same as or closest to the input L values, and extracting the address pointers of the third layer corresponding to the N L values;
step s 403: inquiring all delta B values pointed by the address pointers of the third layer corresponding to the N L values to obtain a delta B value which is the same as the input delta B and a corresponding V value, and if the delta B value which is the same as the input delta B does not exist, obtaining all delta B values pointed by the address pointers corresponding to the N L values and the corresponding V values; the obtained V value and a group of delta B value, L value and Hp value corresponding to the V value are the screening results;
it can be understood that an L value is obtained according to the address pointer corresponding to the Hp value, and then a Δ B value is obtained according to the pointer of the L value, where the V value corresponding to the Δ B value is the heating speed when the inputs are the Hp value, the L value, and the Δ B value.
Step s 501: dividing the screening results into a plurality of large groups according to different Hp values, and dividing each large group into a plurality of small groups according to different L values;
step s 502: drawing an empirical regression curve of the delta B and the V of each group according to the delta B and the V in each group, and obtaining a first output quantity VBi corresponding to an input delta B value on the empirical regression curve of the delta B and the V by a curve interpolation method; i is more than or equal to 1 and less than or equal to N1, and N1 is the total number of subgroups;
step s 503: drawing an empirical regression curve of L and VBi in each large group according to the first output quantity VBi of each small group and the L value corresponding to the small group, and obtaining a second output quantity VLj corresponding to the L value input on the empirical regression curve of each L and VBi by a curve interpolation method; j is more than or equal to 1 and less than or equal to N2, and N2 is the total number of the large groups;
step s 504: and drawing an empirical regression curve of Hp and VLj according to the second output VLj of each large group and the Hp value corresponding to the large group, and obtaining a third output V3 corresponding to the Hp value input on the empirical regression curve of Hp and VLj by a curve interpolation method, wherein V3 is the heating speed corresponding to the input process parameter.
It should be noted that, when the process parameters include 2 or 4 or other types of parameters, the number of layers in the relationship data table included in the preset parameter database is also decreased or increased accordingly. When the screening results are grouped, the grouping times are correspondingly reduced or increased, and the final grouping result needs to ensure that the obtained variables in the minimum group only contain V and one parameter; for example, when the process parameters include 4, assuming that the fourth parameter is X, each large group is divided into a plurality of small groups according to the difference of L values, and each small group is further divided into a plurality of minimum groups according to the difference of Δ B, where two variables included in each minimum group are X and V.
The present invention is not particularly limited, and the order of grouping is determined according to the screening order in the screening process. For example, the Hp value is in the first layer, the Hp value meeting the condition is screened firstly during screening, and then the L value is screened according to the Hp value, so that the large group is divided according to the Hp value, and the small group is divided according to the L value; and if the L values meeting the conditions are screened firstly during screening, and then the Hp values are screened according to the L values, then the large groups are divided according to the L values, and the small groups are divided according to the Hp values. The grouping order is also determined in the above manner when the included parameter types of the process parameters increase or decrease.
Wherein N is 4. It can be understood that the value of N determines the number of the screening results and the proximity of each set of data in the screening results to the input process parameters, and if N is too large, a large error may exist between the relationship obtained by the analysis in the step s103 and the relationship between the input process plant parameters, so that the accuracy of the heating speed obtained by final inference is low, and if N is too small, the number of samples in the screening results is too small, and the accuracy is also insufficient. Of course, the present invention is not limited to specific values of N.
The invention provides a process parameter reasoning method, which comprises the steps of receiving input process parameters, screening partial data which are close to the input process parameters from a plurality of groups of process parameters which are obtained in advance from a preset parameter database and the corresponding parameters to be inferred, analyzing the relationship between each process parameter and the parameters to be inferred in the partial data, and obtaining the parameters to be inferred corresponding to the input process parameters according to the relationship. Therefore, the invention does not limit the parameter range of the process parameters, and as long as the data in the preset parameter database is enough, the process parameters in any range can be inferred by the method to the corresponding parameters to be inferred, so that the application range is wide; the calculation amount of the method depends on the number of the screening results and the types of the process parameters, the iteration times are less, and the calculation amount is small, so that the calculation time is reduced.
The invention also provides a process parameter reasoning device, which is shown in fig. 5, and fig. 5 is a schematic structural diagram of the process parameter reasoning device provided by the invention. The device includes:
the receiving module 1 is used for receiving input process parameters;
the screening module 2 is used for screening data in a preset parameter database according to the input process parameters to obtain a screening result of which the proximity degree of the preset parameter database to the input process parameters is within a preset range, wherein the preset parameter database comprises a plurality of groups of pre-obtained process parameters and corresponding parameters to be inferred;
and the reasoning module 3 is used for analyzing the relationship between each process parameter and the parameter to be inferred according to the screening result, obtaining the parameter to be inferred corresponding to the input process parameter according to the relationship, and processing the parameter to be inferred according to the subsequent parameter to be inferred.
The preset parameter database comprises a plurality of groups of data which are obtained in advance, each group of data comprises a group of process parameters and corresponding parameters to be inferred, the process parameters comprise a plate thickness Hp value of a processed outer plate of a line heating plate, a length L value of a flame path and a target shrinkage delta B value, and the parameters to be inferred are heating speed V; the screening module 2 specifically comprises:
the first comparison unit is used for comparing the input process parameters with data in a preset parameter database to determine N groups of data which are the same as or closest to the input Hp value in the preset parameter database;
the second comparison unit is used for selecting N groups of data with the same or the closest L value to the input L value from the N groups of data to obtain N ^2 groups of data;
the third comparison unit is used for selecting a plurality of groups of data with the delta B value being the same as the input delta B from the N ^2 groups of data as a screening result, and if the data with the same as the input delta B does not exist, all the N ^2 groups of data are used as the screening result;
the inference module 3 specifically includes:
the grouping unit is used for dividing the screening results into a plurality of large groups according to different Hp values and dividing each large group into a plurality of small groups according to different L values;
the first curve analysis unit is used for drawing an empirical regression curve of the delta B and the V of each group according to the delta B and the V in each group, and obtaining a first output value VBi corresponding to an input delta B value on the empirical regression curve of the delta B and the V by a curve interpolation method; i is more than or equal to 1 and less than or equal to N1, and N1 is the total number of subgroups;
the second curve analysis unit is used for drawing an empirical regression curve of L and VBi in each large group according to the first output quantity VBi of each small group and the L value corresponding to the small group, and obtaining a second output quantity VLj corresponding to the L value input on the empirical regression curve of each L and VBi through a curve interpolation method; j is more than or equal to 1 and less than or equal to N2, and N2 is the total number of the large groups;
and the third curve analysis unit is used for drawing an empirical regression curve of Hp and VLj according to the second output quantity VLj of each large group and the Hp value corresponding to the large group, and obtaining a third output quantity V3 corresponding to the Hp value input on the empirical regression curve of Hp and VLj by a curve interpolation method, wherein V3 is the heating speed corresponding to the input process parameter.
In another embodiment for line heating, the preset parameter database comprises 3 layers of relational data tables, and each layer of relational data table is a data table with two rows and multiple columns; the first row of the first layer relational data table stores the Hp values obtained in advance, and the second row stores the address pointers of the next layer relational data table corresponding to all the Hp values; a first row in the second-layer relational data table stores L values obtained in advance, and a second row stores address pointers of a next-layer relational data table corresponding to each L value; the first row in the third layer of relational data table stores a delta B value obtained in advance, and the second row of relational data table stores a corresponding V value;
the screening module 2 specifically comprises:
the first comparison unit is used for acquiring N Hp values which are the same as or closest to the input Hp value from the first-layer relational data table, and extracting address pointers of the second-layer relational data table corresponding to the N Hp values;
the second comparison unit is used for inquiring all L values pointed by the address pointers of the second layer corresponding to the N Hp values, acquiring N L values which are the same as or closest to the input L value, and extracting the address pointers of the third layer corresponding to the N L values;
a third comparing unit, configured to query all Δ B values pointed by the address pointers of the third layer corresponding to the N L values, to obtain Δ B values identical to the input Δ B and V values corresponding to the Δ B values, and if there is no Δ B value identical to the input Δ B, to obtain all Δ B values pointed by the address pointers corresponding to the N L values and V values corresponding to the Δ B values; the obtained V value and a group of delta B value, L value and Hp value corresponding to the V value are the screening results;
the inference module 3 specifically includes:
the grouping unit is used for dividing the screening results into a plurality of large groups according to different Hp values and dividing each large group into a plurality of small groups according to different L values;
the first curve analysis unit is used for drawing an empirical regression curve of the delta B and the V of each group according to the delta B and the V in each group, and obtaining a first output value VBi corresponding to an input delta B value on the empirical regression curve of the delta B and the V by a curve interpolation method; i is more than or equal to 1 and less than or equal to N1, and N1 is the total number of subgroups;
the second curve analysis unit is used for drawing an empirical regression curve of L and VBi in each large group according to the first output quantity VBi of each small group and the L value corresponding to the small group, and obtaining a second output quantity VLj corresponding to the L value input on the empirical regression curve of each L and VBi through a curve interpolation method; j is more than or equal to 1 and less than or equal to N2, and N2 is the total number of the large groups;
and the third curve analysis unit is used for drawing an empirical regression curve of Hp and VLj according to the second output quantity VLj of each large group and the Hp value corresponding to the large group, and obtaining a third output quantity V3 corresponding to the Hp value input on the empirical regression curve of Hp and VLj by a curve interpolation method, wherein V3 is the heating speed corresponding to the input process parameter.
The invention provides a process parameter reasoning device, which is used for receiving input process parameters, screening partial data which is closer to the input process parameters from a plurality of groups of process parameters which are obtained in advance from a preset parameter database and the corresponding parameters to be reasoned, analyzing the relationship between each process parameter and the parameters to be reasoned in the partial data, and further obtaining the parameters to be reasoned corresponding to the input process parameters according to the relationship. Therefore, the invention does not limit the parameter range of the process parameters, and as long as the data in the preset parameter database is enough, the process parameters in any range can be inferred by the method to the corresponding parameters to be inferred, so that the application range is wide; the calculation amount of the method depends on the number of the screening results and the types of the process parameters, the iteration times are less, and the calculation amount is small, so that the calculation time is reduced.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
It is further noted that, in the present specification, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. A process parameter reasoning method is characterized by comprising the following steps:
receiving an input of process parameters;
screening data in a preset parameter database according to the input process parameters to obtain a screening result of which the proximity degree of the preset parameter database to the input process parameters is in a preset range, wherein the preset parameter database comprises a plurality of groups of pre-obtained process parameters and corresponding parameters to be inferred;
and analyzing the relationship between each process parameter and the parameter to be inferred according to the screening result, and obtaining the parameter to be inferred corresponding to the input process parameter according to the relationship for subsequent processing according to the parameter to be inferred.
2. The method of claim 1, wherein the process parameters include a plate thickness Hp of a processed outer plate of a line heating plate, a length L of a flame path, a target shrinkage Δ B; the parameter to be inferred is heating speed.
3. The method according to claim 2, wherein the database of preset parameters comprises a plurality of sets of data obtained in advance, each set of data comprises a set of Hp value, L value and Δ B value and V value corresponding to the Hp value, L value and Δ B value;
the process of screening data in a preset parameter database according to the input process parameters to obtain a screening result in the preset parameter database, wherein the proximity degree of the screening result to the input process parameters is within a preset range, analyzing the relationship between each process parameter and the heating speed according to the screening result, and obtaining the heating speed corresponding to the input process parameters according to the relationship specifically comprises the following steps:
comparing the input process parameters with the data in the preset parameter database to determine N groups of data which are the same as or closest to the input Hp value in the preset parameter database;
selecting N groups of data with L same as or closest to the input L value from the N groups of data to obtain N ^2 groups of data;
selecting a plurality of groups of data with the same delta B value as the input delta B from the N ^2 groups of data as the screening result, and if the data with the same delta B value as the input delta B does not exist, taking all the N ^2 groups of data as the screening result;
dividing the screening results into a plurality of large groups according to different Hp values, and dividing each large group into a plurality of small groups according to different L values;
drawing an empirical regression curve of the delta B and the V of each group according to the delta B and the V in each group, and obtaining a first output value VBi corresponding to the input delta B value on the empirical regression curve of the delta B and the V by a curve interpolation method; i is more than or equal to 1 and less than or equal to N1, and N1 is the total number of subgroups;
drawing an empirical regression curve of L and VBi in each large group according to the first output quantity VBi of each small group and the L value corresponding to the small group, and obtaining a second output quantity VLj corresponding to the input L value on the empirical regression curve of L and VBi by a curve interpolation method; j is more than or equal to 1 and less than or equal to N2, and N2 is the total number of the large groups;
and drawing an empirical regression curve of Hp and VLj according to the second output quantity VLj of each large group and the Hp value corresponding to the large group, and obtaining a third output quantity V3 corresponding to the input Hp value on the empirical regression curve of Hp and VLj by a curve interpolation method, wherein V3 is the heating speed corresponding to the input process parameter.
4. The method according to claim 3, wherein the database of predetermined parameters comprises 3 layers of relational data tables, each layer of the relational data tables being data tables of two rows and multiple columns; the first row of the first layer relational data table stores the Hp values obtained in advance, and the second row stores the address pointers of the next layer relational data table corresponding to all the Hp values; a first row in the second-layer relational data table stores L values obtained in advance, and a second row stores address pointers of a next-layer relational data table corresponding to each L value; the first row in the third layer of relational data table stores a delta B value obtained in advance, and the second row of relational data table stores a corresponding V value;
the process of screening data in a preset parameter database according to the input process parameters to obtain a screening result in the preset parameter database, wherein the proximity degree of the screening result to the input process parameters is within a preset range, analyzing the relationship between each process parameter and the heating speed according to the screening result, and obtaining the heating speed corresponding to the input process parameters according to the relationship specifically comprises the following steps:
acquiring N Hp values which are the same as or closest to the input Hp value from a first-layer relational data table, and extracting address pointers of a second-layer relational data table corresponding to the N Hp values;
inquiring all L values pointed by the address pointers of the second layer corresponding to the N Hp values, acquiring N L values which are the same as or closest to the input L values, and extracting the address pointers of the third layer corresponding to the N L values;
querying all delta B values pointed by the address pointers of the third layer corresponding to the N L values to obtain a delta B value same as the input delta B and a corresponding V value, and if the delta B value same as the input delta B does not exist, obtaining all delta B values pointed by the address pointers corresponding to the N L values and a corresponding V value; the obtained V value and a group of delta B value, L value and Hp value corresponding to the V value are the screening result;
dividing the screening results into a plurality of large groups according to different Hp values, and dividing each large group into a plurality of small groups according to different L values;
drawing an empirical regression curve of the delta B and the V of each group according to the delta B and the V in each group, and obtaining a first output value VBi corresponding to the input delta B value on the empirical regression curve of the delta B and the V by a curve interpolation method; i is more than or equal to 1 and less than or equal to N1, and N1 is the total number of subgroups;
drawing an empirical regression curve of L and VBi in each large group according to the first output quantity VBi of each small group and the L value corresponding to the small group, and obtaining a second output quantity VLj corresponding to the input L value on the empirical regression curve of L and VBi by a curve interpolation method; j is more than or equal to 1 and less than or equal to N2, and N2 is the total number of the large groups;
and drawing an empirical regression curve of Hp and VLj according to the second output quantity VLj of each large group and the Hp value corresponding to the large group, and obtaining a third output quantity V3 corresponding to the input Hp value on the empirical regression curve of Hp and VLj by a curve interpolation method, wherein V3 is the heating speed corresponding to the input process parameter.
5. A method according to claim 3 or 4, wherein N is 4.
6. A process parameter reasoning device is characterized by comprising:
the receiving module is used for receiving the input process parameters;
the screening module is used for screening data in a preset parameter database according to the input process parameters to obtain a screening result of which the proximity degree of the preset parameter database to the input process parameters is in a preset range, wherein the preset parameter database comprises a plurality of groups of pre-obtained process parameters and corresponding parameters to be inferred;
and the reasoning module is used for analyzing the relationship between each process parameter and the parameter to be inferred according to the screening result, obtaining the parameter to be inferred corresponding to the input process parameter according to the relationship, and carrying out subsequent processing according to the parameter to be inferred.
7. The device according to claim 6, wherein the preset parameter database comprises a plurality of sets of data obtained in advance, each set of data comprises a set of process parameters and corresponding parameters to be inferred, the process parameters comprise a plate thickness Hp value of a processed outer plate of a line heating plate, a length L value of a flame path and a target shrinkage Delta B value, and the parameters to be inferred are heating speed V; the screening module specifically comprises:
the first comparison unit is used for performing parameter comparison on the input process parameters and data in the preset parameter database, and determining N groups of data which are the same as or closest to the input Hp value in the preset parameter database;
the second comparison unit is used for selecting N groups of data with L same as or closest to the input L value from the N groups of data to obtain N ^2 groups of data;
a third comparing unit, configured to select, from the N ^2 group of data, a plurality of groups of data having a Δ B value identical to an input Δ B as the screening result, and if there is no data identical to the input Δ B, take all of the N ^2 group of data as the screening result;
the reasoning module specifically comprises:
the grouping unit is used for dividing the screening results into a plurality of large groups according to different Hp values and dividing each large group into a plurality of small groups according to different L values;
the first curve analysis unit is used for drawing an empirical regression curve of the delta B and the V of each group according to the delta B and the V in each group, and obtaining a first output value VBi corresponding to the input delta B value on the empirical regression curve of the delta B and the V by a curve interpolation method; i is more than or equal to 1 and less than or equal to N1, and N1 is the total number of subgroups;
a second curve analysis unit, configured to draw an empirical regression curve of L and VBi in each large group according to the first output quantity VBi of each small group and the L value corresponding to the small group, and obtain a second output quantity VLj corresponding to the input L value on the empirical regression curve of L and VBi by using a curve interpolation method; j is more than or equal to 1 and less than or equal to N2, and N2 is the total number of the large groups;
and a third curve analysis unit, configured to draw an empirical regression curve of Hp and VLj according to each of the large groups of second output quantities VLj and Hp values corresponding to the large groups, and obtain a third output quantity V3 corresponding to the input Hp value on the empirical regression curve of Hp and VLj by using a curve interpolation method, where V3 is a heating speed corresponding to the input process parameter.
8. The apparatus according to claim 6, wherein the database of predetermined parameters comprises 3 layers of relational data tables, each layer of the relational data tables being data tables of two rows and a plurality of columns; the first row of the first layer relational data table stores the Hp values obtained in advance, and the second row stores the address pointers of the next layer relational data table corresponding to all the Hp values; a first row in the second-layer relational data table stores L values obtained in advance, and a second row stores address pointers of a next-layer relational data table corresponding to each L value; the first row in the third layer of relational data table stores a delta B value obtained in advance, and the second row of relational data table stores a corresponding V value;
the screening module specifically comprises:
the first comparison unit is used for acquiring N Hp values which are the same as or closest to the input Hp value from the first-layer relational data table, and extracting address pointers of the second-layer relational data table corresponding to the N Hp values;
a second comparing unit, configured to query all L values pointed by the address pointers of the second layer corresponding to the N Hp values, obtain N L values that are the same as or closest to the input L value, and extract address pointers of a third layer corresponding to the N L values;
a third comparing unit, configured to query all Δ B values pointed by address pointers of a third layer corresponding to the N L values, to obtain Δ B values identical to the input Δ B and V values corresponding to the Δ B values, and if there is no Δ B value identical to the input Δ B, to obtain all Δ B values pointed by address pointers corresponding to the N L values and V values corresponding to the Δ B values; the obtained V value and a group of delta B value, L value and Hp value corresponding to the V value are the screening result;
the reasoning module specifically comprises:
the grouping unit is used for dividing the screening results into a plurality of large groups according to different Hp values and dividing each large group into a plurality of small groups according to different L values;
the first curve analysis unit is used for drawing an empirical regression curve of the delta B and the V of each group according to the delta B and the V in each group, and obtaining a first output value VBi corresponding to the input delta B value on the empirical regression curve of the delta B and the V by a curve interpolation method; i is more than or equal to 1 and less than or equal to N1, and N1 is the total number of subgroups;
a second curve analysis unit, which draws an empirical regression curve of L and VBi in each large group according to the first output quantity VBi of each small group and the L value corresponding to the small group, and obtains a second output quantity VLj corresponding to the input L value on the empirical regression curve of L and VBi by a curve interpolation method; j is more than or equal to 1 and less than or equal to N2, and N2 is the total number of the large groups;
and a third curve analysis unit, which draws an empirical regression curve of Hp and VLj according to the second output quantity VLj of each large group and the Hp value corresponding to the large group, and obtains a third output quantity V3 corresponding to the input Hp value on the empirical regression curve of Hp and VLj by a curve interpolation method, wherein the V3 is the heating speed corresponding to the input process parameter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710192283.8A CN106960100B (en) | 2017-03-28 | 2017-03-28 | Technological parameter reasoning method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710192283.8A CN106960100B (en) | 2017-03-28 | 2017-03-28 | Technological parameter reasoning method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106960100A CN106960100A (en) | 2017-07-18 |
| CN106960100B true CN106960100B (en) | 2021-01-26 |
Family
ID=59471523
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710192283.8A Active CN106960100B (en) | 2017-03-28 | 2017-03-28 | Technological parameter reasoning method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106960100B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102682139A (en) * | 2011-03-17 | 2012-09-19 | 齐亮 | Method for forming shell plate curve of ship body |
| KR101246065B1 (en) * | 2010-11-25 | 2013-03-26 | 삼성중공업 주식회사 | Determination system of heating shape and position for triangle heating and method thereof |
| CN104359415A (en) * | 2014-10-31 | 2015-02-18 | 广东工业大学 | Measuring method and system of angular deformation for line heating and cooling |
-
2017
- 2017-03-28 CN CN201710192283.8A patent/CN106960100B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101246065B1 (en) * | 2010-11-25 | 2013-03-26 | 삼성중공업 주식회사 | Determination system of heating shape and position for triangle heating and method thereof |
| CN102682139A (en) * | 2011-03-17 | 2012-09-19 | 齐亮 | Method for forming shell plate curve of ship body |
| CN104359415A (en) * | 2014-10-31 | 2015-02-18 | 广东工业大学 | Measuring method and system of angular deformation for line heating and cooling |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106960100A (en) | 2017-07-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111208782B (en) | Data processing method and device for machine tool spindle state prediction | |
| CN111242464A (en) | Ceramic bathroom quality management method based on SPC | |
| Tang et al. | Bicriteria scheduling on a single batching machine with job transportation and deterioration considerations | |
| JP2023511464A (en) | SYSTEM, METHOD, AND COMPUTER-READABLE MEDIUM FOR ANALYSIS OF PRODUCT FAILURE FACTORS | |
| CN113642938A (en) | Intelligent production management method and system | |
| EP4080429A1 (en) | Technology readiness level determination method and system based on science and technology big data | |
| CN106960100B (en) | Technological parameter reasoning method and device | |
| CN114626745A (en) | Scheduling plan generation method, system, medium and electronic terminal | |
| US20220129447A1 (en) | Method, system, device and medium for querying product history | |
| CN110909888A (en) | Method, device and equipment for constructing generic decision tree and readable storage medium | |
| CN112199407B (en) | Data packet ordering method, device, equipment and storage medium | |
| CN111626497B (en) | People flow prediction method, device, equipment and storage medium | |
| Kreitlein et al. | E| Benchmark–A Pioneering Method for Process Planning and Sustainable Manufacturing Strategies | |
| CN112861064B (en) | Social credit evaluation source data processing method, system, terminal and medium | |
| CN109884893B (en) | Multi-process variable dynamic time lag estimation method | |
| CN111857192A (en) | Method and device for regulating ejector | |
| JP6236422B2 (en) | Information processing device | |
| CN110991641A (en) | Oil reservoir type analysis method and device and electronic equipment | |
| CN116451916A (en) | Simulation method and device for production process and storage medium | |
| CN113822537A (en) | Product matching method and device | |
| CN115990698A (en) | Welding method, device, equipment and medium | |
| CN116402158A (en) | Machine learning model screening method | |
| CN105787580A (en) | Periodic report refreshing completion time prediction method and device | |
| CN117406677A (en) | Wear-resistant steel quality data processing method, device, equipment and medium | |
| CN116467298A (en) | Data compliance detection method, device and medium based on artificial intelligence |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |