CN116679639A - Optimization method and system of metal product production control system - Google Patents
Optimization method and system of metal product production control system Download PDFInfo
- Publication number
- CN116679639A CN116679639A CN202310606647.8A CN202310606647A CN116679639A CN 116679639 A CN116679639 A CN 116679639A CN 202310606647 A CN202310606647 A CN 202310606647A CN 116679639 A CN116679639 A CN 116679639A
- Authority
- CN
- China
- Prior art keywords
- production
- metal product
- production line
- product production
- value
- 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.)
- Granted
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 868
- 239000002184 metal Substances 0.000 title claims abstract description 346
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000005457 optimization Methods 0.000 title claims abstract description 21
- 238000009826 distribution Methods 0.000 claims abstract description 48
- 230000003993 interaction Effects 0.000 claims description 73
- 230000006399 behavior Effects 0.000 claims description 28
- 230000002452 interceptive effect Effects 0.000 claims description 21
- 230000002787 reinforcement Effects 0.000 claims description 21
- 238000003860 storage Methods 0.000 claims description 17
- 238000005259 measurement Methods 0.000 claims description 5
- 230000003111 delayed effect Effects 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 238000013461 design Methods 0.000 description 19
- 238000004891 communication Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000001351 cycling effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 229910001220 stainless steel Inorganic materials 0.000 description 1
- 239000010935 stainless steel Substances 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
- G05B19/41865—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by job scheduling, process planning, material flow
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/32—Operator till task planning
- G05B2219/32252—Scheduling production, machining, job shop
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the application provides an optimization method and system of a metal product production control system, which are characterized in that the production cost value of a plurality of metal product production lines in a production logic network is obtained, namely, the influence of production distribution times on the distribution state of the production lines is combined in the direct production distribution behavior, so that the accuracy of production control distribution is improved.
Description
Technical Field
The embodiment of the application relates to the technical field of machine learning, in particular to an optimization method and system of a metal product production control system.
Background
The metal product industry includes structural metal product manufacture, metal tool manufacture, container and metal packaging container manufacture, stainless steel and similar commodity metal product manufacture, and the like. With the development of technology, metal products are increasingly widely applied to various fields of industry, agriculture and human life, and great value is created for society. With the continuous upgrading and industrialization of the consumption structure and the acceleration of the urban process, the method drives the rapid growth of industries such as steel and other metal products.
Based on the above, in the process of controlling the production of the metal product, production control allocation of a production line needs to be performed by combining the issued production event of the metal product, and how to improve the reliability of the production control allocation of the production event of the target metal product, so that the optimization accuracy of the production control system of the metal product is higher, which is a technical problem to be solved currently.
Disclosure of Invention
In order to at least overcome the above-mentioned shortcomings in the prior art, an object of an embodiment of the present application is to provide a method and a system for optimizing a metal product production control system.
In a first aspect, an embodiment of the present application provides a method for optimizing a metal product production control system, applied to an optimizing system of a metal product production control system, the method including:
Receiving a production control allocation task of a metal article production control system, the production control allocation task configured to instruct production control allocation of a target metal article production event;
acquiring a production logic network formed by a metal product production line and a production line interaction site;
according to the production routing node of the target metal product production event, a plurality of production line cost values respectively corresponding to a plurality of metal product production lines in the production logic network are obtained, and the production line cost values reflect the production distribution times of direct production distribution behaviors between a reference production line interaction site which is most matched with the target metal product production event and the metal product production lines;
obtaining a plurality of production line entropy values corresponding to the plurality of metal product production lines respectively;
configuring a plurality of production allocation features respectively corresponding to the plurality of metal product production lines according to the production line cost values and the production line entropy values, wherein the production allocation features respectively reflect allocation consumption parameter values of the target metal product production event allocated to the plurality of metal product production lines;
and based on the production distribution characteristics, carrying out production control distribution on the target metal product production event according to a reinforcement learning model.
In a possible implementation manner of the first aspect, the obtaining, by the production routing node according to the target metal product production event, a plurality of production line cost values corresponding to a plurality of metal product production lines in the production logic network includes:
for a first metal product line of the plurality of metal product lines, obtaining one or more production nodes via which the target metal product production event is distributed from the reference line interaction site to the first metal product line;
according to a production line cost estimation network, taking a production load level of a first production node in the one or more production nodes and a corresponding recorded production resource amount when the target metal product production event is respectively distributed to the first production node as network loading data, and generating a production line cost value corresponding to the first production node;
outputting the added value of the production line cost value corresponding to the one or more production nodes as the production line cost value corresponding to the first metal product production line;
taking the production load level of a first production node in the one or more production nodes and the corresponding recorded production resource amount when the target metal product production event is respectively distributed to the first production node as network loading data, generating a production line cost value corresponding to the first production node, comprising:
Determining a metric value A1 according to the production load level of the first production node, wherein the metric value A1 is not less than 0 and less than 1;
outputting a fusion value of the metric value A1 and a first weight coefficient of the production line cost estimation network as a metric value A2;
and fusing the added value of the metric value A2 and the metric value A1 with the production resource quantity to generate a production line cost value corresponding to the first production node.
In a possible implementation manner of the first aspect, the determining the metric value A1 according to the production load level of the first production node includes:
outputting a power of a target value of the production load level as the measurement value A1 if the production load level is not less than 0 and not more than 1, wherein the target value reflects a second weight coefficient of the production line cost estimation network;
and if the production load level is greater than 1 and not greater than m, outputting a power of a target value of a subtraction value of m and the production load level as the measurement value A1.
In a possible implementation manner of the first aspect, before the obtaining, by the production routing node according to the target metal product production event, a plurality of production line cost values corresponding to a plurality of metal product production lines in the production logic network, the method further includes:
Obtaining first a priori gathered data for production control allocation of a priori metal product production event, the first a priori gathered data comprising one or more reference production nodes via which the a priori metal product production event is allocated to a metal product production line;
determining a production load level of a first reference production node of the one or more reference production nodes;
determining the corresponding recorded production resource amount when the prior metal product production events are respectively distributed to the first reference production node;
and iteratively updating the production line cost estimation network by taking the production load level of the first reference production node carrying the production line cost value and the production resource quantity which is correspondingly recorded when the prior metal product production event is respectively distributed to the first reference production node as one network learning data in a network learning data sequence.
In a possible implementation manner of the first aspect, the obtaining a plurality of production line cost values corresponding to a plurality of metal product production lines in the production logic network respectively includes:
acquiring a first target period value, wherein the first target period value is a remaining period parameter value for adjusting the production line cost value between two related stations in the production logic network at the next interval;
If the first target period value is not smaller than the threshold period value, acquiring the cost values of the multiple production lines which are adjusted in the last round;
if the first target period value is smaller than the threshold period value, estimating a production line cost value corresponding to each of the plurality of metal product production lines under a current adjustment period node according to a production line cost estimation network, and generating a plurality of production line cost values;
the production line cost values are regularly adjusted according to the first period parameter value.
In a possible implementation manner of the first aspect, the obtaining a plurality of production line entropy values corresponding to the plurality of metal product production lines respectively includes:
taking the production characteristic vector of the target metal product production event and the production line label data of a second metal product production line in the plurality of metal product production lines as network loading data, and estimating a production line entropy value of the second metal product production line according to a production line entropy value estimation network generated according to a local random gradient descent strategy;
the production feature vector includes: production event features, production trigger features, and production progress features.
In a possible implementation manner of the first aspect, the configuring a plurality of production allocation features corresponding to the plurality of metal product production lines according to the plurality of production line cost values and the plurality of production line entropy values includes:
acquiring the initial residual production task quantity of the target metal product production event and the station identification of the reference production line interaction station from the production control distribution task;
aiming at a third metal product production line in the metal product production lines, acquiring a second target period value and a first production task amount which are correspondingly recorded in the process that the target metal product production event is distributed to the third metal product production line according to the production line cost value corresponding to the third metal product production line;
acquiring a third target period value and an executable production task amount which are delayed to be executed after the target metal product production event is distributed to the third metal product production line according to the production line entropy value of the third metal product production line;
and configuring production allocation characteristics corresponding to the third metal product production line according to the station identification of the reference production line interaction station, the initial residual production task quantity, the second target period value, the first production task quantity, the third target period value and the executable production task quantity.
In a possible implementation manner of the first aspect, the performing, based on the plurality of production allocation features, production control allocation of the target metal product production event according to a reinforcement learning model includes:
loading the plurality of production allocation features into the reinforcement learning model to generate a first production allocation feature reflecting a minimum allocation consumption parameter value of a backward production line interaction site in a direct production allocation behavior of the plurality of metal product production lines to which the target metal product production event is allocated;
if the backward production line interaction site of the reference production line interaction site is a metal product production line in the plurality of metal product production lines, determining the backward production line interaction site as a metal product production line allocated for the target metal product production event, and outputting the allocation behavior of the target metal product production event as the target metal product production event from the initial current production line interaction site to the backward production line interaction site via the executed behavior;
if the backward production line interactive site is not a metal product production line of the plurality of metal product production lines, determining the backward production line interactive site as a current production line interactive site, and determining a backward production line interactive site for the current production line interactive site until the backward production line interactive site determined for the current production line interactive site is a metal product production line of the plurality of metal product production lines based on:
According to the current production line interaction site, a plurality of reference production line cost values corresponding to the plurality of metal product production lines are obtained, wherein the plurality of reference production line cost values are production line cost values of direct production distribution behaviors of the target metal product production event from the current production line interaction site to the plurality of metal product production lines respectively;
configuring a plurality of reference production distribution characteristics respectively corresponding to the plurality of metal product production lines according to the cost values of the plurality of reference production lines and the entropy values of the plurality of production lines;
inputting the multiple reference production allocation features into a reinforcement learning model to generate a backward production line interaction site determined for the current production line interaction site;
the step of obtaining a plurality of reference production line cost values corresponding to the plurality of metal product production lines respectively according to the current production line interaction station comprises the following steps:
acquiring a fourth target period value, wherein the fourth target period value is a remaining period parameter value for adjusting the production line cost value between two related stations in the production logic network at the next interval;
if the fourth target period value is equal to 0 and the target metal product production event is not distributed to the metal product production lines, estimating a production line cost value corresponding to each of the plurality of metal product production lines under the current adjustment period node according to a production line cost estimation network, and generating the plurality of reference production line cost values;
If the fourth target period value is greater than 0 and the target metal product production event is not assigned to a metal product line, the multiple reference line cost values for the last round of adjustment are obtained.
In a possible implementation manner of the first aspect, before the performing production control allocation on the target metal product production event according to the reinforcement learning model based on the plurality of production allocation features, the method further includes:
obtaining second priori gathering data for production control distribution of a priori metal product production event, wherein the second priori gathering data comprises a production routing node of the priori metal product production event and a plurality of priori metal product production lines in the production logic network; according to the production routing node of the prior metal product production event, a plurality of reference production line cost values corresponding to the prior metal product production lines respectively are obtained, and the reference production line cost values reflect the production distribution times of direct production distribution behaviors between a production line interaction site which is most matched with the prior metal product production event and the metal product production lines;
Obtaining a plurality of entropy values of reference production lines corresponding to the prior metal product production lines respectively;
configuring a plurality of reference production distribution characteristics respectively corresponding to the prior metal product production lines according to the cost values of the reference production lines and the entropy values of the reference production lines;
and iteratively updating the reinforcement learning model according to the production allocation characteristics carrying the backward production line interaction site with the minimum allocation consumption parameter value in the direct production allocation behavior reflecting that the prior metal product production event is allocated to the plurality of metal product production lines and the plurality of reference production allocation characteristics.
In a second aspect, an embodiment of the present application further provides an optimization system of a metal product production control system, the optimization system of a metal product production control system including a processor and a machine-readable storage medium having stored therein machine-executable instructions loaded and executed by the processor to implement the optimization method of a metal product production control system in any one of the possible implementations of the first aspect.
In any aspect of the foregoing, the accuracy of production control allocation is improved by obtaining the production line cost values corresponding to the metal product production lines in the production logic network, that is, by combining the influence of the production allocation times on the allocation status of the production lines in the direct production allocation behavior, and by configuring the production allocation features reflecting the allocation consumption parameter values of the target metal product production event allocated to the metal product production lines according to the production line cost values and the corresponding production line entropy values of the metal product production lines, the influence of the production line entropy value status of each metal product production line on the production cost of the target metal product production event is combined, so that the reliability of production control allocation on the target metal product production event is improved, and the optimization accuracy of the metal product production control system is further improved.
Drawings
For a clearer description of the technical solutions of the embodiments of the present application, reference will be made to the accompanying drawings, which are needed to be activated in the embodiments, and it should be understood that the following drawings only illustrate some embodiments of the present application, and therefore should not be considered as limiting the scope, and other related drawings can be extracted by those skilled in the art without the inventive effort.
FIG. 1 is a schematic flow chart of an optimizing method of a metal product production control system according to an embodiment of the present application;
fig. 2 is a schematic block diagram of an optimizing system of a metal product production control system for implementing the optimizing method of the metal product production control system according to an embodiment of the present application.
Detailed Description
The following description is presented to enable one of ordinary skill in the art to make and use the application and is provided in the context of a particular application and its requirements. It will be apparent to those having ordinary skill in the art that various changes can be made to the disclosed embodiments and that the general principles defined herein may be applied to other embodiments and applications without departing from the principles and scope of the application. Therefore, the present application is not limited to the described embodiments, but is to be accorded the widest scope consistent with the claims.
Fig. 1 is a schematic flow chart of an optimizing method of a metal product production control system according to an embodiment of the present application, and the optimizing method of the metal product production control system is described in detail below.
Step 101, receiving a production control allocation task of a metal product production control system, wherein the production control allocation task is configured to instruct production control allocation of a target metal product production event;
102, obtaining a production logic network formed by a metal product production line and a production line interaction station;
step 103, according to the production route node of the target metal product production event, obtaining a plurality of production line cost values corresponding to a plurality of metal product production lines in the production logic network respectively, wherein the plurality of production line cost values reflect the production distribution times of direct production distribution behaviors between a reference production line interaction site which is most matched with the target metal product production event and the plurality of metal product production lines;
104, obtaining a plurality of production line entropy values corresponding to the plurality of metal product production lines respectively;
step 105, configuring a plurality of production allocation features corresponding to the plurality of metal product production lines according to the plurality of production line cost values and the plurality of production line entropy values, wherein the plurality of production allocation features respectively reflect allocation consumption parameter values of the target metal product production event allocated to the plurality of metal product production lines;
And 106, performing production control distribution on the target metal product production event according to the reinforcement learning model based on the production distribution characteristics.
The embodiment can acquire a production control distribution task, determine a production logic network, determine a plurality of production line cost values respectively corresponding to a reference production line interaction site which is most matched with the target metal product production event and the plurality of metal product production lines, determine production line entropy values of the plurality of metal product production lines, configure a plurality of production distribution characteristics respectively reflecting distribution consumption parameter values of the target metal product production event distributed to the plurality of metal product production lines, configure a reinforcement learning model and distribute production control of the target metal product production event according to the reinforcement learning model.
The direct production allocation behavior is a production allocation behavior with a minimum value of an estimated allocation consumption parameter between a reference line interaction site and a metal product line that best matches the target metal product production event. The direct production allocation behavior may be directly estimated from a priori gathered data of production control allocations. The plurality of production line entropy values respectively reflect production load occupancy values of the plurality of metal product production lines in the production process.
In addition, a plurality of production distribution characteristics respectively reflecting the distribution consumption parameter values of the target metal product production event distributed to the plurality of metal product production lines are configured according to the plurality of production line cost values and the plurality of production line entropy values respectively corresponding to the plurality of metal product production lines, so that the influence of the production line entropy value state of each metal product production line on the production cost of the target metal product production event is combined, the reliability of production control distribution of the target metal product production event is improved, and the optimization accuracy of a metal product production control system is higher.
In an exemplary design concept, step 103 may include:
for a first metal product line of the plurality of metal product lines, obtaining one or more production nodes via which the target metal product production event is distributed from the reference line interaction site to the first metal product line;
According to the production line cost estimation network, taking the production load level of a first production node in the one or more production nodes and the corresponding recorded production resource quantity of the target metal product production event when passing through the first production node as network loading data to generate a production line cost value corresponding to the first production node;
and outputting the added value of the production line cost value corresponding to the one or more production nodes as the production line cost value corresponding to the first metal product production line.
The first metal product line may be any one of a plurality of metal product lines.
The production line cost value of the first production node is estimated according to the production line cost estimation network, one or more production nodes through which the target metal product production event is distributed to the first metal product production line from the reference production line interaction site are obtained, namely, the direct production distribution behavior from the reference production line interaction site to the first metal product production line is divided into one or more production nodes with finer granularity, the production load level of the first production node in the one or more production nodes and the corresponding recorded production resource quantity when the target metal product production event passes through the first production node are used as network loading data, the production line cost value of the first production node is estimated, the accuracy of the distribution consumption parameter value of the target metal product production event distributed to the metal product production line is higher, the reliability of production control distribution of the target metal product production event is improved, and the optimization accuracy of a metal product production control system is higher.
In an exemplary design concept, the determining a production line cost value corresponding to a first production node may be described in the following embodiments:
determining a metric value A1 according to the production load level of the first production node, wherein the metric value A1 is not less than 0 and less than 1; outputting a fusion value of the metric value A1 and a first weight coefficient of the production line cost estimation network as a metric value A2; and fusing the added value of the metric value A2 and the metric value A1 with the production resource quantity to generate a production line cost value corresponding to the first production node.
The first weight coefficient of the line cost estimation network may be determined from past production control data.
In an exemplary design concept, the measurement value A1 is determined according to the production load level of the first production node, see the following embodiments:
outputting a power of the target value of the production load level as the metric value A1 if the production load level is not less than 0 and not more than 1, wherein the target value reflects a second weight coefficient of the production line cost estimation network;
if the production load level is greater than 1 and not greater than m, outputting a power of a target value of a subtraction value of m and the production load level as the metric value A1.
The second weight coefficient of the line cost estimation network may be determined from past production control data.
In an exemplary design concept, prior to step 103, the above embodiment may further include the steps of:
obtaining first a priori gathered data for production control allocation of a priori metal product production event, the first a priori gathered data comprising one or more reference production nodes via which the a priori metal product production event is allocated to a metal product production line;
determining a production load level of a first reference production node of the one or more reference production nodes;
determining a corresponding recorded production resource amount when the prior metal product production event passes through the first reference production node;
and training the production line cost estimation network by taking the production load level of the first reference production node carrying the production line cost value and the corresponding recorded production resource quantity when the prior metal product production event passes through the first reference production node as one network learning data in a network learning data sequence.
In an exemplary design concept, the above embodiment may further include the steps of:
and in response to the reinforcement learning model, performing production control allocation of the target metal product production event, determining a fourth target period value and an executable production task amount for the target metal product production event to wait after allocation to a fourth metal product production line of the plurality of metal product production lines if the target metal product production event is allocated to the fourth metal product production line. The fourth metal product line may be any one of a plurality of metal product lines.
In an exemplary design concept, step 103 may include:
acquiring a first target period value, wherein the first target period value is a remaining period parameter value for adjusting the production line cost value between two related stations in the production logic network at the next interval;
and if the first target period value is not smaller than the threshold period value, acquiring the cost values of the production lines which are adjusted in the last round.
And if the first target period value is smaller than the threshold period value, estimating the production line cost value corresponding to each of the plurality of metal product production lines under the current adjustment period node according to the production line cost estimation network, and generating the plurality of production line cost values.
That is, a first target period value is obtained, if the first target period value is not less than a threshold period value, the plurality of line cost values estimated by the line cost estimation network last time are obtained, and if the first target period value is less than the threshold period value, the plurality of line cost values estimated by the current adjustment period node of the line cost estimation network are obtained.
The threshold period value reflects an estimated target period value for which a target metal product production event is assigned to a metal product production line; the estimated target period value may be obtained by a priori gathering data of the production control assignments.
In one exemplary design approach, the plurality of line cost values are regularly adjusted based on the first periodic parameter value.
In an exemplary design concept, step 104 may include:
and taking the production characteristic vector of the target metal product production event and the production line label data of a second metal product production line in the plurality of metal product production lines as network loading data, and estimating the production line entropy value of the second metal product production line according to a production line entropy value estimation network generated according to a local random gradient descent strategy.
In an alternative design concept, the production feature vector may include: production event features, production trigger features, and production progress features.
The second metal product line may be any one of the plurality of metal product lines.
In an exemplary design concept, step 105 may include:
acquiring the initial residual production task quantity of the target metal product production event and the station identification of the interaction station of the reference production line from the production control distribution task;
aiming at a third metal product production line in the metal product production lines, acquiring a second target period value and a first production task amount which are correspondingly recorded in the process that the target metal product production event is distributed to the third metal product production line according to the production line cost value corresponding to the third metal product production line;
Acquiring a third target period value and an executable production task amount which are delayed to be executed after the target metal product production event is distributed to the third metal product production line according to the production line entropy value of the third metal product production line;
and configuring production allocation characteristics corresponding to the third metal product production line according to the station identification of the reference production line interaction station, the initial remaining production task quantity, the second target period value, the first production task quantity, the third target period value and the executable production task quantity. The third metal product line may be any one of a plurality of metal product lines.
In an exemplary design concept, step 106 may include:
inputting the plurality of production allocation features into the reinforcement learning model to generate a first production allocation feature reflecting a backward production line interaction site with a minimum allocation consumption parameter value in a direct production allocation behavior in which the target metal product production event is allocated to the plurality of metal product production lines;
if the backward production line interaction site of the reference production line interaction site is a metal product production line of the plurality of metal product production lines, determining the backward production line interaction site as a metal product production line allocated for the target metal product production event, and outputting the target metal product production event as an allocation of the target metal product production event from the initial current production line interaction site to the backward production line interaction site via an executed action;
If the backward production line interactive site is not a metal product production line of the plurality of metal product production lines, determining the backward production line interactive site as a current production line interactive site, and determining the backward production line interactive site for the current production line interactive site until the backward production line interactive site determined for the current production line interactive site is a metal product production line of the plurality of metal product production lines based on:
according to the current production line interaction site, a plurality of reference production line cost values corresponding to the plurality of metal product production lines are obtained, wherein the plurality of reference production line cost values are the production line cost values of direct production distribution behaviors of the target metal product production event from the current production line interaction site to the plurality of metal product production lines respectively;
configuring a plurality of reference production distribution characteristics corresponding to the plurality of metal product production lines respectively according to the cost values of the plurality of reference production lines and the entropy values of the plurality of production lines;
the plurality of reference production allocation features are input into a reinforcement learning model to generate a backward production line interaction site determined for the current production line interaction site.
The method comprises the steps of inputting the production allocation features into the reinforcement learning model, generating a backward production line interaction site with the smallest allocation consumption parameter value in the direct production allocation behavior of the plurality of metal product production lines reflecting that the target metal product production event is allocated to the plurality of metal product production lines, outputting the backward production line interaction site as a current node, configuring a plurality of reference production allocation features corresponding to the plurality of metal product production lines respectively according to the plurality of reference production line cost values and the plurality of production line entropy values, inputting the plurality of reference production allocation features into the reinforcement learning model, generating a new backward production line interaction site, sequentially cycling until the backward production line interaction site is a metal product production line, determining the backward production line interaction site as a metal product production line allocated to the target metal product production event, and outputting the target metal product production event as the allocation behavior of the target metal product production event through execution behavior from the initial current production line interaction site to the backward production line interaction site.
Re-acquiring a plurality of reference production line cost values of direct production distribution behaviors from each backward production line interaction site to the plurality of metal product production lines respectively through each backward production line interaction site, and re-acquiring a plurality of reference production distribution characteristics corresponding to the plurality of metal product production lines respectively according to the plurality of reference production line cost values and the charging production line entropy values, namely considering the influence of the production line states on production control distribution when each backward production line interaction site is distributed to one production line interaction site; the backward production line interaction site is selected according to the reconfigured multiple reference production allocation features, so that the accuracy of determining the backward production line interaction site is higher, namely the accuracy of calculating the allocation consumption parameter values from the current production line interaction site to the next production line interaction site is higher.
In an exemplary design concept, according to the current production line interaction site, the obtaining a plurality of reference production line cost values corresponding to the plurality of metal product production lines respectively may refer to the following embodiments:
acquiring a fourth target period value, wherein the fourth target period value is a remaining period parameter value for adjusting the production line cost value between two related stations in the production logic network at the next interval;
if the fourth target period value is equal to 0 and the target metal product production event is not distributed to the metal product production line, estimating a production line cost value corresponding to each of the plurality of metal product production lines under the current adjustment period node according to a production line cost estimation network, and generating the plurality of reference production line cost values;
if the fourth target period value is greater than 0 and the target metal product production event is not assigned to a metal product line, the multiple reference line cost values for the last round of adjustment are obtained.
Prior to step 106, the above embodiment further comprises the steps of:
obtaining second priori gathering data for production control distribution of a priori metal product production event, wherein the second priori gathering data comprises a production routing node of the priori metal product production event and a plurality of priori metal product production lines in the production logic network;
According to the production routing node of the prior metal product production event, a plurality of reference production line cost values corresponding to the prior metal product production lines respectively are obtained, and the reference production line cost values reflect the production distribution times of direct production distribution behaviors between a production line interaction site closest to the prior metal product production event and the metal product production lines;
obtaining a plurality of entropy values of reference production lines corresponding to the prior metal product production lines respectively;
configuring a plurality of reference production allocation characteristics corresponding to the prior metal product production lines respectively according to the cost values of the reference production lines and the entropy values of the reference production lines;
and iteratively updating the reinforcement learning model according to the production allocation characteristics carrying the backward production line interaction site with the minimum allocation consumption parameter value in the direct production allocation behavior reflecting that the prior metal product production event is allocated to the plurality of metal product production lines and the plurality of reference production allocation characteristics.
Fig. 2 illustrates a hardware architecture diagram of an optimizing system 100 of a metal product production control system for implementing the optimizing method of the metal product production control system according to an embodiment of the present application, and as shown in fig. 2, the optimizing system 100 of the metal product production control system may include a processor 110, a machine-readable storage medium 120, a bus 130, and a communication unit 140.
In an exemplary design, the optimizing system 100 of the metal product production control system may be an optimizing system of a single metal product production control system or may be an optimizing system group of metal product production control systems. The set of optimization systems for the metal product production control system may be centralized or distributed (e.g., the optimization system 100 for the metal product production control system may be a distributed system). In an exemplary design, the optimization system 100 of the metal product production control system may be local or remote. For example, the optimization system 100 of the metal article production control system may access information and/or data stored in the machine-readable storage medium 120 via a network. As another example, the optimization system 100 of the metal article production control system may be directly connected to the machine-readable storage medium 120 to access stored information and/or data. In an exemplary design concept, the optimization system 100 of the metal product production control system may be implemented on a cloud platform. For example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, or the like, or any combination thereof.
The machine-readable storage medium 120 may store data and/or instructions. In an exemplary design, machine-readable storage medium 120 may store data obtained from an external terminal. In an exemplary design, machine-readable storage medium 120 may store data and/or instructions for use by optimization system 100 of a metal article production control system to perform or use to perform the exemplary methods described herein. In an exemplary design, machine-readable storage medium 120 may include mass storage, removable storage, volatile read-write memory, read-only memory (ROM), and the like, or any combination thereof. Exemplary mass storage devices may include magnetic disks, optical disks, solid state disks, and the like. Exemplary removable memory may include flash drives, floppy disks, optical disks, memory cards, compact disks, tape, and the like. Exemplary volatile read-write memory can include Random Access Memory (RAM). Exemplary RAM may include active random access memory (DRAM), double data rate synchronous active random access memory (DDR SDRAM), passive random access memory (SRAM), thyristor random access memory (T-RAM), zero capacitance random access memory (Z-RAM), and the like. Exemplary read-only memory may include mask read-only memory (MROM), programmable read-only memory (PROM), erasable programmable read-only memory (PEROM), electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM), digital versatile disk read-only memory, and the like. In an exemplary design, machine-readable storage medium 120 may be implemented on a cloud platform. For example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc., or any combination thereof.
In a specific implementation, the one or more processors 110 execute computer executable instructions stored by the machine-readable storage medium 120, so that the processor 110 may perform an optimization method of the metal product production control system as in the above method embodiment, where the processor 110, the machine-readable storage medium 120, and the communication unit 140 are connected through the bus 130, and where the processor 110 may be used to control the transceiving actions of the communication unit 140.
The specific implementation process of the processor 110 may refer to the above-mentioned method embodiments executed by the optimizing system 100 of the metal product production control system, and the implementation principle and technical effects are similar, which are not repeated herein.
In addition, the embodiment of the application also provides a readable storage medium, wherein computer executable instructions are preset in the readable storage medium, and when a processor executes the computer executable instructions, the optimization method of the metal product production control system is realized.
Similarly, it should be noted that in order to simplify the description of the present disclosure and thereby aid in understanding one or more embodiments of the application, various features are sometimes grouped together in a single embodiment, figure, or description thereof. Similarly, it should be noted that in order to simplify the description of the present disclosure and thereby aid in understanding one or more embodiments of the application, various features are sometimes grouped together in a single embodiment, figure, or description thereof.
Claims (10)
1. A method of optimizing a metal product production control system, the method comprising:
receiving a production control allocation task of a metal article production control system, the production control allocation task configured to instruct production control allocation of the target metal article production event;
acquiring a production logic network formed by a metal product production line and a production line interaction site;
according to the production routing node of the target metal product production event, a plurality of production line cost values respectively corresponding to a plurality of metal product production lines in the production logic network are obtained, and the production line cost values reflect the production distribution times of direct production distribution behaviors between a reference production line interaction site which is most matched with the target metal product production event and the metal product production lines;
obtaining a plurality of production line entropy values corresponding to the plurality of metal product production lines respectively;
configuring a plurality of production allocation features respectively corresponding to the plurality of metal product production lines according to the production line cost values and the production line entropy values, wherein the production allocation features respectively reflect allocation consumption parameter values of the target metal product production event allocated to the plurality of metal product production lines;
And based on the production distribution characteristics, carrying out production control distribution on the target metal product production event according to a reinforcement learning model.
2. The method for optimizing a metal product production control system according to claim 1, wherein the obtaining, by the production routing node according to the target metal product production event, a plurality of production line cost values corresponding to a plurality of metal product production lines in the production logic network respectively includes:
for a first metal product line of the plurality of metal product lines, obtaining one or more production nodes via which the target metal product production event is distributed from the reference line interaction site to the first metal product line;
according to a production line cost estimation network, taking a production load level of a first production node in the one or more production nodes and a corresponding recorded production resource amount when the target metal product production event is respectively distributed to the first production node as network loading data, and generating a production line cost value corresponding to the first production node;
outputting the added value of the production line cost value corresponding to the one or more production nodes as the production line cost value corresponding to the first metal product production line;
Taking the production load level of a first production node in the one or more production nodes and the corresponding recorded production resource amount when the target metal product production event is respectively distributed to the first production node as network loading data, generating a production line cost value corresponding to the first production node, comprising:
determining a metric value A1 according to the production load level of the first production node, wherein the metric value A1 is not less than 0 and less than 1;
outputting a fusion value of the metric value A1 and a first weight coefficient of the production line cost estimation network as a metric value A2;
and fusing the added value of the metric value A2 and the metric value A1 with the production resource quantity to generate a production line cost value corresponding to the first production node.
3. The method of optimizing a metal product production control system according to claim 2, wherein the determining the metric value A1 in accordance with the production load level of the first production node comprises:
outputting a power of a target value of the production load level as the measurement value A1 if the production load level is not less than 0 and not more than 1, wherein the target value reflects a second weight coefficient of the production line cost estimation network;
And if the production load level is greater than 1 and not greater than m, outputting a power of a target value of a subtraction value of m and the production load level as the measurement value A1.
4. The method of optimizing a metal product production control system according to claim 2, wherein before the production routing node according to the target metal product production event obtains a plurality of production line cost values corresponding to a plurality of metal product production lines in the production logic network, the method further comprises:
obtaining first a priori gathered data for production control allocation of a priori metal product production event, the first a priori gathered data comprising one or more reference production nodes via which the a priori metal product production event is allocated to a metal product production line;
determining a production load level of a first reference production node of the one or more reference production nodes;
determining the corresponding recorded production resource amount when the prior metal product production events are respectively distributed to the first reference production node;
and iteratively updating the production line cost estimation network by taking the production load level of the first reference production node carrying the production line cost value and the production resource quantity which is correspondingly recorded when the prior metal product production event is respectively distributed to the first reference production node as one network learning data in a network learning data sequence.
5. The method for optimizing a metal product production control system according to claim 1, wherein the obtaining a plurality of production line cost values corresponding to a plurality of metal product production lines in the production logic network respectively comprises:
acquiring a first target period value, wherein the first target period value is a remaining period parameter value for adjusting the production line cost value between two related stations in the production logic network at the next interval;
if the first target period value is not smaller than the threshold period value, acquiring the cost values of the multiple production lines which are adjusted in the last round;
if the first target period value is smaller than the threshold period value, estimating a production line cost value corresponding to each of the plurality of metal product production lines under a current adjustment period node according to a production line cost estimation network, and generating a plurality of production line cost values;
the production line cost values are regularly adjusted according to the first period parameter value.
6. The method for optimizing a metal product production control system according to claim 1, wherein the obtaining a plurality of production line entropy values corresponding to the plurality of metal product production lines, respectively, comprises:
Taking the production characteristic vector of the target metal product production event and the production line label data of a second metal product production line in the plurality of metal product production lines as network loading data, and estimating a production line entropy value of the second metal product production line according to a production line entropy value estimation network generated according to a local random gradient descent strategy;
the production feature vector includes: production event features, production trigger features, and production progress features.
7. The method of optimizing a metal product production control system according to claim 1, wherein configuring a plurality of production allocation features respectively corresponding to the plurality of metal product production lines according to the plurality of production line cost values and the plurality of production line entropy values comprises:
acquiring the initial residual production task quantity of the target metal product production event and the station identification of the reference production line interaction station from the production control distribution task;
aiming at a third metal product production line in the metal product production lines, acquiring a second target period value and a first production task amount which are correspondingly recorded in the process that the target metal product production event is distributed to the third metal product production line according to the production line cost value corresponding to the third metal product production line;
Acquiring a third target period value and an executable production task amount which are delayed to be executed after the target metal product production event is distributed to the third metal product production line according to the production line entropy value of the third metal product production line;
and configuring production allocation characteristics corresponding to the third metal product production line according to the station identification of the reference production line interaction station, the initial residual production task quantity, the second target period value, the first production task quantity, the third target period value and the executable production task quantity.
8. The method of optimizing a metal product production control system of claim 1, wherein the assigning production control to the target metal product production event in accordance with a reinforcement learning model based on the plurality of production assignment features comprises:
loading the plurality of production allocation features into the reinforcement learning model to generate a first production allocation feature reflecting a minimum allocation consumption parameter value of a backward production line interaction site in a direct production allocation behavior of the plurality of metal product production lines to which the target metal product production event is allocated;
If the backward production line interaction site of the reference production line interaction site is a metal product production line in the plurality of metal product production lines, determining the backward production line interaction site as a metal product production line allocated for the target metal product production event, and outputting the allocation behavior of the target metal product production event as the target metal product production event from the initial current production line interaction site to the backward production line interaction site via the executed behavior;
if the backward production line interactive site is not a metal product production line of the plurality of metal product production lines, determining the backward production line interactive site as a current production line interactive site, and determining a backward production line interactive site for the current production line interactive site until the backward production line interactive site determined for the current production line interactive site is a metal product production line of the plurality of metal product production lines based on:
according to the current production line interaction site, a plurality of reference production line cost values corresponding to the plurality of metal product production lines are obtained, wherein the plurality of reference production line cost values are production line cost values of direct production distribution behaviors of the target metal product production event from the current production line interaction site to the plurality of metal product production lines respectively;
Configuring a plurality of reference production distribution characteristics respectively corresponding to the plurality of metal product production lines according to the cost values of the plurality of reference production lines and the entropy values of the plurality of production lines;
inputting the multiple reference production allocation features into a reinforcement learning model to generate a backward production line interaction site determined for the current production line interaction site;
the step of obtaining a plurality of reference production line cost values corresponding to the plurality of metal product production lines respectively according to the current production line interaction station comprises the following steps:
acquiring a fourth target period value, wherein the fourth target period value is a remaining period parameter value for adjusting the production line cost value between two related stations in the production logic network at the next interval;
if the fourth target period value is equal to 0 and the target metal product production event is not distributed to the metal product production lines, estimating a production line cost value corresponding to each of the plurality of metal product production lines under the current adjustment period node according to a production line cost estimation network, and generating the plurality of reference production line cost values;
if the fourth target period value is greater than 0 and the target metal product production event is not assigned to a metal product line, the multiple reference line cost values for the last round of adjustment are obtained.
9. The method of optimizing a metal product production control system of claim 1, wherein the method further comprises, prior to production control allocation of the target metal product production event in accordance with a reinforcement learning model based on the plurality of production allocation features:
obtaining second priori gathering data for production control distribution of a priori metal product production event, wherein the second priori gathering data comprises a production routing node of the priori metal product production event and a plurality of priori metal product production lines in the production logic network; according to the production routing node of the prior metal product production event, a plurality of reference production line cost values corresponding to the prior metal product production lines respectively are obtained, and the reference production line cost values reflect the production distribution times of direct production distribution behaviors between a production line interaction site which is most matched with the prior metal product production event and the metal product production lines;
obtaining a plurality of entropy values of reference production lines corresponding to the prior metal product production lines respectively;
configuring a plurality of reference production distribution characteristics respectively corresponding to the prior metal product production lines according to the cost values of the reference production lines and the entropy values of the reference production lines;
And iteratively updating the reinforcement learning model according to the production allocation characteristics carrying the backward production line interaction site with the minimum allocation consumption parameter value in the direct production allocation behavior reflecting that the prior metal product production event is allocated to the plurality of metal product production lines and the plurality of reference production allocation characteristics.
10. An optimization system for a metal article production control system, comprising a processor and a machine-readable storage medium having stored therein machine-executable instructions loaded and executed by the processor to implement the method of optimizing a metal article production control system of any one of claims 1-9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310606647.8A CN116679639B (en) | 2023-05-26 | 2023-05-26 | Optimization method and system of metal product production control system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310606647.8A CN116679639B (en) | 2023-05-26 | 2023-05-26 | Optimization method and system of metal product production control system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116679639A true CN116679639A (en) | 2023-09-01 |
CN116679639B CN116679639B (en) | 2024-01-05 |
Family
ID=87782919
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310606647.8A Active CN116679639B (en) | 2023-05-26 | 2023-05-26 | Optimization method and system of metal product production control system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116679639B (en) |
Citations (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102053613A (en) * | 2010-12-31 | 2011-05-11 | 北京三博中自科技有限公司 | Energy real-time monitoring system of industrial process equipment and monitoring method thereof |
JP2012074007A (en) * | 2010-08-31 | 2012-04-12 | Fuji Electric Co Ltd | Operation condition optimization system of plant, operation condition optimization method of plant and operation condition optimization program of plant |
CN103324165A (en) * | 2013-05-27 | 2013-09-25 | 西北工业大学 | Process route optimization method considering production line stability |
US20140128996A1 (en) * | 2012-11-05 | 2014-05-08 | Rockwell Automation Technologies, Inc. | Secure models for model-based control and optimization |
JP2015118526A (en) * | 2013-12-18 | 2015-06-25 | 株式会社神戸製鋼所 | Operation data retrieval support method and operation data retrieval support apparatus |
CN105630608A (en) * | 2015-12-28 | 2016-06-01 | 黑龙江科技大学 | Method for achieving multiprocessor scheduling through combined cross entropy |
CN108898285A (en) * | 2018-06-11 | 2018-11-27 | 北华大学 | A kind of beer packaging production line operational efficiency quantitative evaluating method based on comentropy |
CN109032099A (en) * | 2018-09-04 | 2018-12-18 | 山东建筑大学 | Engineering machinery assemble production line online awareness system |
CN109570353A (en) * | 2017-09-29 | 2019-04-05 | 长城汽车股份有限公司 | The molding die component and auto production line of side-wall outer-plate |
US20200104678A1 (en) * | 2018-09-27 | 2020-04-02 | Google Llc | Training optimizer neural networks |
CN111240281A (en) * | 2019-12-31 | 2020-06-05 | 清研讯科(北京)科技有限公司 | Production control method and system |
CN111797992A (en) * | 2020-05-25 | 2020-10-20 | 华为技术有限公司 | Machine learning optimization method and device |
CN112162861A (en) * | 2020-09-29 | 2021-01-01 | 广州虎牙科技有限公司 | Thread allocation method and device, computer equipment and storage medium |
CN113099729A (en) * | 2018-10-26 | 2021-07-09 | 陶氏环球技术有限责任公司 | Deep reinforcement learning for production scheduling |
CN113506082A (en) * | 2021-06-17 | 2021-10-15 | 沈阳新松虚拟现实产业技术研究院有限公司 | VR-based digital factory production line supervision method and system |
CN114218867A (en) * | 2021-12-20 | 2022-03-22 | 暨南大学 | Special equipment flow control method and system based on entropy optimization safety reinforcement learning |
CN114219159A (en) * | 2021-12-20 | 2022-03-22 | 湖南大学 | Production line scheduling method based on federal learning and attention mechanism |
US20220114301A1 (en) * | 2021-12-23 | 2022-04-14 | Intel Corporation | Methods and devices for a collaboration of automated and autonomous machines |
CN114862192A (en) * | 2022-05-07 | 2022-08-05 | 北京航空航天大学 | Low-cost production task allocation and resource scheduling method for heterogeneous industrial internet system |
CN114971258A (en) * | 2022-05-20 | 2022-08-30 | 安徽工业大学 | Industrial process optimization method and system based on coke quality prediction |
CN115239675A (en) * | 2022-07-29 | 2022-10-25 | 平安科技(深圳)有限公司 | Training method of classification model, image classification method and device, equipment and medium |
CN115237074A (en) * | 2022-07-23 | 2022-10-25 | 中南大学 | Self-adaptive dynamic scheduling control system and method for scrap steel feeding |
CN115409836A (en) * | 2022-10-31 | 2022-11-29 | 深圳仕上电子科技有限公司 | Metal workpiece detection and cleaning control method and system based on image processing |
CN115422486A (en) * | 2022-09-15 | 2022-12-02 | 亢朝侠 | Cloud service online page optimization method based on artificial intelligence and big data system |
CN115618269A (en) * | 2022-12-12 | 2023-01-17 | 江门市润宇传感器科技有限公司 | Big data analysis method and system based on industrial sensor production |
US20230076192A1 (en) * | 2020-02-07 | 2023-03-09 | Deepmind Technologies Limited | Learning machine learning incentives by gradient descent for agent cooperation in a distributed multi-agent system |
CN115857461A (en) * | 2023-03-02 | 2023-03-28 | 东莞正大康地饲料有限公司 | Piglet premixed feed production online monitoring method and system |
-
2023
- 2023-05-26 CN CN202310606647.8A patent/CN116679639B/en active Active
Patent Citations (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012074007A (en) * | 2010-08-31 | 2012-04-12 | Fuji Electric Co Ltd | Operation condition optimization system of plant, operation condition optimization method of plant and operation condition optimization program of plant |
CN102053613A (en) * | 2010-12-31 | 2011-05-11 | 北京三博中自科技有限公司 | Energy real-time monitoring system of industrial process equipment and monitoring method thereof |
US20140128996A1 (en) * | 2012-11-05 | 2014-05-08 | Rockwell Automation Technologies, Inc. | Secure models for model-based control and optimization |
CN103324165A (en) * | 2013-05-27 | 2013-09-25 | 西北工业大学 | Process route optimization method considering production line stability |
JP2015118526A (en) * | 2013-12-18 | 2015-06-25 | 株式会社神戸製鋼所 | Operation data retrieval support method and operation data retrieval support apparatus |
CN105630608A (en) * | 2015-12-28 | 2016-06-01 | 黑龙江科技大学 | Method for achieving multiprocessor scheduling through combined cross entropy |
CN109570353A (en) * | 2017-09-29 | 2019-04-05 | 长城汽车股份有限公司 | The molding die component and auto production line of side-wall outer-plate |
CN108898285A (en) * | 2018-06-11 | 2018-11-27 | 北华大学 | A kind of beer packaging production line operational efficiency quantitative evaluating method based on comentropy |
CN109032099A (en) * | 2018-09-04 | 2018-12-18 | 山东建筑大学 | Engineering machinery assemble production line online awareness system |
US20200104678A1 (en) * | 2018-09-27 | 2020-04-02 | Google Llc | Training optimizer neural networks |
CN113099729A (en) * | 2018-10-26 | 2021-07-09 | 陶氏环球技术有限责任公司 | Deep reinforcement learning for production scheduling |
CN111240281A (en) * | 2019-12-31 | 2020-06-05 | 清研讯科(北京)科技有限公司 | Production control method and system |
US20230076192A1 (en) * | 2020-02-07 | 2023-03-09 | Deepmind Technologies Limited | Learning machine learning incentives by gradient descent for agent cooperation in a distributed multi-agent system |
CN111797992A (en) * | 2020-05-25 | 2020-10-20 | 华为技术有限公司 | Machine learning optimization method and device |
CN112162861A (en) * | 2020-09-29 | 2021-01-01 | 广州虎牙科技有限公司 | Thread allocation method and device, computer equipment and storage medium |
CN113506082A (en) * | 2021-06-17 | 2021-10-15 | 沈阳新松虚拟现实产业技术研究院有限公司 | VR-based digital factory production line supervision method and system |
CN114218867A (en) * | 2021-12-20 | 2022-03-22 | 暨南大学 | Special equipment flow control method and system based on entropy optimization safety reinforcement learning |
CN114219159A (en) * | 2021-12-20 | 2022-03-22 | 湖南大学 | Production line scheduling method based on federal learning and attention mechanism |
US20220114301A1 (en) * | 2021-12-23 | 2022-04-14 | Intel Corporation | Methods and devices for a collaboration of automated and autonomous machines |
CN114862192A (en) * | 2022-05-07 | 2022-08-05 | 北京航空航天大学 | Low-cost production task allocation and resource scheduling method for heterogeneous industrial internet system |
CN114971258A (en) * | 2022-05-20 | 2022-08-30 | 安徽工业大学 | Industrial process optimization method and system based on coke quality prediction |
CN115237074A (en) * | 2022-07-23 | 2022-10-25 | 中南大学 | Self-adaptive dynamic scheduling control system and method for scrap steel feeding |
CN115239675A (en) * | 2022-07-29 | 2022-10-25 | 平安科技(深圳)有限公司 | Training method of classification model, image classification method and device, equipment and medium |
CN115422486A (en) * | 2022-09-15 | 2022-12-02 | 亢朝侠 | Cloud service online page optimization method based on artificial intelligence and big data system |
CN115409836A (en) * | 2022-10-31 | 2022-11-29 | 深圳仕上电子科技有限公司 | Metal workpiece detection and cleaning control method and system based on image processing |
CN115618269A (en) * | 2022-12-12 | 2023-01-17 | 江门市润宇传感器科技有限公司 | Big data analysis method and system based on industrial sensor production |
CN115857461A (en) * | 2023-03-02 | 2023-03-28 | 东莞正大康地饲料有限公司 | Piglet premixed feed production online monitoring method and system |
Also Published As
Publication number | Publication date |
---|---|
CN116679639B (en) | 2024-01-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2022037337A1 (en) | Distributed training method and apparatus for machine learning model, and computer device | |
CN110737664B (en) | Method and device for synchronizing block chain link points | |
CN109408590B (en) | Method, device and equipment for expanding distributed database and storage medium | |
CN103914556A (en) | Large-scale graph data processing method | |
CN115062868B (en) | Pre-polymerization type vehicle distribution path planning method and device | |
CN108241531A (en) | A kind of method and apparatus for distributing resource for virtual machine in the cluster | |
JP6059495B2 (en) | System and method for parallel edge partitioning in AND / OR graph search | |
US8443073B2 (en) | Automated performance prediction for service-oriented architectures | |
CN113608751B (en) | Operation method, device and equipment of reasoning service platform and storage medium | |
CN116679639B (en) | Optimization method and system of metal product production control system | |
CN111880926B (en) | Load balancing method and device and computer storage medium | |
CN112417748A (en) | Method, system, equipment and medium for scheduling automatic driving simulation task | |
CN116896591A (en) | Scheduling method and device for network data analysis model and computer equipment | |
CN114581220B (en) | Data processing method and device and distributed computing system | |
CN112884180A (en) | Logistics distribution point addressing method and device, electronic equipment and storage medium | |
CN116954878A (en) | Method, apparatus, device, storage medium and program product for managing container clusters | |
CN111836274B (en) | Service processing method and device | |
CN115878303A (en) | Resource scheduling method and device and electronic equipment | |
CN114581221B (en) | Distributed computing system and computer device | |
CN111898900A (en) | Space domain allocation method and device based on block chain | |
CN109842683A (en) | More copies are executed with equalization methods, server and the container cloud platform of double mode | |
CN116681973B (en) | Image processing method, device, system, equipment and computer storage medium | |
CN114095356B (en) | Method and device for configuring node task strategy in real time | |
CN116886777B (en) | Service flow distribution method and device for container arrangement platform | |
CN111343266B (en) | Route decision method and device |
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 |