CN117709684A - Production management method, system, equipment and medium based on real-time productivity estimation - Google Patents
Production management method, system, equipment and medium based on real-time productivity estimation Download PDFInfo
- Publication number
- CN117709684A CN117709684A CN202410155030.3A CN202410155030A CN117709684A CN 117709684 A CN117709684 A CN 117709684A CN 202410155030 A CN202410155030 A CN 202410155030A CN 117709684 A CN117709684 A CN 117709684A
- Authority
- CN
- China
- Prior art keywords
- productivity
- real
- time
- data
- capacity
- 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.)
- Pending
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 106
- 238000007726 management method Methods 0.000 title claims abstract description 56
- 238000013528 artificial neural network Methods 0.000 claims abstract description 56
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 44
- 238000010219 correlation analysis Methods 0.000 claims abstract description 41
- 238000004364 calculation method Methods 0.000 claims abstract description 35
- 230000002068 genetic effect Effects 0.000 claims abstract description 30
- 238000005457 optimization Methods 0.000 claims abstract description 27
- 238000012545 processing Methods 0.000 claims abstract description 24
- 238000012549 training Methods 0.000 claims abstract description 16
- 238000003860 storage Methods 0.000 claims description 8
- 238000012937 correction Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 238000012800 visualization Methods 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 55
- 238000007619 statistical method Methods 0.000 description 16
- 230000008569 process Effects 0.000 description 15
- 239000002245 particle Substances 0.000 description 14
- 230000006870 function Effects 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 7
- 238000009826 distribution Methods 0.000 description 7
- 238000012417 linear regression Methods 0.000 description 7
- 230000002159 abnormal effect Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 238000011161 development Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000002994 raw material Substances 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000012098 association analyses Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Classifications
-
- 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
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of production management, in particular to a production management method, a system, equipment and a medium based on real-time productivity estimation, wherein the method comprises the following steps: the method comprises the steps of carrying out digital processing on a plurality of data generated in a production process based on edge calculation to obtain a data set to be processed and real-time acquisition data, carrying out gray correlation analysis on the data set to be processed, taking main influence factors in first capacity influence factors as self-variable data, carrying out model optimization on a back propagation neural network according to a genetic algorithm, inputting the data set to be processed and the self-variable data into the back propagation neural network for model training to obtain a real-time capacity estimating model, and inputting data corresponding to second capacity influence factors in the real-time acquisition data into the real-time capacity estimating model for capacity estimating to obtain estimated capacity. The method can effectively improve the accuracy of productivity estimation, ensure the instantaneity of productivity estimation, facilitate management of a manager and improve production efficiency.
Description
Technical Field
The invention relates to the technical field of production management, in particular to a production management method, system, equipment and medium based on real-time productivity estimation.
Background
Along with the rapid development of manufacturing industry, the real-time productivity prediction production management system becomes particularly important for production managers, and can help enterprises to realize the fine management and optimization of the production process and improve the production efficiency.
At present, the productivity of a production line is estimated usually through a particle swarm optimization algorithm, a linear regression analysis method and a multivariate statistical analysis method, but the particle swarm optimization algorithm is mainly dependent on individual and global optimal solutions in the particle searching process, so that the global optimal solutions can not be found, the estimated result is not accurate enough, the selection of the parameters of the particle swarm optimization algorithm can achieve a satisfactory convergence effect only through repeated tests and adjustment and a large number of iterative computations, and the calculation complexity of the algorithm is high and more calculation and time are needed when large-scale data are processed due to more and complex data generated in the production process of the production line; the linear regression analysis method is generally based on the assumption of a linear relation, and in the actual production process, when productivity is low in one day due to various factors affecting productivity, fitting capacity of a regression model is limited, so that accuracy of productivity estimation is greatly affected; the multivariate statistical analysis method is sensitive to abnormal values, when abnormal values exist in influencing variables, the calculation and model establishment of the multivariate statistical analysis can be greatly influenced, the multivariate statistical analysis is easily influenced by data distribution, the assumption of the multivariate statistical analysis is normal distribution, but when actual data does not meet the assumption due to a plurality of emergency situations in the production process, the prediction result can be invalid, and the prediction is inaccurate.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a production management method, a system, equipment and a medium based on real-time productivity estimation, which can effectively improve the accuracy of productivity estimation, ensure the real-time performance of productivity estimation, facilitate management of managers and improve production efficiency.
In a first aspect, an embodiment of the present invention provides a production management method based on real-time productivity estimation, including:
acquiring a plurality of data generated in the production process in real time, and performing digital processing on the plurality of data based on edge calculation to obtain a data set to be processed and real-time acquisition data, wherein the data set to be processed is composed of a plurality of the real-time acquisition data;
gray correlation analysis is carried out on the data set to be processed, a first productivity influence factor is determined, the first productivity influence factors are ordered according to influence degrees, and main influence factors in the first productivity influence factors are used as self-variable data;
model optimization is carried out on the back propagation neural network according to a genetic algorithm, and the data set to be processed and the self-variable data are input into the back propagation neural network for model training, so that a real-time productivity estimation model is obtained;
Gray correlation analysis is carried out on the real-time collected data, and a second productivity influence factor is determined;
inputting data corresponding to the second productivity influence factor in the real-time collected data into the real-time productivity estimation model to perform productivity estimation, and obtaining estimated productivity.
According to some embodiments of the first aspect of the present invention, after inputting the data corresponding to the second capacity influencing factor in the real-time collected data into the real-time capacity estimation model to perform capacity estimation, the method includes:
acquiring a preset productivity threshold;
comparing the preset capacity threshold with the estimated capacity to obtain a capacity comparison result;
and according to the productivity comparison result, confirming whether the estimated productivity meets a preset result.
According to some embodiments of the first aspect of the invention, the method further has:
and generating a warning signal and carrying out warning processing under the condition that the difference value between the estimated productivity and the preset productivity threshold exceeds the preset difference value threshold.
According to some embodiments of the first aspect of the present invention, the capacity is used as a parent sequence of the gray correlation analysis, each influence factor affecting the capacity is used as a child sequence of the gray correlation analysis, the gray correlation analysis is performed on the data set to be processed, a first capacity influence factor is determined, the first capacity influence factors are ranked according to the influence degree, and a main influence factor in the first capacity influence factors is used as self-variable data, where the method includes:
Calculating the absolute difference value of the parent sequence interval numerical value and the child sequence interval numerical value in the data set to be processed to obtain a first absolute difference value;
determining a correlation coefficient sequence according to a preset resolution coefficient and the first absolute difference value;
determining the association degree between each subsequence and each parent sequence according to the association coefficient sequence;
and determining a first productivity influence factor according to the association degree, sorting the first productivity influence factors according to the influence degree, and taking the main influence factor in the first productivity influence factor as self-variable data.
According to some embodiments of the first aspect of the present invention, the model optimization is performed on a back propagation neural network according to a genetic algorithm, and the data to be processed and the self-variable data are input into the back propagation neural network for model training, so as to obtain a real-time productivity estimation model, where the model comprises:
obtaining output values of all layers in the back propagation neural network;
determining an error value of each node in each layer of the back propagation neural network according to the output value;
and carrying out weight correction on the error value through the genetic algorithm so as to carry out model optimization on the back propagation neural network, and obtaining a real-time productivity estimation model.
In a second aspect, an embodiment of the present invention provides a production management system based on real-time productivity estimation, including:
the system comprises an internet of things module based on edge calculation, a data processing module and a data processing module, wherein the internet of things module is used for acquiring multiple items of data generated in a production process in real time and carrying out digital processing on the multiple items of data based on the edge calculation to obtain a data set to be processed and real-time acquired data, and the data set to be processed is formed by a plurality of real-time acquired data;
the first gray correlation analysis module is used for carrying out gray correlation analysis on the data set to be processed, determining first productivity influence factors, sequencing the first productivity influence factors according to influence degrees, and taking main influence factors in the first productivity influence factors as self-variable data;
the genetic algorithm module is used for carrying out model optimization on the back propagation neural network according to a genetic algorithm;
the back propagation neural network module is used for inputting the data set to be processed and the self-variable data into the back propagation neural network for model training to obtain a real-time productivity estimation model;
the second gray correlation analysis module is used for carrying out gray correlation analysis on the real-time acquired data and determining a second productivity influence factor;
The determining module is used for inputting data corresponding to the second productivity influence factor in the real-time collected data into the real-time productivity estimation model to perform productivity estimation, and estimated productivity is obtained. According to some embodiments of the second aspect of the present invention, the production management system based on real-time productivity estimation further has:
the acquisition module is used for acquiring a preset productivity threshold;
and the threshold comparison visualization module is used for comparing the preset capacity threshold with the estimated capacity to obtain a capacity comparison result, and determining whether the estimated capacity meets the preset result according to the capacity comparison result.
According to some embodiments of the second aspect of the present invention, the production management system based on real-time productivity estimation further has:
and the warning module is used for generating a warning signal and carrying out warning processing under the condition that the difference value between the estimated capacity and the preset capacity threshold exceeds the preset difference value threshold.
In a third aspect, an embodiment of the present invention provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing when executing the computer program: the method as described in the first aspect above.
In a fourth aspect, embodiments of the present invention also provide a computer-readable storage medium storing computer-executable instructions for performing the method according to the first aspect.
The method has the beneficial effects that based on the edge computing technology, a data set to be processed and real-time acquisition data are acquired in the Internet of things, wherein the data set to be processed is a plurality of data sets generated in the production process, and the real-time acquisition data are real-time data generated in the production process; carrying out grey correlation analysis on the data set to be processed, and determining that the main influencing factor of productivity is self-variable data; under the condition that the data to be processed is a historical training sample, model optimization is carried out on the back propagation neural network according to a genetic algorithm, and the data to be processed and the self-variable data are input into the back propagation neural network for model training, so that a real-time productivity estimation model is obtained; and inputting the real-time acquired data into a real-time productivity estimation model to perform productivity estimation, so as to obtain estimated productivity. The production management method based on the real-time productivity estimation can effectively improve the accuracy of productivity estimation, ensure the real-time performance of productivity estimation, facilitate management of managers and improve production efficiency.
Drawings
FIG. 1 is a flow chart of a first method for production management based on real-time capacity estimation according to an embodiment of the present invention;
FIG. 2 is a flow chart of a second method for production management based on real-time capacity estimation according to an embodiment of the first aspect of the present invention;
FIG. 3 is a flow chart of a third method for production management based on real-time capacity estimation according to an embodiment of the first aspect of the present invention;
FIG. 4 is a flow chart of a schematic structure of a production management system based on real-time productivity estimation according to a second embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the third aspect of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The production management method, system, equipment and medium based on real-time productivity estimation provided by the embodiment of the invention are described in detail below by means of specific embodiments and application scenes thereof with reference to the accompanying drawings.
Example 1:
referring to fig. 1, fig. 1 shows a production management method based on real-time productivity estimation according to an embodiment of a first aspect of the present invention, where the method is applied to an electronic device and executed by the electronic device. In other words, the method may be performed by software or hardware installed in an electronic device, the method comprising the steps of:
step S110, collecting multiple data generated in the production process in real time, and carrying out digital processing on the multiple data based on edge calculation to obtain a data set to be processed and real-time collected data.
The data set to be processed is composed of a plurality of real-time acquisition data, namely the data set to be processed is a historical training sample data set. In the step, the edge computing technology is integrated into the gateway of the Internet of things to realize the networking of machine data, and real-time digital acquisition can be carried out on the actual capacity data of the equipment while the operation and storage pressure of the server are lightened.
Specifically, the data acquisition mode of the internet of things based on the edge computing technology is responsible for reducing the operation and storage pressure of a server, and simultaneously carries out real-time digital acquisition on data in the production process, wherein the acquired data comprises at least one of the number of production equipment, the number of workers, the speed of the workers, the raw material consumption, the number of orders, the total time and the number of production lines on a production line as factors affecting the productivity.
Step S120, gray correlation analysis is carried out on the data set to be processed, a first productivity influence factor is determined, the first productivity influence factors are ordered according to influence degrees, and main influence factors in the first productivity influence factors are used as self-variable data.
In the step, the gray correlation analysis is used for determining the correlation degree of productivity and each factor influencing the productivity, sequencing the influence degree of each factor influencing the productivity and determining the main influence factor as an independent variable.
And step S130, performing model optimization on the back propagation neural network according to a genetic algorithm, and inputting the data set to be processed and the self-variable data into the back propagation neural network for model training to obtain a real-time productivity estimation model.
In the step, a genetic algorithm is utilized to optimize the back propagation neural network so as to reduce the search range, optimize parameters such as network weight and the like, and determine the optimal parameters of the network and the parameters of the genetic algorithm; and training the back propagation neural network model by utilizing the determined parameters of the back propagation neural network and the genetic algorithm continuously by utilizing the historical data so as to generate a real-time productivity estimation model, and carrying out real-time productivity estimation according to factors affecting productivity of the real-time productivity estimation model as input.
It should be noted that, the Back Propagation neural network is also called a BP neural network, BP is an abbreviation of Back Propagation, and is translated into Back Propagation, and the Back Propagation neural network is a multi-layer feedforward neural network trained according to an error Back Propagation algorithm.
The model based on grey correlation analysis and genetic algorithm for optimizing the back propagation neural network is used, so that the productivity estimation during verification can be ensured, and the accuracy of the productivity estimation can be ensured; in addition, the genetic algorithm can adjust the input of the back propagation neural network and the weight of the parameters according to the factors influencing the productivity and the influence degree thereof, and has certain expansibility.
Specifically, the back propagation neural network performs learning of a single individual, and is basically a local search method from the search method, and searches for the next solution from the neighbors of the solution space each time, so that the back propagation neural network can be trapped into a local optimal solution in the process of searching the solution; the genetic algorithm is used for learning and adapting individual population, the overall is a global searching method, a group of solutions are set in a solution space each time, the next searching points are determined by selecting, crossing and mutating methods, the defects of the neural network can be overcome, and the genetic neural network with better solving effect is obtained by combining the two methods; the genetic algorithm will optimize the BP neural network and correct the weight.
And step S140, gray correlation analysis is carried out on the real-time acquired data, and a second productivity influence factor is determined.
In the step, the association degree of each factor of productivity and influence on productivity is determined through gray association analysis, and a second productivity influence factor is obtained.
And step S150, inputting data corresponding to the second productivity influence factor in the real-time collected data into a real-time productivity estimation model for productivity estimation, and obtaining estimated productivity.
In the step, according to the real-time productivity estimation model, the productivity estimation is carried out on the real-time acquired data input in real time, and the estimated productivity is obtained.
In the method, the gray correlation algorithm is adopted to analyze the data set to be processed in the production process, the importance of each production factor acting in the productivity estimation process is determined through the correlation degree, and the genetic algorithm is used to optimize the back propagation neural network, so that the defects of slow training convergence and easiness in sinking into local minima can be overcome, and the productivity estimation can be rapidly and accurately carried out. In addition, the genetic algorithm may generate different back propagation neural network inputs and weights to obtain more accurate predicted values based on factors affecting productivity and the extent of their impact.
Along with the rapid development of manufacturing industry, the real-time productivity prediction production management system becomes particularly important for production managers, and can help enterprises to realize the fine management and optimization of the production process and improve the production efficiency.
At present, the productivity of a production line is estimated usually through a particle swarm optimization algorithm, a linear regression analysis method and a multivariate statistical analysis method, in the particle swarm optimization algorithm, each particle represents a solution, and the position and the speed are updated according to the guidance of an individual optimal solution and a global optimal solution until the optimal solution is found, so that the productivity estimation is realized; in the linear regression analysis, a regression model is built based on historical data, capacity data and independent variables possibly influencing capacity (such as production equipment, labor efficiency, raw material consumption, order quantity and the like), and a new independent variable value is given to be substituted into the regression model, so that corresponding capacity estimation can be performed; in multivariate statistical analysis, linear relation between productivity and independent variables is analyzed through dimension reduction and extraction, covariance matrix is calculated, characteristic values and characteristic vectors are obtained, main components are selected, a model is built, and productivity estimation is carried out.
However, the particle swarm optimization algorithm mainly depends on individual and global optimal solutions in the particle searching process, so that the global optimal solution may not be found, the estimated result is not accurate enough, the selection of the particle swarm optimization algorithm parameters needs to be subjected to repeated tests and adjustment and a large amount of iterative computation to achieve a satisfactory convergence effect, and the calculation complexity of the algorithm is high and more calculation and time are needed when large-scale data are processed due to more and complex data generated in the production process of the production line; the linear regression analysis method is generally based on the assumption of a linear relation, and in the actual production process, when productivity is low in one day due to various factors affecting productivity, fitting capacity of a regression model is limited, so that accuracy of productivity estimation is greatly affected; the multivariate statistical analysis method is sensitive to abnormal values, when abnormal values exist in influencing variables, the calculation and model establishment of the multivariate statistical analysis can be greatly influenced, the multivariate statistical analysis is easily influenced by data distribution, the assumption of the multivariate statistical analysis is normal distribution, but when actual data does not meet the assumption due to a plurality of emergency situations in the production process, the prediction result can be invalid, and the prediction is inaccurate.
Therefore, the production management method based on real-time productivity estimation provided by the embodiment of the invention acquires multiple data generated in the production process in real time, performs digital processing on the multiple data based on edge calculation to obtain a data set to be processed and real-time acquired data, performs gray correlation analysis on the data set to be processed, determines first productivity influence factors, ranks the first productivity influence factors according to influence degrees, takes main influence factors in the first productivity influence factors as self-variable data, performs model optimization on a back propagation neural network according to a genetic algorithm, inputs the data set to be processed and the self-variable data into the back propagation neural network for model training to obtain a real-time productivity estimation model, performs gray correlation analysis on the real-time acquired data, determines second productivity influence factors, and inputs data corresponding to the second productivity influence factors in the real-time acquired data into the real-time productivity estimation model for productivity estimation to obtain estimated productivity. The production management method based on the real-time productivity estimation can effectively improve the accuracy of productivity estimation, ensure the real-time performance of productivity estimation, facilitate management of managers and improve production efficiency.
Example 2:
referring to fig. 2, fig. 2 illustrates another production management method based on real-time productivity estimation according to the first embodiment of the present invention, which is executed by an electronic device. In other words, the method may be performed by software or hardware installed in an electronic device, the method comprising the steps of:
step S210, collecting multiple data generated in the production process in real time, and carrying out digital processing on the multiple data based on edge calculation to obtain a data set to be processed and real-time collected data.
This step may be described in step S110 in the embodiment of fig. 1, and will not be described herein.
Step S220, calculating the absolute difference value of the parent sequence interval value and the child sequence interval value in the data set to be processed to obtain a first absolute difference value.
In this step, the capacity is used as a parent sequence for gray correlation analysis, and each influencing factor affecting the capacity is used as a child sequence for gray correlation analysis. Examples of parent and child sequences are: the parent sequence: capacity, subsequence: the number of workers on the production line.
In the step, in gray correlation analysis, the factor influencing productivity is essentially a non-time sequence, so that the adoption of the original data interval is reasonable; and carrying out data normalization processing by taking the productivity as a parent sequence of gray correlation analysis and each influencing factor as a child sequence of gray correlation analysis. Calculating a first absolute difference value of the numerical value after the parent sequence interval and the numerical value after the child sequence interval in the same sample according to the following calculation formula :
,
Wherein,representing the parent sequence->Representing the subsequence->=1,2……n。
It should be noted that, the interval processing is to convert the data interval distribution into 0 to 1, so as to realize normalization of the data; in the present application, the influencing factors belong to a non-time series, and are essentially index series, so that the compartmentalization process is selected.
Exemplary, absolute difference calculation: the sequence A after interval is: {0.0,0.25,0.5,0.75,1.0}, the sequence B after interval is: {0.0,0.2,0.6,0.8,1.0}, then, calculating absolute difference values of numerical values at corresponding positions in the sequences A and B, and taking an average value as a correlation degree. The specific calculation steps are as follows: 0.0-0.0=0.0, |0.25-0.2|=0.05, |0.5-0.6|=0.1, |0.75-0.8|=0.05, |1.0-1.0|=0.0.
Step S230, determining a correlation coefficient sequence according to the preset resolution coefficient and the first absolute difference value.
Specifically, the correlation coefficient sequence is calculated using the following calculation formula:
wherein,representing a sequence of association coefficients,/-, for>Representing the minimum difference of the two poles, i.e. +.on the curve>Points and->Is at +.>On the curve, each corresponding point is associated with +.>A minimum value of the distances of the respective points; />Representing a first-stage minimum difference; />Representing the maximum difference of the two poles, and the meaning of the maximum difference is similar to that of the minimum difference; / >Representing a preset resolution factor, < >>The smaller the resolution, the greater the resolution, for weakening +.>The influence of distortion due to excessive numerical value is used for improving the significance of the difference between the correlation coefficients, and the application adopts +.>. On the whole, calculateCorresponds to->Is a correlation coefficient of: />,/>… n, n represents a positive integer.
Step S240, determining the association degree between each sub-sequence and each parent sequence according to the association coefficient sequence.
In this step, the association degree calculation formula is as follows:wherein->Representing the degree of association between each subsequence and parent sequence,/->Representing a sequence of correlation coefficients. And determining each factor influencing productivity and the influence degree order thereof according to the association degree calculation formula, and giving different weights according to the influence degree of the association degree when the subsequent back propagation neural network is input, wherein the weights given by the larger association degree are larger.
Step S250, determining first capacity influencing factors according to the association degree, sorting the first capacity influencing factors according to the influence degree, and taking main influencing factors in the first capacity influencing factors as self-variable data.
In the step, the collected data are subjected to gray correlation analysis to determine factors influencing productivity, the influence degree of the factors is ordered, and main influence factors are determined to serve as independent variables.
Step S260, obtaining output values of each layer in the back propagation neural network.
In this step, the network is provided with m layers, i.e. 1, 2. Order theRepresents the output of the j-th node in the m-th layer, and +.>Is equal to x j I.e. the j-th input. Let->Representing from->To->The realization steps of the connection weight and the BP neural network are as follows: randomly giving each weight->And threshold->Giving an initial value, taking a data pair (x k ，T k ) The input vector is added to the input layer (m=0) so that there is +.>The calculation formula of the output value of each layer of nodes is as follows: />The method comprises the steps of carrying out a first treatment on the surface of the Wherein, (x) k ，T k ) The data pair is the data collected by the front end, and factors and productivity affecting the productivity in the existing data set are used for training the BP neural network>Representing the output of the j-th node in the m-1 layer.
Step S270, determining the error value of each node in each layer of the back propagation neural network according to the output value.
In this step, the calculation formula of the error value of each node of the output layer is as follows:the calculation formula of the error value of each node of the previous layers is as follows: />。
And step S280, carrying out weight correction on the error value through a genetic algorithm to carry out model optimization on the back propagation neural network, and obtaining a real-time productivity estimation model.
In this step, the calculation formula of the correction weight using the genetic algorithm is as follows:=/>，+/>。
step S291, gray correlation analysis is performed on the real-time collected data to determine a second capacity influencing factor.
This step may be described in step S140 of the embodiment of fig. 1, and will not be described herein.
Step S292, inputting the data corresponding to the second capacity influencing factor in the real-time collected data into the real-time capacity estimating model for capacity estimation, and obtaining the estimated capacity.
This step may be described in step S150 in the embodiment of fig. 1, and will not be described herein.
At present, along with the rapid development of manufacturing industry, a real-time productivity estimation production management system is particularly important for production managers, and can help enterprises to realize fine management and optimization of production processes and improve production efficiency. However, the productivity is affected by a plurality of factors, and a common production management system often depends on manual statistics of various factors and inputs the factors into the system, so that productivity estimation cannot be accurately performed, and a large error exists; moreover, such production systems cannot be monitored and estimated in real time, and have a large delay. Meanwhile, the traditional productivity estimation generally depends on manual experience, and large errors are easy to generate, so that the resource allocation is unbalanced.
According to the production management method based on real-time productivity estimation, a plurality of data generated in a production process are collected in real time, the plurality of data are subjected to digital processing based on edge calculation, a data set to be processed and real-time collected data are obtained, absolute differences of parent sequence interval numerical values and child sequence interval numerical values in the data set to be processed are calculated, a first absolute difference value is obtained, a correlation coefficient sequence is determined according to a preset resolution coefficient and the first absolute difference value, the degree of correlation between each child sequence and each parent sequence is determined according to the correlation coefficient sequence, a first productivity influence factor is determined according to the correlation coefficient sequence, the first productivity influence factor is ranked according to the influence degree, main influence factors in the first productivity influence factor are used as self-variable data, output values of all layers in a back propagation neural network are obtained, the error value of each node in all layers in the back propagation neural network is determined, weight correction is carried out on the error value through a genetic algorithm, model optimization is carried out on the back propagation neural network, a real-time estimated productivity model is obtained, the real-time collected data are input into the real-time estimated productivity model, and the estimated productivity is estimated. The production management method based on the real-time productivity estimation can improve the accuracy of the productivity estimation, further effectively improve the efficiency of a production line and labor, improve the real-time performance of the productivity estimation, facilitate a manager to quickly adjust against emergency conditions, reduce loss and save time and cost.
Example 3:
referring to fig. 3, fig. 3 illustrates another production management method based on real-time productivity estimation according to the embodiment of the first aspect of the present invention, which is executed by an electronic device. In other words, the method may be performed by software or hardware installed in an electronic device, the method comprising the steps of:
step S310, collecting multiple data generated in the production process in real time, and digitizing the multiple data based on edge calculation to obtain real-time collected data.
This step may be described in step S110 in the embodiment of fig. 1, and will not be described herein.
Step S320, gray correlation analysis is performed on the real-time collected data to determine a second capacity influencing factor.
This step may be described in step S120 in the embodiment of fig. 1, and will not be described herein.
Step S330, inputting the data corresponding to the second productivity influence factor in the real-time collected data into the real-time productivity estimation model for productivity estimation, and obtaining the estimated productivity.
Step S340, obtaining a preset capacity threshold.
In this step, the preset capacity threshold is a capacity threshold set by the production line before the production process is performed.
Step S350, comparing the preset capacity threshold with the estimated capacity to obtain a capacity comparison result.
Step S360, according to the productivity comparison result, whether the estimated productivity meets the preset result is confirmed.
In this step, the preset capacity threshold is compared with the preset capacity threshold, and the obtained capacity comparison result is fed back to the administrator.
In step S370, when the difference between the estimated capacity and the preset capacity threshold exceeds the preset difference threshold, a warning signal is generated and a warning process is performed.
It should be noted that the design of threshold comparison is responsible for comparing the capacity with a set threshold, so as to feed back the capacity to an administrator, and if the difference between the estimated capacity and the preset capacity threshold is greater than or equal to the preset difference threshold, the administrator is fed back with alarm information.
At present, the productivity of a production line is estimated usually through a particle swarm optimization algorithm, a linear regression analysis method and a multivariate statistical analysis method, but the particle swarm optimization algorithm is mainly dependent on individual and global optimal solutions in the particle searching process, so that the global optimal solutions can not be found, the estimated result is not accurate enough, the selection of the parameters of the particle swarm optimization algorithm can achieve a satisfactory convergence effect only through repeated tests and adjustment and a large number of iterative computations, and the calculation complexity of the algorithm is high and more calculation and time are needed when large-scale data are processed due to more and complex data generated in the production process of the production line; the linear regression analysis method is generally based on the assumption of a linear relation, and in the actual production process, when productivity is low in one day due to various factors affecting productivity, fitting capacity of a regression model is limited, so that accuracy of productivity estimation is greatly affected; the multivariate statistical analysis method is sensitive to abnormal values, when abnormal values exist in influencing variables, the calculation and model establishment of the multivariate statistical analysis can be greatly influenced, the multivariate statistical analysis is easily influenced by data distribution, the assumption of the multivariate statistical analysis is normal distribution, but when actual data does not meet the assumption due to a plurality of emergency situations in the production process, the prediction result can be invalid, and the prediction is inaccurate.
According to the production management method based on real-time productivity estimation, a plurality of items of data generated in a production process are collected in real time, the plurality of items of data are subjected to digital processing based on edge calculation, real-time collected data are obtained, gray correlation analysis is conducted on the real-time collected data, second productivity influence factors are determined, data corresponding to the second productivity influence factors in the real-time collected data are input into a real-time productivity estimation model to be subjected to productivity estimation, estimated productivity is obtained, a preset productivity threshold is obtained, the preset productivity threshold is compared with the estimated productivity to obtain a productivity comparison result, whether the estimated productivity meets the preset result is confirmed according to the productivity comparison result, and warning signals are generated and warning processing is conducted under the condition that the difference between the estimated productivity and the preset productivity threshold exceeds the preset difference threshold. The production management method based on the real-time productivity estimation can effectively improve the accuracy of productivity estimation, ensure the real-time performance of productivity estimation, facilitate management of managers and improve production efficiency.
In a second aspect, fig. 4 is a schematic structural diagram of a production management system based on real-time productivity estimation according to an embodiment of the present application, where the production management system based on real-time productivity estimation includes: the system comprises an internet of things module 410 based on edge calculation, a first gray correlation analysis module 420, a genetic algorithm module 430, a back propagation neural network module 440, a second gray correlation analysis module 450 and a determination module 460.
The internet of things module 410 based on edge calculation is configured to collect multiple items of data generated in a production process in real time and digitally process the multiple items of data based on edge calculation to obtain a data set to be processed and real-time collected data, where the data set to be processed is formed by multiple real-time collected data;
the first gray correlation analysis module 420 is configured to perform gray correlation analysis on the data set to be processed, determine first capacity influence factors, rank the first capacity influence factors according to the influence degree, and take main influence factors in the first capacity influence factors as self-variable data;
a genetic algorithm module 430 for model optimization of the back propagation neural network according to a genetic algorithm;
the back propagation neural network module 440 inputs the data set to be processed and the self-variable data into the back propagation neural network for model training to obtain a real-time productivity estimation model;
the second gray correlation analysis module 450 is configured to perform gray correlation analysis on the real-time collected data, and determine a second capacity influencing factor;
the determining module 460 is configured to input data corresponding to the second capacity influencing factor in the real-time collected data into the real-time capacity estimating model for capacity estimation, so as to obtain estimated capacity.
In one possible implementation, referring to fig. 4, the production management system based on real-time productivity estimation further includes: the acquisition module 470 and the threshold contrast visualization module 480.
An obtaining module 470, configured to obtain a preset capacity threshold;
the threshold comparison visualization module 480 is configured to compare a preset capacity threshold with the estimated capacity to obtain a capacity comparison result, and determine whether the estimated capacity meets the preset result according to the capacity comparison result.
In one possible implementation, referring to fig. 4, the production management system based on real-time productivity estimation further includes: a warning module 490.
The warning module 490 is configured to generate a warning signal and perform warning processing when a difference between the estimated capacity and a preset capacity threshold exceeds a preset difference threshold.
The production management system based on real-time productivity estimation provided in the embodiment of the present application may execute each method in the foregoing method embodiment, and implement the functions and beneficial effects of each method in the foregoing method embodiment, which are not described herein again.
Optionally, as shown in fig. 5, an electronic device 700 according to the third embodiment of the present invention further includes a processor 710 and a memory 720, where the memory 720 stores a program or an instruction that can be executed on the processor 710, and the program or the instruction implements each process of the embodiment of the production management method according to the first embodiment based on real-time productivity estimation when executed by the processor 710, and can achieve the same technical effects, so that repetition is avoided and redundant description is omitted here.
It should be noted that, the electronic device in the embodiment of the present invention includes: a server, a terminal, or other devices besides a terminal.
The above electronic device structure does not constitute a limitation of the electronic device, and the electronic device may include more or less components than illustrated, or may combine some components, or may be different in arrangement of components, for example, an input unit, may include a graphics processor (Graphics Processing Unit, GPU) and a microphone, and a display unit may configure a display panel in the form of a liquid crystal display, an organic light emitting diode, or the like. The user input unit includes at least one of a touch panel and other input devices. Touch panels are also known as touch screens. Other input devices may include, but are not limited to, a physical keyboard, function keys (e.g., volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, and so forth, which are not described in detail herein.
The memory may be used to store software programs as well as various data. The memory may mainly include a first memory area storing programs or instructions and a second memory area storing data, wherein the first memory area may store an operating system, application programs or instructions (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like. Further, the memory may include volatile memory or nonvolatile memory, or the memory may include both volatile and nonvolatile memory. The nonvolatile Memory may be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable EPROM (EEPROM), or a flash Memory. The volatile memory may be random access memory (Random Access Memory, RAM), static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (ddr SDRAM), enhanced SDRAM (Enhanced SDRAM), synchronous DRAM (SLDRAM), and Direct RAM (DRRAM).
The processor may include one or more processing units; optionally, the processor integrates an application processor that primarily processes operations involving an operating system, user interface, application programs, and the like, and a modem processor that primarily processes wireless communication signals, such as a baseband processor. It will be appreciated that the modem processor described above may not be integrated into the processor.
The embodiment of the invention also provides a readable storage medium, on which a program or an instruction is stored, which when executed by a processor, implements each process of the production management method embodiment based on real-time productivity estimation in the first aspect, and can achieve the same technical effect, so that repetition is avoided, and no further description is given here.
The processor is a processor in the electronic device in the above embodiment. A readable storage medium includes a computer readable storage medium such as ROM, RAM, magnetic or optical disk, etc.
It should be noted that, in this document, 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 one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element. Furthermore, it should be noted that the scope of the methods and apparatus in the embodiments of the present invention is not limited to performing the functions in the order shown or discussed, but may also include performing the functions in a substantially simultaneous manner or in an opposite order depending on the functions involved, e.g., the described methods may be performed in an order different from that described, and various steps may also be combined. Additionally, features described with reference to certain examples may be combined in other examples.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part in the form of a computer software product stored on a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) including instructions for causing a terminal (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method of the embodiments of the present invention.
In the description of embodiments of the present invention, the terms "first," "second," "third," "fourth" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first", "a second", "a third" and a fourth "may explicitly or implicitly include one or more such feature. In the description of the present invention, unless otherwise indicated, the meaning of "a plurality" is two or more.
In the description of embodiments of the invention, a particular feature, structure, material, or characteristic may be combined in any suitable manner in one or more embodiments or examples.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (10)
1. A production management method based on real-time productivity estimation is characterized by comprising the following steps:
acquiring a plurality of data generated in the production process in real time, and performing digital processing on the plurality of data based on edge calculation to obtain a data set to be processed and real-time acquisition data, wherein the data set to be processed is composed of a plurality of the real-time acquisition data;
gray correlation analysis is carried out on the data set to be processed, a first productivity influence factor is determined, the first productivity influence factors are ordered according to influence degrees, and main influence factors in the first productivity influence factors are used as self-variable data;
model optimization is carried out on the back propagation neural network according to a genetic algorithm, and the data set to be processed and the self-variable data are input into the back propagation neural network for model training, so that a real-time productivity estimation model is obtained;
Gray correlation analysis is carried out on the real-time collected data, and a second productivity influence factor is determined;
inputting data corresponding to the second productivity influence factor in the real-time collected data into the real-time productivity estimation model to perform productivity estimation, and obtaining estimated productivity.
2. The production management method based on real-time productivity estimation according to claim 1, wherein the inputting the data corresponding to the second productivity influencing factor in the real-time collected data into the real-time productivity estimation model for productivity estimation, after obtaining the estimated productivity, comprises:
acquiring a preset productivity threshold;
comparing the preset capacity threshold with the estimated capacity to obtain a capacity comparison result;
and according to the productivity comparison result, confirming whether the estimated productivity meets a preset result.
3. The production management method based on real-time capacity estimation according to claim 2, further comprising:
and generating a warning signal and carrying out warning processing under the condition that the difference value between the estimated productivity and the preset productivity threshold exceeds the preset difference value threshold.
4. The production management method based on real-time productivity estimation according to claim 1, wherein the productivity is used as a parent sequence of the gray correlation analysis, each influencing factor influencing the productivity is used as a child sequence of the gray correlation analysis, the gray correlation analysis is performed on the data set to be processed to determine a first productivity influencing factor, the first productivity influencing factors are ranked according to the influence degree, and a main influencing factor of the first productivity influencing factors is used as self-variable data, and the production management method comprises the steps of:
Calculating the absolute difference value of the parent sequence interval numerical value and the child sequence interval numerical value in the data set to be processed to obtain a first absolute difference value;
determining a correlation coefficient sequence according to a preset resolution coefficient and the first absolute difference value;
determining the association degree between each subsequence and each parent sequence according to the association coefficient sequence;
and determining a first productivity influence factor according to the association degree, sorting the first productivity influence factors according to the influence degree, and taking the main influence factor in the first productivity influence factor as self-variable data.
5. The production management method based on real-time productivity estimation according to claim 1, wherein the model optimizing the back propagation neural network according to the genetic algorithm, and inputting the data to be processed and the self-variable data to the back propagation neural network for model training, to obtain a real-time productivity estimation model, comprises:
obtaining output values of all layers in the back propagation neural network;
determining an error value of each node in each layer of the back propagation neural network according to the output value;
and carrying out weight correction on the error value through the genetic algorithm so as to carry out model optimization on the back propagation neural network, and obtaining a real-time productivity estimation model.
6. A production management system based on real-time capacity estimation, comprising:
the system comprises an internet of things module based on edge calculation, a data processing module and a data processing module, wherein the internet of things module is used for acquiring multiple items of data generated in a production process in real time and carrying out digital processing on the multiple items of data based on the edge calculation to obtain a data set to be processed and real-time acquired data, and the data set to be processed is formed by a plurality of real-time acquired data;
the first gray correlation analysis module is used for carrying out gray correlation analysis on the data set to be processed, determining first productivity influence factors, sequencing the first productivity influence factors according to influence degrees, and taking main influence factors in the first productivity influence factors as self-variable data;
the genetic algorithm module is used for carrying out model optimization on the back propagation neural network according to a genetic algorithm;
the back propagation neural network module is used for inputting the data set to be processed and the self-variable data into the back propagation neural network for model training to obtain a real-time productivity estimation model;
the second gray correlation analysis module is used for carrying out gray correlation analysis on the real-time acquired data and determining a second productivity influence factor;
And the determining module is used for inputting data corresponding to the second productivity influence factor in the real-time acquired data into the real-time productivity estimation model to perform productivity estimation, so as to obtain estimated productivity.
7. The production management system based on real-time capacity estimation according to claim 6, further comprising:
the acquisition module is used for acquiring a preset productivity threshold;
and the threshold comparison visualization module is used for comparing the preset capacity threshold with the estimated capacity to obtain a capacity comparison result, and determining whether the estimated capacity meets the preset result according to the capacity comparison result.
8. The production management system based on real-time capacity estimation according to claim 7, further comprising:
and the warning module is used for generating a warning signal and carrying out warning processing under the condition that the difference value between the estimated capacity and the preset capacity threshold exceeds the preset difference value threshold.
9. An electronic device, comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing when executing the computer program: the production management method based on real-time capacity estimation according to any one of claims 1 to 5.
10. A computer-readable storage medium storing computer-executable instructions for performing the production management method based on real-time capacity estimation according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410155030.3A CN117709684A (en) | 2024-02-04 | 2024-02-04 | Production management method, system, equipment and medium based on real-time productivity estimation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410155030.3A CN117709684A (en) | 2024-02-04 | 2024-02-04 | Production management method, system, equipment and medium based on real-time productivity estimation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117709684A true CN117709684A (en) | 2024-03-15 |
Family
ID=90151854
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410155030.3A Pending CN117709684A (en) | 2024-02-04 | 2024-02-04 | Production management method, system, equipment and medium based on real-time productivity estimation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117709684A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113779881A (en) * | 2021-09-10 | 2021-12-10 | 中国石油大学(北京) | Method, device and equipment for predicting capacity of dense water-containing gas reservoir |
| CN116956049A (en) * | 2023-09-19 | 2023-10-27 | 中国联合网络通信集团有限公司 | Training method, device, equipment and storage medium of industrial productivity prediction model |
-
2024
- 2024-02-04 CN CN202410155030.3A patent/CN117709684A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113779881A (en) * | 2021-09-10 | 2021-12-10 | 中国石油大学(北京) | Method, device and equipment for predicting capacity of dense water-containing gas reservoir |
| CN116956049A (en) * | 2023-09-19 | 2023-10-27 | 中国联合网络通信集团有限公司 | Training method, device, equipment and storage medium of industrial productivity prediction model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108959728B (en) | Radio frequency device parameter optimization method based on deep learning | |
| WO2020164740A1 (en) | Methods and systems for automatically selecting a model for time series prediction of a data stream | |
| JP7293504B2 (en) | Data evaluation using reinforcement learning | |
| CN113783717A (en) | Smart city network flow prediction method and system | |
| CN116307215A (en) | Load prediction method, device, equipment and storage medium of power system | |
| CN116187835A (en) | Data-driven-based method and system for estimating theoretical line loss interval of transformer area | |
| CN115051929A (en) | Network fault prediction method and device based on self-supervision target perception neural network | |
| CN117709684A (en) | Production management method, system, equipment and medium based on real-time productivity estimation | |
| CN113626177A (en) | CPU resource load condition prediction method based on LSTM | |
| CN115392493A (en) | Distributed prediction method, system, server and storage medium | |
| CN112579580B (en) | Pre-alarm method based on industrial big data prediction | |
| CN115392441A (en) | Method, apparatus, device and medium for on-chip adaptation of quantized neural network model | |
| CN114943328A (en) | SARIMA-GRU time sequence prediction model based on BP neural network nonlinear combination | |
| CN114970711A (en) | Construction method and application of power load prediction model based on Seq2Seq | |
| CN114186646A (en) | Block chain abnormal transaction identification method and device, storage medium and electronic equipment | |
| JP2022088341A (en) | Apparatus learning device and method | |
| Amrane et al. | On the use of ensembles of metamodels for estimation of the failure probability | |
| Li et al. | Prediction of financial economic growth trend based on PVAR model | |
| US20230106295A1 (en) | System and method for deriving a performance metric of an artificial intelligence (ai) model | |
| CN115878783B (en) | Text processing method, deep learning model training method and sample generation method | |
| CN113591781B (en) | Image processing method and system based on service robot cloud platform | |
| JP7154468B2 (en) | Information processing device, information processing method and information processing program | |
| CN115391160B (en) | Abnormal change detection method, device, equipment and storage medium | |
| US11783194B1 (en) | Evolutionary deep learning with extended Kalman filter for modeling and data assimilation | |
| Ortiz-Bejar et al. | Forecasting without context problem |
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 |