CN116629454B - Method and system for predicting production efficiency of servo screw press based on neural network - Google Patents
Method and system for predicting production efficiency of servo screw press based on neural network Download PDFInfo
- Publication number
- CN116629454B CN116629454B CN202310882619.9A CN202310882619A CN116629454B CN 116629454 B CN116629454 B CN 116629454B CN 202310882619 A CN202310882619 A CN 202310882619A CN 116629454 B CN116629454 B CN 116629454B
- Authority
- CN
- China
- Prior art keywords
- production efficiency
- historical
- data
- function
- influence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 353
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 75
- 238000000034 method Methods 0.000 title claims abstract description 49
- 230000006870 function Effects 0.000 claims abstract description 228
- 238000012545 processing Methods 0.000 claims abstract description 180
- 230000004913 activation Effects 0.000 claims abstract description 46
- 238000007781 pre-processing Methods 0.000 claims description 24
- 238000012795 verification Methods 0.000 claims description 23
- 210000002569 neuron Anatomy 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 14
- 238000003754 machining Methods 0.000 claims description 13
- 238000010606 normalization Methods 0.000 claims description 13
- 238000004140 cleaning Methods 0.000 claims description 8
- 230000005484 gravity Effects 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 8
- 230000007812 deficiency Effects 0.000 claims 1
- 238000010200 validation analysis Methods 0.000 claims 1
- 238000003860 storage Methods 0.000 description 26
- 230000008901 benefit Effects 0.000 description 6
- 238000007405 data analysis Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 4
- 230000003213 activating effect Effects 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 4
- 238000005457 optimization Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B30—PRESSES
- B30B—PRESSES IN GENERAL
- B30B15/00—Details of, or accessories for, presses; Auxiliary measures in connection with pressing
- B30B15/26—Programme control arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/0985—Hyperparameter optimisation; Meta-learning; Learning-to-learn
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- 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
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Manufacturing & Machinery (AREA)
- Computational Linguistics (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Primary Health Care (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Quality & Reliability (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Mechanical Engineering (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a method for predicting the production efficiency of a servo screw press based on a neural network, which comprises the following steps: step 101, acquiring historical processing data of a processed historical workpiece of a servo screw press, inputting the historical processing data into a neural network, and according to the number of data types of the historical processing data; step 102, generating a plurality of production efficiency influence functions according to data types, forming a production efficiency prediction model by the plurality of production efficiency influence functions according to historical processing data, taking the plurality of production efficiency influence functions as a first hidden layer, taking the production efficiency prediction model as a second hidden layer, and setting a production efficiency activation function as a third hidden layer; step 103, setting a production efficiency loss function, setting the production efficiency loss function at an output layer, and adjusting a production efficiency prediction model through the production efficiency loss function; step 104, iterating steps 101 to 103 until the error between the predicted value of the production efficiency and the actual value of the production efficiency is minimum.
Description
Technical Field
The invention belongs to the technical field of production efficiency prediction of a servo screw press, and particularly relates to a method and a system for predicting the production efficiency of the servo screw press based on a neural network.
Background
The servo numerical control screw press is a screw press combining servo control and numerical control technologies. The screw press is accurately controlled and automatically operated by adopting a servo driving system and a numerical control system.
The servo numerical control screw press has the following characteristics and advantages:
and (3) accurate control: the servo system can provide high-precision position control and force control, so that the pressure application of the press is more accurate and stable.
And (3) automation operation: by means of a numerical control system, an automatic working flow can be realized, and the production efficiency and consistency are improved, wherein the working flow comprises program control, parameter setting, continuous operation of multiple procedures and the like.
Multifunction: the servo numerical control screw press generally has a plurality of working modes and functional options, and can adapt to different pressure requirements and workpiece requirements.
Data recording and analysis: the numerical control system can record working data and related parameters of the press machine, is used for monitoring and analyzing the production process, and helps to optimize production efficiency and quality control.
Programming flexibility: the numerical control programming language is used, so that the working program of the press can be flexibly adjusted and modified to adapt to different workpiece processing requirements.
The servo numerical control screw press is widely applied to the fields of automobile part manufacturing, aerospace, electronic devices, plastic molds and the like so as to meet the pressure processing requirements of high precision and high efficiency.
In the prior art, some technologies for analyzing production data of a servo numerical control screw press exist, but the accuracy of the technologies is not high enough, so that the effect of data analysis is greatly reduced, and meanwhile, no technology capable of predicting production efficiency according to historical data exists.
Disclosure of Invention
In order to solve the technical problems, the invention provides a method for predicting the production efficiency of a servo screw press based on a neural network, which comprises the following steps:
step 101, acquiring historical processing data of a processed historical workpiece of the servo screw press, taking the historical processing data as input, inputting the input data into a neural network, and setting the number of neurons of an input layer according to the number of data types of the historical processing data;
step 102, generating a plurality of production efficiency influence functions according to data types, forming a production efficiency prediction model by a plurality of production efficiency influence functions according to the historical processing data, taking the plurality of production efficiency influence functions as a first layer hidden layer, taking the production efficiency prediction model as a second layer hidden layer, and setting a production efficiency activation function as a third layer hidden layer;
Step 103, setting a production efficiency loss function, setting the production efficiency loss function at an output layer, receiving a production efficiency predicted value output by the third hidden layer by the output layer, and adjusting the production efficiency prediction model through the production efficiency loss function;
step 104, iterating the step 101 to the step 103 until the error between the predicted production efficiency value and the actual production efficiency value is minimum, obtaining the current machining data of the current workpiece, and inputting the current machining data into the trained neural network so as to complete the production efficiency prediction.
Further, the historical processing data includes:
historical pressure, workpiece historical processing location, historical yield, machine historical temperature, and grinding tool historical wear level.
Further, the production efficiency influence function is specifically:
，
，
，
，
，
wherein a, b and c are adjustment factors, P is the historical pressure, T is the historical temperature of the machine, R is the historical processing position of the workpiece, L is the historical yield, W is the historical wear degree of the grinding tool,
and the influence function of the historical pressure on the production efficiencyInfluence function of the history temperature of the machine on the production efficiency +. >Influence function of the historical processing position of the workpiece on the production efficiency +.>Influence function of said historical yield on production efficiency +.>And the influence function of the historic wear degree of the grinding tool on the production efficiency +.>Respectively located in corresponding neurons in the second hidden layer.
Further, the production efficiency prediction model includes:
,
wherein, E is a predicted value of production efficiency,as a function of the influence of said historical pressure on the production efficiency,/->As a function of the influence of the historical temperature of the machine on the production efficiency +.>As a function of the influence of the historical processing position of the workpiece on the production efficiency +.>As a function of the influence of the historical yield on the production efficiency,/->Is a function of the influence of the historical wear degree of the grinding tool on the production efficiency.
Further, the production efficiency activation function is:
,
and inputting the production efficiency predicted value generated by the production efficiency predicted model into the production efficiency activating function, wherein the production efficiency activating function firstly transforms the production efficiency predicted value through a sine function, then retains the positive value of the production efficiency predicted value and cuts off the negative value to zero.
Further, the production efficiency loss function is:
,
wherein , and />For the weight, for controlling the mean square error and the mean absolute error, respectively, by adjusting +.> and />To adjust the specific gravity of mean square error and mean absolute error, < >>For the production efficiency true value, N is the number of samples.
Further, step 101 further includes performing a data preprocessing operation on the historical processing data, specifically:
and carrying out data preprocessing operation on the historical processing data, wherein the data preprocessing operation comprises data cleaning, missing value processing and feature scaling, so that consistency and usability of the historical processing data are ensured.
Further, step 101 further includes performing normalization processing on the preprocessed historical processing data, specifically:
and carrying out standardization processing on the preprocessed historical processing data to eliminate the influence of different scales or units, wherein the standardization processing comprises zero-mean processing or unit variance processing so as to ensure that the historical processing data has comparability and consistency.
Further, step 101 further includes dividing the normalized historical processing data into a training set and a verification set, so as to avoid the problem of overfitting or insufficient generalization performance.
Further, the method further comprises the following steps:
and verifying the trained neural network according to the verification set, and adjusting the super parameters of the trained neural network according to the verification result to obtain better prediction performance.
The invention also provides a servo screw press production efficiency prediction system based on the neural network, which comprises the following steps:
the historical data acquisition module is used for acquiring historical processing data of the processed historical workpiece of the servo screw press, taking the historical processing data as input, inputting the input data into a neural network, and setting the number of neurons of an input layer according to the number of data types of the historical processing data;
the method comprises the steps of setting a neural network module, wherein the neural network module is used for generating a plurality of production efficiency influence functions according to data types, processing data according to history, forming a production efficiency prediction model by the plurality of production efficiency influence functions, taking the plurality of production efficiency influence functions as a first layer hidden layer, taking the production efficiency prediction model as a second layer hidden layer, and setting a production efficiency activation function as a third layer hidden layer;
the prediction module is used for setting a production efficiency loss function, setting the production efficiency loss function at an output layer, receiving a production efficiency predicted value output by the third hidden layer by the output layer, and adjusting the production efficiency prediction model through the production efficiency loss function;
And the iteration output module is used for iterating the historical data acquisition module, the neural network setting module and the prediction module until the error between the production efficiency predicted value and the production efficiency true value is minimum, acquiring current processing data of the current workpiece, and inputting the current processing data into the trained neural network so as to complete production efficiency prediction.
Further, the historical processing data includes:
historical pressure, workpiece historical processing location, historical yield, machine historical temperature, and grinding tool historical wear level.
Further, the production efficiency influence function is specifically:
,
,
,
,
,
wherein a, b and c are adjustment factors, P is the historical pressure, T is the historical temperature of the machine, R is the historical processing position of the workpiece, L is the historical yield, W is the historical wear degree of the grinding tool,
and the influence function of the historical pressure on the production efficiencyInfluence function of the history temperature of the machine on the production efficiency +.>Influence function of the historical processing position of the workpiece on the production efficiency +.>Influence function of said historical yield on production efficiency +.>And the influence function of the historic wear degree of the grinding tool on the production efficiency +. >Respectively located in corresponding neurons in the second hidden layer.
Further, the production efficiency prediction model includes:
,
wherein, E is a predicted value of production efficiency,as a function of the influence of said historical pressure on the production efficiency,/->As a function of the influence of the historical temperature of the machine on the production efficiency +.>As a function of the influence of the historical processing position of the workpiece on the production efficiency +.>As a function of the influence of the historical yield on the production efficiency,/->Is a function of the influence of the historical wear degree of the grinding tool on the production efficiency.
Further, the production efficiency activation function is:
,
and inputting the production efficiency predicted value generated by the production efficiency predicted model into the production efficiency activating function, wherein the production efficiency activating function firstly transforms the production efficiency predicted value through a sine function, then retains the positive value of the production efficiency predicted value and cuts off the negative value to zero.
Further, the production efficiency loss function is:
,
wherein , and />For the weight, respectivelyIn controlling the mean square error and the average absolute error, by adjusting +.>Andto adjust the specific gravity of mean square error and mean absolute error, < >>For the production efficiency true value, N is the number of samples.
Further, the historical data obtaining module further includes a data preprocessing operation for the historical processing data, specifically:
and carrying out data preprocessing operation on the historical processing data, wherein the data preprocessing operation comprises data cleaning, missing value processing and feature scaling, so that consistency and usability of the historical processing data are ensured.
Further, the historical data obtaining module further includes standardized processing for the preprocessed historical processing data, specifically:
and carrying out standardization processing on the preprocessed historical processing data to eliminate the influence of different scales or units, wherein the standardization processing comprises zero-mean processing or unit variance processing so as to ensure that the historical processing data has comparability and consistency.
Furthermore, the historical processing data after the standardized processing is divided into a training set and a verification set in the historical data acquisition module, so that the problem of insufficient performance of overfitting or generalization is avoided.
Further, the method further comprises the following steps:
and verifying the trained neural network according to the verification set, and adjusting the super parameters of the trained neural network according to the verification result to obtain better prediction performance.
In general, the above technical solutions conceived by the present invention have the following beneficial effects compared with the prior art:
the method comprises the steps of obtaining historical processing data of a processed historical workpiece, inputting the historical processing data into a neural network, generating a plurality of production efficiency influence functions, forming a production efficiency prediction model by the plurality of production efficiency influence functions according to the historical processing data, taking the plurality of production efficiency influence functions as a first hidden layer, taking the production efficiency prediction model as a second hidden layer, and setting a production efficiency activation function as a third hidden layer; setting a production efficiency loss function, setting the production efficiency loss function at an output layer, receiving a production efficiency predicted value output by the third hidden layer by the output layer, and adjusting the production efficiency prediction model through the production efficiency loss function; after iteration, until the error between the production efficiency predicted value and the production efficiency true value is minimum, current processing data of a current workpiece are obtained, and the current processing data are input into a trained neural network so as to complete production efficiency prediction.
Drawings
FIG. 1 is a flow chart of the method of embodiment 1 of the present invention;
fig. 2 is a block diagram of a system of embodiment 2 of the present invention.
Detailed Description
In order to better understand the above technical solutions, the following detailed description will be given with reference to the accompanying drawings and specific embodiments.
The method provided by the invention can be implemented in a terminal environment, wherein the terminal can comprise one or more of the following components: processor, storage medium, and display screen. Wherein the storage medium has stored therein at least one instruction that is loaded and executed by the processor to implement the method described in the embodiments below.
The processor may include one or more processing cores. The processor connects various parts within the overall terminal using various interfaces and lines, performs various functions of the terminal and processes data by executing or executing instructions, programs, code sets, or instruction sets stored in the storage medium, and invoking data stored in the storage medium.
The storage medium may include a random access Memory (Random Access Memory, RAM) or a Read-Only Memory (ROM). The storage medium may be used to store instructions, programs, code sets, or instructions.
The display screen is used for displaying a user interface of each application program.
In addition, it will be appreciated by those skilled in the art that the structure of the terminal described above is not limiting and that the terminal may include more or fewer components, or may combine certain components, or a different arrangement of components. For example, the terminal further includes components such as a radio frequency circuit, an input unit, a sensor, an audio circuit, a power supply, and the like, which are not described herein.
Example 1
As shown in fig. 1, an embodiment of the present invention provides a method for predicting production efficiency of a servo screw press based on a neural network, including:
step 101, acquiring historical processing data of the processed historical workpiece of the servo screw press (the historical processing data can be converted into a format suitable for analysis, for example, the data is arranged into a structured data table (such as CSV or Excel format), or the historical processing data is converted into a format required by a specific data analysis tool or programming language, such as JSON, SQL and the like), the historical processing data is taken as input and is input into a neural network, and the neuron number of an input layer is set according to the number of data types of the historical processing data;
Specifically, the historical processing data includes:
historical pressure, workpiece historical processing location, historical yield, machine historical temperature, and grinding tool historical wear level.
Specifically, step 101 further includes performing a data preprocessing operation on the historical processing data, specifically:
and carrying out data preprocessing operation on the historical processing data, wherein the data preprocessing operation comprises data cleaning, missing value processing and feature scaling, so that consistency and usability of the historical processing data are ensured.
Specifically, step 101 further includes performing normalization processing on the preprocessed historical processing data, specifically:
the pre-processed historical process data is normalized to eliminate the effects of different scales or units, wherein the normalization process includes zero-mean or unit-variance (necessary conversion and processing of the data, e.g., formatting date and time data, unifying different units of data, or performing operations such as data normalization to ensure consistency and comparability of the data) to ensure comparability and consistency of the historical process data.
Specifically, step 101 further includes dividing the normalized historical processing data into a training set and a verification set, so as to avoid the problem of over-fitting or insufficient generalization performance.
Step 102, generating a plurality of production efficiency influence functions according to data types, forming a production efficiency prediction model by a plurality of production efficiency influence functions according to the historical processing data, taking the plurality of production efficiency influence functions as a first layer hidden layer, taking the production efficiency prediction model as a second layer hidden layer, and setting a production efficiency activation function as a third layer hidden layer;
specifically, the production efficiency influence function is specifically:
,
,
，
，
，
wherein a, b and c are adjustment factors,
and the influence function of the historical pressure on the production efficiencyInfluence function of the history temperature of the machine on the production efficiency +.>Influence function of the historical processing position of the workpiece on the production efficiency +.>Influence function of said historical yield on production efficiency +.>And the influence function of the historic wear degree of the grinding tool on the production efficiency +.>Respectively located in corresponding neurons in the second hidden layer.
Specifically, the production efficiency prediction model includes:
，
Wherein E is a predicted value of production efficiency, P is the historical pressure, T is the historical temperature of the machine, R is the historical processing position of the workpiece, L is the historical yield, W is the historical wear degree of the grinding tool,as a function of the influence of said historical pressure on the production efficiency,/->As a function of the influence of the historical temperature of the machine on the production efficiency +.>As a function of the influence of the historical processing position of the workpiece on the production efficiency +.>As a function of the influence of the historical yield on the production efficiency,/->Is a function of the influence of the historical wear degree of the grinding tool on the production efficiency.
Specifically, the production efficiency activation function is:
，
inputting a production efficiency predicted value generated by the production efficiency prediction model into the production efficiency activation function, wherein the production efficiency activation function firstly transforms the production efficiency predicted value through a sine function, then reserves a positive value of the production efficiency predicted value and cuts off a negative value to zero; the production efficiency activation function has the following benefits:
nonlinear: the introduction of the sine function enables the activation function to be nonlinear, and the nonlinear relation between the input and the output can be captured better;
smoothness: the sine function is continuous and conductive in the whole real number domain, and has a smooth characteristic;
Zero truncation: similar to ReLU, when the input is less than zero, the activation function output is zero.
The production efficiency activation function may help the network learn patterns with periodicity or volatility when applied to hidden layers or output layers in the neural network.
Step 103, setting a production efficiency loss function, setting the production efficiency loss function at an output layer, receiving a production efficiency predicted value output by the third hidden layer by the output layer, and adjusting the production efficiency prediction model through the production efficiency loss function;
specifically, the production efficiency loss function is:
，
wherein , and />For the weight, for controlling the mean square error and the mean absolute error, respectively, by adjusting +.> and />To adjust the specific gravity of mean square error and mean absolute error, < >>For the production efficiency true value, N is the number of samples.
Step 104, iterating the step 101 to the step 103 until the error between the predicted production efficiency value and the actual production efficiency value is minimum, obtaining the current machining data of the current workpiece, and inputting the current machining data into the trained neural network so as to complete the production efficiency prediction.
Specifically, the method further comprises the following steps:
and verifying the trained neural network according to the verification set, and adjusting the super parameters of the trained neural network according to the verification result to obtain better prediction performance.
Specifically, the method further comprises the following steps:
an appropriate optimization algorithm, such as Gradient Descent (Gradient Descent) or adaptive moment estimation (Adam), is selected to minimize the loss function and update the weights of the neural network.
Example 2
As shown in fig. 2, the embodiment of the invention further provides a system for predicting the production efficiency of a servo screw press based on a neural network, which comprises:
for example, the data is organized into a structured data table (e.g., CSV or Excel format), or the historical machining data is converted into a format required by a particular data analysis tool or programming language, such as JSON, SQL, etc.), the historical machining data is input into a neural network, and the number of neurons of an input layer is set according to the number of data types of the historical machining data;
Specifically, the historical processing data includes:
historical pressure, workpiece historical processing location, historical yield, machine historical temperature, and grinding tool historical wear level.
Specifically, the historical data obtaining module further includes a data preprocessing operation for the historical processing data, specifically:
and carrying out data preprocessing operation on the historical processing data, wherein the data preprocessing operation comprises data cleaning, missing value processing and feature scaling, so that consistency and usability of the historical processing data are ensured.
Specifically, the historical data obtaining module further includes standardized processing for the preprocessed historical processing data, specifically:
the pre-processed historical process data is normalized to eliminate the effects of different scales or units, wherein the normalization process includes zero-mean or unit-variance (necessary conversion and processing of the data, e.g., formatting date and time data, unifying different units of data, or performing operations such as data normalization to ensure consistency and comparability of the data) to ensure comparability and consistency of the historical process data.
Specifically, the historical processing data after standardized processing is divided into a training set and a verification set in the historical data acquisition module, so that the problem of insufficient fitting or generalization performance is avoided.
The method comprises the steps of setting a neural network module, wherein the neural network module is used for generating a plurality of production efficiency influence functions according to data types, processing data according to history, forming a production efficiency prediction model by the plurality of production efficiency influence functions, taking the plurality of production efficiency influence functions as a first layer hidden layer, taking the production efficiency prediction model as a second layer hidden layer, and setting a production efficiency activation function as a third layer hidden layer;
specifically, the production efficiency influence function is specifically:
，
，
，
，
，
wherein a, b and c are adjustment factors,
and the influence function of the historical pressure on the production efficiencyInfluence function of the history temperature of the machine on the production efficiency +.>Influence function of the historical processing position of the workpiece on the production efficiency +.>Influence function of said historical yield on production efficiency +.>And the influence function of the historic wear degree of the grinding tool on the production efficiency +.>Respectively located in corresponding neurons in the second hidden layer.
Specifically, the production efficiency prediction model includes:
，
wherein E is a predicted value of production efficiency, P is the historical pressure, T is the historical temperature of the machine, R is the historical processing position of the workpiece, L is the historical yield, W is the historical wear degree of the grinding tool,as a function of the influence of said historical pressure on the production efficiency,/->As a function of the influence of the historical temperature of the machine on the production efficiency +.>As a function of the influence of the historical processing position of the workpiece on the production efficiency +.>As a function of the influence of the historical yield on the production efficiency,/->Is a function of the influence of the historical wear degree of the grinding tool on the production efficiency.
Specifically, the production efficiency activation function is:
，
the production efficiency prediction value generated by the production efficiency prediction model is input into the production efficiency activation function, the production efficiency activation function firstly transforms the production efficiency prediction value through a sine function, then the positive value of the production efficiency prediction value is reserved, and the negative value is truncated to zero, and the production efficiency activation function has the following advantages:
nonlinear: the introduction of the sine function enables the activation function to be nonlinear, and the nonlinear relation between the input and the output can be captured better;
Smoothness: the sine function is continuous and conductive in the whole real number domain, and has a smooth characteristic;
zero truncation: similar to ReLU, when the input is less than zero, the activation function output is zero.
The production efficiency activation function may help the network learn patterns with periodicity or volatility when applied to hidden layers or output layers in the neural network.
The prediction module is used for setting a production efficiency loss function, setting the production efficiency loss function at an output layer, receiving a production efficiency predicted value output by the third hidden layer by the output layer, and adjusting the production efficiency prediction model through the production efficiency loss function;
specifically, the production efficiency loss function is:
，
wherein , and />For the weight, for controlling the mean square error and the mean absolute error, respectively, by adjusting +.> and />To adjust the specific gravity of mean square error and mean absolute error, < >>For the production efficiency true value, N is the number of samples.
And the iteration output module is used for iterating the historical data acquisition module, the neural network setting module and the prediction module until the error between the production efficiency predicted value and the production efficiency true value is minimum, acquiring current processing data of the current workpiece, and inputting the current processing data into the trained neural network so as to complete production efficiency prediction.
Specifically, the method further comprises the following steps:
and verifying the trained neural network according to the verification set, and adjusting the super parameters of the trained neural network according to the verification result to obtain better prediction performance.
Specifically, the method further comprises the following steps:
an appropriate optimization algorithm, such as Gradient Descent (Gradient Descent) or adaptive moment estimation (Adam), is selected to minimize the loss function and update the weights of the neural network.
Example 3
The embodiment of the invention also provides a storage medium which stores a plurality of instructions for realizing the method for predicting the production efficiency of the servo screw press based on the neural network.
Alternatively, in this embodiment, the storage medium may be located in any one of the computer terminals in the computer terminal group in the computer network, or in any one of the mobile terminals in the mobile terminal group.
Alternatively, in the present embodiment, the storage medium is configured to store program code for performing the steps of: step 101, acquiring historical processing data of the processed historical workpiece of the servo screw press (the historical processing data can be converted into a format suitable for analysis, for example, the data is arranged into a structured data table (such as CSV or Excel format), or the historical processing data is converted into a format required by a specific data analysis tool or programming language, such as JSON, SQL and the like), the historical processing data is taken as input and is input into a neural network, and the neuron number of an input layer is set according to the number of data types of the historical processing data;
Specifically, the historical processing data includes:
historical pressure, workpiece historical processing location, historical yield, machine historical temperature, and grinding tool historical wear level.
Specifically, step 101 further includes performing a data preprocessing operation on the historical processing data, specifically:
and carrying out data preprocessing operation on the historical processing data, wherein the data preprocessing operation comprises data cleaning, missing value processing and feature scaling, so that consistency and usability of the historical processing data are ensured.
Specifically, step 101 further includes performing normalization processing on the preprocessed historical processing data, specifically:
the pre-processed historical process data is normalized to eliminate the effects of different scales or units, wherein the normalization process includes zero-mean or unit-variance (necessary conversion and processing of the data, e.g., formatting date and time data, unifying different units of data, or performing operations such as data normalization to ensure consistency and comparability of the data) to ensure comparability and consistency of the historical process data.
Specifically, step 101 further includes dividing the normalized historical processing data into a training set and a verification set, so as to avoid the problem of over-fitting or insufficient generalization performance.
Step 102, generating a plurality of production efficiency influence functions according to data types, forming a production efficiency prediction model by a plurality of production efficiency influence functions according to the historical processing data, taking the plurality of production efficiency influence functions as a first layer hidden layer, taking the production efficiency prediction model as a second layer hidden layer, and setting a production efficiency activation function as a third layer hidden layer;
specifically, the production efficiency influence function is specifically:
，
，/>
，
，
，
wherein a, b and c are adjustment factors,
and the influence function of the historical pressure on the production efficiencyInfluence function of the history temperature of the machine on the production efficiency +.>Influence function of the historical processing position of the workpiece on the production efficiency +.>Influence function of said historical yield on production efficiency +.>And the influence function of the historic wear degree of the grinding tool on the production efficiency +.>Respectively located in corresponding neurons in the second hidden layer.
Specifically, the production efficiency prediction model includes:
，
Wherein E is a predicted value of production efficiency, P is the historical pressure, T is the historical temperature of the machine, R is the historical processing position of the workpiece, L is the historical yield, W is the historical wear degree of the grinding tool,as a function of the influence of said historical pressure on the production efficiency,/->As a function of the influence of the historical temperature of the machine on the production efficiency +.>As a function of the influence of the historical processing position of the workpiece on the production efficiency +.>As a function of the influence of the historical yield on the production efficiency,/->Is a function of the influence of the historical wear degree of the grinding tool on the production efficiency.
Specifically, the production efficiency activation function is:
，
the production efficiency prediction value generated by the production efficiency prediction model is input into the production efficiency activation function, the production efficiency activation function firstly transforms the production efficiency prediction value through a sine function, then the positive value of the production efficiency prediction value is reserved, and the negative value is truncated to zero, and the production efficiency activation function has the following advantages:
nonlinear: the introduction of the sine function enables the activation function to be nonlinear, and the nonlinear relation between the input and the output can be captured better;
smoothness: the sine function is continuous and conductive in the whole real number domain, and has a smooth characteristic;
Zero truncation: similar to ReLU, when the input is less than zero, the activation function output is zero.
The production efficiency activation function may help the network learn patterns with periodicity or volatility when applied to hidden layers or output layers in the neural network.
Step 103, setting a production efficiency loss function, setting the production efficiency loss function at an output layer, receiving a production efficiency predicted value output by the third hidden layer by the output layer, and adjusting the production efficiency prediction model through the production efficiency loss function;
specifically, the production efficiency loss function is:
，
wherein , and />For the weight, for controlling the mean square error and the mean absolute error, respectively, by adjusting +.> and />To adjust the specific gravity of mean square error and mean absolute error, < >>For the production efficiency true value, N is the number of samples.
Step 104, iterating the step 101 to the step 103 until the error between the predicted production efficiency value and the actual production efficiency value is minimum, obtaining the current machining data of the current workpiece, and inputting the current machining data into the trained neural network so as to complete the production efficiency prediction.
Specifically, the method further comprises the following steps:
and verifying the trained neural network according to the verification set, and adjusting the super parameters of the trained neural network according to the verification result to obtain better prediction performance.
Specifically, the method further comprises the following steps:
an appropriate optimization algorithm, such as Gradient Descent (Gradient Descent) or adaptive moment estimation (Adam), is selected to minimize the loss function and update the weights of the neural network.
Example 4
The embodiment of the invention also provides electronic equipment, which comprises a processor and a storage medium connected with the processor, wherein the storage medium stores a plurality of instructions, and the instructions can be loaded and executed by the processor so that the processor can execute the method for predicting the production efficiency of the servo screw press based on the neural network.
Specifically, the electronic device of the present embodiment may be a computer terminal, and the computer terminal may include: one or more processors, and a storage medium.
The storage medium may be used to store a software program and a module, for example, a method for predicting production efficiency of a servo screw press based on a neural network in the embodiments of the present invention, and the processor executes various functional applications and data processing by running the software program and the module stored in the storage medium, that is, implements the method for predicting production efficiency of a servo screw press based on a neural network. The storage medium may include a high-speed random access storage medium, and may also include a non-volatile storage medium, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state storage medium. In some examples, the storage medium may further include a storage medium remotely located with respect to the processor, and the remote storage medium may be connected to the terminal through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The processor may invoke the information stored in the storage medium and the application program via the transmission system to perform the following steps: step 101, acquiring historical processing data of the processed historical workpiece of the servo screw press (the historical processing data can be converted into a format suitable for analysis, for example, the data is arranged into a structured data table (such as CSV or Excel format), or the historical processing data is converted into a format required by a specific data analysis tool or programming language, such as JSON, SQL and the like), the historical processing data is taken as input and is input into a neural network, and the neuron number of an input layer is set according to the number of data types of the historical processing data;
specifically, the historical processing data includes:
historical pressure, workpiece historical processing location, historical yield, machine historical temperature, and grinding tool historical wear level.
Specifically, step 101 further includes performing a data preprocessing operation on the historical processing data, specifically:
and carrying out data preprocessing operation on the historical processing data, wherein the data preprocessing operation comprises data cleaning, missing value processing and feature scaling, so that consistency and usability of the historical processing data are ensured.
Specifically, step 101 further includes performing normalization processing on the preprocessed historical processing data, specifically:
the pre-processed historical process data is normalized to eliminate the effects of different scales or units, wherein the normalization process includes zero-mean or unit-variance (necessary conversion and processing of the data, e.g., formatting date and time data, unifying different units of data, or performing operations such as data normalization to ensure consistency and comparability of the data) to ensure comparability and consistency of the historical process data.
Specifically, step 101 further includes dividing the normalized historical processing data into a training set and a verification set, so as to avoid the problem of over-fitting or insufficient generalization performance.
Step 102, generating a plurality of production efficiency influence functions according to data types, forming a production efficiency prediction model by a plurality of production efficiency influence functions according to the historical processing data, taking the plurality of production efficiency influence functions as a first layer hidden layer, taking the production efficiency prediction model as a second layer hidden layer, and setting a production efficiency activation function as a third layer hidden layer;
Specifically, the production efficiency influence function is specifically:
，
，
，
，
，
wherein a, b and c are adjustment factors,
and the influence function of the historical pressure on the production efficiencyInfluence function of the history temperature of the machine on the production efficiency +.>Influence function of the historical processing position of the workpiece on the production efficiency +.>Influence function of said historical yield on production efficiency +.>And the influence function of the historic wear degree of the grinding tool on the production efficiency +.>Respectively located in corresponding neurons in the second hidden layer.
Specifically, the production efficiency prediction model includes:
，
wherein E is a predicted value of production efficiency, P is the historical pressure, T is the historical temperature of the machine, R is the historical processing position of the workpiece, L is the historical yield, W is the historical wear degree of the grinding tool,as a function of the influence of said historical pressure on the production efficiency,/->As a function of the influence of the historical temperature of the machine on the production efficiency +.>As a function of the influence of the historical processing position of the workpiece on the production efficiency +.>As a function of the influence of the historical yield on the production efficiency,/->Is a function of the influence of the historical wear degree of the grinding tool on the production efficiency.
Specifically, the production efficiency activation function is:
，
The production efficiency prediction value generated by the production efficiency prediction model is input into the production efficiency activation function, the production efficiency activation function firstly transforms the production efficiency prediction value through a sine function, then the positive value of the production efficiency prediction value is reserved, and the negative value is truncated to zero, and the production efficiency activation function has the following advantages:
nonlinear: the introduction of the sine function enables the activation function to be nonlinear, and the nonlinear relation between the input and the output can be captured better;
smoothness: the sine function is continuous and conductive in the whole real number domain, and has a smooth characteristic;
zero truncation: similar to ReLU, when the input is less than zero, the activation function output is zero.
The production efficiency activation function may help the network learn patterns with periodicity or volatility when applied to hidden layers or output layers in the neural network.
Step 103, setting a production efficiency loss function, setting the production efficiency loss function at an output layer, receiving a production efficiency predicted value output by the third hidden layer by the output layer, and adjusting the production efficiency prediction model through the production efficiency loss function;
Specifically, the production efficiency loss function is:
，
wherein , and />Are weights for controlling mean square error and averageAbsolute error by adjusting-> and />To adjust the specific gravity of mean square error and mean absolute error, < >>For the production efficiency true value, N is the number of samples.
Step 104, iterating the step 101 to the step 103 until the error between the predicted production efficiency value and the actual production efficiency value is minimum, obtaining the current machining data of the current workpiece, and inputting the current machining data into the trained neural network so as to complete the production efficiency prediction.
Specifically, the method further comprises the following steps:
and verifying the trained neural network according to the verification set, and adjusting the super parameters of the trained neural network according to the verification result to obtain better prediction performance.
Specifically, the method further comprises the following steps:
an appropriate optimization algorithm, such as Gradient Descent (Gradient Descent) or adaptive moment estimation (Adam), is selected to minimize the loss function and update the weights of the neural network.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
In the foregoing embodiments of the present invention, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the embodiments provided in the present invention, it should be understood that the disclosed technology may be implemented in other manners. The system embodiments described above are merely exemplary, and for example, the division of the units is merely a logic function division, and there may be another division manner in actual implementation, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or partly in the form of a software product or all or part of the technical solution, which is stored in a storage medium, and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a random access Memory (RAM, randomAccess Memory), a removable hard disk, a magnetic disk, or an optical disk, or the like, which can store program codes.
It is apparent that the above examples are given by way of illustration only and are not limiting of the embodiments. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary here nor is it exhaustive of all embodiments. While still being apparent from variations or modifications that may be made by those skilled in the art are within the scope of the invention.
Claims (12)
1. A method for predicting the production efficiency of a servo screw press based on a neural network is characterized by comprising the following steps:
step 101, acquiring historical processing data of a processed historical workpiece of the servo screw press, taking the historical processing data as input, inputting the input data into a neural network, and setting the number of neurons of an input layer according to the number of data types of the historical processing data;
step 102, generating a plurality of production efficiency influence functions according to data types, forming a production efficiency prediction model according to the historical processing data and by a plurality of production efficiency influence functions, taking the plurality of production efficiency influence functions as a first layer hidden layer, taking the production efficiency prediction model as a second layer hidden layer, and setting a production efficiency activation function as a third layer hidden layer, wherein the production efficiency influence functions specifically are as follows:
，
，
，
，
，
wherein a, b and c are adjustment factors, P is historical pressure, T is machine historical temperature, R is workpiece historical processing position, L is historical yield, W is grinding tool historical wear degree,
and the influence function of the historical pressure on the production efficiencyInfluence function of the history temperature of the machine on the production efficiency +. >Influence function of the historical processing position of the workpiece on the production efficiency +.>Influence function of said historical yield on production efficiency +.>And the influence function of the historic wear degree of the grinding tool on the production efficiency +.>Respectively located in corresponding neurons in the second hidden layer;
the production efficiency prediction model comprises:
，
wherein, E is a predicted value of production efficiency,as a function of the influence of said historical pressure on the production efficiency,/->As a function of the influence of the historical temperature of the machine on the production efficiency +.>As a function of the influence of the historical processing position of the workpiece on the production efficiency +.>As a function of the influence of the historical yield on the production efficiency,/->An influence function of the historical wear degree of the grinding tool on production efficiency;
the production efficiency activation function is as follows:
,
inputting a production efficiency predicted value generated by the production efficiency prediction model into the production efficiency activation function, wherein the production efficiency activation function firstly transforms the production efficiency predicted value through a sine function, then reserves a positive value of the production efficiency predicted value and cuts off a negative value to zero;
step 103, setting a production efficiency loss function, setting the production efficiency loss function at an output layer, receiving a production efficiency predicted value output by the third hidden layer by the output layer, and adjusting the production efficiency prediction model through the production efficiency loss function, wherein the production efficiency loss function is as follows:
,
wherein , and />For the weight, for controlling the mean square error and the mean absolute error, respectively, by adjusting +.> and />To adjust the specific gravity of mean square error and mean absolute error, < >>N is the number of samples for the production efficiency true value;
step 104, iterating the step 101 to the step 103 until the error between the predicted production efficiency value and the actual production efficiency value is minimum, obtaining the current machining data of the current workpiece, and inputting the current machining data into the trained neural network so as to complete the production efficiency prediction.
2. The method for predicting production efficiency of a neural network-based servo screw press of claim 1, wherein the historical processing data comprises:
historical pressure, workpiece historical processing location, historical yield, machine historical temperature, and grinding tool historical wear level.
3. The method for predicting production efficiency of a neural network-based servo screw press according to claim 1, wherein step 101 further comprises performing a data preprocessing operation on the historical processing data, specifically:
and carrying out data preprocessing operation on the historical processing data, wherein the data preprocessing operation comprises data cleaning, missing value processing and feature scaling, so that consistency and usability of the historical processing data are ensured.
4. The method for predicting production efficiency of a neural network-based servo screw press according to claim 3, wherein step 101 further comprises performing normalization processing on the preprocessed historical processing data, specifically:
and carrying out standardization processing on the preprocessed historical processing data to eliminate the influence of different scales or units, wherein the standardization processing comprises zero-mean processing or unit variance processing so as to ensure that the historical processing data has comparability and consistency.
5. The method for predicting production efficiency of a neural network-based servo screw press according to claim 4, wherein step 101 further comprises dividing the normalized historical processing data into a training set and a validation set to avoid the problem of overfitting or generalization performance deficiency.
6. The neural network-based servo screw press production efficiency prediction method as set forth in claim 5, further comprising:
and verifying the trained neural network according to the verification set, and adjusting the super parameters of the trained neural network according to the verification result to obtain better prediction performance.
7. A servo screw press production efficiency prediction system based on neural network is characterized by comprising:
the historical data acquisition module is used for acquiring historical processing data of the processed historical workpiece of the servo screw press, taking the historical processing data as input, inputting the input data into a neural network, and setting the number of neurons of an input layer according to the number of data types of the historical processing data;
the method comprises the steps of setting a neural network module, generating a plurality of production efficiency influence functions according to data types, processing data according to history, forming a production efficiency prediction model by the plurality of production efficiency influence functions, taking the plurality of production efficiency influence functions as a first layer hidden layer, taking the production efficiency prediction model as a second layer hidden layer, setting a production efficiency activation function as a third layer hidden layer, wherein the production efficiency influence functions are specifically as follows:
，
，
，
，
，
wherein a, b and c are adjustment factors, P is historical pressure, T is machine historical temperature, R is workpiece historical processing position, L is historical yield, W is grinding tool historical wear degree,
and the influence function of the historical pressure on the production efficiency Influence function of the history temperature of the machine on the production efficiency +.>Influence function of the historical processing position of the workpiece on the production efficiency +.>Influence function of said historical yield on production efficiency +.>And the influence function of the historic wear degree of the grinding tool on the production efficiency +.>Respectively located in corresponding neurons in the second hidden layer;
the production efficiency prediction model comprises:
,
wherein, E is a predicted value of production efficiency,as a function of the influence of said historical pressure on the production efficiency,/->As a function of the influence of the historical temperature of the machine on the production efficiency +.>As a function of the influence of the historical processing position of the workpiece on the production efficiency +.>As a function of the influence of the historical yield on the production efficiency,/->An influence function of the historical wear degree of the grinding tool on production efficiency;
the production efficiency activation function is as follows:
,
inputting a production efficiency predicted value generated by the production efficiency prediction model into the production efficiency activation function, wherein the production efficiency activation function firstly transforms the production efficiency predicted value through a sine function, then reserves a positive value of the production efficiency predicted value and cuts off a negative value to zero;
the prediction module is configured to set a production efficiency loss function, set the production efficiency loss function at an output layer, and receive a production efficiency predicted value output by the third hidden layer by the output layer, and adjust the production efficiency prediction model through the production efficiency loss function, where the production efficiency loss function is:
,
wherein , and />For the weight, for controlling the mean square error and the mean absolute error, respectively, by adjusting +.> and />To adjust the specific gravity of mean square error and mean absolute error, < >>N is the number of samples for the production efficiency true value;
and the iteration output module is used for iterating the historical data acquisition module, the neural network setting module and the prediction module until the error between the production efficiency predicted value and the production efficiency true value is minimum, acquiring current processing data of the current workpiece, and inputting the current processing data into the trained neural network so as to complete production efficiency prediction.
8. The neural network-based servo screw press production efficiency prediction system of claim 7, wherein the historical process data comprises:
historical pressure, workpiece historical processing location, historical yield, machine historical temperature, and grinding tool historical wear level.
9. The neural network-based servo screw press production efficiency prediction system of claim 7, wherein the historical data acquisition module further comprises a data preprocessing operation for the historical processing data, specifically:
And carrying out data preprocessing operation on the historical processing data, wherein the data preprocessing operation comprises data cleaning, missing value processing and feature scaling, so that consistency and usability of the historical processing data are ensured.
10. The neural network-based servo screw press production efficiency prediction system of claim 9, wherein the historical data acquisition module further comprises a step of performing standardized processing on the preprocessed historical processing data, specifically:
and carrying out standardization processing on the preprocessed historical processing data to eliminate the influence of different scales or units, wherein the standardization processing comprises zero-mean processing or unit variance processing so as to ensure that the historical processing data has comparability and consistency.
11. The neural network-based servo screw press production efficiency prediction system of claim 9, wherein the obtaining historical data module further comprises dividing the standardized historical processing data into a training set and a verification set, so as to avoid the problems of over-fitting or insufficient generalization performance.
12. The neural network-based servo screw press production efficiency prediction system of claim 11, further comprising:
And verifying the trained neural network according to the verification set, and adjusting the super parameters of the trained neural network according to the verification result to obtain better prediction performance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310882619.9A CN116629454B (en) | 2023-07-19 | 2023-07-19 | Method and system for predicting production efficiency of servo screw press based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310882619.9A CN116629454B (en) | 2023-07-19 | 2023-07-19 | Method and system for predicting production efficiency of servo screw press based on neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116629454A CN116629454A (en) | 2023-08-22 |
| CN116629454B true CN116629454B (en) | 2023-10-03 |
Family
ID=87617345
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310882619.9A Active CN116629454B (en) | 2023-07-19 | 2023-07-19 | Method and system for predicting production efficiency of servo screw press based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116629454B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111199305A (en) * | 2018-11-20 | 2020-05-26 | 上海烟草集团有限责任公司 | Neural network-based production energy consumption prediction method and system, electronic terminal and storage medium |
| CN111274737A (en) * | 2020-02-25 | 2020-06-12 | 山东大学 | Method and system for predicting remaining service life of mechanical equipment |
| CN111950784A (en) * | 2020-08-04 | 2020-11-17 | 广东工业大学 | Productivity prediction method integrating attention mechanism |
| CN112163297A (en) * | 2020-09-30 | 2021-01-01 | 厦门科灿信息技术有限公司 | Equipment health prediction system |
| CN115130671A (en) * | 2022-08-31 | 2022-09-30 | 合肥中科迪宏自动化有限公司 | Training method of equipment comprehensive efficiency prediction model, storage medium and electronic equipment |
| CN115392534A (en) * | 2022-07-13 | 2022-11-25 | 广东工业大学 | FFM (fringe field model) -based workshop equipment comprehensive efficiency prediction method and system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112292703A (en) * | 2018-06-27 | 2021-01-29 | 西门子股份公司 | Equipment management method, device, system and storage medium |
| WO2021188354A1 (en) * | 2020-03-14 | 2021-09-23 | DataRobot, Inc. | Automated and adaptive design and training of neural networks |
-
2023
- 2023-07-19 CN CN202310882619.9A patent/CN116629454B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111199305A (en) * | 2018-11-20 | 2020-05-26 | 上海烟草集团有限责任公司 | Neural network-based production energy consumption prediction method and system, electronic terminal and storage medium |
| CN111274737A (en) * | 2020-02-25 | 2020-06-12 | 山东大学 | Method and system for predicting remaining service life of mechanical equipment |
| CN111950784A (en) * | 2020-08-04 | 2020-11-17 | 广东工业大学 | Productivity prediction method integrating attention mechanism |
| CN112163297A (en) * | 2020-09-30 | 2021-01-01 | 厦门科灿信息技术有限公司 | Equipment health prediction system |
| CN115392534A (en) * | 2022-07-13 | 2022-11-25 | 广东工业大学 | FFM (fringe field model) -based workshop equipment comprehensive efficiency prediction method and system |
| CN115130671A (en) * | 2022-08-31 | 2022-09-30 | 合肥中科迪宏自动化有限公司 | Training method of equipment comprehensive efficiency prediction model, storage medium and electronic equipment |
Non-Patent Citations (3)
| Title |
|---|
| 发动机机匣生产线关键设备效能评估及预测性维护;杨瑞元;《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》(第1期);C039-279 * |
| 基于Elman神经网络模型的IGBT寿命预测;刘子英;朱琛磊;;半导体技术(第05期);全文 * |
| 基于时间――高斯混合模型的刀具磨损预测;黄文;鲁娟;马俊燕;李康;廖小平;周刚;;工具技术(第10期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116629454A (en) | 2023-08-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3620868B1 (en) | Apparatus, method, program, and recording medium to output recommended control conditions | |
| CN116009480B (en) | Fault monitoring method, device and equipment of numerical control machine tool and storage medium | |
| CN109176168B (en) | Tool sharpening tune machine device and method | |
| WO2006110248A1 (en) | Control system and method | |
| JPH07191737A (en) | Apparatus and method for control of factory | |
| CN113534741A (en) | Control method and system for milling thin-walled workpiece | |
| US10303161B2 (en) | Apparatus and method for dynamic operation of machines | |
| US20220397888A1 (en) | System and method for determining operational configuration of an asset | |
| CN116629454B (en) | Method and system for predicting production efficiency of servo screw press based on neural network | |
| CN117608241B (en) | Method, system, device and medium for updating digital twin model of numerical control machine tool | |
| CN115062674B (en) | Tool arrangement and tool changing method and device based on deep learning and storage medium | |
| CN114755974B (en) | Complex structural member machining path optimization method and device, terminal and storage medium | |
| Braatz et al. | μ-sensitivities as an aid for robust identification | |
| CN116339145A (en) | Machine parameter regulation and control method, equipment and storage medium | |
| CN116134387B (en) | Method and system for determining the compression ratio of an AI model for an industrial task | |
| KR20230081136A (en) | Platform system for recommendating cutting tool according usage | |
| CN113330469A (en) | Method for optimizing a modular system of technical functional units for a process plant | |
| Walker et al. | Digital Twin Technology for CNC Machining: A Review | |
| CN109145254A (en) | A kind of calculation method and calculating equipment of accuracy rate | |
| Jankovič et al. | The concept of automatic generation of hydraulic press cycle | |
| CN117684243B (en) | Intelligent electroplating control system and control method | |
| CN117648876B (en) | TPMS gradient hierarchical structure inverse design manufacturing method based on performance and BALANCE-CGAN | |
| WO2018223015A1 (en) | Methods, devices, and systems for part-to-build | |
| EP4068012A1 (en) | Learning processing apparatus, control apparatus, learning processing method, control method, learning program and control program | |
| CN116594349B (en) | Machine tool prediction method, machine tool prediction device, terminal equipment and computer readable storage medium |
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 |