CN111880489A - Regression scheduling method for complex manufacturing system - Google Patents
Regression scheduling method for complex manufacturing system Download PDFInfo
- Publication number
- CN111880489A CN111880489A CN202010647895.3A CN202010647895A CN111880489A CN 111880489 A CN111880489 A CN 111880489A CN 202010647895 A CN202010647895 A CN 202010647895A CN 111880489 A CN111880489 A CN 111880489A
- Authority
- CN
- China
- Prior art keywords
- scheduling
- manufacturing system
- matrix
- hidden layer
- complex manufacturing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 109
- 238000000034 method Methods 0.000 title claims abstract description 31
- 239000011159 matrix material Substances 0.000 claims abstract description 49
- 238000012549 training Methods 0.000 claims abstract description 29
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 13
- 230000008569 process Effects 0.000 claims description 12
- 230000004913 activation Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 238000005457 optimization Methods 0.000 abstract description 4
- 230000003044 adaptive effect Effects 0.000 abstract description 2
- 238000012545 processing Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 6
- 238000010801 machine learning Methods 0.000 description 6
- 238000007493 shaping process Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 4
- 239000004065 semiconductor Substances 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
- G05B19/41865—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by job scheduling, process planning, material flow
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/32—Operator till task planning
- G05B2219/32252—Scheduling production, machining, job shop
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- General Factory Administration (AREA)
Abstract
The invention discloses a regression scheduling method of a complex manufacturing system, which realizes the acquisition of real-time generation of scheduling rule weight combination adaptive to a production state by constructing a regression scheduling model of the complex manufacturing system based on an extreme learning machine, achieves the aim of multi-target optimization of the manufacturing system, and meanwhile utilizes a proportional-integral-derivative (PID) gradient descent algorithm to solve an output weight matrix of a hidden layer node to complete the training of the extreme learning machine model, thereby improving the generalization performance of the algorithm under the conditions of reducing the computational overload and improving the computational efficiency.
Description
Technical Field
The invention relates to the technical field of dynamic scheduling of complex manufacturing systems, in particular to a regression scheduling method of a complex manufacturing system.
Background
Complex manufacturing systems are complex in production process, consisting of multiple associated production processes. When the manufacturing process is relatively stable, the original scheduling strategy can continuously ensure the optimization of the production performance of the system; when the manufacturing system is disturbed by machine faults and the like, the manufacturing environment changes, the previously adopted scheduling strategy fails, and finally the expected production performance cannot be obtained. Therefore, how to dynamically determine an effective scheduling strategy based on the status of the production process is a key to improving the operational performance of a complex manufacturing system. The method for dynamically adjusting the scheduling strategy according to the production state of the manufacturing system is the dynamic scheduling of the complex manufacturing system.
The data-driven modeling method is high in accuracy and calculation efficiency, the data-driven idea is applied to the field of production scheduling, accumulated production data are analyzed by an effective method, relevant knowledge is extracted and used for guiding production, and the method is an important direction for researching the dynamic scheduling problem of the manufacturing system.
Machine learning is an effective method for data-driven modeling, and a data-driven dynamic scheduling model obtained based on machine learning can be roughly divided into two modes: (1) the classification mode is that one scheduling strategy is selected from the existing scheduling strategy set in the dynamic scheduling process to meet the scheduling target of the manufacturing system; (2) the regression model is a set of specific parameter values given to a scheduling strategy containing parameters in the dynamic scheduling process to achieve the scheduling objective of the manufacturing system.
The dynamic scheduling of the complex manufacturing system is a multi-constraint and multi-objective optimization problem, and a simple heuristic scheduling rule is usually used as a scheduling strategy based on a classification scheduling model, so that the multi-objective and multi-constraint requirements of the manufacturing system are difficult to meet; the output weight matrix is obtained by utilizing a ridge regression method based on a traditional extreme learning machine regression model and obtaining an optimal regularization coefficient through a trial and error method, the method is heavy in calculation burden and low in calculation efficiency, and therefore the generalization performance is reduced, and therefore the design of an efficient regression model has very important significance for dynamic scheduling of a complex manufacturing system.
Disclosure of Invention
In view of this, the present invention provides a regression scheduling method for a complex manufacturing system, which can dynamically calculate and generate a scheduling policy of the complex manufacturing system according to a production state of the complex manufacturing system.
The invention provides a regression scheduling method for a complex manufacturing system, which comprises the following steps:
step 2, establishing a regression scheduling model of the complex manufacturing system based on the extreme learning machine model, wherein the regression scheduling model is as shown in a formula (1):
wherein, Yl×N∈Rl×NIs the output matrix of the regression scheduling model, b is the number of hidden layer nodes, Hb×NFor a hidden layer node matrix, Wl×bOutputting a weight matrix, X, for the hidden layer nodem×N∈Rm×NIs an input matrix, X, of the regression scheduling modeli=[xim,m=1,2,...,M]TTo input a momentMatrix Xm×NThe ith input in (1), Yi=[yil,l=1,2,...,L]TIs an output matrix Yl×NThe ith output of (1); a isb×mAs an input weight matrix to the hidden layer nodes, bb×NA bias matrix that is a hidden layer node; phi is a selectable nonlinear activation function; a isb×mAnd bb×NAll are random variables and keep unchanged after being generated;
step 3, training the regression scheduling model by adopting the training samples, and solving an output weight matrix of hidden layer nodes to finish the training of the regression scheduling model; the process of solving the output weight matrix of the hidden layer node is realized by adopting a gradient descent algorithm based on proportional-integral-differential, and comprises the following steps:
step 3.1, randomly generating output weight matrix of hidden layer nodeSetting iteration step length alpha, iteration times K and parameter K in proportional-integral-differential gradient descent algorithmp,ki,kdInitializing the gradient matrix g0Number k of the hidden layer node is 0, and k is initialized to 1;
step 3.2, calculating a gradient matrix of the kth iterative training by adopting a formula (2);
wherein, Yl×NAn ideal output matrix of the regression scheduling model;
step 3.3, calculating an output weight matrix of the hidden layer node by adopting a formula (3);
step 3.4, adding 1 to K, and executing step 3.2 when K is less than or equal to K; otherwise, finishing training and outputting the output weight matrix of the hidden layer node
And 4, inputting the current production state of the complex manufacturing system into the trained regression scheduling model of the complex manufacturing system to obtain the weight coefficient of the combination rule so as to form a scheduling strategy of the complex manufacturing system.
Has the advantages that:
the invention realizes the acquisition of the real-time generation of the scheduling rule weight combination adaptive to the production state by constructing the regression scheduling model of the complex manufacturing system based on the extreme learning machine, achieves the aim of multi-objective optimization of the manufacturing system, simultaneously utilizes the output weight matrix of the hidden layer node obtained based on the proportional-integral-derivative (PID) gradient descent algorithm to complete the training of the extreme learning machine model, can improve the generalization performance of the algorithm under the conditions of reducing the excessive calculation burden and improving the calculation efficiency, optimizes the network structure, improves the calculation efficiency, can meet the requirement of the dynamic scheduling modeling of the complex manufacturing system to a certain extent, and simultaneously provides a new way for more accurately carrying out the dynamic scheduling modeling of the complex manufacturing system.
Drawings
FIG. 1 is a diagram of a prior art dynamic scheduling framework for a complex manufacturing system.
Detailed Description
The invention is described in detail below by way of example with reference to the accompanying drawings.
In general, data-driven complex manufacturing system dynamic scheduling can be generally divided into two phases, namely an offline learning phase of the scheduling model and an online application phase of the scheduling model. Off-line learning of the scheduling model, namely analyzing historical data by adopting a machine learning method according to a scheduling target in an off-line state, establishing the scheduling model and realizing mapping between a production state and a scheduling strategy; and (3) online application of the scheduling model, namely applying the scheduling model to give a scheduling strategy matched with a real-time state during online dynamic scheduling. The complex manufacturing system dynamic scheduling framework is shown in fig. 1 and comprises four parts, namely data acquisition, data processing, machine learning and online dynamic scheduling, wherein the data acquisition is used for acquiring production state data including a manufacturing system state, a scheduling strategy and a performance index from a manufacturing system/simulation model; data processing, namely generating an optimal sample set of a scheduling strategy by adopting methods such as test design and the like based on the acquired production data; machine learning, namely establishing a corresponding scheduling model by adopting machine learning according to different scheduling targets on the basis of the optimal sample set, and storing the scheduling model in a scheduling model library; and (3) carrying out online dynamic scheduling, calling a corresponding scheduling model according to a scheduling target during online scheduling, taking the real-time production state data of the manufacturing system as input, and outputting a scheduling strategy meeting the scheduling target.
The invention provides a regression scheduling method of a complex manufacturing system, which has the basic idea that: the method comprises the steps of establishing a complex manufacturing system regression scheduling model based on an extreme learning machine model, training the complex manufacturing system regression scheduling model by using a historical data training sample of a complex manufacturing system production state and a combination rule weight coefficient, solving an output weight matrix of a hidden layer node by using a proportional-integral-derivative (PID) gradient descent algorithm, and scheduling the complex manufacturing system by using the trained complex manufacturing system regression scheduling model.
The regression scheduling method of the complex manufacturing system comprises the following specific steps:
The following is an example of a complex system, and a semiconductor manufacturing line is taken as an example, the semiconductor manufacturing line includes 104 equipment groups and 228 equipment groups, and 9 products are involved, each of which has hundreds of processes, and the complex system is a typical complex manufacturing system. The production state of the complex manufacturing system is a set of data representing the production state characteristics of the manufacturing system, and describes the states of the workpieces and the production equipment in each processing area in the complex manufacturing system, as shown in table 1.
Name of field | Data item | Data type |
Px_release | Product x feeding amount of scheduling period | Shaping machine |
Eqpy_queue | Y-queue leader of processing area | Shaping machine |
Eqpy_WIP | WIP number integer of processing zone y | |
1/3_lessProcessed WIP | Workpiece number integer with medium processing amount lower than 1/3 | Shaping machine |
2/3_lessProcessed | The processing amount in WIP is less than 2Number of workpieces/3 but higher than 1/3 | Shaping machine |
TABLE 1 set of production states
Generally speaking, the scheduling policy of the complex manufacturing system includes multiple types, so the scheduling policy of the complex manufacturing system may be converted into a combination of a set of scheduling rules, and the scheduling process, i.e., the generation process of the scheduling policy, is converted into a generation process of a combination manner of the scheduling rules, where the combination manner of the scheduling rules may be determined by using a combination rule weight coefficient, and thus the generation process of the scheduling policy is converted into a solution process of the combination rule weight coefficient. For example, taking the semiconductor production line as an example, In combination with a production target to be optimized, three scheduling rules, i.e., First In First Out (FIFO), shortest processing time First (SPT), and Critical Ratio First (CR), may be selected for combination, each scheduling rule has a weight coefficient, and the finally generated scheduling policy is a weighted sum of the three scheduling rules.
Step 2, establishing a regression scheduling model of the complex manufacturing system based on the extreme learning machine model, wherein the regression scheduling model is as shown in a formula (1):
Fn×N=Wn×b·Hb×N
Hb×N=φ(ab×m·Xm×N+bb×N) (1);
wherein, Fn×N∈Rn×NRepresenting the output matrix of the extreme learning machine model, N representing the number of samples, b representing the number of hidden layer nodes, Hb×NRepresenting a hidden layer node matrix, Wn×bRepresenting the hidden layer node output weight matrix, Xm×N∈Rm×NRepresenting input data of the extreme learning machine model, m representing the feature dimension of each input sample vector, ab×mRepresenting the input weight matrix between the input to the hidden layer nodes, bb×NA bias matrix representing hidden layer nodes; phi is a selectable nonlinear activation function; a isb×mAnd bb×NAll are randomly generated and remain unchanged after generation.
And 3, training the extreme learning machine model by adopting the training sample, and solving an output weight matrix of the hidden layer node based on a proportional-integral-derivative (PID) gradient descent algorithm so as to finish the training of the extreme learning machine model.
The training sample is divided into two parts, one part is used for training, and the other part is used for testing; and after the training of the extreme learning machine model is finished, testing by using the test sample.
The process of solving the output weight matrix of the hidden layer node is realized by adopting a descending algorithm based on proportional-integral-differential gradient, and comprises the following steps:
definition of, Yn×NAn ideal output matrix of the extreme learning machine model;
step 3.1, randomly generating output weight matrix of hidden layer nodeArtificially setting iteration step length alpha, iteration times m and parameter k in proportional-integral-derivative (PID) algorithmp,ki,kdGradient matrix g0Number k of the hidden layer node is 0, and k is 1;
step 3.2, calculating a gradient matrix of the kth iterative training by adopting a formula (2);
step 3.3, calculating an output weight matrix of the hidden layer node by adopting a formula (3);
step 1.4, adding k to 1, and executing step 1.2 when k is less than or equal to m; otherwise, finishing training and outputting the output weight matrix of the hidden layer nodeThe flow is ended.
In the process, the output weight matrix of the hidden layer node is solved by adopting a gradient descent algorithm based on proportional-integral-derivative (PID), so that the model is easy to converge, the learning speed of the extreme learning machine model is increased, and the training time of the model is shortened.
And 4, inputting the current production state of the complex manufacturing system into the trained regression scheduling model of the complex manufacturing system to obtain the weight coefficient of the combination rule so as to form a scheduling strategy of the complex manufacturing system.
In summary, the above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (1)
1. The regression scheduling method of the complex manufacturing system is characterized by comprising the following steps of:
step 1, collecting historical data of a production state of a complex manufacturing system and a weight coefficient of a combined scheduling rule to form a training sample, wherein the sample is { X }i,Yi|Xi∈Rm,Yi∈Rl1, 2.., N }, wherein X is Xi=[xim,m=1,2,...,M]TRepresenting the production state, x, of a complex manufacturing systemimM is the mth feature of the production state, M being the total number of features; y isi=[yil,l=1,2,...,L]TRepresenting production state X of a current complex manufacturing systemiThe corresponding optimal scheduling strategy is the weight coefficient of the combined scheduling rule, yilThe weight of the first combined scheduling rule is L, and the total number of the combined scheduling rules is L; i is a sample number, and N is the total number of samples in the training sample set; r is a real number domain;
step 2, establishing a regression scheduling model of the complex manufacturing system based on the extreme learning machine model, wherein the regression scheduling model is as shown in a formula (1):
wherein, Yl×N∈Rl×NIs the output matrix of the regression scheduling model, b is the number of hidden layer nodes, Hb×NFor a hidden layer node matrix, Wl×bOutputting a weight matrix, X, for the hidden layer nodem×N∈Rm×NIs an input matrix, X, of the regression scheduling modeli=[xim,m=1,2,...,M]TFor an input matrix Xm×NThe ith input in (1), Yi=[yil,l=1,2,...,L]TIs an output matrix Yl×NThe ith output of (1); a isb×mAs an input weight matrix to the hidden layer nodes, bb×NA bias matrix that is a hidden layer node; phi is a selectable nonlinear activation function; a isb×mAnd bb×NAll are random variables and keep unchanged after being generated;
step 3, training the regression scheduling model by adopting the training samples, and solving an output weight matrix of hidden layer nodes to finish the training of the regression scheduling model; the process of solving the output weight matrix of the hidden layer node is realized by adopting a gradient descent algorithm based on proportional-integral-differential, and comprises the following steps:
step 3.1, randomly generating output weight matrix of hidden layer nodeSetting iteration step length alpha, iteration times K and parameter K in proportional-integral-differential gradient descent algorithmp,ki,kdInitializing the gradient matrix g0Number k of the hidden layer node is 0, and k is initialized to 1;
step 3.2, calculating a gradient matrix of the kth iterative training by adopting a formula (2);
wherein, Yl×NFor the regression scheduling modelThe ideal output matrix of (2);
step 3.3, calculating an output weight matrix of the hidden layer node by adopting a formula (3);
step 3.4, adding 1 to K, and executing step 3.2 when K is less than or equal to K; otherwise, finishing training and outputting the output weight matrix of the hidden layer node
And 4, inputting the current production state of the complex manufacturing system into the trained regression scheduling model of the complex manufacturing system to obtain the weight coefficient of the combination rule so as to form a scheduling strategy of the complex manufacturing system.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010647895.3A CN111880489B (en) | 2020-07-07 | 2020-07-07 | Regression scheduling method for complex manufacturing system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010647895.3A CN111880489B (en) | 2020-07-07 | 2020-07-07 | Regression scheduling method for complex manufacturing system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111880489A true CN111880489A (en) | 2020-11-03 |
CN111880489B CN111880489B (en) | 2022-12-09 |
Family
ID=73151304
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010647895.3A Active CN111880489B (en) | 2020-07-07 | 2020-07-07 | Regression scheduling method for complex manufacturing system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111880489B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113050567A (en) * | 2021-03-17 | 2021-06-29 | 北京理工大学 | Dynamic scheduling method for intelligent manufacturing system |
CN113780883A (en) * | 2021-09-26 | 2021-12-10 | 无锡唯因特数据技术有限公司 | Production workshop scheduling method and device and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107466816A (en) * | 2017-07-24 | 2017-12-15 | 东华大学 | A kind of irrigation method based on dynamic multilayer extreme learning machine |
CN109445906A (en) * | 2018-10-11 | 2019-03-08 | 北京理工大学 | A kind of virtual machine quantity required prediction technique |
CN110018882A (en) * | 2019-03-29 | 2019-07-16 | 北京理工大学 | A kind of virtual machine performance prediction technique based on width study |
US20190279090A1 (en) * | 2018-03-09 | 2019-09-12 | Tata Consultancy Services Limited | Method and system for incorporating regression into stacked auto encoder (sae) |
-
2020
- 2020-07-07 CN CN202010647895.3A patent/CN111880489B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107466816A (en) * | 2017-07-24 | 2017-12-15 | 东华大学 | A kind of irrigation method based on dynamic multilayer extreme learning machine |
US20190279090A1 (en) * | 2018-03-09 | 2019-09-12 | Tata Consultancy Services Limited | Method and system for incorporating regression into stacked auto encoder (sae) |
CN109445906A (en) * | 2018-10-11 | 2019-03-08 | 北京理工大学 | A kind of virtual machine quantity required prediction technique |
CN110018882A (en) * | 2019-03-29 | 2019-07-16 | 北京理工大学 | A kind of virtual machine performance prediction technique based on width study |
Non-Patent Citations (1)
Title |
---|
邹伟东 等: "基于压缩动量项的增量型 ELM 虚拟机能耗预测", 《自动化学报》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113050567A (en) * | 2021-03-17 | 2021-06-29 | 北京理工大学 | Dynamic scheduling method for intelligent manufacturing system |
CN113780883A (en) * | 2021-09-26 | 2021-12-10 | 无锡唯因特数据技术有限公司 | Production workshop scheduling method and device and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN111880489B (en) | 2022-12-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112907057B (en) | Production scheduling optimization method and system based on improved MOPSO algorithm | |
CN111880489B (en) | Regression scheduling method for complex manufacturing system | |
CN108764540A (en) | Water supply network pressure prediction method based on parallel LSTM series connection DNN | |
CN110009528B (en) | Parameter self-adaptive updating method based on optimal structure multi-dimensional Taylor network | |
CN109599866B (en) | Prediction-assisted power system state estimation method | |
Saric et al. | Estimation of machining time for CNC manufacturing using neural computing | |
CN110018675A (en) | Nonlinear system modeling method based on LWDNN-ARX model | |
Parnianifard et al. | New adaptive surrogate-based approach combined swarm optimizer assisted less tuning cost of dynamic production-inventory control system | |
Savsar et al. | A neural network procedure for kanban allocation in JIT production control systems | |
CN111522240B (en) | MTMLP-ARX model of four-rotor aircraft, identification method, system and storage medium | |
CN115906959A (en) | Parameter training method of neural network model based on DE-BP algorithm | |
CN110766144B (en) | Scalar coupling constant prediction system between atoms based on multi-layer decomposition fuzzy neural network | |
Rafik et al. | Learning and Predictive Energy Consumption Model based on LSTM recursive neural networks | |
Balázs et al. | Comparative investigation of various evolutionary and memetic algorithms | |
Dai et al. | Multi-objectives design optimization based on multi-objectives Gaussian processes for System-in-Package | |
Rospawan et al. | Adaptive predictive PID control using recurrent fuzzy broad learning system for accurate setpoint tracking of digital nonlinear time-delay dynamic systems | |
CN113050567B (en) | Dynamic scheduling method for intelligent manufacturing system | |
Zhiyuan et al. | Research on the evaluation of enterprise competitiveness based on the wavelet neural network forecasting system | |
Fard et al. | Evolutionary short term load forecasting based on ANN: A comparison with the most well-known algorithms | |
Cassandras et al. | Clustering methods for multiresolution simulation modeling | |
Bubnicki | Learning process in a class of computer integrated manufacturing systems with parametric uncertainties | |
Balázs et al. | Comparative analysis of various evolutionary and memetic algorithms | |
Srivastava et al. | A novel recurrent neural network model for the identification of time-delayed nonlinear system | |
Jones et al. | Optimization techniques applied to neural networks: Line search implementation for back propagation | |
Hwangy et al. | Regression Modeling in Back-Propagation and Projection Pursuit Learning3 |
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 |