CN109240227A - A kind of field control method based on Time-sharing control handover control - Google Patents
A kind of field control method based on Time-sharing control handover control Download PDFInfo
- Publication number
- CN109240227A CN109240227A CN201811027145.5A CN201811027145A CN109240227A CN 109240227 A CN109240227 A CN 109240227A CN 201811027145 A CN201811027145 A CN 201811027145A CN 109240227 A CN109240227 A CN 109240227A
- Authority
- CN
- China
- Prior art keywords
- control
- controller
- model
- field
- instruction
- 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
- 238000000034 method Methods 0.000 title claims abstract description 19
- 230000007613 environmental effect Effects 0.000 claims abstract description 28
- 230000005540 biological transmission Effects 0.000 claims abstract description 5
- 238000012986 modification Methods 0.000 claims 1
- 230000004048 modification Effects 0.000 claims 1
- 238000004088 simulation Methods 0.000 claims 1
- 238000010801 machine learning Methods 0.000 abstract description 11
- 239000000463 material Substances 0.000 abstract description 4
- 230000000750 progressive effect Effects 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 2
- 238000003780 insertion Methods 0.000 description 2
- 230000037431 insertion Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 239000000498 cooling water Substances 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
- G05B19/41865—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by job scheduling, process planning, material flow
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N99/00—Subject matter not provided for in other groups of this subclass
-
- 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
- G05B2219/32015—Optimize, process management, optimize production line
-
- 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)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Automation & Control Theory (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The present invention provides a kind of field control methods based on Time-sharing control handover control, include the following steps: 1) model training: holding signal path is signal receiving end-field controller-instruction transmitting terminal, obtain instruction and received data that field controller is sent, and according to the instruction of the transmission got and received data training environment model, environmental model training, which finishes, then enters next step, otherwise repeats the step;2) Developing Tactics;3) intervention control.The present invention is by way of the progressive handover of three stages, it is capable of providing the basis of on-line training, the investment of manpower and material resources is effectively reduced convenient for enterprise, shortens the R&D cycle, the handover of inserting controller of the smooth completion control of energy from traditional field controller to machine learning.
Description
Technical field
The present invention relates to a kind of field control methods based on Time-sharing control handover control.
Background technique
Currently, application of the machine learning in Industry Control gradually increases, however local manufacturing enterprises commonly encounter most instantly
Burden is that data are seriously deficient, this causes the controller of machine learning to be difficult to train completion, and a kind of half-way house is to be divided to two
Phase carries out, and first acquires a period of time data, the code of machine learning controller is completed at the same time, then according to a small amount of number collected
According to being trained, obtain phase machine learning controller, it come into operation, continue in use acquire data it is longer when
Between, then according to the data re -training machine learning model (and adjustment algorithm) all acquired, obtain the second stage of machine learning control
The second stage of machine learning controller is used to finally control by device processed.However, in this way, a R&D cycle is too long, two
Manpower and material resources investment is high, not as good as directly engagement operator for enterprise.
To solve the above problems, our company devises a kind of showing based on Time-sharing control handover control as shown in Figure 1
In addition station control system (is applied for a patent), the field control system can guarantee from hardware from traditional field controller to
Machine learning is that the controller of control core carries out control handover and can be the process gradually substituted, however it is specific how
Control handover is completed, the prior art does not provide technical inspiration.
Summary of the invention
In order to solve the above technical problems, the present invention provides a kind of field control sides based on Time-sharing control handover control
Method should be capable of providing based on the field control method that Time-sharing control joins control by way of the progressive handover of three stages
The basis of line training is effectively reduced the investment of manpower and material resources convenient for enterprise, shortens the R&D cycle.
The present invention is achieved by the following technical programs.
A kind of field control method based on Time-sharing control handover control provided by the invention, includes the following steps:
1) model training: holding signal path is signal receiving end-field controller-instruction transmitting terminal, obtains scene control
The instruction and received data that device processed is sent, and according to the instruction of the transmission got and received data training environment model,
Environmental model training, which finishes, then enters next step, otherwise repeats the step;
2) Developing Tactics: every that signal path is then switched to signal reception by the n times field controller signal transmitting and receiving period
Signal path is switched to signal receiving end-inserting controller-instruction transmitting terminal every time by end-inserting controller-instruction transmitting terminal
When for one controller period of insertion, each cycle interpolation enters controller using received data and environmental model according to a upper control
The analogue data that the device period processed is calculated updates Policy model, and Policy model, which updates, to be completed then to enter next step, otherwise weighs
The multiple step;
3) intervention control: holding signal path is signal receiving end-inserting controller-instruction transmitting terminal, inserting controller
It is controlled by Policy model, and updates Policy model in every M signal transmitting and receiving period.
The training environment model uses RNN algorithm.
The input of the environmental model be controller send instruction, export for n-signal later receive and dispatch the period it is received
Data.
In the step 2), Policy model updates in the following way:
A. the mistake for the analogue data that environmental model is calculated in currently received data and upper controller period was calculated
Difference;
B. the value function of Policy model is updated according to error amount;
C. currently received data are substituting in Policy model and calculate current output order value;
D. current output order value is substituting in environmental model and calculates analogue data;
E. current output order is sent, next controller period is then waited to receive data.
It is that received data and environmental model were counted according to the upper controller period that the Policy model, which updates the condition completed,
Similarity is greater than 90% between obtained analogue data.
The M and N is 10~20.
The M and N are equal.
The condition that the environmental model training finishes is that environmental model sends instruction according to field controller and mould is calculated
Similarity is greater than 90% between quasi- data and received data.
The beneficial effects of the present invention are: by way of the progressive handover of three stages, it is capable of providing the basis of on-line training,
The investment of manpower and material resources is effectively reduced convenient for enterprise, shortens the R&D cycle, can smooth completion control from traditional field control
Device to machine learning inserting controller handover.
Detailed description of the invention
Fig. 1 is the connection schematic diagram of field control system applied by the present invention.
Specific embodiment
Be described further below technical solution of the present invention, but claimed range be not limited to it is described.
The present invention is applied to the field control system based on Time-sharing control handover control as shown in Figure 1, and specifically one
Field control method of the kind based on Time-sharing control handover control, includes the following steps:
1) model training: holding signal path is signal receiving end-field controller-instruction transmitting terminal, obtains scene control
The instruction and received data that device processed is sent, and according to the instruction of the transmission got and received data training environment model,
Environmental model training, which finishes, then enters next step, otherwise repeats the step;
2) Developing Tactics: every that signal path is then switched to signal reception by the n times field controller signal transmitting and receiving period
Signal path is switched to signal receiving end-inserting controller-instruction transmitting terminal every time by end-inserting controller-instruction transmitting terminal
When for one controller period of insertion, each cycle interpolation enters controller using received data and environmental model according to a upper control
The analogue data that the device period processed is calculated updates Policy model, and Policy model, which updates, to be completed then to enter next step, otherwise weighs
The multiple step;
3) intervention control: holding signal path is signal receiving end-inserting controller-instruction transmitting terminal, inserting controller
It is controlled by Policy model, and updates Policy model in every M signal transmitting and receiving period.
The training environment model uses RNN algorithm.
The input of the environmental model be controller send instruction, export for n-signal later receive and dispatch the period it is received
Data.
In the step 2), Policy model updates in the following way:
A. the mistake for the analogue data that environmental model is calculated in currently received data and upper controller period was calculated
Difference;
B. the value function of Policy model is updated according to error amount;
C. currently received data are substituting in Policy model and calculate current output order value;
D. current output order value is substituting in environmental model and calculates analogue data;
E. current output order is sent, next controller period is then waited to receive data.
It is that received data and environmental model were counted according to the upper controller period that the Policy model, which updates the condition completed,
Similarity is greater than 90% between obtained analogue data.
The M and N is 10~20.M and N is substantially that control instruction acts in true environment and generates specific feedback
Action time, therefore specifically should be different according to environment used, be most preferably 17 such as in power plant cooling water emission control.
The M and N are equal.
The condition that the environmental model training finishes is that environmental model sends instruction according to field controller and mould is calculated
Similarity is greater than 90% between quasi- data and received data.
The present invention completes intensified learning model to the control of existing field controller essentially by three phases
Handover, environmental model and Policy model execute in inserting controller, are that the first step first passes through acquisition in real time now generally speaking
There are data to carry out the modeling of environmental model, greatly reduces time and cost needed for collecting data, after the completion of environmental model just
Can be to tactful model modeling, but Policy model needs practical interaction, therefore second step is exactly to be trained by interaction appropriate
Policy model (actually can also be by environmental model virtual training, but result like that is not accurate enough), therefore using tactful mould
Type is directly interacted with true environment within the appropriate period to be a at low cost and (can be based on field control by system fault tolerance mechanism
The stability of device) mode that receives, when all training of environmental model and Policy model are completed, then the control of intensive control module has been
It is mature and can come into operation, third step is entered at this time to be completed control and join to efficiently use intensified learning and adjusted according to environment
The advantage of control strategy realizes more adaptable, the higher field control of accuracy based on machine learning techniques.
Claims (8)
1. a kind of field control method based on Time-sharing control handover control, characterized by the following steps:
1) model training: holding signal path is signal receiving end-field controller-instruction transmitting terminal, obtains field controller
The instruction of transmission and received data, and according to the instruction of the transmission got and received data training environment model, environment
Model training, which finishes, then enters next step, otherwise repeats the step;
2) Developing Tactics: every that signal path is then switched to signal receiving end-by the n times field controller signal transmitting and receiving period
Inserting controller-instruction transmitting terminal, when signal path being switched to signal receiving end-inserting controller-instruction transmitting terminal every time
To be inserted into a controller period, each cycle interpolation enters controller using received data and environmental model according to a upper control
The analogue data that the device period is calculated updates Policy model, and Policy model, which updates, to be completed then to enter next step, otherwise repeats
The step;
3) intervention control: holding signal path is signal receiving end-inserting controller-instruction transmitting terminal, and inserting controller passes through
Policy model control, and Policy model is updated in every M signal transmitting and receiving period.
2. the field control method as described in claim 1 based on Time-sharing control handover control, it is characterised in that: the instruction
Practice environmental model and uses RNN algorithm.
3. the field control method as described in claim 1 based on Time-sharing control handover control, it is characterised in that: the ring
The input of border model is the instruction that controller is sent, and exports and receives and dispatches period received data for n-signal later.
4. the field control method as described in claim 1 based on Time-sharing control handover control, it is characterised in that: the step
It is rapid 2) in, Policy model updates in the following way:
A. the error amount for the analogue data that environmental model is calculated in currently received data and upper controller period was calculated;
B. the value function of Policy model is updated according to error amount;
C. currently received data are substituting in Policy model and calculate current output order value;
D. current output order value is substituting in environmental model and calculates analogue data;
E. current output order is sent, next controller period is then waited to receive data.
5. the field control method as described in claim 1 based on Time-sharing control handover control, it is characterised in that: the plan
The condition that slightly model modification is completed is the simulation number that received data and environmental model were calculated according to the upper controller period
Similarity is greater than 90% between.
6. the field control method as described in claim 1 based on Time-sharing control handover control, it is characterised in that: the M
It is 10~20 with N.
7. the field control method as described in claim 1 based on Time-sharing control handover control, it is characterised in that: the M
It is equal with N.
8. the field control method as described in claim 1 based on Time-sharing control handover control, it is characterised in that: the ring
The condition that border model training finishes is, environmental model sends instruction according to field controller and analogue data and received is calculated
Similarity is greater than 90% between data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811027145.5A CN109240227B (en) | 2018-09-04 | 2018-09-04 | On-site control method based on time-sharing control handover control right |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811027145.5A CN109240227B (en) | 2018-09-04 | 2018-09-04 | On-site control method based on time-sharing control handover control right |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109240227A true CN109240227A (en) | 2019-01-18 |
CN109240227B CN109240227B (en) | 2021-04-13 |
Family
ID=65061170
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811027145.5A Active CN109240227B (en) | 2018-09-04 | 2018-09-04 | On-site control method based on time-sharing control handover control right |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109240227B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6745169B1 (en) * | 1995-07-27 | 2004-06-01 | Siemens Aktiengesellschaft | Learning process for a neural network |
CN102479339A (en) * | 2010-11-24 | 2012-05-30 | 香港理工大学 | Method and system for forecasting short-term wind speed of wind farm based on hybrid neural network |
CN102983819A (en) * | 2012-11-08 | 2013-03-20 | 南京航空航天大学 | Imitating method of power amplifier and imitating device of power amplifier |
CN103410660A (en) * | 2013-05-14 | 2013-11-27 | 湖南工业大学 | Wind power generation variable pitch self-learning control method based on support vector machine |
CN108388205A (en) * | 2017-02-03 | 2018-08-10 | 发那科株式会社 | Learning model construction device and control information optimize device |
US20180240020A1 (en) * | 2017-02-17 | 2018-08-23 | Aman Madaan | Segmentation platform |
-
2018
- 2018-09-04 CN CN201811027145.5A patent/CN109240227B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6745169B1 (en) * | 1995-07-27 | 2004-06-01 | Siemens Aktiengesellschaft | Learning process for a neural network |
CN102479339A (en) * | 2010-11-24 | 2012-05-30 | 香港理工大学 | Method and system for forecasting short-term wind speed of wind farm based on hybrid neural network |
CN102983819A (en) * | 2012-11-08 | 2013-03-20 | 南京航空航天大学 | Imitating method of power amplifier and imitating device of power amplifier |
CN103410660A (en) * | 2013-05-14 | 2013-11-27 | 湖南工业大学 | Wind power generation variable pitch self-learning control method based on support vector machine |
CN108388205A (en) * | 2017-02-03 | 2018-08-10 | 发那科株式会社 | Learning model construction device and control information optimize device |
US20180240020A1 (en) * | 2017-02-17 | 2018-08-23 | Aman Madaan | Segmentation platform |
Non-Patent Citations (1)
Title |
---|
陈海列,高平: "一种基于DCS的分时训练神经网络", 《自动化应用》 * |
Also Published As
Publication number | Publication date |
---|---|
CN109240227B (en) | 2021-04-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108890652B (en) | Transformer substation inspection robot and transformer substation equipment inspection method | |
CN108121216B (en) | Automated workshop virtual debugging method based on digital factory | |
CN108322348B (en) | Transformer substation monitoring information automatic acceptance signal triggering operation subsystem and method | |
CN109444582B (en) | Automatic debugging method for power distribution terminal side based on field mobile terminal and analog master station | |
CN105791025B (en) | A kind of automotive networking communication system forward design method | |
CN108416866A (en) | A kind of patrol task processing method and equipment | |
CN104574204B (en) | Intelligent measuring and control device ICD models automatic identification system and method | |
CN111046943A (en) | Method and system for automatically identifying state of isolation switch of transformer substation | |
CN102073274B (en) | Expert fuzzy incremental type self-adapting parameter on-line setting and optimizing system and method thereof | |
CN106503320B (en) | A kind of sequence model dispatching method of distributing emulation system | |
CN110225100B (en) | Virtual-real mapping system for intelligent assembly production line | |
CN109240227A (en) | A kind of field control method based on Time-sharing control handover control | |
CN109500812A (en) | A kind of robotic programming method positioned in real time by visual pattern | |
CN110765616A (en) | Simulation closed loop test system and method for power plant computer monitoring system | |
CN116560322B (en) | Production control method and system for assembly production line | |
CN105359462B (en) | The method and apparatus that test SDN builds new flow velocity rate | |
CN110516899B (en) | Real-time simulation network-level and provincial-level AGC combined closed-loop test method and device | |
CN110165716A (en) | A method of improving hydraulic turbine primary frequency modulation contributing electricity | |
CN109188904A (en) | A kind of on-line training field control method based on machine learning | |
CN109325593A (en) | A kind of method of field control Time-sharing control handover control | |
CN109366485A (en) | A kind of field control method of online machine learning | |
CN111025135B (en) | Cloud platform-based test method for on-site FTU coordination control capability | |
CN109358531A (en) | A kind of field control system based on Time-sharing control handover control | |
CN106773790A (en) | Based on the synchronous many iteration electric power of clobal axle and communication joint simulation method | |
CN109324508B (en) | Sewage treatment control method based on reinforcement learning |
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 |