CN108319149A - A kind of electromagnetic switch wisdom control system with Self-learning control pattern - Google Patents
A kind of electromagnetic switch wisdom control system with Self-learning control pattern Download PDFInfo
- Publication number
- CN108319149A CN108319149A CN201810253280.5A CN201810253280A CN108319149A CN 108319149 A CN108319149 A CN 108319149A CN 201810253280 A CN201810253280 A CN 201810253280A CN 108319149 A CN108319149 A CN 108319149A
- Authority
- CN
- China
- Prior art keywords
- control
- switch
- control system
- targe
- module
- 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
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
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
Abstract
The present invention provides a kind of electromagnetic switch wisdom control system with Self-learning control pattern comprising：Control targe identification module provides interim control targe for the intelligent process control of electromagnetic switch；Parameter identification module, which characterizes the dynamic control process of switch；State characteristic extracting module receives the current magnetic state information of switch of the contactor main body characteristic parameter observation model output from Parameter identification module；The behavioral characteristics of the electromagnetic system and contact system of state characteristic extracting module real-time characterization contactor；And control strategy computing module, it exports control decision, the control targe and behavioral characteristics that control strategy computing module is switched by real-time reception, obtain a series of movements or a decision that switch will execute, output makes switch kinetic-control system distribute the suction counter-force of switch kinetic-control system according to current working to switch kinetic-control system.Entire control system, which has, to be learnt and adapts to unknown or uncertain system ability.
Description
Technical field
The invention belongs to electric fields, and in particular to a kind of electromagnetic switch wisdom control system with Self-learning control pattern
System.
Background technology
With the development of modern society, requirement of the people to electric energy and power grid is also higher and higher, and currently, domestic and international correlation is ground
Study carefully research and building-up work that mechanism is just carrying out intelligent grid.And intelligent switch electric appliance is the very important composition of intelligent grid
Part, is the material base for building intelligent grid, and the intelligence of device for switching has adapted to the needs of intelligent grid.
Intelligent electric appliance should be with my diagnostic function and adaptive, self study control ability.Adaptively, the control of self study
Ability is that the pith of actuating function is intelligently executed in intelligent electric appliance, can significantly optimize the operation controllability of electric appliance, can
It in the time and spatially realizes speed change or becomes the functions such as track actuating, this can only carry out single division behaviour relative to traditional electric appliance
Work is a kind of technical breakthrough.Adaptively, the control ability of self study requires switch can be according to actual working environment and work
Condition carries out self_adaptive adjusting to action process so that switch performance is in optimum state and always with certain artificial intelligence
Energy characteristic, has the function of the wisdom such as perception, thinking judgement and dynamic optimization.
But intelligent switch is once installed in place mostly at present, the motion process of electromagnetic mechanism just control by basic solidification, intelligence
The relevant parameter of system obtains offline in laboratory, and off-line method is broadly divided into modelling by mechanism and research technique, obtained control
System rule has certain effect to the fixation ontology under same running environment.However, being instructed to connect always with the parameter obtained offline
The operating status of the online time-varying of tentaculum, limitation is big, and the adjusting of kinetic characteristic cannot be carried out according to different operating modes.With existing
For the rapid development of power electronic technique, control technology, microcomputer technology, the control system based on microcomputer is widely used to electric power
Equipment, to improve its performance and automatization level.Adaptive, Self-learning control technology is applied to the actuating control of electromagnetic switch
System realizes the requirement of electromagnetic switch optimal movement characteristic, can increase substantially the property indices of modern electromagnetic switch.
In recent years, some institution of higher learning and research institution start to have carried out theory to the intelligent control problem of switch both at home and abroad
Property and feasibility research, achieve some interim achievements to varying degrees.But contactor specification type is various,
Contact capacity covers the wide scope of 9-2650A, and the characteristics such as counter-force are not quite similar, and operating condition is increasingly complicated, to be widely used in
For the contactor of the new energy fields such as photovoltaic, wind-powered electricity generation, the complex works such as high/low temperature, voltage fluctuation, mechanical oscillation are faced
Condition greatly increases period and the difficulty of modelling by mechanism and test adjustment.Modelling by mechanism relies on contactor actual parameter and vacation
If condition, even if being directed to identical ontology, as self-characteristic changes after contactor long-play, the mould established offline
Type also will gradually disconnect with realistic model, and laboratory is difficult to reappear actual condition.In short, device for switching due to body construction,
The problems such as running environment, load diversity, system complexity, encounter the technology bottle for being difficult to go beyond in design and control
Neck is badly in need of seeking new breakthrough in theoretical and control method.
By the process control of artificial intelligence control theory and design method insertion switch electric appliance, a kind of solve the problems, such as of can yet be regarded as
Idea and method.Over 60 years, artificial intelligence experienced from outburst to severe winter again to the course of cruelty growth, along with man-machine friendship
Change, machine learning, the promotion of the artificial intelligence technologys such as pattern-recognition so that structure is a kind of to have memory and thinking arbitration functions
Machine is possibly realized, and benefits from the application of intelligent algorithm, and the design forms artificial intelligence technology insertion switch electric appliance new
Generation wisdom control system enables switch according to the current trace information of contact system, work condition environment, automatically adjusts electromagnetic system
The magnetic state information for uniting current and control decision form the wisdom electromagnetic switch with self study, adaptation function.To realize
The variation that power grid, environment and control require can be adapted to automatically, and the electric equipment products for being in optimal operating condition always establish base
Plinth.
Invention content
It is an object of the invention to design a kind of self study wisdom control system of electromagnetic switch, the control system is by four
Module forms, including control targe identification, state feature extraction, Parameter identification and control strategy calculate.Modules
On the basis of closed-loop control, in conjunction with intelligent algorithm, makes it have study and adapts to unknown or uncertain system ability,
Meet electromagnetic switch different running environment, the stability control of different working modes requirement, and formed one kind have from become it is excellent,
The intelligent switching control and operational mode of the features such as self study, self-adjusting, multivariable feedback and multiple-objection optimization.
To achieve the above object, the present invention uses following technical scheme：A kind of electromagnetism with Self-learning control pattern is opened
Close wisdom control system comprising：Control targe identification module, the control targe identification module are the intelligent mistake of electromagnetic switch
Process control provides interim control targe；It is inputted connects with the output of state characteristic extracting module；Parameter identification module,
The Parameter identification module characterizes the dynamic control process of switch；The Parameter identification module includes switch dynamic
Control system and contactor main body characteristic parameter observation model；State characteristic extracting module, the state characteristic extracting module connect
Receive the current magnetic state information of switch of the contactor main body characteristic parameter observation model output from Parameter identification module；Shape
The behavioral characteristics of the electromagnetic system and contact system of state characteristic extracting module real-time characterization contactor；And control strategy calculates mould
Block, the control strategy computing module export control decision, the control mesh that control strategy computing module is switched by real-time reception
Mark and behavioral characteristics, obtain a series of movements or a decision that switch will execute, and output makes out to switch kinetic-control system
Close the suction counter-force that kinetic-control system distributes switch kinetic-control system according to current working.
In an embodiment of the present invention, the control targe identification module includes control targe prediction model；Control targe
Prediction model constantly rolls the local suboptimal solution sought under current working, in time refreshing or Correction and Control Model of Target Recognition, begins
New control targe is established on the optimal basis of switch actual working state eventually.
In an embodiment of the present invention, the state characteristic extracting module realizes the more of switch electromagnetic system and contact system
Variable feedback；The state characteristic extracting module includes switching trace data model, switch movement Arc Modelling and evaluation network；
Switch current track information, the switch that the evaluation network exports the switching trace data model move electric arc mould
The switch current arc energy information of type output and the current magnetic state of switch of contactor main body characteristic parameter observation model output
Information carries out state evaluation；The switch current track information of the switching trace data model output, switch movement simultaneously
The switch current arc energy information of Arc Modelling output and the switch of contactor main body characteristic parameter observation model output are current
State input quantity of the magnetic state information as control strategy computing module.
In an embodiment of the present invention, the control strategy computing module includes autonomous learning device；The Self-learning control
Device is made of neural network with each parameter closed loop controller is switched, and the depth closed loop control of switch is realized using multilayer neural network
System；It includes closed-loop current control device, voltage close loop controller and magnetic linkage closed-loop controller to switch each parameter closed loop controller.
Compared with prior art, the present invention has the following advantages：
（1）, the present invention machine learning thought is introduced into the intelligent control of contactor, establish self study wisdom control model, collection
In be embodied in the control targe identification of switch, state feature extraction, Parameter identification and control strategy and calculate etc., make intelligence
Can switch have the function of self study, self-adjusting, adaptive, solve the parameter that existing intelligent control obtains offline with contactor
Always the limitation of the online time-varying operation of contactor is instructed.
（2）, the present invention system control scheme combine closed-loop control Qualitative Knowledge ability to express and machine learning calculate
The quantitative learning ability of method.Enable entire control system non-thread present in contactor motion process efficiently against connecing
The influence of the factors such as property, uncertainty, close coupling, reaches very high control accuracy to control targe.
（3）, control system using multivariable feedback, multiple-objection optimization control model, by roll line solver obtain
The closed-loop optimization sequence of optimum control forms wisdom control system.Model in system is not excellent using a constant overall situation
Change target, but use the finite time-domain optimisation strategy of time rolls forward formula, carries out online repeatedly.Always new excellent
Change and establish on the basis of actual, control is made to keep actual optimal.
Description of the drawings
Fig. 1 is the cardinal principle block diagram of the present invention.
Specific implementation mode
Explanation is further explained to the present invention in the following with reference to the drawings and specific embodiments.
The present invention provides a kind of electromagnetic switch wisdom control system with Self-learning control pattern comprising：Control mesh
Identification module is marked, the control targe identification module provides interim control targe for the intelligent process control of electromagnetic switch；
It is inputted connects with the output of state characteristic extracting module；Parameter identification module, the Parameter identification module characterize
The dynamic control process of switch；The Parameter identification module includes switch kinetic-control system and contactor main body characteristic ginseng
Discharge observation model；State characteristic extracting module, the state characteristic extracting module receive connecing from Parameter identification module
The current magnetic state information of switch of tentaculum main body characteristic parameter observation model output；State characteristic extracting module real-time characterization contacts
The electromagnetic system of device and the behavioral characteristics of contact system；And control strategy computing module, the control strategy computing module are defeated
Go out control decision, the control targe and behavioral characteristics that control strategy computing module is switched by real-time reception, obtaining switch will
The a series of movements of execution or a decision, output make switch kinetic-control system according to current work to switch kinetic-control system
The suction counter-force of condition distribution switch kinetic-control system.
In an embodiment of the present invention, the control targe identification module includes control targe prediction model；Control targe
Prediction model constantly rolls the local suboptimal solution sought under current working, in time refreshing or Correction and Control Model of Target Recognition, begins
New control targe is established on the optimal basis of switch actual working state eventually.
In an embodiment of the present invention, the state characteristic extracting module realizes the more of switch electromagnetic system and contact system
Variable feedback；The state characteristic extracting module includes switching trace data model, switch movement Arc Modelling and evaluation network；
Switch current track information, the switch that the evaluation network exports the switching trace data model move electric arc mould
The switch current arc energy information of type output and the current magnetic state of switch of contactor main body characteristic parameter observation model output
Information carries out state evaluation；The switch current track information of the switching trace data model output, switch movement simultaneously
The switch current arc energy information of Arc Modelling output and the switch of contactor main body characteristic parameter observation model output are current
State input quantity of the magnetic state information as control strategy computing module.
In an embodiment of the present invention, the control strategy computing module includes autonomous learning device；The Self-learning control
Device is made of neural network with each parameter closed loop controller is switched, and the depth closed loop control of switch is realized using multilayer neural network
System；It includes closed-loop current control device, voltage close loop controller and magnetic linkage closed-loop controller to switch each parameter closed loop controller.
Fig. 1 is the electromagnetic switch self study wisdom control system functional block diagram of a specific embodiment of the invention.At dotted line
For boundary, whole system is substantially considered as a closed-loop control network, has four signal processing links.Control targe identifies
Link provides interim control targe for the intelligent process control of electromagnetic switch and system starts the first step of operation.Its
Input is state feature extraction step, its two big system of real-time characterization contactor（Electromagnetic system and contact system）Dynamic it is special
Sign, and introduce a new node on behalf on off state in output end and evaluate network, performance of switch is commented with this
Estimate, convenient for the control targe of real-time optimization switch.State feature extraction step receives the contact from Parameter identification link
Device main body characteristic parameter inputs, and Parameter identification can determine the link of description system according to the input and output of contactor.
Finally, the control decision of contactor by control strategy calculate link output, it by real-time reception switch control targe and shape
State feature, a series of movements or a decision that acquisition switch will execute, output make its basis work as to switch kinetic-control system
The suction counter-force of preceding operating mode reasonable distribution switch kinetic-control system.
System Working Principle：
The control thinking of self study is contained in the four processes of working-flow, link is calculated with control strategy in Fig. 1
For indicate.Wherein, Self-learning Controller is made of neural network with each parameter closed loop controller is switched, and utilizes multilayer nerve net
Network realizes the depth closed-loop control of switch, each closed loop controller provides heuristic letter in the training process of neural network
Breath, training learning process have contained knowledge evolution, so that controller or control strategy is approached with arbitrary accuracy arbitrarily complicated non-
Linear function, and with study and adapt to unknown or uncertain system ability.The operation principle of links is such as in system
Under：
1, control targe identifies：Control targe is sometimes and indefinite or can not be straight in complicated either intelligence control system
It connects to obtain.But control targe of the intelligent switch in control is generally directly given in the form of specified rate mostly at present, Bu Nengsui
Load, the variation of working environment and use classes and change.And it is provided with control targe in this wisdom control system and identifies mould
Block, which makes the control targe of switch be not limited to an optimal solution, but constantly rolls the part sought under current working
Suboptimal solution so that the control targe of intelligent switch can take the uncertainty caused by model mismatch, time-varying, interference etc. into account, and
When refresh or Correction and Control Model of Target Recognition, new control targe is established always the base optimal in switch actual working state
On plinth.Control targe is set in real time according to current working condition is switched, and how to reach control targe as switch,
Then determined by the controller of system.
2, state feature extraction：The sensor assembly of the module being functionally similar in control system realizes switch electromagnetism
The multivariable feedback of system and contact system.Unlike, deep learning algorithm is utilized in the state feature extraction in this system
The characteristics of being very suitable for dimensionality reduction and feature extraction is formd more abstract by characteristic parameters such as the low layer closed loop electricity of switch, magnetic
High-level characteristic indicate, such as contact motion track information, mechanism kinematic velocity information, arc energy power information.As a result, just
In the state space concept being introduced into modern control theory, no longer evade the high characteristic parameter of dimension, is conducive to wisdom control system
It is analysis integrated with control.The higher layer conditions feature of switch is extracted by data model, and one is being introduced to the output end of control targe
A new node on behalf on off state evaluates network, in this way, state characteristic extracting module express simultaneously switch behavioral characteristics with
And implement the action effect switched after control strategy assessment, this is convenient for carrying out real-time optimization to the control targe of switch.
3, Parameter identification：The module includes the execution module in wisdom control system, characterizes the dynamic control of switch
Process processed, it needs to control the input and output of signal according to electromagnetic switch to determine the model of description system, and neural network has
The ability for having fitting complex nonlinear function, can be used for System Discrimination, solve complicated non-linear in switch motion process
Dynamical system is difficult to linear function or the problem established according to priori.
4, control strategy calculates：The module is the controller of entire wisdom control system, is different from traditional single closed loop
Control strategy of its output of control structure is only single controlled quentity controlled variable, and the control strategy for combining self-learning algorithm is calculated in addition to defeated
Go out except single controlled quentity controlled variable, also include a series of movements or a decision that switch will execute, this improves switch severe
Fault-tolerant ability under environment operation.It is calculated different from existing control strategy and needs to obtain the mistake between specified rate and system output
Difference, Self-learning Controller are to carry out control strategy calculating by obtaining system mode.It is dynamic according to current working reasonable distribution switch
The suction counter-force of state control system, realizes the multivariable coordinated control of switch, to meet switch in different running environment, different operating
The stability control requirement of pattern.
In the early stage of whole system exploitation, the desired input of Neural Network Self-learning controller or output are typically
It is not available, and closed loop controller connected in parallel can export stable control signal whithin a period of time, this is to self-study
It practises controller and didactic education resource is provided.Whole process and the process of operator training are very similar, start he with from
The mode of line receives the introduction of general knowledge and basic operation, then it is expected that he can be achieved a solution challenge with online mode
Technical ability.
The above are preferred embodiments of the present invention, all any changes made according to the technical solution of the present invention, and generated function is made
When with range without departing from technical solution of the present invention, all belong to the scope of protection of the present invention.
Claims (4)
1. a kind of electromagnetic switch wisdom control system with Self-learning control pattern, it is characterised in that：Including：
Control targe identification module, the control targe identification module provide stage for the intelligent process control of electromagnetic switch
Control targe；
It is inputted connects with the output of state characteristic extracting module；
Parameter identification module, the Parameter identification module include switch kinetic-control system and contactor main body characteristic
Parameter observation model；
State characteristic extracting module, the state characteristic extracting module receive the contactor ontology from Parameter identification module
The current magnetic state information of switch of characteristic parameter observation model output；The electromagnetism of state characteristic extracting module real-time characterization contactor
The behavioral characteristics of system and contact system；
Control strategy computing module, the control strategy computing module export control decision, and control strategy computing module passes through reality
When receive the control targe and behavioral characteristics of switch, obtain a series of movements that will execute of switch or a decision, output is to opening
Kinetic-control system is closed, switch kinetic-control system is made to distribute the suction counter-force of switch kinetic-control system according to current working.
2. the electromagnetic switch wisdom control system according to claim 1 with Self-learning control pattern, it is characterised in that：
The control targe identification module includes control targe prediction model；Control targe prediction model constantly rolls and seeks current working
Under local suboptimal solution, refresh in time or Correction and Control Model of Target Recognition, new control targe established always real in switch
On the optimal basis of border working condition.
3. the electromagnetic switch wisdom control system according to claim 1 with Self-learning control pattern, it is characterised in that：
The state characteristic extracting module realizes the multivariable feedback of switch electromagnetic system and contact system；The state feature extraction mould
Block includes switching trace data model, switch movement Arc Modelling and evaluation network；The evaluation network is to the switching trace
The switch current arc energy letter that the switch current track information of data model output, the switch movement Arc Modelling export
Breath and the current magnetic state information of switch of contactor main body characteristic parameter observation model output carry out state evaluation；It is described simultaneously to open
Close the switch current track information of track data model output, the switch current arc of the switch movement Arc Modelling output
Energy information and the current magnetic state information of switch of contactor main body characteristic parameter observation model output are calculated as control strategy
The state input quantity of module.
4. the electromagnetic switch wisdom control system according to claim 1 with Self-learning control pattern, it is characterised in that：
The control strategy computing module includes autonomous learning device；The Self-learning Controller is by neural network and each parameter closed loop of switch
Controller forms, and the depth closed-loop control of switch is realized using multilayer neural network；Switching each parameter closed loop controller includes
Closed-loop current control device, voltage close loop controller and magnetic linkage closed-loop controller.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810253280.5A CN108319149B (en) | 2018-03-26 | 2018-03-26 | Electromagnetic switch intelligent control system with self-learning control mode |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810253280.5A CN108319149B (en) | 2018-03-26 | 2018-03-26 | Electromagnetic switch intelligent control system with self-learning control mode |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108319149A true CN108319149A (en) | 2018-07-24 |
| CN108319149B CN108319149B (en) | 2020-11-27 |
Family
ID=62899282
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810253280.5A Active CN108319149B (en) | 2018-03-26 | 2018-03-26 | Electromagnetic switch intelligent control system with self-learning control mode |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108319149B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110932399A (en) * | 2019-11-28 | 2020-03-27 | 上海宏力达信息技术股份有限公司 | Intelligent switch regulation and control method and system for power grid, storage medium and terminal |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5262960A (en) * | 1991-04-04 | 1993-11-16 | Sundstrand Corporation | Expert electrical power simulator |
| CN102543578A (en) * | 2012-02-07 | 2012-07-04 | 福州大学 | Real-time control intelligent alternating-current contactor with full-closed loops |
| CN104636554A (en) * | 2015-02-06 | 2015-05-20 | 福州大学 | Novel design method of intelligent alternating-current and direct-current electromagnetic apparatus |
| CN106229180A (en) * | 2016-08-20 | 2016-12-14 | 福州大学 | A kind of Switch without electric arc adaptive control system and control method thereof |
| CN106503330A (en) * | 2016-10-20 | 2017-03-15 | 国网冀北电力有限公司电力科学研究院 | On-load switch electromagnetic mechanism approximate calculation method and device |
| CN106951664A (en) * | 2017-04-26 | 2017-07-14 | 福州大学 | The automatic modeling and hardware-in-loop simulation method of a kind of contactor |
| CN106952782A (en) * | 2017-04-19 | 2017-07-14 | 福州大学 | Contactor velocity close-loop control method based on neutral net |
-
2018
- 2018-03-26 CN CN201810253280.5A patent/CN108319149B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5262960A (en) * | 1991-04-04 | 1993-11-16 | Sundstrand Corporation | Expert electrical power simulator |
| CN102543578A (en) * | 2012-02-07 | 2012-07-04 | 福州大学 | Real-time control intelligent alternating-current contactor with full-closed loops |
| CN104636554A (en) * | 2015-02-06 | 2015-05-20 | 福州大学 | Novel design method of intelligent alternating-current and direct-current electromagnetic apparatus |
| CN106229180A (en) * | 2016-08-20 | 2016-12-14 | 福州大学 | A kind of Switch without electric arc adaptive control system and control method thereof |
| CN106503330A (en) * | 2016-10-20 | 2017-03-15 | 国网冀北电力有限公司电力科学研究院 | On-load switch electromagnetic mechanism approximate calculation method and device |
| CN106952782A (en) * | 2017-04-19 | 2017-07-14 | 福州大学 | Contactor velocity close-loop control method based on neutral net |
| CN106951664A (en) * | 2017-04-26 | 2017-07-14 | 福州大学 | The automatic modeling and hardware-in-loop simulation method of a kind of contactor |
Non-Patent Citations (7)
| Title |
|---|
| C.T.CHI: "A Study of Closing Adaptive Control in Electronically Controlled Intelligent Contactor", 《THE 27TH INTERNATIONAL CONFERENCE ON ELECTRICAL CONTACTS》 * |
| TANG LONGFEI,ECT.: "Contactor Modeling Technology Based on an Artificial Neural Network", 《IEEE TRANSACTIONS ON MAGNETICS》 * |
| ZHIHONGXU,ECT.: "Digital Closed Loop Control Technology for the AC Contactors", 《THE 27TH INTERNATIONAL CONFERENCE ON ELECTRICAL CONTACTS》 * |
| 刘帼巾等: "神经网络法优化接触器电磁系统", 《电工技术学报》 * |
| 周志凯等: "基于虚拟样机技术的大功率直流接触器动态特性的仿真研究", 《低压电器》 * |
| 庄杰榕: "交流接触器集成化智能控制及仿真", 《第五届电工产品可靠性与电接触国际会议》 * |
| 许志红: "交流接触器智能化控制与设计技术的研究及实现", 《中国优秀博硕士学位论文全文数据库 (博士) 工程科技Ⅱ辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110932399A (en) * | 2019-11-28 | 2020-03-27 | 上海宏力达信息技术股份有限公司 | Intelligent switch regulation and control method and system for power grid, storage medium and terminal |
| CN110932399B (en) * | 2019-11-28 | 2022-02-25 | 上海宏力达信息技术股份有限公司 | Intelligent switch regulation and control method and system for power grid, storage medium and terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108319149B (en) | 2020-11-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xi et al. | Smart generation control based on multi-agent reinforcement learning with the idea of the time tunnel | |
| CN110535146A (en) | The Method for Reactive Power Optimization in Power of Policy-Gradient Reinforcement Learning is determined based on depth | |
| CN109946975A (en) | A kind of intensified learning optimal track control method of unknown servo-system | |
| CN107798199B (en) | Hydroelectric generating set parameter closed-loop identification method | |
| CN114217524B (en) | Power grid real-time self-adaptive decision-making method based on deep reinforcement learning | |
| CN110048606A (en) | DC-DC boost converter Dynamic sliding mode voltage control method based on two type Adaptive Fuzzy Neural-network of section | |
| CN109254530A (en) | MFA control method based on grinding process basis circuit | |
| CN110266228A (en) | Surface permanent magnetic Synchronous Machine Models forecast Control Algorithm based on BP neural network | |
| CN106532691A (en) | Adaptive dynamic programming-based frequency compound control method of single-region power system | |
| CN104852639A (en) | Parameter self-tuning speed controller of permanent magnet synchronous motor based on neural network | |
| CN110535396A (en) | Surface permanent magnetic Synchronous Machine Models forecast Control Algorithm based on BP neural network | |
| CN113970667A (en) | Non-invasive load monitoring method based on midpoint value of prediction window | |
| CN108319149A (en) | A kind of electromagnetic switch wisdom control system with Self-learning control pattern | |
| CN110518847A (en) | Surface permanent magnetic Synchronous Machine Models forecast Control Algorithm based on BP neural network | |
| CN110109356A (en) | The learning-oriented optimized control method and system of process in zinc electrolyzing model-free adaption | |
| CN110518860A (en) | Permanent magnet synchronous motor model predictive control method based on BP neural network and switch list | |
| Cai et al. | Design and simulation of intelligent PID controller based on particle swarm optimization | |
| Mufa’ary et al. | Comparison of FLC and ANFIS Methods to Keep Constant Power Based on Zeta Converter | |
| CN110888323A (en) | Control method for intelligent optimization of switching system | |
| Ma et al. | Research on fault location in DC distribution network based on adaptive artificial bee colony slime mould algorithm | |
| CN105353614A (en) | Control method for speed-variable frequency-constant double-feed bidirectional tidal power generator system | |
| Zhao et al. | Particle swarn optimized adaptive dynamic programming | |
| Liu et al. | NUMERICAL DYNAMIC MODELING AND DATA DRIVEN CONTROL VIA LEAST SQUARE TECHNIQUES AND HEBBIAN LEARNING ALGORITHM. | |
| Wang et al. | A fast converged voltage control method based on deep reinforcement learning | |
| Kamarposhti et al. | The Control of Buck Boost DC-DC Converters for DC Motor Drives on variable DC Voltage by Using Neural Network |
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 |