CN109521669A - A kind of turning table control methods of self-tuning based on intensified learning - Google Patents
A kind of turning table control methods of self-tuning based on intensified learning Download PDFInfo
- Publication number
- CN109521669A CN109521669A CN201811343796.5A CN201811343796A CN109521669A CN 109521669 A CN109521669 A CN 109521669A CN 201811343796 A CN201811343796 A CN 201811343796A CN 109521669 A CN109521669 A CN 109521669A
- Authority
- CN
- China
- Prior art keywords
- control parameter
- turntable
- intensified learning
- turning table
- rotating platform
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000012549 training Methods 0.000 claims abstract description 34
- 238000013461 design Methods 0.000 claims description 18
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 241000208340 Araliaceae Species 0.000 claims description 4
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims description 4
- 235000003140 Panax quinquefolius Nutrition 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 4
- 235000008434 ginseng Nutrition 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 6
- 238000004519 manufacturing process Methods 0.000 abstract description 4
- 238000004088 simulation Methods 0.000 abstract description 4
- 230000002411 adverse Effects 0.000 abstract description 3
- 230000006399 behavior Effects 0.000 abstract 1
- 239000013643 reference control Substances 0.000 abstract 1
- 230000026676 system process Effects 0.000 abstract 1
- 238000004458 analytical method Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 238000011031 large-scale manufacturing process Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000005094 computer simulation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000005086 pumping Methods 0.000 description 1
- 238000012360 testing method Methods 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
- G05B11/00—Automatic controllers
- G05B11/01—Automatic controllers electric
- G05B11/36—Automatic controllers electric with provision for obtaining particular characteristics, e.g. proportional, integral, differential
- G05B11/42—Automatic controllers electric with provision for obtaining particular characteristics, e.g. proportional, integral, differential for obtaining a characteristic which is both proportional and time-dependent, e.g. P. I., P. I. D.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
Abstract
The present invention is a kind of turning table control methods of self-tuning and its system based on intensified learning, solves the problems, such as that traditionally turning table control parameter needs professional to adjust for a long time.The described method includes: establishing intensified learning loop model by the behavior aggregate and Reward Program of formulating Self-tuning System process.Meanwhile invalid frequency of training is reduced as initial value using reference control parameter.Analogue simulation training is combined with true environment fine tuning, and then while can be quickly obtained trained intensified learning network, avoids mathematical modeling error bring adverse effect.This method has the advantage that can automatically adjust turntable control parameter, effectively increases the production efficiency of turntable.
Description
Technical field
The present invention is a kind of turning table control methods of self-tuning based on intensified learning, belongs to field of intelligent control.
Background technique
Turntable, also referred to as inertial navigation testing experiment equipment, it can accurately simulate its building ring for inertia device and system
The method of operation under border, to provide accurate full and accurate calibration foundation for the development of inertia device and system, production and maintenance.For
Guarantee turntable even running, the calculation amount of turning table control algorithm should be as small as possible, to meet the needs of high real-time.
Turntable mainly uses pid algorithm to carry out SERVO CONTROL at this stage.This method is simple due to calculating, and is convenient for engineering
Change and uses.Since the limitation of pid algorithm itself often requires to use more to guarantee that turntable can reach higher precision
Ring control and feedforward control.The complexity for increasing control structure brings the numerous adverse effect of control parameter, simultaneously because not
With differing greatly for performance indicator between turntable and structure, thus experienced professional is needed to carry out prolonged control ginseng
Number is adjusted, and can guarantee the reliable and stable of turning table control.The process to repeat will cause a large amount of human cost timely
Between cost consumption, thus be unfavorable for turntable large-scale production demand.
Summary of the invention
The present invention is exactly directed to above in the prior art insufficient and designs and provide a kind of turntable based on intensified learning
Control parameter automatic setting method needs professional to adjust control parameter for a long time when the purpose is to solve the production of traditional turntable
The problem of, production efficiency is effectively increased, the progress of turntable large-scale production is conducive to.
The purpose of the present invention is what is be achieved through the following technical solutions:
The method of turning table control parameter self-tuning of this kind based on intensified learning, it is characterised in that: the step of this method such as
Under:
Step 1: being missed using the speed precision of turntable, position precision design objective as overshoot, the stable state in pid algorithm
Difference, rise time index, according to pid algorithm calculation formula calculate overshoot, steady-state error, Kp corresponding to the rise time,
Tri- parameters of Ki, Kd, and these three parameters are known as rotating platform control system with reference to control parameter by Kp, Ki, Kd;
Step 2: mount model is chosen, by the design objective and rotating platform control system of turntable under simulated computer environment
It is input in rotating platform control system structural model with reference to control parameter;
Step 3: choosing operating mode under simulated computer environment, the operating mode is intensified learning network training
Mode or turning table control parameter training mode, intensified learning network training mode are used to execute intensified learning on network and evaluation net
Network carries out simulated training, to obtain the execution network and evaluation network that can satisfy under different use conditions;Turning table control ginseng
Number training mode is used to carry out simulated training to turning table control parameter, to obtain the control ginseng that can satisfy turntable performance indicator
Number, in which:
If choosing intensified learning network training mode, firstly, variation range and frequency of training to turret design index into
Row setting, variation range are ± the 10% of design objective, and the frequency of training in the variation range is selected as 50~100 times, so
Afterwards, rotating platform control system is adjusted by intensified learning mode with reference to control parameter, obtains the practical control of rotating platform control system
Parameter processed;
If choosing turning table control parameter training mode, firstly, the design objective of directly value turntable, frequency of training are selected as
400~500 times, then, rotating platform control system is adjusted by intensified learning mode with reference to control parameter, obtains turntable control
The practical control parameter of system processed;
Step 4: the practical control parameter of rotating platform control system is input in turntable pid control algorithm, then to turntable reality
Border operating condition is detected, and is finely adjusted to the practical control parameter of rotating platform control system, until reaching technical indicator.
Further, under the intensified learning network training mode described in step 2, variation range be design objective ±
5%.
The invention has the benefit that
1. the method for the present invention is on the basis of the control of the traditional servo of rotating platform control system structural model, in conjunction with intensified learning
Self-tuning System circuit, realizes the adjusting to turning table control parameter, and this method can be avoided the problem of needing personnel to adjust for a long time, have
Effect improves the efficiency of turntable debugging process, is conducive to the large-scale production of turntable;
2. executing network and evaluation network using analog simulation mode training intensified learning, a large amount of can be quickly obtained
Sample is practised, training network bring training time length in practical operation is avoided, easily causes the problems such as equipment damage.
3. using with reference to control parameter as initial value, reduce invalid frequency of training.
4. being avoided by the way of being finely adjusted turning table control parameter trained under simulated environment on practical turntable
Mathematical modeling error bring adverse effect, to ensure that the satisfaction that the performance indicator of practical turntable requires.
Detailed description of the invention
Fig. 1 is the system structure diagram of turning table control methods of self-tuning of the present invention
Fig. 2 is the schematic diagram of turning table control methods of self-tuning of the present invention
Fig. 3 is the program flow diagram of turning table control methods of self-tuning of the present invention.
Specific embodiment
Technical solution of the present invention is further described below with reference to drawings and examples:
Referring to figure 1, the system for realizing the method for the present invention includes turntable, fft analysis instrument and PC machine.The turntable packet
Motion-control module and turntable communication interface are included, the motion-control module includes driver and motor, for driving turntable frame
Body is moved;The turntable communication interface includes RS232, RS422, RS485 serial line interface, CAN interface, network interface interface
Equal external interfaces, for carrying out the transmission of turntable data and instruction between turntable and PC machine.The fft analysis instrument turns for analyzing
The frequency domain characteristic of bench control system, analysis result can be transmitted to PC machine by USB interface or network interface.The fft analysis instrument
The pumping signal of generation is transmitted to PC machine by output interface, for the desired locations of turntable movement to be arranged;Input interface is used for
The analog signal that reception turntable is input to driver can analyze turntable through input compared with signal between output interface
The performance of control system.PC machine includes PC machine communication interface, control parameter Self-tuning System software and signal conversion module.The PC machine
Communication interface includes serial line interface, network interface interface and USB interface, between PC machine and turntable and fft analysis instrument data and
The transmission of instruction, wherein the matching with turntable communication interface can be realized by the corresponding interface converter;The control parameter is certainly
Software is adjusted for realizing turning table control methods of self-tuning of the present invention;The signal conversion module includes D/A module
And A/D module, wherein A/D module is used to convert the number that PC machine can identify for the desired locations of fft analysis instrument output interface
Signal, D/A module are used to the calculated motor control signal of control algolithm being input to driver.
Referring to shown in attached drawing 2~3, turning table control methods of self-tuning of the present invention is carried out using above system
Steps are as follows:
Step 1: being missed using the speed precision of turntable, position precision design objective as overshoot, the stable state in pid algorithm
Difference, rise time index, according to pid algorithm calculation formula calculate overshoot, steady-state error, Kp corresponding to the rise time,
Tri- parameters of Ki, Kd, and these three parameters are known as rotating platform control system with reference to control parameter by Kp, Ki, Kd;
Step 2: mount model is chosen, by the design objective and rotating platform control system of turntable under simulated computer environment
It is input in rotating platform control system structural model with reference to control parameter;
The simulated computer environment is V-REP emulation platform, and the platform is by calling physical engine, to realize to power
The simulation of the physical quantitys such as square, frictional force.The mount model includes three axis, twin shaft and single axle table model, passes through unified machine
People's descriptor format (URDF) is designed.
After starting emulation platform, mount model is placed in simulating scenes, is turned by quality, the maximum to each component of turntable
Speed is configured, and achievees the purpose that the true turntable running environment of simulation.
When emulation, using rotating platform control system with reference to control parameter as the control parameter in rotating platform control system structural model
Initial value;
Step 3: choosing operating mode under simulated computer environment, the operating mode is intensified learning network training
Mode or turning table control parameter training mode, in which:
If choosing intensified learning network training mode, firstly, variation range and frequency of training to turret design index into
Row setting, variation range are ± the 10% of design objective, and the frequency of training in the variation range is selected as 50~100 times, so
Afterwards, rotating platform control system is adjusted by intensified learning mode with reference to control parameter, obtains the practical control of rotating platform control system
Parameter processed;
If choosing turning table control parameter training mode, firstly, the design objective of directly value turntable, frequency of training are selected as
400~500 times, then, rotating platform control system is adjusted by intensified learning mode with reference to control parameter, obtains turntable control
The practical control parameter of system processed;
The simulated computer environment includes intensified learning Self-tuning System circuit, and then realizes the tune to turning table control parameter
Section.The effect in intensified learning Self-tuning System circuit is that the effect generated to existing turning table control parameter is evaluated, and then basis turns
Platform control result determines the trend that control parameter is adjusted, and then obtains optimal control parameter.Step 4: by rotating platform control system reality
Border control parameter is input in turntable pid control algorithm, it is contemplated that by the mathematical model that uses when computer simulation training with
True environment has a certain difference, therefore trained turning table control parameter needs certain amendment that can guarantee really to turn
Platform performance indicator is qualified, so, it next needs to detect turntable practical operation situation, and to rotating platform control system reality
Control parameter is finely adjusted, until reaching technical indicator.
Claims (2)
1. a kind of method of the turning table control parameter self-tuning based on intensified learning, it is characterised in that: the step of this method is as follows:
Step 1: using the speed precision of turntable, position precision design objective as in pid algorithm overshoot, steady-state error, on
Time index is risen, overshoot, steady-state error, Kp, Ki, Kd corresponding to the rise time are calculated according to pid algorithm calculation formula
Three parameters, and these three parameters are known as rotating platform control system with reference to control parameter by Kp, Ki, Kd;
Step 2: choosing mount model under simulated computer environment, the design objective of turntable and rotating platform control system are referred to
Control parameter is input in rotating platform control system structural model;
Step 3: choosing operating mode under simulated computer environment, the operating mode is intensified learning network training mode
Or turning table control parameter training mode, in which:
If intensified learning network training mode is chosen, firstly, the variation range and frequency of training to turret design index are set
It sets, variation range is ± the 10% of design objective, and the frequency of training in the variation range is selected as 50~100 times, then, leads to
It crosses intensified learning mode and rotating platform control system is adjusted with reference to control parameter, obtain the practical control ginseng of rotating platform control system
Number;
If choosing turning table control parameter training mode, firstly, the design objective of directly value turntable, frequency of training are selected as 400
~500 times, then, rotating platform control system is adjusted by intensified learning mode with reference to control parameter, obtains turning table control
The practical control parameter of system;
Step 4: the practical control parameter of rotating platform control system is input in turntable pid control algorithm, then to the practical fortune of turntable
Market condition is detected, and is finely adjusted to the practical control parameter of rotating platform control system, until reaching technical indicator.
2. the method for the turning table control parameter self-tuning according to claim 1 based on intensified learning, it is characterised in that:
Under intensified learning network training mode described in step 2, variation range is ± the 5% of design objective.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811343796.5A CN109521669A (en) | 2018-11-12 | 2018-11-12 | A kind of turning table control methods of self-tuning based on intensified learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811343796.5A CN109521669A (en) | 2018-11-12 | 2018-11-12 | A kind of turning table control methods of self-tuning based on intensified learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109521669A true CN109521669A (en) | 2019-03-26 |
Family
ID=65776429
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811343796.5A Pending CN109521669A (en) | 2018-11-12 | 2018-11-12 | A kind of turning table control methods of self-tuning based on intensified learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109521669A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112631120A (en) * | 2019-10-09 | 2021-04-09 | Oppo广东移动通信有限公司 | PID control method, device and video coding and decoding system |
CN114371648A (en) * | 2021-12-27 | 2022-04-19 | 九江冠成仿真技术有限公司 | Core board of high-precision turntable control system |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8189608B2 (en) * | 2009-12-31 | 2012-05-29 | Sonicwall, Inc. | Wireless extender secure discovery and provisioning |
CN103853046A (en) * | 2014-02-14 | 2014-06-11 | 广东工业大学 | Adaptive learning control method of piezoelectric ceramics driver |
CN105763272A (en) * | 2016-05-09 | 2016-07-13 | 中国科学院云南天文台 | Self-calibration radio environment testing platform and testing method thereof |
CN106208867A (en) * | 2016-07-16 | 2016-12-07 | 沈阳航空航天大学 | A kind of control system for permanent-magnet synchronous motor pid parameter optimization method |
CN107346138A (en) * | 2017-06-16 | 2017-11-14 | 武汉理工大学 | A kind of unmanned boat method for lateral control based on enhancing learning algorithm |
CN108008627A (en) * | 2017-12-13 | 2018-05-08 | 中国石油大学(华东) | A kind of reinforcement learning adaptive PID control method of parallel optimization |
-
2018
- 2018-11-12 CN CN201811343796.5A patent/CN109521669A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8189608B2 (en) * | 2009-12-31 | 2012-05-29 | Sonicwall, Inc. | Wireless extender secure discovery and provisioning |
CN103853046A (en) * | 2014-02-14 | 2014-06-11 | 广东工业大学 | Adaptive learning control method of piezoelectric ceramics driver |
CN105763272A (en) * | 2016-05-09 | 2016-07-13 | 中国科学院云南天文台 | Self-calibration radio environment testing platform and testing method thereof |
CN106208867A (en) * | 2016-07-16 | 2016-12-07 | 沈阳航空航天大学 | A kind of control system for permanent-magnet synchronous motor pid parameter optimization method |
CN107346138A (en) * | 2017-06-16 | 2017-11-14 | 武汉理工大学 | A kind of unmanned boat method for lateral control based on enhancing learning algorithm |
CN108008627A (en) * | 2017-12-13 | 2018-05-08 | 中国石油大学(华东) | A kind of reinforcement learning adaptive PID control method of parallel optimization |
Non-Patent Citations (3)
Title |
---|
刘红伟: "永磁同步电机控制策略及算法融合研究", 《中国博士学位论文全文数据库 工程科技Ⅱ辑》 * |
李学斌等: "船舶六自由度转台电液伺服系统参数整定方法", 《船海工程》 * |
高瑞娟: "基于改进强化学习的PID参数整定原理及应用", 《现代电子技术》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112631120A (en) * | 2019-10-09 | 2021-04-09 | Oppo广东移动通信有限公司 | PID control method, device and video coding and decoding system |
CN112631120B (en) * | 2019-10-09 | 2022-05-17 | Oppo广东移动通信有限公司 | PID control method, device and video coding and decoding system |
CN114371648A (en) * | 2021-12-27 | 2022-04-19 | 九江冠成仿真技术有限公司 | Core board of high-precision turntable control system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109521669A (en) | A kind of turning table control methods of self-tuning based on intensified learning | |
CN103197596B (en) | A kind of digital control processing parameters self-adaptive fuzzy control rule optimization method | |
CN104122089B (en) | Turboshaft engine speed torque signal simulation device and simulation method | |
CN110007645B (en) | Feeding system hybrid modeling method based on dynamics and deep neural network | |
CN104199283A (en) | Test system and control method for electro-hydraulic servo online self-adjusting fuzzy PID control | |
CN104898647A (en) | Automatic calibration simulation testing system for ECU stand | |
CN110045654B (en) | Rotary table state space control system and method based on RT operating system and FPGA | |
CN108087539A (en) | A kind of transmission model calibration system and its calibration method based on hardware-in-loop simulation | |
CN111396248A (en) | Wind turbine generator set intelligent yaw control method based on short-term wind direction prediction | |
CN102509152A (en) | Switched reluctance motor on-line modeling method based RBF neural network | |
CN111290373B (en) | Fan controller detection system and detection method | |
CN106154838A (en) | The residual oscillation suppressing method of a kind of positioner and system | |
CN107972036B (en) | Industrial robot dynamics control system and method based on TensorFlow | |
CN106898229A (en) | A kind of automation-control experiment platform | |
CN107608231A (en) | A kind of adjustment method of target seeker stabilized platform | |
CN116090552A (en) | Training and reasoning performance test method for artificial intelligent accelerator card product | |
CN102830629A (en) | Testing device and method of power electronic control system | |
CN105319971B (en) | GMA self-adaptation control methods based on fiber grating and device | |
CN107728481B (en) | Closed-loop modeling method and device based on model predictive control | |
CN110134082A (en) | The Virtual Debugging method of auto production line based on digital factory | |
CN201374136Y (en) | Multifunctional teaching device | |
CN104571086B (en) | Temperature controller emulation test method based on transmission function | |
CN109272852B (en) | Content-extensible linear position control teaching experiment system and method | |
CN116562156A (en) | Training method, device, equipment and storage medium for control decision model | |
CN210402906U (en) | Production line letter sorting system based on Niagara development platform |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190326 |
|
RJ01 | Rejection of invention patent application after publication |