CN108958020A - A kind of temprature control method based on RBF neural dynamic prediction PID - Google Patents
A kind of temprature control method based on RBF neural dynamic prediction PID Download PDFInfo
- Publication number
- CN108958020A CN108958020A CN201810530150.1A CN201810530150A CN108958020A CN 108958020 A CN108958020 A CN 108958020A CN 201810530150 A CN201810530150 A CN 201810530150A CN 108958020 A CN108958020 A CN 108958020A
- Authority
- CN
- China
- Prior art keywords
- temperature control
- output
- pid controller
- control system
- temperature
- 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
- 230000001537 neural effect Effects 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 title claims abstract description 16
- 238000005070 sampling Methods 0.000 claims abstract description 10
- 230000000694 effects Effects 0.000 claims abstract description 8
- 238000012887 quadratic function Methods 0.000 claims description 11
- 230000006870 function Effects 0.000 claims description 7
- 230000007613 environmental effect Effects 0.000 claims description 5
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 238000010615 ring circuit Methods 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 4
- 238000010168 coupling process Methods 0.000 description 4
- 238000005859 coupling reaction Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 238000012369 In process control Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000005530 etching Methods 0.000 description 2
- 238000010965 in-process control Methods 0.000 description 2
- 210000005036 nerve Anatomy 0.000 description 2
- 238000005096 rolling process Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 230000018199 S phase Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000001629 suppression 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
- 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)
- Control Of Temperature (AREA)
Abstract
The invention discloses a kind of temprature control methods based on RBF neural dynamic prediction PID.Temperature control system used by the temprature control method includes PID controller, predictive controller, RBF neural and control object model of the multiple groups work under identical operating condition, the specific steps are as follows: setting reference input, the input of PID controller and adjustment target temperature;Temperature control system output is described in the form of quadratic equation;Being arranged the sampling time is less than lag time, predicts the value of multiple future time instances on the basis of current time, after multiple repairing weld, takes 3~4 sampled values nearest apart from lag time, be weighted and averaged as predicted value;The nonlinear discrete model that temperature control system is indicated using formula is adaptively adjusted PID controller parameter by RBF neural, carries out temperature control.The principle of the invention is simple, control effect is good, can be improved temperature controlled precision and uniformity.
Description
Technical field
The invention belongs to automatic control technology field, especially a kind of temperature based on RBF neural dynamic prediction PID
Control method.
Background technique
With the rapid development of space flight and aviation, the demand to the climate environment simulation test equipment of aircraft is increasingly compeled
It cuts, to climate environment simulation test control system, higher requirements are also raised.And the control of climatic environmental changes system temperature
Precision is an important indicator ring of environmental control system.Large-scale weather border analog temperature control system is wanted for temperature uniformity
It asks and mostly uses multi-air duct structure, so that temperature control has the characteristics that big inertia, time lag, coupling.This kind of system is used
Traditional control algorithm is difficult to meet control accuracy requirement, also there is certain influence on system stability.
In Process Control System, PID control and cas PID control frequently become load using extremely wide
Change, the complex control system with coupled characteristic and hysteresis characteristic are difficult to reach ideal control effect, and it is low, warm to there is control precision
Spend the problem of uniformity difference.
Summary of the invention
The purpose of the present invention is to provide a kind of control precision height, temperature uniformity are good based on RBF neural dynamic
The temprature control method of predictive PID controls large-scale environment temperature field temperature to realize.
The technical solution for realizing the aim of the invention is as follows: a kind of temperature control based on RBF neural dynamic prediction PID
Method processed, used temperature control system include PID controller of the multiple groups work under identical operating condition, predictive controller, RBF
Neural network and control object model, the specific steps are as follows:
Step 1, setting reference input, the input of PID controller and adjustment target temperature;
Step 2 describes temperature control system output in the form of quadratic equation;
Step 3, setting sampling time are less than lag time, and multiple future time instances are predicted on the basis of current time
It is worth, after multiple repairing weld, takes 3~4 sampled values nearest apart from lag time, be weighted and averaged as predicted value;
Step 4, the nonlinear discrete model that temperature control system is indicated using formula, it is adaptive by RBF neural
PID controller parameter should be adjusted, temperature control is carried out.
Further, setting reference input, the input of PID controller described in step 1 and adjustment target temperature, specifically such as
Under:
Reference input is ri, the input of PID controller is e (k+T)=ri- y (k+T), wherein y (k+T) is predictive controller
Output, i=1,2,3 ... n indicate that every group #, e (k+T) are the prediction deviation value for indicating reference input and prediction output, and T is stagnant
Time afterwards, k indicate current time, and the parameter of PID controller adaptively adjusts output by RBF neural, PID controller
Output adjusts target temperature by actuator.
Further, system output is described in the form of quadratic equation described in step 2, specific as follows:
Current time is denoted as k, and current PID controller output is u (k)=Δ u (k)+u (k-T), when Δ u (k) indicates previous
The deviation of etching system output and the output of current time system;
Temperature control system output valve is y (k), then the output of previous moment PID controller is u (k-T), temperature control system
Output is y (k-T), and the output of subsequent time PID controller is u (k+T), and temperature control system output is y (k+T), and so on;
Using Quadratic function temperature control system output valve, quadratic function is that temperature control system was exported about the time
The function of t, as follows:
Wherein, a, b, c are respectively quadratic coefficients, coefficient of first order, the constant term of quadratic function;
It is obtained by above-mentioned equation:
A=(y (k)-y (k-T))/2T
From above formula it is found that y (k+T) and e (k+T) passes through kth moment and kth-T moment output valve and output valve
Change rate is calculated.
Further, the setting sampling time described in step 3 is less than lag time, specifically: setting sampling time t is less than stagnant
Time T afterwards, and t < T < 8t.
Further, the nonlinear discrete model for indicating temperature control system described in step 4 using formula, passes through RBF
Neural network adaptively adjusts PID controller parameter, carries out temperature control, specific as follows:
For the temperature control system of multiple-input and multiple-output, due to the intrinsic coupled characteristic in temperature field, per temperature control all the way
It channel processed all can be by other control channel temperature profile effects, other control channels to the temperature of the road temperature control channel
External disturbance is regarded in coupling as, 5 road temperature control channels is equipped with, under the nonlinear discrete model of temperature control system passes through
Formula indicates:
yi(k)=f [u (k-1) ..., u (k-T), yi(k-1),…,yi(k-T)]
U (k)=[u1(k),u2(k),…,u5(k)]
Y (k)=[y1(k),…,y5(k)]
In formula, u (k), y (k) respectively indicate PID controller output and temperature control system output, f []=f1(·)×
f2(·)…fn() is the nonlinear dependence described between the current return air temperature in the i-th air duct of environmental chamber and control amount input and disturbance
System, T indicate lag time;
PID controller parameter is adaptively adjusted by RBF neural, the input of the i-th tunnel RBF neural is u1(k),u2
(k),…,u5(k),yi(k-1), it exports as yi(k), wherein ui(k) it is exported as PID controller, passes through holding in inner ring circuit
Row device exports current supply air temperature, and actuator includes by-passing valve, mixed flow pipeline, surface air cooler;
Temperature control system error is expressed as:
ei(k)=ri(k)-yi(k)
The output of incremental timestamp device are as follows:
ui(k)=ui(k-1)+kpxe1+kixe2+kdxe3
Wherein:
Performance index function E (k) is indicated are as follows:
The proportionality coefficient kp of PID controlleri, differential coefficient kIi, integral coefficient kdiIt is obtained according to gradient descent method:
In above formula, η1, η2, η3Indicate learning rate.
Compared with prior art, the present invention its remarkable advantage are as follows: (1) use RBF neural Adaptive PID Control and move
The mode that state PREDICTIVE CONTROL PID is combined forms organic whole based on PID control structure, improves temperature control precision, mention simultaneously
High-temperature uniformity;(2) principle is simple, control effect is good, can be improved temperature control precision and uniformity.
Detailed description of the invention
Fig. 1 is the schematic diagram that dynamic prediction controls in the present invention.
Fig. 2 is the output curve diagram of control system in the present invention.
Fig. 3 is dynamic prediction control schematic diagram in the present invention.
Fig. 4 is the structure chart of neural network dynamic predictive PID Control system in the present invention.
Fig. 5 is RBF neural identification output and reality output curve comparison figure in the present invention.
Fig. 6 is RBF neural training error curve graph in the present invention.
Fig. 7 is 70 DEG C of hot test temperature profiles of neural network dynamic PREDICTIVE CONTROL in the present invention.
Specific embodiment
The present invention is recompensed for large time delay problem in temperature control using dynamic prediction Control PID algorithm, and for
Nonlinear coupling relationship between multiple-input and multiple-output is rung by the None-linear approximation ability and quick part of RBF neural
Should be able to force adaptive adjust PID controller parameter, mention high control precision and temperature uniformity.
Control system introduces predictive controller feedback output, while PID controller parameter on the basis of conventional PID controllers
By RBF neural self-adaptive sites, improves temperature control precision and uniformity, dynamic prediction PID structure are as shown in Figure 1.
In Fig. 1, r indicates that reference input, y (k+T) indicate under current control amount u effect that the prediction after lag time T is defeated
Out, e (k+T) indicates the prediction deviation value of reference input and prediction output.
As can be seen that dynamic prediction PID control is with traditional PI D the difference is that dynamic prediction PID control value of feedback is
Predicted value of the current return value after prediction lag time T, this is because delay system input action has to pass through system
Intrinsic lag time can just be embodied in output, can thus be converted into traditional PI D to the control system without delay component
The control of system is able to solve large time delay control problem common in process control.
The thought of dynamic prediction PID control is: in the section of very little, the output of controlled device is the song varied less
Line, can be in the form of quadratic equation come the output of approximate description system, and with time recursion, quadratic equation curve is constantly followed
The curve of output of variation.
The present invention is based on the temprature control method of RBF neural dynamic prediction PID, used temperature control system packet
PID controller, predictive controller, RBF neural and control object model of the multiple groups work under identical operating condition are included, specifically
Steps are as follows:
Step 1, setting reference input, the input of PID controller and adjustment target temperature;
Reference input is ri, the input of PID controller is e (k+T)=ri- y (k+T), wherein y (k+T) is predictive controller
Output, i=1,2,3 ... n indicate that every group #, e (k+T) are the prediction deviation value for indicating reference input and prediction output, and T is stagnant
Time afterwards, k indicate current time, and the parameter of PID controller adaptively adjusts output by RBF neural, PID controller
Output adjusts target temperature by actuator.
Step 2 describes temperature control system output in the form of quadratic equation;
It is illustrated in figure 2 control system curve of output, and the curve of output will be described with quadratic equation form.
Current time is denoted as k, and current PID controller output is u (k)=Δ u (k)+u (k-T), when Δ u (k) indicates previous
The deviation of etching system output and the output of current time system;
Temperature control system output valve is y (k), then the output of previous moment PID controller is u (k-T), temperature control system
Output is y (k-T), and the output of subsequent time PID controller is u (k+T), and temperature control system output is y (k+T), and so on;
Using Quadratic function temperature control system output valve, quadratic function is that temperature control system was exported about the time
The function of t, as follows:
Using Quadratic function temperature control system output valve, quadratic function is that temperature control system was exported about the time
The function of t, as follows:
Wherein, a, b, c are respectively quadratic coefficients, coefficient of first order, the constant term of quadratic function;
It is obtained by above-mentioned equation:
A=(y (k)-y (k-T))/2T
From above formula it is found that y (k+T) and e (k+T) passes through kth moment and kth-T moment output valve and output valve
Change rate is calculated, it can be seen that this method principle is simple, and calculation amount is small, has engineering application value.
Step 3, setting sampling time are less than lag time, and multiple future time instances are predicted on the basis of current time
It is worth, after multiple repairing weld, takes 3~4 sampled values nearest apart from lag time, be weighted and averaged as predicted value;
According to PREDICTIVE CONTROL rolling optimization thinking, sampling time t can be set less than lag time T, can generally set t < T <
8t.It can predict the value of multiple future time instances on the basis of current time k in this way, after multiple repairing weld, take k+T recently several times
Sampled value weighted average is used as predicted value, and so as time goes by, prediction time domain constantly moves forward, such dynamic prediction
As a result it is more nearly actual value, so that control action is established in effective predicted value, lag bring, which is effectively reduced, to be influenced.Rolling
Dynamic prediction schematic diagram is as shown in Figure 3:
In Fig. 3, k indicates current time, and single prediction step t is chosen as T=2t, predicts selection of times 6 times, then k+3t
The predicting lag time value at moment can use the weighted average of vertical line selection in figure, it may be assumed that
Y (k+3t+T)=α1y(k+5t)+α2y(k+t+4t)+α3y(k+2t+3t)+α4y(k+3t+T)
α in above formula1α2α3α4For constant coefficient, it is bigger that the nearlyr weighting coefficient of range prediction point distance may be selected.In this way by adopting
Sample can form prediction output matrix
Dynamic prediction PID controller adds dynamic prediction feedback loop on the basis of traditional PID control, in addition to conventional proportional
Coefficient, integral coefficient, differential coefficient first have to clear lag time T outside needing to adjust, and then also need to adjust Single-step Prediction step
Long t and prediction step number, it is however generally that, step-length t is selected bigger, then forecasting accuracy is higher, but calculation amount is also increase accordingly, usually
Lag time 1/8th may be selected to a quarter.
Step 4, the nonlinear discrete model that temperature control system is indicated using formula, it is adaptive by RBF neural
PID controller parameter should be adjusted, temperature control is carried out.
For the temperature control system of multiple-input and multiple-output, due to the intrinsic coupled characteristic in temperature field, per temperature control all the way
It channel processed all can be by other control channel temperature profile effects.We can be other channels to the temperature of the pipelines control structure
Coupling regards that external disturbance, such nonlinear discrete model can pass through RBF neural None-linear approximation as.If
There are 5 road temperature control channels, the nonlinear discrete model of temperature control system is indicated by following formula:
yi(k)=f [u (k-1) ..., u (k-T), yi(k-1),...,yi(k-T)]
U (k)=[u1(k),u2(k),...,u5(k)]
Y (k)=[y1(k),...,y5(k)]
In formula, u (k), y (k) respectively indicate PID controller output and temperature control system output, f []=f1(·)×
f2(·)…fn() is the nonlinear dependence described between the current return air temperature in the i-th air duct of environmental chamber and control amount input and disturbance
System, T indicate lag time;
PID controller parameter is adaptively adjusted by RBF neural, the input of the i-th tunnel RBF neural is u1(k),u2
(k),…,u5(k),yi(k-1), it exports as yi(k), wherein ui(k) it is exported as PID controller, passes through holding in inner ring circuit
Row device exports current supply air temperature, and actuator includes by-passing valve, mixed flow pipeline, surface air cooler;
Temperature control system error is expressed as:
ei(k)=ri(k)-yi(k)
The output of incremental timestamp device are as follows:
ui(k)=ui(k-1)+kpxe1+kixe2+kdxe3
Wherein:
Performance index function E (k) is indicated are as follows:
The proportionality coefficient kp of PID controlleri, differential coefficient kIi, integral coefficient kdiIt is obtained according to gradient descent method:
In above formula, η1, η2, η3Indicate learning rate.
Embodiment 1
Temprature control method based on RBF neural dynamic prediction PID in the present embodiment, based on certain 5 input, 5 output mould
Quasi- temperature control system, structure having the same for each channel, including the mutually isostructural RBF neural unit in five tunnels,
Therefore third I/O channel is selected to be introduced as representative herein.
In order to obtain RBF neural identified parameters, has chosen 600 groups of sampled datas and system is trained.RBF nerve
E-learning rate is 0.25, factor of momentum selection 0.05.Using typical three-layer network model, the number of hidden nodes is 11, instruction
Practice the corresponding output of data and deviation map difference is as shown in Figure 5 and Figure 6.
PID controller is serials control in dynamic prediction PID, is only controlled using dynamic prediction major loop control, inner ring
Road is used for disturbance suppression, improves response speed.Wherein major loop ratio, integral, differential are respectively 4,0.025,0, subloop ratio
Example, integral, differential inner ring are 15,0.1,0, prediction step 5, time step 10s.
It can be seen that by Fig. 5 and Fig. 6 through training, obtain nerve network system parameter, preferably pick out institute's phase
The overall situation room nonlinear model of prestige, while there is good generalization ability.It is adaptive by the RBF neural parameter of acquisition
Pid parameter should be adjusted, big room temperature simulation curve is as shown in Figure 7.
From heating curve, chamber temperature can be good at following set temperature in the temperature rise period.Stablizing rank
Section, 70.6 DEG C of maximum temperature, 69.5 DEG C of minimum temperature, temperature control precision is stablized at ± 0.6 DEG C, and temperature fluctuation range is smaller,
Precision is high, has the advantages that precision is high, uniformity is good.
Claims (5)
1. a kind of temprature control method based on RBF neural dynamic prediction PID, which is characterized in that used temperature control
System processed includes PID controller, predictive controller, RBF neural and control object of the multiple groups work under identical operating condition
Model, the specific steps are as follows:
Step 1, setting reference input, the input of PID controller and adjustment target temperature;
Step 2 describes temperature control system output in the form of quadratic equation;
Step 3, setting sampling time are less than lag time, and the value of multiple future time instances is predicted on the basis of current time, more
After secondary sampling, 3~4 sampled values nearest apart from lag time are taken, are weighted and averaged as predicted value;
Step 4, the nonlinear discrete model that temperature control system is indicated using formula, are adaptively adjusted by RBF neural
Whole PID controller parameter carries out temperature control.
2. the temprature control method according to claim 1 based on RBF neural dynamic prediction PID, which is characterized in that
Setting reference input, the input of PID controller described in step 1 and adjustment target temperature, specific as follows:
Reference input is ri, the input of PID controller is e (k+T)=ri- y (k+T), wherein y (k+T) is that predictive controller is defeated
Out, i=1,2,3 ... n indicate that every group #, e (k+T) are the prediction deviation value for indicating reference input and prediction output, and T is lag
Time, k indicate current time, and the parameter of PID controller adaptively adjusts output by RBF neural, PID controller it is defeated
Target temperature is adjusted by actuator out.
3. the temprature control method according to claim 1 based on RBF neural dynamic prediction PID, which is characterized in that
System output is described in the form of quadratic equation described in step 2, specific as follows:
Current time is denoted as k, and current PID controller output is u (k)=Δ u (k)+u (k-T), and Δ u (k) indicates previous moment system
The deviation of system output and the output of current time system;
Temperature control system output valve is y (k), then the output of previous moment PID controller is u (k-T), temperature control system output
For y (k-T), the output of subsequent time PID controller is u (k+T), and temperature control system output is y (k+T), and so on;
Using Quadratic function temperature control system output valve, quadratic function is that temperature control system is exported about time t's
Function, as follows:
Wherein, a, b, c are respectively quadratic coefficients, coefficient of first order, the constant term of quadratic function;
It is obtained by above-mentioned equation:
A=(y (k)-y (k-T))/2T
From above formula it is found that the variation that y (k+T) and e (k+T) passes through kth moment and kth-T moment output valve and output valve
Rate is calculated.
4. the temprature control method according to claim 1 based on RBF neural dynamic prediction PID, which is characterized in that
The setting sampling time described in step 3 is less than lag time, specifically: setting sampling time t is less than lag time T, and t < T < 8t.
5. the temprature control method according to claim 1 based on RBF neural dynamic prediction PID, which is characterized in that
The nonlinear discrete model for being indicated temperature control system described in step 4 using formula is adaptively adjusted by RBF neural
PID controller parameter carries out temperature control, specific as follows:
It is logical per the control of temperature all the way due to the intrinsic coupled characteristic in temperature field for the temperature control system of multiple-input and multiple-output
Road can all couple temperature of other control channels to the road temperature control channel by other control channel temperature profile effects
External disturbance is regarded in effect as, is equipped with 5 road temperature control channels, and the nonlinear discrete model of temperature control system passes through following formula table
Show:
yi(k)=f [u (k-1) ..., u (k-T), yi(k-1),...,yi(k-T)]
U (k)=[u1(k),u2(k),…,u5(k)]
Y (k)=[y1(k),…,y5(k)]
In formula, u (k), y (k) respectively indicate PID controller output and temperature control system output, f []=f1(·)×f2
(·)…fn() is the nonlinear dependence described between the current return air temperature in the i-th air duct of environmental chamber and control amount input and disturbance
System, T indicate lag time;
PID controller parameter is adaptively adjusted by RBF neural, the input of the i-th tunnel RBF neural is u1(k),u2
(k),…,u5(k),yi(k-1), it exports as yi(k), wherein ui(k) it is exported as PID controller, passes through holding in inner ring circuit
Row device exports current supply air temperature, and actuator includes by-passing valve, mixed flow pipeline, surface air cooler;
Temperature control system error is expressed as:
ei(k)=ri(k)-yi(k)
The output of incremental timestamp device are as follows:
ui(k)=ui(k-1)+kpxe1+kixe2+kdxe3
Wherein:
Performance index function E (k) is indicated are as follows:
The proportionality coefficient kp of PID controlleri, differential coefficient kIi, integral coefficient kdiIt is obtained according to gradient descent method:
In above formula, η1, η2, η3Indicate learning rate.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810530150.1A CN108958020A (en) | 2018-05-29 | 2018-05-29 | A kind of temprature control method based on RBF neural dynamic prediction PID |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810530150.1A CN108958020A (en) | 2018-05-29 | 2018-05-29 | A kind of temprature control method based on RBF neural dynamic prediction PID |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108958020A true CN108958020A (en) | 2018-12-07 |
Family
ID=64492328
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810530150.1A Pending CN108958020A (en) | 2018-05-29 | 2018-05-29 | A kind of temprature control method based on RBF neural dynamic prediction PID |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108958020A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109597449A (en) * | 2019-01-30 | 2019-04-09 | 杭州庆睿科技有限公司 | A kind of ultrasonic wave separating apparatus temprature control method neural network based |
CN110262582A (en) * | 2019-07-30 | 2019-09-20 | 中原工学院 | A kind of barotor temprature control method based on improvement RBF neural |
CN110554715A (en) * | 2019-10-25 | 2019-12-10 | 攀钢集团攀枝花钢铁研究院有限公司 | RBF neural network-based PID control method for hydrolysis process temperature of titanyl sulfate plus seed crystal |
CN111665716A (en) * | 2020-05-27 | 2020-09-15 | 广东工业大学 | Mathematical modeling method for extracting traditional Chinese medicine/natural plant |
CN112034701A (en) * | 2020-09-11 | 2020-12-04 | 长虹美菱股份有限公司 | PID control system and PID control algorithm based on temperature sampling |
CN112648717A (en) * | 2020-12-16 | 2021-04-13 | 桂林电子科技大学 | Terminal control method of multi-zone variable air volume air conditioning system based on radial basis function neural network |
CN113589862A (en) * | 2021-06-24 | 2021-11-02 | 苏州环美生物医疗科技有限公司 | Intelligent PID temperature control method and control chip based on network topology |
CN114167724A (en) * | 2021-11-30 | 2022-03-11 | 中车长春轨道客车股份有限公司 | Dual-controller weighted control method based on black box model |
CN114779626A (en) * | 2021-11-13 | 2022-07-22 | 黄河交通学院 | Networked prediction control method based on neural network and weighted average algorithm |
CN116627028A (en) * | 2023-07-21 | 2023-08-22 | 阳谷新太平洋电缆有限公司 | Control method for crosslinked cable production line |
CN117316356A (en) * | 2023-10-24 | 2023-12-29 | 中国民航大学 | Feedforward compensation regulation and control method for composite material component autoclave molding process parameters |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106528705A (en) * | 2016-10-26 | 2017-03-22 | 桂林电子科技大学 | Repeated record detection method and system based on RBF neural network |
CN107045289A (en) * | 2017-06-05 | 2017-08-15 | 杭州电子科技大学 | A kind of nonlinear neural network optimization PID control method of electric furnace temperature |
CN107544263A (en) * | 2017-10-31 | 2018-01-05 | 河北农业大学 | Liquid manure proportioning control system and method are adjusted based on parameter self-tuning PID |
CN107608215A (en) * | 2017-10-18 | 2018-01-19 | 齐鲁工业大学 | A kind of mapping UAV Attitude control method based on RBF neural |
CN107703754A (en) * | 2017-10-30 | 2018-02-16 | 王才旺 | A kind of simulating turntable control system based on Neural Network PID Control |
-
2018
- 2018-05-29 CN CN201810530150.1A patent/CN108958020A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106528705A (en) * | 2016-10-26 | 2017-03-22 | 桂林电子科技大学 | Repeated record detection method and system based on RBF neural network |
CN107045289A (en) * | 2017-06-05 | 2017-08-15 | 杭州电子科技大学 | A kind of nonlinear neural network optimization PID control method of electric furnace temperature |
CN107608215A (en) * | 2017-10-18 | 2018-01-19 | 齐鲁工业大学 | A kind of mapping UAV Attitude control method based on RBF neural |
CN107703754A (en) * | 2017-10-30 | 2018-02-16 | 王才旺 | A kind of simulating turntable control system based on Neural Network PID Control |
CN107544263A (en) * | 2017-10-31 | 2018-01-05 | 河北农业大学 | Liquid manure proportioning control system and method are adjusted based on parameter self-tuning PID |
Non-Patent Citations (2)
Title |
---|
董文波: ""基于神经网络的锅炉蒸汽温度控制系统"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
马平等: ""锅炉过热蒸汽温度动态预测PID控制算法"", 《热力发电》 * |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109597449A (en) * | 2019-01-30 | 2019-04-09 | 杭州庆睿科技有限公司 | A kind of ultrasonic wave separating apparatus temprature control method neural network based |
CN110262582A (en) * | 2019-07-30 | 2019-09-20 | 中原工学院 | A kind of barotor temprature control method based on improvement RBF neural |
CN110554715A (en) * | 2019-10-25 | 2019-12-10 | 攀钢集团攀枝花钢铁研究院有限公司 | RBF neural network-based PID control method for hydrolysis process temperature of titanyl sulfate plus seed crystal |
CN111665716B (en) * | 2020-05-27 | 2022-07-05 | 广东工业大学 | Mathematical modeling method for extracting traditional Chinese medicine/natural plant |
CN111665716A (en) * | 2020-05-27 | 2020-09-15 | 广东工业大学 | Mathematical modeling method for extracting traditional Chinese medicine/natural plant |
CN112034701A (en) * | 2020-09-11 | 2020-12-04 | 长虹美菱股份有限公司 | PID control system and PID control algorithm based on temperature sampling |
CN112648717A (en) * | 2020-12-16 | 2021-04-13 | 桂林电子科技大学 | Terminal control method of multi-zone variable air volume air conditioning system based on radial basis function neural network |
CN113589862A (en) * | 2021-06-24 | 2021-11-02 | 苏州环美生物医疗科技有限公司 | Intelligent PID temperature control method and control chip based on network topology |
CN114779626A (en) * | 2021-11-13 | 2022-07-22 | 黄河交通学院 | Networked prediction control method based on neural network and weighted average algorithm |
CN114167724A (en) * | 2021-11-30 | 2022-03-11 | 中车长春轨道客车股份有限公司 | Dual-controller weighted control method based on black box model |
CN114167724B (en) * | 2021-11-30 | 2023-10-27 | 中车长春轨道客车股份有限公司 | Dual-controller authority-dividing control method based on black box model |
CN116627028A (en) * | 2023-07-21 | 2023-08-22 | 阳谷新太平洋电缆有限公司 | Control method for crosslinked cable production line |
CN116627028B (en) * | 2023-07-21 | 2023-09-29 | 阳谷新太平洋电缆有限公司 | Control method for crosslinked cable production line |
CN117316356A (en) * | 2023-10-24 | 2023-12-29 | 中国民航大学 | Feedforward compensation regulation and control method for composite material component autoclave molding process parameters |
CN117316356B (en) * | 2023-10-24 | 2024-05-17 | 中国民航大学 | Feedforward compensation regulation and control method for composite material component autoclave molding process parameters |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108958020A (en) | A kind of temprature control method based on RBF neural dynamic prediction PID | |
CN105892296B (en) | A kind of fractional order dynamic matrix control method of industry heating furnace system | |
CN109581877B (en) | Magnetic levitation ball sliding mode variable structure control method based on ESO | |
CN107632524B (en) | Communication machine room temperature model prediction control method and system | |
CN111520878A (en) | Air conditioner temperature control system based on RBF neural network and control method thereof | |
CN115494892B (en) | Decoupling control method for air inlet environment simulation system of high-altitude simulation test bed | |
CN106292785A (en) | Aluminum-bar heating furnace ignition temperature automaton based on neutral net | |
CN105388764A (en) | Electro-hydraulic servo PID control method and system based on dynamic matrix feed-forward prediction | |
CN109612289A (en) | Fuzzy Temperature Control of Electric Resistance Furnace method based on PREDICTIVE CONTROL | |
CN109062040A (en) | Predictive PID method based on the optimization of system nesting | |
CN110763420A (en) | Low-temperature wind tunnel flow field active disturbance rejection control method | |
CN107037727A (en) | A kind of big envelope curve adaptive gain dispatching method of depopulated helicopter | |
CN105425580A (en) | Construction method of ITAE (Integral Time absolute error) optimal N-type system | |
CN110779268A (en) | Refrigerator internal temperature control system algorithm | |
CN102393645A (en) | Control method of high-speed electro-hydraulic proportional governing system | |
CN109298735A (en) | The feed-forward and feedback composite control method of differential scanning calorimeter constant heating rates sintering process | |
CN103576552A (en) | Active-disturbance-rejection control method and device of single-input single-output pure lag self-balancing object | |
CN112859617B (en) | iPI model-free adaptive global nonsingular rapid terminal sliding mode control method | |
CN110488888A (en) | A kind of resistance-heated furnace temprature control method based on adaptive iteration study | |
CN103605284A (en) | Dynamic matrix control optimization-based waste plastic cracking furnace pressure controlling method | |
CN106651090B (en) | Normalized man-machine system flight quality prediction method | |
CN103809434B (en) | The multistage PID controller design method of the compound root locus of Longitudinal Flight model cluster | |
CN103353161B (en) | System and method for controlling variable air volume air conditioner pressure independent type terminal device | |
CN113703478A (en) | Fixed wing height setting control method, electronic equipment and storage medium | |
CN103823374B (en) | Aircraft multiloop model bunch compound root locus compensating controller method for designing |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181207 |