CN108803335A - A kind of out of order removing method of DC servo motor control - Google Patents
A kind of out of order removing method of DC servo motor control Download PDFInfo
- Publication number
- CN108803335A CN108803335A CN201810657507.2A CN201810657507A CN108803335A CN 108803335 A CN108803335 A CN 108803335A CN 201810657507 A CN201810657507 A CN 201810657507A CN 108803335 A CN108803335 A CN 108803335A
- Authority
- CN
- China
- Prior art keywords
- output
- layer
- input
- hidden layer
- matrix
- 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)
- Control Of Electric Motors In General (AREA)
- Feedback Control In General (AREA)
Abstract
Out of order removing method in being controlled the invention discloses a kind of DC servo motor, timestamp generator, logic comparator and predictive controller are added on the basis of traditional network control system, timestamp generator is that data packet adds timestamp, and select newest control signal for the control to DC servo motor using logic comparator in controller and actuator both ends, this paper presents the concepts of predictive controller simultaneously, when system occurs out of order using the real-time control performance of the further lifting system of prediction algorithm, system resource waste is avoided.
Description
Technical field
The present invention relates to the network-controls of DC servo motor, propose a kind of solve due to out of order caused by network transmission
The method of problem, and specifically describe relevant data packet rearrangement method.
Background technology
Network control system (NCS) refers to being connected controlled device, sensor, controller and actuator by network
The feedback control system come is convenient for remote operation and control.But the introducing of network also brings a series of ask to the control of motor
Topic, the wherein out of order runnability for largely affecting motor of data packet.
For the network-control of DC servo motor, Li Ruoqiong et al. was in the situation to time delay present in system in 2014
It is analyzed;Xu Pei et al. had studied influence of the packet loss to system control performance in 2013.But the above research does not all have
The influence for considering data packet disorder, it is difficult to meet the actual demand of engineering.
Invention content
In order to make up the deficiency of existing research, the present invention proposes a kind of improved data packet rearrangement method to solve motor control
Out of order problem in system, to improve the control performance of motor.
Technical solution of the present invention is as follows:
Timestamp generator is added on network control system, patrols for a kind of out of order removing method of DC servo motor control
Comparator and predictive controller are collected, realizes and resequences to out of order data packet;Specifically include following steps:
Step S1, the setting of timestamp generator in sensor side are used for that the collected data of sensor are marked, with
It is convenient subsequently to carry out out of order judgement;
Logic comparator is respectively set in controller and actuator both ends in step S2;Two logic comparators are respectively to control
The timestamp for the data that device and actuator processed receive is compared, and by the timestamp of newly arrived data and is stored in register
In the timestamps of data be compared, judge whether to occur out of order;If the timestamp of newly arrived data packet is newly in original
The timestamp of data packet, then it is out of order without occurring, otherwise occur out of order;If it is judged that not occurring out of order, deposit is updated
Data in device, otherwise the data in register remain unchanged;After register updates, then controller is sent to control signal
Actuator, actuator are applied to the control of DC servo motor by signal is controlled;
Step S3, predictive controller setting obtain prediction output in controller end, predictive controller by prediction algorithm
Signal u is simultaneously sent to controller end, generates control signal and is used for actuator, ensures the continuity of whole system;Prediction output
Signal u is the input of the motor system model of state equation description, and rolling optimization is carried out to system.
Prediction algorithm is the prediction algorithm based on extreme learning machine (ELM), specifically includes following steps:
(1), it using single hidden layer Architecture of Feed-forward Neural Network, determines the neuron number of hidden layer, is randomly provided and solid
Determine the connection weight w and hidden layer neuron threshold value b of input layer and implicit interlayer;
It is connected entirely between input layer and hidden layer, hidden layer and output layer neuron, input layer has n neuron, corresponding n a
Input quantity;The promising single layer of hidden layer shares l neuron;Output layer has m neuron, corresponding m output quantity;Input layer with
Connection weight w between hidden layer is:
In formula (1), wijIndicate the connection weight between i-th of input layer and j-th of hidden layer neuron, l tables
Show the neuron number of hidden layer;N indicates input layer number;
Connection weight β between hidden layer and output layer is:
Wherein βjkIndicate the connection weight between k-th of neuron of j-th of neuron of hidden layer and output layer;L indicates hidden
Neuron number containing layer;M indicates output layer neuron number;
The threshold value b of hidden layer neuron is:
Wherein biIndicate the threshold value of i-th of hidden layer neuron;
(2), it determines the activation primitive g (x) of hidden layer neuron, calculates hidden layer output matrix H;
(3), output layer weights are calculated;T ' is the transposition of matrix T;H is the hidden layer output square of neural network
Battle array;
(4), to known control signal, (prediction output signal u) is grouped processing, normalizes as neural network
Input matrix;
(5), the control signal of next step is predicted;
(6), using prediction result as the input of system model.
The motor system model described with state equation:
X (k) ∈ R in formulanIt is state variable, u (k) and y (k) are respectively that system is output and input, matrix A, B and cTIt is
Dimension is the normal matrix of n × n;M step variations (M=0,1,2 ...) have occurred in the input of system from the k moment, calculate defeated in prediction
Enter the system mode at u (k), u (k+1) ... u (k+i) ... u (k+M-1) the effect lower following P moment, u (k+i) is pre- by ELM algorithms
It measures;
System mode prediction is expressed as:
X (k)=Fxx(k)+GxU(k) (15)
Wherein,
Wherein FxAnd GxIt is the coefficient matrix of x (k) and u (k) respectively, is made of A and B;P is indicated to the following P sampling instant
Do P prediction;
Formula (15) prediction obtained the system mode of system future time instance, by export and state relational expression y (k)=
cTX (k) predicts the output of system, and sends output to controller end, carries out the rolling optimization of system.
The rolling optimization of system specifically includes following steps:
It is expressed as determining the M controlled quentity controlled variable u (k) from the k moment, u (k+1 ..., u (k+ in the state optimization problem at k moment
M-1), making controlled device in the case where M controlled quentity controlled variable acts on, the state at the following P moment is calmed, and levels off to x=0, optimization performance
Index is expressed as vector form:
Wherein, Qx, RxIt is state weight matrix and control weighting matrix;When not considering constraint, bonding state prediction model
Find out the analytical expression of optimal solution:The performance indicator at J (k) table k moment;
Thus instant controlled quentity controlled variable is found out:
Wherein feedback oscillator
It is the coefficient matrix G of u (k)xTransposition.
More preferably, step (2) specifically includes following steps
Training set input matrix X and output matrix Y with Q training sample are respectively:
xijIndicate the input value of i-th of input layer, j-th of sample, yijIndicate the output of i-th of output layer, j-th of sample
Value;
The activation primitive of hidden layer neuron is g (x), as neural network after training sample is normalized
Input, while predicting input of the output signal as system model:
U=[u1 u2…ui…uQ]n×Q(6)
Wherein,
In formula (7), uiIndicate i-th of input layer sample, uijIndicate the input of i-th of input layer, j-th of input sample;
The output T that neural network is obtained by ELM structure charts is:
T=[t1 t2…tj…tQ]m×Q (8)
Wherein, wi=[wi1, wi2... win];xj=[u1j, u2j..., unj]T;
tjIndicate the output of j-th of sample, tmjIndicate the output of m-th of output layer, j-th of sample;
Formula (9) is expressed as:
H β=T ' (10)
Wherein, T ' is the transposition of matrix T;H is the hidden layer output matrix of neural network;
More preferably, step (3) calculates output layer weightsSpecifically include following steps:
The threshold value b of fixed randomly selected input weight w and hidden layer, then train network be equal to ask linear system H β=
The least square solution β of T ',
Xie Wei:
Wherein, H+For the Moor-Penrose generalized inverses of hidden layer output matrix H.
Beneficial effects of the present invention include:
The present invention discloses a kind of DC servo motor and controls out of order removing method, and solve during motor control out of order asks
Topic is added one and is predicted system input based on the predictive controller of extreme learning machine, protected using prediction acquired results
The operation of card system, to solve the problems, such as because of system idle waiting caused by data packet disorder, pace of learning is fast, Generalization Capability
It is good.
Description of the drawings
Present invention will be further explained below with reference to the attached drawings and examples;
Fig. 1 is out of order elimination system construction drawing in DC servo motor control;
Fig. 2 is the mono- hidden layer Architecture of Feed-forward Neural Network figures of ELM.
Specific implementation mode
The invention will be further described below in conjunction with the accompanying drawings and by specific embodiment, and following embodiment is descriptive
, it is not restrictive, protection scope of the present invention cannot be limited with this.
In order to make technological means, creation characteristic, workflow, application method reached purpose and effect of the present invention, and it is
The evaluation method is set to be easy to understand with reference to specific embodiments the present invention is further explained.
As shown in Figure 1, a kind of DC servo motor controls out of order removing method, it is added on the basis of network control system
Timestamp generator, logic comparator and predictive controller are realized and are resequenced to out of order data packet, and out of order bring not is solved
Profit influences, and improves the control performance of motor;Specifically include following steps:
Step S1, the setting of timestamp generator in sensor side are used for that the collected data of sensor are marked, with
It is convenient subsequently to carry out out of order judgement;
Logic comparator is respectively set in controller and actuator both ends in step S2;Two logic comparators are respectively to control
The timestamp for the data that device and actuator processed receive is compared, and by the timestamp of newly arrived data and is stored in register
In the timestamps of data be compared, judge whether to occur out of order;If the timestamp of newly arrived data packet is newly in original
The timestamp of data packet, then it is out of order without occurring, otherwise occur out of order;If it is judged that not occurring out of order, deposit is updated
Data in device, otherwise the data in register remain unchanged;After register updates, then controller is sent to control signal
Actuator, actuator are applied to the control of DC servo motor by signal is controlled;
Step S3, predictive controller setting obtain prediction output in controller end, predictive controller by prediction algorithm
Signal u is simultaneously sent to controller end, generates control signal and is used for actuator, ensures the continuity of whole system;Prediction output
Signal u is the input of the motor system model of state equation description, and rolling optimization is carried out to system.
Prediction algorithm is the prediction algorithm based on extreme learning machine (ELM), specifically includes following steps:
(1), using single hidden layer Architecture of Feed-forward Neural Network (as shown in Figure 2), the neuron number of hidden layer is determined, with
Machine is arranged and fixes the connection weight w and hidden layer neuron threshold value b of input layer and implicit interlayer;
It is connected entirely between input layer and hidden layer, hidden layer and output layer neuron, input layer has n neuron, corresponding n a
Input quantity;The promising single layer of hidden layer shares l neuron;Output layer has m neuron, corresponding m output quantity;Input layer with
Connection weight w between hidden layer is:
In formula (1), wijIndicate the connection weight between i-th of input layer and j-th of hidden layer neuron;
Connection weight β between hidden layer and output layer is:
Wherein βjkIndicate the connection weight between k-th of neuron of j-th of neuron of hidden layer and output layer;L indicates hidden
Neuron number containing layer;
The threshold value b of hidden layer neuron is:
Wherein biIndicate the threshold value of i-th of hidden layer neuron;
(2), it determines the activation primitive g (x) of hidden layer neuron, calculates output matrix H;
Training set input matrix X and output matrix Y with Q training sample are respectively:
xijIndicate the input value of i-th of input layer, j-th of sample, yijIndicate the output of i-th of output layer, j-th of sample
Value;
The activation primitive of hidden layer neuron is g (x), as neural network after training sample is normalized
Input, while predicting input of the output signal as system model:
U=[u1 u2…ui…uQ]n×Q (6)
Wherein,
In formula (7), uiIndicate i-th of input layer sample, uijIndicate the input of i-th of input layer, j-th of input sample;
As shown in Fig. 2, the output T for obtaining neural network by ELM structure charts is:
T=[t1 t2…tj…tQ]m×Q (8)
Wherein, wi=[wi1, wi2..., win];xj=[u1j, u2j..., unj]T;
tjIndicate the output of j-th of sample, tmjIndicate the output of m-th of output layer, j-th of sample;
Formula (9) is expressed as:
H β=T ' (10)
Wherein, T ' is the transposition (T of matrix TT);H is the hidden layer output matrix of neural network;
(3), output layer weights are calculated
The threshold value b of fixed randomly selected input weight w and hidden layer, then train network be equal to ask linear system H β=
The least square solution β of T ',
Xie Wei:
Wherein, H+For the Moor-Penrose generalized inverses of hidden layer output matrix H;
(4), to known control signal, (prediction output signal u) is grouped processing, and normalization is as god
Input matrix through network;
(5), the control signal of next step is predicted;
(6), using prediction result as the input of system model.
The motor system model described with state equation:
X (k) ∈ R in formulanIt is state variable and can surveys in real time, u (k) and y (k) is respectively that system is output and input, matrix
A, B and cTIt is the normal matrix that dimension is n × n;M step variations (M=0,1,2 ...), prediction has occurred in the input of system from the k moment
Future u (k), the system mode at u (k+1) ... u (k+i) ... u (k+M-1) a moment, u (k+i) are predicted by ELM algorithms under effect
It obtains;System mode prediction is expressed as:
X (K)=Fxx(κ)+GxU(κ) (15)
Wherein,
Wherein FxAnd GxIt is the coefficient matrix of x (k) and u (k) respectively, is made of A and B;The system that P indicates the following P moment
State does P prediction at the time of in following P sampling period point;
Formula (15) prediction obtained the system mode of system future time instance, by export and state relational expression y (k)=
cTX (k) predicts the output of system, and sends output to controller end, carries out the rolling optimization of system.
The rolling optimization of system specifically includes following steps:
It is expressed as determining the M controlled quentity controlled variable u (k) from the k moment, u (k+1) ..., u (k in the state optimization problem at k moment
+ M-1), making controlled device in the case where M controlled quentity controlled variable acts on, the state at the following P moment is calmed, and levels off to x=0, optimization property
Energy index is expressed as vector form:
Wherein, Qx, RxIt is state weight matrix and control weighting matrix;When not considering constraint, bonding state prediction model
Find out the analytical expression of optimal solution:J (k) indicates the performance indicator at k moment
It is the coefficient matrix G of u (k)xTransposition;
Thus instant controlled quentity controlled variable is found out:
Wherein feedback oscillator
Those skilled in the art can be modified to the present invention or the think of of modification designed but do not depart from the present invention
Think and range.Therefore, if these modifications and changes of the present invention belongs to the claims in the present invention and its equivalent technical scope
Within, then the present invention is also intended to include these modifications and variations.
Claims (6)
1. a kind of DC servo motor controls out of order removing method, which is characterized in that
Timestamp generator, logic comparator and predictive controller are added on network control system, realizes to out of order data packet
Rearrangement;Specifically include following steps:
Step S1, the setting of timestamp generator in sensor side are used for that the collected data of sensor are marked;
Logic comparator is respectively set in controller and actuator both ends in step S2;Two logic comparators are respectively to controller
The timestamp of the data received with actuator is compared, and by the timestamps of newly arrived data and is stored in register
The timestamp of data is compared, and judges whether to occur out of order;If the timestamp of newly arrived data packet is newly in original data
The timestamp of packet, then it is out of order without occurring, otherwise occur out of order;If it is judged that not occurring out of order, update in register
Data, otherwise the data in register remain unchanged;After register updates, then control signal is sent to execution by controller
Device, actuator are applied to the control of DC servo motor by signal is controlled;
Step S3, predictive controller setting obtain prediction output signal u in controller end, predictive controller by prediction algorithm
And it is sent to controller end, it generates control signal and is used for actuator;Predict that output signal u is the department of electrical engineering of state equation description
The input of system model carries out rolling optimization to system.
2. a kind of DC servo motor according to claim 1 controls out of order removing method, which is characterized in that
The prediction algorithm is the prediction algorithm based on extreme learning machine, specifically includes following steps:
(1), it using single hidden layer Architecture of Feed-forward Neural Network, determines the neuron number of hidden layer, be randomly provided and fix defeated
Enter the connection weight w and hidden layer neuron threshold value b of layer and implicit interlayer;
It is connected entirely between input layer and hidden layer, hidden layer and output layer neuron, input layer has n neuron, corresponding n input
Amount;The promising single layer of hidden layer shares l neuron;Output layer has m neuron, corresponding m output quantity;Input layer with it is implicit
Layer between connection weight w be:
In formula (1), wijIndicate that the connection weight between i-th of input layer and j-th of hidden layer neuron, l indicate hidden
Neuron number containing layer;It indicates
Connection weight β between hidden layer and output layer is:
Wherein βjkIndicate the connection weight between k-th of neuron of j-th of neuron of hidden layer and output layer;L indicates hidden layer
Neuron number;M indicates output layer neuron number;
The threshold value b of hidden layer neuron is:
Wherein biIndicate the threshold value of i-th of hidden layer neuron;
(2), it determines the activation primitive g (x) of hidden layer neuron, calculates output matrix H;
(3), output layer weights are calculatedT ' is the transposition of matrix T;H is the hidden layer output matrix of neural network;
(4), processing is grouped to known control signal, normalizes the input matrix as neural network;
(5), the control signal of next step is predicted;
(6), using prediction result as the input of system model.
3. a kind of DC servo motor according to claim 1 controls out of order removing method, which is characterized in that
The motor system model described with state equation:
X (k) ∈ R in formulanIt is state variable, u (k) and y (k) are respectively that system is output and input, matrix A, B and cTBe dimension be n
The normal matrix of × n;M step variations have occurred in the input of system from the k moment, calculate in prediction input u (k), u (k+1) ... u (k+
I) ... the system mode at u (k+M-1) the effects lower following P moment, u (k+i) are predicted to obtain by ELM algorithms;
System mode prediction is expressed as:
X (k)=Fxx(k)+GxU(k) (15)
Wherein,
Wherein FxAnd GxIt is the coefficient matrix of x (k) and u (k) respectively, is made of A and B;P expressions are P to the following P sampling instant
Secondary prediction;
Formula (15) prediction obtains the system mode of system future time instance, by exporting relational expression y (k)=c with stateTX (k) comes
The output of system is predicted, and sends output to controller end, carries out the rolling optimization of system.
4. a kind of DC servo motor according to claim 1 controls out of order removing method, which is characterized in that
The rolling optimization of system specifically includes following steps:
It is expressed as determining the M controlled quentity controlled variable u (k) from the k moment, u (k+1) ..., u (k+M- in the state optimization problem at k moment
1), making controlled device in the case where M controlled quentity controlled variable acts on, the state at the following P moment is calmed, and levels off to x=0, is optimized performance and is referred to
Mark is expressed as vector form:
Wherein, Qx, RxIt is state weight matrix and control weighting matrix;When not considering constraint, bonding state prediction model is found out
The analytical expression of optimal solution:The performance indicator at J (k) table k moment;
Thus instant controlled quentity controlled variable is found out:
Wherein feedback oscillator
It is the coefficient matrix G of u (k)xTransposition.
5. a kind of DC servo motor according to claim 2 controls out of order removing method, which is characterized in that
Step (2) specifically includes following steps
Training set input matrix X and output matrix Y with Q training sample are respectively:
xijIndicate the input value of i-th of input layer, j-th of sample, yijIndicate the output valve of i-th of output layer, j-th of sample;
The activation primitive of hidden layer neuron is g (x), as the defeated of neural network after training sample is normalized
Enter, while predicting input of the output signal as system model:
U=[u1 u2 …ui…uQ]n×Q (6)
Wherein,
In formula (7), uiIndicate i-th of input layer sample, uijIndicate the input of i-th of input layer, j-th of input sample;
The output T that neural network is obtained by ELM structure charts is:
T=[t1 t2 …tj… tQ]m×Q (8)
Wherein, wi=[wi1, wi2... win];xj=[u1j, u2j..., unj]T;
tjIndicate the output of j-th of sample, tmjIndicate the output of m-th of output layer, j-th of sample;
Formula (9) is expressed as:
H β=T ' (10)
Wherein, T ' is the transposition of matrix T;H is the hidden layer output matrix of neural network;
6. a kind of DC servo motor according to claim 2 controls out of order removing method, which is characterized in that
Step (3) calculates output layer weightsSpecifically include following steps:
The threshold value b of fixed randomly selected input weight w and hidden layer, then train network to be equal to and ask linear system H β=T's '
Least square solution β,
Xie Wei:
Wherein, H+For the Moor-Penrose generalized inverses of hidden layer output matrix H.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810657507.2A CN108803335B (en) | 2018-06-25 | 2018-06-25 | Method for eliminating control disorder of direct current servo motor |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810657507.2A CN108803335B (en) | 2018-06-25 | 2018-06-25 | Method for eliminating control disorder of direct current servo motor |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108803335A true CN108803335A (en) | 2018-11-13 |
CN108803335B CN108803335B (en) | 2021-05-11 |
Family
ID=64084857
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810657507.2A Active CN108803335B (en) | 2018-06-25 | 2018-06-25 | Method for eliminating control disorder of direct current servo motor |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108803335B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114237042A (en) * | 2021-11-25 | 2022-03-25 | 季华实验室 | Smart finger pulse feeling pre-pressure control method and system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101437239A (en) * | 2008-12-11 | 2009-05-20 | 吉林大学 | Real time sensor signal network transmission method based on linear prediction filtering |
US20100202298A1 (en) * | 2009-02-10 | 2010-08-12 | Microsoft Corporation | Network coordinate systems using ip information |
CN105353622A (en) * | 2015-11-30 | 2016-02-24 | 浙江工业大学 | Dynamic matrix control method for networked system having time delay and packet loss |
CN107122790A (en) * | 2017-03-15 | 2017-09-01 | 华北电力大学 | Non-intrusion type load recognizer based on hybrid neural networks and integrated study |
CN107145935A (en) * | 2017-05-12 | 2017-09-08 | 东华大学 | A kind of Smith Prediction Control methods based on modified neutral net |
-
2018
- 2018-06-25 CN CN201810657507.2A patent/CN108803335B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101437239A (en) * | 2008-12-11 | 2009-05-20 | 吉林大学 | Real time sensor signal network transmission method based on linear prediction filtering |
US20100202298A1 (en) * | 2009-02-10 | 2010-08-12 | Microsoft Corporation | Network coordinate systems using ip information |
CN105353622A (en) * | 2015-11-30 | 2016-02-24 | 浙江工业大学 | Dynamic matrix control method for networked system having time delay and packet loss |
CN107122790A (en) * | 2017-03-15 | 2017-09-01 | 华北电力大学 | Non-intrusion type load recognizer based on hybrid neural networks and integrated study |
CN107145935A (en) * | 2017-05-12 | 2017-09-08 | 东华大学 | A kind of Smith Prediction Control methods based on modified neutral net |
Non-Patent Citations (2)
Title |
---|
DANIEL L´OPEZ-ECHEVARR´IA等: "Neural Network and Observer-Based Compensation of Networked Control Systems with Packet Dropout", 《IFAC》 * |
张奇智 等: "网络控制***中的时戳预测函数控制", 《控制理论与应用》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114237042A (en) * | 2021-11-25 | 2022-03-25 | 季华实验室 | Smart finger pulse feeling pre-pressure control method and system |
CN114237042B (en) * | 2021-11-25 | 2022-06-03 | 季华实验室 | Pulse feeling pre-pressure control method and system |
Also Published As
Publication number | Publication date |
---|---|
CN108803335B (en) | 2021-05-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109086804B (en) | Hydraulic equipment early failure prediction method based on fusion of multi-source state monitoring information and reliability characteristics | |
CN109270842B (en) | Bayesian network-based regional heat supply model prediction control system and method | |
US20200241487A1 (en) | Networked control system time-delay compensation method based on predictive control | |
CN102496069B (en) | Cable multimode safe operation evaluation method based on fuzzy analytic hierarchy process (FAHP) | |
CN110472778A (en) | A kind of short-term load forecasting method based on Blending integrated study | |
CN112053560B (en) | Short-time traffic flow prediction method, system and storage medium based on neural network | |
RU2602973C1 (en) | Neural network training controller with genetic algorithm | |
CN108038300A (en) | Optical fiber state evaluating method based on improved membership function combination neutral net | |
CN112437131B (en) | Data dynamic acquisition and transmission method considering data correlation in Internet of things | |
CN108280998A (en) | Short-time Traffic Flow Forecasting Methods based on historical data dynamic select | |
CN108256689A (en) | A kind of neural network prediction method of non-crystaline amorphous metal thermoplastic forming performance | |
CN111555297B (en) | Unified time scale voltage control method with tri-state energy unit | |
CN116224947B (en) | Deep learning industrial control system capable of actively integrating domain rule knowledge | |
CN106055579A (en) | Vehicle performance data cleaning system based on artificial neural network, and method thereof | |
CN110705920A (en) | Method for predicting cement strength by using multi-hidden-layer BP network under big data environment | |
CN109766574A (en) | A kind of double pantograph Active Control Methods of the H ∞ control based on estimation | |
CN106650959A (en) | Power distribution network repair ability assessment method based on improved grey clustering | |
Putra et al. | Multiagent architecture for bridge capacity measurement system using wireless sensor network and weight in motion | |
CN112784920A (en) | Cloud-side-end-coordinated dual-anti-domain self-adaptive fault diagnosis method for rotating part | |
CN111445695A (en) | Running state monitoring method of intelligent vehicle road system | |
CN108803335A (en) | A kind of out of order removing method of DC servo motor control | |
CN106845711A (en) | Processing method and processing device for power supply reliability data | |
CN111489027A (en) | Hydroelectric generating set waveform data trend prediction method and system | |
CN116663744A (en) | Energy consumption prediction method and system for near-zero energy consumption building | |
Shafabakhsh et al. | Determining the relative importance of parameters affecting concrete pavement thickness |
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 |