CN110278571A - It is a kind of based on simple forecast-correction link distributed signal tracking - Google Patents
It is a kind of based on simple forecast-correction link distributed signal tracking Download PDFInfo
- Publication number
- CN110278571A CN110278571A CN201910544597.9A CN201910544597A CN110278571A CN 110278571 A CN110278571 A CN 110278571A CN 201910544597 A CN201910544597 A CN 201910544597A CN 110278571 A CN110278571 A CN 110278571A
- Authority
- CN
- China
- Prior art keywords
- node
- moment
- correction link
- state
- signal tracking
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/22—Traffic simulation tools or models
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/04—Arrangements for maintaining operational condition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W84/00—Network topologies
- H04W84/18—Self-organising networks, e.g. ad-hoc networks or sensor networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a kind of based on simple forecast-correction link distributed signal tracking, comprising: constructs the network structure topological diagram of multiple agent;The cost function of multi-agent system is constructed according to the network structure topological diagram;Using the variation of each node k moment and k-1 moment node state as prediction direction, substitutes into predictor formula and obtain the predicted value of each node k+1 moment node state;It is corrected using predicted value of the gradient descent method to each node k+1 moment node state, obtains the optimal estimation value of each node k+1 moment node state;Time update is carried out, continues to calculate the subsequent time i.e. optimal estimation value of k+2 moment node state.Compared with prior art, the present invention can be improved the robustness and adaptivity of system, reduce computational complexity, improve operation efficiency, improve real-time.
Description
Technical field
The present invention relates to a kind of based on simple forecast-correction link distributed signal tracking, belongs to control and letter
Cease technical field.
Background technique
The set that multi-agent system is made of multiple multiagent systems to intercouple, each multiagent system have one
Fixed independence, and can be communicated by the environment around perception with other intelligent bodies.With development in recent years, more intelligence
The distributed collaborative control of energy system system has become a hot spot of control field research.Distributed AC servo system Relatively centralized formula
For control, have many advantages, such as that small cost, high reliablity, flexibility is high, scalability is high, has extensive engineering background and answers
Use prospect.Distributed tracking method is widely used in solving to want to minimize the global object letter of continuous time-varying between different intelligent body
On number.With the continuous development of sensor technology, sensor can be very good to acquire the variation of continuous time varying signal,
Matter or discrete sampling.Therefore constant problem when continuous time-varying problem discretization can be resolved into a series of by us.When
The method of constant optimization problem is mainly or based on gradient descent method or Newton method when preceding solution, it is contemplated that has in actual conditions
A little reference signals are continually changing, therefore the independent optimal value for being gone to calculate each moment with static gradient method or Newton method
It is unrealistic.In recent years, the method for prediction link being added in conjunction with dynamic optimization thought and in optimization algorithm can be fine
The continually changing reference signal of tracking, solve the problems, such as that calculating time cost is continuously increased, while predicting the addition of link
Improve the convergence effect of optimization algorithm.
Compared to centralized Prediction-correction algorithm, each intelligent body is only needed using certainly in Distributed Predictive-correcting algorithm
It is logical to reduce intelligent body for the information of oneself neighbours' intelligent body, the information without knowing other all intelligent bodies of whole network
The load of letter, and the robustness and privacy of system can also be improved.But for current already present Distributed Predictive-
Correcting algorithm, prediction link need to carry out inversion calculation to the Hessen matrix of cost function, cause it to predict that link calculates complicated
Degree is high, can reduce the operation efficiency of algorithm, influence real-time performance;And for multi-agent system, if cost function
Hessen matrix do not carry out additional processing to matrix first before inverting, then the new matrix that matrix inversion obtains is frequently not adjacent
Matrix is connect, distributed algorithm design can not be directly carried out.Therefore in the Distributed Design of algorithm how the inverse fortune of processing array
It calculates and simplifies prediction link and be also a matter of concern.
Summary of the invention
The object of the present invention is to provide a kind of based on simple forecast-correction link distributed signal tracking, until
It can solve one of above-mentioned technical problem less.
In order to solve the above technical problems, the present invention adopts the following technical scheme that:
It is a kind of based on simple forecast-correction link distributed signal tracking, comprising: step S1, construct it is how intelligent
The network structure topological diagram of body, the network structure topological diagram include n node, and each node respectively represents an intelligent body,
And obtain the neighbor information of the point set of the network structure topological diagram, side collection and each node, wherein n is integer and n >=2;
Step S2 constructs the cost function of multi-agent system;Simultaneously sampling time interval h is arranged in step S3, definition node original state;
Step S4, using the variation of each node k moment node state and k-1 moment node state as prediction direction, prediction is each
The variation tendency of the reference signal at node k+1 moment substitutes into predictor formula and obtains the prediction of each node k+1 moment node state
Value;Step S5, using the predicted value of k+1 moment node state as the initial value of correction link, and using gradient descent method to every
The predicted value of a node k+1 moment node state is corrected, and obtains the optimal estimation of each node k+1 moment node state
Value;Step S6 carries out time update, according to the information at k+1 moment and k moment, repeating said steps S4 and the step S5, and meter
Calculation obtains the optimal estimation value at each node k+2 moment.
It is above-mentioned based in simple forecast-correction link distributed signal tracking, in the step S1, the net
Network structural topology chart is shown as G=(V, E), wherein the set of V={ 1 ..., n } expression node, E=(i, j) | j ∈ Ni,j
=1...n } indicate side set, NiIndicate the set of the neighbor node of node i, j ∈ Ni, i ∈ Nj, the network structure topology
Figure is Connected undigraph, and node j is father node, and node i is child node.
It is above-mentioned based in simple forecast-correction link distributed signal tracking, it is described more in the step S2
Intelligent body cost function has following form:Wherein, xiIt indicates i-th
The state of intelligent body, xi∈Rp;x∈RnpIt is the stacking of each Decision-making of Agent variable;fi(xi;T) expression and intelligent body itself
Related function;gi,j(xi,xj;T) function that two intelligent bodies being in communication with each other in network structure topological diagram are constituted is indicated;t
Indicate the time;The set on E expression side;P is an integer, indicates the dimension of vector.
It is above-mentioned based in simple forecast-correction link distributed signal tracking, the step S3 is specifically included:
Original state x (the t of intelligent body is set0),
It is above-mentioned based in simple forecast-correction link distributed signal tracking, it is described pre- in the step S4
Survey formula are as follows: xk+1|k=xk+hpk, wherein xk+1|kFor the predicted value of k+1 moment node state, xk+1|k∈Rnp, pkIndicate prediction
The variable of reference signal change direction,
It is above-mentioned based in simple forecast-correction link distributed signal tracking, in the step S5, school is set
The initial value of positive linkThe initial value of correction link is the output valve x for predicting linkk+1|k, i.e.,
It is above-mentioned based in simple forecast-correction link distributed signal tracking, in the step S5, i-th
The correction link of intelligent body indicates are as follows:Wherein, γ is step-length;I-th
Intelligent body is about xiThe expression formula of gradient be
It is above-mentioned based in simple forecast-correction link distributed signal tracking, in the step S5, obtain through
Output valve after crossing η step correction, which is then optimal estimation value, i.e.,
Compared with prior art, the present invention devises a kind of based on the tracking of simple forecast-correction link distributed signal
Method, this method only passes through the team collaboration of local neighbours, and each node has only used the information of neighbor node, improves whole
The robustness and adaptivity of a system;Simultaneously also achieve prediction link simplification, avoid inverse of a matrix calculate and its
The calculating process of his complexity largely reduces the computation complexity of prediction link, and calculation amount is small, and real-time performance is good, and
And it ensure that convergence effect.
Detailed description of the invention
Fig. 1 is the flow chart of method provided in an embodiment of the present invention.
The present invention is further illustrated with reference to the accompanying drawings and detailed description.
Specific embodiment
The embodiment of the present invention provide it is a kind of based on simple forecast-correction link distributed signal tracking, such as Fig. 1 institute
Show, comprising:
Step S1, constructs the network structure topological diagram of multiple agent, and network structure topological diagram includes n node, Mei Gejie
Point respectively represents an intelligent body, and obtains the neighbor information of the point set of network structure topological diagram, side collection and each node,
In, n is integer and n >=2;
The present embodiment is directed to the multi-agent system with n intelligent body.
In step S1, network structure topological diagram includes n node, is expressed as G=(V, E), wherein V={ 1 ..., n } table
Show the set of node, E=(i, j) | j ∈ Ni, j=1...n } indicate side set, NiIndicate the collection of the neighbor node of node i
It closes, j ∈ Ni, i ∈ Nj, network structure topological diagram is Connected undigraph, and node j is father node, and node i is child node.
Step S2 constructs the cost function of multi-agent system;
As a kind of optional embodiment of the present embodiment, the cost function of the multi-agent system in step S2 has such as
Lower form:
Wherein, xiIndicate the state of i-th of intelligent body, xi∈Rp;x∈RnpIt is the stacking of each Decision-making of Agent variable, than
Such as, x=(x1 T;...;xn T)T;fi(xi;T) only function related with intelligent body itself is indicated;gi,j(xi,xj;T) network is indicated
The function that two intelligent bodies being in communication with each other in structural topology figure are constituted;∑ indicates summation operation;T representing matrix or vector
Transposition operation;T indicates the time;The set on E expression side;P is an integer, indicates the dimension of vector.
Simultaneously sampling time interval h is arranged in step S3, definition node original state;
Step S3 is specifically included: the original state x (t of intelligent body is arranged0), original state, which can appoint, to be taken,
In the present embodiment, sampling time interval is generally required less than 1, between adjusting the sampling time generally according to actual conditions
Every.
Step S4, using the variation of each node k moment node state and k-1 moment node state as prediction direction,
It predicts the variation tendency of the reference signal at each node k+1 moment, substitutes into predictor formula and obtain each node k+1 moment node shape
The predicted value of state;
As a kind of optional embodiment of the present embodiment, in step S4, predictor formula are as follows: xk+1|k=xk+hpk, wherein
xk+1|kFor the predicted value of k+1 moment node state, xk+1|k∈Rnp, pkIndicate the variable in prediction reference signal intensity direction,
Step S5 using the predicted value of k+1 moment node state as the initial value of correction link, and utilizes gradient descent method
The predicted value of each node k+1 moment node state is corrected, optimal the estimating of each node k+1 moment node state is obtained
Evaluation;
In step S5, the initial value of correction link is setThe initial value of correction link is the output valve for predicting link
xk+1|k, i.e.,
The correction link of i-th of intelligent body indicates are as follows:
Wherein, γ is step-length, and for step-length according to the model debugging of practical application, what should not usually be adjusted is too big, prevents iteration not
Convergence;The value range of step-length is generally configured according to actual needs, is adjusted generally according to 3 times, i.e., 0.00001,
0.00003,0.0001,0.0003,0.001,0.003,0.01,0.03,0.1,0.3 ... determine that range is finely tuned again later.
I-th of intelligent body is about xiGradient expression formula are as follows:
The output valve after η walks correction is obtained, which is then optimal estimation value, i.e.,
Step S6 carries out time update according to the information at k+1 moment and k moment and is repeated in step S4 and step S5,
The each node subsequent time i.e. optimal estimation value at k+2 moment is calculated.
The present embodiment the method only passes through the team collaboration of local neighbours, and each node has only used the letter of neighbor node
Breath, improves the robustness and adaptivity of whole system;The simplification of prediction link is also achieved simultaneously, avoids matrix
Inverse calculating and other complicated calculating processes, largely reduce the computation complexity of prediction link, calculation amount is small, in real time
Performance is good, and ensure that convergence effect.
These are only the preferred embodiment of the present invention, is not intended to restrict the invention, for those skilled in the art
For member, the invention may be variously modified and varied.It is all within creativeness spirit of the invention and principle, it is made any
Modification, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.
Claims (8)
1. a kind of based on simple forecast-correction link distributed signal tracking characterized by comprising
Step S1, constructs the network structure topological diagram of multiple agent, and the network structure topological diagram includes n node, Mei Gejie
Point respectively represents an intelligent body, and obtains neighbours' letter of the point set of the network structure topological diagram, side collection and each node
Breath, wherein n is integer and n >=2;
Step S2 constructs the cost function of multi-agent system;
Simultaneously sampling time interval h is arranged in step S3, definition node original state;
Step S4, using the variation of each node k moment node state and k-1 moment node state as prediction direction, prediction
The variation tendency of the reference signal at each node k+1 moment substitutes into predictor formula and obtains each node k+1 moment node state
Predicted value;
Step S5, using the predicted value of k+1 moment node state as the initial value of correction link, and using gradient descent method to every
The predicted value of a node k+1 moment node state is corrected, and obtains the optimal estimation of each node k+1 moment node state
Value;
Step S6 carries out time update, according to the information at k+1 moment and k moment, repeating said steps S4 and the step S5,
The optimal estimation value at each node k+2 moment is calculated.
2. according to claim 1 based on simple forecast-correction link distributed signal tracking, feature exists
In in the step S1, the network structure topological diagram is expressed as G=(V, E), wherein V={ 1 ..., n } indicates node
Set, E=(i, j) | j ∈ Ni, j=1...n } indicate side set, NiIndicate the set of the neighbor node of node i, j ∈ Ni, i
∈Nj, the network structure topological diagram is Connected undigraph, and node j is father node, and node i is child node.
3. according to claim 1 or 2 based on simple forecast-correction link distributed signal tracking, feature
It is, in the step S2, the cost function of the multi-agent system has following form:
Wherein, xiIndicate the state of i-th of intelligent body, xi∈Rp;x∈RnpIt is the stacking of each Decision-making of Agent variable;fi(xi;
T) only function related with intelligent body itself is indicated;gi,j(xi,xj;T) two be in communication with each other in network structure topological diagram are indicated
The function that intelligent body is constituted;T indicates the time;The set on E expression side;P is an integer, indicates the dimension of vector.
4. it is according to any one of claims 1 to 3 based on simple forecast-correction link distributed signal tracking,
It is characterized in that, the step S3 is specifically included: the original state x (t of intelligent body is arranged0),
5. it is according to any one of claims 1 to 4 based on simple forecast-correction link distributed signal tracking,
It is characterized in that, in the step S4, the predictor formula are as follows: xk+1|k=xk+hpk, wherein xk+1|kFor k+1 moment node shape
The predicted value of state, xk+1|k∈Rnp, pkIndicate the variable in prediction reference signal intensity direction,
6. it is according to any one of claims 1 to 5 based on simple forecast-correction link distributed signal tracking,
It is characterized in that, the initial value of correction link is arranged in the step S5The initial value of correction link is prediction link
Output valve xk+1|k, i.e.,
7. according to claim 6 based on simple forecast-correction link distributed signal tracking, feature exists
In in the step S5, the correction link of i-th of intelligent body is indicated are as follows:
Wherein, γ is step-length;I-th of intelligent body is about xiGradient expression formula are as follows:
8. according to claim 7 based on simple forecast-correction link distributed signal tracking, feature exists
In, in the step S5, obtain by η walk correction after output valve, which is then optimal estimation value, i.e.,
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910544597.9A CN110278571B (en) | 2019-06-21 | 2019-06-21 | Distributed signal tracking method based on simple prediction-correction link |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910544597.9A CN110278571B (en) | 2019-06-21 | 2019-06-21 | Distributed signal tracking method based on simple prediction-correction link |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110278571A true CN110278571A (en) | 2019-09-24 |
CN110278571B CN110278571B (en) | 2022-05-24 |
Family
ID=67961644
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910544597.9A Active CN110278571B (en) | 2019-06-21 | 2019-06-21 | Distributed signal tracking method based on simple prediction-correction link |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110278571B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111385305A (en) * | 2020-03-18 | 2020-07-07 | 东北大学秦皇岛分校 | Distributed image encryption/decryption method based on average consistency |
CN111953515A (en) * | 2020-07-07 | 2020-11-17 | 西南大学 | Double-acceleration distributed asynchronous optimization method based on Nesterov gradient method and gravity method |
CN113848701A (en) * | 2021-06-11 | 2021-12-28 | 东北大学秦皇岛分校 | Distributed average tracking method of uncertainty directed network |
CN115378813A (en) * | 2022-08-12 | 2022-11-22 | 大连海事大学 | Distributed online optimization method based on differential privacy mechanism |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090086745A1 (en) * | 2007-10-01 | 2009-04-02 | Samsung Electronics Co., Ltd | Method and a system for matching between network nodes |
CN105634828A (en) * | 2016-03-03 | 2016-06-01 | 厦门大学 | Method for controlling distributed average tracking of linear differential inclusion multi-agent systems |
CN106502097A (en) * | 2016-11-18 | 2017-03-15 | 厦门大学 | A kind of distributed average tracking method based on time delay sliding formwork control |
CN107451704A (en) * | 2017-09-15 | 2017-12-08 | 厦门大学 | A kind of multiple time delay intelligent grid economic load dispatching method based on PREDICTIVE CONTROL |
US20180120775A1 (en) * | 2016-11-01 | 2018-05-03 | Mitsubishi Electric Research Laboratories, Inc | Multi-Agent Control System and Method |
-
2019
- 2019-06-21 CN CN201910544597.9A patent/CN110278571B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090086745A1 (en) * | 2007-10-01 | 2009-04-02 | Samsung Electronics Co., Ltd | Method and a system for matching between network nodes |
CN105634828A (en) * | 2016-03-03 | 2016-06-01 | 厦门大学 | Method for controlling distributed average tracking of linear differential inclusion multi-agent systems |
US20180120775A1 (en) * | 2016-11-01 | 2018-05-03 | Mitsubishi Electric Research Laboratories, Inc | Multi-Agent Control System and Method |
CN106502097A (en) * | 2016-11-18 | 2017-03-15 | 厦门大学 | A kind of distributed average tracking method based on time delay sliding formwork control |
CN107451704A (en) * | 2017-09-15 | 2017-12-08 | 厦门大学 | A kind of multiple time delay intelligent grid economic load dispatching method based on PREDICTIVE CONTROL |
Non-Patent Citations (2)
Title |
---|
YAO MENG ET AL.: ""An optimal cooperative tracking algorithm for discrete-time multi-agent systems"", 《2015 34TH CHINESE CONTROL CONFERENCE(CCC)》 * |
高瑜隆: ""多智能体系统的分布式模型预测控制"", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技I辑》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111385305A (en) * | 2020-03-18 | 2020-07-07 | 东北大学秦皇岛分校 | Distributed image encryption/decryption method based on average consistency |
CN111385305B (en) * | 2020-03-18 | 2021-12-21 | 东北大学秦皇岛分校 | Distributed image encryption/decryption method based on average consistency |
CN111953515A (en) * | 2020-07-07 | 2020-11-17 | 西南大学 | Double-acceleration distributed asynchronous optimization method based on Nesterov gradient method and gravity method |
CN113848701A (en) * | 2021-06-11 | 2021-12-28 | 东北大学秦皇岛分校 | Distributed average tracking method of uncertainty directed network |
CN113848701B (en) * | 2021-06-11 | 2023-11-14 | 东北大学秦皇岛分校 | Distributed average tracking method for uncertainty directional network |
CN115378813A (en) * | 2022-08-12 | 2022-11-22 | 大连海事大学 | Distributed online optimization method based on differential privacy mechanism |
CN115378813B (en) * | 2022-08-12 | 2023-08-15 | 大连海事大学 | Distributed online optimization method based on differential privacy mechanism |
Also Published As
Publication number | Publication date |
---|---|
CN110278571B (en) | 2022-05-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110278571A (en) | It is a kind of based on simple forecast-correction link distributed signal tracking | |
Chen et al. | An intelligent robust networking mechanism for the Internet of Things | |
Wang et al. | DEFSI: Deep learning based epidemic forecasting with synthetic information | |
Yang et al. | Fault tolerant cooperative control for a class of nonlinear multi-agent systems | |
Liao | Hybrid improved differential evolution and wavelet neural network with load forecasting problem of air conditioning | |
Strumberger et al. | Monarch butterfly optimization algorithm for localization in wireless sensor networks | |
CN103971160A (en) | Particle swarm optimization method based on complex network | |
Yan et al. | An improved multihop-based localization algorithm for wireless sensor network using learning approach | |
WO2021036414A1 (en) | Co-channel interference prediction method for satellite-to-ground downlink under low earth orbit satellite constellation | |
Keshavarz-Mohammadiyan et al. | Consensus-based distributed unscented target tracking in wireless sensor networks with state-dependent noise | |
CN106529818A (en) | Water quality evaluation prediction method based on fuzzy wavelet neural network | |
CN103813355A (en) | Identification method for anomalous points of cooperative synchronization in distributed network | |
Pan et al. | Multi-source transfer ELM-based Q learning | |
CN109818792B (en) | Controller based on second-order linear system time-varying coupling complex dynamic network model | |
Pal et al. | Path length prediction in MANET under AODV routing: Comparative analysis of ARIMA and MLP model | |
Goudarzi et al. | Real-time and intelligent flood forecasting using UAV-assisted wireless sensor network | |
Swain et al. | A reinforcement learning-based cluster routing scheme with dynamic path planning for mutli-uav network | |
CN112770256B (en) | Node track prediction method in unmanned aerial vehicle self-organizing network | |
CN113784380A (en) | Topology prediction method adopting graph attention network and fusion neighborhood | |
Yu et al. | Distributed soft clustering algorithm for IoT based on finite time average consensus | |
He et al. | Short-term power load forecasting based on self-adapting PSO-BP neural network model | |
Wang et al. | A machine learning based connectivity restoration strategy for industrial IoTs | |
CN116781139A (en) | Flow prediction satellite path selection method and system based on reinforcement learning | |
Zhang et al. | Clique-based cooperative multiagent reinforcement learning using factor graphs | |
Yang et al. | Consensus of second-order multi-agent systems with nonsymmetric interconnection and heterogeneous delays |
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 |