CN107045785B - A method of the short-term traffic flow forecast based on grey ELM neural network - Google Patents

A method of the short-term traffic flow forecast based on grey ELM neural network Download PDF

Info

Publication number
CN107045785B
CN107045785B CN201710068325.7A CN201710068325A CN107045785B CN 107045785 B CN107045785 B CN 107045785B CN 201710068325 A CN201710068325 A CN 201710068325A CN 107045785 B CN107045785 B CN 107045785B
Authority
CN
China
Prior art keywords
network
neural network
output
input
grey
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.)
Active
Application number
CN201710068325.7A
Other languages
Chinese (zh)
Other versions
CN107045785A (en
Inventor
钱伟
车凯
王瑞
黄凯征
王俊峰
刘海波
李冰锋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Henan University of Technology
Original Assignee
Henan University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Henan University of Technology filed Critical Henan University of Technology
Priority to CN201710068325.7A priority Critical patent/CN107045785B/en
Publication of CN107045785A publication Critical patent/CN107045785A/en
Application granted granted Critical
Publication of CN107045785B publication Critical patent/CN107045785B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Quality & Reliability (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Tourism & Hospitality (AREA)
  • Game Theory and Decision Science (AREA)
  • General Business, Economics & Management (AREA)
  • Development Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention proposes a kind of methods of short-term traffic flow forecast based on grey ELM neural network, and step: being grouped collected data, the dimension innovation sequence such as obtain, then add up, and equal after add up ties up innovation sequence；Equal dimension innovation sequence after adding up is handled, the input matrix and target output matrix of network are obtained；The first random weight and threshold value for generating network, sets network parameter, and the input matrix collection of the network of generation and target output matrix collection are inputted neural network, training network；Input test data obtain the prediction output result of network；The accumulated value that dimension innovation sequence is subtracted etc. with neural network forecast result obtains actual prediction as a result, completing prediction.The beneficial effect is that: input data of the invention have passed through the processing of gray model, and otherness is smaller, so that the precision of prediction of grey ELM neural network is greatly increased.

Description

A method of the short-term traffic flow forecast based on grey ELM neural network
Technical field
The present invention relates to the technical fields of short-term traffic flow forecast, particularly relate to a kind of based on grey ELM neural network Short-term traffic flow forecast method.
Background technique
With the development of economy, automobile demand is continuously increased, highway communication flow is consequently increased, and is thus brought A series of traffic problems.In the case where not changing current road network, road network is dredged by intelligent traffic control system realization It leads and controls, be the effective way for solving traffic problems.Accurate forecasting traffic flow is the basis of traffic flow dredged and controlled, It is the important component of Intelligent traffic management systems.
Traffic flow itself has very strong uncertainty, is complexity, changeable, is easy by random disturbance, and advise Rule property is unobvious, with the introducing of different prediction techniques, many prediction models also occurs to the prediction of short-term traffic flow, but It is existing prediction technique genuine feature more demanding to data fluctuation and volatile.
Summary of the invention
The present invention is existing to solve the problems, such as, proposes a kind of short-term traffic flow forecast based on grey ELM neural network Method.
The technical scheme of the present invention is realized as follows:
A method of the short-term traffic flow forecast based on grey ELM neural network, step include:
A. grey processing is carried out to data, collected data is grouped according to formula (3), it may be assumed that set collected number According to for Q, then
Q=(q1, q2..., qm), (m ∈ N+) (1)
N group, every group of M+1 data are classified as, and are met
N+M=m, (n ∈ N+, M ∈ N+) (2)
For pth group therein, it is denoted as:
After formula (3) such as obtains at the dimension innovation sequence, add up according to formula (4) and (5), the grade reform after add up Cease sequence, it may be assumed that rightIn data carry out grey processing, obtaining one-accumulate sequence is
Wherein
B. input matrix collection and target output matrix collection are generated, to the equal dimension innovation sequence after adding up according to formula (6), formula (7), formula (8) is handled, and obtains the input matrix and target output matrix of network, it may be assumed that is chosenPreceding M be used as ELM nerve net The input of network, the M+1 desired outputs as network then have
To n group data are divided into above, the input matrix collection x and target output matrix collection y for the network being made of it are respectively
X=[X1, X2..., Xn] (7)
Y=[Y1, Y2..., Yn] (8)
If the reality output matrix T of network is
T=[T1, T2..., Tn] (9)；
C. ELM neural network model is established, first the random weight and threshold value for generating network, sets network parameter, network Parameter setting is as follows:
WijConnection weight between input layer and hidden layer, wherein i=1,2 ..., l, j=1,2 ..., M, l ∈ N+For Hidden neuron number, and remember Wi=(Wi1, Wi2..., WiM)；
BiFor the threshold value of i-th of hidden layer of node；
β is the weight of hidden layer and output layer, wherein βiFor the connection weight of i-th of node of hidden layer and output node layer Value；
The input matrix set target output matrix collection generated in step b is inputted neural network, to net by d. network training Network is trained, network training principle are as follows:
Hidden layer excitation function is sigmoid function, expression formula are as follows:
When input is xpWhen, had according to neural networks principles:
The input of i-th of node of hidden layer is
neti=WiXp+Bi (11)
The output of i-th of node of hidden layer is
si=f (neti)=f (WiXp+Bi) (12)
The output of the output layer of network is
When input is x, remembers that the output matrix of hidden layer is H, have
For output layer, then there is output equation are as follows:
T=H β (15)
The purpose of network training is exactly to find optimal output layer weightThe reality output T of network is set infinitely to approach mesh Mark output y, obtains optimal output layer weight i.e. by the method for seeking minimum norm least-squares solution:
Wherein, H+It is inverse for Moore-penrose,
E. emulation testing utilizes the best initial weights solvedAcquire optimal output layer weightUnder ELM neural network Prediction outputAre as follows:
F. prediction result regressive is reduced into actual prediction as a result, subtracting the equal dimension that step b is obtained with neural network forecast result The accumulated value of innovation sequence obtains actual prediction as a result, obtaining actual prediction value by regressiveAre as follows:
This measurement method is first grouped initial data, and the dimension innovation sequence such as building, equity dimension innovation sequence carries out Grey processing obtains grey treated equal dimension innovation sequence；Then equity dimension innovation sequence is handled, and constitutes input matrix Collection and target output matrix collection, are trained ELM neural network, to obtain prediction of the ELM neural network to grey data As a result；It is restored finally by regressive, obtains the true predictive result of short-term traffic flow.
The beneficial effects of the present invention are: ELM neural network has, training is simple, the fast feature of training speed, to traffic flow Data use gray model accumulation process, reduce traffic flow data randomness, caused by more effective reduction is because of data fluctuation itself Error.Through simulating, verifying, compared to existing some prediction techniques, it to be a kind of effective that the method increase precision of predictions Short-time Traffic Flow Forecasting Methods.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with It obtains other drawings based on these drawings.
Fig. 1 is the architecture diagram of ELM neural network；
Fig. 2 is the comparison diagram of Grey Model result and actual value；
Fig. 3 is the comparison diagram of ELM neural network prediction result and actual value；
Fig. 4 is the comparison diagram using the prediction result and actual value of gray model and BP combination of network model；
Fig. 5 is the prediction result of the built-up pattern with inertial factor and the comparison diagram of actual value；
Fig. 6 is the prediction result of BAYESIAN combined model and the comparison diagram of actual value；
Fig. 7 is the prediction result of grey ELM network and the comparison diagram of actual value；
Fig. 8 is the prediction result Comprehensive Correlation figure of six kinds of prediction models.
Wherein X-axis is time point, and y-axis is the magnitude of traffic flow；Blue curve is predicting traffic flow amount, and red curve is practical hands over Through-current capacity.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on Embodiment in the present invention, those of ordinary skill in the art are obtained every other under that premise of not paying creative labor Embodiment shall fall within the protection scope of the present invention.
A method of the short-term traffic flow forecast based on grey ELM neural network, step include:
A. grey processing is carried out to data, collected data is grouped according to formula (3), it may be assumed that set collected number According to for Q, then
Q=(q1, q2..., qm), (m ∈ N+) (11)
N group, every group of M+1 data are classified as, and are met
N+M=m, (n ∈ N+, M ∈ N+) (2)
For pth group therein, it is denoted as:
After formula (3) such as obtains at the dimension innovation sequence, add up according to formula (4) and (5), the grade reform after add up Cease sequence, it may be assumed that rightIn data carry out grey processing, obtaining one-accumulate sequence is
Wherein
B. input matrix collection and target output matrix collection are generated, to the equal dimension innovation sequence after adding up according to formula (6), formula (7), formula (8) is handled, and obtains the input matrix and target output matrix of network, it may be assumed that is chosenPreceding M be used as ELM nerve net The input of network, the M+1 desired outputs as network then have
To n group data are divided into above, the input matrix collection x and target output matrix collection y for the network being made of it are respectively
X=[X1, X2..., Xn] (7)
Y=[Y1, Y2..., Yn] (8)
If the reality output matrix T of network is
T=[T1, T2..., Tn] (9)；
C. ELM neural network model is established, first the random weight and threshold value for generating network, sets network parameter, network Parameter setting is as follows:
WijConnection weight between input layer and hidden layer, wherein i=1,2 ..., l, j=1,2 ..., M, l ∈ N+For Hidden neuron number, and remember Wi=(Wi1, Wi2..., WiM)；
BiFor the threshold value of i-th of hidden layer of node；
β is the weight of hidden layer and output layer, wherein βiFor the connection weight of i-th of node of hidden layer and output node layer Value；
The input matrix set target output matrix collection generated in step b is inputted neural network, to net by d. network training Network is trained, network training principle are as follows:
Hidden layer excitation function is sigmoid function, expression formula are as follows:
When input is xpWhen, had according to neural networks principles:
The input of i-th of node of hidden layer is
neti=WiXp+Bi (11)
The output of i-th of node of hidden layer is
si=f (neti)=f (WiXp+Bi) (12)
The output of the output layer of network is
When input is x, remembers that the output matrix of hidden layer is H, have
For output layer, then there is output equation are as follows:
T=H β (15)
The purpose of network training is exactly to find optimal output layer weightThe reality output T of network is set infinitely to approach mesh Mark output y, obtains optimal output layer weight i.e. by the method for seeking minimum norm least-squares solution:
Wherein, H+It is inverse for Moore-penrose,
E. emulation testing utilizes the best initial weights solvedAcquire optimal output layer weightUnder ELM neural network Prediction outputAre as follows:
F. prediction result regressive is reduced into actual prediction as a result, subtracting the equal dimension that step b is obtained with neural network forecast result The accumulated value of innovation sequence obtains actual prediction as a result, obtaining actual prediction value by regressiveAre as follows:
This measurement method is first grouped initial data, and the dimension innovation sequence such as building, equity dimension innovation sequence carries out Grey processing obtains grey treated equal dimension innovation sequence；Then equity dimension innovation sequence is handled, and constitutes input matrix Collection and target output matrix collection, are trained ELM neural network, to obtain prediction of the ELM neural network to grey data As a result；It is restored finally by regressive, obtains the true predictive result of short-term traffic flow.
Concrete application citing: use the road Jiaozuo City Pu Ji and people road intersection east-west direction one-way traffic flow for research Object, acquires daily 4 points to 12 points of every 5 minutes magnitudes of traffic flow totally 96 data, and continuous acquisition 4 days.The data of first three days For network training, the 4th day data on flows is for testing grey ELM network model.
Choose M=4, then the neuron number that outputs and inputs of network is 4 and 1 respectively, according to ELM neural networks principles, The number for waiting dimension innovation sequence for choosing the traffic generating that hidden neuron number is acquisition in single day, is 92, then the mould of network Type is 4-92-1；Before training, weight W is first given, and given threshold value is B=1, hidden layer excitation function selects sigmoid function, meter Output matrix is calculated, then training network is predicted with trained network.
As seen in Figure 2, for gray model when carrying out prediction of short-term traffic volume, error is larger, and works as data itself not When steady, prediction result is often very poor, and by the training to historical data, precision of prediction has further to be mentioned ELM neural network It is high；Grey ELM neural network is on the basis of ELM neural network, by adding up to data, so that test data and training number According to otherness it is smaller, effectively increase precision of prediction；Compared with other built-up patterns, precision of prediction also increases.
Grey ELM neural network prediction model can be seen that in precision of prediction to the comparison of six kinds of prediction models by Fig. 8 On to be much better than gray model and ELM neural network prediction model, also superior to existing some built-up patterns, in worst error, Mean absolute error, average relative error aspect are all reduced.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention Within mind and principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.

Claims (2)

1. a kind of method of the short-term traffic flow forecast based on grey ELM neural network, it is characterised in that: its step includes:
A. to data carry out grey processing, collected data are grouped according to formula (3), it may be assumed that set collected data as Q, then
Q=(q1,q2,…,qm),(m∈N+) (1)
N group, every group of M+1 data are classified as, and are met
N+M=m, (n ∈ N+,M∈N+) (2)
For pth group therein, it is denoted as:
After formula (3) such as obtains at the dimension innovation sequence, add up according to formula (4) and (5), the equal reforms breath sequence after being added up Column, it may be assumed that rightIn data carry out grey processing, obtaining one-accumulate sequence is
Wherein
B. input matrix collection and target output matrix collection are generated, to the equal dimension innovation sequence after adding up according to formula (6), formula (7), formula (8) it handles, obtains the input matrix and target output matrix of network, it may be assumed that choosePreceding M as the defeated of ELM neural network Enter, the M+1 desired outputs as network, then has
To n group data are divided into above, the input matrix collection X and target output matrix collection Y for the network being made of it are respectively
X=[X1,X2,…,Xn] (7)
Y=[Y1,Y2,…,Yn] (8)
If the reality output matrix T of network is
T=[T1,T2,…,Tn] (9)；
C. ELM neural network model is established, first the random weight and threshold value for generating network, sets network parameter, network parameter It is provided that
WijConnection weight between input layer and hidden layer, wherein i=1,2 ..., l, j=1,2 ..., M, l ∈ N+For hidden layer Neuron number, and remember Wi=(Wi1,Wi2,…,WiM)；
BiFor the threshold value of i-th of hidden layer of node；
β is the weight of hidden layer and output layer, wherein βiFor the connection weight of i-th of node of hidden layer and output node layer；
D. network training, by the input matrix set target output matrix collection generated in step b input neural network, to network into Row training, network training principle are as follows:
Hidden layer excitation function is sigmoid function, expression formula are as follows:
When input is XpWhen, had according to neural networks principles:
The input of i-th of node of hidden layer is
neti=WiXp+Bi (11)
The output of i-th of node of hidden layer is
si=f (neti)=f (WiXp+Bi) (12)
The output of the output layer of network is
When input is X, remembers that the output matrix of hidden layer is H, have
For output layer, then there is output equation are as follows:
T=H β (15)
The purpose of network training is exactly to find optimal output layer weightThe reality output T of network is set infinitely to approach target output Y obtains optimal output layer weight i.e. by the method for seeking minimum norm least-squares solution:
Wherein, H+It is inverse for Moore-penrose,
E. emulation testing utilizes the optimal output layer weight solvedAcquire optimal output layer weightUnder ELM nerve net The prediction of network exportsAre as follows:
F. prediction result regressive is reduced into actual prediction as a result, being ceased with the equal reforms that neural network forecast result subtracts that step b obtains The accumulated value of sequence obtains actual prediction as a result, obtaining actual prediction value by regressiveAre as follows:
2. the method for the short-term traffic flow forecast according to claim 1 based on grey ELM neural network, feature exist In: initial data is grouped, the dimension innovation sequence such as building, equity dimension innovation sequence carries out grey processing, obtains at grey Equal dimension innovation sequence after reason；Then equity dimension innovation sequence is handled, and constitutes input matrix collection and target output matrix collection, ELM neural network is trained, to obtain ELM neural network to the prediction result of grey data；Also finally by regressive Original obtains the true predictive result of short-term traffic flow.
CN201710068325.7A 2017-02-08 2017-02-08 A method of the short-term traffic flow forecast based on grey ELM neural network Active CN107045785B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710068325.7A CN107045785B (en) 2017-02-08 2017-02-08 A method of the short-term traffic flow forecast based on grey ELM neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710068325.7A CN107045785B (en) 2017-02-08 2017-02-08 A method of the short-term traffic flow forecast based on grey ELM neural network

Publications (2)

Publication Number Publication Date
CN107045785A CN107045785A (en) 2017-08-15
CN107045785B true CN107045785B (en) 2019-10-22

Family

ID=59544180

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710068325.7A Active CN107045785B (en) 2017-02-08 2017-02-08 A method of the short-term traffic flow forecast based on grey ELM neural network

Country Status (1)

Country Link
CN (1) CN107045785B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109000640B (en) * 2018-05-25 2021-09-28 东南大学 Vehicle GNSS/INS integrated navigation method based on discrete grey neural network model
CN109462853B (en) * 2018-11-05 2022-01-14 武汉虹信技术服务有限责任公司 Network capacity prediction method based on neural network model
CN110210648B (en) * 2019-04-30 2023-05-23 南京航空航天大学 Gray long-short term memory network-based control airspace strategic flow prediction method
CN110691396B (en) * 2019-09-29 2020-09-29 南昌航空大学 Unmanned aerial vehicle ad hoc network routing method and system adopting gray Markov model
CN113112011B (en) * 2020-01-13 2024-02-27 中移物联网有限公司 Data prediction method and device
CN111625762B (en) * 2020-04-29 2023-07-25 广东省能源集团贵州有限公司 Fan fault diagnosis method
CN113285896B (en) * 2021-04-30 2023-04-07 南京邮电大学 Time-varying channel prediction method based on stack type ELM
CN113341305A (en) * 2021-05-12 2021-09-03 西安建筑科技大学 Analog circuit fault prediction method based on fusion modeling

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010042973A1 (en) * 2008-10-15 2010-04-22 National Ict Australia Limited Tracking the number of vehicles in a queue
CN102663224A (en) * 2012-03-07 2012-09-12 吉首大学 Comentropy-based integrated prediction model of traffic flow
CN102693633A (en) * 2012-06-07 2012-09-26 浙江大学 Short-term traffic flow weighted combination prediction method
CN103258243A (en) * 2013-04-27 2013-08-21 杭州电子科技大学 Tube explosion predicting method based on grey neural network
CN105469611A (en) * 2015-12-24 2016-04-06 大连理工大学 Short-term traffic flow prediction model method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010042973A1 (en) * 2008-10-15 2010-04-22 National Ict Australia Limited Tracking the number of vehicles in a queue
CN102663224A (en) * 2012-03-07 2012-09-12 吉首大学 Comentropy-based integrated prediction model of traffic flow
CN102693633A (en) * 2012-06-07 2012-09-26 浙江大学 Short-term traffic flow weighted combination prediction method
CN103258243A (en) * 2013-04-27 2013-08-21 杭州电子科技大学 Tube explosion predicting method based on grey neural network
CN105469611A (en) * 2015-12-24 2016-04-06 大连理工大学 Short-term traffic flow prediction model method

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
Optimized and meta-optimized neural networks for short-term traffic flow prediction: A genetic approach;E I Vlahogianni Et al.;《TRANSPORTATION RESEARCH PART C》;20050420;第3卷(第13期);第211-234页 *
Vehicle detection in driving simulation using extreme learning machine;Wentao Zhu Et al.;《Neurocomputing》;20131025;第128卷;第160-165页 *
基于ELM算法的短时交通流预测研究;季雪美等;《青岛大学学报（工程技术版）》;20151231;第30卷(第4期);第58-61页 *
基于组合模型的短时交通流预测;钱伟等;《计算机仿真》;20150215;第32卷(第2期);第175-178页 *

Also Published As

Publication number Publication date
CN107045785A (en) 2017-08-15

Similar Documents

Publication Publication Date Title
CN107045785B (en) A method of the short-term traffic flow forecast based on grey ELM neural network
Li et al. Prediction for tourism flow based on LSTM neural network
Zhu et al. Traffic volume forecasting based on radial basis function neural network with the consideration of traffic flows at the adjacent intersections
Huang et al. Neural network modeling of salinity variation in Apalachicola River
Kisi et al. Forecasting daily lake levels using artificial intelligence approaches
CN108205889A (en) Freeway traffic flow Forecasting Methodology based on convolutional neural networks
Tayyab et al. Discharge forecasting by applying artificial neural networks at the Jinsha River Basin, China
Yuan et al. Parameter identification of nonlinear Muskingum model with backtracking search algorithm
CN106448151A (en) Short-time traffic flow prediction method
CN106789214A (en) It is a kind of based on the just remaining pair network situation awareness method and device of string algorithm
CN107704973A (en) Water level prediction method based on neutral net Yu local Kalman filtering mixed model
CN106803135A (en) The Forecasting Methodology and device of a kind of photovoltaic power generation system output power
Tian et al. A network traffic hybrid prediction model optimized by improved harmony search algorithm
Chen et al. CRML: A convolution regression model with machine learning for hydrology forecasting
Dong et al. Uncertainty-based multi-objective decision making with hierarchical reliability analysis under water resources and environmental constraints
Zhang et al. Prediction of Tourist Quantity Based on RBF Neural Network.
CN109117998B (en) Multi-agent configuration method
Zhang et al. Road section traffic flow prediction method based on the traffic factor state network
Bin et al. A forecasting method of short-term electric power load based on BP neural network
CN113707338A (en) Scenic spot epidemic situation risk prediction and current limiting method, device, equipment and storage medium
Xiaojian et al. A traffic flow forecasting model based on BP neural network
He et al. Short-term power load forecasting based on self-adapting PSO-BP neural network model
Li et al. A short-term power load forecasting method based on BP neural network
CN114021847A (en) Short-term power load prediction method and system based on ISGU (inverse transformation and aggregation) hybrid model
Wang et al. Hybrid neural network modeling for multiple intersections along signalized arterials-current situation and some new results

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