CN114091579A - Urban rail transit passenger flow early warning system and method - Google Patents
Urban rail transit passenger flow early warning system and method Download PDFInfo
- Publication number
- CN114091579A CN114091579A CN202111292372.2A CN202111292372A CN114091579A CN 114091579 A CN114091579 A CN 114091579A CN 202111292372 A CN202111292372 A CN 202111292372A CN 114091579 A CN114091579 A CN 114091579A
- Authority
- CN
- China
- Prior art keywords
- passenger flow
- urban rail
- rail transit
- early warning
- data set
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 74
- 238000012843 least square support vector machine Methods 0.000 claims abstract description 36
- 230000002068 genetic effect Effects 0.000 claims abstract description 32
- 230000006870 function Effects 0.000 claims description 40
- 238000012549 training Methods 0.000 claims description 30
- 238000012706 support-vector machine Methods 0.000 claims description 24
- 238000005457 optimization Methods 0.000 claims description 23
- 239000000427 antigen Substances 0.000 claims description 15
- 108091007433 antigens Proteins 0.000 claims description 15
- 102000036639 antigens Human genes 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 12
- 230000009977 dual effect Effects 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 239000002245 particle Substances 0.000 description 3
- 241001123248 Arma Species 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000001364 causal effect Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 241000282461 Canis lupus Species 0.000 description 1
- 241000544061 Cuculus canorus Species 0.000 description 1
- 241000255581 Drosophila <fruit fly, genus> Species 0.000 description 1
- 241000728173 Sarima Species 0.000 description 1
- 241000255588 Tephritidae Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000005291 chaos (dynamical) Methods 0.000 description 1
- 238000010367 cloning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 108090000623 proteins and genes Proteins 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- 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"
-
- G06Q50/40—
Abstract
The invention discloses an urban rail transit passenger flow early warning system and a method, wherein the method comprises the following steps: acquiring a multidimensional information data set to be detected of urban rail transit, and matching the multidimensional information data set to be detected with a plurality of preset passenger flow modes to obtain a passenger flow mode corresponding to the multidimensional information data set to be detected; inputting the multidimensional information data set to be detected into an urban rail transit passenger flow early warning model corresponding to a passenger flow mode, and outputting predicted urban rail transit passenger flow corresponding to the multidimensional information data set to be detected through the urban rail transit passenger flow early warning model; the urban rail transit passenger flow early warning model adopts an optimized least square support vector machine based on an immune genetic algorithm; and determining the early warning level of the urban rail according to the predicted urban rail traffic passenger flow. The method is based on the immune genetic algorithm to optimize the least square support vector machine algorithm to generate the model, and can accurately predict the urban rail transit passenger flow by applying the model, thereby avoiding the generation of the peak passenger flow treading risk.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an urban rail passenger flow early warning system and method.
Background
The short-term urban rail transit passenger flow prediction is used as an important link for management and control of an urban rail transit system, provides decision basis for urban rail transit real-time operation and passenger flow organization, and has very important practical significance for improving traffic management service level and control capacity. The short-term passenger flow prediction can be divided into three prediction modes of linear prediction, nonlinear prediction and combined prediction according to data characteristics. However, the trend characteristics of short-term passenger flow are not obvious compared with medium-and-long-term passenger flow, researchers often need to perform joint auxiliary prediction on the short-term passenger flow by means of other relevant real-time data, such as factors of weather change, holidays, major activities, surrounding traffic conditions and the like, urban rail passenger flow prediction models under the multi-modal data often need data support of multiple platforms, although prediction accuracy is improved, prediction efficiency is low, and researchers can easily ignore timeliness of short-term prediction. The multi-mode prediction model is suitable for medium and long-term prediction, auxiliary suggestions are provided for urban rail transit planning and construction, problems of operation cost rise, long prediction time and the like can be caused due to the fact that multi-mode data prediction needs support of a plurality of platforms, and potential passenger flow trampling risks are caused due to the fact that short-term urban rail transit passenger flow prediction is inaccurate due to the fact that multi-mode data combined decision cannot respond quickly.
Thus, there is still a need for improvement and development of the prior art.
Disclosure of Invention
The invention aims to solve the technical problems that in the prior art, a multi-mode data combined decision cannot respond quickly, so that short-term urban rail transit passenger flow prediction is not accurate, and potential passenger flow trampling risks are caused.
The technical scheme adopted by the invention for solving the problems is as follows:
in a first aspect, an embodiment of the present invention provides an urban rail transit passenger flow early warning method, where the method includes:
acquiring a multidimensional information data set to be detected of urban rail transit, and matching the multidimensional information data set to be detected with a plurality of preset passenger flow modes to obtain a passenger flow mode corresponding to the multidimensional information data set to be detected;
inputting the multi-dimensional information dataset to be detected into an urban rail transit passenger flow early warning model corresponding to the passenger flow mode, and outputting predicted urban rail transit passenger flow corresponding to the multi-dimensional information dataset to be detected through the urban rail transit passenger flow early warning model; the urban rail transit passenger flow early warning model adopts an optimized least square support vector machine based on an immune genetic algorithm;
and determining the early warning level of the urban rail according to the predicted urban rail traffic passenger flow.
In one implementation, the plurality of passenger flow patterns are generated by:
acquiring a multi-dimensional information historical data set of urban rail transit, and performing optimal clustering grouping on the multi-dimensional information historical data set based on a BWP-KMEANS clustering algorithm to obtain a plurality of passenger flow modes.
In one implementation, the training process of the urban rail transit passenger flow early warning model corresponding to the passenger flow mode is as follows:
acquiring a training data set corresponding to each passenger flow mode; the training data set comprises a training multidimensional information data set corresponding to the passenger flow mode and real urban rail transit passenger flow corresponding to the passenger flow mode;
inputting the training multidimensional information data set corresponding to the passenger flow mode into an initial model corresponding to the passenger flow mode to obtain model output data corresponding to the passenger flow mode;
performing mean square error operation on the model output data and the real urban rail transit passenger flow to obtain a mean square error value corresponding to the passenger flow mode;
and training the initial model based on the mean square error value to obtain an urban rail transit passenger flow early warning model corresponding to the passenger flow mode.
In one implementation, the initial model is generated by:
constructing input data and target data of urban rail transit;
based on the input data, optimizing the target data by adopting an optimization function of a least square support vector machine algorithm to obtain a regression prediction function of the least square support vector machine;
and optimizing the regression prediction function of the least square support vector machine by adopting an immune genetic algorithm to obtain an initial model.
In one implementation, the optimizing the target data by using an optimization function of a least squares support vector machine algorithm based on the input data to obtain a regression prediction function of the least squares support vector machine includes:
acquiring parameters of the target data;
converting the optimization function into a dual space based on a Lagrangian algorithm;
and performing matrix mixing operation on the parameters and the input data based on a dual space to obtain a least square support vector machine regression prediction function.
In one implementation, the optimizing the least squares support vector machine regression prediction function by using an immune genetic algorithm to obtain an initial model specifically includes:
setting a least squares support vector machine antigen and antibody population based on the input data;
performing immune algorithm iteration on the antigen and the antibodies in the antibody population to obtain the information entropy between the antibodies and the affinity between the antibodies and the antigen;
ranking the affinities;
and based on a genetic algorithm, carrying out coding operation on a plurality of antibodies corresponding to the affinity and ranked in the front to obtain an initial model.
In one implementation, the determining the early warning level of the urban rail according to the prediction of the urban rail transit passenger flow includes:
when the predicted urban rail transit passenger flow exceeds a preset threshold value, setting the early warning level of the urban rail as a high risk;
and when the predicted urban rail transit passenger flow is lower than the preset threshold value, setting the early warning level of the urban rail as low risk.
In a second aspect, an embodiment of the present invention further provides an urban rail transit passenger flow early warning system, where the system includes: the passenger flow mode acquisition module is used for acquiring a multi-dimensional information data set to be detected of the urban rail transit, and matching the multi-dimensional information data set to be detected with a plurality of preset passenger flow modes to obtain a passenger flow mode corresponding to the multi-dimensional information data set to be detected;
the predicted urban rail transit passenger flow determining module is used for inputting the multi-dimensional information data set to be detected into an urban rail transit passenger flow early warning model corresponding to the passenger flow mode and outputting the predicted urban rail transit passenger flow corresponding to the multi-dimensional information data set to be detected through the urban rail transit passenger flow early warning model; the urban rail transit passenger flow early warning model adopts an optimized least square support vector machine based on an immune genetic algorithm;
and the early warning level determining module of the urban rail is used for determining the early warning level of the urban rail according to the predicted urban rail traffic passenger flow.
In a third aspect, an embodiment of the present invention further provides an intelligent terminal, including a memory, and one or more programs, where the one or more programs are stored in the memory, and configured to be executed by one or more processors includes a processor configured to execute the method for early warning of passenger flow in urban rail transit.
In a fourth aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium, where instructions in the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the urban rail transit passenger flow early warning method as described in any one of the above.
The invention has the beneficial effects that: firstly, acquiring a multidimensional information data set to be detected of urban rail transit, and matching the multidimensional information data set to be detected with a plurality of preset passenger flow modes to obtain a passenger flow mode corresponding to the multidimensional information data set to be detected; then inputting the multi-dimensional information data set to be detected into an urban rail transit passenger flow early warning model corresponding to the passenger flow mode, and outputting predicted urban rail transit passenger flow corresponding to the multi-dimensional information data set to be detected through the urban rail transit passenger flow early warning model; the urban rail transit passenger flow early warning model adopts an optimized least square support vector machine based on an immune genetic algorithm; finally, determining the early warning level of the urban rail according to the predicted urban rail traffic passenger flow; therefore, the model is generated based on the least square support vector machine algorithm optimized by the immune genetic algorithm, and can be used for accurately predicting urban rail transit passenger flow, so that the generation of the peak passenger flow treading risk is avoided.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of an urban rail transit passenger flow early warning method provided by an embodiment of the present invention
Fig. 2 is a schematic block diagram of an urban rail transit passenger flow early warning system provided in an embodiment of the present invention.
Fig. 3 is a schematic block diagram of an internal structure of an intelligent terminal according to an embodiment of the present invention.
Detailed Description
The invention discloses an urban rail transit passenger flow early warning system and a method, and in order to make the purpose, technical scheme and effect of the invention clearer and clearer, the invention is further described in detail below by referring to the attached drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or wirelessly coupled. As used herein, the term "and/or" includes all or any element and all combinations of one or more of the associated listed items.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
In the prior art, in recent years, researchers at home and abroad mostly use various algorithms to improve the existing models or combined models for optimization, for example, a deep learning method and an optimization algorithm are combined, and the BP network is optimized by introducing a self-adaptive mutation operator to improve the optimization capability of a particle swarm algorithm; introducing an improved genetic algorithm to optimize a wavelet neural network for predicting a traffic flow time sequence; performing phase space deconstruction on the time sequence through a chaos theory, and optimizing a BP neural network by using a cuckoo algorithm for prediction; optimizing the initial weight and wavelet factor of the WNN neural network by utilizing a wolf algorithm, and avoiding the defect of local optimization; an echo state network optimized based on an improved fruit fly optimization algorithm (ESN-IFOA) is provided; the prediction algorithm optimizes the weight of the neural network through the performance of the optimization algorithm or combines the weight with the weight to achieve the effect of rapid convergence, improves the prediction accuracy of traffic flow, has high requirements on data by deep learning, and easily causes the situations of low prediction speed and overfitting along with the increase of the number of network layers and the data.
The support vector machine has good generalization capability when small sample prediction is carried out, so that the method is suitable for the field of traffic flow prediction, wherein the least square support vector machine is a standard SVM improved algorithm, the training speed of the model can be effectively improved, and the prediction accuracy of the model can be improved by optimizing SVM parameters by adopting algorithms such as particle swarm, ant colony, genetic algorithm and the like. When the LSSVM is used for prediction, the accuracy of the prediction is high in sensitivity to parameter selection, so that a proper parameter combination is an important factor for improving the performance.
Based on this, many scholars have also conducted a series of studies on the optimization of parameters of SVMs. A series of prediction modes for optimizing parameters of the support vector machine are provided, wherein an LSSVM is optimized by using an immune algorithm, the LSSVM is optimized by using a drosophila algorithm, the support vector machine is optimized by using a particle swarm, the optimization is performed by using an ant colony algorithm, and the like.
However, since multi-modal data acquisition needs to be supported by multiple platforms, it is difficult to acquire multi-modal data and perform joint prediction in actual production, and therefore, relevant experts propose a corresponding feature extraction algorithm to perform feature extraction on single-time sequence data and then perform combined prediction with a related machine learning algorithm. For example: performing wavelet decomposition on passenger flow to obtain characteristics, performing characteristic prediction through an SVM (support vector machine), and finally re-fusing the predicted characteristics to obtain passenger flow; an EMD-BPN mixed prediction method combining Empirical Mode Decomposition (EMD) and a back propagation neural network (BPN) is provided, and data after modal classification is used as input and is put into the BP neural network for training prediction; time series data were analyzed by SARIMA. Reversely calculating Gaussian white noise to predict indexes; decomposing data into a linear part and a nonlinear part, and respectively performing training prediction through an ARIMA and a wavelet neural network; the problem of sample size is solved, wavelet packet algorithm is utilized to decompose the time sequence and have enough single branch, and phase space reconstruction is carried out on different frequency band sequences, so that LSTM model construction is completed, and the purpose of prediction is achieved; a prediction method model combining a denoising scheme and a support vector machine is provided to improve prediction accuracy, but in order to obtain enough features, the method needs a large amount of data support to perform global feature analysis and extraction, but passenger flow data is used as a special time sequence data type, and passenger flows in different periods have different features, so that the features are not obvious during global feature extraction, and passenger flow patterns need to be classified. In conclusion, due to the fact that multi-mode data combined decision-making cannot respond quickly, short-term urban rail transit passenger flow prediction is inaccurate, and potential passenger flow tread risks are caused
In order to solve the problems in the prior art, the embodiment provides an urban rail traffic early warning system and an urban rail traffic early warning method, a model is generated by the method based on a least square support vector machine algorithm optimized by an immune genetic algorithm, the urban rail traffic passenger flow can be accurately predicted by applying the model, and the generation of a peak passenger flow treading risk is avoided. When the method is specifically implemented, firstly, a multi-dimensional information data set to be detected of the urban rail transit is obtained, and the multi-dimensional information data set to be detected is matched with a plurality of preset passenger flow modes to obtain a passenger flow mode corresponding to the multi-dimensional information data set to be detected; then inputting the multi-dimensional information data set to be detected into an urban rail transit passenger flow early warning model corresponding to the passenger flow mode, and outputting predicted urban rail transit passenger flow corresponding to the multi-dimensional information data set to be detected through the urban rail transit passenger flow early warning model; the urban rail transit passenger flow early warning model adopts an optimized least square support vector machine based on an immune genetic algorithm; and finally, determining the early warning level of the urban rail according to the predicted urban rail traffic passenger flow.
Exemplary method
The embodiment provides an urban rail transit passenger flow early warning method which can be applied to an intelligent terminal for data processing. As shown in fig. 1 in detail, the method includes:
s100, acquiring a multi-dimensional information data set to be detected of urban rail transit, and matching the multi-dimensional information data set to be detected with a plurality of preset passenger flow modes to obtain a passenger flow mode corresponding to the multi-dimensional information data set to be detected;
specifically, the characteristics are not obvious when the global characteristics of the urban rail transit multidimensional information data set to be detected are extracted, the passenger flow modes are classified firstly, so that the characteristics of the classified urban rail transit multidimensional information data set to be detected are obvious, the characteristics are conveniently extracted through the models, each passenger flow mode corresponds to one model, if N passenger flow modes exist, N models exist, and therefore the urban rail transit multidimensional information data set to be detected and the passenger flow modes need to be matched. The specific process is as follows: and comparing the similarity of the initial sampling points in the multi-dimensional information data set to be detected of the urban rail transit with the sampling points in a plurality of passenger flow modes, and selecting a group of passenger flow modes with the highest similarity as the passenger flow modes corresponding to the multi-dimensional information data set to be detected.
In one implementation, the generation of the plurality of passenger flow patterns is as follows: acquiring a multi-dimensional information historical data set of urban rail transit, and performing optimal clustering grouping on the multi-dimensional information historical data set based on a BWP-KMEANS clustering algorithm to obtain a plurality of passenger flow modes.
In practice, a large amount of historical data is acquired, in this embodiment, a multi-dimensional information historical data set of the urban rail transit is acquired, the data is preprocessed, and abnormal data is removed, for example: this document acquires passenger flow data for station a for 17 hours per day (92 days in total for 6-8 months), and proposes abnormal data (blank data, change in station operating time due to activity or overhaul). Finally, 84 days of effective data are obtained. Then, performing KMEANS clustering on the data by using a BWP index to obtain a plurality of passenger flow modes, such as: and taking the clustering group when the K value is maximum as an optimal group, wherein the K value is selected to be in a range of 2-10. When K is 4, the best clustering group is obtained, so that the station is determined to have 4 passenger flow modes.
After obtaining the multidimensional information dataset to be measured, the following steps can be performed as shown in fig. 1: s200, inputting the multi-dimensional information data set to be detected into an urban rail transit passenger flow early warning model corresponding to the passenger flow mode, and outputting predicted urban rail transit passenger flow corresponding to the multi-dimensional information data set to be detected through the urban rail transit passenger flow early warning model; the urban rail transit passenger flow early warning model adopts an optimized least square support vector machine based on an immune genetic algorithm;
specifically, the multi-dimensional information dataset to be detected has corresponding passenger flow modes, and each passenger flow mode has a corresponding urban rail transit passenger flow early warning model, so that the multi-dimensional information dataset to be detected is firstly input into the urban rail transit passenger flow early warning model corresponding to the passenger flow mode corresponding to the multi-dimensional information dataset to be detected, and therefore accurate predicted urban rail transit passenger flow corresponding to the multi-dimensional information dataset to be detected can be obtained. The urban rail transit passenger flow early warning model is accurate because the urban rail transit passenger flow early warning model adopts a least square support vector machine algorithm optimized based on an immune genetic algorithm, the least square support vector machine algorithm is optimized after the immune algorithm and the genetic algorithm are fused, the method combines the advantage of global optimization of the immune algorithm and the characteristic of rapid convergence of the genetic algorithm, and the aim of improving the precision of predicting the urban rail transit passenger flow is achieved through a rapid iteration process.
In one implementation, the training process of the urban rail transit passenger flow early warning model corresponding to the passenger flow mode is as follows: acquiring a training data set corresponding to each passenger flow mode; the training data set comprises a training multidimensional information data set corresponding to the passenger flow mode and real urban rail transit passenger flow corresponding to the passenger flow mode; inputting the training multidimensional information data set corresponding to the passenger flow mode into an initial model corresponding to the passenger flow mode to obtain model output data corresponding to the passenger flow mode; performing mean square error operation on the model output data and the real urban rail transit passenger flow to obtain a mean square error value corresponding to the passenger flow mode; and training the initial model based on the mean square error value to obtain an urban rail transit passenger flow early warning model corresponding to the passenger flow mode.
Specifically, each passenger flow mode corresponds to one model, so the training data set also corresponds to the passenger flow mode, the N passenger flow modes correspond to the N training data sets, and the training data set corresponding to each passenger flow mode is obtained first; the training data set comprises a training multidimensional information data set corresponding to the passenger flow mode and real urban rail transit passenger flow corresponding to the passenger flow mode; real urban rail transit passenger flow is a training target value, and mean square error operation is carried out on the model output data and the real urban rail transit passenger flow to obtain a mean square error value corresponding to the passenger flow mode; and training the initial model based on the mean square error value to obtain an urban rail transit passenger flow early warning model corresponding to the passenger flow mode. In the iterative training process, when the mean square error value is smaller than a preset threshold (such as 0.000001), stopping iteration to obtain the urban rail transit passenger flow early warning model corresponding to the passenger flow mode.
In one implementation, the neural network model and the initial model in the invention are generated based on a function, and the specific generation mode is as follows: constructing input data and target data of urban rail transit; based on the input data, optimizing the target data by adopting an optimization function of a least square support vector machine algorithm to obtain a regression prediction function of the least square support vector machine; and optimizing the regression prediction function of the least square support vector machine by adopting an immune genetic algorithm to obtain an initial model.
Specifically, firstly, input data and target data of urban rail transit are constructed; wherein the target data corresponds to input data. Such as: among the traffic short-term prediction models, the support vector machine, the RBF neural network, the ARMA moving average, and the like are relatively good models in terms of prediction performance. The least square vector machine is used as one of improved optimization models of the support vector machine, after inequality constraints are changed into equality constraints, a quadratic programming problem is directly converted into a linear programming problem, the calculation complexity is greatly reduced, the solving speed and precision of the model are improved, and the approximation is faster and more accurate to a certain degree through single-attribute prediction compared with the approximation by using methods such as ARMA and the like. The main idea of the model is as follows:
given a training set:
{(xi,yi)}i＝1,2,…,l,xi∈Rn,yi∈R}
in the formula xiFor input data, for predicting an n-dimensional dataset, yiAnd L is the number of samples sampled, wherein the number is the target data, namely the predicted value of the short-time passenger flow. And then based on the input data, optimizing the target data by adopting an optimization function of a least square support vector machine algorithm to obtain a regression prediction function of the least square support vector machine.
In this embodiment, according to the GRANGE causal relationship, it is found that the passenger flow in the first 4 hours and the passenger flow in the 5 th hour have the GRANGE causal relationship, so that the passenger flow in the first 4 hours is used as input, the passenger flow in the 5 th hour is used as output, and a nonlinear change f (x) is selected to convert the input into the output, so as to obtain the regression prediction function f (x).
In order to obtain a least squares support vector machine regression prediction function, the step of optimizing the target data by using an optimization function of a least squares support vector machine algorithm based on the input data to obtain the least squares support vector machine regression prediction function comprises the following steps: acquiring parameters of the target data; converting the optimization function into a dual space based on a Lagrangian algorithm; and performing matrix mixing operation on the parameters and the input data based on a dual space to obtain a least square support vector machine regression prediction function.
Specifically, an optimization function of a least squares support vector machine is adopted to optimize the target value:
in the formula: w is a weight vector; b is a deviation; e is the error between the true value and the estimated value; gamma is a punishment factor, the punishment force and the model accuracy can be adjusted by increasing and reducing the value of gamma, when gamma is too small, the prediction accuracy is reduced, and when gamma is too large, overfitting is caused, so that the generalization capability is insufficient; l is the sample volume; e.g. of the typeiIs the i-th component of the error term; y isiThe ith sample value which is an output value; phi (x)i) Mapping the sample data from the low-dimensional space to a kernel function corresponding to the high-dimensional space; x is the number ofiThe input ith sample value. Firstly, acquiring parameters of the target data; the parameters are as follows: w (weight vector); b (deviation); e (error between true and predicted values), then converting the optimization function to dual space based on Lagrangian algorithm; the dual space is the following equation:
in the formula, alpha is Lagrange multiplier; alpha is alphaiIs the ith component of the lagrange multiplier. And finally, performing matrix mixing operation on the parameters and the input data based on a dual space to obtain a regression prediction function of a least square support vector machine, wherein the regression prediction function comprises the following steps:
for w, b and e respectivelyi、αiTaking the derivative and let it be 0, yields the following formula:
elimination of omega and eiRewritten as a matrix form:
in the formula: 1l＝[1 1 … 1](ii) a K is a kernel function; i is an identity matrix; b ═ b1 b2 … bl]α＝[α1 α2 … αl]；Y＝[y1 y2 … yl]。
According to the matrix equation, alpha and b can be obtained, and finally, the function of the LSSVM regression prediction is obtained as follows:
in the formula: k (x, x)i) For the kernel function, we use the gaussian (Gauss) radial basis kernel function, whose functional form is as follows:
in the formula: sigma is the bandwidth of a Gaussian kernel, the performance of LSSVM regression prediction is greatly influenced, the smaller sigma is, the more sensitive the error tolerance is, the correlation among sample data points is weakened, the machine learning process is relatively complex, and the model popularization capability is reduced; the larger the sigma is, the stronger the correlation between sample data points is, the machine is easy to generate an over-learning phenomenon, and the prediction precision cannot be guaranteed.
It follows that the test results of LSSVM depend mainly on the specific model parameters γ and σ.
After the least square support vector machine regression prediction function is obtained, the least square support vector machine regression prediction function is optimized by adopting the immune genetic algorithm to obtain an initial model, namely the model parameters gamma and sigma are adjusted by combining the good optimization and the rapid convergence of the immune genetic algorithm. Correspondingly, the specific process of optimizing the regression prediction function of the least square support vector machine by adopting the immune genetic algorithm to obtain the initial model comprises the following steps: setting a least squares support vector machine antigen and antibody population based on the input data; performing immune algorithm iteration on the antigen and the antibodies in the antibody population to obtain the information entropy between the antibodies and the affinity between the antibodies and the antigen; ranking the affinities; and based on a genetic algorithm, carrying out coding operation on a plurality of antibodies corresponding to the affinity and ranked in the front to obtain an initial model. The specific process in this embodiment includes the following steps:
step1, reading the matching data, setting LSSVM antigen and antibody group (target problem and initial solution), and randomly generating N antibodies and M memory banks to form an initial antibody group according to the binary coding rule, wherein the antibody group is a random combination;
step 2: performing immune algorithm iteration on the antibody and the antigen, controlling the concentration of the antibody by calculating the affinity of the antibody and the antigen (mean square error (MSE) is taken as an index), and calculating the difference value between a target value and a predicted value and the affinity between the antigens, wherein the calculation formula of the affinity is as follows:
in the formula: q (x)i) -affinity between antibody antigens; q (x)i,xj) -affinity between antibodies;
e-entropy of information between antibodies.
Step 3: sorting the affinities, selecting m antibodies with the highest affinities, and performing cloning operation;
step 4: by calculating the expected value e of antibody vyEliminating antibodies of low expected value, i.e., promoting high affinity, low density individuals. The calculation formula is as follows:
cv＝-qk/N,ev＝Q(xi)/cv (8)
wherein c isvFor antibody density, N is the population number, qkRepresents an antibody having a large affinity for antibody k.
Step 5: according to the affinity of different antibodies and antigens and a genetic algorithm calculation mode, exchanging and changing gene sequences according to a certain cross variation probability to generate N new antibodies;
step 6: judging the convergence condition and the iteration times of the model, if the convergence condition or the maximum times is reached, returning to the end, otherwise, entering step 2.
After the predicted urban rail transit passenger flow is obtained, the following steps as shown in fig. 1 can be executed: s300, determining the early warning level of the urban rail according to the predicted urban rail traffic passenger flow. Correspondingly, the step of determining the early warning level of the urban rail according to the predicted urban rail traffic passenger flow comprises the following steps:
s301, when the predicted urban rail transit passenger flow exceeds a preset threshold value, setting the early warning level of the urban rail as a high risk;
s302, when the predicted urban rail transit passenger flow is lower than the preset threshold value, the early warning level of the urban rail is set to be low risk.
Specifically, the preset threshold is specifically determined according to the bearing capacity of the actual urban rail transit equipment, for example, 1000 ten thousand daily average passenger flows of the beijing subway are used as the preset threshold of the beijing subway. And once the predicted urban rail transit passenger flow exceeds the preset threshold value, setting the early warning level of the urban rail to be high risk, and taking corresponding safety measures such as current limiting and the like so as to avoid the generation of the risk of trampling by the passenger flow at the peak. Certainly, when the predicted urban rail transit passenger flow is lower than the preset threshold, the early warning level of the urban rail transit is set to be low risk, and the passenger flow enters and exits normally.
Exemplary device
As shown in fig. 2, an embodiment of the present invention provides an urban rail transit passenger flow early warning system, including: the passenger flow mode acquisition module is used for acquiring a multi-dimensional information data set to be detected of the urban rail transit, and matching the multi-dimensional information data set to be detected with a plurality of preset passenger flow modes to obtain a passenger flow mode corresponding to the multi-dimensional information data set to be detected;
the predicted urban rail transit passenger flow determining module is used for inputting the multi-dimensional information data set to be detected into an urban rail transit passenger flow early warning model corresponding to the passenger flow mode and outputting the predicted urban rail transit passenger flow corresponding to the multi-dimensional information data set to be detected through the urban rail transit passenger flow early warning model; the urban rail transit passenger flow early warning model adopts an optimized least square support vector machine based on an immune genetic algorithm;
and the early warning level determining module of the urban rail is used for determining the early warning level of the urban rail according to the predicted urban rail traffic passenger flow.
Based on the above embodiment, the present invention further provides an intelligent terminal, and a schematic block diagram thereof may be as shown in fig. 3. The intelligent terminal comprises a processor, a memory, a network interface, a display screen and a temperature sensor which are connected through a system bus. Wherein, the processor of the intelligent terminal is used for providing calculation and control capability. The memory of the intelligent terminal comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the intelligent terminal is used for being connected and communicated with an external terminal through a network. The computer program is executed by a processor to realize the urban rail transit passenger flow early warning method. The display screen of the intelligent terminal can be a liquid crystal display screen or an electronic ink display screen, and the temperature sensor of the intelligent terminal is arranged inside the intelligent terminal in advance and used for detecting the operating temperature of internal equipment.
It will be understood by those skilled in the art that the schematic diagram in fig. 3 is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation to the intelligent terminal to which the solution of the present invention is applied, and a specific intelligent terminal may include more or less components than those shown in the figure, or combine some components, or have different arrangements of components.
In one embodiment, an intelligent terminal is provided that includes a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for: acquiring a multidimensional information data set to be detected of urban rail transit, and matching the multidimensional information data set to be detected with a plurality of preset passenger flow modes to obtain a passenger flow mode corresponding to the multidimensional information data set to be detected;
inputting the multi-dimensional information dataset to be detected into an urban rail transit passenger flow early warning model corresponding to the passenger flow mode, and outputting predicted urban rail transit passenger flow corresponding to the multi-dimensional information dataset to be detected through the urban rail transit passenger flow early warning model; the urban rail transit passenger flow early warning model adopts an optimized least square support vector machine based on an immune genetic algorithm;
and determining the early warning level of the urban rail according to the predicted urban rail traffic passenger flow.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, databases, or other media used in embodiments provided herein may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
In summary, the invention discloses an urban rail passenger flow early warning system and method, and the method comprises the following steps: acquiring a multidimensional information data set to be detected of urban rail transit, and matching the multidimensional information data set to be detected with a plurality of preset passenger flow modes to obtain a passenger flow mode corresponding to the multidimensional information data set to be detected; inputting the multidimensional information data set to be detected into an urban rail transit passenger flow early warning model corresponding to a passenger flow mode, and outputting predicted urban rail transit passenger flow corresponding to the multidimensional information data set to be detected through the urban rail transit passenger flow early warning model; the urban rail transit passenger flow early warning model adopts an optimized least square support vector machine based on an immune genetic algorithm; and determining the early warning level of the urban rail according to the predicted urban rail traffic passenger flow. The method is based on the immune genetic algorithm to optimize the least square support vector machine algorithm to generate the model, and can accurately predict the urban rail transit passenger flow by applying the model, thereby avoiding the generation of the peak passenger flow treading risk.
Based on the above embodiments, the present invention discloses a method for urban rail transit passenger flow early warning, it should be understood that the application of the present invention is not limited to the above examples, and it will be obvious to those skilled in the art that modifications and changes can be made according to the above description, and all such modifications and changes are intended to fall within the scope of the appended claims.
Claims (10)
1. An urban rail transit passenger flow early warning method is characterized by comprising the following steps:
acquiring a multidimensional information data set to be detected of urban rail transit, and matching the multidimensional information data set to be detected with a plurality of preset passenger flow modes to obtain a passenger flow mode corresponding to the multidimensional information data set to be detected;
inputting the multi-dimensional information dataset to be detected into an urban rail transit passenger flow early warning model corresponding to the passenger flow mode, and outputting predicted urban rail transit passenger flow corresponding to the multi-dimensional information dataset to be detected through the urban rail transit passenger flow early warning model; the urban rail transit passenger flow early warning model adopts an optimized least square support vector machine based on an immune genetic algorithm;
and determining the early warning level of the urban rail according to the predicted urban rail traffic passenger flow.
2. The urban rail transit passenger flow early warning method according to claim 1, wherein the generation manner of the plurality of passenger flow patterns is as follows:
acquiring a multi-dimensional information historical data set of urban rail transit, and performing optimal clustering grouping on the multi-dimensional information historical data set based on a BWP-KMEANS clustering algorithm to obtain a plurality of passenger flow modes.
3. The urban rail transit passenger flow early warning method according to claim 2, wherein the training process of the urban rail transit passenger flow early warning model corresponding to the passenger flow mode is as follows:
acquiring a training data set corresponding to each passenger flow mode; the training data set comprises a training multidimensional information data set corresponding to the passenger flow mode and real urban rail transit passenger flow corresponding to the passenger flow mode;
inputting the training multidimensional information data set corresponding to the passenger flow mode into an initial model corresponding to the passenger flow mode to obtain model output data corresponding to the passenger flow mode;
performing mean square error operation on the model output data and the real urban rail transit passenger flow to obtain a mean square error value corresponding to the passenger flow mode;
and training the initial model based on the mean square error value to obtain an urban rail transit passenger flow early warning model corresponding to the passenger flow mode.
4. The urban rail transit passenger flow early warning method according to claim 3, wherein the generation mode of the initial model is as follows:
constructing input data and target data of urban rail transit;
based on the input data, optimizing the target data by adopting an optimization function of a least square support vector machine algorithm to obtain a regression prediction function of the least square support vector machine;
and optimizing the regression prediction function of the least square support vector machine by adopting an immune genetic algorithm to obtain an initial model.
5. The urban rail transit passenger flow early warning method according to claim 4, wherein the optimizing the target data by using an optimization function of a least squares support vector machine algorithm based on the input data to obtain a regression prediction function of the least squares support vector machine comprises:
acquiring parameters of the target data;
converting the optimization function into a dual space based on a Lagrangian algorithm;
and performing matrix mixing operation on the parameters and the input data based on a dual space to obtain a least square support vector machine regression prediction function.
6. The urban rail transit passenger flow early warning method according to claim 4, wherein the specific process of optimizing the least squares support vector machine regression prediction function by using the immune genetic algorithm to obtain the initial model comprises:
setting a least squares support vector machine antigen and antibody population based on the input data;
performing immune algorithm iteration on the antigen and the antibodies in the antibody population to obtain the information entropy between the antibodies and the affinity between the antibodies and the antigen;
ranking the affinities;
and based on a genetic algorithm, carrying out coding operation on a plurality of antibodies corresponding to the affinity and ranked in the front to obtain an initial model.
7. The urban rail transit passenger flow early warning method according to claim 1, wherein the determining the early warning level of the urban rail according to the predicted urban rail transit passenger flow comprises:
when the predicted urban rail transit passenger flow exceeds a preset threshold value, setting the early warning level of the urban rail as a high risk;
and when the predicted urban rail transit passenger flow is lower than the preset threshold value, setting the early warning level of the urban rail as low risk.
8. An urban rail transit passenger flow early warning system, characterized in that, the system includes:
the passenger flow mode acquisition module is used for acquiring a multi-dimensional information data set to be detected of the urban rail transit, and matching the multi-dimensional information data set to be detected with a plurality of preset passenger flow modes to obtain a passenger flow mode corresponding to the multi-dimensional information data set to be detected;
the predicted urban rail transit passenger flow determining module is used for inputting the multi-dimensional information data set to be detected into an urban rail transit passenger flow early warning model corresponding to the passenger flow mode and outputting the predicted urban rail transit passenger flow corresponding to the multi-dimensional information data set to be detected through the urban rail transit passenger flow early warning model; the urban rail transit passenger flow early warning model adopts an optimized least square support vector machine based on an immune genetic algorithm;
and the early warning level determining module of the urban rail is used for determining the early warning level of the urban rail according to the predicted urban rail traffic passenger flow.
9. An intelligent terminal comprising a memory, and one or more programs, wherein the one or more programs are stored in the memory, and wherein the one or more programs being configured to be executed by the one or more processors comprises instructions for performing the method of any of claims 1-7.
10. A non-transitory computer-readable storage medium, wherein instructions in the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111292372.2A CN114091579A (en) | 2021-11-03 | 2021-11-03 | Urban rail transit passenger flow early warning system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111292372.2A CN114091579A (en) | 2021-11-03 | 2021-11-03 | Urban rail transit passenger flow early warning system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114091579A true CN114091579A (en) | 2022-02-25 |
Family
ID=80298716
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111292372.2A Pending CN114091579A (en) | 2021-11-03 | 2021-11-03 | Urban rail transit passenger flow early warning system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114091579A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663224A (en) * | 2012-03-07 | 2012-09-12 | 吉首大学 | Comentropy-based integrated prediction model of traffic flow |
| CN106372722A (en) * | 2016-09-18 | 2017-02-01 | 中国科学院遥感与数字地球研究所 | Subway short-time flow prediction method and apparatus |
| CN107180278A (en) * | 2017-05-27 | 2017-09-19 | 重庆大学 | A kind of real-time passenger flow forecasting of track traffic |
| CN109284877A (en) * | 2018-11-19 | 2019-01-29 | 福州大学 | Based on AIGA-WLSSVM Buried Pipeline rate prediction method |
| CN109308543A (en) * | 2018-08-20 | 2019-02-05 | 华南理工大学 | The short-term passenger flow forecasting of subway based on LS-SVM and real-time big data |
| CN110147903A (en) * | 2019-04-19 | 2019-08-20 | 合肥工业大学 | For predicting the method, system and storage medium of the volume of the flow of passengers at scenic spot |
| CN110276474A (en) * | 2019-05-22 | 2019-09-24 | 南京理工大学 | A kind of track traffic station passenger flow forecasting in short-term |
| CN112085368A (en) * | 2020-09-02 | 2020-12-15 | 西南交通大学 | Equipment energy production configuration and layout optimization method based on immune genetic algorithm |
| CN113177657A (en) * | 2021-04-20 | 2021-07-27 | 上海大学 | Rail transit passenger flow prediction method and device |
-
2021
- 2021-11-03 CN CN202111292372.2A patent/CN114091579A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663224A (en) * | 2012-03-07 | 2012-09-12 | 吉首大学 | Comentropy-based integrated prediction model of traffic flow |
| CN106372722A (en) * | 2016-09-18 | 2017-02-01 | 中国科学院遥感与数字地球研究所 | Subway short-time flow prediction method and apparatus |
| CN107180278A (en) * | 2017-05-27 | 2017-09-19 | 重庆大学 | A kind of real-time passenger flow forecasting of track traffic |
| CN109308543A (en) * | 2018-08-20 | 2019-02-05 | 华南理工大学 | The short-term passenger flow forecasting of subway based on LS-SVM and real-time big data |
| CN109284877A (en) * | 2018-11-19 | 2019-01-29 | 福州大学 | Based on AIGA-WLSSVM Buried Pipeline rate prediction method |
| CN110147903A (en) * | 2019-04-19 | 2019-08-20 | 合肥工业大学 | For predicting the method, system and storage medium of the volume of the flow of passengers at scenic spot |
| CN110276474A (en) * | 2019-05-22 | 2019-09-24 | 南京理工大学 | A kind of track traffic station passenger flow forecasting in short-term |
| CN112085368A (en) * | 2020-09-02 | 2020-12-15 | 西南交通大学 | Equipment energy production configuration and layout optimization method based on immune genetic algorithm |
| CN113177657A (en) * | 2021-04-20 | 2021-07-27 | 上海大学 | Rail transit passenger flow prediction method and device |
Non-Patent Citations (3)
| Title |
|---|
| 付华等: "基于IGA-LSSVM的煤矿瓦斯涌出量预测模型研究", 《中国安全科学学报》 * |
| 谭一帆: "综合客运枢纽客流预警研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 马海云著: "《面向过程的软件设计及优化技术》", 西安电子科技大学出版社 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ma et al. | Short-term traffic flow forecasting by selecting appropriate predictions based on pattern matching | |
| Han et al. | A combined online-learning model with K-means clustering and GRU neural networks for trajectory prediction | |
| Li et al. | Graph CNNs for urban traffic passenger flows prediction | |
| Zhang et al. | Traffic accident prediction based on LSTM-GBRT model | |
| Guo et al. | Short‐term passenger flow forecast of urban rail transit based on GPR and KRR | |
| CN110674636B (en) | Power consumption behavior analysis method | |
| Lu et al. | Lane-level traffic speed forecasting: A novel mixed deep learning model | |
| CN105046323B (en) | Regularization-based RBF network multi-label classification method | |
| Wang et al. | M2TNet: Multi-modal multi-task Transformer network for ultra-short-term wind power multi-step forecasting | |
| Liu et al. | Explanatory prediction of traffic congestion propagation mode: A self-attention based approach | |
| CN114493191A (en) | Driving behavior modeling analysis method based on network appointment data | |
| Xue et al. | Multi long-short term memory models for short term traffic flow prediction | |
| Brahimi et al. | Modelling on car-sharing serial prediction based on machine learning and deep learning | |
| Bleu-Laine et al. | Predicting adverse events and their precursors in aviation using multi-class multiple-instance learning | |
| CN108053646B (en) | Traffic characteristic obtaining method, traffic characteristic prediction method and traffic characteristic prediction system based on time sensitive characteristics | |
| Wang et al. | A multiple-parameter approach for short-term traffic flow prediction | |
| Shuai et al. | Short-term traffic flow prediction of expressway: a hybrid method based on singular spectrum analysis decomposition | |
| Cai et al. | A K-nearest neighbor locally search regression algorithm for short-term traffic flow forecasting | |
| CN114091579A (en) | Urban rail transit passenger flow early warning system and method | |
| CN116187506A (en) | Short-term wind power combination probability prediction method and system considering meteorological classification | |
| Ye et al. | Demand forecasting of online car‐hailing by exhaustively capturing the temporal dependency with TCN and Attention approaches | |
| CN114723147A (en) | New energy power prediction method based on improved wavelet transform and neural network | |
| Cai et al. | Hybrid variational autoencoder for time series forecasting | |
| CN114970882A (en) | Model prediction method and model system suitable for multiple scenes and multiple tasks | |
| Chen et al. | Improved Long Short-Term Memory-Based Periodic Traffic Volume Prediction Method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220225 |
|
| RJ01 | Rejection of invention patent application after publication |