CN113344240A - Shared bicycle flow prediction method and system - Google Patents
Shared bicycle flow prediction method and system Download PDFInfo
- Publication number
- CN113344240A CN113344240A CN202110455341.8A CN202110455341A CN113344240A CN 113344240 A CN113344240 A CN 113344240A CN 202110455341 A CN202110455341 A CN 202110455341A CN 113344240 A CN113344240 A CN 113344240A
- Authority
- CN
- China
- Prior art keywords
- shared bicycle
- clustering
- station
- vehicle
- stations
- 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 52
- 239000011159 matrix material Substances 0.000 claims abstract description 52
- 239000000203 mixture Substances 0.000 claims abstract description 37
- 238000013508 migration Methods 0.000 claims abstract description 9
- 230000005012 migration Effects 0.000 claims abstract description 9
- 238000012546 transfer Methods 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 4
- 230000006870 function Effects 0.000 description 9
- 238000009826 distribution Methods 0.000 description 8
- 238000007476 Maximum Likelihood Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
Images
Classifications
-
- 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"
-
- 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
- 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
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0645—Rental transactions; Leasing transactions
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Accounting & Taxation (AREA)
- General Business, Economics & Management (AREA)
- Finance (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Artificial Intelligence (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Game Theory and Decision Science (AREA)
- General Engineering & Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The utility model provides a prediction method and a system for the flow of a shared bicycle, the proposal comprises the following steps: based on the shared bicycle station information and the historical travel information, clustering stations by using a Gaussian mixture clustering model of hierarchical iteration; according to the clustering result, dividing the urban area into an upper layer space structure and a lower layer space structure, and performing area reconstruction on the shared bicycle stations to obtain a bicycle borrowing and returning vector matrix of each shared bicycle station area; obtaining the city meteorological feature matrix based on the meteorological data; inputting the vehicle borrowing and returning vector matrix and the urban meteorological feature matrix into a pre-trained deep space-time residual error network to obtain the prediction results of the vehicle borrowing quantity and the vehicle returning quantity of each shared bicycle area in future time; the scheme introduces a class-to-class migration trend, and the obtained result is more stable and has good robustness; meanwhile, the shared bicycle station after spatial reconstruction has more time and spatial correlation, and the prediction precision can be well increased.
Description
Technical Field
The disclosure belongs to the technical field of shared bicycle flow prediction, and particularly relates to a shared bicycle flow prediction method and system.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The shared bicycle is a novel bicycle using mode generated under the background of ' Internet plus ' and a shared economic society, the problem that residents go out for the last kilometer ' is well solved, and the service is quite popular in cities at present. With the shared bicycle system, people can easily rent or return bicycles at a location provided by the shared bicycle company. The occurrence of the shared single-vehicle service enables the utilization rate of public transportation to be improved to a great extent, and the problem of road traffic jam is effectively relieved.
However, the shared bicycle is convenient for the user to go out, and meanwhile, some adverse factors influencing the riding experience of the customer exist. Among the most obvious disadvantages are the unbalanced use caused by the uneven distribution of the bicycle stations in the different service areas. The number of users with vehicle borrowing requirements in some areas is large, and the number of the single vehicles required to be provided by the station exceeds the reserve number of the station; in some regions, the number of users needing to use the vehicle is small due to remote geographic positions, so that the single vehicle at the station is always in an idle state, and resource waste is caused. The problem of shared bicycle inter-site supply imbalance is generally caused by the way a user travels: users typically leave the home in the morning and go to work at work. The bicycles in the residential area can be used excessively at this time; in contrast, bicycles at work sites are heavily piled up in the morning, resulting in an oversupply. The afternoon situation is completely reversed. Both situations can lead to an unbalanced supply of bicycles at different stations in the city at different times. In order to solve the problem, a shared bicycle operator can use a truck to continuously transport shared bicycles among various stations, and the shared bicycle operator can manually schedule the shared bicycles to recover the normal operation of the shared bicycle system, which is obviously a solution that wastes time and labor and addresses the symptoms rather than the root cause.
Therefore, researchers at home and abroad have long and intensive research on the problem of forecasting the usage of the shared bicycle system: lin et al introduced a bicycle sharing strategy design problem that included a bicycle garage storage system and an inventory center based model. The method relates to various aspects of design work, such as the number and the positions of stations sharing a bicycle system, the creation of a bicycle lane, the creation of a bicycle travel route and the like, and a better solution is found for urban traffic throughput and a balancing and rebalancing strategy thereof. Li, zheng et al proposed a hierarchical predictive model to predict the number of bicycles to be rented or returned in the future, which is more focused on the macroscopic traffic flow in the bicycle sharing system than the micro travel destination and duration, which is of vital reference value to the research methods of bicycle sharing system analysis and travel prediction. The prediction model proposed in the literature firstly performs double-layer clustering on bicycle stations by using GC and K-means, and then predicts the number of rented and returned bicycles by using an inference model based on multiple similarities. Compared with the prediction without adopting a clustering mode, the new prediction model has the advantage that the prediction accuracy is improved. The GC or K-means clustering needs to preset the number of clusters, namely the K value, the clustering effect depends on the selection of the initial center value, and the initial center value has certain randomness, so that the result of each prediction has certain deviation and is unstable. Based on previous work, the W.Jia, Y.Tan et al combine the AP clustering with the multi-similarity reference model to predict the borrowing and returning number of the shared bicycle system in the future time, and the clustering does not need to specify the clustering number, so that the obtained result is more stable than the previous research, but is limited by the limitation of the multi-similarity reference model, and the prediction precision still has a great space for improvement.
The inventor finds that although the existing method provides a great number of solutions for the usage prediction problem of the shared bicycle system, a series of problems still exist, and particularly the existing method cannot effectively capture the characteristics influencing the usage of the shared bicycle, so that the usage of each station of the shared bicycle in future time of a city is unreasonably predicted, and the accuracy is not high.
Disclosure of Invention
In order to solve the problems, the scheme clusters shared single-vehicle stations through a Gaussian mixture clustering model based on hierarchical iteration, introduces a migration trend between classes, and obtains a more stable result with good robustness; meanwhile, the shared bicycle station after spatial reconstruction has more time and spatial correlation, and the prediction precision can be well increased.
According to a first aspect of the embodiments of the present disclosure, there is provided a shared bicycle traffic prediction method, including:
acquiring shared bicycle station information, historical travel information and corresponding meteorological data of a city;
based on the shared bicycle station information and the historical travel information, clustering stations by using a Gaussian mixture clustering model of hierarchical iteration;
according to the clustering result, dividing the urban area into an upper layer space structure and a lower layer space structure, and performing area reconstruction on the shared bicycle stations to obtain a bicycle borrowing and returning vector matrix of each shared bicycle station area;
obtaining the city meteorological feature matrix based on the meteorological data;
and inputting the vehicle borrowing and returning vector matrix and the urban meteorological feature matrix into a pre-trained deep space-time residual error network to obtain the prediction results of the vehicle borrowing quantity and the vehicle returning quantity of each shared bicycle area in future time.
Further, the clustering the sites by using the hierarchical iterative gaussian mixture clustering model specifically includes:
clustering the shared single-vehicle stations by adopting a Gaussian mixture clustering method by using the information of the shared single-vehicle stations to obtain a plurality of initial clusters;
according to historical travel information, calculating the number of vehicle transfer between every two clusters of each type to obtain a transfer trend matrix, and performing norm processing on the transfer trend matrix;
and clustering by adopting a Gaussian mixture clustering method based on the norm of the migration trend matrix and the shared bicycle station information to obtain a new cluster, and repeatedly executing the process until the new clustering result tends to be stable.
Further, the dividing of the urban area into an upper space structure and a lower space structure, and the area reconstruction of the shared single-vehicle station specifically include: taking the original shared bicycle station as a lower-layer space architecture, and taking the shared bicycle station where the clustering center after hierarchical iterative clustering is positioned as a new space point to form an upper-layer space architecture; the lower layer space comprises all the shared single vehicle stations, and the upper layer space only reserves the shared single vehicle stations corresponding to the clustering centers.
According to a second aspect of embodiments of the present disclosure, there is provided a shared bicycle flow prediction system, comprising:
the system comprises a data acquisition unit, a data processing unit and a data processing unit, wherein the data acquisition unit is used for acquiring shared bicycle station information, historical travel information and corresponding meteorological data of a city;
the clustering unit is used for clustering the stations by utilizing a Gaussian mixture clustering model of hierarchical iteration based on the shared bicycle station information and the historical travel information;
the region reconstruction unit is used for dividing the urban region into an upper layer space structure and a lower layer space structure according to the clustering result, performing region reconstruction on the shared bicycle stations and obtaining a bicycle borrowing and returning vector matrix of each shared bicycle station region;
the meteorological feature acquisition unit is used for acquiring the urban meteorological feature matrix based on the meteorological data;
and the prediction unit is used for inputting the vehicle borrowing and returning vector matrix and the urban meteorological feature matrix into a pre-trained deep space-time residual error network to obtain the prediction results of the vehicle borrowing quantity and the vehicle returning quantity of each shared vehicle area in future time.
According to a third aspect of the embodiments of the present disclosure, there is provided an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the memory, wherein the processor implements the method for predicting the shared bicycle traffic when executing the program.
According to a fourth aspect of embodiments of the present disclosure, there is provided a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method of shared bicycle flow prediction as described.
Compared with the prior art, the beneficial effect of this disclosure is:
the scheme disclosed by the disclosure provides a shared bicycle flow prediction model based on Gaussian mixture clustering-space-time residual error network, and the usage of the shared bicycle stations of the city at the future time is predicted based on historical data. Compared with the traditional clustering method, the core idea of clustering the shared single-vehicle stations by the Gaussian mixture clustering model based on hierarchical iteration is the biggest difference that the migration trend among classes is introduced, and the obtained result is more stable and has good robustness. In addition, the shared bicycle station after spatial reconstruction has more time and spatial correlation, and the prediction precision can be well increased. Meanwhile, the robustness and stability of the method disclosed by the invention are proved through a large number of experiments.
Advantages of additional aspects of the disclosure will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the disclosure.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure and are not to limit the disclosure.
Fig. 1 is a flowchart of a method for predicting a shared bicycle traffic flow according to a first embodiment of the disclosure;
fig. 2 is a schematic diagram of gaussian mixed clustering based on hierarchical iteration according to a first embodiment of the present disclosure;
fig. 3 is a schematic view of spatial reconstruction of a shared single-vehicle station according to a first embodiment of the disclosure.
Detailed Description
The present disclosure is further described with reference to the following drawings and examples.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present disclosure. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict.
The first embodiment is as follows:
the embodiment aims to provide a method for predicting the flow of a shared bicycle.
In order to overcome the defects of the prior art, the invention provides a sharing bicycle flow prediction model based on Gaussian mixture clustering-space-time residual error network, which is used for predicting the usage amount of each sharing bicycle station in a future time city. Firstly, a Gaussian mixture clustering model based on hierarchical iteration is provided for sharing a single vehicle station for clustering and dividing the single vehicle station into a plurality of different areas. The model introduces a class-to-class migration trend on the basis of a traditional Gaussian mixture clustering model (only considering position information), obtains a stable clustering effect by performing certain times of clustering iteration, and has good robustness. Secondly, a space hierarchy architecture division method is provided, and an urban area is divided into an upper layer space architecture and a lower layer space architecture on the basis of a clustering result obtained previously. Specifically, the station position of a shared bicycle is used as a lower-layer space framework, the space positions of the shared bicycle are reconstructed in batches, and the station where the clustering center after hierarchical iterative clustering is located is used as a new space point, namely an upper-layer space framework, so that the relative information of the space positions is reserved, and the mutual correlation between the center information can be maintained. On the basis, a deep space-time residual error network (ST-ResNet) is used for carrying out batch prediction on the use condition of a shared single-vehicle region (a new region formed by a cluster center) in the future period of the city, a residual error convolution unit is designed to model traffic flow characteristics according to time continuity, periodicity and trend, then the three residual error networks are dynamically aggregated, and a final prediction result is obtained by combining meteorological factors. As shown in fig. 1, a method for predicting a shared bicycle traffic includes:
step 1: acquiring shared bicycle station information, historical travel information and corresponding meteorological data of a city;
the station information comprises longitude and latitude information of a station. The historical trip information includes a trip duration, a departure time, an end time, a departure station ID, a departure station name, a departure station latitude, longitude, an end station ID, an end station name, an end station latitude, longitude, a single vehicle ID, a user type, and the like of each trip.
Step 2: based on the shared bicycle station information and the historical travel information, clustering stations by using a Gaussian mixture clustering model of hierarchical iteration;
wherein the step 2 comprises:
step 2.1, clustering the shared single-vehicle station information by adopting a Gaussian mixture clustering method according to the geographic position to obtain a plurality of initial clusters;
2.2, calculating the vehicle transfer quantity between every two clusters according to historical travel data to obtain a rented transfer trend matrix, and performing norm processing on the transfer trend matrix;
and 2.3, clustering by adopting a Gaussian mixture clustering method based on the norm and the geographic position of the rented migration trend matrix to obtain a new cluster, and repeatedly executing the steps 2.2-2.3 until a new clustering result tends to be stable.
And step 3: according to the clustering result, dividing the urban area into an upper layer space structure and a lower layer space structure, and performing area reconstruction on the shared bicycle stations to obtain a bicycle borrowing and returning vector matrix of each shared bicycle station area;
wherein, the step 3 specifically comprises:
and according to the clustering result, performing region reconstruction on the shared single-vehicle station. Recalculating the number of the shared bicycles borrowed and returned per hour in each shared bicycle station area by taking the obtained clustering center as a center, and generating a two-dimensional matrix based on the obtained result;
and 4, step 4: obtaining the city meteorological feature matrix based on the meteorological data;
wherein the meteorological features matrix comprises: weather, temperature and wind speed, extracting hourly weather eigenvalues, described using a 0-1 matrix, defining four types of weather (common): sunny days, rainy days, snowy days and foggy days. The rows of the matrix represent hourly meteorological features and the columns of the matrix are divided into 7 columns: the time stamp (the hour to which the record belongs), the sunny day, the rainy day, the snowy day, the foggy day, the temperature value and the wind speed value are marked as 1 if the weather happens in the hour, and are marked as 0 if the weather happens in the hour. Temperature and wind speed we describe using numerical values in degrees celsius and miles (mph), respectively.
And 5: and inputting the vehicle borrowing and returning vector matrix and the urban meteorological feature matrix into a pre-trained deep space-time residual error network to obtain the prediction results of the vehicle borrowing quantity and the vehicle returning quantity of each shared bicycle area in future time.
The following briefly explains the gaussian mixture model and gaussian mixture model clustering:
(one) Gaussian mixture model
A Gaussian Mixture Model (GMM) is a Model that accurately quantizes objects using a Gaussian probability density function (normal distribution curve) and decomposes one object into a plurality of objects based on the Gaussian probability density function (normal distribution curve). In this section, a gaussian mixture model is selected to cluster shared single-vehicle sites, with classification based on the characteristics of each site. Each GMM consists of K Gaussian distributions, each Gaussian is called a 'Component', and the components linearly add together to form the probability density function of the GMM, as shown in equation (1):
wherein K components of GMM correspond to K cluster, pikIs the influence factor of each Gaussian distribution (Component) on data points, μkFor the mean of each class, sigmakIs a covariance matrix.
Now there are N data points and assuming they obey a certain distribution (denoted p (x)), now the inner set of parameters π is determinedk、μkSum-sigmakThe probability distribution it determines yields the probability maximum for the given data points, which is actually shown in equation (2):
we refer to this product as a likelihood function, and usually the probability of a single point is small, and many small numbers are multiplied together to easily cause underflow of floating point numbers in a computer, so taking the logarithm of p (x) and converting the product into a sum, as shown in equation (3):
thereby obtaining the log-likelihood function. Then we maximize this function, i.e. find a set of parameters pik、μkSum-sigmakThe likelihood function is maximized.
Because addition exists in the logarithmic function, the maximum value is directly solved by a method of directly solving the equation, and in order to solve the problem, the method adopts a random point selection method in GMM: the method is divided into two steps of E step and M step, namely EM algorithm. The EM algorithm is a maximum likelihood estimation method for solving probability model parameters from incomplete data or a data set with data loss (hidden variables exist). E, calculating the maximum likelihood estimation value of the hidden variable by using the existing estimation value of the hidden variable; step M, maximizing the maximum likelihood value obtained in step E to calculate the value of the parameter; iterate until convergence.
The EM algorithm flow comprises the following steps:
(1) estimate the probability that data is generated by each "Component" (and not the probability that each "Component" is selected): for each data xiIn other words, the probability it generates from the kth Component is:
wherein N (x)i|μk,Σk) Is the posterior probability.
(2) The values of the parameters μ and Σ can be obtained by derivation by maximum likelihood estimation, letting the parameter be 0:
wherein the content of the first and second substances,
and pikCan be estimated as Nk/N。
(3) The first two steps of the iteration are repeated until the values of the likelihood functions converge.
(II) Gaussian mixture model clustering
Gaussian Mixture Model clustering (Gaussian Mixture Model Cluster) is the "inverse process" of generating data samples with GMM on the basis of a Gaussian Mixture Model: given the cluster number K, the parameters (i.e. mean vector) mu, covariance matrix Σ, and weight pi of each mixture component are derived by a certain parameter estimation method through a given data set, and each multivariate Gaussian distribution component corresponds to a cluster after clustering. And after the parameter estimation process is finished, for each sample point, calculating the posterior probability of the sample point belonging to each cluster according to the Bayesian theorem, and dividing the sample into the clusters with the maximum posterior probability. GMM, a Clustering method that gives the probability that a sample point belongs to each cluster, is called Soft Clustering, as opposed to the Clustering method that gives the cluster division of sample points directly from K-means, etc.
Example two:
the embodiment aims to provide a shared bicycle flow prediction system.
A shared bicycle flow prediction system, comprising:
the system comprises a data acquisition unit, a data processing unit and a data processing unit, wherein the data acquisition unit is used for acquiring shared bicycle station information, historical travel information and corresponding meteorological data of a city;
the clustering unit is used for clustering the stations by utilizing a Gaussian mixture clustering model of hierarchical iteration based on the shared bicycle station information and the historical travel information;
the region reconstruction unit is used for dividing the urban region into an upper layer space structure and a lower layer space structure according to the clustering result, performing region reconstruction on the shared bicycle stations and obtaining a bicycle borrowing and returning vector matrix of each shared bicycle station region;
the meteorological feature acquisition unit is used for acquiring the urban meteorological feature matrix based on the meteorological data;
and the prediction unit is used for inputting the vehicle borrowing and returning vector matrix and the urban meteorological feature matrix into a pre-trained deep space-time residual error network to obtain the prediction results of the vehicle borrowing quantity and the vehicle returning quantity of each shared vehicle area in future time.
Further, the clustering the sites by using the hierarchical iterative gaussian mixture clustering model specifically includes:
clustering the shared single-vehicle stations by adopting a Gaussian mixture clustering method by using the information of the shared single-vehicle stations to obtain a plurality of initial clusters;
according to historical travel information, calculating the number of vehicle transfer between every two clusters of each type to obtain a transfer trend matrix, and performing norm processing on the transfer trend matrix;
and clustering by adopting a Gaussian mixture clustering method based on the norm of the migration trend matrix and the shared bicycle station information to obtain a new cluster, and repeatedly executing the process until the new clustering result tends to be stable.
Further, the dividing of the urban area into an upper space structure and a lower space structure, and the area reconstruction of the shared single-vehicle station specifically include: and taking the original shared bicycle station as a lower-layer space architecture, and taking the shared bicycle station where the clustering center after hierarchical iterative clustering is positioned as a new space point to form an upper-layer space architecture. The lower-layer space comprises all the shared bicycle stations, only the shared bicycle stations corresponding to the clustering centers are reserved in the upper-layer space, the obtained clustering centers are used as centers, the number of the borrowed bicycles and the number of the returned bicycles in the preset time period of each shared bicycle station area are recalculated, and the vehicle borrowing and returning vector matrix is obtained.
In further embodiments, there is also provided:
an electronic device comprising a memory and a processor, and computer instructions stored on the memory and executed on the processor, the computer instructions when executed by the processor performing the method of embodiment one. For brevity, no further description is provided herein.
It should be understood that in this embodiment, the processor may be a central processing unit CPU, and the processor may also be other general purpose processors, digital signal processors DSP, application specific integrated circuits ASIC, off-the-shelf programmable gate arrays FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and so on. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may include both read-only memory and random access memory, and may provide instructions and data to the processor, and a portion of the memory may also include non-volatile random access memory. For example, the memory may also store device type information.
A computer readable storage medium storing computer instructions which, when executed by a processor, perform the method of embodiment one.
The method in the first embodiment may be directly implemented by a hardware processor, or may be implemented by a combination of hardware and software modules in the processor. The software modules may be located in ram, flash, rom, prom, or eprom, registers, among other storage media as is well known in the art. The storage medium is located in a memory, and a processor reads information in the memory and completes the steps of the method in combination with hardware of the processor. To avoid repetition, it is not described in detail here.
Those of ordinary skill in the art will appreciate that the various illustrative elements, i.e., algorithm steps, described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The shared bicycle flow prediction method and the shared bicycle flow prediction system provided by the embodiment can be realized, and have a wide application prospect.
The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure, and various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.
Although the present disclosure has been described with reference to specific embodiments, it should be understood that the scope of the present disclosure is not limited thereto, and those skilled in the art will appreciate that various modifications and changes can be made without departing from the spirit and scope of the present disclosure.
Claims (10)
1. A method for predicting a shared bicycle traffic, comprising:
acquiring shared bicycle station information, historical travel information and corresponding meteorological data of a city;
based on the shared bicycle station information and the historical travel information, clustering stations by using a Gaussian mixture clustering model of hierarchical iteration;
according to the clustering result, dividing the urban area into an upper layer space structure and a lower layer space structure, and performing area reconstruction on the shared bicycle stations to obtain a bicycle borrowing and returning vector matrix of each shared bicycle station area;
obtaining the city meteorological feature matrix based on the meteorological data;
and inputting the vehicle borrowing and returning vector matrix and the urban meteorological feature matrix into a pre-trained deep space-time residual error network to obtain the prediction results of the vehicle borrowing quantity and the vehicle returning quantity of each shared bicycle area in future time.
2. The method for predicting the shared bicycle flow according to claim 1, wherein the clustering of the stations by using the hierarchical iterative gaussian mixture clustering model specifically comprises:
clustering the shared single-vehicle stations by adopting a Gaussian mixture clustering method by using the information of the shared single-vehicle stations to obtain a plurality of initial clusters;
according to historical travel information, calculating the number of vehicle transfer between every two clusters of each type to obtain a transfer trend matrix, and performing norm processing on the transfer trend matrix;
and clustering by adopting a Gaussian mixture clustering method based on the norm of the migration trend matrix and the shared bicycle station information to obtain a new cluster, and repeatedly executing the process until the new clustering result tends to be stable.
3. The method for predicting the flow of the shared bicycle according to claim 1, wherein the step of dividing the urban area into an upper space structure and a lower space structure and performing area reconstruction on the shared bicycle station specifically comprises the steps of: taking the original shared bicycle station as a lower-layer space architecture, and taking the shared bicycle station where the clustering center after hierarchical iterative clustering is positioned as a new space point to form an upper-layer space architecture; the lower-layer space comprises all the shared bicycle stations, and the upper-layer space only reserves the shared bicycle stations corresponding to the clustering centers.
4. The method of claim 1, wherein the station information includes longitude and latitude information of a station, and the historical trip information includes trip duration, departure time, end time, departure station ID, departure station name, departure station latitude, longitude, end station ID, end station name, end station latitude, longitude, vehicle ID, and user type for each trip.
5. The method of claim 1, wherein the weather signature matrix comprises: weather, temperature and wind speed, wherein the weather comprises sunny days, rainy days, snowy days and foggy days.
6. A shared bicycle flow prediction system, comprising:
the system comprises a data acquisition unit, a data processing unit and a data processing unit, wherein the data acquisition unit is used for acquiring shared bicycle station information, historical travel information and corresponding meteorological data of a city;
the clustering unit is used for clustering the stations by utilizing a Gaussian mixture clustering model of hierarchical iteration based on the shared bicycle station information and the historical travel information;
the region reconstruction unit is used for dividing the urban region into an upper layer space structure and a lower layer space structure according to the clustering result, performing region reconstruction on the shared bicycle stations and obtaining a bicycle borrowing and returning vector matrix of each shared bicycle station region;
the meteorological feature acquisition unit is used for acquiring the urban meteorological feature matrix based on the meteorological data;
and the prediction unit is used for inputting the vehicle borrowing and returning vector matrix and the urban meteorological feature matrix into a pre-trained deep space-time residual error network to obtain the prediction results of the vehicle borrowing quantity and the vehicle returning quantity of each shared vehicle area in future time.
7. The system for predicting shared bicycle traffic of claim 6, wherein the clustering of the stations by using the hierarchical iterative Gaussian mixture clustering model specifically comprises:
clustering the shared single-vehicle stations by adopting a Gaussian mixture clustering method by using the information of the shared single-vehicle stations to obtain a plurality of initial clusters;
according to historical travel information, calculating the number of vehicle transfer between every two clusters of each type to obtain a transfer trend matrix, and performing norm processing on the transfer trend matrix;
and clustering by adopting a Gaussian mixture clustering method based on the norm of the migration trend matrix and the shared bicycle station information to obtain a new cluster, and repeatedly executing the process until the new clustering result tends to be stable.
8. The system for predicting the flow of the shared single vehicle according to claim 6, wherein the step of dividing the urban area into an upper space structure and a lower space structure and performing area reconstruction on the shared single vehicle station specifically comprises the steps of: taking the original shared bicycle station as a lower-layer space architecture, and taking the shared bicycle station where the clustering center after hierarchical iterative clustering is positioned as a new space point to form an upper-layer space architecture; the lower-layer space comprises all the shared bicycle stations, and the upper-layer space only reserves the shared bicycle stations corresponding to the clustering centers.
9. An electronic device comprising a memory, a processor, and a computer program stored and executed on the memory, wherein the processor, when executing the program, implements a method of shared bicycle flow prediction as claimed in any one of claims 1-5.
10. A non-transitory computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method of shared bicycle flow prediction as claimed in any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110455341.8A CN113344240A (en) | 2021-04-26 | 2021-04-26 | Shared bicycle flow prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110455341.8A CN113344240A (en) | 2021-04-26 | 2021-04-26 | Shared bicycle flow prediction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113344240A true CN113344240A (en) | 2021-09-03 |
Family
ID=77468681
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110455341.8A Pending CN113344240A (en) | 2021-04-26 | 2021-04-26 | Shared bicycle flow prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113344240A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114639202A (en) * | 2022-03-16 | 2022-06-17 | 无锡易百客科技有限公司 | Campus shared electric bicycle management system and method based on Internet |
| CN115271833A (en) * | 2022-09-28 | 2022-11-01 | 湖北省楚天云有限公司 | Shared bicycle demand prediction method and prediction system |
| CN116362527A (en) * | 2023-06-02 | 2023-06-30 | 北京阿帕科蓝科技有限公司 | Vehicle scheduling method, device, computer equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109345001A (en) * | 2018-09-07 | 2019-02-15 | 山东师范大学 | The unbalance method of adjustment of shared bicycle website and system based on Pareto multiobjective selection |
| CN111429000A (en) * | 2020-03-23 | 2020-07-17 | 成都信息工程大学 | Shared bicycle pick-and-return site recommendation method and system based on site clustering |
| CN112419716A (en) * | 2020-11-13 | 2021-02-26 | 东南大学 | Layout configuration method for shared single-vehicle facilities in track station transfer influence area |
| CN112466117A (en) * | 2020-11-24 | 2021-03-09 | 南通大学 | Road network short-term traffic flow prediction method based on deep space-time residual error network |
-
2021
- 2021-04-26 CN CN202110455341.8A patent/CN113344240A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109345001A (en) * | 2018-09-07 | 2019-02-15 | 山东师范大学 | The unbalance method of adjustment of shared bicycle website and system based on Pareto multiobjective selection |
| CN111429000A (en) * | 2020-03-23 | 2020-07-17 | 成都信息工程大学 | Shared bicycle pick-and-return site recommendation method and system based on site clustering |
| CN112419716A (en) * | 2020-11-13 | 2021-02-26 | 东南大学 | Layout configuration method for shared single-vehicle facilities in track station transfer influence area |
| CN112466117A (en) * | 2020-11-24 | 2021-03-09 | 南通大学 | Road network short-term traffic flow prediction method based on deep space-time residual error network |
Non-Patent Citations (4)
| Title |
|---|
| J ZHANG: "predicting citywide crowd flows using deep spatio-temporal residual networks", 《ARTIFICIAL INTELLIGENCE》 * |
| LIN J R: "A hub location inventory model for bicycle sharing system design: Formulation and solution", 《COMPUTERS & INDUSTRIAL ENGINEERING》 * |
| YEXIN LI: "Traffic Prediction in a bike-sharing system", 《IN SIGSPATIAL》 * |
| 贾文祯: "共享单车系统分析与流量预测方法研究", 《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅱ辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114639202A (en) * | 2022-03-16 | 2022-06-17 | 无锡易百客科技有限公司 | Campus shared electric bicycle management system and method based on Internet |
| CN115271833A (en) * | 2022-09-28 | 2022-11-01 | 湖北省楚天云有限公司 | Shared bicycle demand prediction method and prediction system |
| CN115271833B (en) * | 2022-09-28 | 2023-08-25 | 湖北省楚天云有限公司 | Method and system for predicting demand of shared bicycle |
| CN116362527A (en) * | 2023-06-02 | 2023-06-30 | 北京阿帕科蓝科技有限公司 | Vehicle scheduling method, device, computer equipment and storage medium |
| CN116362527B (en) * | 2023-06-02 | 2023-12-05 | 北京阿帕科蓝科技有限公司 | Vehicle scheduling method, device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113344240A (en) | Shared bicycle flow prediction method and system | |
| CN108960476B (en) | AP-TI clustering-based shared bicycle flow prediction method and device | |
| Miao et al. | Data-driven distributionally robust vehicle balancing using dynamic region partitions | |
| KR20190020852A (en) | Methods and systems for transport capacity scheduling | |
| Hajbabaie et al. | Traffic signal timing optimization: Choosing the objective function | |
| Li et al. | Deep learning based parking prediction on cloud platform | |
| Li et al. | Origin and destination forecasting on dockless shared bicycle in a hybrid deep-learning algorithms | |
| CN114021883A (en) | Dispatching method for subway transfer shared bicycle in peak period | |
| Guo et al. | An online learning collaborative method for traffic forecasting and routing optimization | |
| CN116934056B (en) | Vehicle scheduling method, device, computer equipment and storage medium | |
| Lin et al. | A stacking model for variation prediction of public bicycle traffic flow | |
| CN111861017A (en) | Charging station network optimization method based on current network data | |
| Kong et al. | The method and application of big data mining for mobile trajectory of taxi based on MapReduce | |
| CN116828515A (en) | Edge server load prediction method based on space-time diagram convolution under Internet of vehicles | |
| Khan et al. | Electric kickboard demand prediction in spatiotemporal dimension using clustering-aided bagging regressor | |
| CN113140108B (en) | Cloud traffic situation prediction method in internet-connected intelligent traffic system | |
| Rasaizadi et al. | Short‐Term Prediction of Traffic State for a Rural Road Applying Ensemble Learning Process | |
| Yu et al. | A hybrid prediction approach for road tunnel traffic based on spatial-temporary data fusion | |
| CN107194505A (en) | A kind of method and system that bus travel amount is predicted based on city big data | |
| Li et al. | Multigraph aggregation spatiotemporal graph convolution network for ride-hailing pick-up region prediction | |
| Wang et al. | TL-FCM: A hierarchical prediction model based on two-level fuzzy c-means clustering for bike-sharing system | |
| CN111121806B (en) | Travel mode planning method and device, computer equipment and storage medium | |
| CN114418606B (en) | Network vehicle order demand prediction method based on space-time convolution network | |
| CN110146102B (en) | Path planning method, device, equipment and storage medium | |
| Zheng et al. | A multiple factor bike usage prediction model in bike-sharing system |
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: 20210903 |
|
| RJ01 | Rejection of invention patent application after publication |