CN107798877B - Method and system for predicting traffic volume based on highway charging data - Google Patents
Method and system for predicting traffic volume based on highway charging data Download PDFInfo
- Publication number
- CN107798877B CN107798877B CN201711176723.7A CN201711176723A CN107798877B CN 107798877 B CN107798877 B CN 107798877B CN 201711176723 A CN201711176723 A CN 201711176723A CN 107798877 B CN107798877 B CN 107798877B
- Authority
- CN
- China
- Prior art keywords
- data
- time
- traffic demand
- time sequence
- expressway
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Abstract
The invention discloses a method for predicting traffic volume between an entrance and an exit of an expressway based on expressway toll data, which comprises the following steps: the method comprises the steps that networking charging data of a highway in a preset time period are subjected to centralized processing according to a time granularity threshold, and the centralized networking charging data are obtained; converting the integrated networking charging data into at least one OD traffic demand matrix according to a certain index rule; and generating an OD traffic demand time sequence by using the at least one OD traffic demand matrix, and predicting the OD traffic demand in the future time period by using an autoregressive integrated moving average model (ARIMA). The invention can help managers to master the development situation of traffic demand in time, early warn the situation of sudden increase of traffic demand and the like, and prejudge whether the capacity of the highway is overloaded, thereby laying a data foundation for efficient development of daily business of the highway, reasonable formulation of policy measures, capital release of construction projects and the like and providing decision reference data.
Description
Technical Field
The invention relates to the technical field of highway networking charging data analysis, in particular to a method and a system for predicting traffic volume based on highway charging data.
Background
In recent years, the construction of the expressway in China is vigorously developed, and the expressway plays an important role in promoting the regional economic development. With the increasing utilization rate of the expressway, the corresponding problems of traffic congestion, traffic safety, environmental pollution and the like of the expressway are gradually shown, and the problems in the operation of the expressway are urgently solved.
Meanwhile, in the operation of the highway, a large amount of networking charging data is generated, and the data stores detailed information of all vehicles entering and leaving the highway, including information of entering time and place, leaving time and place, model plate numbers and the like. Although the data can provide comprehensive information, the data does not play a due role in daily operation and management. How to utilize emerging data processing and data mining technology to convert 'sleeping data' into an important medium for enhancing information service, avoiding accidents and optimizing management of a highway becomes an important problem in the field of intelligent transportation.
Disclosure of Invention
The invention provides a method and a system for predicting traffic volume based on highway toll data, which aim to solve the problem of predicting OD traffic volume of a highway.
In order to solve the above-mentioned problems, according to an aspect of the present invention, there is provided a method of predicting a traffic volume based on highway toll data, the method including:
the method comprises the steps of performing centralized processing on networking charging data of a highway in a preset time period according to a time granularity threshold value, and acquiring the centralized networking charging data, wherein the networking charging data comprises the following steps: time data, entrance code data, exit code data and direction of the vehicle entering the highway;
converting the integrated networking charging data into at least one OD traffic demand matrix according to a certain index rule;
and generating an OD traffic demand time sequence by using the at least one OD traffic demand matrix, and predicting the OD traffic demand in the future time period by using an autoregressive integrated moving average model (ARIMA).
Preferably, the collecting and processing the networking charging data of the expressway within the preset time period according to the time granularity threshold value to obtain collected and counted networking charging data includes:
converting time data of vehicles entering the expressway in the networking charging data from a field format into a float type to acquire first data;
performing precision truncation processing on time data of vehicles entering the expressway in the first data according to a time granularity threshold to obtain second data;
converting time data of vehicles entering the expressway in the second data into a field format to obtain third data;
and respectively adding a timestamp to the time data of the third data when the vehicle enters the expressway according to a preset period threshold value to serve as an index of the time data of the vehicle entering the expressway, and converting the third data added with the index to obtain the integrated networking charging data.
Preferably, wherein
Converting time data of vehicles entering the expressway in the networking charging data from a field format into a float type by using a cast function, and acquiring first data;
and converting the time data in the second data into a field format by using a cast function, acquiring third data, and converting the third data into aggregated networking charging data which meets the time granularity requirement and is grouped based on database entry codes and exit codes by using a COUNT function and a group by statement.
Preferably, the converting the collected networked charging data into at least one OD traffic demand matrix according to a certain index rule includes:
numbering the entrance and the exit of the expressway according to the driving direction of the vehicle, forming OD pairs according to the numbers, and establishing an index mapping relation between each OD pair in an OD traffic demand matrix and an entrance code and an exit code in a database;
and extracting and collecting the collected networking charging data and putting the networking charging data into corresponding positions of the OD traffic demand matrixes to form at least one OD traffic demand matrix within a preset period threshold, wherein the row number of each OD traffic demand matrix corresponds to the number of an OD pair, and the column number of each OD traffic demand matrix corresponds to the time stamp of the time data when the vehicle enters the expressway.
Preferably, the generating an OD traffic demand time series by using the at least one OD traffic demand matrix, and predicting OD traffic volumes in a future time period by using an autoregressive integrated moving average model ARIMA includes:
converting the at least one OD traffic demand matrix data into historical traffic demand time-series data of each OD pair;
respectively carrying out trend line removing processing on the historical traffic demand time sequence data of each OD pair to obtain a stationary time sequence;
establishing a test model, calculating an information criterion AIC value, and selecting a model with the minimum AIC value as a prediction model;
and carrying out reduction difference processing by using an autoregressive integrated moving average model ARIMA to predict OD traffic volume of a future time period.
Preferably, wherein
Adopting an armax function to establish a test model, calculating an information criterion AIC value by using an AIC function, and selecting the model with the minimum AIC value as a prediction model;
and (3) performing reduction difference processing by using a prediction function and applying an autoregressive integrated moving average model ARIMA to predict OD traffic volume of a future time period.
Preferably, the performing a trendline removing process on the time series of each OD traffic demand to obtain a stationary time series includes:
respectively carrying out trend line removing processing on the time series of each OD traffic demand and judging whether the time series after the trend line removing processing is a stable time series,
if the time sequence after the trend line removing processing is not the stable time sequence, carrying out differential processing on the time sequence after the trend line removing processing, judging whether the time sequence after the differential processing is the stable time sequence, if the time sequence after the differential processing is not the stable time sequence, continuing the differential processing until the time sequence after the differential processing is the stable time sequence or the differential times reach a preset differential times threshold value, and obtaining the stable time sequence; otherwise, the time sequence after the trend line removing processing is the stable time sequence.
Preferably, wherein
Respectively carrying out trend line removing processing on the time sequence of each OD traffic demand by using a detrend function;
judging whether the time sequence subjected to the trend line removing processing is a stable time sequence or not by utilizing an adftest function;
and carrying out differential processing on the time series after the trend line removing processing by using a diff function.
According to another aspect of the present invention, there is provided a system for predicting traffic volume based on highway toll data, the system including:
the integrated networking charging data acquisition unit is used for integrating the networking charging data of the expressway in the preset time period according to the time granularity threshold value and acquiring the integrated networking charging data, wherein the networking charging data comprises: time data, entrance code data, exit code data and direction of the vehicle entering the highway;
an OD traffic demand matrix determining unit, configured to convert the collected networking charging data into at least one OD traffic demand matrix according to a certain index rule;
and the traffic demand prediction unit is used for generating an OD traffic demand time sequence by using the at least one OD traffic demand matrix, and predicting the OD traffic demand in the future time period by adopting an autoregressive integrated moving average model ARIMA.
Preferably, the integrated networked charging data obtaining unit performs integrated processing on the networked charging data of the expressway in the preset time period according to the time granularity threshold value to obtain integrated networked charging data, and includes:
converting time data of vehicles entering the expressway in the networking charging data from a field format into a float type to acquire first data;
performing precision truncation processing on time data of vehicles entering the expressway in the first data according to a time granularity threshold to obtain second data;
converting time data of vehicles entering the expressway in the second data into a field format to obtain third data;
and respectively adding a timestamp to the time data of the third data when the vehicle enters the expressway according to a preset period threshold value to serve as an index of the time data of the vehicle entering the expressway, and converting the third data added with the index to obtain the integrated networking charging data.
Preferably, wherein
Converting time data of vehicles entering the expressway in the networking charging data from a field format into a float type by using a cast function, and acquiring first data;
and converting the time data in the second data into a field format by using a cast function, acquiring third data, and converting the third data into aggregated networking charging data which meets the time granularity requirement and is grouped based on database entry codes and exit codes by using a COUNT function and a group by statement.
Preferably, the OD traffic demand matrix determining unit converts the aggregated networking charging data into at least one OD traffic demand matrix according to a certain index rule, and includes:
numbering the entrance and the exit of the expressway according to the driving direction of the vehicle, forming OD pairs according to the numbers, and establishing an index mapping relation between each OD pair in an OD traffic demand matrix and an entrance code and an exit code in a database;
and extracting and collecting the collected networking charging data and putting the networking charging data into corresponding positions of the OD traffic demand matrixes to form at least one OD traffic demand matrix within a preset period threshold, wherein the row number of each OD traffic demand matrix corresponds to the number of an OD pair, and the column number of each OD traffic demand matrix corresponds to the time stamp of the time data when the vehicle enters the expressway.
Preferably, the traffic demand prediction unit, which generates an OD traffic demand time series by using the at least one OD traffic demand matrix and predicts OD traffic volumes of future time periods by using an autoregressive integrated moving average model ARIMA, includes:
converting the at least one OD traffic demand matrix data into historical traffic demand time-series data of each OD pair;
respectively carrying out trend line removing processing on the historical traffic demand time sequence data of each OD to obtain a stationary time sequence;
establishing a test model, calculating an information criterion AIC value, and selecting a model with the minimum AIC value as a prediction model;
and carrying out reduction difference processing by using an autoregressive integrated moving average model ARIMA to predict OD traffic volume of a future time period.
Preferably, wherein
Adopting an armax function to establish a test model, calculating an information criterion AIC value by using an AIC function, and selecting the model with the minimum AIC value as a prediction model;
and (3) performing reduction difference processing by using a prediction function and applying an autoregressive integrated moving average model ARIMA to predict OD traffic volume of a future time period.
Preferably, the performing a trendline removing process on the time series of each OD traffic demand to obtain a stationary time series includes:
respectively carrying out trend line removing processing on the time series of each OD traffic demand and judging whether the time series after the trend line removing processing is a stable time series,
if the time sequence after the trend line removing processing is not the stable time sequence, carrying out differential processing on the time sequence after the trend line removing processing, judging whether the time sequence after the differential processing is the stable time sequence, if the time sequence after the differential processing is not the stable time sequence, continuing the differential processing until the time sequence after the differential processing is the stable time sequence or the differential times reach a preset differential times threshold value, and obtaining the stable time sequence; otherwise, the time sequence after the trend line removing processing is the stable time sequence.
Preferably, wherein
Respectively carrying out trend line removing processing on the time sequence of each OD traffic demand by using a detrend function;
judging whether the time sequence subjected to the trend line removing processing is a stable time sequence or not by utilizing an adftest function;
and carrying out differential processing on the time series after the trend line removing processing by using a diff function.
The method and the system for predicting the traffic volume based on the highway toll data extract an OD traffic demand matrix from the highway networking toll data, predict future OD traffic demands, help managers to master the development situation of the traffic demands in time, early warn the situations of sudden increase of the traffic demands and the like, pre-judge whether the capacity of the highway is overloaded or not, lay a data foundation for efficient development of daily business of the highway, reasonable formulation of policy measures, capital release of construction projects, increase of social and economic benefits and the like, and provide decision reference data.
Drawings
A more complete understanding of exemplary embodiments of the present invention may be had by reference to the following drawings in which:
fig. 1 is a flow chart of a method 100 for predicting traffic volume based on highway toll data according to an embodiment of the present invention;
FIG. 2 is a flow diagram of a method 200 of obtaining aggregated networking charging data, according to an embodiment of the present invention;
FIG. 3 is a flow chart of a method 300 of obtaining an OD traffic demand matrix map according to an embodiment of the invention;
FIG. 4 is a flow chart of a method 400 of predicting OD traffic demand for a future time period in accordance with an embodiment of the present invention;
FIG. 5 is a schematic diagram illustrating the conversion of an OD traffic demand matrix into a time series during day-by-day prediction according to an embodiment of the present invention;
FIG. 6 is a schematic diagram illustrating the OD traffic demand matrix converted into a time series when predicting in units of time granularity according to an embodiment of the present invention; and
fig. 7 is a schematic diagram illustrating a system 700 for predicting traffic volume based on highway toll data according to an embodiment of the present invention.
Detailed Description
The exemplary embodiments of the present invention will now be described with reference to the accompanying drawings, however, the present invention may be embodied in many different forms and is not limited to the embodiments described herein, which are provided for complete and complete disclosure of the present invention and to fully convey the scope of the present invention to those skilled in the art. The terminology used in the exemplary embodiments illustrated in the accompanying drawings is not intended to be limiting of the invention. In the drawings, the same units/elements are denoted by the same reference numerals.
Unless otherwise defined, 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. Further, it will be 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 relevant art and will not be interpreted in an idealized or overly formal sense.
Fig. 1 is a flow chart of a method 100 for predicting traffic volume based on highway toll data according to an embodiment of the present invention. As shown in fig. 1, a method 100 for predicting traffic volume based on highway toll data according to an embodiment of the present invention performs centralized processing on networking toll data of a highway within a preset time period according to a time granularity threshold, and obtains the centralized networking toll data; converting the integrated networking charging data into at least one OD traffic demand matrix according to a certain index rule; the method comprises the steps of generating an OD traffic demand time sequence by utilizing the at least one OD traffic demand matrix, predicting OD traffic demand in a future time period by adopting an autoregressive integral sliding average model ARIMA, helping a manager to master the development situation of traffic demand in time, carrying out early warning on the situation of sudden increase of traffic demand and the like, and prejudging whether the capacity of the expressway is overloaded or not, thereby laying a data foundation for efficient development of daily business of the expressway, reasonable formulation of policy measures, capital release of construction projects, increase of social and economic benefits and the like, and providing decision reference data. In an embodiment of the present invention, in step 101, the networking charging data of the expressway within a preset time period is subjected to aggregation processing according to a time granularity threshold, and aggregated networking charging data is obtained, where the networking charging data includes: time data, entrance code data, exit code data, and direction of entry of the vehicle into the highway.
Fig. 2 is a flow diagram of a method 200 of obtaining aggregated networking charging data, according to an embodiment of the invention. As shown in fig. 2, the networked toll data of the expressway in the preset time period is subjected to aggregation processing according to the time granularity threshold, the acquired aggregated networked toll data is started from step 1011, and time data of vehicles entering the expressway in the networked toll data is converted into a float type from a field format in step 1011 to acquire first data.
Preferably, in step 1012, the precision cut-off processing is performed on the time data of the vehicle entering the expressway in the first data according to the time granularity threshold, so as to obtain the second data.
Preferably, in step 1013, the time data of the vehicle entering the highway in the second data is converted into a field format, and the third data is obtained.
Preferably, in step 1014, a timestamp is added to the time data of the vehicle entering the highway according to the preset period threshold value, and the third data after the addition of the timestamp is used as an index of the time data of the vehicle entering the highway, and the third data after the addition of the index is converted to obtain the aggregated networking charging data.
Preferably, wherein
Converting time data of vehicles entering the expressway in the networking charging data from a field format into a float type by using a cast function, and acquiring first data;
and converting the time data in the second data into a field format by using a cast function, acquiring third data, and converting the third data into aggregated networking charging data which meets the time granularity requirement and is grouped based on database entry codes and exit codes by using a COUNT function and a group by statement.
Preferably, the aggregated networked charging data is converted into at least one OD traffic demand matrix according to certain indexing rules in step 102.
Fig. 3 is a flowchart of a method 300 for obtaining an OD traffic demand matrix map according to an embodiment of the invention. As shown in fig. 3, the method 300 for obtaining an OD traffic demand matrix map starts at step 301, and at step 1021, the entrance and exit of the highway are numbered according to the driving direction of the vehicle, and OD pairs are formed according to the numbers, and an index mapping relationship between each OD pair in the OD traffic demand matrix and the entrance code and the exit code in the database is established.
Preferably, in step 1022, according to the index mapping relationship, the collected and counted networking charging data is extracted and placed in a corresponding position of the OD traffic demand matrix, so as to form at least one OD traffic demand matrix within a preset period threshold, wherein the row number of each OD traffic demand matrix corresponds to the number of the OD pair, and the column number of each OD traffic demand matrix corresponds to the timestamp of the time data when the vehicle enters the expressway.
Preferably, the at least one OD traffic demand matrix is used to generate an OD traffic demand time series, and an autoregressive integrated moving average model ARIMA is used to predict OD traffic demand in a future time period in step 103.
Fig. 4 is a flow chart of a method 400 of predicting OD traffic demand for a future time period in accordance with an embodiment of the present invention. As shown in fig. 4, the OD traffic demand time series is generated by using the at least one OD traffic demand matrix, and the OD traffic volume prediction for the future time period is performed by using an autoregressive integrated moving average model ARIMA, starting from step 1031, and the at least one OD traffic demand matrix data is converted into historical traffic demand time series data of each OD pair in step 1031.
In step 1032, the historical traffic demand time series data of each OD pair is subjected to a trendline removing process to obtain a stationary time series.
In step 1033, a test model is built, the AIC value of the information criterion is calculated, and the model with the minimum AIC value is selected as a prediction model.
The autoregressive integrated moving average model ARIMA is used to perform a reduction difference process at step 1034 to predict OD traffic volume in a future time period.
Preferably, wherein
Adopting an armax function to establish a test model, calculating an information criterion AIC value by using an AIC function, and selecting the model with the minimum AIC value as a prediction model;
and (3) performing reduction difference processing by using a prediction function and applying an autoregressive integrated moving average model ARIMA to predict OD traffic volume of a future time period.
Preferably, the performing a trendline removing process on the time series of each OD traffic demand to obtain a stationary time series includes:
respectively carrying out trend line removing processing on the time series of each OD traffic demand and judging whether the time series after the trend line removing processing is a stable time series,
if the time sequence after the trend line removing processing is not the stable time sequence, carrying out differential processing on the time sequence after the trend line removing processing, judging whether the time sequence after the differential processing is the stable time sequence, if the time sequence after the differential processing is not the stable time sequence, continuing the differential processing until the time sequence after the differential processing is the stable time sequence or the differential times reach a preset differential times threshold value, and obtaining the stable time sequence; otherwise, the time sequence after the trend line removing processing is the stable time sequence.
Preferably, wherein
Respectively carrying out trend line removing processing on the time sequence of each OD traffic demand by using a detrend function;
judging whether the time sequence subjected to the trend line removing processing is a stable time sequence or not by utilizing an adftest function;
and carrying out differential processing on the time series after the trend line removing processing by using a diff function.
The following specifically exemplifies embodiments of the present invention
In an embodiment of the present invention, a method of predicting traffic volume based on highway toll data first counts networked toll data into aggregated data with a time granularity threshold of every 10 seconds, every 5 minutes, every hour, or other time interval according to vehicle entry highway time, entrance ID and exit ID, and according to time granularity requirements. The method specifically comprises the following steps:
step 1: the field format of the network charging data into the high speed time is converted from datatime to float using the cast function, and after conversion, its bits represent that the day is the next few days from 1 month 1 day 1900.
Step 2: for convenience of centralized counting, the converted entry time t needs to be counted according to the time granularityentAnd (3) performing precision truncation processing:
t＝[tent×24×60×60÷tdur]×tdur÷60÷60÷24，
wherein, the [ alpha ], [ beta ]]The method is characterized in that the inner number of the method is rounded and rounded, t is the entering time after precision truncation, and tdurTo convert to time granularity in seconds; t is tentIs the entry time before the precision phase.
Step 3: converting the entry time field format after precision truncation into datatime from float by using cast function to obtain the precision ddurTime of (d).
Step 4: adding a time stamp tag to the time series of each day as an index of time:
wherein the content of the first and second substances,
indicating rounding down its inner digits.
Step 5: converting the networking charging data into time granularity d according to different import and export IDs (identity), directions and the intercepted entry time BY using a count function and a SELECT statement containing a GROUP BY clausedurThe aggregate data of (1).
Secondly, the collected networking charging data is converted into an OD traffic demand matrix according to a certain index rule so as to carry out subsequent OD traffic demand prediction or historical traffic flow simulation and other applications. The OD traffic demand matrix generation specifically comprises the following steps:
step 1: and numbering the import and export ramps from upstream to downstream in sequence according to the positions of the import and export ramps connected on the expressway, and corresponding the numbers to the IDs of the import and export ramps in the networking toll database one by one. And then, the import and export which can form the OD pairs are listed in pairs according to the number in the previous step, and the numbers are numbered from 1, and the number is marked as K. Thus, an index mapping relation between the rows of the OD traffic demand matrix and the ID of the entrance and the exit of the expressway is established.
Step 2: according to the index relation, an MATLAB construction statement is used for reading an SQLServer database, collected networking charging data is extracted and placed in corresponding positions of a matrix, an OD traffic demand matrix of each day is formed, the row number R of the matrix corresponds to the number of OD pairs, and the column number C corresponds to the timestamp of the day.
And finally, predicting the OD traffic demand at a certain time point or time period in the future by using an autoregressive moving average model, namely ARIMA (p, d, q), and utilizing the time sequence of the historical OD traffic demands. The method specifically comprises the following steps:
step 1: the OD traffic demand matrix is converted into a time sequence of historical demands, and the two modes are divided. The first is predicted in days. Fig. 5 is a schematic diagram illustrating conversion of an OD traffic demand matrix into a time series during day-by-day prediction according to an embodiment of the present invention. As shown in fig. 5, the whole matrix is regarded as a time step, R × C time series are formed by using a certain amount of historical OD traffic demand matrix data in the same row and the same column in a time increasing order, and the converted time series of the individual OD demand for traffic is 1 day apart and has a length of N. The mode is suitable for predicting the OD traffic demand all day long, and the effect is reduced when short-time traffic flow prediction is carried out. The second is prediction in units of temporal granularity. Fig. 6 is a schematic diagram illustrating the OD traffic demand matrix converted into a time series when predicting in units of time granularity according to the embodiment of the present invention.
As shown in fig. 6, with a single time granularity as a time step, one or more pieces of historical OD traffic demand matrix data are grouped into R time series in the same row according to a time increasing order, and the converted time series of the single OD on the traffic demand are spaced by the single time granularity and have a length of N × C. The mode is suitable for short-term traffic flow prediction, and the effect is not good when the prediction is carried out on all-day traffic flow.
Step 2: and (3) performing trend line removing processing on the time sequence data of the OD traffic demand by using a detrend function of the MATLAB, judging whether the time sequence data is a stable time sequence or not by using an addtest function, and if not, differentiating the time sequence by using a diff function, and judging whether the time sequence is stable or not until the time sequence becomes the stable sequence or the preset difference upper limit is reached. The number of differences d is noted.
Step 3: and (3) setting respective upper limits P and Q of the lag lengths of the autoregressive corresponding partial correlation coefficient PACF and the moving average corresponding autocorrelation coefficient ACF, establishing models with different lag lengths through an armax function, calculating AIC values by using AIC functions, selecting the model with the minimum AIC value as a prediction model, and recording the autoregressive order P and the moving average order Q.
Step 4: and (3) performing 1-step or multi-step prediction on the OD traffic demand data by using a prediction function and applying an ARIMA (p, d, q) model, and performing reduction differential processing to obtain the OD traffic demand data at a certain time point or time period in the future. And respectively storing the p, d and q of each time sequence so as to skip the steps of model identification and order determination when prediction is carried out again in a period of time, and directly applying the ARIMA model for prediction, thereby saving time.
Fig. 7 is a schematic diagram illustrating a system 700 for predicting traffic volume based on highway toll data according to an embodiment of the present invention. As shown in fig. 7, a system 700 for predicting traffic volume based on highway toll data according to an embodiment of the present invention includes: the system comprises an integrated networking charging data acquisition unit 701, an OD traffic demand matrix determination unit 702 and a traffic demand prediction unit 703. Preferably, in the collected networked charging data obtaining unit 701, the collected networked charging data of the expressway in the preset time period is collected and processed according to the time granularity threshold, so as to obtain the collected networked charging data, where the collected networked charging data includes: time data, entrance code data, exit code data, and direction of entry of the vehicle into the highway.
Preferably, the collected and counted networking charging data obtaining unit 701 performs a collection processing on the networking charging data of the expressway within a preset time period according to a time granularity threshold, and obtains the collected and counted networking charging data, including:
converting time data of vehicles entering the expressway in the networking charging data from a field format into a float type to acquire first data;
performing precision truncation processing on time data of vehicles entering the expressway in the first data according to a time granularity threshold to obtain second data;
converting time data of vehicles entering the expressway in the second data into a field format to obtain third data;
and respectively adding a timestamp to the time data of the third data when the vehicle enters the expressway according to a preset period threshold value to serve as an index of the time data of the vehicle entering the expressway, and converting the third data added with the index to obtain the integrated networking charging data.
Preferably, wherein
Converting time data of vehicles entering the expressway in the networking charging data from a field format into a float type by using a cast function, and acquiring first data;
and converting the time data in the second data into a field format by using a cast function, acquiring third data, and converting the third data into aggregated networking charging data which meets the time granularity requirement and is grouped based on database entry codes and exit codes by using a COUNT function and a group by statement.
Preferably, at the OD traffic demand matrix determining unit 702, the aggregated networked charging data is converted into at least one OD traffic demand matrix according to a certain index rule.
Preferably, the OD traffic demand matrix determining unit 702 converts the collected networked charging data into at least one OD traffic demand matrix according to a certain index rule, including:
numbering the entrance and the exit of the expressway according to the driving direction of the vehicle, forming OD pairs according to the numbers, and establishing an index mapping relation between each OD pair in an OD traffic demand matrix and an entrance code and an exit code in a database;
and extracting and collecting the collected networking charging data and putting the networking charging data into corresponding positions of the OD traffic demand matrixes to form at least one OD traffic demand matrix within a preset period threshold, wherein the row number of each OD traffic demand matrix corresponds to the number of an OD pair, and the column number of each OD traffic demand matrix corresponds to the time stamp of the time data when the vehicle enters the expressway.
Preferably, in the traffic demand prediction unit 703, the OD traffic demand time series is generated by using the at least one OD traffic demand matrix, and an autoregressive integrated moving average model ARIMA is used to predict the OD traffic demand in the future time period.
Preferably, the traffic demand prediction unit 703, generating an OD traffic demand time series by using the at least one OD traffic demand matrix, and predicting OD traffic volumes in a future time period by using an autoregressive integrated moving average model ARIMA, includes:
converting the at least one OD traffic demand matrix data into historical traffic demand time-series data of each OD pair;
respectively carrying out trend line removing processing on the historical traffic demand time sequence data of each OD to obtain a stationary time sequence;
establishing a test model, calculating an information criterion AIC value, and selecting a model with the minimum AIC value as a prediction model;
and carrying out reduction difference processing by using an autoregressive integrated moving average model ARIMA to predict OD traffic volume of a future time period.
Preferably, wherein
Adopting an armax function to establish a test model, calculating an information criterion AIC value by using an AIC function, and selecting the model with the minimum AIC value as a prediction model;
and (3) performing reduction difference processing by using a prediction function and applying an autoregressive integrated moving average model ARIMA to predict OD traffic volume of a future time period.
Preferably, the performing a trendline removing process on the time series of each OD traffic demand to obtain a stationary time series includes:
respectively carrying out trend line removing processing on the time series of each OD traffic demand and judging whether the time series after the trend line removing processing is a stable time series,
if the time sequence after the trend line removing processing is not the stable time sequence, carrying out differential processing on the time sequence after the trend line removing processing, judging whether the time sequence after the differential processing is the stable time sequence, if the time sequence after the differential processing is not the stable time sequence, continuing the differential processing until the time sequence after the differential processing is the stable time sequence or the differential times reach a preset differential times threshold value, and obtaining the stable time sequence; otherwise, the time sequence after the trend line removing processing is the stable time sequence.
Preferably, wherein
Respectively carrying out trend line removing processing on the time sequence of each OD traffic demand by using a detrend function;
judging whether the time sequence subjected to the trend line removing processing is a stable time sequence or not by utilizing an adftest function;
and carrying out differential processing on the time series after the trend line removing processing by using a diff function.
The system 700 for predicting traffic volume based on highway toll data according to the embodiment of the present invention corresponds to the method 100 for predicting traffic volume based on highway toll data according to another embodiment of the present invention, and will not be described herein again.
The invention has been described with reference to a few embodiments. However, other embodiments of the invention than the one disclosed above are equally possible within the scope of the invention, as would be apparent to a person skilled in the art from the appended patent claims.
Generally, all terms used in the claims are to be interpreted according to their ordinary meaning in the technical field, unless explicitly defined otherwise herein. All references to "a/an/the [ device, component, etc ]" are to be interpreted openly as referring to at least one instance of said device, component, etc., unless explicitly stated otherwise. The steps of any method disclosed herein do not have to be performed in the exact order disclosed, unless explicitly stated.
Claims (14)
1. A method for predicting traffic volume based on highway toll data, the method comprising:
the method comprises the steps of performing centralized processing on networking charging data of a highway in a preset time period according to a time granularity threshold value, and acquiring the centralized networking charging data, wherein the networking charging data comprises the following steps: time data, entrance code data, exit code data and direction of the vehicle entering the highway;
converting the integrated networking charging data into at least one OD traffic demand matrix according to a certain index rule;
generating an OD traffic demand time sequence by using the at least one OD traffic demand matrix, and predicting OD traffic demand in a future time period by using an autoregressive integrated moving average model (ARIMA);
wherein, the converting the collected networking charging data into at least one OD traffic demand matrix according to a certain index rule comprises:
numbering the entrance and the exit of the expressway according to the driving direction of the vehicle, forming OD pairs according to the numbers, and establishing an index mapping relation between each OD pair in an OD traffic demand matrix and an entrance code and an exit code in a database;
and extracting and collecting the collected networking charging data and putting the networking charging data into corresponding positions of the OD traffic demand matrixes to form at least one OD traffic demand matrix within a preset period threshold, wherein the row number of each OD traffic demand matrix corresponds to the number of an OD pair, and the column number of each OD traffic demand matrix corresponds to the time stamp of the time data when the vehicle enters the expressway.
2. The method according to claim 1, wherein the aggregating the networking charging data of the expressway within the preset time period according to the time granularity threshold to obtain the aggregated networking charging data comprises:
converting time data of vehicles entering the expressway in the networking charging data from a field format into a float type to acquire first data;
performing precision truncation processing on time data of vehicles entering the expressway in the first data according to a time granularity threshold to obtain second data;
converting time data of vehicles entering the expressway in the second data into a field format to obtain third data;
and respectively adding a timestamp to the time data of the third data when the vehicle enters the expressway according to a preset period threshold value to serve as an index of the time data of the vehicle entering the expressway, and converting the third data added with the index to obtain the integrated networking charging data.
3. The method of claim 2,
converting time data of vehicles entering the expressway in the networking charging data from a field format into a float type by using a cast function, and acquiring first data;
and converting the time data in the second data into a field format by using a cast function, acquiring third data, and converting the third data into aggregated networking charging data which meets the time granularity requirement and is grouped based on database entry codes and exit codes by using a COUNT function and a group by statement.
4. The method of claim 1, wherein the generating an OD traffic demand time series using the at least one OD traffic demand matrix and predicting OD traffic volumes for future time periods using an autoregressive integrated moving average model ARIMA comprises:
converting the at least one OD traffic demand matrix data into historical traffic demand time-series data of each OD pair;
respectively carrying out trend line removing processing on the historical traffic demand time sequence data of each OD pair to obtain a stationary time sequence;
establishing a test model, calculating an information criterion AIC value, and selecting a model with the minimum AIC value as a prediction model;
and carrying out reduction difference processing by using an autoregressive integrated moving average model ARIMA to predict OD traffic volume of a future time period.
5. The method of claim 4,
adopting an armax function to establish a test model, calculating an information criterion AIC value by using an AIC function, and selecting the model with the minimum AIC value as a prediction model;
and (3) performing reduction difference processing by using a prediction function and applying an autoregressive integrated moving average model ARIMA to predict OD traffic volume of a future time period.
6. The method of claim 4, wherein the separately de-trending each OD traffic demand time series to obtain a stationary time series comprises:
respectively carrying out trend line removing processing on the time series of each OD traffic demand and judging whether the time series after the trend line removing processing is a stable time series,
if the time sequence after the trend line removing processing is not the stable time sequence, carrying out differential processing on the time sequence after the trend line removing processing, judging whether the time sequence after the differential processing is the stable time sequence, if the time sequence after the differential processing is not the stable time sequence, continuing the differential processing until the time sequence after the differential processing is the stable time sequence or the differential times reach a preset differential times threshold value, and obtaining the stable time sequence; otherwise, the time sequence after the trend line removing processing is the stable time sequence.
7. The method of claim 6,
respectively carrying out trend line removing processing on the time sequence of each OD traffic demand by using a detrend function;
judging whether the time sequence subjected to the trend line removing processing is a stable time sequence or not by utilizing an adftest function;
and carrying out differential processing on the time series after the trend line removing processing by using a diff function.
8. A system for predicting traffic volume based on highway toll data, the system comprising:
the integrated networking charging data acquisition unit is used for integrating the networking charging data of the expressway in the preset time period according to the time granularity threshold value and acquiring the integrated networking charging data, wherein the networking charging data comprises: time data, entrance code data, exit code data and direction of the vehicle entering the highway;
an OD traffic demand matrix determining unit, configured to convert the collected networking charging data into at least one OD traffic demand matrix according to a certain index rule;
the traffic demand prediction unit is used for generating an OD traffic demand time sequence by using the at least one OD traffic demand matrix, and predicting OD traffic demand in a future time period by adopting an autoregressive integral moving average model ARIMA;
wherein, the OD traffic demand matrix determining unit converts the collected networking charging data into at least one OD traffic demand matrix according to a certain index rule, including:
numbering the entrance and the exit of the expressway according to the driving direction of the vehicle, forming OD pairs according to the numbers, and establishing an index mapping relation between each OD pair in an OD traffic demand matrix and an entrance code and an exit code in a database;
and extracting and collecting the collected networking charging data and putting the networking charging data into corresponding positions of the OD traffic demand matrixes to form at least one OD traffic demand matrix within a preset period threshold, wherein the row number of each OD traffic demand matrix corresponds to the number of an OD pair, and the column number of each OD traffic demand matrix corresponds to the time stamp of the time data when the vehicle enters the expressway.
9. The system according to claim 8, wherein the aggregated networked charging data obtaining unit performs aggregated processing on the networked charging data of the expressway within the preset time period according to the time granularity threshold value to obtain the aggregated networked charging data, and includes:
converting time data of vehicles entering the expressway in the networking charging data from a field format into a float type to acquire first data;
performing precision truncation processing on time data of vehicles entering the expressway in the first data according to a time granularity threshold to obtain second data;
converting time data of vehicles entering the expressway in the second data into a field format to obtain third data;
and respectively adding a timestamp to the time data of the third data when the vehicle enters the expressway according to a preset period threshold value to serve as an index of the time data of the vehicle entering the expressway, and converting the third data added with the index to obtain the integrated networking charging data.
10. The system of claim 9,
converting time data of vehicles entering the expressway in the networking charging data from a field format into a float type by using a cast function, and acquiring first data;
and converting the time data in the second data into a field format by using a cast function, acquiring third data, and converting the third data into aggregated networking charging data which meets the time granularity requirement and is grouped based on database entry codes and exit codes by using a COUNT function and a group by statement.
11. The system of claim 8, wherein the traffic demand prediction unit generates an OD traffic demand time series using the at least one OD traffic demand matrix, and predicts OD traffic volumes for future time periods using an autoregressive integrated moving average model ARIMA, comprising:
converting the at least one OD traffic demand matrix data into historical traffic demand time-series data of each OD pair;
respectively carrying out trend line removing processing on the historical traffic demand time sequence data of each OD to obtain a stationary time sequence;
establishing a test model, calculating an information criterion AIC value, and selecting a model with the minimum AIC value as a prediction model;
and carrying out reduction difference processing by using an autoregressive integrated moving average model ARIMA to predict OD traffic volume of a future time period.
12. The system of claim 11,
adopting an armax function to establish a test model, calculating an information criterion AIC value by using an AIC function, and selecting the model with the minimum AIC value as a prediction model;
and (3) performing reduction difference processing by using a prediction function and applying an autoregressive integrated moving average model ARIMA to predict OD traffic volume of a future time period.
13. The system of claim 11, wherein the separately de-trending each OD traffic demand time series to obtain a stationary time series comprises:
respectively carrying out trend line removing processing on the time series of each OD traffic demand and judging whether the time series after the trend line removing processing is a stable time series,
if the time sequence after the trend line removing processing is not the stable time sequence, carrying out differential processing on the time sequence after the trend line removing processing, judging whether the time sequence after the differential processing is the stable time sequence, if the time sequence after the differential processing is not the stable time sequence, continuing the differential processing until the time sequence after the differential processing is the stable time sequence or the differential times reach a preset differential times threshold value, and obtaining the stable time sequence; otherwise, the time sequence after the trend line removing processing is the stable time sequence.
14. The system of claim 13,
respectively carrying out trend line removing processing on the time sequence of each OD traffic demand by using a detrend function;
judging whether the time sequence subjected to the trend line removing processing is a stable time sequence or not by utilizing an adftest function;
and carrying out differential processing on the time series after the trend line removing processing by using a diff function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711176723.7A CN107798877B (en) | 2017-11-22 | 2017-11-22 | Method and system for predicting traffic volume based on highway charging data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711176723.7A CN107798877B (en) | 2017-11-22 | 2017-11-22 | Method and system for predicting traffic volume based on highway charging data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107798877A CN107798877A (en) | 2018-03-13 |
| CN107798877B true CN107798877B (en) | 2020-10-02 |
Family
ID=61536373
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711176723.7A Active CN107798877B (en) | 2017-11-22 | 2017-11-22 | Method and system for predicting traffic volume based on highway charging data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107798877B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108615360B (en) * | 2018-05-08 | 2022-02-11 | 东南大学 | Traffic demand day-to-day evolution prediction method based on neural network |
| CN108805623B (en) * | 2018-06-11 | 2021-08-24 | 浙江工业大学 | Supply prediction method for fast-selling products |
| CN112102613A (en) * | 2020-08-12 | 2020-12-18 | 东南大学 | Method and system for predicting passenger and cargo traffic volume on expressway |
| CN113537555B (en) * | 2021-06-03 | 2023-04-11 | 太原理工大学 | Traffic sub-region model prediction sliding mode boundary control method considering disturbance |
| CN113593243B (en) * | 2021-09-28 | 2021-12-07 | 中铁第五勘察设计院集团有限公司 | Method and device for predicting toll vehicle type traffic volume in toll road operation period |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4038970B2 (en) * | 2000-08-01 | 2008-01-30 | 住友電気工業株式会社 | Link generation traffic calculation method, OD traffic estimation method and apparatus |
| CN102097004A (en) * | 2011-01-31 | 2011-06-15 | 上海美慧软件有限公司 | Mobile phone positioning data-based traveling origin-destination (OD) matrix acquisition method |
| CN103854472A (en) * | 2012-12-05 | 2014-06-11 | 深圳先进技术研究院 | Taxi cloud-intelligent scheduling method and system |
| CN104183119A (en) * | 2014-08-19 | 2014-12-03 | 中山大学 | Real-time traffic flow distribution prediction system based on road section OD backstepping |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8949394B1 (en) * | 2003-07-10 | 2015-02-03 | Sprint Communications Company L.P. | Forecasting link utilization between points of presence in an IP network |
| CN103903452B (en) * | 2014-03-11 | 2015-10-07 | 东南大学 | Forecasting Approach for Short-term Traffic Flow |
| CN104183134B (en) * | 2014-08-27 | 2016-02-10 | 重庆大学 | The highway short-term traffic flow forecast method of vehicle is divided based on intelligence |
| CN106327864A (en) * | 2015-07-10 | 2017-01-11 | 北京大学 | Traffic flow estimation method based on network charging data of highway |
| CN105761492B (en) * | 2016-05-04 | 2018-07-13 | 北京大学 | A kind of a wide range of highway network Dynamic Assignment method based on network flow |
| CN107123267A (en) * | 2017-06-29 | 2017-09-01 | 中国路桥工程有限责任公司 | A kind of Freeway Traffic Volume Prediction system and method |
-
2017
- 2017-11-22 CN CN201711176723.7A patent/CN107798877B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4038970B2 (en) * | 2000-08-01 | 2008-01-30 | 住友電気工業株式会社 | Link generation traffic calculation method, OD traffic estimation method and apparatus |
| CN102097004A (en) * | 2011-01-31 | 2011-06-15 | 上海美慧软件有限公司 | Mobile phone positioning data-based traveling origin-destination (OD) matrix acquisition method |
| CN103854472A (en) * | 2012-12-05 | 2014-06-11 | 深圳先进技术研究院 | Taxi cloud-intelligent scheduling method and system |
| CN104183119A (en) * | 2014-08-19 | 2014-12-03 | 中山大学 | Real-time traffic flow distribution prediction system based on road section OD backstepping |
Non-Patent Citations (1)
| Title |
|---|
| 基于OLAM的高速公路交通量多维预测研究;钱超等;《交通运输系统工程与信息》;20130415;第13卷(第2期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107798877A (en) | 2018-03-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107798877B (en) | Method and system for predicting traffic volume based on highway charging data | |
| CN107798876B (en) | Road traffic abnormal jam judging method based on event | |
| CN103150900B (en) | Traffic jam event automatic detecting method based on videos | |
| CN103886756B (en) | Based on the freeway network method for detecting operation state of OBU | |
| CN110751828B (en) | Road congestion measuring method and device, computer equipment and storage medium | |
| Wang et al. | A big data approach for smart transportation management on bus network | |
| CN101488284A (en) | Intelligent management system for road traffic condition instant prediction | |
| CN107085943B (en) | Short-term prediction method and system for road travel time | |
| CN103903433A (en) | Real-time dynamic judgment method and device for road traffic state | |
| CN113808401B (en) | Traffic congestion prediction method, device, equipment and storage medium | |
| CN108648445B (en) | Dynamic traffic situation prediction method based on traffic big data | |
| CN101540099A (en) | Method and system for judging road traffic states | |
| CN106203662A (en) | A kind of based on ETC system realize highway pass through reservation method | |
| CN110414715B (en) | Community detection-based passenger flow volume early warning method | |
| CN101465059A (en) | Early-warning system for inspecting and judging urban road traffic safety situation | |
| CN106887141A (en) | Continuous Traffic node congestion degree forecast model based on queueing theory, system and method | |
| CN113643539A (en) | Intelligent traffic control system based on big data | |
| CN111915920A (en) | Smart city public transport system construction coordination management system based on cloud computing | |
| Gore et al. | Congestion index and reliability-based freeway level of service | |
| CN114997709A (en) | Road waste water recycling governing system | |
| Yao et al. | Period division-based Markov models for short-term traffic flow prediction | |
| CN106355541A (en) | Method for recognizing difficulty in taking taxi on basis of GPS data | |
| Huang et al. | Real time delay estimation for signalized intersection using transit vehicle positioning data | |
| CN105931463A (en) | Method for calculating road traffic performance index based on traffic surface radar | |
| Song et al. | Public transportation service evaluations utilizing seoul transportation card data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |