CN108961762A - A kind of urban road traffic flow amount prediction technique based on multifactor fusion - Google Patents

Abstract

A kind of urban road traffic flow amount prediction technique based on multifactor fusion, belong to traffic flow forecasting field, solves the problems, such as that the existing urban road traffic flow amount prediction technique based on multiple linear regression analysis causes prediction accuracy poor because the magnitude of traffic flow influence factor of consideration is less.Compared with the existing urban road traffic flow amount prediction technique based on multiple linear regression analysis, urban road traffic flow amount prediction technique of the present invention based on multifactor fusion additionally considers a variety of magnitude of traffic flow influence factors, including but not limited to government's large-scale activity information, civil large-scale activity information, the interim planning information of urban road and municipal administration watering clean road information, and then improve the prediction accuracy of itself to a certain extent.

Description

A prediction method of urban road traffic flow based on multi-factor fusion

technical field

The invention relates to a method for predicting urban road traffic flow, which belongs to the field of traffic flow prediction.

Background technique

Traffic problems have become a global urban common problem, seriously affecting the city's operational efficiency and economic development, and traffic congestion, as the main manifestation of urban traffic diseases, has caused a huge impact on people's life and work. The economic, safety and environmental losses caused by traffic congestion have already attracted widespread attention from the society, and the research on related scientific issues of urban traffic congestion phenomena and laws has become one of the hotspots in the field of transportation. Based on this background, various urban road traffic flow forecasting methods emerged as the times require.

The urban road traffic flow forecasting method based on multiple linear regression analysis is a common urban road traffic flow forecasting method. The basic principle of this method is: take the urban road traffic flow as the dependent variable, take the influencing factors of the traffic flow as the independent variables, and use multiple independent variables to explain the changes of the dependent variable, finally according to the multiple influencing factors of the traffic flow and the urban road traffic The linear relationship between flows predicts urban road traffic flow in future time periods.

However, the existing urban road traffic flow forecasting methods based on multiple linear regression analysis usually only include holidays and bad weather, and do not take into account other traffic flow influencing factors. This leads to the poor prediction accuracy of the existing urban road traffic flow prediction method based on multiple linear regression analysis, which needs to be further improved.

Contents of the invention

In order to solve the problem that the existing urban road traffic flow prediction method based on multiple linear regression analysis has poor prediction accuracy due to the consideration of fewer traffic flow influencing factors, the present invention proposes an urban road traffic flow based on multi-factor fusion method of prediction.

The urban road traffic flow prediction method based on multi-factor fusion of the present invention comprises:

Step 1. Obtain the target road traffic flow data set within a predetermined historical time period;

Step 2, extracting a plurality of target road traffic flow data samples from the target road traffic flow data set within a predetermined historical time period;

Step 3. Obtain the traffic flow influencing factors existing in the corresponding time period of each target road traffic flow data sample;

Factors affecting traffic flow include holiday information, bad weather information, large-scale government event information, private large-scale event information, temporary urban road planning information, and municipal watering and road cleaning information;

Step 4. According to the mentioned plurality of target road traffic flow data samples and their corresponding traffic flow influencing factors, generate a change curve of the target road traffic flow with the traffic flow influencing factors in the historical time period;

Step 5: Predict the target road traffic flow in the future time period according to the change curve of the target road traffic flow with the traffic flow influencing factors in the historical time period and the determined existing traffic flow influencing factors in the future time period.

Preferably, in step 2, a support vector machine algorithm is used to extract a plurality of target road traffic flow data samples from the target road traffic flow data set within a predetermined historical time period.

Preferably, in step 4, according to the mentioned plurality of target road traffic flow data samples and their corresponding traffic flow influencing factors, and based on the single factor polynomial regression model, generate the target road traffic flow in the historical time period with the traffic flow influencing factors change curve.

Preferably, the single factor polynomial regression model is based on polynomial smooth support vector sequential regression algorithm to optimize the change curve of the target road traffic flow with the traffic flow influencing factors in the historical time period generated by it.

Preferably, the single factor polynomial regression model corrects the change curve of the target road traffic flow with traffic flow influencing factors in the historical time period based on the difference between the predicted target road traffic flow and its corresponding actual target road traffic flow.

Preferably, the time period corresponding to each target road traffic flow data sample includes the previous day, the current day and the next day when the target road traffic flow data sample occurs.

Compared with the existing urban road traffic flow prediction method based on multiple linear regression analysis, the urban road traffic flow prediction method based on multi-factor fusion in the present invention additionally considers a variety of traffic flow influencing factors, including but not limited to government large-scale Activity information, civil large-scale event information, urban road temporary planning information, and municipal watering and road cleaning information, thereby improving its own prediction accuracy to a certain extent.

Description of drawings

Hereinafter, the urban road traffic flow prediction method based on multi-factor fusion will be described in more detail based on the embodiments and with reference to the accompanying drawings, wherein:

Fig. 1 is the flowchart of the urban road traffic flow prediction method based on multi-factor fusion described in the embodiment.

Detailed ways

The method for predicting urban road traffic flow based on multi-factor fusion according to the present invention will be further described below in conjunction with the accompanying drawings.

Embodiment: The present embodiment will be described in detail below in conjunction with FIG. 1 .

The urban road traffic flow prediction method based on multi-factor fusion described in this embodiment includes:

Step 1. Obtain the target road traffic flow data set within a predetermined historical time period;

Step 2, extracting a plurality of target road traffic flow data samples from the target road traffic flow data set within a predetermined historical time period;

Step 3. Obtain the traffic flow influencing factors existing in the corresponding time period of each target road traffic flow data sample;

Factors affecting traffic flow include holiday information, bad weather information, large-scale government event information, private large-scale event information, temporary urban road planning information, and municipal watering and road cleaning information;

Step 4. According to the mentioned plurality of target road traffic flow data samples and their corresponding traffic flow influencing factors, generate a change curve of the target road traffic flow with the traffic flow influencing factors in the historical time period;

Step 5: Predict the target road traffic flow in the future time period according to the change curve of the target road traffic flow with the traffic flow influencing factors in the historical time period and the determined existing traffic flow influencing factors in the future time period.

In step 2 of this embodiment, a support vector machine algorithm is used to extract a plurality of target road traffic flow data samples from the target road traffic flow data set within a predetermined historical time period.

In Step 4 of this embodiment, according to the mentioned multiple target road traffic flow data samples and their corresponding traffic flow influencing factors, and based on the single factor polynomial regression model, generate the target road traffic flow in the historical time period with the traffic flow influencing factors. Curve. The single factor polynomial regression model in this embodiment is based on the polynomial smoothing support vector sequential regression algorithm to optimize the change curve of the target road traffic flow with the traffic flow influencing factors in the historical time period generated by it.

The single-factor polynomial regression model in this embodiment corrects the change curve of the target road traffic flow with traffic flow influencing factors in the historical time period based on the difference between the predicted target road traffic flow and the corresponding actual target road traffic flow.

The time period corresponding to each target road traffic flow data sample includes the previous day, the current day and the next day when the target road traffic flow data sample occurs. Such setting fully takes into account the advance and delay of the traffic flow influencing factors themselves, and further improves the prediction accuracy of the urban road traffic flow prediction method.

Although the invention is described herein with reference to specific embodiments, it should be understood that these embodiments are merely illustrative of the principles and applications of the invention. It is therefore to be understood that numerous modifications may be made to the exemplary embodiments and that other arrangements may be devised without departing from the spirit and scope of the invention as defined by the appended claims. It shall be understood that different dependent claims and features described herein may be combined in a different way than that described in the original claims. It will also be appreciated that features described in connection with individual embodiments can be used in other described embodiments.

Claims (6)

1. a kind of urban road traffic flow amount prediction technique based on multifactor fusion, which is characterized in that the urban road is handed over Through-current capacity prediction technique includes:

Step 1: obtaining the target road traffic flow data set in the order history period；

Step 2: extracting multiple target road traffic in the target road traffic flow data set in the order history period Data on flows sample；

Step 3: obtain each target road traffic flow data sample correspond to the existing magnitude of traffic flow in the period influence because Element；

Magnitude of traffic flow influence factor includes holiday information, bad weather information, government's large-scale activity information, civil large-scale activity The interim planning information of information, urban road and municipal administration watering clean road information；

Step 4: according to the multiple target road traffic flow data samples mentioned and corresponding magnitude of traffic flow influence because Element, generate historical time section in the target road magnitude of traffic flow with magnitude of traffic flow influence factor change curve；

Step 5: according to the target road magnitude of traffic flow in historical time section with the change curve and not of magnitude of traffic flow influence factor Come in the period to determine existing magnitude of traffic flow influence factor, the target road magnitude of traffic flow in future time section be carried out pre- It surveys.

2. the urban road traffic flow amount prediction technique based on multifactor fusion as described in claim 1, which is characterized in that step Rapid two uses algorithm of support vector machine extracts multiple in the target road traffic flow data set in the order history period Target road traffic flow data sample.

3. the urban road traffic flow amount prediction technique based on multifactor fusion as claimed in claim 2, which is characterized in that step The multiple target road traffic flow data samples and corresponding magnitude of traffic flow influence factor that rapid four basis is mentioned, and base In the target road magnitude of traffic flow in single-factor polynomial regression model, generation historical time section with the change of magnitude of traffic flow influence factor Change curve.

4. the urban road traffic flow amount prediction technique based on multifactor fusion as claimed in claim 3, which is characterized in that single In the historical time section that factor polynomial regression model generates it based on moving-polynomial smoother type supporting vector ordinal regression algorithm The target road magnitude of traffic flow is optimized with the change curve of magnitude of traffic flow influence factor.

5. the urban road traffic flow amount prediction technique based on multifactor fusion as claimed in claim 4, which is characterized in that single The corresponding realistic objective road traffic flow of the target road magnitude of traffic flow of the factor polynomial regression model based on prediction Difference is modified the target road magnitude of traffic flow in historical time section with the change curve of magnitude of traffic flow influence factor.

6. the urban road traffic flow amount prediction technique based on multifactor fusion as claimed in claim 5, which is characterized in that every A target road traffic flow data sample corresponding period includes before the target road traffic flow data sample occurs One day, same day and one day after.