CN108596381B - Urban parking demand prediction method based on OD data - Google Patents

Abstract

The invention provides a parking demand prediction method based on OD data. The method comprises the following steps: obtaining OD data and parking data of a research area, and dividing parking cells of the research area according to the OD data and the parking data; performing regression analysis based on the OD data, the parking data and the parking cell partition scheme of the research area to construct a parking demand prediction model; and forecasting the parking demand of the target parking area according to the parking demand forecasting model. According to the invention, the parking demand prediction model is obtained by performing regression analysis on the OD data and the parking data of the known region, then the parking demand prediction model is used for estimating the parking demand of the unknown region, the traditional large-scale investigation is abandoned, the manpower and material resources are saved, the method has the advantages of simplicity, convenience, rapidness and accuracy, and references and technical support can be provided for parking planning, parking resource allocation and parking problem solution.

Description

Prediction method of urban parking demand based on OD data

technical field

The invention relates to the field of urban traffic and parking planning, in particular to an urban parking demand prediction method based on OD (origin, destination, starting point, ending point) data.

Background technique

In recent years, with the rapid growth of the number of motor vehicles, in many cities, the supply of parking spaces is in short supply, and the difficulty of parking has become a "growing trouble" in most cities. Parking experience has also become a key indicator of people's travel happiness.

Accurate forecasting of parking demand is the premise and foundation of the planning and construction of urban parking facilities. Excessive forecasting of parking demand will result in waste of land and capital, but too small forecast will lead to inability to meet urban parking demand, restrict social and economic development, and cause serious problems. Traffic problems.

At present, the method for calculating parking demand in the prior art needs to be based on relatively complete urban land report data, requires a lot of manpower, material resources and financial resources to conduct urban traffic travel survey, and the survey takes a long time.

Therefore, how to quickly and accurately predict the urban parking demand is an urgent problem to be solved.

SUMMARY OF THE INVENTION

Embodiments of the present invention provide a method for predicting urban parking demand based on OD data, so as to solve the problems of the prior art.

In order to achieve the above objects, the present invention adopts the following technical solutions.

A parking demand forecasting method based on OD data, comprising:

Obtain the OD data and parking data of the research area, and divide the research area into parking cells according to the OD data and the parking data;

Based on the OD data, parking data and parking area division scheme of the research area, regression analysis is performed to build a parking demand forecasting model;

According to the parking demand prediction model, the parking demand prediction is performed on the target parking area.

Further, the described acquisition of OD data and parking data of the research area includes:

Obtain the OD data of the car in the study area, the OD data includes: serial number, longitude and latitude of travel starting point O point, travel end point D point longitude and latitude, start time, end time and travel distance;

Obtain parking data of cars in the study area, including parking lot IDs, peak hours, and peak parking counts.

Further, the method also includes:

Perform preprocessing on the errors and abnormal values in the OD data and parking data, filter out the data whose field value is empty, and delete the data whose points O or D are not in the research area according to the latitude and longitude coordinate range of the research area. For data whose travel distance is less than 500 meters, delete data with travel time t _j ≤ 0min; delete data with travel speed V _j ≥ 100 km/h.

Further, the described research area is divided into parking cells according to the OD data and parking data, including:

Divide the research area into multiple grids, and the grid side lengths increase from the set value in turn, and count the number of peak parking lots, the number of points O, and the number of points D in each grid under the condition of different grid side lengths. The grids whose number of points and the number of D points are greater than the specified numerical threshold are defined as valid grids, and the total number of valid grids is denoted as m;

The correlation analysis is carried out on the number of peak parking, the number of O points and the number of D points in each valid grid, and the correlation coefficient r _O between the number of peak parking and the number of O points under the condition of different grid side lengths is calculated, and the peak value under the condition of different grid side lengths is calculated. The correlation coefficient r _D between the number of parking lots and the number of points D is determined by comprehensively considering the number of valid samples m, r _O and r _D to determine the optimal format side length, and the research area is divided into parking cells according to the optimal format side length.

Further, the regression analysis is carried out based on the OD data, parking data and the parking cell division scheme of the research area, and a parking demand prediction model is constructed, including:

After the research area is divided into parking areas according to the optimal format side length, each grid in the research area is defined as a parking area, and the number of peak parking areas and the number of O points in each parking area are determined respectively. and the number of D points;

According to the peak parking number and the number of O points in each parking area, establish a univariate linear fitting function P _i =αx _Oi between the peak parking number of the parking area and the number of O points in the study area;

According to the peak parking number and the number of D points in each parking area, establish a univariate linear fitting function P _i =βx _Di between the peak parking number of the parking area and the number of D points in the study area;

Among them, Pi is the number of peak parking spaces in the parking area _i , x _Oi is the number of points O in the parking area, x _Di is the number of points D in the parking area, and α and β are the regression coefficients of the number of points O and D in the parking area, respectively;

The univariate linear fitting function between the peak parking numbers corresponding to the research area, the number of points O and the number of points D, respectively, is used as a parking demand prediction model.

Further, the predicting the parking demand of the target parking area according to the parking demand prediction model includes:

Input the number of 0 points of the target parking area into the univariate linear fitting function between the number of peak parking and the number of 0 points in the parking demand prediction model, to obtain the peak parking number of the target parking area;

and / or,

The number of D points in the target parking area is input into the univariate linear fitting function between the number of peak parking and the number of D points in the parking demand prediction model to obtain the peak parking number in the target parking area.

It can be seen from the technical solutions provided by the above embodiments of the present invention that the method of the embodiment of the present invention obtains a parking demand prediction model by performing regression analysis on OD data and parking data in a known area, and then uses the parking demand prediction model to estimate the unknown Regional parking demand, abandoning the traditional large-scale survey, saving manpower and material resources, has the advantages of simplicity, speed and accuracy, and can provide reference and technical support for parking planning, parking resource allocation, and parking problem solving.

Additional aspects and advantages of the present invention will be set forth in part in the following description, which will be apparent from the following description, or may be learned by practice of the present invention.

Description of drawings

In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the following briefly introduces the accompanying drawings used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without any creative effort.

1 is a process flow diagram of a method for predicting urban parking demand based on OD data provided by an embodiment of the present invention;

Fig. 2 is a kind of correlation analysis diagram of the number of parking lots and the number of starting points and the number of D points with different grid side length conditions provided by an embodiment of the present invention;

3 is a schematic diagram of a parking cell division provided by an embodiment of the present invention;

FIG. 4 is a scatter diagram of the number of peak parking and the number of starting points of a parking area provided by an embodiment of the present invention;

FIG. 5 is a scatter diagram of the number of peak parking spaces and the number of D points in a parking area provided by an embodiment of the present invention.

Detailed ways

Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, but not to be construed as a limitation of the present invention.

It will be understood by those skilled in the art that the singular forms "a", "an", "the" and "the" as used herein can include the plural forms as well, unless expressly stated otherwise. It should be further understood that the word "comprising" used in the description of the present invention refers to the presence of stated features, integers, steps, operations, elements and/or components, but does not exclude the presence or addition of one or more other features, Integers, steps, operations, elements, components and/or groups thereof. It will be understood that when we refer to an element as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Furthermore, "connected" or "coupled" as used herein may include wirelessly connected or coupled. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.

It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It should also be understood that terms such as those defined in the general dictionary should be understood to have meanings consistent with their meanings in the context of the prior art and, unless defined as herein, are not to be taken in an idealized or overly formal sense. explain.

In order to facilitate the understanding of the embodiments of the present invention, the following will take several specific embodiments as examples for further explanation and description in conjunction with the accompanying drawings, and each embodiment does not constitute a limitation to the embodiments of the present invention.

A processing flow of a method for predicting parking demand based on OD data provided by an embodiment of the present invention is shown in FIG. 1 , and includes the following processing steps:

Step S110, acquire OD data and parking data of the research area.

Step S120, dividing the parking area, and determining the optimal parking area dividing scheme according to the research purpose.

In step S130, based on the OD data and parking data obtained in step S110 and the parking cell division obtained in step 2, regression analysis is performed to construct a parking demand prediction model.

Step S140: Predict the parking demand for the target parking area according to the parking demand prediction model.

In some embodiments, the specific implementation of obtaining the OD data and parking data of the research area in the aforementioned step S110 includes:

Extract the OD data of car trips in the study area. The OD data refers to a series of data describing trips, mainly including: number, O point (travel start point) latitude and longitude, D point (travel end point) longitude and latitude, start time, end time, itinerary distance. As shown in Table 1 below.

Table 1

Extract the parking data of car travel in the study area. The parking data mainly includes the parking lot ID, peak hours, and peak parking numbers.

In some embodiments, the errors and abnormal values in the OD data and parking data extracted in the foregoing step 1 are preprocessed, and the data whose field value is empty is filtered out. And according to the latitude and longitude coordinate range of the studied area, delete the data whose starting point or point D is not in this area. Considering that the service radius of public transport stations is 500m, the data with travel distance less than 500m are deleted.

The travel time and travel speed can be calculated according to the start time t _Oj , the end time t _Dj and the travel distance L _j of a trip (set as j), assuming that the travel time is t _j and the travel speed is V _j , then t _j = t _Dj -t _Oj , V _j =L _j /t _j .

According to the regeneration data calculated in the above steps, delete the data of travel time t _j ≤ 0min; delete the data of travel speed V _j ≥ 100km/h.

Those skilled in the art should understand that the above threshold values of travel time t _j and travel speed V _j are only examples, and other existing or possible future threshold values of travel time t _j and travel speed V _j are applicable to the present invention. The embodiments of the invention should also be included within the protection scope of the present invention, and are incorporated herein by reference.

In some embodiments, the specific implementation of determining the optimal parking cell division scheme according to the research purpose of performing parking cell division in the aforementioned step S120 includes:

Step 1. Divide the research area into several grids, and each grid is used as a basic research unit for parking demand forecasting, and is used to determine the unit to which the travel start point and travel end point belong.

Step 2. The grid side length R is set from the set value (such as 200m), and is incremented in units of 100m, and the number of valid grids under different R, the number of peak stops in each grid, the number of starting points and the number of D points are counted.

Step 3. Define the grids with the number of peak stops, the number of starting points, and the number of D points greater than the specified numerical threshold as valid grids, and count the number of valid grids, denoted as m. The grids for which the peak parking numbers are not obtained are defined as unknown grids.

Step 4: Calculate the correlation coefficients r _O and r _D of the number of peak stops and the number of starting points, and the number of peak stops and the number of D points under different R conditions.

Step 5: According to the research purpose, consider the effective sample numbers m, r _O and r _D to determine the optimal format side length, and divide the research area into parking cells according to the optimal format side length.

Step 6. Under the determined optimal grid side length, define each grid as a parking area, and determine the number of peak parking, the number of starting points, and the number of D points in each parking area.

2 is a correlation analysis diagram of the number of parking spaces, the number of starting points, and the number of D points with different grid side length conditions provided by an embodiment of the present invention. In some embodiments, when the optimal grid side length is selected in the foregoing step 5, the effective sample The numbers m, r _O and r _D at least satisfy: m ≥ 10, r _O and r _D ≥ 0.85.

Those skilled in the art should understand that the above-mentioned threshold values for the number of valid samples of m, r _O , and r _D are only examples, and other existing or possible threshold values for the number of valid samples in the future may be applicable to the embodiments of the present invention, It should also be included within the scope of the present invention and is hereby incorporated by reference.

In some embodiments, the specific implementation of performing regression analysis to construct a parking demand prediction model based on OD data, parking data and parking cell division in the aforementioned step S130 includes:

According to the peak parking number and the number of O points in each parking area, a univariate linear fitting function P _i =αx _Oi is established between the peak parking number of the parking area and the number of O points in the study area. The univariate linear fitting function P _i = αx _Oi is suitable for all parking areas in the study area;

According to the peak parking number and the number of D points in each parking area, a univariate linear fitting function P _i =βx _Di between the peak parking number of the parking area and the number of D points in the study area is established. The univariate linear fitting function P _i = βx _Di is suitable for all parking areas in the study area;

Among them, Pi is the number of peak parking spaces in the parking area _i , x _Oi is the number of points O in the parking area, x _Di is the number of points D in the parking area, and α and β are the regression coefficients of the number of points O and D in the parking area, respectively;

The univariate linear fitting function between the peak parking numbers corresponding to the research area, the number of points O and the number of points D, respectively, is used as a parking demand prediction model.

FIG. 3 is a schematic diagram of a parking area division provided by an embodiment of the present invention, FIG. 4 is a scatter diagram of the number of parking areas at the peak of the parking area and the number of starting points, and FIG. 5 is a scatter diagram of the number of parking areas at the peak of the parking area and the number of D points. Each grid in the study area is defined as a parking area, and the number of peak parking, the number of points O and the number of points D of each parking area are determined respectively. As shown in Figure 3. Analyze the relationship between the number of peak parking, the number of starting points, and the number of points D, as shown in Figure 4 and Figure 5, and observe the scatter plots of the number of peak parking, the number of points O, and the number of points D in the parking area, and you can preliminarily determine the peak parking in the parking area. There is a certain positive correlation between the number of O points and the number of D points, so the next correlation analysis is carried out.

According to the results of correlation analysis, according to the peak parking number and the number of O points in each parking area, a univariate linear fitting function P _i = 0.009x _Oi between the peak parking number and the number of O points in each parking area is established, a univariate linear regression model The R-square of the fitting result of the relationship between the number of peak stops and the number of O points is 0.913, and the relative error is 58.45%.

According to the peak parking number and the number of D points in each parking area, a univariate linear fitting function P _i = 0.009x _Di between the peak parking number of the parking area and the number of D points is established. The R-square of the fitting result of point number relationship is 0.965, and the relative error is 40.69%.

Among them, Pi is the number of peak parking spaces in the parking area _i , x _Oi is the number of points O in the parking area, x _Di is the number of points D in the parking area, and α and β are the regression coefficients of the number of points O and D in the parking area, respectively;

The univariate linear fitting function between the peak parking numbers corresponding to the research area, the number of points O and the number of points D, respectively, is used as a parking demand prediction model.

In some embodiments, the specific implementation of predicting the parking demand in the target parking area according to the parking demand prediction model in the foregoing step S140 includes:

Using the finally obtained parking demand prediction model and the number of points O and D in the target parking area, the peak parking number Pi in the target parking area can be _predicted , so as to predict the parking demand of the target parking area.

Input the number of 0 points of the target parking area into the univariate linear fitting function between the number of peak parking and the number of 0 points in the parking demand prediction model, to obtain the peak parking number of the target parking area;

and / or,

The number of D points in the target parking area is input into the univariate linear fitting function between the number of peak parking and the number of D points in the parking demand prediction model to obtain the peak parking number in the target parking area.

In some embodiments, a model with a larger R-square may be selected as the final parking demand prediction model by comparing the R-squares of the two univariate linear regression results.

To sum up, the method of the embodiment of the present invention obtains a parking demand prediction model by performing regression analysis on OD data and parking data in a known area, and then uses the parking demand prediction model to estimate the parking demand in an unknown area, and abandons the traditional large-scale parking requirements. The survey saves manpower, material and financial resources, has the advantages of simplicity, speed and accuracy, and can provide reference and technical support for parking planning, parking resource allocation, and parking problem solving.

Those of ordinary skill in the art can understand that the accompanying drawing is only a schematic diagram of an embodiment, and the modules or processes in the accompanying drawing are not necessarily necessary to implement the present invention.

Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the apparatus or system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and reference may be made to some descriptions of the method embodiments for related parts. The apparatus and system embodiments described above are only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, It can be located in one place, or it can be distributed over multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

The above description is only a preferred embodiment of the present invention, but the protection scope of the present invention is not limited to this. Substitutions should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims (3)

1. a parking demand forecasting method based on OD data, is characterized in that, comprises: Obtain the OD data and parking data of the study area, and divide the study area into parking cells according to the OD data and the parking data; the OD data include: serial number, travel starting point O point latitude and longitude, travel destination point D point longitude and latitude, start time , end time and travel distance; the parking data includes parking lot ID, peak time, and peak parking number; the division process includes: dividing the research area into a plurality of grids, the grid side lengths increase sequentially from the set value, and count different grids respectively. The number of peak parking, the number of points O, the number of points D in each grid under the condition of side length, the grid whose peak parking number, the number of points O, and the number of points D are all greater than the specified numerical threshold is defined as an effective grid, and the total number of valid grids is recorded as m; perform correlation analysis on the number of peak parking, the number of points O and the number of points D in each valid grid, calculate the correlation coefficient r _O between the number of peak parking and the number of points O under the conditions of different grid side lengths, and calculate the conditions of different grid side lengths. The correlation coefficient r _{D between the number of parking at the lower peak and the number of points D} , comprehensively consider the number of valid samples m, r _O and r _D to determine the optimal format side length, and divide the study area into parking cells according to the optimal format side length; Based on the OD data, parking data and parking cell division scheme of the research area, regression analysis is performed to build a parking demand forecasting model, including: After the research area is divided into parking areas according to the optimal format side length, each grid in the research area is defined as a parking area, and the number of peak parking areas and the number of O points in each parking area are determined respectively. and the number of D points; According to the peak parking number and the number of O points in each parking area, establish a univariate linear fitting function P _i =αx _Oi between the peak parking number of the parking area and the number of O points in the study area; According to the peak parking number and the number of D points in each parking area, establish a univariate linear fitting function P _i =βx _Di between the peak parking number of the parking area and the number of D points in the study area; Among them, Pi is the number of peak parking spaces in the parking area _i , x _Oi is the number of points O in the parking area, x _Di is the number of points D in the parking area, and α and β are the regression coefficients of the number of points O and D in the parking area, respectively; The univariate linear fitting function between the number of peak parking spaces corresponding to the research area and the number of points O and point D is used as a parking demand forecasting model; According to the parking demand prediction model, the parking demand prediction is performed on the target parking area. 2. The method according to claim 1, wherein the method further comprises: Perform preprocessing on the errors and abnormal values in the OD data and parking data, filter out the data whose field value is empty, and delete the data whose points O or D are not in the research area according to the latitude and longitude coordinate range of the research area. For data whose travel distance is less than 500 meters, delete data with travel time t _j ≤ 0min; delete data with travel speed V _j ≥ 100 km/h. 3. The method according to claim 1 or 2, wherein the predicting the parking demand for the target parking area according to the parking demand prediction model comprises: Input the number of 0 points of the target parking area into the univariate linear fitting function between the number of peak parking and the number of 0 points in the parking demand prediction model, to obtain the peak parking number of the target parking area; and / or, The number of D points in the target parking area is input into the univariate linear fitting function between the number of peak parking and the number of D points in the parking demand prediction model to obtain the peak parking number in the target parking area.

Images