Method for predicting traffic performance index value of urban road construction

Abstract

The invention relates to an urban road construction traffic performance index value prediction method, and belongs to the technical field of prediction or optimization specially suitable for administrative or management purposes. The method comprises the following steps: firstly, setting nine traffic performance index values, and taking one type of intersection or one category of main road as a road form; then traffic performance index values of traffic forms before and during construction of historical traffic construction engineering are obtained, and data sets are constructed respectively; predicting the data set before historical construction to construct a prediction data set, constructing regression models for the prediction set before historical construction and the data set during historical construction, and obtaining an error model according to the two regression models; and finally, predicting the traffic performance index value before actual road construction engineering construction to construct an actual prediction set, inputting the length of each main road in the actual prediction set into the error model to obtain an error set, and finally, correcting the actual prediction set through the error set to predict the actual traffic performance index value.

Description

Technical Field

[0001] The present invention relates to a method for predicting traffic performance index values ​​of urban road construction, and belongs to the technical field of prediction or optimization (G06Q10 / 04) specially suitable for administrative or management purposes. Background Art

[0002] With the continuous advancement of my country's urbanization process, traffic demand has increased dramatically, and various traffic problems have emerged one after another, among which the most prominent is traffic congestion. In order to alleviate this problem, a series of road construction projects have emerged, such as urban rail transit, bus rapid transit systems and road reconstruction. However, these new or expanded road projects will inevitably occupy the originally limited road resources during implementation. Urban road construction causes lanes to be occupied, reducing the road space for vehicles to pass, and thus reducing the overall traffic capacity of the road section, resulting in increased vehicle delays. At the same time, the traffic flow of sections and intersections around the construction area increases, which is easy to form traffic bottlenecks, aggravating traffic congestion in the construction area, and even affecting the surrounding road network. In addition, traffic congestion will increase automobile exhaust emissions and further pollute the environment.

[0003] During road construction, traffic performance index values ​​are often used to reflect the capacity, traffic conditions and congestion level of roads, intersections or the entire traffic network to help traffic management departments plan, monitor and make decisions. However, if the traffic performance index values ​​during construction can be quickly and accurately predicted before actual construction, the impact of road construction on traffic can be effectively evaluated, and the construction plan can be quickly adjusted based on these data, which will help alleviate the pressure of traffic congestion and reduce the impact on citizens' travel.

[0004] Therefore, in order to avoid excessive congestion during the construction of traffic projects, a prediction method for the traffic performance index values ​​of urban road construction is urgently needed as a basis for construction management and adjustment to ensure smooth traffic and environmental friendliness. Summary of the invention

[0005] The technical problem to be solved by the present invention is: how to quickly and accurately predict the traffic performance index value during construction before road construction.

[0006] The technical solution proposed by the present invention to solve the above technical problems is: a method for predicting the traffic performance index value of urban road construction, comprising the following steps:

[0007] Step 1: Set nine traffic performance index values, including traffic volume, traffic speed, traffic density, traffic delay, congestion index, accident rate, injury rate, dust rate and noise level;

[0008] The intersections are divided into three types: cross, T and Y. The main roads are divided into different categories according to the number of lanes on the main roads.

[0009] Treat one type of intersection or one category of main roads as a road form;

[0010] Step 2: Obtain nine historical pre-construction traffic performance index values ​​of the road type within the construction impact range before the construction of the historical road construction project;

[0011] Step 2.1: Classify the historical pre-construction traffic performance index values ​​obtained in step 2 according to the length X of each main road, and construct a historical pre-construction data set {S1}={X,s1(i,j)};

[0012] s1(i,j) is the jth historical pre-construction traffic performance index value of the i-th road form;

[0013] Step 2.2: Input the historical pre-construction data set {S1} into the multivariate linear regression model, and predict the historical pre-construction traffic performance index value to generate the historical pre-construction prediction set {S2}={X,s2(i,j)};

[0014] s2(i,j) is the predicted value of the jth historical pre-construction traffic performance index of the i-th road form;

[0015] Step 2.3: Obtain the nine historical construction traffic performance index values ​​of the road types within the construction impact range during the construction of the historical road construction project, and construct a historical construction data set {S3}={X,s3(i,j)};

[0016] s3(i,j) is the traffic performance index value of the jth historical construction period of the i-th road form;

[0017] Step 3: Map the historical pre-construction traffic performance index values ​​of the road forms in the historical pre-construction prediction set {S2} into a first scatter plot, and perform curve fitting on the first scatter plot to obtain a first regression function n ij =(X,t ij );

[0018] t ij is the first regression function value of the jth historical pre-construction traffic performance index of the i-th road form;

[0019] The first regression function is used to construct the first regression model ;

[0020] Step 3.1: Map the historical construction traffic performance index values ​​of the road forms in the historical construction dataset {S3} into a second scatter plot, perform curve fitting on the second scatter plot to obtain a second regression function, and construct a second regression model with the second regression function ;

[0021] Step 3.2: Calculate the error model according to formula (1).

[0022] (1);

[0023] Step 4: Determine the construction impact range of the actual road construction project, and collect nine actual pre-construction traffic performance index values ​​of the road types within the construction impact range before the actual road construction project is constructed;

[0024] Step 4.1: Input the actual pre-construction traffic performance index values ​​collected in step 4 into the multivariate linear regression model for prediction and construct the actual pre-construction prediction set {S4}={X,s4(i,j)};

[0025] s4(i,j) is the predicted value of the jth actual pre-construction traffic performance index of the i-th road form;

[0026] Step 4.2: Input the length X of each main road in the actual pre-construction prediction set {S4} into the error model The error value is obtained, and the error value is used to construct an error data set {S5}={X,s5(i,j)};

[0027] s5(i,j) is the jth traffic performance index error value of the i-th road form;

[0028] Step 4.3: Calculate the actual construction time data set of the actual road construction project according to formula (2): {S6}={X,s6(i,j)}

[0029] {S6}={S4}·{S5} (2);

[0030] s6(i,j) is the traffic performance index value of the jth actual construction of the i-th road type.

[0031] Furthermore, the curve fitting in step 3 and step 3.1 is performed by using ridge regression.

[0032] The beneficial effects of the present invention are as follows: by obtaining the traffic performance index values ​​of the road forms before and during the construction of historical road construction projects, and constructing data sets respectively, and predicting the historical data sets before construction to obtain a prediction set; then constructing an error model through the historical prediction set before construction and the historical data sets during construction; before actual construction, the error model can be used to output the actual traffic performance index values, which can not only quickly adjust the construction plan and reduce the impact on citizens' travel, but also provide strong support for subsequent traffic management during actual construction, alleviate the pressure of traffic congestion, minimize the negative impact on the transportation system, and provide strong support for the sustainable development and traffic management of the city. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The following is a further description of a method for predicting a traffic performance index value of an urban road construction according to the present invention in conjunction with the accompanying drawings.

[0034] Figure 1 It is the first scatter plot of traffic delay at a crossroad in the method for predicting traffic performance index values ​​of urban road construction in the embodiment.

[0035] Figure 2 It is a second scatter plot of traffic delays at crossroads in the method for predicting traffic performance index values ​​of urban road construction in the embodiment. DETAILED DESCRIPTION

[0036] Example

[0037] A method for predicting a traffic performance index value of an urban road construction in this embodiment includes the following steps:

[0038] Step 1: Set nine traffic performance index values, including traffic volume, traffic speed, traffic density, traffic delay, congestion index, accident rate, injury rate, dust rate and noise level;

[0039] The intersections are divided into three types: cross, T and Y. The main roads are divided into different categories according to the number of lanes on the main roads.

[0040] Treat one type of intersection or one category of main roads as a road form;

[0041] Step 2: Obtain nine historical pre-construction traffic performance index values ​​of the road forms within the construction impact range before the construction of the historical road construction project;

[0042] The historical road construction projects obtained in this embodiment are similar to the subsequent actual road construction projects, both of which have three types of intersections: cross, T-shaped, and Y-shaped, and three types of main roads: two-lane, three-lane, and four-lane;

[0043] Step 2.1: Classify the historical pre-construction traffic performance index values ​​obtained in step 2 according to the length X of each main road, and construct the historical pre-construction data set {S1}={X,s1(i,j)};

[0044] s1(i,j) is the jth historical pre-construction traffic performance index value of the i-th road form;

[0045] Step 2.2: Input the historical pre-construction data set {S1} into the multivariate linear regression model, and predict the historical pre-construction traffic performance index values ​​to generate the historical pre-construction prediction set {S2}={X,s2(i,j)};

[0046] This embodiment uses MATLAB software environment as a computing platform, and selects a multiple linear regression model to predict the historical pre-construction traffic performance index values. The multiple linear regression model is a statistical model that predicts the linear relationship between one or more dependent variables based on multiple independent variables. It is an existing and commonly used model.

[0047] s2(i,j) is the predicted value of the jth historical pre-construction traffic performance index of the i-th road form;

[0048] Step 2.3: Obtain the nine historical construction-time traffic performance index values ​​of the road forms within the construction impact range during the construction of the historical road construction project, and construct the historical construction-time data set {S3}={X,s3(i,j)};

[0049] s3(i,j) is the traffic performance index value of the jth historical construction period of the i-th road form;

[0050] Step 3: Map the historical pre-construction traffic performance index values ​​of the road types in the historical pre-construction prediction set {S2} into the first scatter plot, and perform curve fitting on the first scatter plot to obtain the first regression function n ij =(X,t ij );

[0051] t ij is the first regression function value of the jth historical pre-construction traffic performance index of the i-th road form;

[0052] This embodiment uses ridge regression to perform curve fitting. Taking the traffic delay at an intersection as an example, the first scatter plot mapped and the first regression function fitted are as follows: Figure 1 As shown;

[0053] Construct the first regression model using the first regression function ;

[0054] Step 3.1: Map the historical construction traffic performance index values ​​of the road types in the historical construction dataset {S3} into a second scatter plot, perform curve fitting on the second scatter plot to obtain a second regression function, and construct a second regression model with the second regression function ;

[0055] Taking the example of traffic delay at the intersection, the second scatter plot and the fitted second regression function are as follows: Figure 2 Shown

[0056] Step 3.2: Calculate the error model according to formula (1).

[0057] (1);

[0058] Step 4: Determine the construction impact range of the actual road construction project, and collect nine actual pre-construction traffic performance index values ​​of the road forms within the construction impact range before the actual road construction project is constructed;

[0059] Step 4.1: Input the actual pre-construction traffic performance index values ​​collected in step 4 into the multivariate linear regression model for prediction and construct the actual pre-construction prediction set {S4}={X,s4(i,j)};

[0060] s4(i,j) is the predicted value of the jth actual pre-construction traffic performance index of the i-th road form;

[0061] Step 4.2: Input the length X of each main road in the actual pre-construction prediction set {S4} into the error model The error value is obtained, and the error value is used to construct the error data set {S5}={X,s5(i,j)};

[0062] s5(i,j) is the jth traffic performance index error value of the i-th road form;

[0063] In this embodiment, the error model composed of regression functions is obtained by dividing the first regression model and the second regression model in step 3.2. When the length X of each trunk road is input into the error model, the error value of the actual pre-construction traffic performance index value of the jth road form under the length X of each trunk road in the error model can be obtained;

[0064] Step 4.3: Calculate the actual construction time data set of the actual road construction project according to formula (2): {S6}={X,s6(i,j)}

[0065] {S6}={S4}·{S5} (2);

[0066] s6(i,j) is the traffic performance index value of the jth actual construction of the i-th road type.

[0067] The method of this embodiment can predict the traffic performance index value during actual construction based on the historical traffic performance index values ​​of historical road construction projects before the actual road construction project is carried out. This can not only quickly adjust the construction plan and reduce the impact on citizens' travel, but also provide strong support for subsequent traffic management during actual construction and alleviate the pressure of traffic congestion.

[0068] The above description is only a preferred embodiment of the present invention, but the present invention is not limited thereto. All equivalent replacements or equivalent changes of the concepts and technical solutions according to the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting traffic performance index values ​​of urban road construction, characterized in that The steps include: Step 1: Set nine traffic performance index values, including traffic volume, traffic speed, traffic density, traffic delay, congestion index, accident rate, injury rate, dust rate and noise level; The intersections are divided into three types: cross, T and Y. The main roads are divided into different categories according to the number of lanes on the main roads. Treat one type of intersection or one category of main roads as a road form; Step 2: Obtain nine historical pre-construction traffic performance index values ​​of the road type within the construction impact range before the construction of the historical road construction project; Step 2.1: Classify the historical pre-construction traffic performance index values ​​obtained in step 2 according to the length X of each main road, and construct a historical pre-construction data set {S1}={X,s1(i,j)}; s1(i,j) is the jth historical pre-construction traffic performance index value of the i-th road form; Step 2.2: Input the historical pre-construction data set {S1} into the multivariate linear regression model, and predict the historical pre-construction traffic performance index value to generate the historical pre-construction prediction set {S2}={X,s2(i,j)}; s2(i,j) is the predicted value of the jth historical pre-construction traffic performance index of the i-th road form; Step 2.3: Obtain the nine historical construction traffic performance index values ​​of the road types within the construction impact range during the construction of the historical road construction project, and construct a historical construction data set {S3}={X,s3(i,j)}; s3(i,j) is the traffic performance index value of the jth historical construction period of the i-th road form; Step 3: Map the historical pre-construction traffic performance index values ​​of the road forms in the historical pre-construction prediction set {S2} into a first scatter plot, and perform curve fitting on the first scatter plot to obtain a first regression function n ij =(X,t ij ); t ij is the first regression function value of the jth historical pre-construction traffic performance index of the i-th road form; The first regression function is used to construct the first regression model ; Step 3.1: Map the historical construction traffic performance index values ​​of the road forms in the historical construction dataset {S3} into a second scatter plot, perform curve fitting on the second scatter plot to obtain a second regression function, and construct a second regression model with the second regression function ; Step 3.2: Calculate the error model according to formula (1). （1）； Step 4: Determine the construction impact range of the actual road construction project, and collect nine actual pre-construction traffic performance index values ​​of the road types within the construction impact range before the actual road construction project is constructed; Step 4.1: Input the actual pre-construction traffic performance index values ​​collected in step 4 into the multivariate linear regression model for prediction and construct the actual pre-construction prediction set {S4}={X,s4(i,j)}; s4(i,j) is the predicted value of the jth actual pre-construction traffic performance index of the i-th road form; Step 4.2: Input the length X of each main road in the actual pre-construction prediction set {S4} into the error model The error value is obtained, and the error value is used to construct an error data set {S5}={X,s5(i,j)}; s5(i,j) is the jth traffic performance index error value of the i-th road form; Step 4.3: Calculate the actual construction time data set of the actual road construction project according to formula (2): {S6}={X,s6(i,j)} {S6}={S4}·{S5} (2); s6(i,j) is the traffic performance index value of the jth actual construction of the i-th road type.

2. According to claim 1, it is characterized in that: The curve fitting in step 3 and step 3.1 is performed by using ridge regression.

Images