CN108364060A - A kind of classification design method of highway - Google Patents

Abstract

The invention discloses a kind of classification design methods of highway：Establish differential weights Grey BP Neural Network built-up pattern；Utilize differential weights Grey BP Neural Network Combined model forecast nonagricultural population's quantity；Estimation prediction year town site newly increased requirement amount S₁ ⁽ⁿ⁼ⁱ⁾；Determine the town site area S that practical maximum can increase₂；Pass through S₁ ⁽ⁿ⁼ⁱ⁾And S₂Comparative analysis, divide highway type.The present invention carries out classification refinement to highway type, to different types of highway using the new designing concept for adapting to the development of periphery plot, highway programming and distribution is made to develop in harmony with neighboring area urbanization process.

Description

A Classified Design Method for Expressway

technical field

The invention relates to a method for classifying and designing expressways in the transportation industry, and more specifically, relates to a method for classifying and designing expressways.

Background technique

The traditional concept of expressway planning and layout believes that expressway is a closed and independent system whose main function is to ensure the transition of transit and entry-exit traffic. It does not classify and refine its type according to its location, ignoring the layout of expressways and the surrounding areas. relationship between urban development. However, as my country's urbanization process accelerates, more and more expressways split urban land, seriously hindering the expansion of urban construction land.

Contents of the invention

The purpose of the present invention is to overcome the deficiencies in the prior art, provide a kind of expressway classification design method, classify and refine the expressway types, and adopt new design concepts that adapt to the development of surrounding land plots for different types of expressways , so that the planning and layout of the expressway can develop in harmony with the urbanization process of the surrounding areas.

The purpose of the present invention is achieved through the following technical solutions.

A classification design method for expressways, comprising the following steps:

Step 1, establishing an unequal weight gray BP neural network combination model;

Step 2, non-agricultural population prediction: use the unequal weight gray BP neural network combination model to predict the number of non-agricultural population;

Step 3: Estimating the new demand for urban construction land in the forecast year S ₁ ⁽ⁿ⁼ⁱ⁾ : Using the number of non-agricultural population, combining the "Urban Land Classification and Planning and Construction Land Standard GB50137-2011" and local land policies, the forecasted annual urban Increased demand for construction land;

Step 4: Determine the actual maximum increaseable urban construction land area S ₂ : take the current (or planned) expressway line position and the geographical boundary of the regional available construction land as the boundary, draw a closed geometric figure on the map, and obtain the actual maximum Increaseable urban construction land area;

Step five, classify the expressway types through the comparative analysis of S ₁ ⁽ⁿ⁼ⁱ⁾ and S ₂ .

The establishment process of the unequal weight gray BP neural network combination model in step 1:

(1) Establish the original gray prediction model

Create a raw data sequence:

X ⁽⁰⁾ ＝{x ⁽⁰⁾ (1),x ⁽⁰⁾ (2),...,x ⁽⁰⁾ (n)}

According to the following formula

Perform first-order accumulation on the original data sequence to generate a 1-AGO sequence:

X ⁽¹⁾ ＝{x ⁽¹⁾ (1),x ⁽¹⁾ (2),...,x ⁽¹⁾ (n)}

(2) Unequal weight optimization of the original gray prediction model

Using AHP to optimize the original gray forecasting model, using Delphi method to compare and evaluate each factor pairwise, determine the short-term time series weight as λ ₁ and the long-term weight as λ ₂ , and optimize the original gray forecasting model with unequal weights for:

Then generate the unequal weight 1-AGO sequence model:

X' ⁽¹⁾ = {x' ⁽¹⁾ (1),x' ⁽¹⁾ (2),...,x' ⁽¹⁾ (n)}

(4) BP neural network design

Build a three-layer network with an input layer, a hidden layer, and an output layer:

①Select the time series value {x′ ⁽¹⁾ (1),x′ ⁽¹⁾ (2),...,x′ ⁽¹⁾ (m)} in the unequal weighted 1-AGO sequence model (m< n) as the input layer of BP neural network;

② Use x′ ⁽¹⁾ (m+1) as the output layer of the BP neural network;

③ Hidden layer nodes according to the formula Determined by calculation, m is the number of input neurons, p is the number of output neurons, and q is a constant between 1 and 10;

④ Use the trained BP neural network to make predictions, and use cumulative reduction and reduction for the prediction sequence to obtain the predicted value for the future.

The non-agricultural population forecasting process in Step 2 is as follows:

Collect the non-agricultural population of the predicted area in the city over the years to form the original time series:

The improved non-agricultural population accumulation sequence is obtained by using the improved unequal weight gray prediction model for accumulation:

The obtained sequence is input into the trained BP neural network, and the optimal result calculated by the BP neural network is accumulated and restored to obtain the predicted value of non-agricultural population in the forecast year.

Comparative analysis of S ₁ ⁽ⁿ⁼ⁱ⁾ and S ₂ in step five:

(1) If the predicted annual demand for new urban construction land is not less than the actual maximum increaseable urban construction land area 5 years after the completion of the expressway, that is: S ₁ ⁽ⁿ⁼⁵⁾ ≥ S ₂ , then the expressway The road belongs to the "urban expressway";

(2) If the predicted annual demand for urban construction land is not less than the actual maximum increaseable urban construction land area after 5-20 years of expressway completion, that is: S ₁ ^(n=5～20) ≥ _{S 2} , the expressway belongs to the "suburban expressway".

Compared with the prior art, the beneficial effects brought by the technical solution of the present invention are:

(1) The present invention sets up the combination model of unequal weight gray BP neural network, predicts the trend of urban expansion, and draws the newly added demand S ₁ ⁽ⁿ⁼ⁱ⁾ of urban construction land in the forecast year, and draws the present situation (planning) under the influence of highway layout The actual maximum increaseable urban construction land area S ₂ is obtained, and S ₁ ⁽ⁿ⁼ⁱ⁾ and S ₂ are compared to clearly predict the type of expressway in the year, which is optimized for expressway planning and layout Or provide a basis for decision-making on reconstruction and expansion plans.

(2) The present invention is different from the expressways in the traditional sense, classifies and refines the types of expressways, proposes new concepts of "urban expressways" and "suburban expressways", and adopts corresponding concepts to design respectively according to their location characteristics , in the process of planning and design, fully consider the influence of expressway layout and location on the urbanization of surrounding areas, and provide technical support for the planning and site selection of new expressways and the decision-making of existing expressway reconstruction and expansion plans.

Description of drawings

Fig. 1 is a neural network training figure of the present invention;

Fig. 2 is a neural network fitting diagram;

Figure 3 is a planning map of construction land in Tianjin;

Figure 4 is the calculation map of future construction land in Dongli District.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings.

The classification design method of expressways of the present invention proposes new concepts of "urban expressways" and "suburban expressways". Provide technical support for highway planning and reconstruction and expansion plans. The specific process is as follows:

1. Establish an unequal weight gray BP neural network combination model;

(1) Establish the original gray prediction model

Create a raw data sequence:

X ⁽⁰⁾ ＝{x ⁽⁰⁾ (1),x ⁽⁰⁾ (2),...,x ⁽⁰⁾ (n)}(1)

According to the following formula (2)

Perform first-order accumulation on the original data sequence X ⁽⁰⁾ to generate a 1-AGO sequence:

X ⁽¹⁾ ＝{x ⁽¹⁾ (1),x ⁽¹⁾ (2),...,x ⁽¹⁾ (n)}(3)

(2) Unequal weight optimization of the original gray prediction model

The original gray forecasting model only performs equal-weighted accumulation on the original series, and does not consider the influence of time factors on the forecast results. In actual situations, the closer to the forecast year, the greater the amount of information contained in the time series, the more it can reflect the future development. trend, the assigned weight should also be larger. Therefore, the AHP is used to further optimize the original gray forecasting model, experts are invited to use the Delphi method to compare and evaluate each factor pairwise, and determine the weight of the short-term time series as λ ₁ , the weight of the long-term as λ ₂ , and the original gray The unequal weight optimization of the prediction model is:

Then generate the unequal weight 1-AGO sequence model:

X' ⁽¹⁾ ={x' ⁽¹⁾ (1),x' ⁽¹⁾ (2),...,x' ⁽¹⁾ (n)}(5)

(5) BP neural network design

Build a three-layer network with an input layer, a hidden layer, and an output layer:

①Select the time series value {x′ ⁽¹⁾ (1),x′ ⁽¹⁾ (2),...,x′ ⁽¹⁾ (m)} in the unequal weighted 1-AGO sequence model (m< n) as the input layer of BP neural network;

② Use x′ ⁽¹⁾ (m+1) as the output layer of the BP neural network;

③ Hidden layer nodes can be based on the formula Determined by calculation, m is the number of input neurons, p is the number of output neurons, and q is a constant between 1 and 10;

④ Use the trained BP neural network to make predictions, and use cumulative reduction and reduction for the prediction sequence to obtain the predicted value for the future.

2. Prediction of non-agricultural population: use the unequal weight gray BP neural network combination model to predict the number of non-agricultural population.

Collect the non-agricultural population of the predicted area in the city over the years to form the original time series:

Note: It is the value of the non-agricultural population survey over the years.

The improved non-agricultural population accumulation sequence is obtained by using the improved unequal weight gray prediction model for accumulation:

Note: Calculated by formula (4).

The obtained sequence is input into the trained BP neural network, and the optimal result calculated by the BP neural network is accumulated and restored, and the predicted value of the non-agricultural population in the forecast year can be obtained.

3. Estimate the new demand for urban construction land S ₁ ⁽ⁿ⁼ⁱ⁾ in the forecast year:

Using the predicted value of the above-mentioned non-agricultural population, combined with the "Urban Land Classification and Planning and Construction Land Standard GB50137-2011" and local land policies, the predicted annual urban construction land new demand S ₁ ⁽ⁿ⁼ⁱ⁾ is obtained.

4. Determine the actual maximum increaseable urban construction land area S ₂ :

The development direction of urbanization usually expands from the central area suitable for development to the edge of the urban area. Draw a closed geometric figure on the map based on the current (or planned) expressway line position and the relevant geographical boundary of the available construction land in the area. On both sides, the actual maximum urban construction land area S ₂ that can be increased in the city.

5. Classification of Expressway Types:

The design life of general expressways is 20 years. Under normal circumstances, under the condition of ensuring that there are no quality problems and overloading and exceeding the limit, newly-built expressways will not undergo large-scale renovation within 5 years. Therefore, 5 years and 20 years after the completion and operation of the expressway are taken as nodes, and the expressway types are divided through the comparative analysis of S ₁ ⁽ⁿ⁼ⁱ⁾ and S ₂ .

Comparative analysis of S ₁ ⁽ⁿ⁼ⁱ⁾ and S ₂ :

(1) If the estimated new demand for urban construction land is not less than the actual maximum increaseable urban construction land area five years after the expressway is completed, that is: S ₁ ⁽ⁿ⁼⁵⁾ ≥ S ₂ . It shows that this expressway will soon invade the adjacent administrative area and hinder the urbanization process of the surrounding area. It belongs to the "urban expressway" and needs to adopt the concept of "urban expressway" for planning and design.

(2) If the predicted annual demand for urban construction land is not less than the actual maximum increaseable urban construction land area after 5-20 years of expressway completion, that is: S ₁ ^(n=5～20) ≥ _{S 2} . It shows that although the expressway will not hinder the development of the city within a certain period of time, it will enter the city within the design period. It belongs to the "suburban expressway" and needs to be designed with the concept of "suburban expressway".

According to the results of the above schemes, optimize the planning and layout of expressways or optimize the reconstruction and expansion schemes.

Example:

Tianjin Dongli District is located in the main axis of the development of Jinbin, bordering the core area of Binhai New Area in the east and the central urban area in the west. It is an important functional area of the central urban area of Tianjin and Binhai New Area. Tianjin's development model is the twin-city development of the main urban area and the Binhai New Area, and Dongli District between the two urban areas has become the main development area of Tianjin. Jishan Expressway starts from Beijing-Tianjin Expressway and ends at Rongwu Expressway (Jinjin Expressway) in the national expressway network. It passes through Dongli, Jinnan, and Xiqing Districts along the way. According to the planning direction of Tianjin's urban development, the Dongli section of Jishan Expressway is very likely to enter the city within the design period. The method proposed in this patent will be used to judge and classify Jishan Expressway.

Obtain the non-agricultural population of Dongli District in the past 20 years from Tianjin Statistical Yearbook, that is, n=20; use Delphi method to determine the short-term time series weight λ ₁ =0.65, and the long-term weight λ ₂ =0.35.

The optimized unequal weight gray BP neural network combination model is used to predict the non-agricultural population in urban areas. The selected training function is trainlm, the transfer function of the first layer is tansig, the transfer function of the second layer is purelin, and the training error is 0.0001. The learning rate is 0.1.

After repeated experiments, it is determined that the prediction effect is the best when the parameters m=8, p=1, and q=10. The training graph is shown in Figure 1, the fitting graph is shown in Figure 2, and the statistical data and prediction results are shown in Table 1.

Table 1 Population prediction results and errors of Dongli District, Tianjin

According to the development plan of Dongli District, Tianjin, the per capita area of urban construction land in Dongli District is controlled at 100m ² /person, and the value range of urban construction land per capita in Tianjin is obtained from the "Urban Land Classification and Planning and Construction Land Standard GB50137-2011" is 75-100m ² /person, and the planned per capita urban construction land is selected as 90m ² /person. According to the predicted value of the above population, it is calculated that since the Jishan Expressway was completed and opened to traffic 5 years and 10 years later, Tianjin East The urban construction land area increments in Li District are S ₁ ⁽ⁿ⁼⁵⁾ =8.19km ² and S ₁ ⁽ⁿ⁼¹⁰⁾ =18.19km ² respectively.

According to the planning of construction land in Tianjin, the urbanization direction of Dongli District expands from the central urban area to the east, with the Jishan Expressway and the relevant geographical dividing line of available construction land as the boundary. The scope of Dongli District may be developed into construction land as follows: Shown in Figure 3 and Figure 4. It can be seen that S ₂ =16.38km ² .

The comparison shows that: S ₁ ⁽ⁿ⁼⁵⁾ <S ₂ , S ₁ ⁽ⁿ⁼¹⁰⁾ >S ₂ , indicating: (1) After the opening of Jishan Expressway for 5 years, the increase of urban construction land in Dongli District is higher than that of The estimated area is small, and the land around the highway is still not developed into urban construction land; (2) But with the acceleration of urbanization, after 10 years of operation of Jishan Expressway, Dongli District will gradually expand eastward, and the urban construction land The demand has exceeded the calculated area, that is, the layout of the Jishan Expressway will hinder the urbanization process of Dongli District. The effect of expansion.

Although the function and working process of the present invention have been described above in conjunction with the accompanying drawings, the present invention is not limited to the above-mentioned specific functions and working process, and the above-mentioned specific implementation is only illustrative, rather than limiting. Under the enlightenment of the present invention, those skilled in the art can also make many forms without departing from the spirit of the present invention and the scope protected by the claims, and these all belong to the protection of the present invention.

Claims (4)

1. a classification design method of expressway, is characterized in that, comprises the following steps: Step 1, establishing an unequal weight gray BP neural network combination model; Step 2, non-agricultural population prediction: use the unequal weight gray BP neural network combination model to predict the number of non-agricultural population; Step 3: Estimating the new demand for urban construction land in the forecast year S ₁ ⁽ⁿ⁼ⁱ⁾ : Using the number of non-agricultural population, combining the "Urban Land Classification and Planning and Construction Land Standard GB50137-2011" and local land policies, the forecasted annual urban Increased demand for construction land; Step 4: Determine the actual maximum increaseable urban construction land area S ₂ : take the current (or planned) expressway line position and the geographical boundary of the regional available construction land as the boundary, draw a closed geometric figure on the map, and obtain the actual maximum Increaseable urban construction land area; Step five, classify the expressway types through the comparative analysis of S ₁ ⁽ⁿ⁼ⁱ⁾ and S ₂ . 2. the classification design method of expressway according to claim 1, is characterized in that, in the step 1, unequal weight gray BP neural network combined model establishment process: (1) Establish the original gray prediction model Create a raw data sequence: X ⁽⁰⁾ ＝{x ⁽⁰⁾ (1),x ⁽⁰⁾ (2),...,x ⁽⁰⁾ (n)} According to the following formula Perform first-order accumulation on the original data sequence to generate a 1-AGO sequence: X ⁽¹⁾ ＝{x ⁽¹⁾ (1),x ⁽¹⁾ (2),...,x ⁽¹⁾ (n)} (2) Unequal weight optimization of the original gray prediction model Using AHP to optimize the original gray forecasting model, using Delphi method to compare and evaluate each factor pairwise, determine the short-term time series weight as λ ₁ and the long-term weight as λ ₂ , and optimize the original gray forecasting model with unequal weights for: Then generate the unequal weight 1-AGO sequence model: X' ⁽¹⁾ = {x' ⁽¹⁾ (1),x' ⁽¹⁾ (2),...,x' ⁽¹⁾ (n)} (3) BP neural network design Build a three-layer network with an input layer, a hidden layer, and an output layer: ①Select the time series value {x′ ⁽¹⁾ (1),x′ ⁽¹⁾ (2),...,x′ ⁽¹⁾ (m)} in the unequal weighted 1-AGO sequence model (m< n) as the input layer of BP neural network; ② Use x′ ⁽¹⁾ (m+1) as the output layer of the BP neural network; ③ Hidden layer nodes according to the formula Determined by calculation, m is the number of input neurons, p is the number of output neurons, and q is a constant between 1 and 10; ④ Use the trained BP neural network to make predictions, and use cumulative reduction and reduction for the prediction sequence to obtain the predicted value for the future. 3. the classification design method of expressway according to claim 1, is characterized in that, non-agricultural population forecasting process is specially in step 2: Collect the non-agricultural population of the predicted area in the city over the years to form the original time series: The improved non-agricultural population accumulation sequence is obtained by using the improved unequal weight gray prediction model for accumulation: The obtained sequence is input into the trained BP neural network, and the optimal result calculated by the BP neural network is accumulated and restored to obtain the predicted value of non-agricultural population in the forecast year. 4. the classified design method of expressway according to claim 1, is characterized in that, in step 5, S ₁ ⁽ⁿ⁼ⁱ⁾ and S ₂ comparative analysis: (1) If the predicted annual demand for new urban construction land is not less than the actual maximum increaseable urban construction land area 5 years after the completion of the expressway, that is: S ₁ ⁽ⁿ⁼⁵⁾ ≥ S ₂ , then the expressway The road belongs to the "urban expressway"; (2) If the predicted annual demand for urban construction land is not less than the actual maximum increaseable urban construction land area after 5-20 years of expressway completion, that is: S ₁ ^(n=5～20) ≥ _{S 2} , the expressway belongs to the "suburban expressway".