A Safety Evaluation Method for Urban Underground Interchanges Based on Driving Simulation

Abstract

This invention relates to a safety evaluation method for urban underground interchanges based on driving simulation, comprising the following steps: 1) Selecting multiple existing urban underground interchanges in China to create simulation scenario models, and conducting experiments with drivers to collect vehicle operation and running data; 2) Determining driving behavior indicators for each urban underground interchange; 3) Calculating the safety entropy of the driving behavior indicators using sample entropy; 4) Calculating the comprehensive safety evaluation value for each urban underground interchange using an improved grey near-optimal comprehensive evaluation method based on the safety entropy of the driving behavior indicators; 5) Classifying each urban underground interchange according to its comprehensive safety evaluation value using the rank-sum ratio method to obtain the safety level of each urban underground interchange. This provides a theoretical basis for the safety design and management of urban underground interchanges and expands the application scope of driving simulation technology.

Description

Technical Field

[0001] This invention belongs to the field of road traffic safety technology, and relates to a safety evaluation method for urban underground interchanges, and more particularly to a safety evaluation method for urban underground interchanges based on driving simulation. Background Technology

[0002] With the rapid advancement of urbanization in China, limited urban land can no longer meet the ever-expanding demand for surface roads, thus giving rise to urban underground roads. Because urban underground interchanges combine the traffic environment characteristics of both surface interchanges and underground roads, their driving environment is more complex, and safety issues are prominent, urgently requiring research into their driving behavior and safety evaluation.

[0003] There have been various studies on the safety evaluation of interchanges. For example, some researchers have established traffic accident prediction models for merging areas of interchanges; others have found through statistical analysis that the speed of vehicles merging at ramps is the main factor affecting the number of traffic accidents in merging areas of interchanges; some have proposed a method based on a combination of microscopic simulation and traffic conflict identification technology to evaluate the safety of merging and diverging areas of interchanges; some have used the main road and ramps of interchanges as research objects, comparing and analyzing five road safety evaluation methods: human factors engineering theory, traffic conflict technology, standard deviation analysis of operating speed, traffic accident statistics, and analytic hierarchy process (AHP), concluding that human factors engineering theory is the most suitable safety evaluation method for interchange areas; some have classified the types of traffic conflicts in highway interchange areas and used grey clustering to evaluate the safety of merging and diverging areas of interchanges; some have studied the correlation between the standard deviation of vehicle speed and the accident rate in merging areas of urban underground interchanges and classified the safety levels of merging areas; and some have used merging areas of interchanges as research objects, selecting road conditions and vehicle performance as evaluation indicators to construct a fuzzy safety evaluation model to evaluate the safety of merging area ramps.

[0004] However, a review of existing research reveals that while there is a wealth of research on driving behavior and safety assessments at interchanges, most studies focus on above-ground interchanges, with limited consideration given to underground interchanges. Because underground interchanges combine the characteristics of both above-ground interchanges and underground roads, their traffic environment is far more complex than that of above-ground interchanges, making existing research on driving behavior at interchanges less transferable. Furthermore, the limited number of underground interchanges in cities and the difficulty in collecting accident data make conventional safety assessment methods unsuitable.

[0005] Given the aforementioned shortcomings of existing technologies, there is an urgent need to research a safety evaluation method for urban underground interchanges. Summary of the Invention

[0006] Considering that underground interchanges combine the characteristics of above-ground interchanges and underground roads, their traffic environment is more complex than that of above-ground interchanges, and there is relatively little research on underground interchanges, this invention provides a safety evaluation method for urban underground interchanges based on driving simulation, in order to provide a theoretical basis for the safety design and management of urban underground interchanges.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A safety evaluation method for urban underground interchanges based on driving simulation, characterized by the following steps:

[0009] 1) Select several underground interchanges that have been built in China to create simulation scenarios, and conduct experiments with drivers to collect vehicle operation and running data.

[0010] 2) Determine the driving behavior indicators for each of the aforementioned urban underground interchanges;

[0011] 3) Calculate the safety entropy of driving behavior indicators for each of the aforementioned urban underground interchanges using sample entropy;

[0012] 4) Based on the safety entropy of driving behavior indicators for each of the aforementioned urban underground interchanges, the comprehensive safety evaluation value for each of the aforementioned urban underground interchanges is calculated using the improved grey near-optimal comprehensive evaluation method;

[0013] 5) Based on the comprehensive safety evaluation value, the rank-sum ratio method is used to classify each urban underground interchange into different grades to obtain the safety level of each urban underground interchange.

[0014] Preferably, step 4) specifically includes:

[0015] 4.1) The driving behavior indicators of each of the aforementioned urban underground interchanges are used as gray elements in the gray matrix;

[0016] 4.2) The safety entropy of the driving behavior indicators for each of the aforementioned urban underground interchanges is used as the whitening gray element value;

[0017] 4.3) The whitening gray element value is dimensionless to obtain the near-optimal whitening gray value;

[0018] 4.4) Calculate the weights of each of the aforementioned driving behavior indicators using the entropy weight method;

[0019] 4.5) Determine the comprehensive safety evaluation value of each of the urban underground interchanges based on the near-optimal whitening gray value and weight.

[0020] Preferably, step 4.3) specifically comprises:

[0021] 4.3.1) Select the minimum value in the dataset of each driving behavior indicator as the specified smaller value;

[0022] 4.3.2) Divide the specified smaller value by the whitening gray element value to obtain the near-optimal whitening gray value.

[0023] Preferably, step 4.4) specifically comprises:

[0024] 4.4.1) Calculate the characteristic weight of each of the driving behavior indicators, that is, the weight of a certain value of each of the driving behavior indicators in all the values ​​of the driving behavior indicator;

[0025] 4.4.2) Calculate the information entropy value of each driving behavior indicator based on the feature weight of each driving behavior indicator;

[0026] 4.4.3) Calculate the weight of each driving behavior indicator based on the information entropy value of each driving behavior indicator.

[0027] Preferably, step 5) specifically comprises:

[0028] 5.1) The comprehensive safety evaluation value of each of the aforementioned urban underground interchanges is grouped according to its magnitude;

[0029] 5.2) List the frequency, cumulative frequency, average rank, and cumulative frequency of each group, and calculate the probability unit for each group;

[0030] 5.3) Using the probability unit as the independent variable and the comprehensive safety evaluation value as the dependent variable, perform linear regression to obtain the linear regression equation;

[0031] 5.4) The rank-sum ratio is calculated based on the linear regression equation.

[0032] 5.5) Based on the rank-sum comparison, each urban underground interchange is classified and categorized to obtain the safety level of each urban underground interchange.

[0033] Preferably, the linear regression equation is Sj = -0.275 + 0.127Probit, where S j The Probit is a comprehensive security evaluation value, and it is a unit of probability.

[0034] Preferably, step 1) specifically involves: selecting six existing underground interchanges in Chinese cities to create a simulation scenario model, conducting experiments with 20 drivers, and collecting vehicle operation and running data from the experiments.

[0035] Preferably, the vehicle operation and running data collected in the experiment include simulation time, driving distance, speed, steering wheel angle, acceleration, coordinates, lateral displacement, distance from the left edge of the lane, distance from the right edge of the lane, and fuel consumption.

[0036] Preferably, step 2) specifically involves dividing each of the urban underground interchanges into three types of road segments: merging segment, diverging segment, and basic road segment. Speed, acceleration, steering wheel turning, and lateral displacement from the experimental vehicle operation and running data of each road segment of each urban underground interchange are selected as driving behavior indicators. Thus, each urban underground interchange has a total of 12 driving behavior indicators.

[0037] Compared with the prior art, the safety evaluation method for urban underground interchanges based on driving simulation of the present invention has one or more of the following beneficial technical effects:

[0038] 1. This invention uses an indirect evaluation method to study the safety of urban underground interchanges. This evaluation method considers more comprehensive factors and makes the evaluation indicators expandable, which solves the problem that traditional road safety evaluation methods do not consider the differences in the importance of various indicators.

[0039] 2. This invention proposes a gray near-optimal comprehensive evaluation method based on entropy weight and improvement, and uses the rank sum ratio method to divide the evaluation level. Compared with traditional evaluation methods, it has stronger adaptability to road safety evaluation under the complex effects of multiple factors, such as urban underground interchanges.

[0040] 3. This invention applies driving simulation technology to the safety evaluation of urban underground interchanges, thus expanding the application scope of driving simulation technology. Attached Figure Description

[0041] Figure 1 This is a flowchart of the safety evaluation method for urban underground interchanges based on driving simulation, as proposed in this invention.

[0042] Figure 2 This is a schematic diagram of a simulation scenario for an underground interchange in a certain city.

[0043] Figure 3 This is a schematic diagram of the driving simulator used in this invention.

[0044] Figure 4 The ranking results of the overall comprehensive evaluation values ​​of underground interchanges in various cities. Detailed Implementation

[0045] The present invention will be further described below with reference to the accompanying drawings and embodiments. The content of the embodiments is not intended to limit the scope of protection of the present invention.

[0046] Considering that underground interchanges combine the characteristics of above-ground interchanges and underground roads, their traffic environment is more complex than that of above-ground interchanges, and there is relatively little research on underground interchanges, this invention provides a safety evaluation method for urban underground interchanges based on driving simulation, in order to provide a theoretical basis for the safety design and management of urban underground interchanges.

[0047] Figure 1 A flowchart of the safety evaluation method for urban underground interchanges based on driving simulation according to the present invention is shown. Figure 1 As shown, the safety evaluation method for urban underground interchanges based on driving simulation of the present invention includes the following steps:

[0048] First, we selected several existing underground interchanges in China to create simulation scenarios, and then conducted experiments with drivers to collect vehicle operation and running data.

[0049] Specifically, six existing underground interchanges in Chinese cities can be selected as engineering examples. UC-win / Road can be used to model these interchanges in a simulation environment, resulting in simulation models for each of the six urban underground interchanges. The final simulation model of one of these interchanges is shown below. Figure 2 As shown.

[0050] Then, using a driving simulator, its appearance is as follows: Figure 3 As shown, drivers were recruited to conduct experiments, and data on vehicle operation and performance were collected.

[0051] Preferably, 20 drivers were recruited for the experiment, and vehicle operation and running data were collected from each driver. Data collected during experiments where no collision occurred was recorded as one set, resulting in a total of 120 sets of experimental data. This ensures that the collected vehicle operation and running data are more realistic and effective.

[0052] Meanwhile, preferably, the vehicle operation and running data collected in the experiment include simulation time, driving distance, speed, steering wheel angle, acceleration, coordinates, lateral displacement, distance from the left edge of the lane, distance from the right edge of the lane, and fuel consumption.

[0053] II. Determine the driving behavior indicators for each of the aforementioned urban underground interchanges.

[0054] In this invention, each urban underground interchange is divided into three types of road sections: merging section, diverging section, and basic road section. Speed, acceleration, steering wheel turning, and lateral displacement from the experimental vehicle operation and running data of each road section of each urban underground interchange are selected as driving behavior indicators. Therefore, each urban underground interchange has a total of 12 driving behavior indicators.

[0055] Among them, speed and acceleration are longitudinal driving behavior indicators, while steering wheel turning and lateral displacement are lateral driving behavior indicators.

[0056] 3. Calculate the safety entropy of driving behavior indicators for each of the aforementioned urban underground interchanges using sample entropy.

[0057] The sample entropy method used in this invention takes into account that the data output by the driving simulator is a time series, and sample entropy can effectively evaluate the stability of driving behavior indicators.

[0058] Sample entropy can be represented by the formula S(m,r,N), where N is the data length, m is the embedding dimension, r is the similarity tolerance error, and m and r are control variables for calculating sample entropy. Typically, m is 1 or 2. When m>2, the required number of samples increases significantly, so m is chosen to be 2. N ranges from 100 to 5000, and in this invention, it is set to 1000.

[0059] The calculation process of security entropy implemented using Python programming is shown below:

[0060] 1) Let the original data be x(1) and x(2), a total of N numbers;

[0061] 2) Construct vector m, subtracting every two vectors, as shown in formula (1):

[0062] X(i)=[x(i),x(i+1),…,x(i+m-1)] (1)

[0063] In the formula, i = j = 1, 2, ..., N-m+1, ...

[0064] 3) Define d[x(i),x(j)] as the maximum length of the difference between x(i) and x(j), and calculate it as shown in equation (2):

[0065] d[x(i),x(j)]=max|x(i+k)-x(j+k)| (2)

[0066] In the formula, k = 0, 1, 2, ..., m + l.

[0067] 4) B is the ratio of the number of d[x(i),x(j)] less than r to N-m+1.m (r) is The average value is calculated using formulas (3) and (4):

[0068]

[0069]

[0070] In the formula, r = 0.1 to 0.25e, and in this invention, r = 0.25e is used.

[0071] 5) When n < ∞, the calculation formula is as shown in equation (5):

[0072]

[0073] This invention utilizes the safety entropy calculated from driver speed, acceleration, lateral displacement, and steering wheel angle using sample entropy to evaluate the impact of different urban underground interchanges on drivers' lateral and longitudinal driving behaviors. A higher safety entropy indicates a greater impact on driver behavior.

[0074] IV. Based on the safety entropy of driving behavior indicators for each of the aforementioned urban underground interchanges, the comprehensive safety evaluation value for each of the aforementioned urban underground interchanges is calculated using the improved grey near-optimal comprehensive evaluation method.

[0075] The grey near-optimal comprehensive evaluation method is typically chosen to assess road safety. While this method requires minimal computation, no large sample size, and is independent of subjective human judgment, it does not consider the varying importance of different indicators (i.e., it lacks defined weights). Therefore, to ensure the scientific rigor of the evaluation, weights need to be assigned to each indicator. This invention proposes a safety evaluation method for urban underground interchanges based on entropy weighting and an improved grey near-optimal comprehensive evaluation method.

[0076] The calculation process of the entropy weight-improved grey near-optimal comprehensive evaluation method is as follows:

[0077] (1) Establishing the gray matrix

[0078] Suppose there is a scheme M j (j=1,2,…,m), performance indicator C i = (i = 1, 2, ..., n) and the corresponding whitening ash amount R ij Then the gray element of the n-dimensional index of m schemes is called R. n×m The following gray matrix R is obtained. n×m :

[0079]

[0080] In the formula, R ij Let R be the i-th performance indicator for the j-th scheme.m×n The gray elements of a gray matrix.

[0081] (2) Establish the whitening matrix

[0082] Enter the measured assessment indicators to obtain the whitening gray matrix.

[0083]

[0084] In the formula, This is the actual measured value of the i-th assessment indicator (i.e., the whitening gray element value) for the j-th scheme.

[0085] (3) Dimensionless processing

[0086] Because the magnitude and dimensions of each evaluation indicator are different, it is not possible to conduct a comprehensive evaluation directly. It is necessary to perform dimensionless processing on each evaluation indicator and map the different evaluation indicators to the [0,1] interval.

[0087] Grey near-optimal comprehensive evaluation method often uses single-point effect measures to perform dimensionless processing of the indicators. The evaluation indicators selected in this invention are all of the type where smaller is better; therefore, a lower limit effect measure is used to perform dimensionless processing of the evaluation indicators. Furthermore, to avoid situations where the evaluation indicators of a single scheme cannot be dimensionless, the minimum value is selected as the specified smaller value μ based on the existing evaluation indicator dataset. min The dimensionless calculation formula for the evaluation index is:

[0088]

[0089] By replacing the whitening gray element values ​​with the whitening gray element effect measures, a near-optimal whitening gray matrix is ​​obtained:

[0090]

[0091] In the formula Let be the near-optimal whitening gray element value of the i-th evaluation indicator for the j-th scheme, i.e., the effectiveness measurement degree.

[0092] (4) Calculate the weight of each indicator using the entropy weight method.

[0093] Specifically, the characteristic weight is first calculated using the following formula:

[0094]

[0095] In the formula, P ij The characteristic proportion is the proportion of the j-th value of the i-th indicator to the total value of the indicator.

[0096] Next, the information entropy is calculated.

[0097] According to the definition of information entropy in information theory, the information entropy of a set of data can be calculated using equation (11):

[0098]

[0099] In the formula, E i Let be the information entropy value of the i-th indicator.

[0100] Then, the indicator weights are calculated.

[0101] Based on the formula for calculating information entropy, the information entropy of each evaluation indicator is calculated as E1, E2, ..., E n The weights of each indicator are calculated using the obtained information entropy, and the calculation formula is shown in equation (12):

[0102]

[0103] (5) Entropy weight-improved grey near-optimal comprehensive evaluation method

[0104] Substituting the index weights determined using the entropy weight method into the grey near-optimal comprehensive evaluation method, the improved near-optimal degree whitening grey matrix is ​​obtained. As in equation (13):

[0105]

[0106] In the formula, S j The near-optimality of the j-th scheme obtained by the improved grey near-optimal comprehensive evaluation method, i.e., the comprehensive evaluation value, is calculated by the formula shown in equation (14):

[0107]

[0108] Substituting equation (8) into the equation yields the relationship between the comprehensive evaluation value and the evaluation index:

[0109]

[0110] Evaluation index C i Replace the whitening gray element value This will give you a comprehensive safety rating:

[0111]

[0112] V. Based on the aforementioned comprehensive safety evaluation value, the rank-sum ratio method is used to classify each urban underground interchange into different grades to obtain the safety level of each urban underground interchange.

[0113] The Rank-sum ratio (RSR) method can classify and categorize the evaluation objects, and finally make a comprehensive evaluation of the evaluation objects.

[0114] Specifically, groups are formed based on the magnitude of the comprehensive evaluation value, and the frequency f and cumulative frequency ∑f of different groups are listed, along with the average rank. and cumulative frequency Then calculate the corresponding probability unit, Probit value. The last term of the cumulative frequency is denoted as... The correction is made. The calculation process of SRS is as follows: Using Probiti as the independent variable and the comprehensive evaluation value as the dependent variable, a linear regression is performed to obtain the linear regression equation. Then, RSR is substituted into the regression equation instead of the comprehensive evaluation value, and the calculation formula is shown in equation (17):

[0115] RSR = a + b × Probit (17)

[0116] In the formula, a and b are regression parameters. The RSR can be obtained according to formula (17).

[0117] The regression equation needs to be tested. Common regression equation tests include validity tests of the regression coefficient b and goodness-of-fit tests. This invention uses the t-test of the regression coefficient b and the Pearson correlation coefficient test for goodness-of-fit to test the regression equation.

[0118] After sorting the Probit values, the corresponding RSR (Recovery Rate) classification is calculated. Then, based on the comprehensive evaluation value, the evaluation objects are classified into different categories. The number of categories is determined according to the actual situation. Essentially, this step maps the evaluation values ​​to a normal distribution curve and uses the normal distribution's classification method. Table 1 shows the percentile thresholds and corresponding Probit values ​​for commonly used classification scenarios.

[0119] Table 1. Percentile thresholds and corresponding Probit values ​​for common tiering scenarios.

[0120]

[0121]

[0122] The following describes in detail the safety evaluation method for urban underground interchanges based on driving simulation of the present invention using a specific embodiment.

[0123] In this embodiment, the six urban underground interchanges are the Xiamen Wanshishan and Zhonggushan Tunnel Interchange, the Suining Guanyinhu Tunnel, the Hangzhou Zizhi Tunnel Interchange, the Shenzhen Henglongshan Tunnel Interchange, the Qingdao Jiaozhou Bay Tunnel Interchange, and the Nanjing Qing'ao Tunnel Interchange. This allows the selected urban underground interchanges to be located in different regions, making the results more representative and universally applicable.

[0124] The main design parameters of the six selected urban underground interchanges are shown in Table 2.

[0125] Table 2 Statistical Table of Design Parameters for Underground Interchanges

[0126]

[0127]

[0128] Simulation scenario modeling is performed using UC-win / Road. After driving simulation experiments, the vehicle operation and running data can be obtained through the driving simulator. The output data mainly includes simulation time, driving distance, speed, steering wheel angle, acceleration, coordinates, distance from the left edge of the lane, distance from the right edge of the lane, fuel consumption, etc.

[0129] Furthermore, to explore the overall safety of urban underground interchanges, a comprehensive safety evaluation index system was constructed by selecting 12 driving behavior safety entropy indicators from the merging section, diverging section, and basic road section of urban underground interchanges. The calculation process of driving behavior safety entropy is shown in equations (1) to (5).

[0130] Table 3 Comprehensive Evaluation Index System for the Safety of Urban Underground Interchanges

[0131]

[0132] Furthermore, in order to evaluate the safety of the road, an urban underground interchange safety evaluation method based on entropy weight-improved grey near-optimal comprehensive evaluation method is adopted. The calculation process of this comprehensive evaluation method is shown in equations (6) to (16).

[0133] Then, based on the improved grey near-optimal model comprehensive evaluation method, composite grey elements were established for the overall evaluation indicators of the six underground interchanges. These were then substituted into the model to obtain the whitened grey matrix of the impact of each underground interchange on driver safety, as follows (all indicator results are rounded to three decimal places):

[0134]

[0135] Equation (8) maps the overall evaluation index values ​​of each underground interchange to the interval [0,1]. Considering the significant differences between the indicators of the basic road section and the merging / diversion section, the minimum value of each evaluation index of all merging / diversion sections of these 6 underground interchanges is selected as the specified smaller value for the indicators of the merging / diversion sections. For the basic road sections, the minimum value of each evaluation indicator for all basic road sections of these 6 underground interchanges is selected as the designated smaller value. The minimum values ​​for each indicator are specified as follows (all indicator results are rounded to three decimal places):

[0136]

[0137] The optimized whiteness matrix obtained after calculation is as follows:

[0138]

[0139] Next, the weights of each evaluation index are calculated using the entropy weight method (Equations (10) to (12)), and the resulting weight matrix is ​​as follows:

[0140] W=[0.0690.0800.1260.0680.0680.104 0.0830.0700.0800.0770.0710.102]

[0142] After determining the weights of each evaluation index, the specified minimum value and index weights are substituted into formula (16) to obtain the comprehensive safety evaluation model for underground interchanges. The specific model is shown in formula (18):

[0143]

[0144] Finally, the overall security evaluation value of each underground interchange was calculated, and the overall security evaluation value matrix is ​​as follows:

[0145]

[0146] The overall safety evaluation values ​​of the underground interchanges calculated using the above formula are then ranked. The ranking results of the overall safety evaluation values ​​are as follows: Figure 4 As shown.

[0147] After obtaining the overall safety evaluation value of underground interchanges, the overall safety level of urban underground interchanges is classified. First, the evaluation results output by the safety comprehensive evaluation model are grouped, as shown in Table 4.

[0148] Table 4. Distribution of Evaluation Results and Corresponding Probability Units

[0149]

[0150] Note: * Corrected by [1 - (1 / 4n) × 100%].

[0151] The regression equation is calculated according to formula (17):

[0152] RSR j = -0.275 + 0.127 Probit

[0153] The significance value of the t-test for the regression coefficient b is 0.001, and the goodness of fit of the equation is 0.985, which is close to 1. Therefore, the equation can be considered reasonable and has a good fit.

[0154] Based on the regression equation and the principle of reasonable classification, the evaluation level is divided into 5 levels. According to the comprehensive evaluation value from large to small, they are divided into five levels: safe, relatively safe, general, relatively dangerous and dangerous. The classification threshold value of each level is calculated, and the overall safety level of the 6 underground interchanges is classified. The results are shown in Table 5.

[0155] Table 5. Overall Safety Level Classification of Underground Interchanges

[0156]

[0157] According to the safety level classification results, the Qingdao Jiaozhou Bay Tunnel Interchange has the highest overall safety level, described as safe. The Hangzhou Zizhi Tunnel Interchange is relatively safe. The Nanjing Qingao Tunnel Interchange, Suining Guanyinhu Tunnel Interchange, and Shenzhen Henglongshan Tunnel Interchange have average overall safety levels. The Xiamen Wanshishan and Zhonggushan Tunnel Interchange has the lowest overall safety level, described as relatively dangerous. During the operation and management phase, it is necessary to strengthen safety measures for the entire road section to ensure driving safety.

[0158] This invention comprehensively selects driving behavior safety entropy from various sections of urban underground interchanges to construct a comprehensive evaluation index system. It uses evaluation index data from six urban underground interchanges in China as a dataset and employs an entropy-weighted, improved grey near-optimal comprehensive evaluation method to construct a comprehensive safety evaluation method for urban underground interchanges. The evaluation levels are then classified based on the rank-sum ratio method. Results show that the safety levels of each evaluated object are relatively evenly distributed across different levels, and the constructed comprehensive safety evaluation model can effectively evaluate the overall safety of urban underground interchanges.

[0159] The above embodiments of the present invention are merely examples for clearly illustrating the present invention and are not intended to limit the implementation of the present invention. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is impossible to exhaustively list all possible implementations here. All obvious variations or modifications derived from the technical solutions of the present invention are still within the protection scope of the present invention.

Claims

1. A safety evaluation method for urban underground interchanges based on driving simulation, characterized in that, Includes the following steps: 1) Select several existing underground interchanges in China to create simulation scenarios, and conduct experiments with drivers to collect vehicle operation and running data. 2) Determine the driving behavior indicators for each of the aforementioned urban underground interchanges; 3) Calculate the safety entropy of driving behavior indicators for each of the aforementioned urban underground interchanges using sample entropy; 4) Based on the safety entropy of driving behavior indicators for each of the aforementioned urban underground interchanges, calculate the comprehensive safety evaluation value for each of the aforementioned urban underground interchanges using the improved grey near-optimal comprehensive evaluation method; 5) Based on the comprehensive safety evaluation value, the rank-sum ratio method is used to classify each urban underground interchange into different grades to obtain the safety level of each urban underground interchange; Step 4) specifically includes: 4.1) The driving behavior indicators for each of the aforementioned urban underground interchanges are used as gray elements in the gray matrix; 4.2) The safety entropy of the driving behavior indicators for each of the aforementioned urban underground interchanges is used as the whitening gray element value; 4.3) The whitening gray element value is dimensionless to obtain the near-optimal whitening gray value; 4.4) Calculate the weights of each of the aforementioned driving behavior indicators using the entropy weight method; 4.5) Determine the comprehensive safety evaluation value of each of the aforementioned urban underground interchanges based on the near-optimal whitening gray value and weight.

2. The safety evaluation method for urban underground interchanges based on driving simulation according to claim 1, characterized in that, Step 4.3) specifically refers to: 4.3.1) Select the minimum value in the dataset of each driving behavior indicator as the specified smaller value; 4.3.2) Divide the specified smaller value by the whitening gray element value to obtain the near-optimal whitening gray value.

3. The safety evaluation method for urban underground interchanges based on driving simulation according to claim 2, characterized in that, Step 4.4) specifically refers to: 4.4.1) Calculate the characteristic weight of each of the driving behavior indicators, that is, the weight of a certain value of each of the driving behavior indicators in all the values ​​of the driving behavior indicator; 4.4.2) Calculate the information entropy value of each driving behavior indicator based on the feature weight of each indicator; 4.4.3) Calculate the weight of each driving behavior indicator based on the information entropy value of each driving behavior indicator.

4. The safety evaluation method for urban underground interchanges based on driving simulation according to claim 3, characterized in that, Step 5) specifically involves: 5.1) The comprehensive safety evaluation value of each of the aforementioned urban underground interchanges is grouped according to its magnitude; 5.2) List the frequency, cumulative frequency, average rank, and cumulative frequency for each group, and calculate the probability unit for each group; 5.3) Using the probability unit as the independent variable and the comprehensive safety evaluation value as the dependent variable, perform linear regression to obtain the linear regression equation; 5.4) Calculate the rank-sum ratio based on the linear regression equation; 5.5) Based on the rank-sum comparison, each urban underground interchange is classified and categorized to obtain the safety level of each urban underground interchange.

5. The safety evaluation method for urban underground interchanges based on driving simulation according to claim 4, characterized in that the linear regression equation is: ,in, The overall safety evaluation value, This is a unit of probability.

6. The method for safety evaluation of urban underground interchanges based on driving simulation according to any one of claims 1-5, wherein step 1) specifically involves: selecting six urban underground interchanges that have been built in China to model simulation scenarios, and conducting experiments with 20 drivers to collect vehicle operation and running data.

7. The safety evaluation method for urban underground interchanges based on driving simulation according to claim 6, characterized in that, The collected vehicle operation and running data for the experiment include simulation time, driving distance, speed, steering wheel angle, acceleration, coordinates, lateral displacement, distance from the left edge of the lane, distance from the right edge of the lane, and fuel consumption.

8. The safety evaluation method for urban underground interchanges based on driving simulation according to claim 7, characterized in that, Step 2) specifically involves dividing each of the urban underground interchanges into three types of road segments: merging segment, diverging segment, and basic road segment. Speed, acceleration, steering wheel turning, and lateral displacement from the experimental vehicle operation and running data of each road segment of each urban underground interchange are selected as driving behavior indicators. Thus, each urban underground interchange has a total of 12 driving behavior indicators.

Images