Road traffic accident black spot discrimination method based on space-time combination dimension

Abstract

The application provides a road traffic accident black point discrimination method based on a space-time combined dimension, which comprises the following steps: S1. Statistics of the traffic accident data occurring in the target section in N years, and the traffic accident data is subjected to spatial segmentation and time segmentation; S2. Calculation of the space-time overlap rate of the space-time composite point; S3. Calculation of the space-time accident rate of the space-time composite point; S4. Calculation of the final value G of the space-time composite point according to the space-time overlap rate, the space-time accident rate, the number of casualties and the economic loss; S5. Determination of the critical value, and the space-time composite point with the final value G greater than the critical value is judged as a traffic accident black point. Through the above method, the space and time are effectively combined, which is helpful to improve the existing road traffic conditions and improve the road driving safety.

Description

Technical Field

[0001] This invention relates to a method for identifying black spots in road traffic accidents, and more particularly to a method for identifying black spots in road traffic accidents based on a combination of spatiotemporal dimensions. Background Technology

[0002] While enjoying the immense convenience brought by modern transportation systems, people also face the dangers of road traffic accidents, resulting in significant property damage and personal injury. Therefore, researching methods for identifying black spots in road traffic accidents is of great importance for improving existing road traffic conditions and enhancing road safety.

[0003] Traffic accidents exhibit a clustering tendency in both time and space, and traditional methods for identifying accident black spots mostly focus on the spatial dimension. For example, traditional absolute number methods, relative number methods, and regression models first statistically analyze raw accident data, then calculate specific values ​​for relevant indicators at corresponding locations. When these indicators exceed a given threshold, the location is designated as an accident black spot. While temporal statistical analysis is also present, it primarily identifies spatial accident black spots first, followed by a simple explanation of the temporal distribution of traffic accidents. This analysis of accident temporal distribution is mostly aimed at identifying the causes of spatial accident black spots, rather than defining a specific point in time as an accident black spot. Furthermore, currently, no fundamental model can combine indicators such as the number of accidents, the number of vehicles passing through, the number of casualties, and economic losses from a spatiotemporal perspective.

[0004] This shows that there is a relative lack of ideas on using spatiotemporal composite points as accident black spots. Therefore, it is of great significance to study the identification method of road traffic accident black spots from the perspective of spatiotemporal combination. Summary of the Invention

[0005] In view of this, the present invention proposes a basic method for identifying black spots of road traffic accidents based on a combination of spatiotemporal dimensions, which solves the problem that traditional methods for identifying black spots of road traffic accidents do not combine space and time. Moreover, this basic method fully includes elements such as the number of accidents, the number of vehicles passing through, the number of casualties, and economic losses.

[0006] A method for identifying black spots in road traffic accidents based on a combination of spatiotemporal dimensions includes the following steps:

[0007] S1. Collect data on traffic accidents that occurred on the target road segment within N years, and segment the traffic accident data spatially and temporally;

[0008] The traffic accident data includes the number of accidents, the number of vehicles passing through, the number of casualties, and the economic losses.

[0009] The spatial segmentation refers to dividing the length of the target into I units evenly.

[0010] The time segmentation refers to dividing a natural day into J units evenly.

[0011] S2. Calculate the spatiotemporal overlap rate of spatiotemporal composite points;

[0012] S3. Calculate the spatiotemporal accident rate of spatiotemporal composite points;

[0013] S4. Calculate the final value G of the spatiotemporal composite point based on the spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic losses;

[0014] S5. Determine the critical value, and identify spatiotemporal composite points where the final value G is greater than the critical value as traffic accident black spots.

[0015] Furthermore, in step S2, the spatiotemporal overlap rate C of the spatiotemporal composite point is calculated using the following formula. (i,j) :

[0016] C (i,j) =τ i α i +τ j α j (2.1)

[0017] Where, τ i The weight represents the overlap rate of the i-th spatial cell, where i represents the spatial cell index and α represents the empty overlap rate. i τ represents the overlap rate of the i-th spatial unit. j The weight representing the overlap rate of the j-th time unit, where j represents the time unit number, α j This represents the overlap rate of the j-th time unit.

[0018] Furthermore, the spatial overlap rate and temporal overlap rate are calculated using the following formulas:

[0019]

[0020]

[0021] Where α represents the overlap rate, α i B represents the spatial overlap rate of the i-th spatial unit, and B represents the number of accidents. This represents the number of accidents occurring in the nth group within the i-th spatial unit. α represents the number of accidents occurring in the nth group within the jth time period in the i-th spatial segment. j This represents the time overlap rate of the j-th time unit. Let i represent the number of accidents that occurred in the nth group within the j-th time unit, i represent the spatial unit number, j represent the time unit number, n represent the data group number, and N represent the total number of data groups, where n∈N.

[0022] Furthermore, the weights of spatial overlap and temporal overlap are calculated through the following steps:

[0023] S21. Calculate the mean and standard deviation of the number of accidents in spatial and temporal units using the following formulas:

[0024]

[0025]

[0026]

[0027]

[0028] in, This represents the average number of accidents in the i-th spatial unit, and N represents the total number of data sets. This represents the number of accidents occurring in the nth group within the i-th spatial unit. This represents the average value in the j-th time unit. S represents the number of accidents occurring in the nth group within the j-th time unit. i S represents the standard deviation of the i-th spatial unit. j This represents the standard deviation of the j-th spatial unit;

[0029] S22. The weights for calculating the average number of accidents in spatial and temporal units, and the weights for calculating the standard deviation of the number of accidents in spatial and temporal units, are as follows:

[0030] S221. Standardize the mean and standard deviation of the number of accidents to obtain the standardized mean. and and the standardized standard deviation S' i and S′ j The calculation formula is as follows:

[0031]

[0032]

[0033]

[0034]

[0035] in, Represents the standardized average. Represents the standardized average. S' i S represents the standardized standard deviation. i S' j S represents the standardized standard deviation.j F and H represent coefficients, F + H = 1, I represents the total number of spatial units, and J represents the total number of time units. This represents the union of the set of spatial unit averages and the set of time unit averages. This represents the minimum value of the average value within both spatial and temporal units. S represents the maximum value of the average values ​​in both spatial and temporal units. I ∪S J The set of standard deviations of spatial units and the set of standard deviations of time units represents the union of these sets, max(S). I ∪S J ) represents the maximum standard deviation in the spatial and temporal units, min(S) I ∪S J () represents the minimum standard deviation in the spatial and temporal units;

[0036] S222. Calculate the weighting of the standardized mean and standard deviation of the accident data using the following formula:

[0037]

[0038]

[0039]

[0040]

[0041] in, Represents the standardized average. The proportion of the sum of the average values ​​of the standardized spatial units and the average values ​​of the time units. Represents the standardized average. I represents the total number of spatial units. Represents the standardized average. The proportion of the sum of the average values ​​of the standardized spatial units and the average values ​​of the time units. Represents the standardized average. J represents the total number of time units. S represents the standardized standard deviation. i The proportion of the sum of the standard deviations of spatial and temporal units after standardization, S' i S represents the standardized standard deviation. i , S represents the standardized standard deviation. j The proportion of the total standard deviation of the spatial unit and the standard deviation of the time unit after standardization;

[0042] S223. Calculate the average information entropy value of the number of accidents in the standardized spatial and temporal units, as well as the standard deviation information entropy value of the standardized spatial and temporal units, based on the proportions. The calculation formulas are as follows:

[0043]

[0044]

[0045] in, The information entropy value represents the average number of accidents in spatial and temporal units. The information entropy value represents the standard deviation of the number of accidents in spatial and temporal units, where I represents the total number of spatial units, i represents the spatial unit number, J represents the total number of temporal units, and j represents the temporal unit number. Represents the standardized average. The proportion of the sum of the average values ​​of the standardized spatial units and the average values ​​of the time units. Represents the standardized average. The proportion of the sum of the average values ​​of the standardized spatial units and the average values ​​of the time units. S represents the standardized standard deviation. i The proportion of the total standardized spatial and temporal unit standard deviations. S represents the standardized standard deviation. j The proportion of the total standardized spatial and temporal unit standard deviations.

[0046] S224. Calculate the average weight of spatial and temporal units, and the standard deviation weight of spatial and temporal units;

[0047] The calculation formula is as follows:

[0048]

[0049]

[0050] Where W1 represents the average weight of the number of accidents in the spatial unit and the time unit, and W2 represents the standard deviation weight of the number of accidents in the spatial unit and the time unit. The information entropy value represents the average number of accidents in spatial and temporal units. Information entropy values ​​representing the standard deviation of the number of accidents in spatial and temporal units;

[0051] S23. Calculate the spatial overlap rate weight τ for each spatial unit. i and the time overlap rate weight τ of each time unit j The calculation formula is as follows:

[0052]

[0053]

[0054] Where W1 represents the average weight of the number of accidents in spatial and temporal units, W2 represents the standard deviation weight of the number of accidents in spatial and temporal units, I represents the total number of spatial units, i represents the spatial unit number, J represents the total number of temporal units, and j represents the temporal unit number. S represents the average number of accidents in the i-th spatial unit. i This represents the standard deviation of the i-th spatial unit. S represents the average value of the j-th time unit. j This represents the standard deviation of the j-th time unit.

[0055] Furthermore, in step S3, the spatiotemporal accident rate R of each spatiotemporal composite point is calculated using the following formula. (i,j) :

[0056]

[0057] in, This represents the number of accidents occurring in the nth group within the j-th time period in the i-th spatial segment. Let represent the number of vehicles passing through the nth group in the jth time segment within the i-th spatial segment, where i represents the spatial unit number, j represents the time unit number, n represents the data group number, and N represents the total number of data groups, n∈N.

[0058] Furthermore, in step S4, the final value G of the spatiotemporal composite point is calculated using the following formula:

[0059] G (i,j) =K3C″ (i,j) +K4R″ (i,j) +K5D″ (i,j) +K6E″ (i,j) (4.1)

[0060]

[0061]

[0062]

[0063]

[0064] Where K3 represents the weight of the spatiotemporal overlap rate, K4 represents the weight of the spatiotemporal accident rate, K5 represents the weight of the number of casualties, K6 represents the weight of the economic loss, and C (i,j)C' represents the spatiotemporal overlap rate between the i-th spatial unit and the j-th temporal unit. ( ' i,j) Represents the normalized C (i,j) R (i,j) R' represents the spatiotemporal accident rate of the i-th spatial unit and the j-th time unit. ( ' i,j) Represents the normalized R (i,j) D (i,j) D' represents the number of casualties in the i-th spatial unit and the j-th time unit. ( ' i,j) D represents the normalized D (i,j) E (i,j) E' represents the economic loss in the i-th spatial unit and the j-th time unit. ( ' i,j) Represents the normalized E (i,j) .

[0065] Furthermore, the weights of spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic losses are calculated using the following method, with the calculation steps as follows:

[0066] S41. Standardize the spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic losses to obtain C'. (i,j) 、R' (i,j) D' (i,j) , and E' (i,j) The calculation formula is as follows:

[0067]

[0068]

[0069]

[0070]

[0071] Where i represents the spatial unit number, j represents the temporal unit number, and C (i,j) C' represents the spatiotemporal overlap rate of a spatiotemporal composite point. (i,j) Represents the standardized spatiotemporal overlap rate, min{C (I,J)} represents the minimum value in the set of spatiotemporal overlap rates, max{C (I,J)} represents the maximum value in the set of spatiotemporal overlap rates, R (i,j) R' represents the spatiotemporal accident rate of a spatiotemporal composite point. (i,j) Represents the standardized spatiotemporal accident rate, max{R IJ} represents the maximum value in the set of spatiotemporal accident rates, min{R} IJ} represents the minimum value in the set of spatiotemporal accident rates, D(i,j) D' represents the number of casualties at the spatiotemporal composite point. (i,j) Denotes the standardized number of casualties, min{D (I,J)} represents the minimum value in the set of casualties at the spatiotemporal composite point, max{D (I,J)} represents the maximum value in the set of casualties at the spatiotemporal composite point, E (i,j) E' represents the economic loss at the spatiotemporal composite point. (i,j) Represents the standardized economic loss, min{E (I,J)} represents the minimum value in the set of economic losses at the spatiotemporal composite point, max{E (I,J)} represents the maximum value in the set of economic losses at the spatiotemporal composite point, F and H represent coefficients, F+H=1, I represents the total number of spatial units, and J represents the total number of time units;

[0072] S42. Calculate the standardized spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and proportion of economic loss, and obtain the following results: and The calculation formula is as follows:

[0073]

[0074]

[0075]

[0076]

[0077] Where i represents the spatial unit number, I represents the total number of spatial units, j represents the temporal unit number, and J represents the total number of temporal units. C' represents the proportion of the standardized spatiotemporal overlap rate. (i,j) This represents the standardized spatiotemporal overlap rate. R' represents the proportion of the standardized spatiotemporal accident rate. (i,j) This represents the standardized spatiotemporal accident rate. D' represents the proportion of standardized casualties. (i,j) This represents the standardized number of casualties. E' represents the proportion of economic loss after standardization. (i,j) This represents the economic loss after standardization;

[0078] S43. Calculate the information entropy of spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic loss, and obtain e respectively. C e R e D and e E The calculation formula is as follows:

[0079]

[0080]

[0081]

[0082]

[0083] in, i represents the spatial unit number, I represents the total number of spatial units, j represents the temporal unit number, and J represents the total number of temporal units. The proportion representing the standardized spatiotemporal overlap rate. This represents the proportion of the standardized spatiotemporal accident rate. This represents the proportion of standardized casualties. This indicates the proportion of economic losses after standardization.

[0084] S44. Calculate the weights of spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic losses using the following formula:

[0085]

[0086]

[0087]

[0088]

[0089] Where K3 represents the weight of the spatiotemporal overlap rate, K4 represents the weight of the spatiotemporal accident rate, K5 represents the weight of the number of casualties, K6 represents the weight of the economic loss, and e C Information entropy, e, represents the spatiotemporal overlap rate. R Information entropy, e, represents the spatiotemporal accident rate. D The information entropy representing the number of casualties, e E Information entropy represents economic losses.

[0090] Furthermore, in step S5, the critical value is determined through the following steps:

[0091] S51. Sort the final value G of the spatiotemporal composite point in ascending order and divide it into T groups;

[0092] S52. Calculate the cumulative frequency of the final value G of the spatiotemporal composite point, and plot the scatter plot of the final value G;

[0093] L T =P1+P2+P3+…+P t (5.1)

[0094]

[0095] Among them, L T P represents the cumulative frequency of the final value G of group T. t V represents the frequency of the final value G in the t-th group. t Let G represent the number of final values ​​G in the t-th group, and V represent the total number of final values ​​G.

[0096] S53. Fit the hyperbolic tangent function y of the final value G to the scatter plot until the function converges;

[0097] S54. The Y-value corresponding to the minimum radius of curvature of the hyperbolic tangent function is determined as the critical value, where the minimum radius of curvature ρ is:

[0098]

[0099] Where y' represents the first derivative of the hyperbolic tangent function y, and y” represents the second derivative of the hyperbolic tangent function y.

[0100] The beneficial effects of this invention: This invention proposes a new method for identifying black spots in road traffic accidents. By combining spatial and temporal factors and incorporating elements such as the number of accidents, the number of vehicles passing through, the number of casualties, and economic losses, this invention can comprehensively identify black spots in road traffic accidents, which helps to improve existing road traffic conditions and enhance road driving safety. Attached Figure Description

[0101] The present invention will be further described below with reference to the accompanying drawings and embodiments:

[0102] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0103] A method for identifying black spots in road traffic accidents based on a combination of spatiotemporal dimensions includes the following steps:

[0104] S1. Collect data on traffic accidents that occurred on the target road segment within N years, and segment the traffic accident data spatially and temporally;

[0105] The traffic accident data includes the number of accidents, the number of vehicles passing through, the number of casualties, and the economic losses.

[0106] The spatial segmentation refers to dividing the length of the target into I units evenly.

[0107] The time segmentation refers to dividing a natural day into J units evenly.

[0108] S2. Calculate the spatiotemporal overlap rate of spatiotemporal composite points;

[0109] S3. Calculate the spatiotemporal accident rate of spatiotemporal composite points;

[0110] S4. Calculate the final value G of the spatiotemporal composite point based on the spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic losses;

[0111] S5. Determine the critical value, and identify spatiotemporal composite points where the final value G is greater than the critical value as traffic accident black spots.

[0112] In this embodiment, step S1 involves statistically analyzing traffic accident data occurring on the target road segment within N years, and then spatially and temporally segmenting the traffic accident data. The traffic accident data includes the number of accidents, the number of vehicles passing through, the number of casualties, and economic losses. For example, it involves statistically analyzing the number of accidents occurring between 79km and 80km of the Fengdian Tunnel section of the Chengdu-Nanchong Expressway from 2019 to 2022. Temporally, the natural day is divided into 24 time units, with a 1-hour time interval, i.e., 0:00-1:00 (inclusive). The time intervals are 79km-79.1km (inclusive of 79, excluding 79.1), 79.1km-79.2km (inclusive of 79.1, excluding 79.2), and so on. The 80km section is divided into 79.9km-80km.

[0113] In this embodiment, in step S2, the spatiotemporal composite point refers to the point where the time unit and the spatial unit coincide. For example, 79km-79.1km and 0:00-1:00 represent a set of spatiotemporal composite points, denoted as 0K-0; 79.6km-79.7km and 12:00-13:00 represent a set of spatiotemporal composite points, denoted as 6K-12; 79.9km-80km and 23:00-24:00 represent a set of spatiotemporal composite points, denoted as 9K-23; and so on for the remaining spatiotemporal composite points.

[0114] The spatiotemporal overlap rate of spatiotemporal composite points is calculated using the following method:

[0115] C (i,j) =τ i α i +τ j α j (2.1)

[0116] Where, τ i The weight represents the overlap rate of the i-th spatial cell, where i represents the spatial cell index and α represents the empty overlap rate. i τ represents the overlap rate of the i-th spatial unit. jThe weight representing the overlap rate of the j-th time unit, where j represents the time unit number, α j This represents the overlap rate of the j-th time unit.

[0117] Furthermore, the spatial overlap rate and temporal overlap rate are calculated using the following formulas:

[0118]

[0119]

[0120] Where α represents the overlap rate, α i B represents the spatial overlap rate of the i-th spatial unit, and B represents the number of accidents. This represents the number of accidents occurring in the nth group within the i-th spatial unit. α represents the number of accidents occurring in the nth group within the jth time period in the i-th spatial segment. j This represents the time overlap rate of the j-th time unit. Let i represent the number of accidents that occurred in the nth group within the j-th time unit, i represent the spatial unit number, j represent the time unit number, n represent the data group number, and N represent the total number of data groups, where n∈N.

[0121] When analyzing data from 2019 to 2022, N equals 4. Each year, from smallest to largest, represents a set of data, for a total of 4 sets of data. In the spatial unit, This indicates the number of accidents that occurred in the second space unit in 2020.

[0122] When the spatial overlap rate and temporal overlap rate are equal to zero, it will affect subsequent calculations. Adding 0.2% when calculating the spatial overlap rate and temporal overlap rate can ensure the validity of the data and has a small impact on the overall data.

[0123] Furthermore, the weights of spatial overlap and temporal overlap are calculated through the following steps:

[0124] S21. Calculate the mean and standard deviation of the number of accidents in spatial and temporal units using the following formulas:

[0125]

[0126]

[0127]

[0128]

[0129] in, This represents the average number of accidents in the i-th spatial unit, and N represents the total number of data sets. This represents the number of accidents occurring in the nth group within the i-th spatial unit. This represents the average value in the j-th time unit. S represents the number of accidents occurring in the nth group within the j-th time unit. i S represents the standard deviation of the i-th spatial unit. j This represents the standard deviation of the j-th spatial unit;

[0130] S22. The weights for calculating the average number of accidents in spatial and temporal units, and the weights for calculating the standard deviation of the number of accidents in spatial and temporal units, are as follows:

[0131] S221. Standardize the mean and standard deviation of the number of accidents to obtain the standardized mean. and and the standardized standard deviation S' i and S′ j The calculation formula is as follows:

[0132]

[0133]

[0134]

[0135]

[0136] in, Represents the standardized average. Represents the standardized average. S' i S represents the standardized standard deviation. i S' j S represents the standardized standard deviation. j F and H represent coefficients, F + H = 1, I represents the total number of spatial units, and J represents the total number of time units. This represents the union of the set of spatial unit averages and the set of time unit averages. This represents the minimum value of the average value within both spatial and temporal units. S represents the maximum value of the average values ​​in both spatial and temporal units. I ∪S J The set of standard deviations of spatial units and the set of standard deviations of time units represents the union of these sets, max(S). I ∪S J ) represents the maximum standard deviation in the spatial and temporal units, min(S) I ∪S J () represents the minimum standard deviation in the spatial and temporal units;

[0137] Where F+H=1, and the range of F is [0.995, 0.999], and the range of H is [0.001, 0.005]. Setting the coefficients F and H ensures that the standardized mean and standard deviation are no less than 0 and no greater than 1, thus guaranteeing the validity of the data.

[0138] S222. Calculate the weighting of the standardized mean and standard deviation of the accident data using the following formula:

[0139]

[0140]

[0141]

[0142]

[0143] in, Represents the standardized average. The proportion of the sum of the average values ​​of the standardized spatial units and the average values ​​of the time units. Represents the standardized average. I represents the total number of spatial units. Represents the standardized average. The proportion of the sum of the average values ​​of the standardized spatial units and the average values ​​of the time units. Represents the standardized average. J represents the total number of time units. S represents the standardized standard deviation. i The proportion of the sum of the standard deviations of spatial and temporal units after standardization, S' i S represents the standardized standard deviation. i , S represents the standardized standard deviation. j The proportion of the total standard deviation of the spatial unit and the standard deviation of the time unit after standardization;

[0144] S223. Calculate the average information entropy value of the number of accidents in the standardized spatial and temporal units, as well as the standard deviation information entropy value of the standardized spatial and temporal units, based on the proportions. The calculation formulas are as follows:

[0145]

[0146]

[0147] in, The information entropy value represents the average number of accidents in spatial and temporal units. The information entropy value represents the standard deviation of the number of accidents in spatial and temporal units, where I represents the total number of spatial units, i represents the spatial unit number, J represents the total number of temporal units, and j represents the temporal unit number. Represents the standardized average. The proportion of the sum of the average values ​​of the standardized spatial units and the average values ​​of the time units. Represents the standardized average. The proportion of the sum of the average values ​​of the standardized spatial units and the average values ​​of the time units. S represents the standardized standard deviation. i The proportion of the total standardized spatial and temporal unit standard deviations. S represents the standardized standard deviation. j The proportion of the total standardized spatial and temporal unit standard deviations.

[0148] S224. Calculate the average weight of spatial and temporal units, and the standard deviation weight of spatial and temporal units;

[0149] The calculation formula is as follows:

[0150]

[0151]

[0152] Where W1 represents the average weight of the number of accidents in the spatial unit and the time unit, and W2 represents the standard deviation weight of the number of accidents in the spatial unit and the time unit. The information entropy value represents the average number of accidents in spatial and temporal units. Information entropy values ​​representing the standard deviation of the number of accidents in spatial and temporal units;

[0153] S23. Calculate the spatial overlap rate weight τ for each spatial unit. i and the time overlap rate weight τ of each time unit j The calculation formula is as follows:

[0154]

[0155]

[0156] Where W1 represents the average weight of the number of accidents in spatial and temporal units, W2 represents the standard deviation weight of the number of accidents in spatial and temporal units, I represents the total number of spatial units, i represents the spatial unit number, J represents the total number of temporal units, and j represents the temporal unit number. S represents the average number of accidents in the i-th spatial unit. iThis represents the standard deviation of the i-th spatial unit. S represents the average value of the j-th time unit. j This represents the standard deviation of the j-th time unit.

[0157] In this embodiment, in step S3, the spatiotemporal accident rate R of each spatiotemporal composite point is calculated using the following formula. (i,j) :

[0158]

[0159] in, This represents the number of accidents occurring in the nth group within the j-th time period in the i-th spatial segment. Let represent the number of vehicles passing through the nth group in the jth time segment within the i-th spatial segment, where i represents the spatial unit number, j represents the time unit number, n represents the data group number, and N represents the total number of data groups, n∈N.

[0160] In this embodiment, in step S4, the final value G of the spatiotemporal composite point is calculated using the following formula:

[0161] G (i,j) =K3C″ (i,j) +K4R″ (i,j) +K5D″ (i,j) +K6E″ (i,j) (4.1)

[0162]

[0163]

[0164]

[0165]

[0166] Where K3 represents the weight of the spatiotemporal overlap rate, K4 represents the weight of the spatiotemporal accident rate, K5 represents the weight of the number of casualties, K6 represents the weight of the economic loss, and C (i,j) C' represents the spatiotemporal overlap rate between the i-th spatial unit and the j-th temporal unit. ( ' i,j) Represents the normalized C (i,j) R (i,j) R' represents the spatiotemporal accident rate of the i-th spatial unit and the j-th time unit. ( ' i,j) Represents the normalized R (i,j) D (i,h) D' represents the number of casualties in the i-th spatial unit and the j-th time unit. ( ' i,j) D represents the normalized D(i,j) E (i,j) E' represents the economic loss in the i-th spatial unit and the j-th time unit. ( ' i,j) Represents the normalized E (i,j) .

[0167] Furthermore, the weights of spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic losses are calculated using the following method, with the calculation steps as follows:

[0168] S41. Standardize the spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic losses to obtain C'. (i,j) 、R' (i,j) D' (i,j) , and E' (i,j) The calculation formula is as follows:

[0169]

[0170]

[0171]

[0172]

[0173] Where i represents the spatial unit number, j represents the temporal unit number, and C (i,j) C' represents the spatiotemporal overlap rate of a spatiotemporal composite point. (i,j) Represents the standardized spatiotemporal overlap rate, min{C (I,J)} represents the minimum value in the set of spatiotemporal overlap rates, max{C (I,J)} represents the maximum value in the set of spatiotemporal overlap rates, R (i,j) R' represents the spatiotemporal accident rate of a spatiotemporal composite point. (i,j) Represents the standardized spatiotemporal accident rate, max{R IJ} represents the maximum value in the set of spatiotemporal accident rates, min{R} IJ} represents the minimum value in the set of spatiotemporal accident rates, D (i,j) D' represents the number of casualties at the spatiotemporal composite point. (i,j) Denotes the standardized number of casualties, min{D (I,J)} represents the minimum value in the set of casualties at the spatiotemporal composite point, max{D (I,J)} represents the maximum value in the set of casualties at the spatiotemporal composite point, E (i,j) E' represents the economic loss at the spatiotemporal composite point. (i,j) Represents the standardized economic loss, min{E (I,J)} represents the minimum value in the set of economic losses at the spatiotemporal composite point, max{E (I,J)} represents the maximum value in the set of economic losses at the spatiotemporal composite point, F and H represent coefficients, F+H=1, and the range of F is [0.995, 0.999], the range of H is [0.001, 0.005], I represents the total number of spatial units, and J represents the total number of time units;

[0174] S42. Calculate the standardized spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and proportion of economic loss, and obtain the following results: and The calculation formula is as follows:

[0175]

[0176]

[0177]

[0178]

[0179] Where i represents the spatial unit number, I represents the total number of spatial units, j represents the temporal unit number, and J represents the total number of temporal units. C' represents the proportion of the standardized spatiotemporal overlap rate. (i,j) This represents the standardized spatiotemporal overlap rate. R' represents the proportion of the standardized spatiotemporal accident rate. (i,j) This represents the standardized spatiotemporal accident rate. D' represents the proportion of standardized casualties. (i,j) This represents the standardized number of casualties. E' represents the proportion of economic loss after standardization. (i,j) This represents the economic loss after standardization;

[0180] S43. Calculate the information entropy of spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic loss, and obtain e respectively. C e R e D and e E The calculation formula is as follows:

[0181]

[0182]

[0183]

[0184]

[0185] in, i represents the spatial unit number, I represents the total number of spatial units, j represents the temporal unit number, and J represents the total number of temporal units. The proportion representing the standardized spatiotemporal overlap rate. This represents the proportion of the standardized spatiotemporal accident rate. This represents the proportion of standardized casualties. This indicates the proportion of economic losses after standardization.

[0186] S44. Calculate the weights of spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic losses using the following formula:

[0187]

[0188]

[0189]

[0190]

[0191] Where K3 represents the weight of the spatiotemporal overlap rate, K4 represents the weight of the spatiotemporal accident rate, K5 represents the weight of the number of casualties, K6 represents the weight of the economic loss, and e C Information entropy, e, represents the spatiotemporal overlap rate. R Information entropy, e, represents the spatiotemporal accident rate. D The information entropy representing the number of casualties, e E Information entropy represents economic losses.

[0192] In this embodiment, in step S5, the critical value is determined and traffic accident black spots are identified through the following steps:

[0193] S51. Sort the final values ​​G of the spatiotemporal composite points in ascending order. Divide the final values ​​G into T groups according to their distribution. These groups can be either equidistant or non-equidistant. Arrange the Q final values ​​G in ascending order and calculate the difference between adjacent final values ​​G to obtain Q-1 differences. If the Q-1 differences fluctuate around a certain horizontal line, then equidistant groups are used. If there is an abrupt change between the Q-1 differences, then the data corresponding to the abrupt difference are non-equidistant groups. The specific group interval is determined according to actual needs. The smaller the group interval, the higher the accuracy and the more complex the calculation process.

[0194] S52. Calculate the cumulative frequency of the final value G of the spatiotemporal composite point, and plot the scatter plot of the final value G;

[0195] L T =P1+P2+P3+…+P t (5.1)

[0196]

[0197] Among them, L T L represents the cumulative frequency of the final value G of group T. T =1, P t V represents the frequency of the final value G in the t-th group. t Let G represent the number of final values ​​G in the t-th group, and V represent the total number of final values ​​G.

[0198] S53. In Origin, fit the hyperbolic tangent function y of the final value G based on the scatter plot until the function converges. In the Origin interface, select Analysis-Fit-Nonlinear Curve Fitting-Category to create a new function y = a + btanh(cx + cd).

[0199] S54. The Y-value corresponding to the minimum radius of curvature of the hyperbolic tangent function is determined as the critical value, where the minimum radius of curvature ρ is:

[0200]

[0201] Where y' represents the first derivative of the hyperbolic tangent function y, and y” represents the second derivative of the hyperbolic tangent function y.

[0202] Based on step S54, the critical value is determined, and the maximum total value G that is greater than the critical value is identified as a traffic accident black spot.

[0203] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for identifying black spots in road traffic accidents based on a spatiotemporal combination, characterized in that: Includes the following steps: S1. Collect data on traffic accidents that occurred on the target road segment within N years, and segment the traffic accident data spatially and temporally; The traffic accident data includes the number of accidents, the number of vehicles passing through, the number of casualties, and the economic losses. The spatial segmentation refers to dividing the length of the target into I units evenly. The time segmentation refers to dividing a natural day into J units evenly. S2. Calculate the spatiotemporal overlap rate of spatiotemporal composite points; S3. Calculate the spatiotemporal accident rate of spatiotemporal composite points; S4. Calculate the final value of the spatiotemporal composite point based on the spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic losses. ; S5. Determine the critical value, and set the final value. Spatiotemporal composite points exceeding the critical value are identified as traffic accident black spots. In step S2, the spatiotemporal overlap rate of the spatiotemporal composite point is calculated using the following formula. : in, The weight representing the overlap rate of the i-th spatial unit, where i represents the spatial unit number. This represents the overlap rate of the i-th spatial unit. This represents the weight of the overlap rate of the j-th time unit, where j represents the time unit number. This represents the overlap rate of the j-th time unit; Spatial overlap rate and temporal overlap rate are calculated using the following formulas: in, Indicates the overlap rate. B represents the spatial overlap rate of the i-th spatial unit, and B represents the number of accidents. This represents the number of accidents occurring in the nth group within the i-th spatial unit. This represents the number of accidents occurring in the nth group within the j-th time period in the i-th spatial segment. This represents the time overlap rate of the j-th time unit. This represents the number of accidents occurring in the nth group within the j-th time unit, where i represents the spatial unit number, j represents the time unit number, n represents the data group number, and N represents the total number of data groups. .

2. The method for identifying road traffic accident black spots based on a spatiotemporal combination dimension as described in claim 1, characterized in that: The weights of spatial overlap and temporal overlap are calculated using the following steps: S21. Calculate the mean and standard deviation of the number of accidents in spatial and temporal units using the following formulas: in, This represents the average number of accidents in the i-th spatial unit, and N represents the total number of data sets. This represents the number of accidents occurring in the nth group within the i-th spatial unit. This represents the average value in the j-th time unit. This represents the number of accidents occurring in the nth group within the j-th time unit. This represents the standard deviation of the i-th spatial unit. This represents the standard deviation of the j-th time unit; S22. The weights for calculating the average number of accidents in spatial and temporal units, and the weights for calculating the standard deviation of the number of accidents in spatial and temporal units, are as follows: S221. Standardize the mean and standard deviation of the number of accidents to obtain the standardized mean. and and the standard deviation after standardization and The calculation formula is as follows: in, Represents the standardized average. , Represents the standardized average. , Standard deviation after standardization , Standard deviation after standardization , H represents the coefficient, F+H=1, I represents the total number of spatial units, and J represents the total number of time units. This represents the union of the set of spatial unit averages and the set of time unit averages. This represents the minimum value of the average value within both spatial and temporal units. This represents the maximum value of the average value within both the spatial and temporal units. This represents the union of the set of standard deviations for spatial units and the set of standard deviations for temporal units. This represents the maximum standard deviation within the spatial and temporal units. This represents the minimum standard deviation within the spatial and temporal units; S222. Calculate the weighting of the standardized mean and standard deviation of the accident data using the following formula: in, Represents the standardized average. The proportion of the sum of the average values ​​of the standardized spatial units and the average values ​​of the time units. Represents the standardized average. I represents the total number of spatial units. Represents the standardized average. The proportion of the sum of the average values ​​of the standardized spatial units and the average values ​​of the time units. Represents the standardized average. J represents the total number of time units. Standard deviation after standardization The proportion of the total standardized spatial and temporal unit standard deviations. Standard deviation after standardization , Standard deviation after standardization The proportion of the total standard deviation of the spatial unit and the standard deviation of the time unit after standardization; S223. Calculate the average information entropy value of the number of accidents in the standardized spatial and temporal units, as well as the standard deviation information entropy value of the standardized spatial and temporal units, based on the proportions. The calculation formulas are as follows: in, The information entropy value represents the average number of accidents in spatial and temporal units. The information entropy value represents the standard deviation of the number of accidents in spatial and temporal units, where I represents the total number of spatial units, i represents the spatial unit number, J represents the total number of temporal units, and j represents the temporal unit number. Represents the standardized average. The proportion of the sum of the average values ​​of the standardized spatial units and the average values ​​of the time units. Represents the standardized average. The proportion of the sum of the average values ​​of the standardized spatial units and the average values ​​of the time units. Standard deviation after standardization The proportion of the total standardized spatial and temporal unit standard deviations. Standard deviation after standardization The proportion of the total standardized spatial and temporal unit standard deviations. ; S224. Calculate the average weight of spatial and temporal units, and the standard deviation weight of spatial and temporal units; The calculation formula is as follows: in, The average weight representing the number of accidents in spatial and temporal units. The standard deviation weights representing the number of accidents in spatial and temporal units. The information entropy value represents the average number of accidents in spatial and temporal units. Information entropy values ​​representing the standard deviation of the number of accidents in spatial and temporal units; S23. Calculate the spatial overlap rate weights of each spatial unit. and the weight of time overlap rate of each time unit The calculation formula is as follows: in, The average weight representing the number of accidents in spatial and temporal units. The standard deviation weights represent the number of accidents in spatial and temporal units, where I represents the total number of spatial units, i represents the spatial unit number, J represents the total number of temporal units, and j represents the temporal unit number. This represents the average number of accidents in the i-th spatial unit. This represents the standard deviation of the number of accidents in the i-th spatial unit. This represents the average value of the j-th time unit. This represents the standard deviation of the number of accidents in the j-th time unit.

3. The method for identifying road traffic accident black spots based on a spatiotemporal combination dimension as described in claim 1, characterized in that: In step S3, the spatiotemporal accident rate of each spatiotemporal composite point is calculated using the following formula. : in, This represents the number of accidents occurring in the nth group within the j-th time period in the i-th spatial segment. This represents the number of vehicles passing through the nth group within the j-th time segment in the i-th spatial segment, where i represents the spatial unit number, j represents the time unit number, n represents the data group number, and N represents the total number of data groups. .

4. The method for identifying road traffic accident black spots based on a spatiotemporal combination dimension as described in claim 1, characterized in that: In step S4, the final value of the spatiotemporal composite point is calculated using the following formula. : in, The weights representing the spatiotemporal overlap rate The weights representing the spatiotemporal accident rate The weights representing the number of casualties Weights representing economic losses This represents the spatiotemporal overlap rate between the i-th spatial unit and the j-th temporal unit. Represents the normalized result , This represents the spatiotemporal accident rate of the i-th spatial unit and the j-th time unit. Represents the normalized result , This represents the number of casualties in the i-th spatial unit and the j-th time unit. Represents the normalized result , This represents the economic loss in the i-th spatial unit and the j-th time unit. Represents the normalized result .

5. The method for identifying road traffic accident black spots based on a spatiotemporal combination dimension as described in claim 4, characterized in that: The weights of spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic losses are calculated using the following method, with the calculation steps as follows: S41. Standardize the spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic losses to obtain the following results: , , ,and The calculation formula is as follows: Where i represents the spatial unit number and j represents the temporal unit number. The spatiotemporal overlap rate represents the spatiotemporal composite point. This represents the standardized spatiotemporal overlap rate. This represents the minimum value in the set of spatiotemporal overlap rates. This represents the maximum value in the set of spatiotemporal overlap rates. The spatiotemporal accident rate represents the spatiotemporal composite point. This represents the standardized spatiotemporal accident rate. This represents the maximum value in the set of spatiotemporal accident rates. This represents the minimum value in the set of spatiotemporal accident rates. This indicates the number of casualties at the spatiotemporal composite point. This represents the standardized number of casualties. This represents the minimum value in the set of casualties at a spatiotemporal composite point. This represents the maximum value in the set of casualties at a spatiotemporal composite point. This represents the economic loss at the spatiotemporal composite point. This represents the economic loss after standardization. This represents the minimum value in the set of economic losses at the spatiotemporal composite point. This represents the maximum value in the set of economic losses at the spatiotemporal composite point. H represents the coefficient, F+H=1, I represents the total number of spatial units, and J represents the total number of time units; S42. Calculate the standardized spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and proportion of economic loss, and obtain the following results: , , and The calculation formula is as follows: Where i represents the spatial unit number, I represents the total number of spatial units, j represents the temporal unit number, and J represents the total number of temporal units. The proportion representing the standardized spatiotemporal overlap rate. This represents the standardized spatiotemporal overlap rate. This represents the proportion of the standardized spatiotemporal accident rate. This represents the standardized spatiotemporal accident rate. This represents the proportion of standardized casualties. This represents the standardized number of casualties. This indicates the proportion of economic loss after standardization. This represents the economic loss after standardization; S43. Calculate the information entropy of spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic loss, and obtain the following results: , , and The calculation formula is as follows: in, , where i represents the spatial unit number, I represents the total number of spatial units, j represents the temporal unit number, and J represents the total number of temporal units. The proportion representing the standardized spatiotemporal overlap rate. This represents the proportion of the standardized spatiotemporal accident rate. This represents the proportion of standardized casualties. This indicates the proportion of economic losses after standardization. S44. Calculate the weights of spatiotemporal overlap rate, spatiotemporal accident rate, number of casualties, and economic losses using the following formula: in, The weights representing the spatiotemporal overlap rate The weights representing the spatiotemporal accident rate The weights representing the number of casualties Weights representing economic losses Information entropy, representing the spatiotemporal overlap rate, Information entropy represents the spatiotemporal accident rate. Information entropy representing the number of casualties, Information entropy represents economic losses.

6. The method for identifying road traffic accident black spots based on a spatiotemporal combination dimension as described in claim 1, characterized in that: In step S5, the critical value is determined through the following steps: S51. Sort the final value G of the spatiotemporal composite point in ascending order and divide it into T groups; S52. Calculate the cumulative frequency of the final value G of the spatiotemporal composite point, and plot the scatter plot of the final value G; in, This represents the cumulative frequency of the final value G in group T. This represents the frequency of the final value G in the t-th group. This represents the number of final values ​​G in the t-th group. This represents the total number of final values ​​G; S53. Fit the hyperbolic tangent function y of the final value G to the scatter plot until the function converges; S54. The Y value corresponding to the minimum radius of curvature of the hyperbolic tangent function is determined as the critical value, where the minimum radius of curvature... for: in, Let denote the first derivative of the hyperbolic tangent function y. Let represent the second derivative of the hyperbolic tangent function y.

Images