Algorithm for Correction Coefficient of Capacity Model of Special-Shaped Signalized Intersection

By analyzing the characteristics of non-motorized vehicle violations at irregular intersections, a traffic capacity correction model was established, which solved the problem of insufficient traffic capacity prediction accuracy at irregular intersections and achieved more accurate traffic capacity prediction.

CN118379892BActive Publication Date: 2024-11-19HOHAI UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410637266.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-22
Publication Date
2024-11-19
Estimated Expiration
2044-05-22

AI Technical Summary

Technical Problem

In existing technologies, there is limited research on the capacity of irregular signalized intersections, especially the impact of non-motorized vehicles running red lights on the capacity of motorized vehicles has not been fully considered, resulting in insufficient accuracy in capacity prediction.

Method used

By collecting basic information on irregular intersections and data on non-motorized vehicle violations, we analyze the spatiotemporal characteristics of non-motorized vehicle violations, identify key influencing factors, and establish a signal control capacity correction model for irregular intersections under the influence of non-motorized vehicle violations, including correction coefficients for saturated headway and acceleration loss time of motor vehicles.

Benefits of technology

The prediction accuracy of the capacity model for irregular intersections has been improved, and it can more accurately reflect the impact of non-motorized vehicle violations on motorized vehicle capacity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118379892B_ABST
    Figure CN118379892B_ABST
Patent Text Reader

Abstract

The invention relates to a correction coefficient algorithm for the capacity model of a special-shaped signal intersection. For a special-shaped intersection, basic information of the special-shaped intersection, non-motor vehicle violation behavior data, and motor vehicle data under the influence of non-motor vehicle violation behavior are collected; according to the collected non-motor vehicle violation behavior data, the spatiotemporal characteristics of the non-motor vehicle violation behavior are analyzed, and the spatiotemporal characteristics include time characteristics and space characteristics, that is, the analysis is performed from the two perspectives of time characteristics and space characteristics, and the analyzed factors are used as possible influencing factors on the motor vehicle traffic flow, and these possible influencing factors are preliminarily screened to obtain the key influencing factors that affect the saturated motor vehicle traffic flow; based on the motor vehicle data under the influence of non-motor vehicle violation behavior, the screened key influencing factors are respectively analyzed to obtain the influence coefficients of different key influencing factors on the motor vehicle operation characteristics under the influence of non-motor vehicle violation behavior; different influence coefficients are integrated to determine the correction coefficients of the saturated headway and acceleration loss time of the motor vehicle, and a signal control special-shaped intersection capacity correction model under the influence of non-motor vehicle violation behavior is established; and finally, the established capacity correction model is verified. The invention provides an important reference for the organization optimization and signal timing of special-shaped signal intersections.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a correction coefficient algorithm for a capacity model of a special-shaped signalized intersection, and belongs to the technical field of traffic engineering. Background Art

[0002] Intersections are important nodes in the road traffic network, where traffic flows of various directions and types converge, causing serious traffic conflicts. Affected by factors such as terrain and buildings, special-shaped intersections are more common in cities. According to the "China Key City Road Network Structure Portrait Report", special-shaped intersections account for 38.8% of urban intersections. Due to their irregular shape, unsmooth lane lines and unsatisfactory sight distance, compared with conventional intersections, their traffic organization is more difficult, their traffic capacity is more insufficient, and they are more likely to become traffic congestion and accident-prone points. Signal special-shaped intersections often resolve conflicts between motor vehicles and non-motor vehicles through time separation. However, non-motor vehicles have the characteristics of high flexibility and low violation costs, so red light running is frequent. This not only increases the risk of accidents, but also has a serious impact on the traffic capacity of special-shaped intersections.

[0003] Chen Xiaoming et al. studied the capacity of signalized intersections under the influence of pedestrians and non-motorized vehicles respectively; Sun Feng et al. studied the capacity of road sections under the influence of straight and bay-type bus stops based on queuing theory and gap theory; Qin Lihui et al. studied the capacity of urban trunk roads under the interference of parallel bus stops and the capacity of urban main and secondary trunk roads under the interference of bicycles; Ye Xiaofei et al. studied the impact of on-street parking on the capacity of non-motorized lanes; Zhao et al. proposed a quantitative model for the impact of roadside parking on intersection capacity; Cheng Wei et al. and Mathew et al. used gap acceptance theory and modified HCM equation respectively to explore the capacity of roundabouts under mixed traffic flow conditions; Huang Yanguo et al. established a non-motorized vehicle capacity model based on the headway correction coefficient. The capacity model of signalized intersections under vehicle interference; Hong et al. proposed a capacity evaluation method based on VISSIM to explore the capacity of urban expressways after traffic accidents; Wang Jiawen et al. studied the capacity of road sections under the influence of traffic events based on a multi-layer fuzzy model; Alshabibi et al. studied the impact of short-term work zones on the capacity of signalized intersections in New York City; Kota et al. studied the impact of right-turning vehicles under mixed traffic on the capacity of urban uncontrolled intersections. Most of the existing capacity research is about road sections or normal intersections, and there are fewer studies on special-shaped intersections, and even fewer studies on the capacity of special-shaped intersections under the condition of non-motor vehicles running red lights. Therefore, it is of great significance to fill this gap. Summary of the invention

[0004] The invention provides a correction coefficient algorithm for the capacity model of a special-shaped signalized intersection, which provides an important reference for the organization optimization and signal timing of the special-shaped signalized intersection.

[0005] The technical solution adopted by the present invention to solve its technical problem is:

[0006] The correction coefficient algorithm of the capacity model of the special-shaped signalized intersection specifically includes the following steps:

[0007] Step S1: data collection and sorting, for special-shaped intersections, basic information of special-shaped intersections, non-motor vehicle violation behavior data, and motor vehicle data affected by non-motor vehicle violation behavior;

[0008] Step S2: Analyze the spatiotemporal characteristics of the non-motor vehicle violation behavior data collected in step S1, wherein the spatiotemporal characteristics include time characteristics and space characteristics, that is, analyze from the two perspectives of time characteristics and space characteristics, and use the analyzed factors as possible influencing factors on the motor vehicle traffic flow, preliminarily screen these possible influencing factors, and screen out the key influencing factors that affect the saturated motor vehicle traffic flow;

[0009] Step S3: Based on the motor vehicle data collected in step S1 and whether there is an impact of non-motor vehicle violations, the key influencing factors screened in step S2 are analyzed respectively to obtain the influence coefficients of different key influencing factors on the motor vehicle operation characteristics under the impact of non-motor vehicle violations;

[0010] Step S4: Integrate the different influence coefficients in step S3, determine the correction coefficients of the saturated headway and acceleration loss time of motor vehicles, and establish a traffic capacity correction model for signal-controlled special-shaped intersections under the influence of non-motor vehicle violations;

[0011] Step S5: verifying the established capacity correction model;

[0012] As a further preferred embodiment of the present invention, in step S1, the special-shaped intersection refers to an intersection where one of the road angles is less than 75°;

[0013] The basic information of special-shaped intersections includes dividing lanes into small-angle left-turn lanes, large-angle left-turn lanes, small-angle right-turn lanes, and large-angle right-turn lanes according to the road angle and direction;

[0014] It also includes cycle, phase, yellow light duration, green light duration and red light duration, providing various parameters required to calculate the effective green light time;

[0015] The data on non-motor vehicle violation behaviors include the time when non-motor vehicles are in the right of way and the right of no way, the type of non-motor vehicle, the time when non-motor vehicles arrive, the time when non-motor vehicles violate the law, the time when non-motor vehicles stop violating the law, the distance of violation, the duration of violation, and the average saturated headway of motor vehicles;

[0016] Motor vehicle data include the type of motor vehicle, the time when the motor vehicle reaches the stop line, and whether there are any non-motor vehicles violating traffic rules;

[0017] As a further preferred embodiment of the present invention, in step S1, the headway, start loss time, acceleration time and road clearing time of two motor vehicles are calculated by the time when the motor vehicle reaches the stop line, with the signal light phase as the unit;

[0018] The violation distance is the shortest distance between the front end of the non-motor vehicle and the motor vehicle traffic flow;

[0019] The duration of the violation is the difference between the time when the non-motor vehicle violates the law and the time when the non-motor vehicle stops violating the law;

[0020] Based on the collected headway time, statistical analysis is performed on the average headway time of each motor vehicle in each phase. If there is no significant difference in the average headway time of motor vehicles in adjacent positions, it is considered that the headway time remains stable after that position, that is, the average headway time thereafter is the saturated headway time, and the motor vehicle positions when the traffic flow of different types of lanes reaches saturated traffic flow are judged based on this saturated headway time.

[0021] As a further preferred embodiment of the present invention, in step S2, the time characteristics include the start time of the non-motor vehicle violation and the duration ratio of the non-motor vehicle violation, and the spatial characteristics include the relative position of the non-motor vehicle, the number of violations of the non-motor vehicle group and the violation distance;

[0022] The above-mentioned temporal and spatial characteristics are used as possible influencing factors of non-motor vehicle violations on the saturated headway of motor vehicles. The binomial logistic regression model is used for factor screening, and the key influencing factors affecting the saturated traffic flow of motor vehicles include the relative position of non-motor vehicles, the number of violations of non-motor vehicle groups, the duration ratio of violations, and the distance of violations.

[0023] As a further preferred embodiment of the present invention, the violation duration ratio is the ratio of the non-motor vehicle violation duration to the corresponding red light phase duration when the non-motor vehicle violates the law;

[0024] Define the duration ratio of the violation as λ, the duration of the non-motor vehicle violation as t1, in seconds, and the corresponding red light phase duration when the non-motor vehicle violates the law as t2, in seconds. The calculation formula is:

[0025]

[0026] As a further preferred embodiment of the present invention, in step S3, the influence of non-motor vehicle violation on motor vehicle traffic is divided into inner influence and outer influence, and the key influencing factors screened in step S2 are analyzed by K-means clustering method and data fitting method to obtain a saturated headway influence coefficient, wherein the saturated headway influence coefficient includes the influence coefficient of different violation duration ratios on the saturated headway and the influence coefficient of the violation distance on the saturated headway;

[0027] Based on the motor vehicle data collected in step S1 with or without the influence of non-motor vehicle violations, the green light loss time influence coefficient is analyzed by using a data fitting method to obtain the influence coefficient of the motor vehicle acceleration loss time;

[0028] As a further optimization of the present invention, the K-means clustering method is used to classify the influence of different violation duration ratios on the saturated headway, and then the cluster member intervals of different types of lanes are merged to obtain four ranges of violation duration ratios of special-shaped intersections λ, namely λ∈[0,0.22), λ∈[0.22,0.48), λ∈[0.48,0.75) and λ∈[0.75,1.5]; according to the saturated headway of different types of lanes in the absence of violation, the influence coefficients of the violation duration ratios of the four ranges under different types of lanes on the saturated headway are calculated respectively;

[0029] The Weibull function model is used to fit the influence of the violation distance inside and outside different types of lanes on the saturated headway time. The function is expressed as:

[0030]

[0031] In the formula, H is the saturated headway of motor vehicle traffic, in seconds; l is the violation distance, in meters; a, b, c, and k are all parameters;

[0032] The K-means clustering method is used to classify the irregular intersections according to the violation distances of different types of lanes, and four levels are obtained, namely l∈[0,1.3); l∈[1.3,2.2); l∈[2.2,3.2); l∈[3.2,4.5]. The influence coefficients of the saturated headway of different types of lanes in different ranges are calculated respectively.

[0033] A linear model is used to fit the data of the motor vehicle start time with and without the influence of non-motor vehicle violations to determine the motor vehicle start time. The formula is:

[0034] T 启 =a×k+b

[0035] In the formula, T 启is the starting time of the queue vehicle, in seconds; k is the position of the queue vehicle in the queue; a and b are parameters;

[0036] The Gompertz model is used to fit the motor vehicle acceleration process to obtain the acceleration time of motor vehicles in different types of lanes under different combinations of special-shaped intersections and violations. The formula is:

[0037]

[0038] In the formula, V(t) is the speed of the motor vehicle after it starts, in m / s; t is the time series after the motor vehicle starts, in s; a, b, c are parameters;

[0039] The time when the last motor vehicle passes the stop line minus the time when the red light turns on during the traffic phase is collected under the influence of non-motor vehicle violations to obtain the road clearing loss time. The sample expected value is used as the road clearing loss time. The formula is:

[0040] t 清 =E(t a -t b )

[0041] Where: t 清 is the road clearing loss time, unit is s; E is the expected value of the sample; t a The time when the red light turns on during the motor vehicle traffic phase, in seconds; t b The time when the last motor vehicle passes the stop line, in seconds;

[0042] As a further optimization of the present invention, during the actual acceleration process of the motor vehicles queuing at the intersection, it takes a certain amount of time to reach the saturated flow state. The motor vehicles start to accelerate during the yellow light time, so the yellow light time is also a component of the effective green light time. At the same time, due to the influence of non-motor vehicle violations, the saturated headway and acceleration loss time of motor vehicles should also be corrected. Finally, in step S4, the established traffic capacity correction model is:

[0043]

[0044] In the formula, C i is the traffic capacity of a lane, in pcu / h; n is t 加 The number of vehicles passing the stop line within a certain time period; T is the cycle length, A is the yellow light time, t g is the green light time, t 启 is the startup loss time, t 加 is the acceleration loss time, t 清 is the road clearing loss time, h is the saturated headway, both in seconds; is the acceleration loss time correction factor; is the saturated headway correction factor;

[0045] As a further preferred embodiment of the present invention, the correction coefficient of the saturated headway of the motor vehicle is as follows:

[0046]

[0047] In the formula, is the saturated headway correction coefficient; f T is the influence coefficient of different violation duration ratios on the saturated headway of motor vehicles; f L is the influence coefficient of different violation distance ratios on the saturated headway of motor vehicles;

[0048] Acceleration loss time correction factor, the formula is as follows:

[0049]

[0050] In the formula, is the acceleration loss time correction factor; t 加有 is the acceleration loss time under different combinations of non-motor vehicle violation behaviors; t 加无 is the acceleration loss time without the influence of non-motor vehicle violation, and the range of acceleration loss time without the influence of non-motor vehicle violation is λ∈[0,0.22), l∈[3.2,4.5];

[0051] As a further preferred embodiment of the present invention, in step S5, the verification of the capacity correction model includes verifying the actual capacity and the corrected capacity error of the capacity correction model and verifying the actual capacity and the uncorrected capacity error of the capacity correction model.

[0052] Through the above technical solution, compared with the prior art, the present invention has the following beneficial effects:

[0053] The correction coefficient algorithm of the capacity model of special-shaped signalized intersections provided by the present invention fully considers actual working conditions such as the duration of non-motor vehicle violations at special-shaped intersections, the different capacities of turning lanes caused by different turning angles due to intersection angles, and the time lost by green lights, and can improve the prediction accuracy of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0055] Figure 1 Schematic diagram of left-turn lanes at different angles in a preferred embodiment provided by the present invention;

[0056] Figure 2 It is a diagram showing the relationship between headway and queue position in a preferred embodiment provided by the present invention;

[0057] Figure 3 It is a schematic diagram of the inside and outside of a non-motor vehicle running a red light relative to a motor vehicle flow in a preferred embodiment provided by the present invention;

[0058] Figure 4 It is a clustering k-value line graph based on the elbow method in a preferred embodiment provided by the present invention;

[0059] Figure 5 is a fitting result diagram of a special-shaped intersection in a preferred embodiment provided by the present invention;

[0060] Figure 6 It is a linear fitting diagram of the motor vehicle starting law in the preferred embodiment provided by the present invention;

[0061] Figure 7 It is a speed fitting diagram of motor vehicles in different types of lanes in a preferred embodiment provided by the present invention;

[0062] Figure 8 is a frequency distribution histogram of the time lost in clearing roads for motor vehicles in a preferred embodiment provided by the present invention;

[0063] Fig. 9 The invention provides a flow chart of the algorithm for the correction coefficient of the capacity model of the special-shaped signalized intersection. DETAILED DESCRIPTION

[0064] The present invention will now be described in further detail with reference to the accompanying drawings. The specific dimensions used in this embodiment are only for illustrating the technical solution and do not limit the protection scope of the present invention.

[0065] As explained in the background technology, there are very few studies on special-shaped intersections. For example, in existing studies on non-motor vehicle violations, only the different times when the violations occurred are considered, but few people have analyzed and studied the duration of the violations; in existing studies on the capacity of signal-controlled intersections, the different turning angles caused by the intersection angles have not been taken into account, resulting in different capacities of turning lanes; in existing studies on green light loss time, the green light loss time consists of three parts: start-up time, acceleration time, and road clearing time. However, all of these loss times are not taken into account in existing studies on the capacity of signal-controlled intersections. Therefore, this application aims to strengthen the analysis of non-motor vehicle violations, select appropriate methods to analyze the impact of non-motor vehicle violations on motor vehicle operating characteristics, and ultimately obtain a correction model for the capacity of signal-controlled special-shaped intersections under the influence of non-motor vehicle violations.

[0066] Next, the correction coefficient algorithm of the capacity model of the special-shaped signalized intersection provided by this application is explained in detail, and actual examples are interspersed in the method explanation to assist understanding. Specifically, the following steps are included:

[0067] Step S1: data collection and sorting, for special-shaped intersections, basic information of special-shaped intersections, non-motor vehicle violation behavior data, and motor vehicle data affected by non-motor vehicle violation behavior;

[0068] Before collecting data, it is necessary to explain the main components of special-shaped intersections. The "China Key Cities Road Network Structure Portrait Report" points out that oblique cross and oblique T-shaped intersections account for 75.78% of special-shaped intersections and are the main components of special-shaped intersections.

[0069] In the embodiment, this application selected the intersections of Zhanjiang Road and Hanjiang Road (oblique cross shape) and Lhasa Road and Guangzhou Road (oblique T-shaped) in Nanjing, and conducted morning and evening rush hour surveys using video shooting on weekdays with good weather. The survey time and basic characteristics of the intersections are shown in Table 1.

[0070] Table 1 Basic characteristics of special-shaped intersections

[0071]

[0072] Since the significant feature of oblique cross and oblique T-shaped intersections is the different road angles, combined with Table 1, this application defines an irregular intersection as one of the road angles at the intersection is less than 75°. Since the operation of the straight lane in an irregular intersection is not affected by the road angle, this application divides the lanes into four categories according to different angles and turns: small-angle left-turn lanes, large-angle left-turn lanes, small-angle right-turn lanes, and large-angle right-turn lanes for research. Figure 1 Schematic diagrams of left turns at different angles are provided, and investigation and research are carried out on these four types of lanes.

[0073] When collecting information on opposite intersections, it also includes cycle, phase, yellow light duration, green light duration and red light duration, providing various parameters required to calculate the effective green light time.

[0074] According to the video, the data of non-motor vehicles and motor vehicles are sorted, and the signal light phase is used as a unit to facilitate the subsequent analysis of the impact on the average headway. The data on non-motor vehicle violations in step S1 include the type of non-motor vehicle, the arrival time of non-motor vehicles, the time of non-motor vehicle violation, the time when non-motor vehicles stop violating the law, the violation distance, the duration of violation, and the average saturated headway of motor vehicles when non-motor vehicles are in the right of way and the right of no way; the motor vehicle data include the type of motor vehicle, the time when the motor vehicle arrives at the stop line, and whether there are non-motor vehicles violating the law. Among them, the headway, start-up loss time, acceleration time and road clearing time of the two motor vehicles are calculated by the time when the motor vehicle arrives at the stop line. As shown in Table 2, the violation distance in Table 2 is the shortest distance between the front end of the non-motor vehicle and the motor vehicle traffic; the duration of the violation is the difference between the time when the non-motor vehicle violates the law and the time when the non-motor vehicle stops violating the law.

[0075] Table 2 Specific operation status of motor vehicles and non-motor vehicles

[0076]

[0077] Based on the collected headway time, a statistical analysis is performed on the average headway time of each motor vehicle in a queue in each phase. If there is no significant difference in the average headway time of motor vehicles in queues at adjacent positions, it is considered that the headway time remains stable after that position, that is, the average headway time thereafter is the saturated headway time, and the position of motor vehicles when the traffic flow in different types of lanes reaches saturated traffic flow is judged based on this saturated headway time.

[0078] Implementation example Figure 2 The figure shows the relationship between the saturated headway time and queue position of different types of lanes in the special-shaped intersection when there is no non-motor vehicle running a red light. It can be seen from the figure that the small / large angle left-turn lanes of the oblique T-shaped type reach the saturated flow state at the fourth vehicle, and the saturated headway time is 2.18 and 2.15 respectively; the small / large angle right-turn lanes reach the saturated flow state at the third vehicle, and the saturated headway time is 2.17 and 2.13 respectively. The saturated headway time of the oblique cross type is the same as that of the oblique T-shaped type. The saturated headway time of the small / large angle left-turn lane and the small / large angle right-turn lane are 2.26, 2.15, 2.23 and 2.18 respectively.

[0079] If the impact of non-motor vehicle violations on traffic capacity is to be studied, the spatiotemporal characteristics of non-motor vehicle violations should be studied first. Step S2 studies and analyzes the spatiotemporal characteristics of non-motor vehicle violations based on the non-motor vehicle violation data collected in step S1. The spatiotemporal characteristics are analyzed from two perspectives: time characteristics and spatial characteristics. The time characteristics include the start time of the non-motor vehicle violation and the duration ratio of the non-motor vehicle violation. The spatial characteristics include the relative position of the non-motor vehicle, the number of violations of the non-motor vehicle group, and the violation distance. The above five characteristics are used as possible influencing factors on the motor vehicle traffic flow and analyzed.

[0080] The violation duration ratio is the ratio of the duration of the non-motor vehicle violation to the duration of the corresponding red light phase when the non-motor vehicle violates the law;

[0081] Regarding the duration ratio of non-motor vehicle violations, the duration ratio of violations is defined as λ, the duration of non-motor vehicle violations is t1, the unit is s, and the corresponding red light phase duration when non-motor vehicle violations occur is t2, the unit is s, then the calculation formula is:

[0082]

[0083] Specifically, the above-mentioned time characteristics and spatial characteristics are used as possible influencing factors of non-motor vehicle violations on the saturated headway of motor vehicles, and a binomial logistic regression model is used for factor screening. The established binomial logistic regression model must first determine the dependent variable. The binomial logistic model means that there are only two situations. In the actual research of this application, the dependent variable is divided into two types: one is the impact on the saturated headway of motor vehicles, that is, the saturated headway is greater than the saturated headway when there is no violation, and the dependent variable in this case is 1; the other is no impact on the saturated headway of motor vehicles, that is, the saturated headway is equal to the saturated headway when there is no violation, and the dependent variable in this case is 0.

[0084] Then confirm the independent variables and select the starting time of non-motor vehicle violation, the duration ratio of non-motor vehicle violation, the relative position of non-motor vehicle, the number of violations of non-motor vehicle group and the violation distance as independent variables.

[0085] Some relevant test settings are also carried out in the preferred embodiment, mainly selecting the HL test and the sum test of the model coefficients to ensure the accuracy of the model. Finally, the model is run to obtain the output results to determine the key influencing factors affecting the saturated traffic flow of motor vehicles, including the relative position of non-motor vehicles, the number of violations of the non-motor vehicle group, the duration ratio of violations, and the distance of violations.

[0086] Next, step S3 is based on the motor vehicle data collected in step S1 on the presence or absence of non-motor vehicle violations, and analyzes the key influencing factors screened in step S2 to obtain the influence coefficients of different key influencing factors on the motor vehicle operation characteristics under the influence of non-motor vehicle violations.

[0087] First, we analyze the impact of the relative position of non-motor vehicles' violations on the headway time of motor vehicles. The position of non-motor vehicles that run red lights relative to the motor vehicle traffic in the turning lane can be divided into two situations: one is on the inside of the turning radius, that is, close to the inside of the turn; the other is on the outside of the turning radius, that is, close to the outside. Figure 3 The figure shows the inner and outer positions of non-motor vehicles that run red lights in the left-turn lane relative to the motor vehicle traffic. The red non-motor vehicles in the figure are vehicles that are running red lights. Since the two positions have different degrees of impact on the motor vehicle traffic, the impact of non-motor vehicles running red lights on the motor vehicle traffic is divided into inner and outer impacts for study.

[0088] The second is the analysis of the impact of the violation duration ratio on the motor vehicle headway time. The aforementioned violation duration ratio is the ratio of the duration of a non-motor vehicle running a red light to the corresponding red light phase duration when the non-motor vehicle runs a red light. In order to analyze the impact of different violation duration ratios on the motor vehicle headway time, they need to be classified. K-means is widely used because of its advantages such as easy implementation, high operating efficiency and high speed. Therefore, this paper adopts the K-means algorithm to divide the violation duration ratio into k categories, where the k value is determined by the elbow method. The core of the elbow method is to observe the line graph of the sum of squared errors (SSE) changing with the k value, and find the k value corresponding to the point where the SSE begins to stabilize. This value is the number of clusters. The preferred embodiment of the present application selects a k value range of 2-6, calculates each k value to obtain the corresponding SSE value and draws the corresponding line graph. Figure 4 This is the line graph of the intersection of Lhasa Road and Guangzhou Road. Figure 4 It can be determined that the optimal k value for different types of lanes is 4. Repeat the above operation for the intersection of Zhanjiang Road and Hanjiang Road. Since the result is similar to the above intersection, it will not be repeated here.

[0089] After determining that k = 4 is the optimal number of clusters, the K-means algorithm is re-run after setting k = 4 to obtain 4 cluster centers, and then each data point is assigned to the cluster center closest to it, and finally the cluster member of each data point is obtained. Table 3 lists the specific cluster centers and cluster member intervals of the violation duration ratios of the two special-shaped intersections.

[0090] Table 3 Cluster center and cluster member interval table of violation duration ratio

[0091]

[0092] By comparing the cluster member intervals of the two intersections in Table 3, it can be found that the cluster member intervals of different types of lanes are very close. Therefore, this paper merges the cluster member intervals of different types of lanes to obtain four ranges of violation duration ratios at special-shaped intersections: λ∈[0,0.22), λ∈[0.22,0.48), λ∈[0.48,0.75) and λ∈[0.75,1.5]. Based on the saturated headway time of different types of lanes without violations, the influence coefficients of the violation duration ratios of the four ranges under different types of lanes on the saturated headway time are calculated, as shown in Table 4.

[0093] Table 4 Saturated headway influence coefficient under different violation duration ratios

[0094]

[0095] In order to analyze the influence of the violation distance on the headway of motor vehicles, a scatter plot of the violation distance on the inner side of the small-angle left-turn lane and the saturated headway is drawn, and it is observed that its variation law is nonlinear. Therefore, a curve model that conforms to its law can be found through fitting. Common models commonly used to describe the saturated headway of motor vehicles include negative exponential function, displacement negative exponential function and Weibull function model, but negative exponential and displacement negative exponential functions are mostly used to describe discrete traffic. Therefore, this application selects the Weibull function model for fitting, and the function expression is as follows:

[0096]

[0097] In the formula, H is the saturated headway time of motor vehicle traffic flow, unit is s; l is the violation distance, unit is m; a, b, c, k are all parameters.

[0098] Origin software was used for curve fitting. Figure 5 This is the fitting result diagram of the special-shaped intersection. Then the fitting model is analyzed by R 2 The specific test results and parameter estimates are shown in Table 5.

[0099] Table 5 Weibull model test and parameter table of traffic violation distance

[0100]

[0101] Through the fitting curve and data collection accuracy, 0.1m is taken as the violation distance interval value, and the corresponding saturated headway time is calculated. Then, based on the saturated headway time without non-motor vehicle violations, the increase rate of saturated headway time under different violation distances is calculated. However, it can be seen from the calculation that some violation distances have the same increase rate, so the above method is used to perform K-means cluster analysis on them, and the k value is determined to be 4 by the elbow method. Therefore, the embodiment of the present application divides the influence of the violation distance of the special-shaped intersection on the saturated headway time into 4 categories, namely l∈[0,1.3); l∈[1.3,2.2); l∈[2.2,3.2); l∈[3.2,4.5]; the classification results and corresponding influence coefficients are shown in Table 6.

[0102] Table 6 Saturated headway influence coefficient under different violation distances

[0103]

[0104] Regarding the analysis of the green light loss time coefficient, this application first studies the motor vehicle start time, and investigates the start time of queued vehicles in four types of lanes, taking into account the influence of non-motor vehicle violations. The start time is the time when the queued motor vehicle starts to start minus the time when the green light comes on. The study found that the start time conforms to the linear law, so a linear model is used for data fitting, and the formula is:

[0105] T 启 =a×k+b

[0106] In the formula, T 启 is the starting time of the queuing vehicle, in seconds; k is the position of the queuing vehicle in the queue; a and b are parameters.

[0107] Through analysis, it is found that non-motor vehicle violations have almost no effect on the motor vehicle startup loss time and road clearing loss time, and only affect the motor vehicle acceleration loss time. Therefore, the green light loss time influence coefficient only includes the influence coefficient of the motor vehicle acceleration loss time.

[0108] In the preferred embodiment, as shown in the linear function fitting results in Table 7, the average start time intervals of the vehicles in the four types of lanes are basically the same and the maximum difference is 0.04s, which can be ignored. Because the start rules of the vehicles in the queues of different types of lanes are basically the same, the data of the four types of lanes are combined and processed, and the start rules of the vehicles in the queues of the special-shaped intersections are finally analyzed. Figure 6 As shown, the function parameter a is 1.01, the parameter b is 0.79, and the model fit goodness is 0.95, which is a good effect. The same operation is performed on the motor vehicle start time without non-motor vehicle violations, and it is found that its start time is 1.79s, indicating that non-motor vehicle violations have no effect on the motor vehicle start time. Therefore, this application sets the motor vehicle start time to 1.80s.

[0109] Table 7 Fitting parameters of motor vehicle traffic start-up rules

[0110]

[0111] Then the acceleration time of motor vehicles was analyzed. The most mainstream models for describing the queued vehicles starting first and then accelerating to the saturated flow speed are the two-stage acceleration model, the linear reduction acceleration model and the polynomial acceleration model. However, these models all have some problems: the two-stage acceleration model does not conform to the actual situation, often resulting in a sudden change in speed, and the acceleration law under the influence of non-motor vehicle violations cannot be described by two stages; the acceleration of the affected motor vehicles in the linear reduction acceleration model is not a simple linear reduction, but also a process of getting faster and slower; the principle of the polynomial acceleration model is to use the Taylor formula to approximate and solve to form a fitting model, and the actual situation is not analyzed.

[0112] According to actual observations of special-shaped intersections, the acceleration of queued motor vehicles can be divided into three stages: initial acceleration stage: the initial speed is zero and the initial acceleration is small; rapid acceleration stage: the speed increases and the acceleration gradually increases; saturated flow speed stage: the speed gradually stabilizes and the acceleration begins to decrease until it reaches zero. The things described by the growth curve function are also divided into three stages: occurrence, development and maturity, and the acceleration law of each stage coincides with the acceleration law of queued motor vehicles. Therefore, this application selects the S-curve to fit the acceleration process of motor vehicles.

[0113] The most common growth curve models are the Logistic model and the Gompertz model. The difference between the two is whether the two sides of the S-curve are symmetrical. The inflection point of the Logistic model is approximately at 1 / 2 of the maximum value, while the inflection point of the Gompertz model is biased forward. According to actual observation data, the inflection point of the acceleration process of the queued motor vehicles is biased forward, not at 1 / 2 of the maximum value. Therefore, the Gompertz model is used to fit the acceleration law of the queued motor vehicles. The formula is:

[0114]

[0115] In the formula, V(t) is the speed of the motor vehicle after it starts, in m / s; t is the time series after the motor vehicle starts, in s; a, b, and c are parameters.

[0116] In the above model, according to the function extreme value theorem, when time t tends to ∞, the value of V(t) is a, which is the theoretical saturated flow speed of motor vehicles. Curve fitting is performed on different types of lanes at special-shaped intersections to analyze the change of motor vehicle speed over time after starting under the influence of non-motor vehicle violations. Figure 7 This is the fitting curve diagram of different types of lanes at the intersection of Lhasa Road and Guangzhou Road.

[0117] Table 8 shows the test results and parameter estimates of different types of lanes at the intersection of Lhasa Road and Guangzhou Road. It can be seen from the table that the fitting effect is good. The same curve fitting is performed on Zhanjiang Road and Hanjiang Road. Since the results are not much different from those of the intersection, they are not described here.

[0118] Table 8 Speed ​​fitting test and parameters of motor vehicles in different types of lanes

[0119]

[0120] According to the above classification of violation duration ratio and violation distance, the speed of motor vehicles under different violation combinations is fitted to obtain the speed change curves over time under different violation conditions. Then, the inverse function of the fitting function of the above acceleration law is obtained to obtain the acceleration time of motor vehicles in different types of lanes under different special-shaped intersections and different violation influence combinations. The specific results are shown in Table 9.

[0121] Table 9 Acceleration schedules for different types of lanes under different impact combinations

[0122]

[0123] Finally, the analysis of the road clearing loss time for motor vehicles is studied. The time when the last motor vehicle passes the stop line minus the time when the red light turns on during the traffic phase is collected under the influence of non-motor vehicle violations. The sample expected value is used as the road clearing loss time. The formula is:

[0124] t 清 =E(t a -t b )

[0125] Where: t 清 is the road clearing loss time, unit is s; E is the expected value of the sample; t a The time when the red light turns on during the motor vehicle traffic phase, in seconds; t b It is the time when the last motor vehicle passes the stop line, in seconds.

[0126] In the preferred embodiment, according to the above formula, the loss time of motor vehicle road clearing under the influence of non-motor vehicle violation is calculated, and a total of 334 sets of data are obtained, and their distribution is as follows: Figure 8 As shown, t is calculated 清 = 2.30s. Perform the same operation on the road clearance loss time of motor vehicles without the influence of non-motor vehicle violations to calculate t 清 =2.28s. Since the difference between the two is only 0.02s, it can be seen that the non-motor vehicle violation has almost no effect on the road clearance loss time of the motor vehicle, so the present application sets the road clearance loss time of the motor vehicle to 2.29s.

[0127] Step S4, establish a correction model for the traffic capacity of signal-controlled special-shaped intersections under the influence of non-motor vehicle violations. There are four common methods for calculating the traffic capacity of intersections: saturated flow rate, stop line, conflict point and stop line method. Considering China's unique road conditions and the loss time during the start and acceleration of queued vehicles, this application adopts the stop line method and corrects it to calculate the traffic capacity. Through analysis, it can be seen that in the actual intersection queued motor vehicle start acceleration process, it takes a certain amount of time to reach the saturated flow state. The loss time in the stop line method only considers the start loss time, and does not consider the subsequent vehicle acceleration time and road clearing loss time. Since the motor vehicle starts to start and accelerate within the yellow light time, the yellow light time is also a component of the effective green light time. At the same time, due to the influence of non-motor vehicle violations, the saturated headway and acceleration loss time of motor vehicles should also be corrected. The final established traffic capacity correction model is:

[0128]

[0129] In the formula, C i is the traffic capacity of a lane, in pcu / h; n is t 加 The number of vehicles passing the stop line within a certain time period; T is the cycle length, A is the yellow light time, t g is the green light time, t 启 is the startup loss time, t 加 is the acceleration loss time, t 清 is the road clearing loss time, h is the saturated headway, both in seconds; is the acceleration loss time correction factor; is the saturated headway correction factor.

[0130] The impact of non-motor vehicle violations on the saturated headway of motor vehicles comes from the violation duration ratio and violation distance. Since different violation duration ratios and violation distances will have different impacts on the saturated headway of motor vehicles, the influence coefficients of the saturated headway of motor vehicles are weighted according to the two influencing factors to obtain the correction coefficient of the saturated headway of motor vehicles. The formula is:

[0131]

[0132] In the formula, is the saturated headway correction coefficient; f T is the influence coefficient of different violation duration ratios on the saturated headway of motor vehicles; f L is the influence coefficient of different violation distance ratios on the saturated headway of motor vehicles. T 、f L It can be found from Table 4 and Table 6 that The calculation results are shown in Table 10.

[0133] Table 10 Correction coefficients for headway times of different types of lane saturation

[0134]

[0135] It can be concluded from Table 4 and Table 9 that when λ∈[0,0.22) and l∈[3.2,4.5], there is no influence on the acceleration loss time of motor vehicles. Therefore, the acceleration loss time within this range is regarded as the acceleration loss time without the influence of non-motor vehicle violations. The acceleration loss time correction coefficient under different influence combinations is calculated according to the following formula:

[0136]

[0137] In the formula, is the acceleration loss time correction factor; t 加有 is the acceleration loss time under different combinations of non-motor vehicle violation behaviors; t 加无 It is the time lost due to acceleration without the influence of non-motor vehicle violations.

[0138] Table 11 is a schematic table of the acceleration loss time correction coefficients for different types of lanes, specifically:

[0139] Table 11 Correction coefficients for acceleration loss time in different types of lanes

[0140]

[0141]

[0142] Finally, in order to verify the accuracy of the capacity correction model for the capacity of other special-shaped intersections, this application selects the capacity of different types of lanes at two special-shaped intersections for calculation, and compares the calculation results with the actual capacity to calculate the error.

[0143] The intersection of Yuhua Road and Yuhua East Road is an oblique T-shaped intersection, and the intersection of Ninghai Road and Beijing West Road is an oblique cross-shaped intersection. The basic conditions of motor vehicles and non-motor vehicles in different types of lanes at the two special-shaped intersections are investigated, and their traffic capacities are calculated according to the above-mentioned correction formula. The calculation results are shown in Table 12.

[0144] Table 12 Capacity of different types of roads at special-shaped intersections

[0145]

[0146]

[0147] According to Table 12, the error rate of the small-angle right-turn lane at the oblique T-shaped intersection is significantly greater than that of other lanes. The error rate between the measured and corrected capacity of the small-angle left-turn lane at the oblique cross-shaped intersection exceeds 15%. The reason for the error is that the green light time is short. The other lanes meet the general error requirements. Therefore, it can be verified that the capacity correction model can well reflect the impact of non-motor vehicle violations on the capacity of most special-shaped intersections.

[0148] In summary, this application as a whole includes three parts. The first part is to collect data on non-motor vehicle violations, analyze the spatiotemporal characteristics of non-motor vehicle violations and screen out the key factors that affect the saturated traffic flow of motor vehicles; the second part is to study the impact of key factors in non-motor vehicle violations on the operating characteristics of motor vehicles, including the saturated headway of motor vehicles and the time to start acceleration and clearing the road; the third part, based on the second part, establishes a capacity correction model for signal-controlled special-shaped intersections under the influence of non-motor vehicle violations, determines the correction coefficients of different parameters, and verifies the final capacity correction model by example. In the first part, the spatiotemporal characteristics of non-motor vehicle violations are analyzed from the perspectives of temporal characteristics and spatial characteristics, and the analyzed factors are used as possible influencing factors on motor vehicle traffic flow, and these possible influencing factors are preliminarily screened to obtain key influencing factors. In the second part, based on the motor vehicle operation characteristic data with or without the influence of non-motor vehicle violations, the key influencing factors are analyzed respectively, and the influence coefficients of different key influencing factors on the operating characteristics of motor vehicles are obtained, which lays the foundation for the establishment of the capacity correction model under the influence of non-motor vehicle violations in the third part. In the third part, the influence coefficients of different key influencing factors are integrated to determine the correction coefficients of the saturated headway and acceleration loss time of motor vehicles. Finally, the capacity correction model of signal-controlled special-shaped intersections under the influence of non-motor vehicle violations is obtained, and the model is verified through actual intersections.

[0149] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

[0150] The meaning of "and / or" described in this application means that the situations where each exists alone or both exist at the same time are included.

[0151] The term “connection” as used in this application may mean a direct connection between components or an indirect connection between components via other components.

[0152] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. The correction coefficient algorithm of the capacity model of special-shaped signalized intersections is characterized by: The specific steps include: Step S1: data collection and sorting, for special-shaped intersections, basic information of special-shaped intersections, non-motor vehicle violation behavior data, and motor vehicle data affected by non-motor vehicle violation behavior; An irregular intersection is one where the angle of one of the roads is less than 75°; The basic information of special-shaped intersections includes dividing lanes into small-angle left-turn lanes, large-angle left-turn lanes, small-angle right-turn lanes, and large-angle right-turn lanes according to the road angle and direction; It also includes cycle, phase, yellow light duration, green light duration and red light duration, providing various parameters required to calculate the effective green light time; The data on non-motor vehicle violation behaviors include the time when non-motor vehicles are in the right of way and the right of no way, the type of non-motor vehicle, the time when non-motor vehicles arrive, the time when non-motor vehicles violate the law, the time when non-motor vehicles stop violating the law, the distance of violation, the duration of violation, and the average saturated headway of motor vehicles; Motor vehicle data include the type of motor vehicle, the time when the motor vehicle reaches the stop line, and whether there are any non-motor vehicles violating traffic rules; Step S2: Analyze the spatiotemporal characteristics of the non-motor vehicle violation behavior data collected in step S1, wherein the spatiotemporal characteristics include time characteristics and space characteristics, that is, analyze from the two perspectives of time characteristics and space characteristics, and use the analyzed factors as possible influencing factors on the motor vehicle traffic flow, preliminarily screen these possible influencing factors, and screen out the key influencing factors that affect the saturated motor vehicle traffic flow; The temporal features include the start time of the non-motor vehicle violation and the duration ratio of the non-motor vehicle violation, and the spatial features include the relative position of the non-motor vehicle, the number of violations of the non-motor vehicle group and the violation distance; The above-mentioned temporal and spatial characteristics are used as possible influencing factors of non-motor vehicle violations on the saturated headway of motor vehicles. The binomial logistic regression model is used for factor screening, and the key influencing factors affecting the saturated traffic flow of motor vehicles include the relative position of non-motor vehicles, the number of violations of non-motor vehicle groups, the duration ratio of violations, and the distance of violations. The violation duration ratio is the ratio of the duration of the non-motor vehicle violation to the duration of the corresponding red light phase when the non-motor vehicle violates the law; Define the duration ratio of the violation as λ, the duration of the non-motor vehicle violation as t1, in seconds, and the corresponding red light phase duration when the non-motor vehicle violates the law as t2, in seconds. The calculation formula is: Step S3: Based on the motor vehicle data collected in step S1 and whether there is an impact of non-motor vehicle violations, the key influencing factors screened in step S2 are analyzed respectively to obtain the influence coefficients of different key influencing factors on the motor vehicle operation characteristics under the impact of non-motor vehicle violations; Step S4: Integrate the different influence coefficients in step S3, determine the correction coefficients of the saturated headway and acceleration loss time of motor vehicles, and establish a traffic capacity correction model for signal-controlled special-shaped intersections under the influence of non-motor vehicle violations; Step S5: verifying the established capacity correction model.

2. The correction coefficient algorithm for the capacity model of special-shaped signalized intersection according to claim 1 is characterized in that: In step S1, the headway, start loss time, acceleration time and road clearing time of two motor vehicles are calculated based on the time when the motor vehicle reaches the stop line, with the phase of the signal light as the unit; The violation distance is the shortest distance between the front end of the non-motor vehicle and the motor vehicle traffic flow; The duration of the violation is the difference between the time when the non-motor vehicle violates the law and the time when the non-motor vehicle stops violating the law; Based on the collected headway time, a statistical analysis is performed on the average headway time of each motor vehicle in a queue in each phase. If there is no significant difference in the average headway time of motor vehicles in queues at adjacent positions, it is considered that the headway time remains stable after that position, that is, the average headway time thereafter is the saturated headway time, and the position of motor vehicles when the traffic flow in different types of lanes reaches saturated traffic flow is judged based on this saturated headway time.

3. The correction coefficient algorithm for the capacity model of special-shaped signalized intersection according to claim 1 is characterized in that: In step S3, the influence of non-motor vehicle violation on motor vehicle traffic is divided into inner influence and outer influence, and the key influencing factors screened in step S2 are analyzed by K-means clustering method and data fitting method to obtain the saturated headway influence coefficient, which includes the influence coefficient of different violation duration ratios on the saturated headway and the influence coefficient of the violation distance on the saturated headway; Based on the motor vehicle data collected in step S1 with or without the influence of non-motor vehicle violations, the green light loss time influence coefficient is analyzed by using a data fitting method to obtain the motor vehicle acceleration loss time influence coefficient.

4. The correction coefficient algorithm for the capacity model of special-shaped signalized intersection according to claim 3 is characterized by: The K-means clustering method is used to classify the effects of different violation duration ratios on the saturated headway. Then, the cluster member intervals of different types of lanes are merged to obtain four ranges of violation duration ratios λ at special-shaped intersections, namely λ∈[0,0.22), λ∈[0.22,0.48), λ∈[0.48,0.75) and λ∈[0.75,1.5]. According to the saturated headway of different types of lanes without violations, the influence coefficients of the violation duration ratios of the four ranges under different types of lanes on the saturated headway are calculated respectively. The Weibull function model is used to fit the influence of the violation distance inside and outside different types of lanes on the saturated headway time. The function is expressed as: H=a-(a-b)e -(kl)c In the formula, H is the saturated headway of motor vehicle traffic, in seconds; l is the violation distance, in meters; a, b, c, and k are all parameters; The K-means clustering method is used to classify the irregular intersections according to the violation distances of different types of lanes, and four levels are obtained, namely l∈[0,1.3); l∈[1.3,2.2); l∈[2.2,3.2); l∈[3.2,4.5]. The influence coefficients of the saturated headway of different types of lanes in different ranges are calculated respectively. A linear model is used to fit the data of the motor vehicle start time with and without the influence of non-motor vehicle violations to determine the motor vehicle start time. The formula is: T 启 =a×k+b In the formula, T 启 is the starting time of the queuing vehicles, in seconds; k is the position of the queued vehicle in the queue; a and b are parameters; The Gompertz model is used to fit the motor vehicle acceleration process to obtain the acceleration time of motor vehicles in different types of lanes under different combinations of special-shaped intersections and violations. The formula is: In the formula, V(t) is the speed of the motor vehicle after it starts, in m / s; t is the time series after the motor vehicle starts, in s; a, b, c are parameters; The time when the last motor vehicle passes the stop line minus the time when the red light turns on during the traffic phase is collected under the influence of non-motor vehicle violations to obtain the road clearing loss time. The sample expected value is used as the road clearing loss time. The formula is: t 清 =E(t a -t b ) Where: t 清 is the road clearing loss time, unit is s; E is the expected value of the sample; t a The time when the red light turns on during the motor vehicle traffic phase, in seconds; t b It is the time when the last motor vehicle passes the stop line, in seconds.

5. The correction coefficient algorithm for the capacity model of special-shaped signalized intersection according to claim 1 is characterized by: In the actual intersection, when motor vehicles start accelerating, it takes a certain amount of time to reach the saturated flow state. Motor vehicles start accelerating during the yellow light time, so the yellow light time is also a component of the effective green light time. At the same time, due to the influence of non-motor vehicle violations, the saturated headway and acceleration loss time of motor vehicles should also be corrected. Finally, in step S4, the established traffic capacity correction model is: In the formula, C i is the traffic capacity of a lane, in pcu / h; n is t 加 The number of vehicles passing the stop line within a certain time period; T is the cycle length, A is the yellow light time, t g is the green light time, t 启 is the startup loss time, t 加 is the acceleration loss time, t 清 is the road clearing loss time, h is the saturated headway, both in seconds; is the acceleration loss time correction factor; is the saturated headway correction factor.

6. The correction coefficient algorithm for the capacity model of special-shaped signalized intersection according to claim 5 is characterized by: The correction coefficient of the motor vehicle saturated headway is as follows: In the formula, is the saturated headway correction coefficient; f T is the influence coefficient of different violation duration ratios on the saturated headway of motor vehicles; f L is the influence coefficient of different violation distance ratios on the saturated headway of motor vehicles; Acceleration loss time correction factor, the formula is as follows: In the formula, is the acceleration loss time correction factor; t 加有 is the acceleration loss time under different combinations of non-motor vehicle violation behaviors; t 加无 is the acceleration loss time when there is no influence of non-motor vehicle violation. The range of acceleration loss time when there is no influence of non-motor vehicle violation is λ∈[0,0.22) and l∈[3.2,4.5].

7. The correction coefficient algorithm for the capacity model of special-shaped signalized intersection according to claim 6 is characterized by: In step S5, the verification of the capacity correction model includes verifying the error between the actual measured capacity and the corrected capacity and verifying the error between the actual measured capacity and the uncorrected capacity.