Driver ecological driving behavior evaluation method and system based on driving habits

Abstract

The invention relates to the technical field of ecological driving behaviors, and discloses a driver ecological driving behavior evaluation method based on driving habits, which comprises the following steps: establishing a generalized ecological driving behavior evaluation model and determining effective knowledge parameters of the generalized ecological driving behavior evaluation model; acquiring travel data of a driver within a set time, obtaining a driving route, time and specific driving behavior data of the driver, acquiring a vehicle maintenance record within one year, and defining a driving event according to the vehicle maintenance record; threshold values of part of evaluation parameters in the driving event are determined; standardizing the driving event frequency, and converting the driving event frequency into a language Z number through fuzzy scale based on an expert system; inputting the language Z number into a generalized ecological driving behavior evaluation model, and obtaining a truth value of a final target library through efficient batch iteration; and calculating final evaluation data of each driver by applying a Z-number score function so as to feed back the generalized driving behavior of each driver.

Description

Technical Field

[0001] This invention relates to the field of ecological driving behavior technology, and in particular to a method and system for assessing driver ecological driving behavior based on driving habits. Background Technology

[0002] Energy consumption and air pollution caused by road traffic development have become a focus of research in the field of transportation science. Eco-driving, due to its rapid and low-cost environmental benefits, has attracted widespread attention from scholars and has become one of the important measures for energy conservation and emission reduction in road traffic. Currently, a series of studies have explored strategies and training methods for eco-driving behavior. However, according to existing conclusions, unless drivers personally experience the direct benefits of eco-driving behavior, it will be difficult to maintain this behavior in the long term. The fundamental reason is the lack of a comprehensive assessment and feedback mechanism for eco-driving behavior. Therefore, accurately assessing drivers' eco-driving capabilities is imperative. Currently, research on multi-attribute decision-making methods has some shortcomings: 1) the complexity of the decision problem itself and the ambiguity and uncertainty of decision information; 2) excessive subjectivity; and 3) the limitations of single decision-making methods.

[0003] In fact, beyond specific eco-driving operations, eco-driving behavior in a broader sense also includes factors dependent on the driver's subjective awareness, such as vehicle purchase and maintenance, and travel decisions. This is precisely where eco-driving promotion and training need improvement. However, traditional eco-driving theoretical research mostly focuses on a single vehicle in a single trip, and the parameters used to evaluate eco-driving ability are mainly concentrated on emissions and fuel consumption, neglecting the coupling effect of external factors in the driver's actual driving operation, which may lead to abnormal results. Furthermore, even if existing technologies consider a comprehensive environmental assessment framework, factors such as the inaccuracy of knowledge and information, the limitations of acquiring uncertain knowledge, and the lack of research on fuzzy reasoning relationships that can assess the interactions between various factors also increase the challenges of assessment. In summary, current research and technology mainly assess driving behavior on specific trips, lacking assessment of individual driver habits, and technically lacking comprehensive and reasonable modeling of various assessment factors and their coupling relationships. Summary of the Invention

[0004] This invention provides a method and system for assessing driver ecological driving behavior based on driving habits, in order to solve the problem that existing assessment methods cannot provide accurate assessment results of the driver's driving process.

[0005] To achieve the above objectives, the present invention employs the following technical solution: In a first aspect, the present invention provides a method for assessing driver ecological driving behavior based on driving habits, comprising: Establish a generalized ecological driving behavior assessment model and determine the effective knowledge parameters of the generalized ecological driving behavior assessment model; Collect drivers' travel data within a set time period to obtain their driving routes, times, and specific driving behavior data, and collect their vehicle maintenance records for one year, and define driving events accordingly; Determine the threshold values ​​for some evaluation parameters in driving events; Standardize the frequency of driving events and convert them into linguistic Z-numbers using fuzzy scaling based on an expert system; The language Z-number is input into the generalized ecological driving behavior assessment model, and the true value of the final target library is obtained through efficient batch iteration. The Z-number scoring function is applied to calculate the final assessment data of each driver to provide feedback on the generalized driving behavior of each driver.

[0006] Optionally, a generalized ecological driving behavior assessment model can be constructed using Petri nets.

[0007] Optionally, when constructing a generalized ecological driving behavior assessment model using Petri nets, the knowledge parameters of this Z-number Petri net are defined as follows: ; Where P, T, I, O represent the structure of the entire Petri net; : is a finite set of libraries represented by {p1, p2, ..., pn}; : is a finite set of transitions represented by {t1, t2, ..., tm}; is n The m-input relation matrix defines the input flow relation from place to transition; is n The m-output relation matrix defines the input flow relation from transition to place; This refers to the finite set of driving behavior events represented by the corresponding database. Represents an identifier vector , in, It is the truth value of the corresponding place, expressed in terms of the language Z-number. The initial identifier is recorded as , indicating the probability of the driving behavior occurring; Represents a confidence vector . in, It is the confidence level of the corresponding strain transition rule expressed in terms of the language Z-number; Represents a threshold vector . in, It is the occurrence threshold of the corresponding location, expressed in terms of the language Z-number; Similarly, this represents a threshold vector. . in, This is the threshold for the occurrence of the corresponding strain transition, expressed in terms of the Z-number in language; Represents the global weight, where, This corresponds to change, reflecting the relative importance of a change to its subsequent storage locations; Represents the local weights, where, The corresponding term is "warehouse," which reflects the relative importance of a warehouse to its subsequent changes. Describe an indirect action matrix . in, Used to indicate the intensity of indirect influence between driving behavior events; Describes an indirect action coefficient matrix If the corresponding indirect effect is positive, increasing the impact of driving behavior on the driving score, then kij = 1; if the corresponding indirect effect is negative, decreasing the impact of driving behavior on the driving score, then kij = -1; otherwise, kij = 0. For example, if we analyze the indirect effect between rapid deceleration and frequent acceleration, the corresponding kij = 1, indicating that rapid deceleration will enhance the negative impact of frequent acceleration. If we analyze the indirect effect between coasting in neutral and frequent use of air conditioning, the corresponding kij = -1, indicating that coasting in neutral may reduce the score for frequent use of air conditioning. By inputting the language Z-values ​​for each event, the generalized driving behavior assessment score of the driver can be obtained. By defuzzifying the language Z-values, a value of [0, 1] can be obtained to achieve the evaluation.

[0008] Optionally, the definition and related conversion formulas for language Z-numbers are as follows: set up Let be a finite ordered set of linguistic terms, where Representing the potential value of linguistic terms, It is a non-negative integer. and , representing the minimum and maximum values ​​of the language terms, respectively; Language terminology collection If the following requirements are met: ,but To fully preserve the original linguistic information, the discrete language terminology set is expanded into a continuous form: ; in, If is a sufficiently large positive integer, , Then it is called a primitive language term; otherwise it is considered a virtual language term. Fuzzy membership function: Let A collection of linguistic terms, If the values ​​range from 0 to 1, then the fuzzy membership function is a function that ranges from... arrive The mapping is defined as: ; in, , It describes the preferences of decision-makers when choosing an UI; To represent the semantic values ​​of linguistic terms, the following fuzzy membership function was used: ; Let X be a set of utterances. and Let X be two non-empty sets of linguistic terms. Then the set of linguistic Z-numbers in X is defined as follows: ; in, Fuzzy constraints representing the information contained in an uncertain variable x; And indicated The credibility of the language Z-number. The simplified form of the language Z-number is... ,in and These are two non-empty linguistic terms; For the Z-number of two languages and ,if and These are two language scaling functions, each with its own inverse function. and The basic operations of the Z-number in the language can be expressed as follows: ; ; ; ; For a language Z-number ,if and These are two non-empty linguistic terms. and Then, it can be blurred into a clear value using the following method: ; in, , .

[0009] Optionally, the two truth-computation operators involved in the language Z-number Petri net are as follows: ; .

[0010] Optionally, when collecting travel data for a driver within a set time period, real-time data acquisition can be performed using the vehicle's OBD interface and GPS module.

[0011] Optionally, defining driving events includes: Define a sudden deceleration driving event as p 1; Define a rapid acceleration driving event as p 2; Frequent acceleration driving events are defined as p 3; Frequent deceleration driving events are defined as p 4; Define travel as a driving event with poor weather conditions. p 5; Define intense driving operations as driving events p 6; Define FVR as an event that is less than the average driving event. p 7; Define driving events that frequently switch driving modes as p 8; Define frequent driving speed change driving events as p 9; Define frequent use of vehicle air conditioning as a driving event. p 10 ; Define fewer congestion-avoiding driving events as p 11 ; Poor driving habits are defined as driving events. p 12 ; Define driving events with low vehicle maintenance frequency as p 13 ; Define a driving event with a large load in the vehicle's trunk as p 14 ; Define a lower generalized ecological driving behavior score driving event as p 15 .

[0012] Optionally, the method further includes: a step of determining a threshold for a portion of the evaluation parameters, the step comprising: Define the threshold values ​​for acceleration parameters during rapid acceleration and deceleration: Set the acceleration of the vehicle during acceleration and deceleration to ±2 m / s², respectively, as the standard threshold values ​​for rapid acceleration and deceleration. Severe weather thresholds are defined based on road visibility: visibility below 100m; Define the maximum load threshold for the trunk: exceeding 60% of the maximum load capacity of the trunk.

[0013] Optionally, the frequency of driving events is standardized and converted into a linguistic Z-number using fuzzy scaling based on an expert system, specifically: For rapid acceleration events, the percentage of rapid acceleration periods in the total acceleration period for each trip is statistically analyzed, and a conversion analysis is performed. For sudden deceleration events, the proportion of sudden deceleration periods in the total deceleration periods for each trip is statistically analyzed and converted. For frequent acceleration behavior, the proportion of acceleration time to total driving time in each trip was statistically analyzed. For frequent deceleration, the proportion of deceleration time to total driving time in each trip was statistically analyzed; For severe weather events, statistics are compiled on the proportion of trips taken during severe weather to the total number of trips. In response to frequent coasting incidents, the proportion of coasting time in neutral to the total driving time in each trip was statistically analyzed. To address the frequent switching of driving modes, the number of times the driving mode was switched during each trip was counted. In response to the frequent use of vehicle air conditioning, statistics were compiled on the proportion of air conditioning usage time to total driving time. For less avoidable traffic congestion, traffic flow data for travel routes is collected. For vehicles with low maintenance frequency, the vehicle maintenance frequency over a year is statistically analyzed; For vehicles with large trunk loads, the frequency of trips with heavy trunk loads was statistically analyzed over a year.

[0014] Secondly, embodiments of this application provide a driver ecological driving behavior assessment system based on driving habits, the system comprising: The evaluation model building module is configured to: build a generalized ecological driving behavior evaluation model and determine the effective knowledge parameters of the generalized ecological driving behavior evaluation model; The data acquisition module is configured to: collect the driver's travel data within a set time period, including driving route, time and specific driving behavior data, and collect the vehicle maintenance records within one year, and define driving events accordingly; The data processing module is configured to: determine the thresholds of some evaluation parameters in driving events, standardize the frequency of driving events, and convert them into linguistic Z-numbers based on an expert system using fuzzy scaling; The evaluation module is configured to: input the language Z-number into the generalized ecological driving behavior evaluation model, obtain the true value of the final target library through efficient batch iteration; and apply the Z-number scoring function to calculate the final evaluation data of each driver to provide feedback on the generalized driving behavior of each driver.

[0015] Beneficial effects: The present invention provides a driver ecological driving behavior assessment method based on driving habits. It utilizes Petri nets as knowledge acquisition and reasoning tools, and establishes an assessment model in conjunction with an expert knowledge system. It fully considers subjective and objective factors in the real world to solve the problems of the limitations of indicators in traditional assessment methods and the neglect of potential interrelationships between indicators. It proposes a language Z-number Petri net to score the generalized driving behavior of each driver and provide reasonable feedback. Attached Figure Description

[0016] Figure 1 A flowchart of a driver ecological driving behavior assessment method according to a preferred embodiment of the present invention; Figure 2 This is a schematic diagram of the ILZPNs for a generalized eco-driving assessment according to a preferred embodiment of the present invention; Figure 3 A schematic diagram of a generalized eco-driving evaluation index system provided in a preferred embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of the ecological driving behavior assessment system provided in a preferred embodiment of the present invention; Figure 5 The flowchart of the eco-driving behavior assessment method provided in a preferred embodiment of the present invention is shown. Detailed Implementation

[0017] The technical solution of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, the terms "an" or "a" and similar terms do not indicate a quantity limitation, but rather indicate the presence of at least one. The terms "connected" or "linked" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up," "down," "left," "right," etc., are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship also changes accordingly.

[0019] Example 1 This invention first discloses a driver-based method for assessing ecological driving behavior, see [link to relevant documentation]. Figure 1 , 5 This includes the following steps: Step 1: Collect the driver's travel data for one week through the vehicle's OBD interface and GPS module to obtain their driving route, time and specific driving behavior data. Some driving data samples are shown in Table 1. Collect the vehicle maintenance records for one year.

[0020] Table 1. Driving Data Sample

[0021] Step 2: Define the acceleration thresholds for rapid acceleration and deceleration: Set the acceleration of the vehicle during acceleration and deceleration to ±2 m / s², respectively, as the standard thresholds for rapid acceleration and deceleration. Severe weather thresholds are defined based on road visibility: visibility below 100m; Define the maximum load threshold for the trunk: exceeding 60% of the maximum load capacity of the trunk.

[0022] Step 3: Define driving events and standardize event frequencies. Based on the determined evaluation parameter thresholds, convert fuzzy scaling into linguistic Z-numbers using an expert system for evaluation.

[0023] Regarding rapid acceleration events, the percentage of rapid acceleration periods in the total acceleration period for each trip is statistically analyzed, and a conversion analysis is performed. Regarding sudden deceleration events, the percentage of sudden deceleration periods in the total deceleration periods for each trip was statistically analyzed, and a conversion analysis was performed. Regarding frequent acceleration behavior, the proportion of acceleration time to total driving time in each trip was statistically analyzed. Regarding frequent deceleration, statistics were compiled on the proportion of deceleration time to total travel time in each trip. Regarding severe weather events, statistics are compiled on the proportion of trips taken during severe weather to the total number of trips. Regarding frequent coasting in neutral, the proportion of coasting time in neutral to the total driving time in each trip was statistically analyzed. Regarding the frequent switching of driving modes, the number of times the driving mode was switched during each trip was counted. Regarding the frequent use of vehicle air conditioning, statistics were compiled on the proportion of air conditioning usage time to total driving time. Regarding less avoidable traffic congestion, statistics on traffic flow along travel routes are compiled. Regarding the low vehicle maintenance frequency, statistics on vehicle maintenance frequency over a year are provided. Regarding the large load capacity of vehicle trunks, statistics are compiled on the frequency of trips with large loads in the vehicle trunk within one year.

[0024] Step 4: Based on the expert knowledge system, the knowledge parameters of the Petri net are defined through a set of linguistic terms, and a fuzzy scale is established based on this, as shown in Table 2 below: ; .

[0025] Table 2 Fuzzy Scaling Table

[0026] Step 5: Define the knowledge parameters of this Z-number Petri net as follows: ; The knowledge parameters of ILZPNs were obtained through a three-person expert panel discussion and scoring, as shown below: ; ; ; ; ; ; ; Step 5: Assess the generalized eco-driving behavior. The initial authenticity is scored by another expert based on the actual situation, as follows: , , , , , , , , , , .

[0027] Based on the simplified algorithm above, the truth value of p10 has not reached the threshold, and for transition t1, its input truth value has not reached the threshold. Therefore, the model can be further simplified.

[0028] The initial label vector for this simplified ILZPNs can be listed as follows:

[0029] Then we get: ; Step 6: The generalized ecological driving behavior assessment level is converted into the truth value of the termination position p15.

[0030] In the established generalized ecological driving behavior evaluation model, the logical relationships between the influencing factors of vehicle emissions can be represented by the transitional and positional relationships of ILZPN. Assume dj ( ) represents all the propositions constructed in this paper for the evaluation of eco-driving behavior, such as Figure 2 As shown. Therefore, the generation rules ILZPRs in the ecological driving behavior evaluation process are represented as follows: The entire fuzzy inference process can be viewed as a structural model of Petri nets, i.e.: The expression IF d1 AND d2 THEN d6 (λ1, λ2; lw1, lw2; s1; k1; τ; μ; gw) represents the inference relationship between driving behavior events. If driving behavior d1 occurs, then d2 will be inferred to occur. : IF d3 AND d4 THEN d9 (λ1, λ2; lw1, lw2; s1, s2; k1, k2; τ; μ; gw) : IF d6 AND d7 AND d8 AND d9 THEN d12 (λ1, λ2, λ3, λ4; lw1, lw2, lw3, lw4; τ; μ; gw) : IF d10 THEN d15 (λ; lw; τ; μ; gw) : IF d11 THEN d15 (λ; lw; τ; μ; gw) : IF d12 THEN d15 (λ; lw; τ; μ; gw) : IF d13 THEN d15 (λ; lw; τ; μ; gw) : IF d14 THEN d15 (λ; lw; s1, s2; k1, k2; τ; μ; gw).

[0031] The libraries corresponding to the propositions in the above ILZPNs are shown in Table 3. The generalized ecological driving behavior assessment level can be converted into the truth value of the termination position p15.

[0032] Table 3 Driving Event Table

[0033] Step 7: Perform reasoning based on fuzzy knowledge algorithm Make r=1.

[0034] Triggering place vector We can obtain the following: .

[0035] The truth value vector of the input library : ; Triggering place vector : ; Output truth value vector : ; New label vector M1:

[0036] if Then set r=3 and continue running the program:

[0037] if Then set r=3 and continue running the program:

[0038] if Then let r=4 and continue the iteration:

[0039] The iteration ended, and the final result is as follows:

[0040] The final p15 score represents the comprehensive assessment result of driving behavior, with the truth value at p15 being [value missing]. This means that emissions are lower and drivers' driving behavior is more in line with environmentally friendly driving requirements.

[0041] To illustrate the proposed ILZPN inference algorithm, the basic matrix operations are introduced below: Addition Operator :

[0042] in: , , and

[0043] Multiplication Operator :

[0044] in: , , and

[0045] Comparison Operator :

[0046] in: , , and

[0047] The following reasoning algorithm is based on the following knowledge: Input: , , and , ; as well as ; , ,as well as .

[0048] Output: This represents the result of each iteration.

[0049] Step 1: Record the number of loops Let r = 1, where r is the number of iterations; Step 2: Identify possible changes (1) Calculate the possible place vectors. as follows: ; (2) If If the vector contains all zero elements, proceed to Step 5; otherwise, compute the element representing the truth value of the input, as follows: ; in, That is, it includes the operator of ILZWA, and this local weight vector It will be used; otherwise, g jumps to Step 7.

[0050] (3) Then, the possible transition vector F(r) is calculated as follows: ; in and if If it is a non-zero vector, proceed to Step 3; otherwise, jump to Step 5.

[0051] Step 3: Calculate the output of the transition. Calculate the truth value and output the result. ,as follows: ; in, It is a truth-based operator. Step 4: Update the identifiers of all warehouses. ; in .

[0052] Step 5: If If the condition is met, the reasoning process ends immediately; otherwise, set r = r + 1 and continue to Step 2.

[0053] In addition, a novel Petri net model simplification method was established. This method performs a preliminary judgment on the obtained fuzzy knowledge information matrix to filter out untriggerable libraries, thus avoiding "state explosion." The model simplification method is as follows: (1) Remove transitions that cannot occur, i.e., ti satisfies the following conditions: ; (2) Removing the place before the transition ti removed in (1) can be represented as a set for i=1, 2, …, If po is a subsequent location of another transition, retain the location; otherwise, delete the location.

[0054] (3) By checking all the transitions of the established ILZPNs again through (1) and (2) above, a simplified ILZPNs is finally obtained to improve the efficiency of knowledge reasoning.

[0055] In this embodiment, see Figure 5 As shown, the method includes the following steps: defining indirect effects that comprehensively consider both subjective human factors and objective potential influencing factors; constructing a generalized ecological driving behavior evaluation index system, and creating Indirect-effect-incorporated Petri Nets (ILZPNs) with full consideration of indirect effects to visualize the evaluation process; determining the correspondence between evaluation level linguistic variables and fuzzy scales, and using an expert system to capture uncertainty information in the evaluation information based on linguistic Z-numbers to determine the knowledge parameters of ILZPNs; avoiding the "state explosion" problem caused by structural complexity by using a proposed novel Petri net structure simplification method, thereby improving the efficiency of knowledge reasoning; standardizing the knowledge parameters of ILZPNs into a matrix, and using the designed Indirect-effect-incorporated linguistic Z-Number weighted average (ILZWA) operator to transform the indirect effects between repositories into input influences on subsequent changes, thereby realizing the reasoning process of indirect effect information. Through efficient batch iteration, the true value of the final target library was obtained; the Z-number scoring function was applied to calculate the final score of each driver, so as to score the generalized driving behavior of each driver and provide reasonable feedback.

[0056] like Figure 4 As shown in the figure, this application provides a driver ecological driving behavior assessment system based on driving habits, the system comprising: The evaluation model building module is configured to: build a generalized ecological driving behavior evaluation model and determine the effective knowledge parameters of the generalized ecological driving behavior evaluation model; The data acquisition module is configured to: collect the driver's travel data within a set time period, including driving route, time and specific driving behavior data, and collect the vehicle maintenance records within one year, and define driving events accordingly; The data processing module is configured to: determine the thresholds of some evaluation parameters in driving events, standardize the frequency of driving events, and convert them into linguistic Z-numbers based on an expert system using fuzzy scaling; The evaluation module is configured to: input the language Z-number into the generalized ecological driving behavior evaluation model, obtain the true value of the final target library through efficient batch iteration; and apply the Z-number scoring function to calculate the final evaluation data for each driver, providing feedback on each driver's generalized driving behavior.

[0057] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for assessing driver ecological driving behavior based on driving habits, characterized in that, include: Establish a generalized ecological driving behavior assessment model and determine the effective knowledge parameters of the generalized ecological driving behavior assessment model; Collect drivers' travel data within a set time period to obtain their driving routes, times, and specific driving behavior data, and collect their vehicle maintenance records for one year, and define driving events accordingly; Determine the threshold values ​​for some evaluation parameters in driving events; Standardize the frequency of driving events and convert them into linguistic Z-numbers using fuzzy scaling based on an expert system; The language Z-number is input into the generalized ecological driving behavior assessment model, and the true value of the final target library is obtained through efficient batch iteration. The Z-number scoring function is applied to calculate the final evaluation data for each driver, providing feedback on each driver's generalized driving behavior.

2. The driver ecological driving behavior assessment method based on driving habits according to claim 1, characterized in that, A generalized ecological driving behavior assessment model was constructed using Petri nets.

3. The driver ecological driving behavior assessment method based on driving habits according to claim 2, characterized in that, When constructing a generalized ecological driving behavior assessment model using Petri nets, the knowledge parameters of the Z-number Petri net are defined as follows: ； Where P, T, I, O represent the structure of the entire Petri net; : is a finite set of libraries represented by {p1, p2, ..., pn}; : is a finite set of transitions represented by {t1, t2, ..., tm}; is n The m-input relation matrix defines the input flow relation from place to transition; is n The m-output relation matrix defines the input flow relation from transition to place; This refers to the finite set of driving behavior events represented by the corresponding database. Represents an identifier vector , in, It is the truth value of the corresponding place, expressed in terms of the language Z-number. The initial identifier is recorded as , indicating the probability of the driving behavior occurring; Represents a confidence vector . in, It is the confidence level of the corresponding strain transition rule expressed in terms of the language Z-number; Represents a threshold vector . in, It is the occurrence threshold of the corresponding location, expressed in terms of the language Z-number; Similarly, this represents a threshold vector. . in, This is the threshold for the occurrence of the corresponding strain transition, expressed in terms of the Z-number in language; Represents the global weight, where, This corresponds to change, reflecting the relative importance of a change to its subsequent storage locations; Represents the local weights, where, The corresponding term is "warehouse," which reflects the relative importance of a warehouse to its subsequent changes. Describe an indirect action matrix . in, Used to indicate the intensity of indirect influence between driving behavior events; Describes an indirect action coefficient matrix If the corresponding indirect effect is positive, increasing the impact of driving behavior on the driving score, then kij = 1; if the corresponding indirect effect is negative, decreasing the impact of driving behavior on the driving score, then kij = -1; otherwise, kij = 0. For example, if we analyze the indirect effect between rapid deceleration and frequent acceleration, the corresponding kij = 1, indicating that rapid deceleration will enhance the negative impact of frequent acceleration. If we analyze the indirect effect between coasting in neutral and frequent use of air conditioning, the corresponding kij = -1, indicating that coasting in neutral may reduce the score for frequent use of air conditioning. By inputting the language Z-values ​​for each event, the generalized driving behavior assessment score of the driver can be obtained. By defuzzifying the language Z-values, a value of [0, 1] can be obtained to achieve the evaluation.

4. The driver ecological driving behavior assessment method based on driving habits according to claim 3, characterized in that, The definition and related conversion formulas of the language Z-number are as follows: set up Let be a finite ordered set of linguistic terms, where Representing the potential value of linguistic terms, It is a non-negative integer. and , representing the minimum and maximum values ​​of the language terms, respectively; Language terminology collection If the following requirements are met: ,but To fully preserve the original linguistic information, the discrete language terminology set is expanded into a continuous form: ； in, If is a sufficiently large positive integer, , Then it is called a primitive language term; otherwise it is considered a virtual language term. Fuzzy membership function: Let A collection of linguistic terms, If the values ​​range from 0 to 1, then the fuzzy membership function is a function that ranges from... arrive The mapping is defined as: ； in, , It describes the preferences of decision-makers when choosing an UI; To represent the semantic values ​​of linguistic terms, the following fuzzy membership function was used: ； Let X be a set of utterances. and Let X be two non-empty sets of linguistic terms. Then the set of linguistic Z-numbers in X is defined as follows: ； in, Fuzzy constraints representing the information contained in an uncertain variable x; And indicated The credibility of the language Z-number. The simplified form of the language Z-number is... ,in and These are two non-empty linguistic terms; For the Z-number of two languages and ,if and These are two language scaling functions, each with its own inverse function. and The basic operations of the Z-number in the language can be expressed as follows: ； ； ； ； For a language Z-number ,if and These are two non-empty linguistic terms. and Then, it can be blurred into a clear value using the following method: ； in, , .

5. The driver ecological driving behavior assessment method based on driving habits according to claim 4, characterized in that, The two truth-value calculation operators involved in the language Z-number Petri net are as follows: ； 。 6. The driver ecological driving behavior assessment method based on driving habits according to claim 1, characterized in that, When collecting travel data from drivers within a set time period, real-time data acquisition is performed using the vehicle's OBD interface and GPS module.

7. The driver ecological driving behavior assessment method based on driving habits according to claim 1, characterized in that, Defining driving events includes: Define a sudden deceleration driving event as p 1; Define a rapid acceleration driving event as p 2; Frequent acceleration driving events are defined as p 3; Frequent deceleration driving events are defined as p 4; Define travel as a driving event with poor weather conditions. p 5; Define intense driving operations as driving events p 6; Define FVR as an event that is less than the average driving event. p 7; Define driving events that frequently switch driving modes as p 8; Define frequent driving speed change driving events as p 9; Define frequent use of vehicle air conditioning as a driving event. p 10 ; Define fewer congestion-avoiding driving events as p 11 ; Poor driving habits are defined as driving events. p 12 ; Define driving events with low vehicle maintenance frequency as p 13 ; Define a driving event with a large load in the vehicle's trunk as p 14 ; Define a lower generalized ecological driving behavior score driving event as p 15 .

8. The method for assessing driver ecological driving behavior based on driving habits according to claim 1, characterized in that, The method further includes a step of determining threshold values ​​for some evaluation parameters, which includes: Define the threshold values ​​for acceleration parameters during rapid acceleration and deceleration: Set the acceleration of the vehicle during acceleration and deceleration to ±2 m / s², respectively, as the standard threshold values ​​for rapid acceleration and deceleration. Severe weather thresholds are defined based on road visibility: visibility below 100m; Define the maximum load threshold for the trunk: exceeding 60% of the maximum load capacity of the trunk.

9. The driver ecological driving behavior assessment method based on driving habits according to claim 1, characterized in that, Standardize driving event frequencies and convert them into linguistic Z-numbers using fuzzy scaling based on an expert system, specifically: For rapid acceleration events, the percentage of rapid acceleration periods in the total acceleration period for each trip is statistically analyzed, and a conversion analysis is performed. For sudden deceleration events, the proportion of sudden deceleration periods in the total deceleration periods for each trip is statistically analyzed and converted. For frequent acceleration behavior, the proportion of acceleration time to total driving time in each trip was statistically analyzed. For frequent deceleration, the proportion of deceleration time to total driving time in each trip was statistically analyzed; For severe weather events, statistics are compiled on the proportion of trips taken during severe weather to the total number of trips. In response to frequent coasting incidents, the proportion of coasting time in neutral to the total driving time in each trip was statistically analyzed. To address the frequent switching of driving modes, the number of times the driving mode was switched during each trip was counted. In response to the frequent use of vehicle air conditioning, statistics were compiled on the proportion of air conditioning usage time to total driving time. For less avoidable traffic congestion, traffic flow data for travel routes is collected. For vehicles with low maintenance frequency, the vehicle maintenance frequency over a year is statistically analyzed; For vehicles with large trunk loads, the frequency of trips with heavy trunk loads was statistically analyzed over a year.

10. A driver ecological driving behavior assessment system based on driving habits, characterized in that, The system includes: The evaluation model building module is configured to: build a generalized ecological driving behavior evaluation model and determine the effective knowledge parameters of the generalized ecological driving behavior evaluation model; The data acquisition module is configured to: collect the driver's travel data within a set time period, including driving route, time and specific driving behavior data, and collect the vehicle maintenance records within one year, and define driving events accordingly; The data processing module is configured to: determine the thresholds of some evaluation parameters in driving events, standardize the frequency of driving events, and convert them into linguistic Z-numbers based on an expert system using fuzzy scaling; The evaluation module is configured to: input the language Z-number into the generalized ecological driving behavior evaluation model, obtain the true value of the final target library through efficient batch iteration; and apply the Z-number scoring function to calculate the final evaluation data of each driver to provide feedback on the generalized driving behavior of each driver.

Images