CN111126819B - Intelligent analysis method for urban driving condition - Google Patents
Intelligent analysis method for urban driving condition Download PDFInfo
- Publication number
- CN111126819B CN111126819B CN201911312807.8A CN201911312807A CN111126819B CN 111126819 B CN111126819 B CN 111126819B CN 201911312807 A CN201911312807 A CN 201911312807A CN 111126819 B CN111126819 B CN 111126819B
- Authority
- CN
- China
- Prior art keywords
- short
- speed
- short stroke
- automobile
- stroke
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 37
- 239000012634 fragment Substances 0.000 claims abstract description 42
- 238000000034 method Methods 0.000 claims abstract description 35
- 230000001133 acceleration Effects 0.000 claims abstract description 31
- 238000003064 k means clustering Methods 0.000 claims abstract description 14
- 230000009467 reduction Effects 0.000 claims abstract description 5
- 238000012545 processing Methods 0.000 claims description 10
- 238000000354 decomposition reaction Methods 0.000 claims description 5
- 238000012847 principal component analysis method Methods 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000013519 translation Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/80—Technologies aiming to reduce greenhouse gasses emissions common to all road transportation technologies
- Y02T10/84—Data processing systems or methods, management, administration
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Primary Health Care (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Combined Controls Of Internal Combustion Engines (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses an intelligent analysis method for urban driving conditions. The method comprises the following steps: carrying out wavelet analysis noise reduction on the collected automobile running time and speed data, and dividing the data into short-stroke segments; carrying out k-means clustering on the divided short stroke fragments to divide three types of short strokes to obtain external state sequences of the short stroke fragments; dividing the interior of the short stroke segment into 3 states of acceleration, uniform speed and idling to obtain an internal state sequence of the short stroke segment; and establishing a double-layer Markov chain for the short stroke segment to obtain a double-layer Markov model, constructing an automobile driving condition curve by using the double-layer Markov model, evaluating the error of the automobile driving condition curve, and selecting the curve with the minimum error as a final automobile driving condition curve. The method reduces the error of the analysis of the running condition of the automobile and improves the accuracy of the analysis of the running condition of the automobile.
Description
Technical Field
The invention relates to the technical field of wavelet analysis and cluster analysis, in particular to an intelligent analysis method for urban driving conditions.
Background
The automobile running condition reflects the speed-time curve of the urban running vehicle, and is mainly used for vehicle performance calibration and automobile oil consumption calibration, so that whether the accuracy of the automobile running condition reflects the urban characteristics is very important. Most countries carry out optimization to calibrate the automobile by taking the working condition of NEDC (new standard European cycle test) as a reference, but the working condition of NEDC is less and less consistent with the actual condition of China along with the change of the urban road traffic condition of China. The two most main working condition characteristics of the idle ratio and the average speed of the world light vehicle test cycle (WLTC) working condition are not consistent with the actual working condition of China, so that the construction of the urban-based vehicle running working condition to reflect the local running characteristics is necessary.
The method for constructing the driving condition mainly comprises a short-stroke method, a Markov chain method and the like. The traditional short-stroke method cannot reflect the characteristics of urban driving conditions, and the experimental results are different in level and large in difference because of the random selection and generation of the simple Markov-based state segments.
Disclosure of Invention
The invention aims to provide an intelligent analysis method for urban driving conditions, which is small in error, high in accuracy and in accordance with urban road characteristics.
The technical solution for realizing the purpose of the invention is as follows: an intelligent analysis method for urban driving conditions comprises the following steps:
step 1, carrying out wavelet analysis and noise reduction on collected automobile running time and speed data, and dividing the data into short-stroke segments;
step 3, dividing the interior of the short stroke segment into 3 states of acceleration, uniform speed and idling to obtain an interior state sequence of the short stroke segment;
and 4, establishing a double-layer Markov chain for the short stroke segment to obtain a double-layer Markov model, constructing an automobile driving condition curve by using the double-layer Markov model, evaluating the error of the automobile driving condition curve, and selecting the curve with the minimum error as a final automobile driving condition curve.
Further, the wavelet analysis denoising is performed on the acquired automobile running time and speed data in the step 1, and the data are divided into short-stroke segments, which are specifically as follows:
step 1.1, collecting GPS vehicle speed data through a vehicle-mounted terminal;
step 1.2, processing partial lost data and abnormal data;
step 1.3, performing wavelet analysis denoising processing on the data, performing 4-scale decomposition on the signals by using Daubechies-3-order wavelets, and gradually performing multi-scale refinement on the signals through a telescopic translation operation to finally obtain a smooth curve which is subjected to high-frequency time refinement and low-frequency refinement and can automatically adapt to time-frequency signal analysis;
and 1.4, dividing the stroke into short stroke sections of a movement forming section of the automobile from one idling state to the next idling state.
Further, the step 2 of performing k-means clustering on the divided short stroke segments to divide three types of short strokes to obtain external state sequences of the short stroke segments specifically as follows:
step 2.1, respectively calculating each short stroke segment by taking 11 parameters of average speed, maximum speed, average running speed, average acceleration, average deceleration, idle speed time ratio, acceleration time ratio, deceleration time ratio, speed standard deviation, acceleration standard deviation and deceleration standard deviation as characteristic parameters of each short stroke segment;
2.2, reducing the dimension of the short stroke segment characteristics by adopting a principal component analysis method, and selecting the first 5 parameters with contribution rate more than 90% as new characteristic parameters;
and 2.3, carrying out k-means clustering on the characteristic parameters, and dividing the short stroke fragments into a high-speed class, a medium-speed class and a low-speed class to obtain an external state sequence of the short stroke fragments.
Further, the step 3 of dividing the inside of the short stroke segment into 3 states of acceleration, uniform speed and idle speed to obtain an internal state sequence of the short stroke segment specifically includes:
step 3.1, dividing the internal state of the short-stroke fragment, and classifying by adopting a rule-based method, wherein the classification method specifically comprises the following steps:
the method for judging the accelerated segment classification comprises the following steps:
the classification of the deceleration state is:
the other states are judged to be constant speed states;
the above formula is how to judge the state of t time, a t Acceleration of the vehicle at time t, a t+1 Acceleration of the next second at time t, a t+2 Acceleration of the next two seconds at time t;
and 3.2, obtaining the internal state sequence of the short-stroke fragment.
Further, the step 4 of establishing a double-layer markov chain for the short stroke segment to obtain a double-layer markov model, constructing an automobile driving condition curve by using the double-layer markov model, evaluating an error of the automobile driving condition curve, and selecting a curve with the minimum error as a final automobile driving condition curve, wherein the method specifically comprises the following steps:
step 4.1, constructing a Markov chain for the obtained external state sequence, and constructing a Markov chain for the internal fragment state sequence to obtain a double-layer Markov model;
step 4.2, constructing a driving condition within 2000s by adopting a randomization method to obtain an external state class sequence and an internal state fragment sequence;
4.3, selecting an internal state fragment sequence from the original data, requiring that the speed difference between fragments is not more than 0.5m/s, and constructing an automobile driving condition curve;
and 4.4, comparing and evaluating the constructed automobile running condition curve with the original data parameters, and selecting the automobile running condition curve with the average error of less than 5% as a final automobile running condition curve.
Compared with the prior art, the invention has the following remarkable advantages: (1) the data preprocessing is carried out by adopting a wavelet analysis method, so that the driving condition can better accord with the real driving condition of the urban road; (2) and the double-layer Markov model is adopted, so that the error of the analysis of the running condition of the automobile is reduced, and the accuracy of the analysis of the running condition of the automobile is improved.
Drawings
FIG. 1 is a schematic flow chart of an intelligent analysis method for urban driving conditions.
FIG. 2 is a vehicle speed graph before and after wavelet analysis denoising in an embodiment of the invention.
FIG. 3 is a general flowchart of an analysis method for driving conditions of an automobile according to an embodiment of the present invention.
FIG. 4 is a flow chart of a k-means clustering algorithm in the embodiment of the present invention.
FIG. 5 is a diagram illustrating the results of the k-means clustering algorithm in the embodiment of the present invention.
FIG. 6 is a flowchart illustrating a method for predicting a driving condition of an automobile according to an embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and the specific embodiments.
With reference to fig. 1, the invention provides an intelligent analysis method for urban driving conditions, comprising the following steps:
step 1, performing wavelet analysis and noise reduction on the collected automobile running time and speed data, and dividing the data into short-stroke segments, wherein the method specifically comprises the following steps:
step 1.1, collecting GPS vehicle speed data through a vehicle-mounted terminal;
step 1.2, processing partial lost data and abnormal data;
step 1.3, performing wavelet analysis denoising processing on the data, performing 4-scale decomposition on the signals by using Daubechies-3-order wavelets, and gradually performing multi-scale refinement on the signals through a telescopic translation operation to finally obtain a smooth curve which is subjected to high-frequency time refinement and low-frequency refinement and can automatically adapt to time-frequency signal analysis;
and 1.4, dividing the stroke into short stroke sections of a movement forming section of the automobile from one idling state to the next idling state.
step 2.1, respectively calculating each short stroke segment by taking 11 parameters of average speed, maximum speed, average running speed, average acceleration, average deceleration, idle speed time ratio, acceleration time ratio, deceleration time ratio, speed standard deviation, acceleration standard deviation and deceleration standard deviation as characteristic parameters of each short stroke segment;
2.2, reducing the dimension of the short stroke segment characteristics by adopting a principal component analysis method, and selecting the first 5 parameters with contribution rate more than 90% as new characteristic parameters;
and 2.3, carrying out k-means clustering on the characteristic parameters, and dividing the short stroke fragments into a high-speed class, a medium-speed class and a low-speed class to obtain an external state sequence of the short stroke fragments.
Step 3, dividing the interior of the short stroke segment into 3 states of acceleration, uniform speed and idling to obtain an interior state sequence of the short stroke segment, which is specifically as follows:
step 3.1, dividing the internal state of the short-stroke fragment, and classifying by adopting a rule-based method, wherein the classification method specifically comprises the following steps:
the method for judging the accelerated segment classification comprises the following steps:
the classification of the deceleration state is:
the other states are judged to be constant speed states;
the above formula is how to judge the state of the t moment, a t Acceleration of the vehicle at time t, a t+1 Acceleration of the next second at time t，a t+2 Acceleration of the next two seconds at time t
And 3.2, obtaining the internal state sequence of the short-stroke fragment.
step 4.1, constructing a Markov chain for the obtained external state sequence, and constructing a Markov chain for the internal fragment state sequence to obtain a double-layer Markov model;
4.2, constructing a driving working condition within 2000s by adopting a randomization method to obtain an external state category sequence and an internal state fragment sequence;
4.3, selecting an internal state fragment sequence from the original data, requiring that the speed difference between fragments is not more than 0.5m/s, and constructing an automobile driving condition curve;
and 4.4, comparing and evaluating the constructed automobile running condition curve with the original data parameters, and selecting the automobile running condition curve with the average error of less than 5% as a final automobile running condition curve.
Example 1
With reference to fig. 1, the present embodiment includes the following steps:
step 1, performing wavelet analysis and noise reduction on the collected automobile running time and speed data, and dividing the data into short-stroke segments, wherein the method specifically comprises the following steps:
step 1.1, collecting GPS vehicle speed data through a vehicle-mounted terminal;
step 1.2, processing partial lost data and abnormal data;
step 1.3, performing wavelet analysis denoising processing on the data, performing 4-scale decomposition on the signals by using Daubechies-3-order wavelets, and gradually performing multi-scale refinement on the signals through a telescopic translation operation to finally obtain a smooth curve which is capable of automatically adapting to time-frequency signal analysis and has high-frequency time subdivision and low-frequency subdivision, wherein the smooth curve is specifically as follows:
converting the vehicle speed signal into a one-dimensional discrete signal, and processing the signal by using discrete wavelet transform. Using Daubechies-4 order wavelet as mother wavelet psi (t), discretizing scale factor a according to power series, and under the same scale, uniformly discretizing displacement factor tau to obtain odd function of wavelet as
The constants are typically taken to be:
a 0 ＝2,τ 0 ＝1
thus, the wavelet odd function is noted as:
the discrete wavelet transform coefficients are:
wavelet analysis is carried out on the vehicle speed signal, wavelet transformation coefficients are calculated, a threshold denoising processing method is adopted, threshold processing is carried out on high-frequency coefficients obtained by wavelet decomposition, coefficients larger than a threshold are reserved, and coefficients lower than the threshold are set to be zero.
FIG. 2 is a graph showing the results.
Step 1.4, dividing the stroke into short stroke segments of a movement forming interval of the automobile from one idling state to the next idling state, wherein the short stroke segments are as follows:
the short stroke section is a movement forming interval from the beginning of an idling state to the beginning of the next idling state of the automobile, the overall speed-time curve is divided into a plurality of short stroke sections, and finally a short stroke section combination is obtained.
step 2.1, in order to describe the short stroke difference, respectively calculating each short stroke segment, wherein 11 parameters including average speed, maximum speed, average running speed, average acceleration, average deceleration, idle time ratio, acceleration time ratio, deceleration time ratio, speed standard deviation, acceleration standard deviation and deceleration standard deviation are used as characteristic parameters of each short stroke segment, and respectively calculating each short stroke segment;
2.2, reducing the dimension of the short stroke segment characteristics by adopting a principal component analysis method, and selecting the first 5 parameters with contribution rate more than 90% as new characteristic parameters;
the method adopts 11 characteristic parameters, has a plurality of variables, directly utilizes all the characteristic parameters to classify the short stroke, and has great calculation complexity. Meanwhile, partial characteristic parameters have strong correlation, for example, the average speed and the average running speed have strong positive correlation, and the characteristic parameters which are not independent of each other can cause a certain index to be considered too much or too little when the working condition is constructed. By combining the characteristics of more characteristic parameters and overlapping of the parameters, a parameter processing method capable of recombining a group of variables with more quantity and strong correlation into a group of variables with less quantity and weak correlation is needed, 11 characteristic parameter variables are recombined by adopting a principal component analysis method, and the first 5 principal components with contribution rate of more than 90% are selected as new characteristic parameters.
And 2.3, carrying out k-means clustering on the characteristic parameters, and dividing the short stroke fragments into a high-speed class, a medium-speed class and a low-speed class to obtain an external state sequence of the short stroke fragments:
and (5) performing clustering analysis on the five main components by adopting a k-means clustering method. The specific process is shown in fig. 4, the short stroke is finally divided into three categories, namely a high-speed category, a medium-speed category and a low-speed category, the clustering result and the characteristics are shown in table 1, and the clustering result is shown in fig. 5.
TABLE 1 clustering results and characteristics
According to the clustering result, the conversion probability between the short-stroke segments can be obtained, and the conversion probability is represented by a formula:
wherein N is ij Representing the frequency of the current state being i and the next state being j; p is a radical of ij Representing the probability that the current state is i and the next state is j, and l is the number of categories, the transition probability is obtained as shown in table 2:
TABLE 2 transition probabilities between short Stroke segments
Step 3, dividing the interior of the short stroke segment into 3 states of acceleration, uniform speed and idling to obtain an interior state sequence of the short stroke segment, which is specifically as follows:
step 3.1, dividing the short-stroke internal state, and classifying by adopting a rule-based method, wherein the classification method mainly comprises the following steps:
the method for judging the accelerated segment classification comprises the following steps:
the classification of the deceleration state is:
the other states are judged to be constant speed states;
the above formula is how to judge the state of t time, a t Acceleration of the vehicle at time t, a t+1 Acceleration of the next second at time t, a t+2 Acceleration of the next two seconds at time t
Step 3.2, obtaining an internal state sequence of the short-stroke fragment, obtaining the internal fragment conversion probability, and using a formula:
three types of internal state Markov models are obtained, such as Table 3, Table 4 and Table 5.
TABLE 3
TABLE 4
TABLE 5
step 4.1, constructing a Markov chain for the obtained external state sequence, and constructing a Markov chain for the internal fragment state sequence to obtain a double-layer Markov model;
4.2, constructing a driving working condition within 2000s by adopting a randomization method to obtain an external state category sequence and an internal state fragment sequence;
4.3, selecting an internal state fragment sequence from the original data, requiring that the speed difference between fragments is not more than 0.5m/s, and constructing an automobile driving condition curve;
and 4.4, comparing and evaluating the constructed automobile running condition curve with the original data parameters, and selecting the automobile running condition curve with the average error of less than 5% as a final automobile running condition curve.
The method comprises the steps of constructing driving conditions by a constructed external state Markov model and an internal state Markov model together, firstly selecting an external state a, selecting a stroke segment in the a, initially setting the stroke segment to be at a constant speed, sequentially selecting segments with the speed difference of less than 0.5m/s from an internal transfer matrix until the length requirement is met, then selecting a next external state b from the external state transfer matrix, sequentially performing the steps until the final working condition length is met, and generating 100 sections of driving conditions according to the steps.
And comparing the constructed running condition with the original data parameters, taking the average value of the relative errors as a final evaluation index, and finally selecting an automobile running condition curve with the average error of 4.53 percent as a final automobile running condition curve.
Claims (4)
1. An intelligent analysis method for urban driving conditions is characterized by comprising the following steps:
step 1, performing wavelet analysis noise reduction on collected automobile running time and speed data, and dividing the data into short-stroke segments;
step 2, performing k-means clustering on the divided short stroke fragments to divide three types of short strokes to obtain external state sequences of the short stroke fragments;
step 3, dividing the interior of the short stroke segment into 3 states of acceleration, uniform speed and idling to obtain an internal state sequence of the short stroke segment;
step 4, establishing a double-layer Markov chain for the short stroke segment to obtain a double-layer Markov model, constructing an automobile driving condition curve by using the double-layer Markov model, evaluating the error of the automobile driving condition curve, and selecting the curve with the minimum error as a final automobile driving condition curve, wherein the method specifically comprises the following steps:
step 4.1, constructing a Markov chain for the obtained external state sequence, and constructing a Markov chain for the internal fragment state sequence to obtain a double-layer Markov model;
4.2, constructing a driving working condition within 2000s by adopting a randomization method to obtain an external state category sequence and an internal state fragment sequence;
4.3, selecting an internal state fragment sequence from the original data, requiring that the speed difference between fragments is not more than 0.5m/s, and constructing an automobile driving condition curve;
and 4.4, comparing and evaluating the constructed automobile running condition curve with the original data parameters, and selecting the automobile running condition curve with the average error of less than 5% as a final automobile running condition curve.
2. The intelligent urban driving condition analysis method according to claim 1, wherein the wavelet analysis denoising is performed on the collected automobile driving time and speed data in the step 1, and the data are divided into short-stroke segments, specifically as follows:
step 1.1, collecting GPS vehicle speed data through a vehicle-mounted terminal;
step 1.2, processing partial lost data and abnormal data;
step 1.3, performing wavelet analysis denoising processing on the data, performing 4-scale decomposition on the signals by using Daubechies-3-order wavelets, and gradually performing multi-scale refinement on the signals through a telescopic translation operation to finally obtain a smooth curve which is subjected to high-frequency time refinement and low-frequency refinement and can automatically adapt to time-frequency signal analysis;
and 1.4, dividing the stroke into short stroke sections of a movement forming section of the automobile from one idling state to the next idling state.
3. The intelligent analysis method for urban driving conditions according to claim 1, wherein the step 2 performs k-means clustering on the divided short stroke segments to divide three types of short strokes to obtain external state sequences of the short stroke segments, and specifically comprises the following steps:
step 2.1, respectively calculating each short stroke segment by taking 11 parameters of average speed, maximum speed, average running speed, average acceleration, average deceleration, idle speed time ratio, acceleration time ratio, deceleration time ratio, speed standard deviation, acceleration standard deviation and deceleration standard deviation as characteristic parameters of each short stroke segment;
2.2, reducing the dimension of the short stroke segment characteristics by adopting a principal component analysis method, and selecting the first 5 parameters with contribution rate more than 90% as new characteristic parameters;
and 2.3, carrying out k-means clustering on the characteristic parameters, and dividing the short stroke fragments into a high-speed class, a medium-speed class and a low-speed class to obtain an external state sequence of the short stroke fragments.
4. The intelligent analysis method for urban driving conditions according to claim 1, wherein the short-stroke segment is divided into 3 states of acceleration, uniform speed and idling in step 3 to obtain an internal state sequence of the short-stroke segment, and the method specifically comprises the following steps:
step 3.1, dividing the internal state of the short-stroke fragment, and classifying by adopting a rule-based method, wherein the classification method specifically comprises the following steps:
the method for judging the accelerated segment classification comprises the following steps:
the classification of the deceleration state is:
the other states are judged to be constant speed states;
a t acceleration of the vehicle at time t, a t+1 Acceleration of the next second at time t, a t+2 Acceleration of the next two seconds at time t;
and 3.2, obtaining the internal state sequence of the short-stroke fragment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911312807.8A CN111126819B (en) | 2019-12-18 | 2019-12-18 | Intelligent analysis method for urban driving condition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911312807.8A CN111126819B (en) | 2019-12-18 | 2019-12-18 | Intelligent analysis method for urban driving condition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111126819A CN111126819A (en) | 2020-05-08 |
| CN111126819B true CN111126819B (en) | 2022-08-16 |
Family
ID=70499720
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911312807.8A Active CN111126819B (en) | 2019-12-18 | 2019-12-18 | Intelligent analysis method for urban driving condition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111126819B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111832225B (en) * | 2020-07-07 | 2023-01-31 | 重庆邮电大学 | Method for constructing driving condition of automobile |
| CN113076697B (en) * | 2021-04-20 | 2022-11-29 | 潍柴动力股份有限公司 | Typical driving condition construction method, related device and computer storage medium |
| CN113222385B (en) * | 2021-05-06 | 2022-08-16 | 吉林大学 | Method for constructing and evaluating driving condition of electric automobile |
| CN113469240B (en) * | 2021-06-29 | 2024-04-02 | 中国科学技术大学 | Driving condition construction method based on shape similarity and storage medium |
| CN115221669B (en) * | 2021-08-05 | 2023-07-07 | 广州汽车集团股份有限公司 | Method for generating simulated driving cycle, computer device and storage medium |
| CN113744530B (en) * | 2021-09-08 | 2022-12-02 | 河南科技大学 | Construction method of vehicle working condition |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018064931A1 (en) * | 2016-10-08 | 2018-04-12 | 大连理工大学 | Method for estimating travel time distribution of taxi on urban roads when operating states of taxis are considered |
| CN108198425A (en) * | 2018-02-10 | 2018-06-22 | 长安大学 | A kind of construction method of Electric Vehicles Driving Cycle |
-
2019
- 2019-12-18 CN CN201911312807.8A patent/CN111126819B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018064931A1 (en) * | 2016-10-08 | 2018-04-12 | 大连理工大学 | Method for estimating travel time distribution of taxi on urban roads when operating states of taxis are considered |
| CN108198425A (en) * | 2018-02-10 | 2018-06-22 | 长安大学 | A kind of construction method of Electric Vehicles Driving Cycle |
Non-Patent Citations (1)
| Title |
|---|
| 聚类和马尔科夫方法结合的城市汽车行驶工况构建;姜平等;《中国机械工程》;20101210(第23期);第2893-2897页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111126819A (en) | 2020-05-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111126819B (en) | Intelligent analysis method for urban driving condition | |
| CN109946389B (en) | Structural damage identification method based on ensemble empirical mode decomposition and convolutional neural network | |
| US20210255059A1 (en) | Fault diagnosis method of reciprocating machinery based on keyphasor-free complete-cycle signal | |
| CN111381170A (en) | Electric vehicle battery pack health state prediction method and system based on big data | |
| CN109597401B (en) | Equipment fault diagnosis method based on data driving | |
| CN111832225A (en) | Method for constructing driving condition of automobile | |
| CN108269325A (en) | A kind of analysis method and device of driving behavior oil consumption economy | |
| CN111351665B (en) | Rolling bearing fault diagnosis method based on EMD and residual error neural network | |
| CN111310615A (en) | Small target traffic sign detection method based on multi-scale information and residual error network | |
| CN110596506A (en) | Converter fault diagnosis method based on time convolution network | |
| CN116226646A (en) | Method, system, equipment and medium for predicting health state and residual life of bearing | |
| CN112966853A (en) | Urban road network short-term traffic flow prediction method based on space-time residual error mixed model | |
| CN106656669A (en) | Equipment parameter abnormity detection system and method based on self-adaptive setting of threshold | |
| CN112305379A (en) | Mode identification method and system for GIS insulation defect | |
| CN114819315A (en) | Bearing degradation trend prediction method based on multi-parameter fusion health factor and time convolution neural network | |
| CN115290326A (en) | Rolling bearing fault intelligent diagnosis method | |
| CN116665483A (en) | Novel method for predicting residual parking space | |
| CN114563671A (en) | High-voltage cable partial discharge diagnosis method based on CNN-LSTM-Attention neural network | |
| CN110750848A (en) | Method for estimating remaining life of software-hardware degradation system by considering software operation | |
| CN113987927A (en) | Commercial vehicle-mounted quality detection method | |
| CN116484176A (en) | Bearing fault diagnosis method, system and storage medium based on ultra-wavelet | |
| CN110633314A (en) | Internet of vehicles data processing method and device | |
| CN116596129A (en) | Electric vehicle charging station short-term load prediction model construction method | |
| CN116403090A (en) | Small-size target detection method based on dynamic anchor frame and transducer | |
| CN115718886A (en) | Construction method and device for pure electric vehicle running condition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |