CN112035546B - Fuel consumption correlation factor analysis method for vehicle condition signal data - Google Patents
Fuel consumption correlation factor analysis method for vehicle condition signal data Download PDFInfo
- Publication number
- CN112035546B CN112035546B CN202010897512.8A CN202010897512A CN112035546B CN 112035546 B CN112035546 B CN 112035546B CN 202010897512 A CN202010897512 A CN 202010897512A CN 112035546 B CN112035546 B CN 112035546B
- Authority
- CN
- China
- Prior art keywords
- vehicle condition
- signals
- signal
- numerical
- fuel consumption
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
- G06F16/24578—Query processing with adaptation to user needs using ranking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/248—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
Abstract
The invention discloses an oil consumption correlation factor analysis method of vehicle condition signal data, which comprises the following steps: determining vehicle condition signals influencing oil consumption, and screening oil consumption signal data from a vehicle condition signal database of a big data platform; and (3) scoring each numerical vehicle condition signal and each classified vehicle condition signal in the finally output numerical vehicle condition signals and the finally output classified vehicle condition signals by adopting a LightGBM algorithm in the gradient lifting decision tree, wherein the higher the score is, the higher the correlation between the numerical vehicle condition signals and the fuel consumption signals is, sorting the finally output numerical vehicle condition signals and the finally output classified vehicle condition signals from high to low according to the score, and outputting a ranking result. The method can effectively model the complex vehicle condition signals, and the oil consumption analysis is more accurate.
Description
Technical Field
The invention relates to the technical field of automobile oil consumption analysis, in particular to an oil consumption correlation factor analysis method of vehicle condition signal data.
Background
With the rapid development of communication technology, car networking technology and big data technology, more and more cars with intelligent networking function are carried on, and meanwhile, the collection, transmission, storage and processing technology of car condition signals is mature day by day, and each big car manufacturer can carry out different dimensions of analysis on the cars through the car condition signals. The oil consumption is a very important factor when a user purchases a car, how to analyze the oil consumption and find out an important factor influencing the oil consumption has practical significance for reducing the oil consumption. At present, most of analysis and mining work of various vehicle manufacturers on factors influencing oil consumption stays in an experimental stage, and experimental data has the characteristics of small quantity, small latitude and no completeness; meanwhile, the traditional statistical analysis method cannot effectively model the complex vehicle condition signals, so that the experimental result often has no statistical significance and is greatly different from the actual result used by the user.
Disclosure of Invention
The invention aims to provide a fuel consumption correlation factor analysis method of vehicle condition signal data, which can effectively model complex vehicle condition signals and can more accurately analyze fuel consumption.
In order to achieve the above object, the present invention provides a method for analyzing a fuel consumption correlation factor of vehicle condition signal data, comprising the following steps:
determining vehicle condition signals influencing fuel consumption, and screening fuel consumption signal data from a vehicle condition signal database of a big data platform;
filtering data which do not meet the oil consumption value range in the oil consumption signal data, calculating the oil consumption signal density of each oil consumption signal value interval according to the value distribution of the oil consumption signals, and then carrying out layered sampling on the vehicle condition signals according to the sampling proportion;
performing information entropy calculation on each signal in the vehicle condition signal data after the layered sampling, deleting the vehicle condition signal data of which the information entropy is smaller than a first preset value, and reserving the vehicle condition signal data of which the information entropy is larger than or equal to the first preset value; the vehicle condition signals after information entropy filtration comprise numerical vehicle condition signals and category vehicle condition signals;
calculating the Pearson correlation coefficient of the numerical vehicle condition signals and the fuel consumption signals, firstly keeping the numerical signals of which the absolute value of the Pearson correlation coefficient is larger than a second preset value, then calculating the maximum information coefficient of the numerical signals which are smaller than or equal to the second preset value, keeping the numerical signals of which the maximum information coefficient is larger than a third preset value, and taking and collecting the numerical vehicle condition signals kept twice as the final output numerical vehicle condition signals;
calculating the correlation between each type vehicle condition signal and the fuel consumption signal by using a random forest algorithm, reserving the type vehicle condition signals with the average impurity degree larger than a fourth preset value, and counting the type vehicle condition signals as N; calculating the correlation between each type of vehicle condition signal and the fuel consumption signal by using a recursive feature elimination algorithm, sequencing the correlation from high to low of all types of vehicle condition signals, reserving the first N types of vehicle condition signals, and taking and collecting the two reserved types of vehicle condition signals as the finally output type of vehicle condition signals;
and (3) scoring each of the finally output numerical type vehicle condition signal and the finally output category type vehicle condition signal by adopting a LightGBM algorithm in the gradient lifting decision tree, wherein the higher the score is, the higher the correlation between the numerical type vehicle condition signal and the fuel consumption signal is, sorting the finally output numerical type vehicle condition signal and the finally output category type vehicle condition signal from high to low according to the scores, and outputting a ranking result.
Further, before calculating the correlation between each type of vehicle condition signal and the fuel consumption signal by using a random forest algorithm, the following steps are also executed: the category type vehicle condition signal is numerically encoded.
Further, the vehicle condition signals comprise the real torque of the engine, the rotating speed of the engine, the state of a brake pedal, the actual gear, the vehicle speed, the total driving mileage and the temperature of the vehicle.
Further, the oil consumption signal density of each oil consumption signal value interval is calculated according to the numerical distribution of the oil consumption signals, and the method specifically comprises the following steps: and performing histogram statistics according to the numerical distribution of the oil consumption signals, and calculating the data density of each histogram block.
Further, the first preset value is 0.1.
Further, the second preset value is 0.2.
Further, the third preset value is 0.1.
Further, the fourth preset value is 0.01.
Compared with the prior art, the invention has the following advantages:
according to the oil consumption correlation factor analysis method for the vehicle condition signal data, the vehicle condition signals influencing oil consumption are collected by using the large data platform, and the collected data volume is large, the dimensionality is multiple, and the completeness is realized; data which do not meet the oil consumption value range in the oil consumption signal data are filtered, so that the oil consumption signal value has completeness; linear correlation is calculated for the numerical vehicle condition signals, and then nonlinear correlation is calculated for the linear correlation weak signals, so that the linear correlation calculation cost is low, and the algorithm can perform parallel calculation; the LightGBM algorithm in the gradient lifting decision tree is adopted for modeling, effective modeling of complex vehicle condition signals can be achieved, the experimental result has statistical significance, and the oil consumption analysis is more accurate.
Drawings
Fig. 1 is a flowchart of a method for analyzing a fuel consumption correlation factor of vehicle condition signal data according to the present invention.
Detailed Description
The following further describes embodiments of the present invention with reference to the drawings.
Referring to fig. 1, the embodiment discloses a method for analyzing a fuel consumption correlation factor of vehicle condition signal data, which includes the following steps:
determining vehicle condition signals influencing oil consumption, and screening oil consumption signal data from a vehicle condition signal database of a big data platform;
and filtering data which do not meet the oil consumption value range in the oil consumption signal data, calculating the oil consumption signal density of each oil consumption signal value interval according to the value distribution of the oil consumption signals, and then carrying out layered sampling on the vehicle condition signals according to the sampling proportion.
Performing information entropy calculation on each signal in the vehicle condition signal data after the layered sampling, deleting the vehicle condition signal data of which the information entropy is smaller than a first preset value, and reserving the vehicle condition signal data of which the information entropy is larger than or equal to the first preset value; wherein, the vehicle condition signals after the information entropy filtration comprise numerical type vehicle condition signals andcategory type vehicle condition signal. The formula of the information entropy is as follows: h (x) = ∑ Σ i Pilog 2 Pi, wherein H (x) represents the information entropy of the random variable x, and Pi represents the probability of the random variable taking the value i; the larger the information entropy is, the more information the signal contains is, the smaller the information entropy is, the less information the signal contains is, and the condition signal with small information entropy is deleted, wherein the condition signal data with information entropy smaller than a first preset value, such as the state of an engine hood, the state signal of a window motor and the like, is deleted.
The method comprises the steps of calculating the Pearson correlation coefficient of a numerical vehicle condition signal and a fuel consumption signal, firstly keeping the numerical signal of which the absolute value of the Pearson correlation coefficient is larger than a second preset value, then calculating the maximum information coefficient (maximum information coefficient) of the numerical signal which is smaller than or equal to the second preset value, keeping the numerical signal of which the maximum information coefficient is larger than a third preset value, and taking and collecting the numerical vehicle condition signals which are kept twice as the final output numerical vehicle condition signal. In this embodiment, the correlation coefficient values of the two retained numerical type vehicle condition signals are also sorted from large to small. In this embodiment, the second preset value is 0.2, and the third preset value is 0.1. For a numerical vehicle condition signal, according to the formula of Pearson's correlation coefficient: p is a radical of formula XY ＝cov(X,Y)/σ X σ Y Where cov (X, Y) represents the covariance between variable X and variable Y, σ X Represents the standard deviation of the variable X, and σ Y represents the standard deviation of the variable Y; calculating the Pearson correlation coefficient of each numerical value signal and the oil consumption signal, and keeping the absolute value of the Pearson correlation coefficient larger than a second preset value (| p) XY A numerical signal, | > 0.2); the numerical signal with the absolute value of the Pearson correlation coefficient smaller than or equal to the second preset value represents a signal with weaker linear correlation index strength, a nonlinear correlation relationship possibly exists between the signal and the oil consumption signal, and according to a formula of a maximum information coefficient:
Calculating the correlation between each type vehicle condition signal and the fuel consumption signal by using a random forest algorithm, reserving the type vehicle condition signals with the average impurity degree larger than a fourth preset value, and counting the type vehicle condition signals as N; and calculating the correlation between each type of vehicle condition signal and the fuel consumption signal by using a recursive feature elimination algorithm, sequencing the correlation from high to low of all types of vehicle condition signals, reserving the first N types of vehicle condition signals, and taking and collecting the two reserved types of vehicle condition signals as the finally output type of vehicle condition signals. In the present embodiment, the fourth preset value is 0.01. And then calculating the correlation between the category type vehicle condition signals and the fuel consumption signals by utilizing a random forest algorithm of integrated learning, wherein the score of each signal represents the reduction of the average impurity degree (Mean coarse impuity) of the fuel consumption signals by the signal, the score is larger, the correlation with the fuel consumption signals is more shown, category type vehicle condition signals with the average impurity degree larger than a fourth preset value (MDI is larger than 0.01) are reserved, and the total number N of the reserved signals is recorded. And calculating the correlation between each category signal and the fuel consumption signal by using a recursive characteristic elimination algorithm, selecting the signal with the strongest correlation each time, calculating the correlation between the rest signals and the fuel consumption signal until all the rest signals are calculated each time, and keeping the first N category vehicle condition signals. And finally, taking the intersection of the vehicle condition signals which are reserved after the calculation of the two algorithms as the effective type vehicle condition signals.
And (3) scoring each numerical vehicle condition signal and each classified vehicle condition signal in the finally output numerical vehicle condition signals and the finally output classified vehicle condition signals by adopting a LightGBM algorithm in the gradient lifting decision tree, wherein the higher the score is, the higher the correlation between the numerical vehicle condition signals and the fuel consumption signals is, sorting the finally output numerical vehicle condition signals and the finally output classified vehicle condition signals from high to low according to the score, and outputting a ranking result.
Because the LightGBM algorithm in the Gradient Boosting Decision Tree (GBDT) can process both numerical signals and class signals, the algorithm does not need to carry out one-hot coding on the class signals, the generation of a signal list sparse matrix is avoided, and the problem that the 1-VS-many condition which is not processed by the algorithm is effectively solved. And inputting the screened vehicle condition signals (numerical type and classification type) serving as training data into a LightGBM algorithm, wherein the LightGBM algorithm scores each vehicle condition signal, and outputs ranking results of each vehicle condition signal according to the scoring results, and the signals with higher scores show that the signals have higher correlation with the fuel consumption signals.
In the embodiment, before calculating the correlation between each type of vehicle condition signal and the fuel consumption signal by using the random forest algorithm, the following steps are further executed: the category type vehicle condition signal is numerically encoded. Most of the class type vehicle condition signals are switch type signals with sequences or degrees, the signals can be subjected to numerical value coding, and the numerical value represents the strength or the sequence relation of the signals. And the vehicle condition signals of the big data are subjected to hierarchical sampling, and the whole information of the fuel consumption signals is reserved to the maximum extent.
In the present embodiment, the vehicle condition signals include an engine real torque, an engine speed, a brake pedal state, an actual gear, a vehicle speed, a total driving distance, and a vehicle temperature. The method for determining the vehicle condition signal comprises the following steps: through the discussion with the expert in the oil consumption analysis field, the vehicle condition signals influencing the oil consumption are determined from experience, and other common vehicle condition signals are added. The determination method facilitates the discovery of unknown signals from the data, which signals can influence the fuel consumption. In some embodiments, the vehicle condition signals empirically determined to affect the fuel consumption include the actual torque of the engine, the engine speed, the brake pedal state and the actual gear, and in some embodiments, the vehicle condition signals empirically determined to affect the fuel consumption are not limited herein and are set according to the actual conditions. The common vehicle condition signals include vehicle speed, total mileage and vehicle temperature, and in some embodiments, other common vehicle condition signals are also included, which are not limited herein and are set according to actual conditions.
In this embodiment, the method for calculating the fuel consumption signal density of each fuel consumption signal value interval according to the value distribution of the fuel consumption signals specifically includes the following steps: and performing histogram statistics according to the numerical distribution of the oil consumption signals, and calculating the data density of each histogram block.
After data which do not meet the oil consumption value range in the oil consumption signal data are filtered, the following steps are also executed: carrying out statistical analysis on the filtered oil consumption signal; the statistical analysis comprises the following steps: calculating a common statistical index of the oil consumption signal data, and analyzing the statistical index; the common statistical indexes comprise mean, variance, maximum value, minimum value, 1/4 digit number, median, 3/4 digit number and mode. The calculation of the common statistical indexes of the oil consumption signal data is beneficial to the arrangement and understanding of the distribution of the oil consumption signal data, and is convenient for technicians to formulate a method for calculating the oil consumption correlation factor.
According to the oil consumption correlation factor analysis method of the vehicle condition signal data, the vehicle condition signals influencing oil consumption are collected by using the big data platform, the collected data volume is large, and the dimensionality is large; data which do not meet the oil consumption value range in the oil consumption signal data are filtered, so that the oil consumption signal value has completeness; linear correlation is calculated on the numerical vehicle condition signals, then nonlinear correlation is calculated on the linear correlation weak signals, so that the linear correlation calculation cost is low, and the algorithm can perform parallel calculation; the LightGBM algorithm in the gradient lifting decision tree is adopted for modeling, effective modeling of complex vehicle condition signals can be achieved, the experimental result has statistical significance, and the oil consumption analysis is more accurate.
The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations can be devised by those skilled in the art in light of the above teachings. Therefore, the technical solutions that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the prior art according to the concepts of the present invention should be within the scope of protection determined by the claims.
Claims (8)
1. A method for analyzing a fuel consumption correlation factor of vehicle condition signal data is characterized by comprising the following steps:
determining vehicle condition signals influencing fuel consumption, and screening fuel consumption signal data from a vehicle condition signal database of a big data platform;
filtering data which do not meet the oil consumption value range in the oil consumption signal data, calculating the oil consumption signal density of each oil consumption signal value interval according to the value distribution of the oil consumption signals, and then carrying out layered sampling on the vehicle condition signals according to the sampling proportion;
carrying out information entropy calculation on each signal in the vehicle condition signal data after layered sampling, deleting the vehicle condition signal data of which the information entropy is smaller than a first preset value, and reserving the vehicle condition signal data of which the information entropy is larger than or equal to the first preset value; the vehicle condition signals subjected to information entropy filtering comprise numerical vehicle condition signals and category vehicle condition signals;
calculating the Pearson correlation coefficient of the numerical vehicle condition signals and the fuel consumption signals, firstly keeping the numerical signals of which the absolute value of the Pearson correlation coefficient is larger than a second preset value, then calculating the maximum information coefficient of the numerical signals which are smaller than or equal to the second preset value, keeping the numerical signals of which the maximum information coefficient is larger than a third preset value, and taking and collecting the numerical vehicle condition signals kept twice as the final output numerical vehicle condition signals;
calculating the correlation between each type of vehicle condition signal and the fuel consumption signal by using a random forest algorithm, reserving the type of vehicle condition signals with the average impurity degree larger than a fourth preset value, and counting the type of vehicle condition signals as N; calculating the correlation between each type of vehicle condition signal and the fuel consumption signal by using a recursive feature elimination algorithm, sequencing the correlation of all types of vehicle condition signals from high to low, reserving the first N types of vehicle condition signals, and taking and collecting the two reserved types of vehicle condition signals as the finally output type of vehicle condition signals;
and (3) scoring each of the finally output numerical type vehicle condition signal and the finally output category type vehicle condition signal by adopting a LightGBM algorithm in the gradient lifting decision tree, wherein the higher the score is, the higher the correlation between the numerical type vehicle condition signal and the fuel consumption signal is, sorting the finally output numerical type vehicle condition signal and the finally output category type vehicle condition signal from high to low according to the scores, and outputting a ranking result.
2. The method for analyzing the fuel consumption correlation factor of the vehicle condition signal data according to claim 1, wherein before calculating the correlation between each type of vehicle condition signal and the fuel consumption signal by using a random forest algorithm, the following steps are further performed: the category type vehicle condition signal is numerically encoded.
3. The fuel consumption correlation factor analysis method of vehicle condition signal data according to claim 1 or 2, wherein the vehicle condition signal includes an engine real torque, an engine speed, a brake pedal state, an actual gear, a vehicle speed, a total driving mileage, and a vehicle temperature.
4. The oil consumption correlation factor analysis method of the vehicle condition signal data according to claim 3, wherein the oil consumption signal density of each oil consumption signal value interval is calculated according to the numerical distribution of the oil consumption signals, and the method specifically comprises the following steps: and performing histogram statistics according to the numerical distribution of the oil consumption signals, and calculating the data density of each histogram block.
5. The method for analyzing the fuel consumption correlation factor of the vehicle condition signal data according to claim 1, 2 or 4, wherein the first preset value is 0.1.
6. The method for analyzing the fuel consumption correlation factor of the vehicle condition signal data according to claim 5, wherein the second preset value is 0.2.
7. The fuel consumption correlation factor analysis method according to claim 1, 2, 4 or 6, wherein the third preset value is 0.1.
8. The method for analyzing the fuel consumption correlation factor of the vehicle condition signal data according to claim 7, wherein the fourth preset value is 0.01.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010897512.8A CN112035546B (en) | 2020-08-31 | 2020-08-31 | Fuel consumption correlation factor analysis method for vehicle condition signal data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010897512.8A CN112035546B (en) | 2020-08-31 | 2020-08-31 | Fuel consumption correlation factor analysis method for vehicle condition signal data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112035546A CN112035546A (en) | 2020-12-04 |
| CN112035546B true CN112035546B (en) | 2022-10-11 |
Family
ID=73586410
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010897512.8A Active CN112035546B (en) | 2020-08-31 | 2020-08-31 | Fuel consumption correlation factor analysis method for vehicle condition signal data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112035546B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150021772A (en) * | 2013-08-21 | 2015-03-03 | 현대자동차주식회사 | Method for guiding fuel efficient driving and system thereof |
| CN109613905A (en) * | 2018-11-07 | 2019-04-12 | 中国汽车技术研究中心有限公司 | A kind of method and apparatus of the high oil consumption bad working environments of Dynamic Recognition heavy-duty commercial vehicle actual motion |
| CN110705774A (en) * | 2019-09-26 | 2020-01-17 | 汉纳森（厦门）数据股份有限公司 | Vehicle energy consumption analysis prediction method and system |
| CN111311783A (en) * | 2020-02-24 | 2020-06-19 | 重庆众鸿科技有限公司 | Vehicle-mounted oil consumption data analysis method and system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7715961B1 (en) * | 2004-04-28 | 2010-05-11 | Agnik, Llc | Onboard driver, vehicle and fleet data mining |
| GB2543281A (en) * | 2015-10-13 | 2017-04-19 | British Gas Trading Ltd | System for energy consumption prediction |
-
2020
- 2020-08-31 CN CN202010897512.8A patent/CN112035546B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150021772A (en) * | 2013-08-21 | 2015-03-03 | 현대자동차주식회사 | Method for guiding fuel efficient driving and system thereof |
| CN109613905A (en) * | 2018-11-07 | 2019-04-12 | 中国汽车技术研究中心有限公司 | A kind of method and apparatus of the high oil consumption bad working environments of Dynamic Recognition heavy-duty commercial vehicle actual motion |
| CN110705774A (en) * | 2019-09-26 | 2020-01-17 | 汉纳森（厦门）数据股份有限公司 | Vehicle energy consumption analysis prediction method and system |
| CN111311783A (en) * | 2020-02-24 | 2020-06-19 | 重庆众鸿科技有限公司 | Vehicle-mounted oil consumption data analysis method and system |
Non-Patent Citations (2)
| Title |
|---|
| Effects of driving style on the fuel consumption of city buses under different road conditions and vehicle masses;Hongjie Ma等;《Transportation Research Part D》;20151020;205-216 * |
| 工况特征参数对客车燃油经济性的影响;刘炳文等;《科学技术与工程》;20180630;第18卷(第17期);181-186 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112035546A (en) | 2020-12-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112254964A (en) | Rolling bearing fault diagnosis method based on rapid multi-scale convolution neural network | |
| CN108108766A (en) | Driving behavior recognition methods and system based on Fusion | |
| CN108269325A (en) | A kind of analysis method and device of driving behavior oil consumption economy | |
| CN108830254B (en) | Fine-grained vehicle type detection and identification method based on data balance strategy and intensive attention network | |
| CN113158947A (en) | Power battery health scoring method, system and storage medium | |
| CN111612640A (en) | Data-driven vehicle insurance fraud identification method | |
| CN113688558B (en) | Automobile driving condition construction method and system based on large database sample | |
| CN111563555A (en) | Driver driving behavior analysis method and system | |
| CN116340746A (en) | Feature selection method based on random forest improvement | |
| CN113297795A (en) | Method for constructing running condition of pure electric vehicle | |
| CN113581188A (en) | Commercial vehicle driver driving style identification method based on Internet of vehicles data | |
| CN114817371A (en) | XGboost-based vehicle driving condition identification method | |
| CN110400469A (en) | Standard based on people's vehicle association analysis, which is driven, is not inconsistent illegal activities method for early warning | |
| CN112035546B (en) | Fuel consumption correlation factor analysis method for vehicle condition signal data | |
| CN110119891B (en) | Traffic safety influence factor identification method suitable for big data | |
| CN101923650A (en) | Random forest classification method and classifiers based on comparison mode | |
| CN108090635B (en) | Road performance prediction method based on cluster classification | |
| CN112836967B (en) | New energy automobile battery safety risk assessment system | |
| CN112651666A (en) | Driver risk assessment method based on driving mode transfer characteristics | |
| CN113657432A (en) | Commercial vehicle driving behavior risk level identification method based on Internet of vehicles data | |
| CN115689040B (en) | Traffic accident severity prediction method and system based on convolutional neural network | |
| CN115718886A (en) | Construction method and device for pure electric vehicle running condition | |
| CN108346287B (en) | Traffic flow sequence pattern matching method based on influence factor analysis | |
| CN112862276A (en) | Vehicle networking platform and method for defining risk preference of driver by combining longitudinal direction and transverse direction | |
| CN111931768A (en) | Vehicle identification method and system capable of self-adapting to sample distribution |
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 |