LU102193B1 - Method for predicting single-vehicle emission - Google Patents
Method for predicting single-vehicle emission Download PDFInfo
- Publication number
- LU102193B1 LU102193B1 LU102193A LU102193A LU102193B1 LU 102193 B1 LU102193 B1 LU 102193B1 LU 102193 A LU102193 A LU 102193A LU 102193 A LU102193 A LU 102193A LU 102193 B1 LU102193 B1 LU 102193B1
- Authority
- LU
- Luxembourg
- Prior art keywords
- emission
- model
- vehicle
- rvm
- short
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 13
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 8
- 230000001133 acceleration Effects 0.000 claims description 23
- 238000012549 training Methods 0.000 claims description 7
- 239000013598 vector Substances 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 abstract description 3
- 238000012360 testing method Methods 0.000 description 6
- 239000000446 fuel Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 238000000746 purification Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Software Systems (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Medical Informatics (AREA)
- Operations Research (AREA)
- Development Economics (AREA)
- Pure & Applied Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Computational Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Algebra (AREA)
- Probability & Statistics with Applications (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Traffic Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Combined Controls Of Internal Combustion Engines (AREA)
Abstract
The present invention proposes a method for predicting single-vehicle emission, which establishes an RVM model by using the real driving conditions of the vehicle and the corresponding emission data, and introduces mutual information algorithm in the RVM model to achieve key features selection, and introduces a leapfrogging algorithm to optimize the calculation of parameters for a kernel function. By adopting the RVM model, it is only required to collect a small amount of random driving data as an input for vehicles with unknown emission levels, and the RVM model can directly output emission results, which is simple and easy to implement with low cost.
Description
LU102193 | Description | The present invention belongs to the field of transportation and particularly | relates to a method for predicting single-vehicle emission. .
At present, for fuel consumption prediction, major automobile companies | and engine companies have mature simulation software (such as AVL-Cruise) | that can accurately predict fuel consumption in actual vehicle operation. For fleet | of cars emission macro prediction, there are mature models such as MOVES. | The effective correlation between vehicle specific power, vehicle operating | speed, and transient pollutant emission rate was established based on the | VSPbin distribution. Based on a large amount of emission data, by considering | the driving characteristic data and adjustment factors of the fleet of cars, the . current and future vehicle activity levels, technical conditions, and overall | emission levels under management factors can be determined or estimated. | However, for single-vehicle emission prediction, in addition to in-engine | purification process and after-treatment devices, emission is also affected by ‘ operating conditions. Traditional modeling methods are complicated and cannot | truly reflect the actual emission of vehicles. : | The present invention provides a method for predicting single-vehicle | emission. The present invention is implemented by the following technical , solutions. . | A method for predicting single-vehicle emission is provided, wherein the |
LU102193 j single-vehicle emission includes emission in idling segments and emission in . short trips. The emission in each short trip is predicted by an RVM (Relevance | Vector Machine) model. The emission in idling segments is the product of the . idling time and the idling emission rate. | Further, when training the RVM model, part of feature values for a short trip ; is used as an input to the model. The type of feature values that are used as the : input to the model is determined by the mutual information value between the ‘ feature values and the emissions in the short trips. | Further, when training the RVM model, parameters for a kernel function are ! determined by a leapfrogging algorithm. | Further, when predicting the emission in a short trip by the RVM model, the .
input includes the time of the short trip, the relative positive acceleration and the | average speed. . Compared with the prior art, the present invention has the beneficial effect i that the emission prediction method of the present invention is used to predict : single-vehicle emission, and can describe the relationship between the vehicle | ; operating speed and the emission level in the current state. For vehicles whose | emission level is unknown, the emission of those vehicles can be directly output : by the model by collecting a small amount of random driving data at low cost and : using the collected data as the input to the model. It is simple and easy to : | implement. | FIG. 1 is an establishment and prediction flowchart of a single-vehicle | emission model; and | FIG. 2 is a histogram showing the mutual information value between the | feature values and the emissions. |
LU102193 |
In this embodiment, the emission in each short trip is predicted by an RVM | model, wherein the establishment of the RVM model includes the following | steps. |
S1: PEMS (portable emission measurement system) test data from 300 ‘ | groups of heavy trucks is selected, including speed and NOx emission, and the | test data for each group is divided into idling segments and short trips; | 82: The idling emission rate is calculated from the ratio of the total time of | all idling segments to the total emission in the test data; and ; |
83: Feature values in each short trip in the test data for each group are ; calculated, including time (t,), average speed (tz), maximum speed (ts), average : acceleration in the acceleration segment (t,), average deceleration in the | deceleration segment (ts), acceleration ratio (ts), deceleration ratio (t,), . constant-speed ratio (tg), relative positive acceleration (tg) and maximum | acceleration (t,), and corresponding emission C in each short trip are recorded. ; In this embodiment, the average speed, the maximum speed and the ; emission are directly obtained from the raw data. . The average acceleration in the acceleration segment is calculated by: | where, a;is the acceleration at an acceleration point, ais the average | acceleration in the acceleration segment, / is the sampling moment, v; is the | speed of the vehicle at the # second, and T is the total time of operating | conditions in the short trips.
The average deceleration in the deceleration | segment is calculated in a way similar to the average acceleration in the | acceleration segment and will not be repeated here. |
The acceleration (deceleration/constant-speed) ratio is a ratio of the total : time of all acceleration (deceleration/constant-speed) conditions in the short trips ' to the total time of operating conditions in the short trips. |
The relative positive acceleration is calculated by: :
where, vi is the speed of the vehicle at the second, T is the total time of | operating conditions in the short trips, a is the acceleration value greater than | 0m/s?, and x is the distance travelled by the vehicle in the short trips. |
Table 1 ' Time Average | Maximum | Average Average |
11 speedt2 | speed t3 acceleration in | deceleration in | acceleration segment t5 | segment t4 ;
Table 2 |
Accelerat | Decelerati | Constant-spe | Relative Maximum | Emission \ SPFFEST | t6 t7 t8 acceleration | on t10 |
19 |
LU102193 ,
S4: By the mutual information algorithm, the mutual information values MI . between the ten feature values and the emission in each short trip are obtained, | respectively, feature values having the maximum mutual information value in all , the short trips are obtained, and three feature values having the highest | information value are used as the key feature values.
Wherein, the mutual . information value is calculated by: ; where, t is a feature value in a short trip, C is the total emission for a short | travel distance in the short trip, MI(t,C) is the mutual information value between :
the feature value and the total emission, H(t) is the entropy for the feature value t, i and H(t,C) is the combination entropy for t and C. | in this embodiment, the selected three key feature values are the time of | the short trip, the relative positive acceleration and the average speed. | 85: The key feature values are normalized to remove the unit and order of :
magnitude of the features. . S6: An RVM model is trained and established by using key feature values ;
for each short trip, which have been normalized, in the 200 groups selected from , the 300 groups of PEMS test data as the input value and the emission for each ; short trip as the output value.
During the training, the calculation of parameters :
for the kernel function is optimized by the leapfrogging algorithm. | In this embodiment, the step of “calculation of parameters for the kernel | function is optimized by the leapfrogging algorithm” includes the following steps: | S601: the population is initialized.; specifically, P L-dimensional vectors are | randomly generated, where the component w; is the random number in the | interval [0,C], and the number of iterations for the population is set as g’, the | number of sub-populations is set as Q, and the number of iterations for the | sub-populations is set as g. | Wherein, one w;corresponds to one training sample.
L is the number of |
LU102193 ; training samples, and C is the preset upper limit. , | $602: The fitness value for each component w; is calculated. If the | constraintYE_, w;y; = 0 is violated( y;is the corresponding expected value of | w;), the fitness value for this component is defined as an infinitely positive | number, otherwise the fitness value is kept unchanged, and the population is . divided into sub-populations. | S603: For each sub-population, the optimal component in the ] sub-population is found for the update formula for the worst component by the | leapfrogging algorithm, so that a new population is generated by the .
combination of all sub-populations. Then, the process goes back to S602 to , repeat this process up to the number of iterations for the population, and Î returned the component w# having the best fitness. | S604: The non-zero solutions in the component w# are calculated. Such . Ë soiutions are parameters for the kernel function. , In this embodiment, the kernel function may be a common kernel function, ; for example, a polynomial kernel function, an RBF kernel function or a Sigmoid | kernel function. : S7: The key feature values for each short trip, which have been normalized, | in the remaining 100 groups of the 300 groups of PEMS test data is selected as : the input value, the output emission and the actual emission are compared to ; verify the established RVM model. | In this embodiment, during the prediction of single-vehicle emission, the | raw vehicle data is divided into idling segments and short trips; the total time of | the idling segments is obtained and the total time is multiplied by the idling | emission rate to obtain the idling emission K,; the key feature values in each - : | short trip are obtained, normalized and input to the RVM model to obtain the | emission in this short trip; and the ernissions in all short trips are accumulated to | obtain the emission in short trips K,. The sum of K, and K, is the |
LU102193 | single-vehicle emission. |
In the embodiments of the present invention, except for the special . description of the models of the devices, the models of other devices are not , limited, as long as the devices can complete the functions described above.
It may be understood by those skilled in the art that the accompanying | drawings are only schematic views of a preferred embodiment, and the serial . numbers of the foregoing embodiments of the present invention are only for | , description and do not represent the priority of the embodiments.
À What described above is merely a preferred embodiment of the present | invention and not intended to limit the present invention.
Any modifications, À equivalent replacements and improvements without departing from the spirit and | principle of the present invention should fall into the protection scope of the ' present invention. ;
Claims (4)
1. A method for predicting single-vehicle emission, is characterized in that | the single-vehicle emission includes emission in idling segments and . accumulative emission in short trips; the emission in each segment is predicted | by an RVM (Relevance Vector Machine) model; the emission in idling segments | is the product of the idling time and the idling emission rate. |
2. The method for predicting single-vehicle emission according to claim 1, is | characterized in that part of feature values for a short trip is used as an input to | the model when training the RVM model; the type of feature values that are used | as the input to the model is determined by the mutual information value between | the feature values and the emissions in the short trips. |
3. The method for predicting single-vehicle emission according to claim 1, is | | characterized in that parameters for a kernel function are determined by a | leapfrogging algorithm when training the RVM model. |
4. The method for predicting single-vehicle emission according to claim 2, is . characterized in that when predicting the emission in a short trip by the RVM . model, the input includes the time of the short trip, the relative positive | acceleration and the average speed. |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910199075.XA CN109948237B (en) | 2019-03-15 | 2019-03-15 | Method for predicting emission of bicycle |
Publications (1)
Publication Number | Publication Date |
---|---|
LU102193B1 true LU102193B1 (en) | 2020-12-04 |
Family
ID=67010114
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
LU102193A LU102193B1 (en) | 2019-03-15 | 2020-03-12 | Method for predicting single-vehicle emission |
Country Status (3)
Country | Link |
---|---|
CN (1) | CN109948237B (en) |
LU (1) | LU102193B1 (en) |
WO (1) | WO2020187113A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113781771A (en) * | 2021-08-17 | 2021-12-10 | 四川省生态环境科学研究院 | Online operation method of IVE model |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109948237B (en) * | 2019-03-15 | 2023-06-02 | 中国汽车技术研究中心有限公司 | Method for predicting emission of bicycle |
CN112529104B (en) * | 2020-12-23 | 2024-06-18 | 东软睿驰汽车技术(沈阳)有限公司 | Vehicle fault prediction model generation method, fault prediction method and device |
CN113806675B (en) * | 2021-08-06 | 2023-06-23 | 中汽研汽车检验中心(天津)有限公司 | NOx emission and oil consumption characteristic analysis method |
CN113918890B (en) * | 2021-11-24 | 2024-07-09 | 中国汽车技术研究中心有限公司 | Low-load working condition construction method based on mobile window library |
CN114821854B (en) * | 2022-03-28 | 2023-06-09 | 中汽研汽车检验中心(天津)有限公司 | Method for evaluating influence of working condition switching on vehicle fuel consumption |
CN114991922B (en) * | 2022-05-30 | 2024-01-23 | 中国汽车工程研究院股份有限公司 | Real-time early warning method for exceeding of NOx emission of vehicle |
CN117074046B (en) * | 2023-10-12 | 2024-01-02 | 中汽研汽车检验中心(昆明)有限公司 | Automobile laboratory emission test method and device in plateau environment |
Family Cites Families (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4318299B2 (en) * | 2004-03-31 | 2009-08-19 | 株式会社山武 | Method and apparatus for creating prediction model for fuel consumption or CO2 emissions |
JP4853969B2 (en) * | 2006-11-28 | 2012-01-11 | 範幸 杉本 | Carbon dioxide display device for automobile with evaluation device |
WO2010071898A2 (en) * | 2008-12-19 | 2010-06-24 | The Johns Hopkins Univeristy | A system and method for automated detection of age related macular degeneration and other retinal abnormalities |
CN101519073A (en) * | 2009-04-07 | 2009-09-02 | 北京大学 | Method for forecasting running load of hybrid electric vehicle |
FR2984557B1 (en) * | 2011-12-20 | 2014-07-25 | IFP Energies Nouvelles | SYSTEM AND METHOD FOR PREDICTING EMISSIONS OF POLLUTANTS OF A VEHICLE WITH SIMULTANEOUS CALCULATIONS OF CHEMICAL KINETICS AND EMISSIONS |
US20130274952A1 (en) * | 2012-04-16 | 2013-10-17 | Feisel Weslati | Predictive powertrain control using driving history |
US9600541B2 (en) * | 2014-05-02 | 2017-03-21 | Kookmin University Industry Academy Corporation Foundation | Method of processing and analysing vehicle driving big data and system thereof |
CN104680232A (en) * | 2014-10-28 | 2015-06-03 | 芜湖杰诺瑞汽车电器***有限公司 | RVM (Relevance Vector Machine)-based engine failure detecting method |
CN105966195B (en) * | 2016-05-31 | 2018-09-25 | 中国汽车技术研究中心 | Intelligent management system for air quality in vehicle and use method thereof |
CN106446398B (en) * | 2016-09-20 | 2019-08-20 | 中山大学 | Light-duty vehicle rate of discharge calculation method based on traffic circulation data and deterioration rate |
CN106845371B (en) * | 2016-12-31 | 2019-10-25 | 中国科学技术大学 | A kind of city road network automotive emission remote sensing monitoring system |
CN107368913B (en) * | 2017-06-15 | 2020-06-12 | 中国汽车技术研究中心 | Oil consumption prediction method based on least square support vector machine |
JP7162814B2 (en) * | 2017-07-31 | 2022-10-31 | 本田技研工業株式会社 | Control device |
CN107545122A (en) * | 2017-09-27 | 2018-01-05 | 重庆长安汽车股份有限公司 | A kind of simulation system of the vehicle gaseous effluent based on neutral net |
CN107886188B (en) * | 2017-10-18 | 2021-12-10 | 东南大学 | Liquefied natural gas bus tail gas emission prediction method |
CN108596104B (en) * | 2018-04-26 | 2021-01-05 | 安徽大学 | Wheat powdery mildew remote sensing monitoring method with disease characteristic preprocessing function |
CN108629450A (en) * | 2018-04-26 | 2018-10-09 | 东南大学 | A kind of liquefied natural gas bus exhaust emissions prediction technique |
CN108717165A (en) * | 2018-05-28 | 2018-10-30 | 桂林电子科技大学 | Lithium ion battery SOC on-line prediction methods based on data-driven method |
CN108846526A (en) * | 2018-08-08 | 2018-11-20 | 华北电力大学 | A kind of CO2 emissions prediction technique |
CN109443779A (en) * | 2018-11-07 | 2019-03-08 | 中国汽车技术研究中心有限公司 | A kind of dynamic diagnosis extracts diesel vehicle actual motion NOXThe method and apparatus of maximum discharge bad working environments |
CN109948237B (en) * | 2019-03-15 | 2023-06-02 | 中国汽车技术研究中心有限公司 | Method for predicting emission of bicycle |
-
2019
- 2019-03-15 CN CN201910199075.XA patent/CN109948237B/en active Active
-
2020
- 2020-03-12 LU LU102193A patent/LU102193B1/en active IP Right Grant
- 2020-03-12 WO PCT/CN2020/078891 patent/WO2020187113A1/en active Application Filing
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113781771A (en) * | 2021-08-17 | 2021-12-10 | 四川省生态环境科学研究院 | Online operation method of IVE model |
Also Published As
Publication number | Publication date |
---|---|
CN109948237A (en) | 2019-06-28 |
WO2020187113A1 (en) | 2020-09-24 |
CN109948237B (en) | 2023-06-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
LU102193B1 (en) | Method for predicting single-vehicle emission | |
Liu et al. | Reinforcement learning optimized look-ahead energy management of a parallel hybrid electric vehicle | |
Yuhui et al. | Development of a representative driving cycle for urban buses based on the K-means cluster method | |
Kivekäs et al. | Stochastic driving cycle synthesis for analyzing the energy consumption of a battery electric bus | |
Wu et al. | An optimized real-time energy management strategy for the power-split hybrid electric vehicles | |
Li et al. | Research on optimized GA-SVM vehicle speed prediction model based on driver-vehicle-road-traffic system | |
Ganji et al. | A study on look-ahead control and energy management strategies in hybrid electric vehicles | |
Abas et al. | Development of Malaysian urban drive cycle using vehicle and engine parameters | |
CN110543098B (en) | Method for selecting configuration of plug-in hybrid power system | |
Cui et al. | Effects of route guidance strategies on traffic emissions in intelligent transportation systems | |
Pham et al. | Analytical solution to energy management guaranteeing battery life for hybrid trucks | |
Finesso et al. | An unsupervised machine-learning technique for the definition of a rule-based control strategy in a complex HEV | |
CN108629450A (en) | A kind of liquefied natural gas bus exhaust emissions prediction technique | |
Montazeri-Gh et al. | Driving condition recognition for genetic-fuzzy HEV control | |
Deshpande et al. | In-vehicle test results for advanced propulsion and vehicle system controls using connected and automated vehicle information | |
Guo et al. | Integrated multistep Markov-based velocity predictor of energy consumption prediction model for battery electric vehicles | |
JP6874540B2 (en) | Vehicle control method, information processing device, and vehicle control system | |
Guo et al. | Clustered energy management strategy of plug-in hybrid electric logistics vehicle based on Gaussian mixture model and stochastic dynamic programming | |
Zhao et al. | Path-forecasting for HEV optimal energy management (POEM) | |
Yu et al. | A real-time robust ecological-adaptive cruise control strategy for battery electric vehicles | |
Li et al. | Eco-routing for plug-in hybrid electric vehicles | |
KR20200053412A (en) | Method for controlling a power train of a motor vehicle | |
CN112993344A (en) | Neural network-based fuel cell system output performance prediction method and device | |
Asher et al. | The effect of hill planning and route type identification prediction signal quality on hybrid vehicle fuel economy | |
Yuan et al. | Study of driving cycle of city tour bus based on coupled GA-K-means and HMM algorithms: a case study in Beijing |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
FG | Patent granted |
Effective date: 20201204 |