CN107886188B - Liquefied natural gas bus tail gas emission prediction method - Google Patents
Liquefied natural gas bus tail gas emission prediction method Download PDFInfo
- Publication number
- CN107886188B CN107886188B CN201710972889.3A CN201710972889A CN107886188B CN 107886188 B CN107886188 B CN 107886188B CN 201710972889 A CN201710972889 A CN 201710972889A CN 107886188 B CN107886188 B CN 107886188B
- Authority
- CN
- China
- Prior art keywords
- data
- emission
- vehicle
- time
- bus
- 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
- 238000000034 method Methods 0.000 title claims abstract description 50
- 239000007789 gas Substances 0.000 title claims abstract description 47
- 239000003949 liquefied natural gas Substances 0.000 title claims abstract description 21
- 230000008569 process Effects 0.000 claims abstract description 9
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 230000001133 acceleration Effects 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 17
- 238000011160 research Methods 0.000 claims description 16
- 238000013138 pruning Methods 0.000 claims description 14
- 230000001174 ascending effect Effects 0.000 claims description 9
- BUCXEFZXWKUCCY-UHFFFAOYSA-N 4-methyl-3-(2-phenylethyl)-1,2,4-oxadiazol-5-one Chemical compound O1C(=O)N(C)C(CCC=2C=CC=CC=2)=N1 BUCXEFZXWKUCCY-UHFFFAOYSA-N 0.000 claims description 6
- 238000002229 photoelectron microspectroscopy Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 5
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 claims description 4
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000005520 cutting process Methods 0.000 claims description 3
- 230000005484 gravity Effects 0.000 claims description 3
- 238000005096 rolling process Methods 0.000 claims description 3
- 230000006872 improvement Effects 0.000 description 7
- 238000009826 distribution Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000004134 energy conservation Methods 0.000 description 3
- 239000003344 environmental pollutant Substances 0.000 description 3
- 231100000719 pollutant Toxicity 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 238000012216 screening Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 239000000446 fuel Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- BXNJHAXVSOCGBA-UHFFFAOYSA-N Harmine Chemical compound N1=CC=C2C3=CC=C(OC)C=C3NC2=C1C BXNJHAXVSOCGBA-UHFFFAOYSA-N 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 239000000356 contaminant Substances 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000001737 promoting effect 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Navigation (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
Abstract
The invention discloses a liquefied natural gas bus tail gas emission prediction method, which comprises the following steps: selecting a predicted bus route; collecting GPS data and emission data of the public transport vehicle; data preprocessing and identification of idle speed nodes; calculating the specific power parameter of the vehicle, and performing discretization processing; establishing a regression model of the mass emission rate and the specific power parameter of the bus in the road section driving state based on the CART algorithm, and calculating the mass emission rate of the bus in the idle speed state by adopting a mean value estimation method; and establishing an emission quality prediction model of the vehicle in any period of time in the running process. By adopting the scheme, the tail gas emission of the liquefied natural gas bus can be scientifically predicted, so that experience and reference are provided for traffic managers and decision makers.
Description
Technical Field
The invention relates to a traffic energy-saving emission-reducing method, in particular to a liquefied natural gas bus tail gas emission prediction method.
Background
Green traffic is an important target of traffic industry development all the time, and bus priority not only can maximize road utilization rate, but also has important effects on the aspects of energy conservation and emission reduction. In recent years, new energy public transportation represented by pure electric public transportation, hybrid power, liquefied natural gas public transportation and the like is greatly popularized, so that exhaust emission is further reduced, and the method has an important effect on improvement of urban air quality.
In the aspect of research on tail gas emission of buses, related research covers many levels from macro to micro, and the macro research shows that public transport in a certain area has positive influence on energy conservation and emission reduction. The study on the microscopic level mainly comprises (1) the study on some short-time emission characteristics, which mainly comprise the tail gas emission characteristics of buses at bus stations, the tail gas emission characteristics at intersections, the emission characteristic comparison under different fuel types and the like, and shows that the emission of new energy vehicles is far smaller than that of traditional fuel buses; (2) the research on the influence factors of tail gas emission mainly comprises characteristics of a bus driver and passengers, vehicle running conditions, bus station characteristics and running environment. (3) And (3) prediction research of exhaust emission: most of the research on the microscopic level is based on modeling of the specific power of the motor vehicle. This has the advantage that a number of parameters of the vehicle operation are taken into account, including speed, acceleration, load, gradient, etc. The method mainly adopted comprises mean estimation based on vehicle specific power parameters (VSP) and polynomial fitting based on the vehicle specific power parameters, however, a very suitable corresponding relation between the emission and the VSP is not provided. In addition, the existing literature has obvious defects on the emission of the liquefied natural gas public transport, so the invention adopts the idea of big data, improves the defects and carries out real-time prediction research on the tail gas emission of the liquefied natural gas public transport.
Disclosure of Invention
The invention provides a liquefied natural gas bus tail gas emission prediction method aiming at the technical problems in the prior art, which is used for predicting the emission conditions of liquefied natural gas buses in various states and providing reference and decision-making bases for traffic management departments.
In order to achieve the purpose, the technical scheme of the invention is as follows: a liquefied natural gas bus tail gas emission prediction method comprises the following steps:
step 1: inspecting the emission condition of the bus, determining the bus to be researched, and selecting a proper research time period;
step 2: the method comprises the following steps of utilizing the PEMS and the GPS equipment to collect real-time operation condition data and emission data of the bus, wherein the data collected by the PEMS equipment comprise: CO, CO of vehicles2HC and NOXMass discharge rate; the data collected by the GPS equipment comprises: the position information (including longitude and latitude) and the real-time speed of the vehicle, and calculating the real-time acceleration;
and step 3: data cleaning and GPS data preprocessing;
and 4, step 4: identifying the idle node position in the running process of the vehicle based on a point density method;
and 5: calculating a Vehicle Specific Power (VSP) parameter of the Vehicle, and carrying out VSP discretization;
step 6: dividing the running state of the vehicle into a road section running state and an idle node state, and respectively predicting the mass emission rate in the two states by applying a CART algorithm and a mean value estimation method;
and 7: based on the mass emission rate prediction model in the step 6, combining the two states, establishing a vehicle full-line tail gas emission real-time prediction model, solving the model, and predicting the emission quality of the vehicle under different running time lengths;
as an improvement of the present invention, the real-time acceleration calculation formula of the vehicle in step 2 is as follows:
ai＝vi-vi-1
wherein, aiThe unit is m/s which is the acceleration at the ith second2；viThe average speed at the ith second is taken as the unit of m/s; v. ofi-1The average speed at the i-1 second is in m/s.
As an improvement of the present invention, the content of the GPS data preprocessing in step 3 mainly includes the following aspects:
the data collection process is usually interfered by road traffic environment, communication signals and satellite signals along the way, so that the signal loss condition is generated, and the GPS signal loss can generate three conditions of data repetition, data loss and data error;
step 31: the data repetition means that a plurality of data are completely the same in continuous time, and the repeated data of the station part is not influenced too much and is reserved; and the data of the road section is directly removed.
Step 32: the data loss is that the GPS signal cannot be received within a period of time due to the influence of various external environments, GPS positioning information is obtained, and if the loss time is short, a linear interpolation method is adopted for interpolation; if the loss time is longer, directly removing the part of data;
step 33: the data error is that the measured longitude and latitude data has great deviation from the actual data, and obvious error exists, and the data is directly removed.
As an improvement of the present invention, the step 4 specifically includes the following steps:
step 41: arranging the position information of the vehicle according to a time sequence, and calculating the distance L between two adjacent points;
step 42: if the distances L between k-1 adjacent points among the adjacent k (k < ═ 100) points are less than 1m, the points are considered to be possible idle speed nodes, namely bus stops or partial intersections, and the points are added into a list of the idle speed nodes;
step 43: if two groups of possible idle speed nodes exist, according to the time sequence, subtracting the time T2 corresponding to the last point of the former group from the time T1 corresponding to the first point of the latter group to obtain a time difference delta T smaller than T, namely a time difference threshold, and according to the running characteristics of the bus, taking T as 10s, then considering that the two groups of idle speed nodes are both positioned at the same bus stop or intersection; in the case of Δ T > T, if Δ T >0, it is indicated that a secondary stop has occurred; if Δ t < ═ 0, it means that the number of points included in the idle node exceeds k;
step 44: and only one same element in the idle node list is reserved, and all the idle node GPS information is sorted according to time sequence.
As an improvement of the present invention, the step 5 specifically includes the following steps:
step 51: the specific power of the motor vehicle represents the power required to be output when the engine moves one ton of mass (including dead weight), the unit is represented by km/t, various parameters including speed, acceleration, air resistance coefficient, gradient and windward area are contained in a calculation formula of the specific power, and the calculation formula is as follows:
wherein m: the mass of the vehicle and all the people and objects inside the vehicle is kg; v: vehicle speed in m/s;
a: acceleration in m/s2；ε：A mass coefficient; grade: the slope is the ratio of the slope height to the slope length, and the unit is percent;
g: acceleration of gravity in m/s2；CR: a rolling resistance coefficient; cD: a wind resistance coefficient; a: frontal area of vehicle in m2；ρa: air density in kg/m3。
Step 52: in order to facilitate the following research on bus emission, the power parameter needs to be compared for discretization, that is, the VSP is divided according to a certain mode, and the dividing mode is as follows:
as an improvement of the present invention, the CART algorithm in step 6 is specifically as follows:
step 61: in order to eliminate the influence of too large or too small data on the modeling, part of the data is required to be removed: corresponding CO and CO of the same VSP2HC and NOXThe mass discharge rate data are respectively arranged in ascending order; according to each group of CO mass emission rate data after ascending arrangement, removing the data of the first 2.5 percent and the data of the second 2.5 percent of the data of the row, namely, retaining the data of the middle 95 percent for establishing a regression model;
step 62, the CART algorithm is used for establishing a regression model of the mass emission rate vehicle specific power parameter, and mainly comprises two parts, namely selection of an optimal cutting point and tree pruning;
(1) selection of optimal cut points: for each type of data set, a linear model is used for fitting the data set, then the difference between a real target value and a model predicted value is calculated, the squares of the difference are summed to obtain a required total difference (SSE), and finally the feature with the minimum total difference is still selected as a branch feature.
Wherein, yiAnd
representing the actual value and the estimated value, respectively.
(2) Pruning the tree: generally, the Pruning of CART includes Pre-Pruning (Pre-Pruning) and Post-Pruning (Post-Pruning). The invention adopts a pre-pruning method, namely, the tree growth is stopped early according to some principles, and the overfitting condition is prevented by setting the minimum sample parameters in the nodes;
and step 63: calibrating the allowable error in the model and the minimum sample number parameter of segmentation;
step 64: calculating a coefficient of determination (R)2) The effect of the fit was evaluated with the percent mean absolute error (MAPE) and calculated as follows:
wherein,
is the average of each data set.
As a modification of the present invention, the step 6 specifically includes the following steps:
step 65: similar to the method for screening data in step 61, the CO and CO in idle state are respectively selected2HC and NOxArranging the data in an ascending order, and removing 2.5 percent of the data before and after the data are arranged;
and step 66: calculate the average mass emission rate for each emission:
wherein:
of the exhaust gas type iAverage mass emission rate, N is the total number of node idle data of the tail gas type i,
is the value of the mass discharge rate of the first data.
As an improvement of the present invention, the step 7 specifically includes the following steps:
the emission quality of certain exhaust gas in the time T is equal to the sum of the emission quality of certain exhaust gas in the road driving state and the emission quality of certain exhaust gas in the node idling state, and the calculation formula is as follows:
wherein: x is the number ofsThe variable is 0-1, when the vehicle is in a road section driving state, 1 is taken, otherwise 0 is taken; x is the number ofNThe variable is 0-1, when the vehicle is in a node idling state, 1 is taken, otherwise 0 is taken; t isSMeans the total time T that the vehicle is in the driving state of the road section within the time TNThe total time that the vehicle is in the node idling state within the time T is referred to;
the mass emission rate of the tail gas type i in the road section driving state is indicated;
the mass emission rate of the tail gas type i in a node idling state is indicated; emissioniRefers to the total mass of the exhaust i emitted by the vehicle within the time T.
Compared with the prior art, the invention has the following beneficial effects: the invention starts from the idea of big data, applies the CART algorithm, explores the relation between specific power parameters and mass emission rate, and on the basis, establishes a bus full-line exhaust emission prediction model, can scientifically predict the exhaust emission quality of a vehicle in any period of time during running, realizes the effective connection of traffic condition data and exhaust emission data, has positive effect on the research of bus emission characteristic law, provides data support and theoretical basis for the formulation of related policies of bus energy conservation and emission reduction, enables traffic managers and planners to better manage and optimize system operation, and further reduces the pollutant emission amount of the vehicle in the running process The energy-saving and environment-friendly public transport system provides experience reference.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
FIG. 2 is a schematic diagram of the linear interpolation method of the present invention.
FIG. 3 is a diagram of the recognition result of the idle node of the partial running track of the bus.
FIG. 4 is a distribution diagram of the calculation results of the specific power parameters of the 1-way bus according to the present invention.
FIG. 5 is a frequency histogram of the VSP Bin distribution of the present invention.
FIG. 6 is a graphical representation of CO mass emission rate versus VSP for the present invention.
FIG. 7 is a CO of the present invention2Mass emission rates are plotted versus VSP.
FIG. 8 is a graphical representation of HC mass emission rate versus VSP for the present invention.
FIG. 9 is NO of the present inventionXMass emission rates are plotted versus VSP.
Fig. 10 is a diagram of the predicted results of the vehicle of the present invention for different operating periods.
Detailed Description
For the purposes of promoting an understanding and appreciation of the invention, the invention will be further described in connection with specific embodiments thereof;
example 1:
a liquefied natural gas bus tail gas emission prediction method comprises the following steps:
step 1: inspecting the emission condition of the bus, determining the bus to be researched, and selecting a proper research time period;
step 2: the PEMS and the GPS equipment are utilized to collect real-time running condition data and emission data of the bus,the data collected by the PEMS device includes: CO, CO of vehicles2HC and NOXMass discharge rate; the data collected by the GPS equipment comprises: the position information (including longitude and latitude) and the real-time speed of the vehicle, and calculating the real-time acceleration;
and step 3: data cleaning and GPS data preprocessing;
and 4, step 4: identifying the idle node position in the running process of the vehicle based on a point density method;
and 5: calculating a Vehicle Specific Power (VSP) parameter of the Vehicle, and carrying out VSP discretization;
step 6: dividing the running state of the vehicle into a road section running state and an idle node state, and respectively predicting the mass emission rate in the two states by applying a CART algorithm and a mean value estimation method;
and 7: based on the mass emission rate prediction model in the step 6, combining the two states, establishing a vehicle full-line tail gas emission real-time prediction model, solving the model, and predicting the emission quality of the vehicle under different running time lengths;
the real-time acceleration calculation formula of the vehicle in the step 2 is as follows:
ai＝vi-vi-1；
wherein, aiThe unit is m/s which is the acceleration at the ith second2；viThe average speed at the ith second is taken as the unit of m/s; v. ofi-1The average speed at the i-1 second is in m/s.
The content of the GPS data preprocessing in step 3 mainly includes the following aspects:
the data collection process is usually interfered by road traffic environment, communication signals and satellite signals along the way, so that the signal loss condition is generated, and the GPS signal loss can generate three conditions of data repetition, data loss and data error;
step 31: the data repetition means that a plurality of data are completely the same in continuous time, and the repeated data of the station part is not influenced too much and is reserved; and the data of the road section is directly removed.
Step 32: the data loss is that the GPS signal cannot be received within a period of time due to the influence of various external environments, GPS positioning information is obtained, and if the loss time is short, a linear interpolation method is adopted for interpolation; if the loss time is longer, directly removing the part of data;
step 33: the data error is that the measured longitude and latitude data has great deviation from the actual data, and obvious error exists, and the data is directly removed.
The step 4 is specifically as follows:
step 41: arranging the position information of the vehicle according to a time sequence, and calculating the distance L between two adjacent points;
step 42: if the distances L between k-1 adjacent points among the adjacent k (k < ═ 100) points are less than 1m, the points are considered to be possible idle speed nodes, namely bus stops or partial intersections, and the points are added into a list of the idle speed nodes;
step 43: if two groups of possible idle speed nodes exist, according to the time sequence, subtracting the time T2 corresponding to the last point of the former group from the time T1 corresponding to the first point of the latter group to obtain a time difference delta T smaller than T, namely a time difference threshold, and according to the running characteristics of the bus, taking T as 10s, then considering that the two groups of idle speed nodes are both positioned at the same bus stop or intersection; in the case of Δ T > T, if Δ T >0, it is indicated that a secondary stop has occurred; if Δ t < ═ 0, it means that the number of points included in the idle node exceeds k;
step 44: and only one same element in the idle node list is reserved, and all the idle node GPS information is sorted according to time sequence.
The step 5 is specifically as follows:
step 51: the specific power of the motor vehicle represents the power required to be output when the engine moves one ton of mass (including dead weight), the unit is represented by km/t, various parameters including speed, acceleration, air resistance coefficient, gradient and windward area are contained in a calculation formula of the specific power, and the calculation formula is as follows:
wherein m: the mass of the vehicle and all the people and objects inside the vehicle is kg; v: vehicle speed in m/s;
a: acceleration in m/s2(ii) a Epsilon: a mass coefficient; grade: the slope is the ratio of the slope height to the slope length, and the unit is percent;
g: acceleration of gravity in m/s2；CR: a rolling resistance coefficient; cD: a wind resistance coefficient; a: frontal area of vehicle in m2；ρa: air density in kg/m3。
Step 52: in order to facilitate the following research on bus emission, the power parameter needs to be compared for discretization, that is, the VSP is divided according to a certain mode, and the dividing mode is as follows:
the CART algorithm in step 6 is specifically as follows:
step 61: in order to eliminate the influence of too large or too small data on the modeling, part of the data is required to be removed: corresponding CO and CO of the same VSP2HC and NOXThe mass discharge rate data are respectively arranged in ascending order; according to each group of CO mass emission rate data after ascending arrangement, removing the data of the first 2.5 percent and the data of the second 2.5 percent of the data of the row, namely, retaining the data of the middle 95 percent for establishing a regression model;
step 62, the CART algorithm is used for establishing a regression model of the mass emission rate vehicle specific power parameter, and mainly comprises two parts, namely selection of an optimal cutting point and tree pruning;
(1) selection of optimal cut points: for each type of data set, a linear model is used for fitting the data set, then the difference between a real target value and a model predicted value is calculated, the squares of the difference are summed to obtain a required total difference (SSE), and finally the feature with the minimum total difference is still selected as a branch feature.
Wherein, yiAnd
representing the actual value and the estimated value, respectively.
(2) Pruning the tree: generally, the Pruning of CART includes Pre-Pruning (Pre-Pruning) and Post-Pruning (Post-Pruning). The invention adopts a pre-pruning method, namely, the tree growth is stopped early according to some principles, and the overfitting condition is prevented by setting the minimum sample parameters in the nodes;
and step 63: calibrating the allowable error in the model and the minimum sample number parameter of segmentation;
step 64: calculating a coefficient of determination (R)2) The effect of the fit was evaluated with the percent mean absolute error (MAPE) and calculated as follows:
wherein,
average for each data set;
step 65: similar to the method for screening data in step 61, the CO and CO in idle state are respectively selected2HC and NOxArranging the data in an ascending order, and removing 2.5 percent of the data before and after the data are arranged;
and step 66: calculate the average mass emission rate for each emission:
wherein:
refers to the average mass emission rate of the tail gas type i, N is the total number of node idle speed data of the tail gas type i,
is the value of the mass discharge rate of the first data.
The step 7 is specifically as follows:
the emission quality of certain exhaust gas in the time T is equal to the sum of the emission quality of certain exhaust gas in the road driving state and the emission quality of certain exhaust gas in the node idling state, and the calculation formula is as follows:
wherein: x is the number ofsThe variable is 0-1, when the vehicle is in a road section driving state, 1 is taken, otherwise 0 is taken; x is the number ofNThe variable is 0-1, when the vehicle is in a node idling state, 1 is taken, otherwise 0 is taken; t isSMeans the total time T that the vehicle is in the driving state of the road section within the time TNThe total time that the vehicle is in the node idling state within the time T is referred to;
the mass emission rate of the tail gas type i in the road section driving state is indicated;
the mass emission rate of the tail gas type i in a node idling state is indicated; emissioniRefers to the total mass of the exhaust i emitted by the vehicle within the time T.
The application example is as follows:
the method for predicting the liquefied natural gas bus tail gas emission is described with reference to fig. 1, and comprises the following steps.
TABLE 1 basic information of bus routes
as shown in FIG. 2, in a rectangular plane coordinate system, (x)0,y0) And (x)1,y1) Are two known points and now need to estimate x (x is located at x)0And x1Between) the corresponding value of y.
Due to the point (x)0,y0) And point (x)0,y0)，(x1,y1) On the same line, therefore, first calculate the pass (x)0,y0)，(x1,y1) Linear equation for two points:
further, the value of y can be calculated by using x, (x)0,y0) And (x)1,y1) Expressed as:
then
Is an estimate of y obtained by linear interpolation.
And 4, identifying the idle speed position of the node of the vehicle based on a point density method, and compiling an algorithm program by using a matlab, wherein idle speed identification points of the vehicle running track and part of road sections are shown in the attached figure 3.
And 5: and calculating the specific power value of the vehicle, wherein the parameter values in the VSP formula are shown in a table 2 according to the characteristic parameters of the vehicle:
TABLE 2 VSP formula parameter values
Based on the parameters used in Table 1, the specific power calculation formula in the embodiment can be simplified as follows
Substituting the speed and acceleration data into the above equation, the specific power distribution of 51 buses can be obtained, and the result is shown in fig. 4.
The VSP is discretized to obtain a frequency distribution graph of VSP bin, as shown in fig. 5, it can be seen that after discretization, the VSP has the highest ratio of 0 to 1, most values are concentrated near 0, and the probability of VSP values from-11 to-11 is more than 95%, and approximately follows normal distribution. Considering that the state data of the VSP outside the range of [ -11,11] after discretization is less, and the data is not beneficial to finding out a rule or useful and reasonable information, the invention only researches the state of the VSP after discretization taking the values from-11 to 11.
Step 6: screening of useful data for use according to the method of claim 6The CART algorithm regresses emissions data on the road segment. Two important parameters of the model tree, the allowable error drop value is 0, and the minimum sample number of the segmentation is 1000. Respectively establishing CO and CO2HC and NOXThe mass emission rates were modeled by a model tree regression model on discretized VSP, and the results are shown in fig. 6, 7, 8, and 9.
The regression equation of the mass emission rate versus specific power parameter for each pollutant is as follows:
the coefficient of determination and percentage of mean absolute error calculation formula of claim 64, wherein the calculation result for each contaminant is shown in Table 3:
TABLE 3R2And MAPE calculated result
It can be seen that the regression fit of CO among the four pollutants works best, followed by CO2And NOXHC fits are the least effective, but also substantially meet the fit requirement.
For the node idling state, after removing 2.5% of data before and after the node idling state, estimating the total by adopting a mean value estimation method and using the mean value of the emission rates of the exhaust gas quality. The analytical results are shown in table 4:
TABLE 4 mean exhaust emission rate and 95% confidence interval at idle
And 7: and (3) establishing a full-line tail gas emission model of the liquefied natural gas bus based on the step (7), and in order to verify the reliability of the model, performing 5 groups of tests, wherein each group of tests is the average absolute percentage error between the prediction result and the actual result of the continuous running of the bus for 1 minute, 3 minutes, 5 minutes, 8 minutes and 10 minutes, and the result is shown in the attached figure 10.
In summary, the invention provides a liquefied natural gas bus tail gas emission prediction method. The invention starts from the angle of the vehicle running state, divides the vehicle running state into the road section running state and the node idling state, respectively models the two states by adopting a CART algorithm and a mean value estimation method, and establishes a bus full-line exhaust emission quality prediction model on the basis of the two states.
Although the preferred embodiments of the present invention have been described in detail, the present invention is not limited to the details of the embodiments, and various equivalent modifications can be made within the technical spirit of the present invention, and the scope of the present invention is also within the scope of the present invention.
It should be noted that the various features described in the above embodiments may be combined in any suitable manner without departing from the scope of the invention. The invention is not described in detail in order to avoid unnecessary repetition.
In addition, any combination of the various embodiments of the present invention is also possible, and the same should be considered as the disclosure of the present invention as long as it does not depart from the spirit of the present invention.
Claims (6)
1. A liquefied natural gas bus tail gas emission prediction method is characterized by comprising the following steps:
step 1: inspecting the emission condition of the bus, determining the bus to be researched, and selecting a proper research time period;
step 2: the method comprises the following steps of utilizing the PEMS and the GPS equipment to collect real-time operation condition data and emission data of the bus, wherein the data collected by the PEMS equipment comprise: CO, CO of vehicles2HC and NOXMass discharge rate; the data collected by the GPS equipment comprises: the position information and the real-time speed of the vehicle are calculated, and the real-time acceleration is calculated;
and step 3: data cleaning and GPS data preprocessing;
and 4, step 4: identifying the idle node position in the running process of the vehicle based on a point density method;
and 5: calculating the Vehicle Specific Power (VSP) parameters of the Vehicle, and carrying out VSP discretization;
step 6: dividing the running state of the vehicle into a road section running state and an idle node state, and respectively predicting the mass emission rate in the two states by applying a CART algorithm and a mean value estimation method;
and 7: based on the mass emission rate prediction model in the step 6, combining the two states, establishing a vehicle full-line tail gas emission real-time prediction model, solving the model, and predicting the emission quality of the vehicle under different running time lengths;
the mean value estimation in step 6 is specifically as follows:
step 65: respectively mixing CO and CO in an idle state2HC and NOxArranging the data in an ascending order, and removing 2.5 percent of the data before and after the data are arranged;
and step 66: calculate the average mass emission rate for each emission:
wherein:
refers to the average mass emission rate of the tail gas type i, N is the total number of idle node data of the tail gas type i,
a value of mass discharge rate for the nth data;
the step 7 is specifically as follows: the emission quality of certain exhaust gas in the time T is equal to the sum of the emission quality of certain exhaust gas in the road driving state and the emission quality of certain exhaust gas in the node idling state, and the calculation formula is as follows:
wherein: x is the number ofsThe variable is 0-1, when the vehicle is in a road section driving state, 1 is taken, otherwise 0 is taken; x is the number ofNThe variable is 0-1, when the vehicle is in a node idling state, 1 is taken, otherwise 0 is taken; t isSMeans the total time T that the vehicle is in the driving state of the road section within the time TNThe total time that the vehicle is in the node idling state within the time T is referred to;
the mass emission rate of the tail gas type i in the road section driving state is indicated;
the mass emission rate of the tail gas type i in a node idling state is indicated; emissioniRefers to the total mass of the exhaust i emitted by the vehicle within the time T.
2. The liquefied natural gas bus tail gas emission prediction method as claimed in claim 1, wherein the real-time acceleration calculation formula of the vehicle in the step 2 is as follows:
ai＝vi-vi-1；
wherein, aiThe unit is m/s which is the acceleration at the ith second2；viThe average speed at the ith second is taken as the unit of m/s;
vi-1the average speed at the i-1 second is in m/s.
3. The liquefied natural gas bus tail gas emission prediction method according to claim 2, wherein the content of the GPS data preprocessing in the step 3 is specifically as follows:
step 31: the data repetition means that a plurality of data are completely the same in continuous time, and the repeated data of the station part is not influenced too much and is reserved; the repeated data of the road section part is directly removed;
step 32: the data loss is that the GPS signal cannot be received within a period of time due to the influence of various external environments, GPS positioning information is obtained, and if the loss time is short, a linear interpolation method is adopted for interpolation; if the loss time is longer, directly removing the part of data;
step 33: the data error is that the measured longitude and latitude data has great deviation from the actual data, and obvious error exists, and the data is directly removed.
4. The liquefied natural gas bus tail gas emission prediction method as claimed in claim 3, wherein the step 4 is as follows:
step 41: arranging the position information of the vehicle according to a time sequence, and calculating the distance L between two adjacent points;
step 42: if the distances L between k-1 adjacent points among the adjacent k (k < ═ 100) points are less than 1m, the points are considered to be possible idle speed nodes, namely bus stops or partial intersections, and the points are added into a list of the idle speed nodes;
step 43: if two groups of possible idle speed nodes exist, according to the time sequence, subtracting the time T2 corresponding to the last point of the former group from the time T1 corresponding to the first point of the latter group to obtain a time difference delta T smaller than T, namely a time difference threshold, and according to the running characteristics of the bus, taking T as 10s, then considering that the two groups of idle speed nodes are both positioned at the same bus stop or intersection; in the case of Δ T > T, if Δ T >0, it is indicated that a secondary stop has occurred; if Δ t < ═ 0, it means that the number of points included in the idle node exceeds k;
step 44: and only one same element in the idle node list is reserved, and all the idle node GPS information is sorted according to time sequence.
5. The liquefied natural gas bus tail gas emission prediction method as claimed in claim 4, wherein the step 5 is as follows:
step 51: the specific power of the motor vehicle represents the mass of each ton of the engine including dead weight and power required to be output, the unit is represented by km/t, various parameters including speed, acceleration, air resistance coefficient, gradient and windward area are contained in a calculation formula of the specific power, and the calculation formula is as follows:
wherein m: the mass of the vehicle and all the people and objects inside the vehicle is kg; v: vehicle speed in m/s;
a: acceleration in m/s2(ii) a Epsilon: a mass coefficient; grade: the slope is the ratio of the slope height to the slope length, and the unit is percent; g: acceleration of gravity in m/s2；CR: a rolling resistance coefficient; cD: a wind resistance coefficient; a: frontal area of vehicle in m2；ρa: air density in kg/m3；
Step 52: in order to facilitate the following research on bus emission, the power parameter needs to be compared for discretization, that is, the VSP is divided according to a certain mode, and the dividing mode is as follows:
6. the liquefied natural gas bus exhaust emission prediction method according to claim 5, wherein the CART algorithm in the step 6 is specifically as follows:
step 61: in order to eliminate the influence of too large or too small data on the modeling, part of the data is required to be removed: corresponding CO and CO of the same VSP2HC and NOXThe mass discharge rate data are respectively arranged in ascending order; according to each group of CO mass emission rate data after ascending arrangement, removing the data of the first 2.5 percent and the data of the second 2.5 percent of the data of the row, namely, retaining the data of the middle 95 percent for establishing a regression model;
step 62, the CART algorithm is used for establishing a regression model of the mass emission rate vehicle specific power parameter, and mainly comprises two parts, namely selection of an optimal cutting point and tree pruning;
(1) selection of optimal cut points: for each type of data set, firstly, a linear model is used for fitting the data set, then, the difference between a real target value and a model predicted value is calculated, the squares of the difference are summed to obtain a required total difference (SSE), and finally, the feature with the minimum total difference is still selected as a branch feature;
wherein, yiAnd
respectively representing an actual value and an estimated value;
(2) pruning the tree: generally, the Pruning of CART comprises Pre-Pruning Pre-Pruning and Post-Pruning;
and step 63: calibrating the allowable error in the model and the minimum sample number parameter of segmentation;
step 64: calculating the determination coefficient R2And evaluating the fitting effect with the average absolute error percentage MAPE, wherein the calculation formula is as follows:
wherein,
is the average of each data set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710972889.3A CN107886188B (en) | 2017-10-18 | 2017-10-18 | Liquefied natural gas bus tail gas emission prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710972889.3A CN107886188B (en) | 2017-10-18 | 2017-10-18 | Liquefied natural gas bus tail gas emission prediction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107886188A CN107886188A (en) | 2018-04-06 |
| CN107886188B true CN107886188B (en) | 2021-12-10 |
Family
ID=61781739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710972889.3A Active CN107886188B (en) | 2017-10-18 | 2017-10-18 | Liquefied natural gas bus tail gas emission prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107886188B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108629450A (en) * | 2018-04-26 | 2018-10-09 | 东南大学 | A kind of liquefied natural gas bus exhaust emissions prediction technique |
| CN109376331A (en) * | 2018-08-22 | 2019-02-22 | 东南大学 | A kind of city bus emission index estimation method promoting regression tree based on gradient |
| CN109086946B (en) * | 2018-09-11 | 2021-09-17 | 东南大学 | Method for predicting emission of polluted gas of conventional energy and new energy public transport vehicle |
| CN111126655A (en) * | 2019-03-05 | 2020-05-08 | 东南大学 | Toll station vehicle emission prediction method based on vehicle specific power and model tree regression |
| 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 |
| CN112113911B (en) * | 2020-08-18 | 2022-04-12 | 北京理工大学 | Remote sensing big data detection method and system for automobile exhaust emission of ignition engine |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103838971A (en) * | 2014-03-12 | 2014-06-04 | 中国航天系统工程有限公司 | Method for computing dynamical traffic energy consumption and emission of urban road networks |
| CN104715605A (en) * | 2015-02-16 | 2015-06-17 | 北京交通大学 | VSP-distribution-based traffic operation data and emission data coupling method and system |
| CN105957348A (en) * | 2016-07-01 | 2016-09-21 | 东南大学 | Urban bus route node emission estimating method based on GIS and PEMS |
-
2017
- 2017-10-18 CN CN201710972889.3A patent/CN107886188B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103838971A (en) * | 2014-03-12 | 2014-06-04 | 中国航天系统工程有限公司 | Method for computing dynamical traffic energy consumption and emission of urban road networks |
| CN104715605A (en) * | 2015-02-16 | 2015-06-17 | 北京交通大学 | VSP-distribution-based traffic operation data and emission data coupling method and system |
| CN105957348A (en) * | 2016-07-01 | 2016-09-21 | 东南大学 | Urban bus route node emission estimating method based on GIS and PEMS |
Non-Patent Citations (2)
| Title |
|---|
| "中观层次路网机动车排放动态量化评价研究";郝艳召;《中国博士学位论文全文数据库》;20110715;全文 * |
| "Effects of Errors on Vehicle Emission Rates from Portable Emissions Measurement Systems";Gurdas S. Sandhu; H. Christopher Frey;《Transportation Research Record》;20130131;第2340卷(第1期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107886188A (en) | 2018-04-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107886188B (en) | Liquefied natural gas bus tail gas emission prediction method | |
| CN110610260B (en) | Driving energy consumption prediction system, method, storage medium and equipment | |
| Apaydin et al. | Emission control with route optimization in solid waste collection process: A case study | |
| CN107563566B (en) | Inter-bus-station operation time interval prediction method based on support vector machine | |
| CN109086946B (en) | Method for predicting emission of polluted gas of conventional energy and new energy public transport vehicle | |
| CN104574967B (en) | A kind of city based on Big Dipper large area road grid traffic cognitive method | |
| CN110836675B (en) | Decision tree-based automatic driving search decision method | |
| CN106156966A (en) | Logistics monitoring method and equipment | |
| CN110738855B (en) | Road traffic flow condition prediction method in data sparse time period | |
| CN110503139A (en) | A kind of urban area tail gas pollution prediction technique | |
| CN104318315A (en) | City road traffic pollutant discharge monitoring early warning system and method | |
| Bousonville et al. | Estimating truck fuel consumption with machine learning using telematics, topology and weather data | |
| CN109493449B (en) | Freight train cargo state estimation method based on freight train GPS track data and high-speed transaction data | |
| CN114819305B (en) | Path planning method based on carbon emission measurement scale | |
| CN109376331A (en) | A kind of city bus emission index estimation method promoting regression tree based on gradient | |
| CN107832910A (en) | A kind of method that evaluation path traffic characteristic influences on carbon monoxide emission concentration | |
| CN115907076A (en) | Mountain city charging load probability prediction system and method | |
| CN117128966B (en) | Vehicle charging path planning method and device based on multi-factor coupling | |
| CN112486962A (en) | Extraction and combination short segment calculation heavy-duty diesel vehicle NOxMethod of discharging | |
| Cunillera et al. | A literature review on train motion model calibration | |
| CN114604140B (en) | Cloud control-based new energy automobile unified temperature regulation system and method | |
| CN109919393B (en) | Charging load prediction method for electric taxi | |
| Hlasny et al. | Optimal fuel consumption for heavy trucks: A review | |
| Zhang et al. | Neural network based vehicle speed prediction for specific urban driving | |
| CN116562486B (en) | Optimization method and device for urban household garbage transportation route |
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 |