CN112986073B - Particulate matter detection method and device and tail gas detection system - Google Patents
Particulate matter detection method and device and tail gas detection system Download PDFInfo
- Publication number
- CN112986073B CN112986073B CN202110514443.2A CN202110514443A CN112986073B CN 112986073 B CN112986073 B CN 112986073B CN 202110514443 A CN202110514443 A CN 202110514443A CN 112986073 B CN112986073 B CN 112986073B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- particulate matter
- detection
- target
- running
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 325
- 239000013618 particulate matter Substances 0.000 title claims abstract description 278
- 238000000034 method Methods 0.000 claims abstract description 70
- 230000008569 process Effects 0.000 claims abstract description 47
- 238000012360 testing method Methods 0.000 claims abstract description 47
- 230000004927 fusion Effects 0.000 claims abstract description 37
- 239000002245 particle Substances 0.000 claims description 137
- 230000001133 acceleration Effects 0.000 claims description 28
- 238000012549 training Methods 0.000 claims description 10
- 238000004891 communication Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 6
- 239000000126 substance Substances 0.000 claims description 6
- 238000010586 diagram Methods 0.000 description 14
- 210000002569 neuron Anatomy 0.000 description 13
- 238000004088 simulation Methods 0.000 description 10
- 238000000149 argon plasma sintering Methods 0.000 description 8
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000009833 condensation Methods 0.000 description 4
- 230000005494 condensation Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- QGZKDVFQNNGYKY-UHFFFAOYSA-N Ammonia Chemical compound N QGZKDVFQNNGYKY-UHFFFAOYSA-N 0.000 description 3
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 3
- 241000282414 Homo sapiens Species 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 229910002092 carbon dioxide Inorganic materials 0.000 description 2
- 238000003912 environmental pollution Methods 0.000 description 2
- 230000005284 excitation Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000002203 pretreatment Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 1
- 241000196324 Embryophyta Species 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- 238000003915 air pollution Methods 0.000 description 1
- 229910021529 ammonia Inorganic materials 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- 229910002091 carbon monoxide Inorganic materials 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 239000010419 fine particle Substances 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 229910000069 nitrogen hydride Inorganic materials 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000013022 venting Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
- G01N15/02—Investigating particle size or size distribution
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M17/00—Testing of vehicles
- G01M17/007—Wheeled or endless-tracked vehicles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
- G01N15/06—Investigating concentration of particle suspensions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/061—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using biological neurons, e.g. biological neurons connected to an integrated circuit
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Analytical Chemistry (AREA)
- Neurology (AREA)
- Pathology (AREA)
- Molecular Biology (AREA)
- Immunology (AREA)
- Artificial Intelligence (AREA)
- Dispersion Chemistry (AREA)
- General Engineering & Computer Science (AREA)
- Biochemistry (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Testing Of Engines (AREA)
- Sampling And Sample Adjustment (AREA)
- Exhaust Gas After Treatment (AREA)
Abstract
The disclosure relates to a particulate matter detection method, a particulate matter detection device and a tail gas detection system, wherein the method comprises the following steps: in the running process of a vehicle to be detected, simultaneously performing particulate matter detection on vehicle exhaust emitted by the vehicle to be detected through a plurality of target sensors assembled in a particulate matter detection device contained in an exhaust detection system; determining the running state of the vehicle to be tested according to the vehicle running information of the vehicle to be tested; and performing data fusion on the particulate matter detection data acquired by the plurality of target sensors according to the running state, the vehicle running information and a pre-trained target detection result prediction model to acquire particulate matter emission data of the vehicle to be detected in the running process. The vehicle exhaust emission detection method has the advantages that the vehicle can be used for simultaneously detecting the particulate matter emission condition of the vehicle through the multiple sensors in the running process of the vehicle, detection data are fused according to the running condition of the vehicle, and the accuracy of the particulate matter emission detection result in the vehicle exhaust emission drive test process is improved.
Description
Technical Field
The disclosure relates to the field of vehicle environmental protection detection, in particular to a particulate matter detection method, a particulate matter detection device and a tail gas detection system.
Background
The emergence of vehicles promotes the development of human civilization, the development of the vehicle industry and the vehicle transportation industry powerfully promotes the development of world economy, and the travel conditions of people are greatly improved. However, in recent years, the vehicle industry worldwide has developed beyond the routine, the vehicle keeping amount has broken through 10 hundred million, and the problem of environmental pollution is more serious with the sharp increase of the vehicle keeping amount. The vehicle exhaust contains more than 200 harmful substances, including CO (carbon monoxide) gas and CO2(carbon dioxide) gas, NOx(oxynitride) gas, NH3(ammonia) and particulate matter are the major pollutants. Among them, the particulate matter that has currently aroused people to regard is divided into two types: PM2.5 (fine particles, where "fine" takes the first meaning of Chinese characters: small in size, as opposed to "coarse") and PM10 (respirable particles), the former having a diameter of no more than 2.5 microns and being one-thirtieth the diameter of human hair, the latter being coarser. Current eu air quality standards define an annual average of PM2.5 of at most 25 micrograms per cubic meter and PM10 of 40 micrograms per cubic meter. The guidance principle of the United nations world health organization suggests: the annual average values for PM2.5 and PM10 are 10 micrograms per cubic meter and 20 micrograms per cubic meter, respectively. Wherein, PM2.5 mainly gathers in the low-rise space about one meter away from the ground, and has great harm to human beings, animals and plants. According to recent research, the contribution rate of partial urban mobile sources to PM2.5 reaches 30% -50%, and the urban mobile sources become main sources of urban air pollution.
In order to solve the problem of environmental pollution caused by motor vehicles, global governments and social circles make active efforts and take various effective measures to continuously strengthen the pollution emission reduction management and control of the motor vehicles. In addition to the exhaust emission data obtained by the conventional hub test, the emission test requirements of various countries also include the regulations for exhaust emission detection, particularly particulate matter emission detection, under the actual driving state of the vehicle. With the gradual update of particulate matter emission standards, the demand for vehicle particulate matter emission detection devices and instruments that are adapted to different vehicle types and different driving conditions is also increasing.
Disclosure of Invention
The invention aims to provide a particulate matter detection method, a particulate matter detection device and an exhaust gas detection system, and aims to solve the technical problems that in the existing vehicle exhaust road test process, the particulate matter detection method is single in means, the detection method is small in application range and poor in detection precision.
In a first aspect, the present disclosure provides a particulate detection method, the method comprising:
in the running process of the vehicle to be detected, detecting particulate matters in vehicle tail gas discharged by the vehicle to be detected through m target sensors assembled in particulate matter detection equipment, wherein m is larger than 1;
determining the running state of the vehicle to be tested according to the vehicle running information of the vehicle to be tested;
and performing data fusion on the particulate matter detection data acquired by the m target sensors according to the running state, the vehicle running information and a pre-trained target detection result prediction model to acquire the particulate matter emission data of the vehicle to be detected in the running process.
Therefore, the particulate matter detection method provided by the disclosure can be used for simultaneously detecting the particulate matters in the vehicle exhaust emitted by the vehicle to be detected through the plurality of target sensors assembled in the particulate matter detection equipment in the running process of the vehicle to be detected; determining the running state of the vehicle to be tested according to the vehicle running information of the vehicle to be tested; according to the running state, the vehicle running information and a pre-trained target detection result prediction model, data fusion is carried out on the particulate matter detection data collected by the plurality of target sensors to obtain particulate matter emission data of the vehicle to be detected in the running process, so that the accuracy of the particulate matter emission detection result in the vehicle exhaust emission road test process is improved.
In one possible implementation manner, the m target sensors have different particle size detection ranges and/or different particle concentration range ranges, and perform data fusion on the particle detection data acquired by the m target sensors according to the driving state, the vehicle driving information, and a pre-trained target detection result prediction model to acquire the particle emission data of the vehicle to be detected in the driving process, including:
determining the weight of each particle detection data in data fusion according to the running state, the vehicle running information and the target detection result prediction model;
and performing data fusion on the m particle detection data according to the weight to obtain the particle emission data.
In one possible implementation, the driving state includes: the method comprises the following steps of determining the weight of each particulate matter detection data in data fusion according to the running state, the vehicle running information and the target detection result prediction model, wherein the weight comprises the following steps:
determining the weight according to the driving state when the vehicle to be tested is determined to be in any one of the cold starting state, the rapid acceleration state, the rapid deceleration state and the high-speed driving state;
and under the condition that the vehicle to be tested is determined not to be in any one of the cold starting state, the rapid acceleration state, the rapid deceleration state and the high-speed driving state, determining the weight according to the vehicle driving information and the target detection result prediction model.
By adopting the technical scheme, the particulate matter detection method provided by the disclosure can distribute different weights to different particulate matter sensors according to a specific driving state, or distribute different weights to different particulate matter sensors when a vehicle is in a non-specific driving state through vehicle driving information and a prediction model, so that the weight distribution of different particulate matter sensors under all driving scenes of the vehicle is realized, and the accuracy of particulate matter emission detection in the vehicle exhaust emission road measurement process is further improved.
In one possible implementation, the determining the weight according to the driving state includes:
determining that the weight corresponding to the particulate matter detection data collected by a first sensor included in the m target sensors is a preset first weight when the vehicle to be detected is in any one of the cold start state, the rapid acceleration state and the rapid deceleration state; the lower limit value in the particle size detection range of the particulate matter corresponding to the first sensor is smaller than a preset first particle size value, and the upper limit value in the particle concentration range corresponding to the first sensor is larger than a preset concentration value;
determining a weight corresponding to the particulate matter detection data collected by each target sensor except the first sensor in the m target sensors to be a preset second weight, wherein the first weight is greater than the second weight; alternatively, the first and second electrodes may be,
under the condition that the vehicle to be detected is in the high-speed running state, determining that the weight corresponding to the particulate matter detection data collected by the second sensor included in the m target sensors is a preset third weight; the upper limit value in the particle size detection range corresponding to the second sensor is greater than a preset second particle size value, the upper limit value in the particle concentration range corresponding to the second sensor is greater than the preset concentration value, and the second particle size value is greater than the first particle size value;
determining a weight corresponding to the particulate matter detection data collected by each target sensor except the second sensor in the m target sensors as a preset fourth weight, wherein the third weight is greater than the fourth weight.
By adopting the technical scheme, the particulate matter detection method provided by the disclosure can distribute proper weight for particulate matter detection data of the particulate matter sensors with different particulate matter particle size detection ranges and/or different particulate matter concentration range ranges according to the emission characteristics of the vehicle in a specific driving state, and improve the accuracy of particulate matter detection of the vehicle in the specific driving state.
In a possible implementation manner, the target detection result prediction model is a detection result prediction model obtained by training a preset prediction model through vehicle running information of a test vehicle and actual particulate matter emission data corresponding to the vehicle running information of the test vehicle, and the test vehicle and the vehicle to be detected have the same vehicle type.
In one possible implementation manner, the determining the weight according to the vehicle driving information and the target detection result prediction model includes:
taking the vehicle running information as the input of the target detection result prediction model to obtain particulate matter emission prediction data output by the target detection result prediction model;
determining the weight based on a difference between each of the particulate matter detection data and the particulate matter emission prediction data.
Under the condition of adopting the technical scheme, the particulate matter detection method can predict the particulate matter emission data of the vehicle according to the actual particulate matter detection data and the vehicle running information of the vehicle of the same type, and further performs weight distribution on the detection result of the particulate matter sensor with different particulate matter particle size detection ranges and/or different particulate matter concentration range ranges through the particulate matter emission prediction data, so that the accuracy of particulate matter detection of the vehicle in the normal running process is improved.
In a possible implementation manner, the determining the running state of the vehicle to be tested according to the vehicle running information of the vehicle to be tested includes:
collecting the vehicle running information from a vehicle control bus of the vehicle to be tested, wherein the vehicle running information comprises: timing information after the vehicle to be tested is started, the running speed of the vehicle to be tested and the running acceleration of the vehicle to be tested;
and determining the running state according to the vehicle running information.
In one possible implementation, the determining the driving state according to the vehicle driving information includes:
determining that the vehicle to be tested is in the cold start state under the condition that the timing information is less than or equal to a preset time length;
determining that the vehicle to be tested is in the rapid acceleration state or the rapid deceleration state under the condition that the running acceleration is out of a preset acceleration threshold range; and the number of the first and second groups,
and under the condition that the running speed is greater than a preset speed threshold value, determining that the vehicle to be tested is in the high-speed running state.
By adopting the technical scheme, the particulate matter detection method provided by the disclosure can accurately identify multiple preset specific driving states through the vehicle driving information acquired from the vehicle control bus, and further perform weight setting of the particulate matter sensor detection result according to the identified specific driving states.
In a possible implementation manner, before the particulate matter detection is performed on the vehicle exhaust emitted by the vehicle to be detected through m target sensors equipped in the particulate matter detection device during the running process of the vehicle to be detected, the method further comprises the following steps;
determining the m target sensors from h candidate particulate matter sensors according to the vehicle type;
outputting prompt information for representing the completion of the assembly of the particulate matter sensor in the case that the m target sensors are determined to be assembled into n sensor grooves arranged in the particulate matter detection device, wherein n is less than or equal to h, and n is greater than or equal to m; and the number of the first and second groups,
and determining a detection result prediction model corresponding to the vehicle type from a plurality of pre-trained detection result prediction models to serve as the target detection result prediction model.
Under the condition of adopting above-mentioned technical scheme, the particulate matter check out test set that this disclosure provided contains detachable alternative particulate matter sensor and a plurality of sensor groove that are used for assembling particulate matter sensor, to the vehicle of different vehicle types, can place different particulate matter sensor combinations in this sensor groove, and then realizes vehicle particulate matter emission according to different particulate matter sensor combinations and detects, improves vehicle particulate matter emission road test equipment's application scope.
In a possible implementation manner, after the performing data fusion on the particulate matter detection data collected by the m target sensors according to the driving state, the vehicle driving information and a pre-trained target detection result prediction model to obtain the particulate matter emission data of the vehicle to be detected in the driving process, the method further includes;
after the acquired particulate matter emission data reach a preset number, retraining the target detection result prediction model through target vehicle running information and the particulate matter emission data of the preset number to acquire an updated target detection result prediction model; wherein the content of the first and second substances,
the target vehicle travel information includes: and acquiring vehicle running information of the vehicle to be detected, which is acquired in the process of acquiring each particulate matter emission data.
By adopting the technical scheme, the particulate matter detection method provided by the disclosure can update the target detection result prediction model in real time in the vehicle drive test process, improve the pertinence of the prediction result of the target detection result prediction model to the current vehicle under test, and further improve the accuracy of particulate matter detection.
In a second aspect, the present disclosure also provides a particulate matter detection device, the device comprising:
the particulate matter detection module is configured to detect particulate matter in vehicle exhaust emitted by the vehicle to be detected through m target sensors assembled in the particulate matter detection equipment during the running process of the vehicle to be detected; wherein m is greater than 1;
the state determination module is configured to determine the running state of the vehicle to be tested according to the vehicle running information of the vehicle to be tested;
and the data fusion module is configured to perform data fusion on the particulate matter detection data acquired by the m target sensors according to the running state, the vehicle running information and a pre-trained target detection result prediction model so as to acquire the particulate matter emission data of the vehicle to be detected in the running process.
In one possible implementation, the apparatus further includes:
a type determination module configured to acquire a vehicle type of the vehicle under test after determining that the particulate matter detection device is mounted on the vehicle under test;
a sensor determination module configured to determine the m target sensors from h candidate particulate matter sensors according to the vehicle type;
an information output module configured to output prompt information for characterizing completion of assembly of the particulate matter sensor in a case where it is determined that the m target sensors are assembled into n sensor slots provided in the particulate matter detecting apparatus, where n is less than or equal to h, and n is greater than or equal to m; and the number of the first and second groups,
and the model determining module is configured to determine a detection result prediction model corresponding to the vehicle type from a plurality of pre-trained detection result prediction models as the target detection result prediction model.
In one possible implementation, the apparatus further includes;
the model updating module is configured to retrain the target detection result prediction model through target vehicle running information and the particulate matter emission data in the preset quantity after the obtained particulate matter emission data reaches the preset quantity, so as to obtain an updated target detection result prediction model; wherein the content of the first and second substances,
the target vehicle travel information includes: and acquiring vehicle running information of the vehicle to be detected, which is acquired in the process of acquiring each particulate matter emission data.
Compared with the prior art, the beneficial effects of the particulate matter detection device provided by the embodiment of the disclosure are the same as the beneficial effects of the particulate matter detection method provided by the first aspect of the disclosure, and are not repeated herein.
In a third aspect, the present disclosure also provides an exhaust gas detection system, including: the device comprises a heat tracing pipeline, an operation terminal, a tail gas pretreatment unit, a control unit, gas detection equipment and particle detection equipment;
the particle detection equipment, the gas detection equipment, the operation terminal and the tail gas pretreatment unit are all in communication connection with the control unit, n sensor grooves are formed in the particle detection equipment and used for detachably placing particle sensors, and n is larger than 1;
the operation terminal is used for sending a control signal to the control unit;
and outputting the prompt information sent by the control unit;
the control unit includes:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to implement the steps of the particulate matter detection method provided by the first aspect of the embodiments of the present disclosure.
Compared with the prior art, the beneficial effects of the tail gas detection system provided by the embodiment of the disclosure are the same as the beneficial effects of the particulate matter detection method provided by the first aspect of the disclosure, and are not repeated herein.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure. In the drawings:
FIG. 1 is a flow chart illustrating a particulate detection method according to an exemplary embodiment;
FIG. 2 is a flow chart of a method of acquiring particulate matter emission data according to the method of FIG. 1;
FIG. 3 is a data flow diagram illustrating a process of determining sensor weights in accordance with an exemplary embodiment;
FIG. 4 is a schematic diagram illustrating a virtual architecture of a neural network predictive model in accordance with an exemplary embodiment;
FIG. 5 is a flow chart of a method of determining a driving condition according to the method shown in FIG. 1;
FIG. 6 is a schematic diagram illustrating the construction of a particulate detection device according to one exemplary embodiment;
FIG. 7 is a flow chart of another particulate matter detection method according to FIG. 1;
FIG. 8 is a flow chart of yet another particulate detection method according to FIG. 7;
FIG. 9 is a block diagram illustrating a particulate matter detection device according to an exemplary embodiment;
FIG. 10 is a block diagram of another particulate matter detection device according to FIG. 9;
FIG. 11 is a block diagram of yet another particulate matter detection device according to FIG. 9;
FIG. 12 is a schematic diagram of an exhaust detection system according to an exemplary embodiment.
Detailed Description
The following detailed description of specific embodiments of the present disclosure is provided in connection with the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present disclosure, are given by way of illustration and explanation only, not limitation.
In the correlation technique of vehicle exhaust emission road test, adopt single particulate matter sensor to carry out particulate matter detection to the vehicle exhaust who gathers usually, and then regard the detected data as the particulate matter emission data of vehicle. However, there is a certain difference between the concentration range (i.e., the range of the particulate matter concentration range) and the particle size range (i.e., the range of the particulate matter particle size detection range) that can be detected by different particulate matter sensors, and there is also a difference between the number concentration (or the mass concentration) of the particulate matter and the particle size in the exhaust gas of the vehicle emitted by different vehicles in different driving phases. The range of the particle concentration range and the particle size detection range of the single particle sensor cannot meet the requirement of reliable detection of the particle concentration and the particle size of different vehicles in different driving stages, and the application range of particle detection is small. Based on under the circumstances, the mode that adopts a plurality of different particulate matter sensors to detect has still appeared among the correlation technique, fuses the testing result according to the mathematical property of a plurality of testing data itself again, and then regard the particulate matter testing data after will fusing as the particulate matter emission data of vehicle. However, on one hand, the method performs undifferentiated fusion on the detection data from different particulate matter sensors, and still does not consider the difference of emission characteristics of different vehicles in different driving stages, so that the accuracy is low, and on the other hand, the amount of sampling data required by the undifferentiated data fusion is extremely large, and an algorithm with high complexity and large calculation amount needs to analyze and fuse a large amount of sampling data, so that data delay is caused, and the real-time performance of detection is further affected.
To this end, the present disclosure provides a particulate matter detection method, apparatus, and exhaust gas detection system, which specifically include:
fig. 1 is a flowchart illustrating a particulate matter detection method according to an exemplary embodiment, as shown in fig. 1, applied to an exhaust gas detection system installed on a vehicle to be tested, in which a particulate matter detection apparatus is provided, the method including:
and 101, in the running process of the vehicle to be detected, detecting the particles of the vehicle exhaust emitted by the vehicle to be detected through m target sensors assembled in the particle detection equipment.
Wherein m is greater than 1.
For example, the exhaust gas detection system may be installed in a trunk of a vehicle to be tested, and an exhaust pipe of the vehicle to be tested is connected to the exhaust gas detection system through an air passage pipe so as to introduce vehicle exhaust gas emitted by the vehicle to be tested into the particulate matter detection device. The above-described m target sensors are mounted in the particulate matter detecting apparatus "in parallel", and vehicle exhaust gas emitted from the vehicle under test is introduced into each target sensor on average, so that each target sensor detects the emitted exhaust gas to an equal degree. The embodiment of the disclosure relates to a scene of a vehicle exhaust emission road test, in the scene, an exhaust detection system needs to be installed on a vehicle to be tested firstly, and after the installation is completed, a tester drives the vehicle to be tested to normally run. In the running process of the vehicle to be detected, each particulate matter sensor (namely, the target sensor) collects the particulate matter detection data once every preset time (for example, 1 second), so that the particulate matter detection equipment can collect m particulate matter detection data every second in the running process of the vehicle to be detected. The particle detection data may be particle Number concentration data (PN for short).
And 102, determining the running state of the vehicle to be tested according to the vehicle running information of the vehicle to be tested.
For example, while step 101 is executed, the exhaust gas detection system may further collect vehicle running information of the vehicle under test from a CAN (Controller Area Network) of the vehicle under test, where the vehicle running information may include timing information, acceleration information, speed information, and the like. The collection cycle of collecting the vehicle running information is consistent with the detection cycle of detecting the particulate matter by each particulate matter sensor. After the vehicle travel information is acquired, the travel state of the vehicle may be determined based on the vehicle travel information. The particulate matter detection data, the running state, and the vehicle running information described above may be stored in an array, and specifically, the array corresponding to each particulate matter sensor for each cycle (for example, each second during the running of the vehicle) may be represented as [ cycle flag, running state flag, particulate matter detection data, [ vehicle running information array ] ].
And 103, performing data fusion on the particulate matter detection data acquired by the m target sensors according to the running state, the vehicle running information and a pre-trained target detection result prediction model to acquire particulate matter emission data of the vehicle to be detected in the running process.
For example, in a period of an actual vehicle drive test process, m target sensors may acquire m pieces of Particulate Matter detection data, the m pieces of Particulate Matter detection data need to be subjected to data fusion, and unique Particulate Matter emission data corresponding to the period is determined, where the Particulate Matter emission data may be PN or Particulate Matter mass concentration data (PM for short) determined according to the PN. It should be noted that, for a common gasoline passenger vehicle, there is a certain difference in the particulate matter emission characteristics of the vehicle at different stages or different states of the vehicle. For example, in a cold start condition, the particulate matter contained in the exhaust gas emitted from the vehicle is small in particle size and large in concentration, while in a high speed condition, the particulate matter contained in the exhaust gas emitted from the vehicle is large in particle size and large in concentration. The m target sensors have different particulate matter detection characteristics, and the driving state, the vehicle driving information, and the target detection result prediction model are used to determine weight distribution in data fusion of the m particulate matter detection data. Specifically, after the travel state is determined, the travel state may be determined once. If the running state of the vehicle to be tested in the current period is a preset specific running state, directly carrying out data fusion according to a weight distribution mode corresponding to the specific running state; if the running state of the vehicle to be detected in the current period is not the specific running state, determining a weight distribution mode adopted by data fusion according to the vehicle running information and the target detection result prediction model. The weight distribution mode can be understood as the corresponding weight of each particle detection data in the data fusion process.
In summary, the particulate matter detection method provided by the embodiment of the disclosure can detect the particulate matter emission condition of the vehicle through multiple sensors simultaneously in the vehicle driving process, and fuse the detection data according to the driving condition of the vehicle, thereby improving the accuracy of the particulate matter emission detection result in the vehicle exhaust emission road test process.
Fig. 2 is a flowchart of a method for acquiring particulate matter emission data according to fig. 1, wherein, as shown in fig. 2, the m target sensors have different particle size detection ranges and/or different particle concentration range, and step 103 may include:
Illustratively, the driving state includes: the cold start state, the rapid acceleration state, the rapid deceleration state, and the high-speed driving state, in step 1031, may include: step a, under the condition that the vehicle to be tested is determined to be in any one of a cold starting state, a rapid acceleration state, a rapid deceleration state and a high-speed driving state, determining the weight according to the driving state; alternatively, step b is a step of determining the weight based on the vehicle travel information and the target detection result prediction model when it is determined that the travel state is not in any of the cold start state, the rapid acceleration state, the rapid deceleration state, and the high speed travel state.
For example, each target sensor has a different particle size detection range and a different particle concentration range, or each target sensor has a different particle size detection range, or each target sensor has a different particle concentration range. The m target sensors at least include: a first sensor and a second sensor; the lower limit value of the particle size detection range of the particles corresponding to the first sensor is smaller than a preset first particle size value, and the upper limit value of the particle concentration range corresponding to the first sensor is larger than a preset concentration value; the upper limit value of the particle size detection range corresponding to the second sensor is larger than a preset second particle size value, the upper limit value of the particle concentration range corresponding to the second sensor is larger than the preset concentration value, and the second preset particle size value is larger than the first preset particle size value.
Illustratively, the first sensor may be a particulate matter sensor based on the DC (diffusion-Charge) principle, with a range of particle concentration ranges of 3000-2 × 107#/cm3The currently used particle size detection range is 23nm to 70,000nm, and the actual particle size range can be widened to 10nm to 70,000 nm. The second sensor can be a light scattering particle sensor or an acoustic particle sensor, and compared with the first sensor, the upper limit value and the lower limit value of the particle size detection range of the particles of the second sensor are higher, and the upper limit value and the lower limit value of the particle concentration range of the particles are also higher. For example, the range of the particle concentration range of the second sensor can be 2 x 104-2×1012#/cm3Particle size detection range of particulate matterThe circumference is 20,000nm to 200,000 nm. The above m target sensors may further include: light scattering sensors and ionization sensors.
Based on this, the step a may include: when the vehicle to be tested is in any one of the cold start state, the rapid acceleration state and the rapid deceleration state, a first weight corresponding to the particulate matter detection data collected by a first sensor and a second weight corresponding to the particulate matter detection data collected by m-1 target sensors except the first sensor in the m target sensors are determined, wherein the first weight is greater than any one of the m-1 second weights. Or, under the condition that the vehicle to be tested is in the high-speed driving state, determining a third weight corresponding to the particulate matter detection data collected by the second sensor and a fourth weight corresponding to the particulate matter detection data collected by m-1 target sensors except the second sensor, wherein the third weight is greater than any fourth weight in m-1 fourth weights.
For example, in the case where the vehicle is in a specific driving state, the weight used for data fusion of the detection data of the above-described respective target sensors is written down. Taking the example that the m target sensors include only the first sensor and the second sensor, and the third sensor and the fourth sensor, when the vehicle under test is in any one of the cold start state, the rapid acceleration state, and the rapid deceleration state, the weights for fusing the detection data of the four target sensors may be respectively: a. b, c and d, wherein a is a first weight, b is a second weight, the sum of a, b, c and d is equal to 1, a is greater than any one of b, c and d, or greater than the sum of b, c and d. The description of the third weight and the fourth weight in the case where the vehicle under test is in the high-speed travel state is similar to the first weight and the second weight described above.
Illustratively, the corresponding weight of each target sensor in a particular driving state is predetermined by a dynamic adaptive planning algorithm. The equipment required in the early test process includes: tail gas output analogue device andm object sensors corresponding to the vehicle type. Wherein, the tail gas output simulation equipment can output mixed gas similar to vehicle tail gas. The particle size content ratio and the particle concentration data of the particles in the mixed gas are controllable. For example, it is possible to set the number concentration of particles in the mixed gas output by the exhaust gas output simulation device to be 2 × 10, wherein 20% of all particles in the mixed gas output by the exhaust gas output simulation device have a particle size smaller than 23nm, and 80% of all particles in the mixed gas output by the exhaust gas output simulation device have a particle size larger than 23nm5#/cm3(namely, the particle size content ratio and the particle number concentration of the vehicle exhaust gas under the cold starting state of the vehicle are equivalent to the simulation of the vehicle exhaust gas under the cold starting state of the vehicle). Specifically, the method for obtaining the weight may include: taking the mixed gas output by the tail gas output simulation equipment as the input of m sensors to obtain the particulate matter detection data detected by each sensor, wherein the initial weight corresponding to each sensor is 1/m; after the particle detection data of the m sensors are subjected to weighted fusion to obtain fusion data, if the difference value between the fusion data and the actual particle concentration data (which can be directly obtained through the set data of the tail gas output simulation equipment) is larger than a preset difference value, the weight corresponding to each sensor is adjusted on the basis of the initial weight according to the difference value between the particle detection data detected by each sensor and the actual particle concentration data; and repeating the steps until the difference value between the fusion data and the actual particulate matter concentration data is smaller than the preset difference value.
Illustratively, fig. 3 is a data flow diagram illustrating a process of determining sensor weights according to an exemplary embodiment, and as shown in fig. 3, the m target sensors only include the above-mentioned DC principle-based particle sensor as a detection data source a and the light scattering particle sensor as a detection data source B, and the exhaust gas output simulation device is a preset gas particle size content ratio and particle concentration data source. The dynamic self-adaptive planning algorithm comprises two trained state echo nets, and aiming at each sensor, the particle size content ratio and the particle concentration of preset particles stored in tail gas output simulation equipment and a plurality of particle detection data collected by the sensor are used as the input of the two state echo nets to obtain particle detection data of an output end. Secondly, the predicted particulate matter emission data is multiplied by weights (W1 and W2, both initial values are 0.5) to obtain two weighted detection results. And then comparing the sum of the two weighted detection results with standard particulate matter concentration data (namely particulate matter concentration data set in the exhaust output simulation equipment) (namely judging errors), and adjusting the W1 and the W2 according to the comparison results until the error between the sum of the two weighted detection results and the standard particulate matter concentration data is minimized, so as to obtain the current W1 and the current W2 as the weights corresponding to the two sensors. The number of neurons in the input layer of each echo state network is 2+ q (wherein "2" is two sets of data of particle size content ratio and particle concentration, and "q" is the number of detected data), and the number of neurons in the output layer of each echo state network is 1. Compared with the hidden layer of a common neural network model, the reserve pool comprises a relatively large number of neurons, the connection relation between the neurons is generated randomly, and the links between the neurons have sparsity and are more suitable for processing continuous data.
Illustratively, based on the dynamic adaptive programming algorithm, for the particulate matter sensor based on the DC principle and the light scattering particulate matter sensor, through experiments, in the case that the vehicle to be tested is in any one of the cold start state, the rapid acceleration state and the rapid deceleration state, the weight corresponding to the particulate matter sensor based on the DC principle is between 0.6 and 0.8, and thus, the weight corresponding to the light scattering particulate matter sensor may be between 0.2 and 0.4; in the case where the vehicle under test is in the high-speed running state, the weight corresponding to the particulate matter sensor based on the DC principle is between 0.1 and 0.3, and thus, the weight corresponding to the light scattering particulate matter sensor may be between 0.7 and 0.9.
For example, in the case that the vehicle is in a non-specific driving state, first, a prediction model is trained based on the passing particulate matter detection data of the vehicles of the same model and the vehicle driving information, and the result is directly predicted through the trained prediction model and the vehicle driving information collected in real time. And secondly, determining the weight corresponding to each target sensor according to the proximity degree of the detection results and the prediction results of the m target sensors. Specifically, the predictive model may be a neural network model, the input training data may include a set of vehicle driving information, such as driving speed, driving acceleration, and engine speed, and the output training data may include manually approved particulate matter emission data corresponding to each set of vehicle driving information. In this case, the target detection result prediction model is a detection result prediction model obtained by training a preset prediction model with actual vehicle travel information of a test vehicle and actual particulate matter emission data corresponding to the vehicle travel information of the test vehicle, the test vehicle having the same vehicle type as the vehicle to be tested.
Taking input training data including driving speed, driving acceleration and engine speed and output training data being manually approved particulate matter emission data as an example, fig. 4 is a schematic diagram illustrating a virtual structure of a neural network prediction model according to an exemplary embodiment, as shown in fig. 4, the number of neurons in an input layer of the prediction model is 3, and the number of neurons in an output layer is 1. The prediction model comprises a hidden layer, and the embodiment of the disclosure adopts a three-layer multi-input single-output neural network comprising one hidden layer to establish the prediction model. Based on the experiment of the particulate matter emission data, in the embodiment of the present disclosure, the implicit layer number empirical formula may be expressed as formula (1):
wherein L is the number of hidden layer neurons, R is the number of input layer neurons, S is the number of output layer neurons, and P is a constant between 1 and 10. The number of neurons can be calculated to be between 3 and 12 according to the above formula, and the number of cryptic neurons selected in the embodiment of the present disclosure is 5. In addition, the neural network shown in fig. 4 generally adopts Sigmoid differentiable functions and linear functions as the excitation functions of the network, and selects the S-type tangent function tansig as the excitation function of the hidden layer neurons.
Based on this, the step b may include: taking the vehicle running information as the input of the target detection result prediction model to obtain particulate matter emission prediction data output by the target detection result prediction model; the weight is determined based on a difference between each of the particulate matter detection data and the particulate matter emission prediction data. Wherein the greater the difference between the particulate matter detection data and the particulate matter emission prediction data, the less the weight corresponding to its target sensor.
And 1032, performing data fusion on the m particle detection data according to the weight to acquire the particle emission data.
For example, after determining the weight corresponding to each target sensor, the particulate matter detection data of the plurality of target sensors may be fused into one particulate matter emission data for display. The data fusion mode is various, the simplest mode is to calculate the weighted average value of three detection data, and specifically, the calculation formula of the final particulate matter emission data can be expressed as the following formula (2):
wherein D is the particulate matter emission data, a, b and c are three particulate matter detection data a, b and c, v is a corresponding weight v for a, u is a corresponding weight b, and i is a corresponding weight c. It is understood that this is a single-point data fusion method, and three detected values in each period are fused, and the fused value at the current moment is calculated in a short time as the particulate matter emission data.
Fig. 5 is a flow chart of a driving state determination method according to fig. 1, and as shown in fig. 5, the step 102 may include:
Wherein the vehicle travel information may include: timing information after the vehicle to be tested is started, the running speed of the vehicle to be tested and the acceleration of the vehicle to be tested.
In step 1022, the driving status is determined according to the vehicle driving information.
Illustratively, this step 1022 may include: determining that the vehicle to be tested is in the cold starting state under the condition that the timing information is less than or equal to the preset time length; determining that the vehicle to be tested is in the rapid acceleration state or the rapid deceleration state under the condition that the running acceleration is out of a preset acceleration threshold range; and determining that the vehicle to be tested is in the high-speed running state under the condition that the running speed is greater than a preset speed threshold value. The preset time period may be set to 200 seconds, and the acceleration threshold range may be set to-10 m/s2To 10m/s2The speed threshold may be set to 120 km/h.
Fig. 6 is a schematic structural diagram of a particle detector according to an exemplary embodiment, and the structure of the particle detector is illustrated in fig. 6 by taking a particle detector 200 including 2 sensor slots as an example. Specifically, as shown in fig. 6, the particulate matter detecting apparatus 200 includes: a flow splitting assembly 201, a flow converging assembly 202, two identically configured sensor slots 203, and a housing 204. Two venting interfaces 205, 206 are provided in the sensor slot 203 labeled in fig. 6. The flow dividing assembly 201 is used for guiding vehicle exhaust introduced by a main gas line pipe of an exhaust detection system to two target sensors arranged in two sensor grooves respectively. For one of the target sensors 207, the target sensor 207 itself includes an air inlet and an air outlet, when the target sensor 207 is placed, the air inlet of the target sensor 207 needs to be connected to the air inlet 205 on the left side of the sensor slot 203, the air outlet of the target sensor 207 needs to be connected to the air inlet 206 on the right side of the sensor slot 203, and after the connection on both sides is completed, it is determined that the target sensor 207 is placed. The merging component 202 is used for merging the vehicle exhaust detected by the two target sensors and finally guiding the merged exhaust to the exhaust outlet of the exhaust detection system.
Fig. 7 is a flowchart of another particulate matter detection method according to fig. 1, and as shown in fig. 7, based on the particulate matter detection apparatus 200 shown in fig. 6, before the step 101, the method further includes:
and 104, acquiring the vehicle type of the vehicle to be tested after determining that the particulate matter detection device is installed on the vehicle to be tested.
And step 105, determining the m target sensors from the h candidate particulate matter sensors according to the vehicle type.
And 106, under the condition that the m target sensors are assembled into n sensor grooves arranged in the particle detection device, outputting prompt information for representing the completion of the assembly of the particle sensors.
Wherein n is less than or equal to h, and n is greater than or equal to m.
And step 107, determining a detection result prediction model corresponding to the vehicle type from a plurality of pre-trained detection result prediction models as the target detection result prediction model.
For example, after the installation of the exhaust gas detection system is completed, the installer may manually input the vehicle type (or driving scene) of the vehicle to be tested, where the vehicle type includes: light vehicles, motorcycles, heavy vehicles, diesel vehicles and off-road vehicles, and the target sensor is determined according to the type of vehicle. The particle monitoring device of the exhaust gas detection system is provided with n sensor grooves, h detachable sensors are arranged aiming at different vehicle types, and n is smaller than or equal to h. The h detachable sensors may include: particle sensors based on the DC principle, condensation nucleus counting sensors, opacity sensors, light scattering sensors, particle ionization sensors and particle acoustic measurement sensors. Aiming at light vehicles and motorcycles, the particle size of the discharged particles is small, the concentration of the particles is low, and the particle sensor based on the DC principle, the condensation nucleus counting sensor and the like which have small lower limit of the particle size detection range and low lower limit of the particle concentration detection range can be preferentially selected from h detachable sensors. For heavy vehicles, diesel vehicles or non-road vehicles, because the particle size of the emitted particles may be very large and the particle concentration may also be very high under some conditions, sensors such as a light scattering sensor and a particle acoustic measurement sensor, which have a large upper limit of the particle size detection range and a high upper limit of the particle concentration detection range, can be added on the basis of the particle sensor and condensation nucleus counting sensor based on the DC principle.
Illustratively, when assembling the target sensor, it is not necessary to fill all sensor slots, and thus n is greater than or equal to m. Each detachable sensor corresponds to an independent number, the corresponding relation among the vehicle type, the sensor number and the detection result prediction model is stored in a control unit of the system in advance, and after the system determines m target sensors according to the vehicle type and the corresponding relation, the number of the target sensor required to be used in the detection can be output through a screen of an operation terminal of the tail gas detection system. The installer can determine the target sensors according to the numbers and install the target sensors into the sensor grooves, and after all the m target sensors are assembled, prompt information for representing the assembling completion of the particulate matter sensors can be output through the screen of the operation terminal to indicate the formal start of detection.
Illustratively, the h alternative sensors to which the particulate matter detecting apparatus 200 corresponds include five types of particulate matter sensors, sensor a, sensor B, sensor C, sensor D, and sensor E. In the case where it is determined that the vehicle to be tested is a light vehicle, it is determined to use the sensors a and D as target sensors required for the test of the exhaust emission road of the light vehicle. The installer can manually place the sensors a and D in the sensor slots. Alternatively, in another possible embodiment, the h alternative sensors are non-detachably connected in parallel with the exhaust port (including subsequent heat tracing pipeline equipment, condensation equipment, and the like) of the vehicle to be tested through the air passage pipe, and finally are converged into the air outlet (the air outlet is also provided with a filtering device in front of the air outlet). After the tail gas detection system is installed and the type of the vehicle is determined, the target sensor can be connected by controlling the on-off of an air passage pipe between each alternative sensor and the exhaust port of the vehicle. For example, for the five particulate matter sensors described above, sensors A and D may be kept in communication, and sensors B, C and E may be disconnected. In addition, the system can also determine the currently used target detection result prediction model according to the vehicle type and the corresponding relation, and it can be understood that the training data of the detection result prediction model is different for different vehicle types.
FIG. 8 is a flow chart of yet another particulate matter detection method according to FIG. 7, as shown in FIG. 8, after step 103, the method may further comprise:
and 108, after the particulate matter emission data reach the preset number, retraining the target detection result prediction model according to the target vehicle running information and the particulate matter emission data of the preset number to obtain an updated target detection result prediction model.
Wherein the target vehicle travel information includes: and acquiring the vehicle running information of the vehicle to be tested, which is acquired in the process of acquiring each particulate matter emission data.
For example, during the driving process, on one hand, the system can acquire real-time particulate matter emission data every other period, and on the other hand, vehicle driving information can be collected and recorded during the process of acquiring the real-time particulate matter emission data. In this way, a set of training data can be formed by the particulate matter emission data obtained after the end of a period and the corresponding set of vehicle driving information. The target detection result prediction model may be updated (or weight-trained) each time a certain amount of training data is accumulated. It will be appreciated that the entire vehicle emission drive test process will typically last one to two hours, with a detection granularity of once per second being sufficient to complete the update of the predictive model.
In summary, the particulate matter detection method provided by the embodiment of the disclosure can detect the particulate matter emission condition of the vehicle through the plurality of sensors in the vehicle driving process, and fuse the particulate matter detection data according to the driving condition of the vehicle, so that the accuracy of the particulate matter detection result in the vehicle exhaust emission road test result is improved and the application range and the detection efficiency of the vehicle particulate matter emission road test device are widened under the condition of ensuring the real-time monitoring.
FIG. 9 is a block diagram illustrating a particulate matter detection device according to an exemplary embodiment, and as shown in FIG. 9, the device 300 includes:
the particulate matter detection module 310 is configured to perform particulate matter detection on vehicle exhaust emitted by the vehicle to be detected simultaneously through m target sensors equipped in the particulate matter detection device during the running process of the vehicle to be detected; wherein m is greater than 1;
a state determination module 320 configured to determine a driving state of the vehicle to be tested according to the vehicle driving information of the vehicle to be tested;
and the data fusion module 330 is configured to perform data fusion on the particulate matter detection data collected by the m target sensors according to the driving state, the vehicle driving information and a pre-trained target detection result prediction model, so as to obtain the particulate matter emission data of the vehicle to be detected in the driving process.
Fig. 10 is a block diagram of another particulate matter detecting apparatus according to fig. 9, and as shown in fig. 10, the apparatus 300 further includes:
a type determination module 340 configured to acquire a vehicle type of the vehicle under test after the particulate matter detection apparatus is mounted to the vehicle under test;
a sensor determination module 350 configured to determine the m target sensors from the h candidate particulate matter sensors according to the vehicle type;
an information output module 360 configured to output prompt information for characterizing completion of assembly of the particulate matter sensor in a case where it is determined that the m target sensors are assembled into n sensor slots provided in the particulate matter detecting apparatus, where n is less than or equal to h, and n is greater than or equal to m; and the number of the first and second groups,
and a model determining module 370, configured to determine a detection result prediction model corresponding to the vehicle type from a plurality of pre-trained detection result prediction models as the target detection result prediction model.
FIG. 11 is a block diagram of yet another particulate matter detection device according to FIG. 9, the device 300 further including, as shown in FIG. 11;
the model updating module 380 is configured to retrain the target detection result prediction model according to the target vehicle driving information and the particulate matter emission data of the preset number after the particulate matter emission data reaches the preset number, so as to obtain an updated target detection result prediction model; wherein the content of the first and second substances,
the target vehicle travel information includes: and acquiring the vehicle running information of the vehicle to be tested, which is acquired in the process of acquiring each particulate matter emission data. The particulate matter emission data is the particulate matter emission data collected in real time by the particulate matter detection module 310, the state determination module 320, and the information output module 360 described above.
In conclusion, the particulate matter detection device that this disclosed embodiment provided can detect the particulate matter circumstances of discharging of vehicle through a plurality of sensors at the vehicle in-process of traveling to go the situation and fuse particulate matter detection data according to the vehicle, under the circumstances of guaranteeing the monitoring real-time, improve the accuracy of particulate matter detection result in the vehicle exhaust emission way survey result, widen vehicle particulate matter emission way survey equipment's application scope and detection efficiency.
Fig. 12 is a schematic diagram illustrating an exhaust gas detection system according to an exemplary embodiment, and as shown in fig. 12, the exhaust gas detection system 400 includes: a heat tracing line 410, an off-gas pretreatment unit 420, a gas detection device 430, a particulate matter detection device 440, a control unit 450, and an operation terminal 460.
Wherein n sensor slots are provided in the particle detection device 440 for detachably placing a particle sensor, wherein n is larger than 1. The particle detection device 440, the gas detection device 430, the operation terminal 460 and the exhaust gas pretreatment unit 420 are all in communication with the control unit 450 by wire or wirelessly. It should be noted that only the communication connection lines between the control unit 450 and the operation terminal 460 are shown in fig. 12, and the communication connection lines between the control unit 450 and the particulate matter detection device 440, the gas detection device 430 and the exhaust gas pretreatment unit 420 are not shown.
Wherein, the control unit 450 includes:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to implement the steps of the particulate matter detection method provided by the embodiments of the present disclosure.
Optionally, the tail gas pretreatment unit 420 may include: a first pre-treatment module for pre-treating vehicle exhaust entering the gas detection unit 430, and a second pre-treatment module for pre-treating vehicle exhaust entering the particulate matter detection unit 440. One end of the exhaust gas pretreatment unit 420 is communicated with the exhaust port 510 of the vehicle to be tested through the heat tracing pipeline 410 and an air passage pipe, and the other end of the exhaust gas pretreatment unit 420 is respectively communicated with the gas detection device 430 and the particulate matter detection device 440 through the air passage pipe;
the heat tracing pipeline 410 is used for heating the vehicle exhaust gas discharged by the vehicle to be tested so as to maintain the temperature of the vehicle exhaust gas within a preset temperature range;
the operation terminal 460 for transmitting a control signal to the control unit 450; and outputs the prompt information sent by the control unit 450.
In conclusion, the exhaust gas detection system provided by the embodiment of the disclosure can detect the particulate matter emission condition of the vehicle through a plurality of sensors in the vehicle driving process, and fuse the particulate matter detection data according to the driving condition of the vehicle, so that the accuracy of the particulate matter detection result in the vehicle exhaust emission road test result is improved under the condition of ensuring the real-time monitoring, and the application range and the detection efficiency of the vehicle particulate matter emission road test device are widened.
The preferred embodiments of the present disclosure are described in detail with reference to the accompanying drawings, however, the present disclosure is not limited to the specific details of the above embodiments, and various simple modifications may be made to the technical solution of the present disclosure within the technical idea of the present disclosure, and these simple modifications all belong to the protection scope of the present disclosure.
It should be noted that, in the foregoing embodiments, various features described in the above embodiments may be combined in any suitable manner, and in order to avoid unnecessary repetition, various combinations that are possible in the present disclosure are not described again.
Claims (13)
1. A particulate matter detection method, characterized in that the method comprises:
in the running process of a vehicle to be detected, detecting particulate matters in vehicle tail gas emitted by the vehicle to be detected through m target sensors assembled in particulate matter detection equipment, wherein m is larger than 1;
determining the running state of the vehicle to be tested according to the vehicle running information of the vehicle to be tested;
performing data fusion on the particulate matter detection data acquired by the m target sensors according to the running state, the vehicle running information and a pre-trained target detection result prediction model to acquire particulate matter emission data of the vehicle to be detected in the running process;
the method for acquiring the particulate matter emission data of the vehicle to be detected in the driving process comprises the following steps of:
determining the weight of each particle detection data in data fusion according to the running state, the vehicle running information and the target detection result prediction model;
and performing data fusion on the m particle detection data according to the weight to obtain the particle emission data.
2. The particulate matter detection method according to claim 1, wherein the running state includes: the method comprises the following steps of determining the weight of each particulate matter detection data in data fusion according to the running state, the vehicle running information and the target detection result prediction model, wherein the weight comprises the following steps:
determining the weight according to the driving state when the vehicle to be tested is determined to be in any one of the cold starting state, the rapid acceleration state, the rapid deceleration state and the high-speed driving state;
and under the condition that the vehicle to be tested is determined not to be in any one of the cold starting state, the rapid acceleration state, the rapid deceleration state and the high-speed driving state, determining the weight according to the vehicle driving information and the target detection result prediction model.
3. The particulate matter detection method according to claim 2, wherein the determining the weight according to the running state includes:
determining that the weight corresponding to the particulate matter detection data collected by a first sensor included in the m target sensors is a preset first weight when the vehicle to be detected is in any one of the cold start state, the rapid acceleration state and the rapid deceleration state; the lower limit value in the particle size detection range of the particulate matter corresponding to the first sensor is smaller than a preset first particle size value, and the upper limit value in the particle concentration range corresponding to the first sensor is larger than a preset concentration value;
determining a weight corresponding to the particulate matter detection data collected by each target sensor except the first sensor in the m target sensors to be a preset second weight, wherein the first weight is greater than the second weight; alternatively, the first and second electrodes may be,
under the condition that the vehicle to be detected is in the high-speed running state, determining that the weight corresponding to the particulate matter detection data collected by the second sensor included in the m target sensors is a preset third weight; the upper limit value in the particle size detection range corresponding to the second sensor is greater than a preset second particle size value, the upper limit value in the particle concentration range corresponding to the second sensor is greater than the preset concentration value, and the second particle size value is greater than the first particle size value;
determining a weight corresponding to the particulate matter detection data collected by each target sensor except the second sensor in the m target sensors as a preset fourth weight, wherein the third weight is greater than the fourth weight.
4. The particulate matter detection method according to claim 2, wherein the target detection result prediction model is a detection result prediction model obtained by training a preset prediction model with vehicle travel information of a test vehicle and actual particulate matter emission data corresponding to the vehicle travel information of the test vehicle, the test vehicle and the vehicle to be tested having the same vehicle type.
5. The particulate matter detection method according to claim 4, wherein the determining the weight based on the vehicle travel information and the target detection result prediction model includes:
taking the vehicle running information as the input of the target detection result prediction model to obtain particulate matter emission prediction data output by the target detection result prediction model;
determining the weight based on a difference between each of the particulate matter detection data and the particulate matter emission prediction data.
6. The particulate matter detection method according to claim 2, wherein the determining of the running state of the vehicle under test from the vehicle running information of the vehicle under test includes:
collecting the vehicle running information from a vehicle control bus of the vehicle to be tested, wherein the vehicle running information comprises: timing information after the vehicle to be tested is started, the running speed of the vehicle to be tested and the running acceleration of the vehicle to be tested;
and determining the running state according to the vehicle running information.
7. The particulate matter detection method according to claim 6, wherein the determining the running state from the vehicle running information includes:
determining that the vehicle to be tested is in the cold start state under the condition that the timing information is less than or equal to a preset time length;
determining that the vehicle to be tested is in the rapid acceleration state or the rapid deceleration state under the condition that the running acceleration is out of a preset acceleration threshold range; and the number of the first and second groups,
and under the condition that the running speed is greater than a preset speed threshold value, determining that the vehicle to be tested is in the high-speed running state.
8. The particulate matter detection method according to any one of claims 1 to 7, characterized by further comprising, before the particulate matter detection of vehicle exhaust emitted from the vehicle under test by m target sensors equipped in the particulate matter detection apparatus;
after determining that the particulate matter detection device is mounted on the vehicle to be detected, acquiring the vehicle type of the vehicle to be detected;
determining the m target sensors from h candidate particulate matter sensors according to the vehicle type;
outputting prompt information for representing the completion of the assembly of the particulate matter sensor in the case that the m target sensors are determined to be assembled into n sensor grooves arranged in the particulate matter detection device, wherein n is less than or equal to h, and n is greater than or equal to m; and the number of the first and second groups,
and determining a detection result prediction model corresponding to the vehicle type from a plurality of pre-trained detection result prediction models to serve as the target detection result prediction model.
9. The particulate matter detection method according to any one of claims 1 to 7, wherein after the performing data fusion on the particulate matter detection data collected by the m target sensors according to the driving state, the vehicle driving information and a pre-trained target detection result prediction model to obtain the particulate matter emission data of the vehicle to be detected in the driving process, the method further comprises;
after the acquired particulate matter emission data reach a preset number, retraining the target detection result prediction model through target vehicle running information and the particulate matter emission data of the preset number to acquire an updated target detection result prediction model; wherein the content of the first and second substances,
the target vehicle travel information includes: and acquiring vehicle running information of the vehicle to be detected, which is acquired in the process of acquiring each particulate matter emission data.
10. A particulate matter detecting device, characterized in that the device comprises:
the device comprises a particulate matter detection module, a particulate matter detection module and a control module, wherein the particulate matter detection module is configured to detect particulate matters in vehicle exhaust emitted by a vehicle to be detected through m target sensors assembled in particulate matter detection equipment in the running process of the vehicle to be detected, and m is larger than 1;
the state determination module is configured to determine the running state of the vehicle to be tested according to the vehicle running information of the vehicle to be tested;
the data fusion module is configured to perform data fusion on the particulate matter detection data acquired by the m target sensors according to the running state, the vehicle running information and a pre-trained target detection result prediction model so as to acquire particulate matter emission data of the vehicle to be detected in the running process;
wherein the m target sensors have different particle size detection ranges and/or different particle concentration range, and the data fusion module is configured to:
determining the weight of each particle detection data in data fusion according to the running state, the vehicle running information and the target detection result prediction model;
and performing data fusion on the m particle detection data according to the weight to obtain the particle emission data.
11. The particulate matter detecting device according to claim 10, characterized by further comprising:
a type determination module configured to acquire a vehicle type of the vehicle under test after determining that the particulate matter detection device is mounted on the vehicle under test;
a sensor determination module configured to determine the m target sensors from h candidate particulate matter sensors according to the vehicle type;
an information output module configured to output prompt information for characterizing completion of assembly of the particulate matter sensor in a case where it is determined that the m target sensors are assembled into n sensor slots provided in the particulate matter detecting apparatus, where n is less than or equal to h, and n is greater than or equal to m; and the number of the first and second groups,
and the model determining module is configured to determine a detection result prediction model corresponding to the vehicle type from a plurality of pre-trained detection result prediction models as the target detection result prediction model.
12. The particulate matter detection device according to claim 10 or 11, characterized by further comprising;
the model updating module is configured to retrain the target detection result prediction model through target vehicle running information and the particulate matter emission data in the preset quantity after the obtained particulate matter emission data reaches the preset quantity, so as to obtain an updated target detection result prediction model; wherein the content of the first and second substances,
the target vehicle travel information includes: and acquiring vehicle running information of the vehicle to be detected, which is acquired in the process of acquiring each particulate matter emission data.
13. An exhaust gas detection system, comprising: the device comprises a heat tracing pipeline, an operation terminal, a tail gas pretreatment unit, a control unit, gas detection equipment and particle detection equipment;
one end of the tail gas pretreatment unit is communicated with an exhaust port of a vehicle to be detected through the heat tracing pipeline and the air passage pipe, and the other end of the tail gas pretreatment unit is respectively communicated with the gas detection equipment and the particulate matter detection equipment through the air passage pipe;
the particle detection equipment, the gas detection equipment, the operation terminal and the tail gas pretreatment unit are all in communication connection with the control unit, n sensor grooves are formed in the particle detection equipment and used for detachably placing particle sensors, and n is larger than 1;
the operation terminal is used for sending a control signal to the control unit;
and outputting the prompt information sent by the control unit;
the control unit includes:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to implement the steps of the particulate detection method of any one of claims 1 to 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110514443.2A CN112986073B (en) | 2021-05-12 | 2021-05-12 | Particulate matter detection method and device and tail gas detection system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110514443.2A CN112986073B (en) | 2021-05-12 | 2021-05-12 | Particulate matter detection method and device and tail gas detection system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112986073A CN112986073A (en) | 2021-06-18 |
CN112986073B true CN112986073B (en) | 2021-08-17 |
Family
ID=76337562
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110514443.2A Active CN112986073B (en) | 2021-05-12 | 2021-05-12 | Particulate matter detection method and device and tail gas detection system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112986073B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114167002A (en) * | 2021-12-08 | 2022-03-11 | 西安多普多信息科技有限公司 | Method and device for rapidly detecting pollutants |
CN114964824B (en) * | 2022-08-01 | 2022-11-15 | 北京复兰环保科技有限公司 | Vehicle aftertreatment unit testing method, device and system |
CN115691180B (en) * | 2022-10-14 | 2023-12-12 | 东莞理工学院 | Method and device for monitoring particulate matters discharged on basis of vehicle tail gas |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008032548A (en) * | 2006-07-28 | 2008-02-14 | Shimadzu Corp | Light scattering detection device |
TW201504625A (en) * | 2013-07-29 | 2015-02-01 | Weier Well Ltd Company | On-board dynamic exhaust air pollution analysis device and inspection method thereof |
CN104501862A (en) * | 2014-12-09 | 2015-04-08 | 清华大学 | Vehicle-mounted emission testing system of non-road mobile machinery |
CN105551240A (en) * | 2015-12-08 | 2016-05-04 | 东南大学 | Automobile exhaust discharge determination method based on different road conditions |
CN108535158A (en) * | 2018-03-27 | 2018-09-14 | 北京理工大学 | A kind of sound-electric compound dust concentration detecting system and detection method |
EP3382372A1 (en) * | 2017-03-27 | 2018-10-03 | Koninklijke Philips N.V. | Optical particle sensor and sensing method |
CN108872039A (en) * | 2018-09-19 | 2018-11-23 | 中煤科工集团重庆研究院有限公司 | A kind of light scattering flue gas particles on-line checking adaptive calibration method |
CN111610039A (en) * | 2020-06-02 | 2020-09-01 | 中国汽车工程研究院股份有限公司 | Passenger vehicle PM test system and method |
CN112433026A (en) * | 2020-10-12 | 2021-03-02 | 同济大学 | Emission test and evaluation method for non-road mobile machinery |
-
2021
- 2021-05-12 CN CN202110514443.2A patent/CN112986073B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008032548A (en) * | 2006-07-28 | 2008-02-14 | Shimadzu Corp | Light scattering detection device |
TW201504625A (en) * | 2013-07-29 | 2015-02-01 | Weier Well Ltd Company | On-board dynamic exhaust air pollution analysis device and inspection method thereof |
CN104501862A (en) * | 2014-12-09 | 2015-04-08 | 清华大学 | Vehicle-mounted emission testing system of non-road mobile machinery |
CN105551240A (en) * | 2015-12-08 | 2016-05-04 | 东南大学 | Automobile exhaust discharge determination method based on different road conditions |
EP3382372A1 (en) * | 2017-03-27 | 2018-10-03 | Koninklijke Philips N.V. | Optical particle sensor and sensing method |
CN108535158A (en) * | 2018-03-27 | 2018-09-14 | 北京理工大学 | A kind of sound-electric compound dust concentration detecting system and detection method |
CN108872039A (en) * | 2018-09-19 | 2018-11-23 | 中煤科工集团重庆研究院有限公司 | A kind of light scattering flue gas particles on-line checking adaptive calibration method |
CN111610039A (en) * | 2020-06-02 | 2020-09-01 | 中国汽车工程研究院股份有限公司 | Passenger vehicle PM test system and method |
CN112433026A (en) * | 2020-10-12 | 2021-03-02 | 同济大学 | Emission test and evaluation method for non-road mobile machinery |
Non-Patent Citations (2)
Title |
---|
国I~国III重型柴油车尾气PM2.5及其碳质组分的排放特征;何立强等;《环境科学学报》;20150331;第35卷(第3期);第657页第1节至第660页第3节 * |
柴油机颗粒物实时诊断传感器技术研究进展;资新运等;《传感器与微系统》;20181231;第37卷(第7期);第7-11页 * |
Also Published As
Publication number | Publication date |
---|---|
CN112986073A (en) | 2021-06-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112986073B (en) | Particulate matter detection method and device and tail gas detection system | |
Montazeri-Gh et al. | Development of car drive cycle for simulation of emissions and fuel economy | |
US10583792B2 (en) | System for evaluating and/or optimizing the operating behavior of a vehicle | |
Donateo et al. | Building a cycle for real driving emissions | |
CN105806627A (en) | Vehicle-mounted emission testing device and testing method thereof | |
CN107368913B (en) | Oil consumption prediction method based on least square support vector machine | |
CN106596135B (en) | A kind of pure electric automobile actual travel energy consumption testing, evaluation and Forecasting Methodology | |
CN111806239A (en) | Electric vehicle endurance mileage measuring method, electronic device, and storage medium | |
CN107577910B (en) | Vehicle exhaust concentration inversion method based on deep neural network | |
CN105181343B (en) | Suitable for the simple vehicle-mounted emission test method of the heavy-duty car equipped with OBD | |
CN106650287A (en) | Motor vehicle exhaust emission factor estimation method based on MLP neural network | |
CN112964476B (en) | On-line prediction method for actual road particulate matter emission of heavy-duty diesel vehicle | |
CN111639683B (en) | Method for evaluating nitrogen oxide emission grade of diesel vehicle based on OBD technology | |
CN110243384A (en) | Determination method, apparatus, equipment and the medium of actual travel emission testing route | |
CN111339638B (en) | Automobile driving condition construction method based on existing data | |
KR102303077B1 (en) | Method for reducing the exhaust gas emission of a drive system of a vehicle with combustion engine | |
CN113657651A (en) | Diesel vehicle emission prediction method, medium and equipment based on deep migration learning | |
CN104865346A (en) | Method and apparatus for verification of data acquired by vehicle tail gas detection system | |
Mafi et al. | Developing local driving cycle for accurate vehicular CO2 monitoring: A case study of Tehran | |
CN116992240B (en) | Vehicle emission remote evaluation method, device and storage medium | |
CN111896264B (en) | Method and device for generating test working condition of range extender engine and electronic equipment | |
CN115964905B (en) | Method and medium for constructing emission test working conditions of mountain roads in highland | |
Liu et al. | Construction of driving behavior scoring model based on obd terminal data analysis | |
CN114840916A (en) | Construction method for testing working condition of thermal damage performance of parts of whole vehicle and comprehensive judgment method for influence degree of thermal damage on service life of whole vehicle | |
CN113671386A (en) | Method and device for analyzing durability of hydrogen fuel cell |
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 |