WO2023221228A1 - Automobile tail gas concentration testing system based on neural network - Google Patents
Automobile tail gas concentration testing system based on neural network Download PDFInfo
- Publication number
- WO2023221228A1 WO2023221228A1 PCT/CN2022/099945 CN2022099945W WO2023221228A1 WO 2023221228 A1 WO2023221228 A1 WO 2023221228A1 CN 2022099945 W CN2022099945 W CN 2022099945W WO 2023221228 A1 WO2023221228 A1 WO 2023221228A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- gas concentration
- exhaust gas
- circuit
- network
- training
- Prior art date
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 24
- 238000012360 testing method Methods 0.000 title abstract description 7
- 238000009529 body temperature measurement Methods 0.000 claims abstract description 20
- 239000000523 sample Substances 0.000 claims abstract description 19
- 238000012544 monitoring process Methods 0.000 claims abstract description 5
- 238000001514 detection method Methods 0.000 claims description 49
- 239000013598 vector Substances 0.000 claims description 31
- 238000012549 training Methods 0.000 claims description 30
- 238000005457 optimization Methods 0.000 claims description 13
- 238000000034 method Methods 0.000 claims description 12
- 230000008859 change Effects 0.000 claims description 10
- 238000010276 construction Methods 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 9
- 210000002569 neuron Anatomy 0.000 claims description 8
- 238000012937 correction Methods 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 5
- 238000002474 experimental method Methods 0.000 claims description 5
- 230000004044 response Effects 0.000 abstract description 3
- 230000035945 sensitivity Effects 0.000 abstract description 2
- 239000007789 gas Substances 0.000 description 43
- MWUXSHHQAYIFBG-UHFFFAOYSA-N nitrogen oxide Inorganic materials O=[N] MWUXSHHQAYIFBG-UHFFFAOYSA-N 0.000 description 12
- 238000010438 heat treatment Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 6
- 230000002457 bidirectional effect Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 2
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 2
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 2
- 229910002091 carbon monoxide Inorganic materials 0.000 description 2
- 239000003344 environmental pollutant Substances 0.000 description 2
- 229930195733 hydrocarbon Natural products 0.000 description 2
- 150000002430 hydrocarbons Chemical class 0.000 description 2
- 238000002347 injection Methods 0.000 description 2
- 239000007924 injection Substances 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 229910052760 oxygen Inorganic materials 0.000 description 2
- 239000001301 oxygen Substances 0.000 description 2
- 231100000719 pollutant Toxicity 0.000 description 2
- 239000000243 solution Substances 0.000 description 2
- 229910018487 Ni—Cr Inorganic materials 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 239000000956 alloy Substances 0.000 description 1
- 229910045601 alloy Inorganic materials 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 229910002092 carbon dioxide Inorganic materials 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- VNNRSPGTAMTISX-UHFFFAOYSA-N chromium nickel Chemical compound [Cr].[Ni] VNNRSPGTAMTISX-UHFFFAOYSA-N 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 239000011133 lead Substances 0.000 description 1
- 229910000464 lead oxide Inorganic materials 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- XTQHKBHJIVJGKJ-UHFFFAOYSA-N sulfur monoxide Chemical class S=O XTQHKBHJIVJGKJ-UHFFFAOYSA-N 0.000 description 1
- 229910052815 sulfur oxide Inorganic materials 0.000 description 1
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F01—MACHINES OR ENGINES IN GENERAL; ENGINE PLANTS IN GENERAL; STEAM ENGINES
- F01N—GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR MACHINES OR ENGINES IN GENERAL; GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR INTERNAL COMBUSTION ENGINES
- F01N11/00—Monitoring or diagnostic devices for exhaust-gas treatment apparatus, e.g. for catalytic activity
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D23/00—Control of temperature
- G05D23/19—Control of temperature characterised by the use of electric means
- G05D23/1919—Control of temperature characterised by the use of electric means characterised by the type of controller
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0499—Feedforward networks
-
- 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
- G06N3/08—Learning methods
-
- 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
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F01—MACHINES OR ENGINES IN GENERAL; ENGINE PLANTS IN GENERAL; STEAM ENGINES
- F01N—GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR MACHINES OR ENGINES IN GENERAL; GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR INTERNAL COMBUSTION ENGINES
- F01N2560/00—Exhaust systems with means for detecting or measuring exhaust gas components or characteristics
- F01N2560/02—Exhaust systems with means for detecting or measuring exhaust gas components or characteristics the means being an exhaust gas sensor
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F01—MACHINES OR ENGINES IN GENERAL; ENGINE PLANTS IN GENERAL; STEAM ENGINES
- F01N—GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR MACHINES OR ENGINES IN GENERAL; GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR INTERNAL COMBUSTION ENGINES
- F01N2560/00—Exhaust systems with means for detecting or measuring exhaust gas components or characteristics
- F01N2560/06—Exhaust systems with means for detecting or measuring exhaust gas components or characteristics the means being a temperature sensor
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F01—MACHINES OR ENGINES IN GENERAL; ENGINE PLANTS IN GENERAL; STEAM ENGINES
- F01N—GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR MACHINES OR ENGINES IN GENERAL; GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR INTERNAL COMBUSTION ENGINES
- F01N2900/00—Details of electrical control or of the monitoring of the exhaust gas treating apparatus
- F01N2900/04—Methods of control or diagnosing
- F01N2900/0402—Methods of control or diagnosing using adaptive learning
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F01—MACHINES OR ENGINES IN GENERAL; ENGINE PLANTS IN GENERAL; STEAM ENGINES
- F01N—GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR MACHINES OR ENGINES IN GENERAL; GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR INTERNAL COMBUSTION ENGINES
- F01N2900/00—Details of electrical control or of the monitoring of the exhaust gas treating apparatus
- F01N2900/04—Methods of control or diagnosing
- F01N2900/0422—Methods of control or diagnosing measuring the elapsed time
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F01—MACHINES OR ENGINES IN GENERAL; ENGINE PLANTS IN GENERAL; STEAM ENGINES
- F01N—GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR MACHINES OR ENGINES IN GENERAL; GAS-FLOW SILENCERS OR EXHAUST APPARATUS FOR INTERNAL COMBUSTION ENGINES
- F01N2900/00—Details of electrical control or of the monitoring of the exhaust gas treating apparatus
- F01N2900/06—Parameters used for exhaust control or diagnosing
- F01N2900/14—Parameters used for exhaust control or diagnosing said parameters being related to the exhaust gas
- F01N2900/1402—Exhaust gas composition
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/0004—Gaseous mixtures, e.g. polluted air
- G01N33/0006—Calibrating gas analysers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/0004—Gaseous mixtures, e.g. polluted air
- G01N33/0009—General constructional details of gas analysers, e.g. portable test equipment
- G01N33/0027—General constructional details of gas analysers, e.g. portable test equipment concerning the detector
- G01N33/0036—General constructional details of gas analysers, e.g. portable test equipment concerning the detector specially adapted to detect a particular component
- G01N33/0037—NOx
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/0004—Gaseous mixtures, e.g. polluted air
- G01N33/0009—General constructional details of gas analysers, e.g. portable test equipment
- G01N33/0027—General constructional details of gas analysers, e.g. portable test equipment concerning the detector
- G01N33/0036—General constructional details of gas analysers, e.g. portable test equipment concerning the detector specially adapted to detect a particular component
- G01N33/004—CO or CO2
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Definitions
- the invention relates to the technical field of automobile exhaust gas detection, and in particular to a neural network-based automobile exhaust gas concentration detection system.
- Car exhaust contains hundreds of different compounds, among which pollutants include solid suspended particles, carbon monoxide, carbon dioxide, hydrocarbons, nitrogen oxides, lead and sulfur oxides, etc. Car exhaust not only directly harms human health, but also affects the natural ecological environment.
- the exhaust gas concentration value i.e., oxygen concentration, nitrogen oxide concentration
- the exhaust gas concentration value changes extremely weakly during real-time driving of the car, resulting in an extremely weak electrical signal output by the sensor, which cannot be detected in a timely and effective manner; at the same time, due to the In different driving states (that is, when the car is driving, including acceleration, deceleration, etc.), there is a non-linear functional relationship between the sensor output signal and the exhaust gas concentration, resulting in uncontrollable errors in the conversion of electrical signals and concentration, thereby causing concentration detection errors.
- the accuracy is poor. Therefore, the existing high-temperature exhaust gas sensors have problems such as low detection accuracy, poor real-time performance, slow response, and poor stability.
- the purpose of the present invention is to provide a neural network-based vehicle exhaust gas concentration detection system to solve the problems existing in the above background technology and achieve accurate detection of exhaust gas concentration during driving of the vehicle.
- An automobile exhaust gas concentration detection system based on neural network which is characterized by: including a sensor probe, a control module, a temperature control circuit, a timer module, an exhaust gas concentration potential detection circuit, a thermocouple temperature measurement compensation circuit and an analog output circuit;
- the sensor probe is electrically connected to the temperature control circuit, the exhaust gas concentration potential detection circuit, and the thermocouple temperature measurement compensation circuit.
- the temperature control circuit is electrically connected to the timer module.
- the timer module, the exhaust gas concentration potential detection circuit, and the thermocouple measurement circuit are electrically connected.
- the temperature compensation circuit is electrically connected to the control module, and the exhaust gas concentration potential detection circuit and the thermocouple temperature measurement compensation circuit are connected to the ADC interface of the control module, and the DAC interface of the control module is connected to the analog output circuit;
- the control module uses the BP neural network algorithm to convert the collected digital signals into exhaust gas concentration signals to realize the monitoring of exhaust gas concentration, including data preprocessing, data network construction, data network training and data network detection.
- the data preprocessing is to perform a calibration experiment in advance, which is specifically: in a simulated exhaust gas environment, control the change of exhaust gas concentration so that the sensor probe changes from the minimum value V min to the maximum value V max , that is, V min , V 1 , V 2 ,..., V max , where V min ⁇ V max , a total of n values;
- the value of T i takes the value of vector X and vector Y, and these two sets of vectors are normalized and mapped through the above formula;
- the data network is specifically constructed as follows: using a 5-layer neural network topology, that is, the input layer, the first hidden layer, the second hidden layer, the third hidden layer and the output layer; starting from the first hidden layer, the corresponding network layer connections
- the coefficient vectors are: W 1 , W 2 , W 3 , W 4
- the network layer neuron threshold coefficient vectors are: B 1 , B 2 , B 3 , B 4
- the neuron output vectors are: Z 1 , Z 2 , Z 3 , Z 4 ;
- the activation function F(x) is used as the activation function between network layers, where, The activation function F(x) can nonlinearly map the input vector of each layer, so that the data between network layers has a nonlinear relationship, so that the network model has the ability of nonlinear mapping and can fit complex functions.
- the purpose of the relationship is to ensure the accuracy between the sensor output signal and the exhaust gas concentration conversion and avoid uncontrollable errors;
- the data network training is specifically:
- the forward propagation algorithm is completed.
- the output of each layer is specifically:
- the mean square error function is specifically:
- the neural network algorithm is based on the gradient descent strategy and uses the negative gradient direction of the optimization function to adjust the network parameters. Select any parameter v in the neural network to obtain its correction parameter v’, specifically:
- u represents the strength of the correction function
- the preset number of training times is K times, and the preset mean square error function threshold L’;
- the network parameters are corrected once, and the forward propagation algorithm in data network training is performed, and then the value of the mean square error is calculated to see if it meets the preset threshold. If it is met, the parameters in the network are saved and the loop is exited; if If not satisfied, continue training until the corresponding training times are reached;
- the data network detection is to use the trained network parameters and the digital signals detected by the sensor probes, and use the forward propagation algorithm in the data network construction to complete the detection of automobile exhaust concentration.
- the temperature control circuit includes a microcontroller (STM32H7 series), a triac (BTA08), a heating wire and a photoelectric triac driver (MOC3063).
- the microcontroller is internally set with a timer and an output comparison of the timer.
- the mode generates a pulse wave signal with variable frequency and duty cycle, that is, a PWM signal.
- This signal can control the switch drive circuit composed of a triac to adjust the heating wire power; the triac (BTA08) and The heating wire is connected in series in the AC circuit to control the on and off of the heating circuit.
- the photoelectric triac driver (MOC3063) is used to receive the PWM signal and use the PID control algorithm in the PID control module to detect the The temperature is controlled at (750 ⁇ 3)°C.
- the PID control algorithm is specifically:
- p(t) is the PID control feedback quantity
- K p , K i , and K d are the proportional coefficient, integral coefficient, and differential coefficient respectively. Their coefficient values are obtained according to the tuning method of the PID control module
- the PID control algorithm applies the digital signal of the linear combination of proportion, integral and differential to the duty cycle of the PWM signal generated inside the microcontroller to control the heating power; where the deviation signal is the set temperature The difference signal between the value and the actual temperature value.
- the exhaust gas concentration detection system also includes a constant current output circuit.
- the constant current output circuit includes an operational amplifier and a transistor, which is used to improve the anti-interference ability of the circuit and obtain a constant current output of 0 to 10 mA.
- the exhaust gas concentration potential detection circuit uses an amplifier to amplify the sensor probe output signal, thereby effectively avoiding the impact of high-frequency electromagnetic noise on the control module detection circuit.
- thermocouple temperature measurement compensation circuit includes a thermocouple and a temperature sensor integrated chip.
- 80% of the data in the vector (X, Y) are used as training samples and 20% of the data are used as test samples in the data preprocessing.
- the numbers of neurons in the input layer, first hidden layer, second hidden layer, third hidden layer and output layer in the construction of the data network are 1, 20, 20, 10 and 1 respectively.
- u is set to 0.01 in the data network training.
- This application achieves constant control of the sensor operating temperature through the cooperation of the sensor probe, control module, temperature control circuit, timer module, exhaust gas concentration potential detection circuit, thermocouple temperature measurement compensation circuit, and analog output circuit, and effectively solves the problem of The sensor output signal is weak and the nonlinear output data has large fitting errors and poor accuracy.
- the exhaust gas concentration detection system of the present application has high sensitivity, fast response, good stability, strong practicability, easy installation and simple operation, and can be widely used in real-time monitoring of automobile exhaust concentration, thereby effectively reducing environmental pollution and the harm of exhaust gas to the human body.
- Figure 1 is a schematic structural diagram of the detection system in the embodiment of the present application.
- Figure 2 is a temperature control circuit diagram of the detection system in the embodiment of the present application.
- FIG. 3 is a circuit diagram of the exhaust gas concentration potential detection circuit in the embodiment of the present application.
- FIG. 4 is a circuit diagram of the thermocouple temperature measurement compensation circuit in the embodiment of the present application.
- Figure 5 is a constant current output circuit diagram in the embodiment of the present application.
- Figure 6 is a schematic diagram of the network structure in the data network construction of the detection system in the embodiment of the present application.
- 100. Sensor probe 200. Control module; 300. Temperature control circuit; 400. Timer module; 500. Exhaust gas concentration potential detection circuit; 600. Thermocouple temperature measurement compensation circuit; 700. Analog output circuit.
- a neural network-based vehicle exhaust gas concentration detection system is characterized by: including a sensor probe 100, a control module 200, a temperature control circuit 300, a timer module 400, an exhaust gas concentration potential detection circuit 500, Thermocouple temperature measurement compensation circuit 600 and analog output circuit 700; the sensor probe 100 is electrically connected to the temperature control circuit 300, the exhaust gas concentration potential detection circuit 500, and the thermocouple temperature measurement compensation circuit 600.
- the temperature control circuit 300 is electrically connected to the timer module 400.
- thermocouple temperature measurement compensation circuit 600 is electrically connected to the control module 200, and the exhaust gas concentration potential detection circuit 500, the thermocouple temperature measurement compensation circuit 600 are connected to the ADC interface of the control module 200 Connect, the DAC interface of the control module 200 is connected to the analog output circuit 700;
- the temperature control circuit 300 includes a microcontroller (STM32H7 series, such as STM32H7VGT6), a triac (BTA08), a heating wire (using nickel-chromium electrothermal alloy) and a photoelectric triac driver (MOC3063) (as shown in Figure 2).
- the timer is set internally (not the same as the timer module 400), and the output comparison mode of the timer generates a pulse wave signal with variable frequency and duty cycle, that is, a PWM signal. This signal can be controlled by a bidirectional thyristor (BTA08).
- the switch drive circuit realizes the adjustment of the heating wire power; the bidirectional thyristor (BTA08) and the heating wire are connected in series in the AC circuit to control the on and off of the heating circuit.
- the photoelectric bidirectional thyristor driver (MOC3063) is used It receives the PWM signal and uses the PID control algorithm in the PID control module to control the detected temperature at (750 ⁇ 3)°C.
- the PID control algorithm is specifically:
- p(t) is the PID control feedback quantity
- K p , K i , and K d are the proportional coefficient, integral coefficient, and differential coefficient respectively. Their coefficient values are obtained according to the tuning method of the PID control module
- the PID control algorithm applies the digital signal of the deviation signal through the linear combination of proportion, integral and differential to the duty cycle of the PWM signal generated inside the microcontroller to control the heating power; where the deviation signal is the set temperature value and The difference signal of the actual temperature value.
- the exhaust gas concentration potential detection circuit 500 uses an amplifier (such as AD603) to amplify the output signal of the sensor probe 100, thereby effectively avoiding the impact of high-frequency electromagnetic noise on the detection circuit of the control module 200. At the same time, the exhaust concentration potential detection circuit 500 also sets a resistance value of Two resistors of 4.7k, 1M, and multiple capacitors (as shown in Figure 3).
- the thermocouple temperature measurement compensation circuit 600 includes a thermocouple (using a K-type thermocouple) and a temperature sensor integrated chip (AD8495). At the same time, the thermocouple temperature measurement compensation circuit 600 also has four resistors with resistance values of 20k, 22k, 200k, and 4.9k. A reference resistor (as shown in Figure 4).
- the exhaust gas concentration detection system also includes a constant current output circuit.
- the constant current output circuit includes an operational amplifier (LM324), a transistor (9013), and two resistors with a resistance of 1k each (as shown in Figure 5), which are used to improve the circuit resistance. Interference ability, obtain constant current output of 0 ⁇ 10mA.
- the control module 200 uses the BP neural network algorithm to convert the collected digital signals into exhaust gas concentration signals to implement monitoring of exhaust gas concentration, including data preprocessing, data network construction, data network training, and data network detection.
- Data preprocessing is to perform calibration experiments in advance, which is specifically: controlling the changes in exhaust gas concentration in a simulated exhaust gas environment.
- Data preprocessing is performing calibration experiments in advance, which is specifically: controlling changes in exhaust gas concentration in a simulated exhaust gas environment.
- the value of T i takes the value of vector X and vector Y, and these two sets of vectors are normalized and mapped through the above formula;
- the specific construction of the data network is: using a 5-layer neural network topology structure, that is, the input layer, the first hidden layer, the second hidden layer, the third hidden layer and the output layer. Their numbers of neurons are 1, 20, and 20 respectively. , 10, 1; starting from the first hidden layer, the corresponding network layer connection coefficient vectors are: W 1 , W 2 , W 3 , W 4 , and the network layer neuron threshold coefficient vectors are: B 1 , B 2 , B 3 , B 4 , the neuron output vectors are: Z 1 , Z 2 , Z 3 , Z 4 ;
- the activation function F(x) is used as the activation function between network layers, where, The activation function F(x) can nonlinearly map the input vector of each layer, so that the data between network layers has a nonlinear relationship, so that the network model has the ability of nonlinear mapping and can fit complex functions.
- the purpose of the relationship is to ensure the accuracy between the sensor output signal and the exhaust gas concentration conversion and avoid uncontrollable errors;
- the forward propagation algorithm is completed.
- the output of each layer is specifically:
- the mean square error function is specifically:
- the neural network algorithm is based on the gradient descent strategy and uses the negative gradient direction of the optimization function to adjust the network parameters. Select any parameter v in the neural network to obtain its correction parameter v’, specifically:
- u represents the strength of the correction function, u is 0.01;
- the preset training times are K times, and the preset mean square error function thresholds L’, L’ ⁇ 0.01;
- the network parameters are corrected once, and the forward propagation algorithm in data network training is performed, and then the value of the mean square error is calculated to see if it meets the preset threshold. If it is met, the parameters in the network are saved and the loop is exited; if If not satisfied, continue training until the corresponding training times are reached;
- Data network detection is: using the trained network parameters and the digital signals detected by the sensor probe 100, and using the forward propagation algorithm in the data network construction to complete the detection of vehicle exhaust gas concentration.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Automation & Control Theory (AREA)
- Electrochemistry (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- Combined Controls Of Internal Combustion Engines (AREA)
- Investigating Or Analyzing Materials By The Use Of Electric Means (AREA)
Abstract
The present invention provides an automobile tail gas concentration testing system based on a neural network, comprising a sensor probe (100), a control module (200), a temperature control circuit (300), a timer module (400), a tail gas concentration potential testing circuit (500), a thermocouple temperature measurement compensation circuit (600), and an analog quantity output circuit (700). The sensor probe (100) is electrically connected to the temperature control circuit (300), the tail gas concentration potential testing circuit (500), and the thermocouple temperature measurement compensation circuit (600); the temperature control circuit (300) is electrically connected to the timer module (400); the timer module (400), the tail gas concentration potential testing circuit (500), and the thermocouple temperature measurement compensation circuit (600) are electrically connected to the control module (200); the control module (200) is connected to the analog quantity output circuit (700). The system in the present application is high in sensitivity, fast in response, good in stability, and high in practicability, and can be widely applied to real-time monitoring of the automobile tail gas concentration.
Description
本发明涉及汽车尾气检测技术领域,具体涉及一种基于神经网络的汽车尾气浓度检测***。The invention relates to the technical field of automobile exhaust gas detection, and in particular to a neural network-based automobile exhaust gas concentration detection system.
汽车尾气中含有上百种不同的化合物,其中污染物包括固体悬浮颗粒、一氧化碳、二氧化碳、碳氢化合物、氮氧化合物、铅与硫氧化合物等。汽车尾气在直接危害人体身体健康的同时,还会对自然生态环境造成影响。Car exhaust contains hundreds of different compounds, among which pollutants include solid suspended particles, carbon monoxide, carbon dioxide, hydrocarbons, nitrogen oxides, lead and sulfur oxides, etc. Car exhaust not only directly harms human health, but also affects the natural ecological environment.
目前,全球已经开始进行强制实施汽车发动机电喷化、即利用装备电子控制燃油汽车的喷射***,其通过高温尾气传感器实时监测与反馈尾气中的氧浓度与氮氧化物浓度,从而利用控制***通过相应的控制手段从源头上降低一氧化碳、氮氧化物、碳氢化物等典型污染物的排放,达到节能减排、保护环境的目的。然而,除初始启动外、汽车在实时行驶过程中尾气浓度值(即氧浓度、氮氧化合物浓度)改变极其微弱,导致传感器输出的电信号极其微弱,无法及时、有效进行检测;同时由于汽车的行驶状态不同(即汽车行驶过程中包括加速、减速等状态)、传感器输出信号与尾气浓度之间为非线性的函数关系,造成电信号与浓度转换时存在不可控的误差、进而造成浓度检测的精确度差,因此,现有的高温尾气传感器存在检测精度低、实时性差、响应慢、稳定性差的问题。At present, the world has begun to enforce the implementation of electronic injection of automobile engines, that is, the use of electronic equipment to control the injection system of fuel vehicles. It uses high-temperature exhaust gas sensors to monitor and feedback the oxygen concentration and nitrogen oxide concentration in the exhaust gas in real time, thereby using the control system to pass Corresponding control measures reduce emissions of typical pollutants such as carbon monoxide, nitrogen oxides, and hydrocarbons from the source to achieve the goals of energy conservation, emission reduction, and environmental protection. However, except for the initial startup, the exhaust gas concentration value (i.e., oxygen concentration, nitrogen oxide concentration) changes extremely weakly during real-time driving of the car, resulting in an extremely weak electrical signal output by the sensor, which cannot be detected in a timely and effective manner; at the same time, due to the In different driving states (that is, when the car is driving, including acceleration, deceleration, etc.), there is a non-linear functional relationship between the sensor output signal and the exhaust gas concentration, resulting in uncontrollable errors in the conversion of electrical signals and concentration, thereby causing concentration detection errors. The accuracy is poor. Therefore, the existing high-temperature exhaust gas sensors have problems such as low detection accuracy, poor real-time performance, slow response, and poor stability.
发明内容Contents of the invention
针对以上现有技术存在的问题,本发明的目的在于提供一种基于神经网络的汽车尾气浓度检测***,以解决上述背景技术中存在的问题、实现汽车行驶过程中尾气浓度的精确检测。In view of the problems existing in the above prior art, the purpose of the present invention is to provide a neural network-based vehicle exhaust gas concentration detection system to solve the problems existing in the above background technology and achieve accurate detection of exhaust gas concentration during driving of the vehicle.
本发明的目的通过以下技术方案实现:The object of the present invention is achieved through the following technical solutions:
一种基于神经网络的汽车尾气浓度检测***,其特征在于:包括传感器探头、控制模块、温度控制电路、定时器模块、尾气浓度电势检测电路、热电偶测温补偿电路及模拟量输出电路;所述传感器探头与温度控制电路、尾气浓度电势检测电路、热电偶测温补偿电路电连接,所述温度控制电路与定时器模块电连接,所述定时器模块、尾气浓度电势检测电路、热电偶测温补偿电路与控制模块电连接,且尾气浓度电势检测电路、热电偶测温补偿电路与控制模块的ADC接口连接,所述控制模块的DAC接口与模拟量输出电路连接;An automobile exhaust gas concentration detection system based on neural network, which is characterized by: including a sensor probe, a control module, a temperature control circuit, a timer module, an exhaust gas concentration potential detection circuit, a thermocouple temperature measurement compensation circuit and an analog output circuit; The sensor probe is electrically connected to the temperature control circuit, the exhaust gas concentration potential detection circuit, and the thermocouple temperature measurement compensation circuit. The temperature control circuit is electrically connected to the timer module. The timer module, the exhaust gas concentration potential detection circuit, and the thermocouple measurement circuit are electrically connected. The temperature compensation circuit is electrically connected to the control module, and the exhaust gas concentration potential detection circuit and the thermocouple temperature measurement compensation circuit are connected to the ADC interface of the control module, and the DAC interface of the control module is connected to the analog output circuit;
所述控制模块采用BP神经网络算法将采集的数字信号转化为尾气浓度信号、实现对尾气气体浓度的监测,包括数据预处理、数据网络构建、数据网络训练与数据网络检测。The control module uses the BP neural network algorithm to convert the collected digital signals into exhaust gas concentration signals to realize the monitoring of exhaust gas concentration, including data preprocessing, data network construction, data network training and data network detection.
作进一步优化,所述数据预处理为预先进行标定实验,其具体为:在模拟尾气环境下,控制尾气气体浓度变化,使得传感器探头由最小值V
min到最大值V
max进行变化,即V
min,V
1,V
2,……,V
max,其中V
min~V
max、一共n个值;
For further optimization, the data preprocessing is to perform a calibration experiment in advance, which is specifically: in a simulated exhaust gas environment, control the change of exhaust gas concentration so that the sensor probe changes from the minimum value V min to the maximum value V max , that is, V min , V 1 , V 2 ,..., V max , where V min ~ V max , a total of n values;
重复进行上述浓度变化过程m次,并且在变化过程中通过微控制器实时采集尾气浓度电势检测电路输出的模拟信号、记录依次对应的数字序列X
i、其中i∈(0,n×m-1),构成向量X;在同时刻使用专门的浓度检测装置记录尾气气体浓度值,构成数字序列Y
i、其中i∈(0,n×m-1),构成向量Y;然后,计算获得向量X与Y中的最值,分别记为X
min、X
max、Y
min、Y
max;再使用归一法将数据映射到0~1的范围内,从而加快梯度下降时的求解速度与提高模型的训练精度,具体为:
Repeat the above concentration change process m times, and during the change process, the analog signal output by the exhaust gas concentration potential detection circuit is collected in real time through the microcontroller, and the corresponding digital sequence X i is recorded, where i∈(0,n×m-1 ), forming a vector and the maximum value in Y , recorded as X min , Training accuracy, specifically:
式中,T
i值取向量X、向量Y的值,通过上述公式分别将这两组向量作归一化映射;
In the formula, the value of T i takes the value of vector X and vector Y, and these two sets of vectors are normalized and mapped through the above formula;
最后将向量X与Y构成一组2行、N×M-1列的向量、即(X,Y),并将其分成训练样本与测试样本,完成数据预处理;Finally, the vectors
所述数据网络构建具体为:采用5层神经网络拓补结构,即输入层、第一隐藏层、第二隐藏层、第三隐藏层与输出层;从第一隐藏层开始,对应网络层连接系数向量为:W
1、W
2、W
3、W
4,网络层神经元阈值系数向量为:B
1、B
2、B
3、B
4,神经元输出向量为:Z
1、Z
2、Z
3、Z
4;
The data network is specifically constructed as follows: using a 5-layer neural network topology, that is, the input layer, the first hidden layer, the second hidden layer, the third hidden layer and the output layer; starting from the first hidden layer, the corresponding network layer connections The coefficient vectors are: W 1 , W 2 , W 3 , W 4 , the network layer neuron threshold coefficient vectors are: B 1 , B 2 , B 3 , B 4 , and the neuron output vectors are: Z 1 , Z 2 , Z 3 , Z 4 ;
网络层与网络层之间采用F(x)作为激活函数,其中,
激活函数F(x)能够将每一层输入向量进行非线性的映射,使得网络层与网络层之间的数据为非线性关系,从而使网络模型具有非线性映射的能力、实现拟合复杂函数关系的目的,进而确保传感器输出信号与尾气浓度转换之间的精确性、避免出现不可控误差;
F(x) is used as the activation function between network layers, where, The activation function F(x) can nonlinearly map the input vector of each layer, so that the data between network layers has a nonlinear relationship, so that the network model has the ability of nonlinear mapping and can fit complex functions. The purpose of the relationship is to ensure the accuracy between the sensor output signal and the exhaust gas concentration conversion and avoid uncontrollable errors;
所述数据网络训练具体为:The data network training is specifically:
首先根据网络模型,完成正向传播算法,从第一隐藏层开始,每一层的输出具体为:First, according to the network model, the forward propagation algorithm is completed. Starting from the first hidden layer, the output of each layer is specifically:
Z
1=F(X*W
1+B
1);
Z 1 =F(X*W 1 +B 1 );
Z
2=F(Z
1*W
2+B
2);
Z 2 =F(Z 1 *W 2 +B 2 );
Z
3=F(Z
2*W
3+B
3);
Z 3 =F(Z 2 *W 3 +B 3 );
Z
4=F(Z
3*W
4+B
4);
Z 4 =F(Z 3 *W 4 +B 4 );
得到神经网络的预测输出结果向量Z
4;
Obtain the predicted output result vector Z 4 of the neural network;
然后采用均方误差优化网络参数,从而使得误差最小且最为优化,均方误差函数具体为:Then the mean square error is used to optimize the network parameters, so that the error is minimized and optimized. The mean square error function is specifically:
神经网络算法基于梯度下降策略,以优化函数的负梯度方向进行网络参数的调整,选取神经网络中的任意一个参数v,获得其修正参数v’,具体为:The neural network algorithm is based on the gradient descent strategy and uses the negative gradient direction of the optimization function to adjust the network parameters. Select any parameter v in the neural network to obtain its correction parameter v’, specifically:
式中,u表示修正函数的强度;In the formula, u represents the strength of the correction function;
数据网络训练中预设训练次数为K次,且预设均方差函数阈值L’;In data network training, the preset number of training times is K times, and the preset mean square error function threshold L’;
在训练过程中,修正一次网络参数、便进行一次数据网络训练中的正向传播算法,然后计算均方误差的值是否满足预设阈值,若满足,保存网络中的参数、且退出循环;若不满足,则继续训练、直到达到相应的训练次数;During the training process, the network parameters are corrected once, and the forward propagation algorithm in data network training is performed, and then the value of the mean square error is calculated to see if it meets the preset threshold. If it is met, the parameters in the network are saved and the loop is exited; if If not satisfied, continue training until the corresponding training times are reached;
所述数据网络检测为:利用训练完成的网络参数及传感器探头检测到的数字信号,利用数据网络构建中的正向传播算法完成对汽车尾气浓度的检测。The data network detection is to use the trained network parameters and the digital signals detected by the sensor probes, and use the forward propagation algorithm in the data network construction to complete the detection of automobile exhaust concentration.
作进一步优化,所述温度控制电路包括单片机(STM32H7系列)、双向可控硅(BTA08)、加热丝与光电双向可控硅驱动器(MOC3063),所述单片机内部设置定时器、定时器的输出比较模式产生频率、占空比可变的脉冲波信号,即PWM信号,这个信号能够控制由双向可控硅构成的开关驱动电路、实现加热丝功率的调节;所述双向可控硅(BTA08)与加热丝串接在交流回路中、用于控制加热回路的导通与截止,所述光电双向可控硅驱动器(MOC3063)用于接收PWM信号,并运用PID控制模块中PID控制算法将检测到的温度控制在(750±3)℃。For further optimization, the temperature control circuit includes a microcontroller (STM32H7 series), a triac (BTA08), a heating wire and a photoelectric triac driver (MOC3063). The microcontroller is internally set with a timer and an output comparison of the timer. The mode generates a pulse wave signal with variable frequency and duty cycle, that is, a PWM signal. This signal can control the switch drive circuit composed of a triac to adjust the heating wire power; the triac (BTA08) and The heating wire is connected in series in the AC circuit to control the on and off of the heating circuit. The photoelectric triac driver (MOC3063) is used to receive the PWM signal and use the PID control algorithm in the PID control module to detect the The temperature is controlled at (750±3)℃.
作进一步优化,所述PID控制算法具体为:For further optimization, the PID control algorithm is specifically:
式中,p(t)为PID控制反馈量;K
p、K
i、K
d分别为比例系数、积分系数、微分系数,它们的系数值均按照PID控制模块的整定方法得到;
In the formula, p(t) is the PID control feedback quantity; K p , K i , and K d are the proportional coefficient, integral coefficient, and differential coefficient respectively. Their coefficient values are obtained according to the tuning method of the PID control module;
PID控制算法将偏差信号通过比例、积分、微分的线性组合后的数字信号作用到单片机内部产生的PWM信号的占空比上,实现加热 功率进行控制;其中,所述偏差信号为设定的温度值与实际的温度值的差值信号。The PID control algorithm applies the digital signal of the linear combination of proportion, integral and differential to the duty cycle of the PWM signal generated inside the microcontroller to control the heating power; where the deviation signal is the set temperature The difference signal between the value and the actual temperature value.
作进一步优化,所述尾气浓度检测***还包括恒流输出电路,所述恒流输出电路包括运算放大器与三极管,其用于提高电路抗干扰能力、获得0~10mA的恒流输出。For further optimization, the exhaust gas concentration detection system also includes a constant current output circuit. The constant current output circuit includes an operational amplifier and a transistor, which is used to improve the anti-interference ability of the circuit and obtain a constant current output of 0 to 10 mA.
作进一步优化,所述尾气浓度电势检测电路采用放大器对传感器探头输出信号进行放大,从而有效避免高频电磁噪声对控制模块检测电路的影响。For further optimization, the exhaust gas concentration potential detection circuit uses an amplifier to amplify the sensor probe output signal, thereby effectively avoiding the impact of high-frequency electromagnetic noise on the control module detection circuit.
作进一步优化,所述热电偶测温补偿电路包括热电偶与温度传感器集成芯片。For further optimization, the thermocouple temperature measurement compensation circuit includes a thermocouple and a temperature sensor integrated chip.
作进一步优化,所述数据预处理中采用向量(X,Y)中的80%数据作为训练样本、20%数据作为测试样本。For further optimization, 80% of the data in the vector (X, Y) are used as training samples and 20% of the data are used as test samples in the data preprocessing.
作进一步优化,所述数据网络构建中输入层、第一隐藏层、第二隐藏层、第三隐藏层与输出层的神经元个数分别为1、20、20、10、1。For further optimization, the numbers of neurons in the input layer, first hidden layer, second hidden layer, third hidden layer and output layer in the construction of the data network are 1, 20, 20, 10 and 1 respectively.
作进一步优化,所述数据网络训练中u取0.01。For further optimization, u is set to 0.01 in the data network training.
作进一步优化,所述数据网络训练中L'≤0.01。For further optimization, L'≤0.01 in the data network training.
本发明具有如下技术效果:The invention has the following technical effects:
本申请通过传感器探头、控制模块、温度控制电路、定时器模块、尾气浓度电势检测电路、热电偶测温补偿电路、模拟量输出电路的配合,实现了传感器工作温度的恒定控制,同时有效解决了传感器输出信号微弱以及非线性输出的数据拟合误差大、精确性差的问题。本申请尾气浓度检测***灵敏度高、响应快、稳定性好、实用性强,且安装简便、操作简单,能够广泛应用于汽车尾气浓度的实时监测,从而有效减少环境污染以及尾气对于人体的伤害。This application achieves constant control of the sensor operating temperature through the cooperation of the sensor probe, control module, temperature control circuit, timer module, exhaust gas concentration potential detection circuit, thermocouple temperature measurement compensation circuit, and analog output circuit, and effectively solves the problem of The sensor output signal is weak and the nonlinear output data has large fitting errors and poor accuracy. The exhaust gas concentration detection system of the present application has high sensitivity, fast response, good stability, strong practicability, easy installation and simple operation, and can be widely used in real-time monitoring of automobile exhaust concentration, thereby effectively reducing environmental pollution and the harm of exhaust gas to the human body.
图1为本申请实施例中检测***的结构示意图。Figure 1 is a schematic structural diagram of the detection system in the embodiment of the present application.
图2为本申请实施例中检测***的温度控制电路图。Figure 2 is a temperature control circuit diagram of the detection system in the embodiment of the present application.
图3为本申请实施例中尾气浓度电势检测电路图。Figure 3 is a circuit diagram of the exhaust gas concentration potential detection circuit in the embodiment of the present application.
图4为本申请实施例中热电偶测温补偿电路图。Figure 4 is a circuit diagram of the thermocouple temperature measurement compensation circuit in the embodiment of the present application.
图5为本申请实施例中恒流输出电路图。Figure 5 is a constant current output circuit diagram in the embodiment of the present application.
图6为本申请实施例中检测***的数据网络构建中网络结构示意图。Figure 6 is a schematic diagram of the network structure in the data network construction of the detection system in the embodiment of the present application.
其中,100、传感器探头;200、控制模块;300、温度控制电路;400、定时器模块;500、尾气浓度电势检测电路;600、热电偶测温补偿电路;700、模拟量输出电路。Among them, 100. Sensor probe; 200. Control module; 300. Temperature control circuit; 400. Timer module; 500. Exhaust gas concentration potential detection circuit; 600. Thermocouple temperature measurement compensation circuit; 700. Analog output circuit.
下面将结合本发明实施例中的附图,对本发明实施例中的技术方 案进行清楚、完整地描述。The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
实施例:Example:
如图1~6所示,一种基于神经网络的汽车尾气浓度检测***,其特征在于:包括传感器探头100、控制模块200、温度控制电路300、定时器模块400、尾气浓度电势检测电路500、热电偶测温补偿电路600及模拟量输出电路700;传感器探头100与温度控制电路300、尾气浓度电势检测电路500、热电偶测温补偿电路600电连接,温度控制电路300与定时器模块400电连接,定时器模块400、尾气浓度电势检测电路500、热电偶测温补偿电路600与控制模块200电连接,且尾气浓度电势检测电路500、热电偶测温补偿电路600与控制模块200的ADC接口连接,控制模块200的DAC接口与模拟量输出电路700连接;As shown in Figures 1 to 6, a neural network-based vehicle exhaust gas concentration detection system is characterized by: including a sensor probe 100, a control module 200, a temperature control circuit 300, a timer module 400, an exhaust gas concentration potential detection circuit 500, Thermocouple temperature measurement compensation circuit 600 and analog output circuit 700; the sensor probe 100 is electrically connected to the temperature control circuit 300, the exhaust gas concentration potential detection circuit 500, and the thermocouple temperature measurement compensation circuit 600. The temperature control circuit 300 is electrically connected to the timer module 400. Connection, the timer module 400, the exhaust gas concentration potential detection circuit 500, the thermocouple temperature measurement compensation circuit 600 are electrically connected to the control module 200, and the exhaust gas concentration potential detection circuit 500, the thermocouple temperature measurement compensation circuit 600 are connected to the ADC interface of the control module 200 Connect, the DAC interface of the control module 200 is connected to the analog output circuit 700;
温度控制电路300包括单片机(STM32H7系列、例如STM32H7VGT6)、双向可控硅(BTA08)、加热丝(采用镍铬电热合金)与光电双向可控硅驱动器(MOC3063)(如图2所示),单片机内部设置定时器(与定时器模块400不等同)、定时器的输出比较模式产生频率、占空比可变的脉冲波信号,即PWM信号,这个信号能够控制由双向可控硅(BTA08)构成的开关驱动电路、实现加热丝功率的调节;双向可控硅(BTA08)与加热丝串接在交流回路中、用于控制加热回路的导通与截止,光电双向可控硅驱动器(MOC3063)用于接收PWM信号,并运用PID控制模块中的PID控制算法将检测到的温度控制在(750±3)℃。The temperature control circuit 300 includes a microcontroller (STM32H7 series, such as STM32H7VGT6), a triac (BTA08), a heating wire (using nickel-chromium electrothermal alloy) and a photoelectric triac driver (MOC3063) (as shown in Figure 2). The timer is set internally (not the same as the timer module 400), and the output comparison mode of the timer generates a pulse wave signal with variable frequency and duty cycle, that is, a PWM signal. This signal can be controlled by a bidirectional thyristor (BTA08). The switch drive circuit realizes the adjustment of the heating wire power; the bidirectional thyristor (BTA08) and the heating wire are connected in series in the AC circuit to control the on and off of the heating circuit. The photoelectric bidirectional thyristor driver (MOC3063) is used It receives the PWM signal and uses the PID control algorithm in the PID control module to control the detected temperature at (750±3)℃.
PID控制算法具体为:The PID control algorithm is specifically:
式中,p(t)为PID控制反馈量;K
p、K
i、K
d分别为比例系数、积分系数、微分系数,它们的系数值均按照PID控制模块的整定方法得到;
In the formula, p(t) is the PID control feedback quantity; K p , K i , and K d are the proportional coefficient, integral coefficient, and differential coefficient respectively. Their coefficient values are obtained according to the tuning method of the PID control module;
PID控制算法将偏差信号通过比例、积分、微分的线性组合后的数字信号作用到单片机内部产生的PWM信号的占空比上,实现加热功率进行控制;其中,偏差信号为设定的温度值与实际的温度值的差值信号。The PID control algorithm applies the digital signal of the deviation signal through the linear combination of proportion, integral and differential to the duty cycle of the PWM signal generated inside the microcontroller to control the heating power; where the deviation signal is the set temperature value and The difference signal of the actual temperature value.
尾气浓度电势检测电路500采用放大器(如AD603)对传感器探头100输出信号进行放大,从而有效避免高频电磁噪声对控制模块200检测电路的影响,同时尾气浓度电势检测电路500中还设置阻值为4.7k、1M的两个电阻以及多个电容器(如图3所示)。The exhaust gas concentration potential detection circuit 500 uses an amplifier (such as AD603) to amplify the output signal of the sensor probe 100, thereby effectively avoiding the impact of high-frequency electromagnetic noise on the detection circuit of the control module 200. At the same time, the exhaust concentration potential detection circuit 500 also sets a resistance value of Two resistors of 4.7k, 1M, and multiple capacitors (as shown in Figure 3).
热电偶测温补偿电路600包括热电偶(采用K型热电偶)与温度 传感器集成芯片(AD8495),同时热电偶测温补偿电路600中还设置阻值为20k、22k、200k、4.9k的四个基准电阻(如图4所示)。The thermocouple temperature measurement compensation circuit 600 includes a thermocouple (using a K-type thermocouple) and a temperature sensor integrated chip (AD8495). At the same time, the thermocouple temperature measurement compensation circuit 600 also has four resistors with resistance values of 20k, 22k, 200k, and 4.9k. A reference resistor (as shown in Figure 4).
尾气浓度检测***还包括恒流输出电路,恒流输出电路包括运算放大器(LM324)、三极管(9013)以及阻值分别为1k的两个电阻(如图5所示),其用于提高电路抗干扰能力、获得0~10mA的恒流输出。The exhaust gas concentration detection system also includes a constant current output circuit. The constant current output circuit includes an operational amplifier (LM324), a transistor (9013), and two resistors with a resistance of 1k each (as shown in Figure 5), which are used to improve the circuit resistance. Interference ability, obtain constant current output of 0~10mA.
控制模块200采用BP神经网络算法将采集的数字信号转化为尾气浓度信号、实现对尾气气体浓度的监测,包括数据预处理、数据网络构建、数据网络训练与数据网络检测。The control module 200 uses the BP neural network algorithm to convert the collected digital signals into exhaust gas concentration signals to implement monitoring of exhaust gas concentration, including data preprocessing, data network construction, data network training, and data network detection.
数据预处理为预先进行标定实验,其具体为:在模拟尾气环境下,控制尾气气体浓度变化,数据预处理为预先进行标定实验,其具体为:在模拟尾气环境下,控制尾气气体浓度变化,使得传感器探头由最小值V
min到最大值V
max进行变化,即V
min,V
1,V
2,……,V
max,其中V
min~V
max、一共n个值;
Data preprocessing is to perform calibration experiments in advance, which is specifically: controlling the changes in exhaust gas concentration in a simulated exhaust gas environment. Data preprocessing is performing calibration experiments in advance, which is specifically: controlling changes in exhaust gas concentration in a simulated exhaust gas environment. Make the sensor probe change from the minimum value V min to the maximum value V max , that is, V min , V 1 , V 2 ,..., V max , where V min ~ V max , a total of n values;
重复进行上述浓度变化过程m次,并且在变化过程中通过微控制器采集尾气浓度电势检测电路500输出的模拟信号、记录依次对应的数字序列X
i、其中i∈(0,n×m-1),构成向量X;在同时刻使用专门的浓度检测装置记录尾气气体浓度值,构成数字序列Y
i、其中i∈(0,n×m-1),构成向量Y;然后,计算获得向量X与Y中的最值,分别记为X
min、X
max、Y
min、Y
max;再使用归一法将数据映射到0~1的范围内,从而加快梯度下降时的求解速度与提高模型的训练精度,具体为:
The above concentration change process is repeated m times, and during the change process, the analog signal output by the exhaust gas concentration potential detection circuit 500 is collected through the microcontroller, and the corresponding digital sequence X i is recorded, where i∈(0,n×m-1 ), forming a vector and the maximum value in Y , recorded as X min , Training accuracy, specifically:
式中,T
i值取向量X、向量Y的值,通过上述公式分别将这两组向量作归一化映射;
In the formula, the value of T i takes the value of vector X and vector Y, and these two sets of vectors are normalized and mapped through the above formula;
最后将向量X与Y构成一组2行、N×M-1列的向量、即(X,Y),并将其分成训练样本与测试样本,其中80%数据作为训练样本、20%数据作为测试样本,完成数据预处理;Finally, the vectors Test samples and complete data preprocessing;
数据网络构建具体为:采用5层神经网络拓补结构,即输入层、第一隐藏层、第二隐藏层、第三隐藏层与输出层,它们的神经元个数分别为1、20、20、10、1;从第一隐藏层开始,对应网络层连接系数向量为:W
1、W
2、W
3、W
4,网络层神经元阈值系数向量为:B
1、B
2、B
3、B
4,神经元输出向量为:Z
1、Z
2、Z
3、Z
4;
The specific construction of the data network is: using a 5-layer neural network topology structure, that is, the input layer, the first hidden layer, the second hidden layer, the third hidden layer and the output layer. Their numbers of neurons are 1, 20, and 20 respectively. , 10, 1; starting from the first hidden layer, the corresponding network layer connection coefficient vectors are: W 1 , W 2 , W 3 , W 4 , and the network layer neuron threshold coefficient vectors are: B 1 , B 2 , B 3 , B 4 , the neuron output vectors are: Z 1 , Z 2 , Z 3 , Z 4 ;
网络层与网络层之间采用F(x)作为激活函数,其中,
激活函数F(x)能够将每一层输入向量进行非线性的映 射,使得网络层与网络层之间的数据为非线性关系,从而使网络模型具有非线性映射的能力、实现拟合复杂函数关系的目的,进而确保传感器输出信号与尾气浓度转换之间的精确性、避免出现不可控误差;
F(x) is used as the activation function between network layers, where, The activation function F(x) can nonlinearly map the input vector of each layer, so that the data between network layers has a nonlinear relationship, so that the network model has the ability of nonlinear mapping and can fit complex functions. The purpose of the relationship is to ensure the accuracy between the sensor output signal and the exhaust gas concentration conversion and avoid uncontrollable errors;
数据网络训练具体为:The details of data network training are:
首先根据网络模型,完成正向传播算法,从第一隐藏层开始,每一层的输出具体为:First, according to the network model, the forward propagation algorithm is completed. Starting from the first hidden layer, the output of each layer is specifically:
Z
1=F(X*W
1+B
1);
Z 1 =F(X*W 1 +B 1 );
Z
2=F(Z
1*W
2+B
2);
Z 2 =F(Z 1 *W 2 +B 2 );
Z
3=F(Z
2*W
3+B
3);
Z 3 =F(Z 2 *W 3 +B 3 );
Z
4=F(Z
3*W
4+B
4);
Z 4 =F(Z 3 *W 4 +B 4 );
得到神经网络的预测输出结果向量Z
4;
Obtain the predicted output result vector Z 4 of the neural network;
然后采用均方误差优化网络参数,从而使得误差最小且最为优化,均方误差函数具体为:Then the mean square error is used to optimize the network parameters, so that the error is minimized and optimized. The mean square error function is specifically:
神经网络算法基于梯度下降策略,以优化函数的负梯度方向进行网络参数的调整,选取神经网络中的任意一个参数v,获得其修正参数v’,具体为:The neural network algorithm is based on the gradient descent strategy and uses the negative gradient direction of the optimization function to adjust the network parameters. Select any parameter v in the neural network to obtain its correction parameter v’, specifically:
式中,u表示修正函数的强度,u取0.01;In the formula, u represents the strength of the correction function, u is 0.01;
数据网络训练中预设训练次数为K次,且预设均方差函数阈值L’、L'≤0.01;In data network training, the preset training times are K times, and the preset mean square error function thresholds L’, L’≤0.01;
在训练过程中,修正一次网络参数、便进行一次数据网络训练中的正向传播算法,然后计算均方误差的值是否满足预设阈值,若满足,保存网络中的参数、且退出循环;若不满足,则继续训练、直到达到相应的训练次数;During the training process, the network parameters are corrected once, and the forward propagation algorithm in data network training is performed, and then the value of the mean square error is calculated to see if it meets the preset threshold. If it is met, the parameters in the network are saved and the loop is exited; if If not satisfied, continue training until the corresponding training times are reached;
数据网络检测为:利用训练完成的网络参数及传感器探头100检测到的数字信号,利用数据网络构建中的正向传播算法完成对汽车尾气浓度的检测。Data network detection is: using the trained network parameters and the digital signals detected by the sensor probe 100, and using the forward propagation algorithm in the data network construction to complete the detection of vehicle exhaust gas concentration.
Claims (3)
- 一种基于神经网络的汽车尾气浓度检测***,其特征在于:包括传感器探头(100)、控制模块(200)、温度控制电路(300)、定时器模块(400)、尾气浓度电势检测电路(500)、热电偶测温补偿电路(600)及模拟量输出电路(700);所述传感器探头(100)与温度控制电路(300)、尾气浓度电势检测电路(500)、热电偶测温补偿电路(600)电连接,所述温度控制电路(300)与定时器模块(400)电连接,所述定时器模块(400)、尾气浓度电势检测电路(500)、热电偶测温补偿电路(600)与控制模块(200)电连接,且尾气浓度电势检测电路(500)、热电偶测温补偿电路(600)与控制模块(200)的ADC接口连接,所述控制模块(200)的DAC接口与模拟量输出电路(700)连接;A vehicle exhaust gas concentration detection system based on neural network, characterized by: including a sensor probe (100), a control module (200), a temperature control circuit (300), a timer module (400), and an exhaust gas concentration potential detection circuit (500 ), thermocouple temperature measurement compensation circuit (600) and analog output circuit (700); the sensor probe (100) and temperature control circuit (300), exhaust gas concentration potential detection circuit (500), thermocouple temperature measurement compensation circuit (600) is electrically connected, the temperature control circuit (300) is electrically connected with the timer module (400), the timer module (400), the exhaust gas concentration potential detection circuit (500), and the thermocouple temperature measurement compensation circuit (600) ) is electrically connected to the control module (200), and the exhaust gas concentration potential detection circuit (500) and the thermocouple temperature measurement compensation circuit (600) are connected to the ADC interface of the control module (200), and the DAC interface of the control module (200) Connected to the analog output circuit (700);所述控制模块(200)采用BP神经网络算法将采集的数字信号转化为尾气浓度信号、实现对尾气气体浓度的监测,包括数据预处理、数据网络构建、数据网络训练与数据网络检测。The control module (200) uses the BP neural network algorithm to convert the collected digital signals into exhaust gas concentration signals to realize monitoring of exhaust gas concentration, including data preprocessing, data network construction, data network training and data network detection.
- 根据权利要求1所述基于神经网络的汽车尾气浓度检测***,其特征在于:所述数据预处理为预先进行标定实验,其具体为:在模拟尾气环境下,控制尾气气体浓度变化、使得传感器探头(100)输出值从最小值到最大值进行变化,数据预处理为预先进行标定实验,其具体为:在模拟尾气环境下,控制尾气气体浓度变化,使得传感器探头由最小值V min到最大值V max进行变化,即 其中V min~V max、一共n个值; The automobile exhaust gas concentration detection system based on neural network according to claim 1, characterized in that: the data preprocessing is to perform a calibration experiment in advance, specifically: in a simulated exhaust gas environment, control the exhaust gas concentration change so that the sensor probe (100) The output value changes from the minimum value to the maximum value. The data preprocessing is to perform a calibration experiment in advance. The specific steps are: in the simulated exhaust gas environment, control the change of the exhaust gas concentration so that the sensor probe changes from the minimum value V min to the maximum value. V max changes, that is Among them, V min ~ V max , a total of n values;重复进行上述浓度变化过程m次,并且在变化过程中通过微控制器采集尾气浓度电势检测电路(500)输出的模拟信号、记录依次对应的数字序列X i、其中i∈(0,n×m-1),构成向量X;在同时刻使用专门的浓度检测装置记录尾气气体浓度值,构成数字序列Y i、其中i∈(0,n×m-1),构成向量Y;然后,计算获得向量X与Y中的最值,分别记为X min、X max、Y min、Y max;再使用归一法将数据映射到0~1的范围内,从而加快梯度下降时的求解速度与提高模型的训练精度,具体为: Repeat the above concentration change process m times, and during the change process, the microcontroller collects the analog signal output by the exhaust gas concentration potential detection circuit (500) and records the corresponding digital sequence X i , where i∈(0,n×m -1), forming a vector The maximum values in the vectors X and Y are recorded as X min , The training accuracy of the model, specifically:式中,T i值取向量X、向量Y的值,通过上述公式分别将这两组向量作归一化映射; In the formula, the value of T i takes the value of vector X and vector Y, and these two sets of vectors are normalized and mapped through the above formula;最后将向量X与Y构成一组2行、N×M-1列的向量、即(X,Y),并将其分成训练样本与测试样本,完成数据预处理。Finally, the vectors
- 根据权利要求2所述基于神经网络的汽车尾气浓度检测***,其特征在于:所述数据网络构建具体为:采用5层神经网络拓补结构,即输入层、第一隐藏层、第二隐藏层、第三隐藏层与输出层;从第一隐藏层开始,对应网络层连接系数向量为:W 1、W 2、W 3、W 4,网络层神经元阈值系数向量为:B 1、B 2、B 3、B 4,神经元输出向量为:Z 1、Z 2、Z 3、Z 4; The automobile exhaust concentration detection system based on neural network according to claim 2, characterized in that: the data network construction is specifically: adopting a 5-layer neural network topology structure, that is, an input layer, a first hidden layer, and a second hidden layer. , the third hidden layer and the output layer; starting from the first hidden layer, the corresponding network layer connection coefficient vectors are: W 1 , W 2 , W 3 , W 4 , and the network layer neuron threshold coefficient vectors are: B 1 , B 2 , B 3 , B 4 , the neuron output vectors are: Z 1 , Z 2 , Z 3 , Z 4 ;所述数据网络训练具体为:The data network training is specifically:首先根据网络模型,完成正向传播算法,从第一隐藏层开始,每一层的输出具体为:First, according to the network model, the forward propagation algorithm is completed. Starting from the first hidden layer, the output of each layer is specifically:Z 1=F(X*W 1+B 1); Z 1 =F(X*W 1 +B 1 );Z 2=F(Z 1*W 2+B 2); Z 2 =F(Z 1 *W 2 +B 2 );Z 3=F(Z 2*W 3+B 3); Z 3 =F(Z 2 *W 3 +B 3 );Z 4=F(Z 3*W 4+B 4); Z 4 =F(Z 3 *W 4 +B 4 );得到神经网络的预测输出结果向量Z 4; Obtain the predicted output result vector Z 4 of the neural network;然后采用均方误差优化网络参数,从而使得误差最小且最为优化,均方误差函数具体为:Then the mean square error is used to optimize the network parameters, so that the error is minimized and optimized. The mean square error function is specifically:神经网络算法基于梯度下降策略,以优化函数的负梯度方向进行网络参数的调整,选取神经网络中的任意一个参数v,获得其修正参数v’,具体为:The neural network algorithm is based on the gradient descent strategy and uses the negative gradient direction of the optimization function to adjust the network parameters. Select any parameter v in the neural network to obtain its correction parameter v’, specifically:式中,u表示修正函数的强度;In the formula, u represents the strength of the correction function;数据网络训练中预设训练次数为K次,且预设均方差函数阈值L’;In data network training, the preset number of training times is K times, and the preset mean square error function threshold L’;在训练过程中,修正一次网络参数、便进行一次数据网络训练中的正向传播算法,然后计算均方误差的值是否满足预设阈值,若满足,保存网络中的参数、且退出循环;若不满足,则继续训练、直到达到相应的训练次数;During the training process, the network parameters are corrected once, and the forward propagation algorithm in data network training is performed, and then the value of the mean square error is calculated to see if it meets the preset threshold. If it is met, the parameters in the network are saved and the loop is exited; if If not satisfied, continue training until the corresponding training times are reached;所述数据网络检测为:利用训练完成的网络参数及传感器探头(100)检测到的数字信号,利用数据网络构建中的正向传播算法完 成对汽车尾气浓度的检测。The data network detection is: using the trained network parameters and the digital signal detected by the sensor probe (100), and using the forward propagation algorithm in the data network construction to complete the detection of the automobile exhaust concentration.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210535430.8A CN114878639A (en) | 2022-05-17 | 2022-05-17 | Automobile exhaust concentration detection system based on neural network |
CN202210535430.8 | 2022-05-17 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2023221228A1 true WO2023221228A1 (en) | 2023-11-23 |
Family
ID=82675862
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2022/099945 WO2023221228A1 (en) | 2022-05-17 | 2022-06-20 | Automobile tail gas concentration testing system based on neural network |
Country Status (3)
Country | Link |
---|---|
CN (1) | CN114878639A (en) |
LU (1) | LU503532B1 (en) |
WO (1) | WO2023221228A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118032711A (en) * | 2024-04-12 | 2024-05-14 | 华夏天信传感科技(大连)有限公司 | Signal control method and system of laser gas sensor |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102192927A (en) * | 2010-11-05 | 2011-09-21 | 重庆大学 | Air-quality monitoring system based on electronic nose technique, and monitoring method thereof |
US20120048953A1 (en) * | 2009-07-15 | 2012-03-01 | Beihang University | Temperature adjusting device and an intelligent temperature control method for a sand and dust environment testing system |
CN105388937A (en) * | 2015-12-04 | 2016-03-09 | 中国电子科技集团公司第四十八研究所 | Accurate constant-temperature control method and device for gas sensor |
CN112037012A (en) * | 2020-08-14 | 2020-12-04 | 百维金科(上海)信息科技有限公司 | Internet financial credit evaluation method based on PSO-BP neural network |
-
2022
- 2022-05-17 CN CN202210535430.8A patent/CN114878639A/en active Pending
- 2022-06-20 WO PCT/CN2022/099945 patent/WO2023221228A1/en unknown
- 2022-06-20 LU LU503532A patent/LU503532B1/en active IP Right Grant
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120048953A1 (en) * | 2009-07-15 | 2012-03-01 | Beihang University | Temperature adjusting device and an intelligent temperature control method for a sand and dust environment testing system |
CN102192927A (en) * | 2010-11-05 | 2011-09-21 | 重庆大学 | Air-quality monitoring system based on electronic nose technique, and monitoring method thereof |
CN105388937A (en) * | 2015-12-04 | 2016-03-09 | 中国电子科技集团公司第四十八研究所 | Accurate constant-temperature control method and device for gas sensor |
CN112037012A (en) * | 2020-08-14 | 2020-12-04 | 百维金科(上海)信息科技有限公司 | Internet financial credit evaluation method based on PSO-BP neural network |
Non-Patent Citations (2)
Title |
---|
HUANG WEI-JUN; HUA MENG; WU CHEN-HUI: "Design of Automobile Exhaust Detection System based on Improved BP Neural Network", TRANSDUCER AND MICROSYSTEM TECHNOLOGIES, ZHONGGUO DIANZHI KEJI JITUAN GONGSI DI-49 YANJIUSUO, CN, no. 10, 31 October 2018 (2018-10-31), CN , pages 95 - 97, 101, XP009550494, ISSN: 2096-2436, DOI: 10.13873/j.1000-9787(2018)10-0095-03 * |
LIU PING; JIAN JIAWEN; CHEN ZHIYUN: "Design of Detection System for Automobile Exhaust based on Integrated Neural Network", CHINESE JOURNAL OF ENVIRONMENTAL ENGINEERING, CN, vol. 10, no. 4, 30 April 2016 (2016-04-30), CN , pages 1883 - 1887, XP009550493, ISSN: 1673-9108 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118032711A (en) * | 2024-04-12 | 2024-05-14 | 华夏天信传感科技(大连)有限公司 | Signal control method and system of laser gas sensor |
Also Published As
Publication number | Publication date |
---|---|
CN114878639A (en) | 2022-08-09 |
LU503532B1 (en) | 2023-11-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2023221228A1 (en) | Automobile tail gas concentration testing system based on neural network | |
CN106150725B (en) | A kind of broad domain oxygen sensor controller | |
EP2625580B1 (en) | Method for adjusting a temperature of a sensor element | |
CN103016184B (en) | Electronic controlled diesel egr system NOx discharge monitoring method and device | |
US9575030B2 (en) | Digital controller for an oxygen sensor | |
GB2512040A (en) | Using resistance equivalent to estimate heater temperature of an exhaust gas after-treatment component | |
CN110735699B (en) | Nitrogen oxide sensor controller | |
CN105759603A (en) | Voltage transformation circuit control system and method based on automatic optimizing model-free controller | |
CN103604951B (en) | The GES measuring method and its TT&C system of a kind of vehicle speed sensor | |
CN101294890A (en) | Carbon black particulate sensor | |
JP4095138B2 (en) | Method and apparatus for determining the sensitivity of a hydrocarbon sensor for an internal combustion engine | |
CN110865225B (en) | Proportional valve current acquisition method and system and electronic equipment | |
CN112526066B (en) | Gas concentration measuring device and method | |
JP4016814B2 (en) | Oxygen sensor element impedance detector | |
CN102495289A (en) | Test circuit for internal resistance of oxygen sensor of automobile and method for measuring internal resistance of oxygen sensor of automobile | |
CN113250798A (en) | Nitrogen-oxygen sensor | |
Vance et al. | Neural network controller development and implementation for spark ignition engines with high EGR levels | |
CN113608011B (en) | Direct-current high-voltage measuring device with self-calibration function and operation method thereof | |
CN114991920B (en) | Treatment system for nitrogen oxides in automobile exhaust of diesel engine | |
CN203116967U (en) | Air-fuel ratio measuring and control instrument for automobile engine | |
He et al. | Research on DC Electronic Load System Based on BP Neural Network PID Control | |
CN214252181U (en) | Gas concentration measuring device | |
CN108594895B (en) | Electric control system and method for automobile exhaust energy recovery device based on thermoelectric power generation | |
US6476611B2 (en) | Circuit and a method for monitoring and diagnosing an oxygen probe | |
CN114755524B (en) | Vehicle-mounted bidirectional charger measurement and control system applying RBM algorithm model |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22942259 Country of ref document: EP Kind code of ref document: A1 |