CN107472038B - Hybrid electric vehicle energy management method based on HCCI engine - Google Patents

Hybrid electric vehicle energy management method based on HCCI engine Download PDF

Info

Publication number
CN107472038B
CN107472038B CN201710601345.6A CN201710601345A CN107472038B CN 107472038 B CN107472038 B CN 107472038B CN 201710601345 A CN201710601345 A CN 201710601345A CN 107472038 B CN107472038 B CN 107472038B
Authority
CN
China
Prior art keywords
layer
engine
output
hcci engine
neuron
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
Application number
CN201710601345.6A
Other languages
Chinese (zh)
Other versions
CN107472038A (en
Inventor
郑太雄
侯晓康
杨新琴
杨斌
何招
褚良宇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing University of Post and Telecommunications
Original Assignee
Chongqing University of Post and Telecommunications
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Chongqing University of Post and Telecommunications filed Critical Chongqing University of Post and Telecommunications
Priority to CN201710601345.6A priority Critical patent/CN107472038B/en
Publication of CN107472038A publication Critical patent/CN107472038A/en
Application granted granted Critical
Publication of CN107472038B publication Critical patent/CN107472038B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
    • B60L50/00—Electric propulsion with power supplied within the vehicle
    • B60L50/10—Electric propulsion with power supplied within the vehicle using propulsion power supplied by engine-driven generators, e.g. generators driven by combustion engines
    • B60L50/15—Electric propulsion with power supplied within the vehicle using propulsion power supplied by engine-driven generators, e.g. generators driven by combustion engines with additional electric power supply
    • 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
    • Y02T10/60—Other road transportation technologies with climate change mitigation effect
    • Y02T10/70—Energy storage systems for electromobility, e.g. batteries
    • 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
    • Y02T10/60—Other road transportation technologies with climate change mitigation effect
    • Y02T10/7072—Electromobility specific charging systems or methods for batteries, ultracapacitors, supercapacitors or double-layer capacitors

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Electrical Control Of Ignition Timing (AREA)
  • Combined Controls Of Internal Combustion Engines (AREA)

Abstract

The invention discloses a hybrid power energy management method based on an HCCI engine, and relates to the field of new energy automobiles. The present invention utilizes an HCCI engine as a power source for a hybrid vehicle. Considering that the ignition timing of the HCCI engine is not measurable, the method takes the opening and closing timing of an intake valve and an exhaust valve, the rotating speed of the engine, the temperature of an intake manifold and the fuel injection quantity as input, and builds a neural network to predict the ignition timing of the HCCI engine; the opening and closing time of an intake valve and an exhaust valve of the engine is controlled by using a variable valve timing technology, so that the exhaust gas recompression of the HCCI engine is realized, and the mixed gas is subjected to compression ignition; and then the waste gas of the HCCI engine is introduced into the Stirling engine, and the Stirling engine is utilized to do work and also charge the power battery, so that the energy of the fuel is comprehensively utilized, and the requirements of energy conservation and environmental protection are met.

Description

Hybrid electric vehicle energy management method based on HCCI engine
Technical Field
The invention belongs to the field of new energy automobiles, and particularly relates to a hybrid electric vehicle.
Background
With the continuous exhaustion of petroleum and the enhancement of the environmental protection consciousness of people, people put forward more requirements on energy-saving and environment-friendly new energy automobiles. Under such circumstances, alternatives to conventional automobiles such as hybrid automobiles and electric automobiles are emerging. Before the problems of endurance, charging and the like of a battery of a pure electric vehicle are not completely solved, the hybrid electric vehicle is a good choice. The hybrid electric vehicle is simultaneously provided with the engine and the power motor, when the electric quantity of the power battery is insufficient, the engine drives the generator to charge the power battery, the power battery provides energy for the power motor to drive the vehicle to move, and when the vehicle needs high-power output, the engine and the power motor can work cooperatively to drive the vehicle together. The engine of the existing hybrid electric vehicle adopts the traditional engine, namely a spark ignition type gasoline engine (SI) or a compression combustion diesel engine (CI), although the hybrid electric vehicle has the advantages that the engine can work at the best efficiency point, the engine is energy-saving, the thermal efficiency of the two engines is limited, the thermal efficiency of the relatively high diesel engine is more than 40 percent at most, most of heat is taken away by cooling water, waste gas and a cylinder wall, and great energy waste is caused.
The mean charge compression ignition (HCCI) technology was born 1897 and has flameless properties and high dilution capability, allowing combustion to be carried out at lower temperatures, reducing NOx and PM formation, and reducing CO and HC emissions. In addition, HCCI works under the state of no throttle, can reduce the pumping loss of the engine to a great extent, raise the fuel efficiency up to 30%, effectively reduce the fuel consumption. In view of this, HCCI is recognized as a new generation combustion technology, which is a promising technology, and can further improve fuel efficiency while reducing emissions, and for this reason the present invention contemplates using an HCCI engine as a power source for charging a battery of a hybrid vehicle.
At present, the hybrid electric vehicle can be classified into a plug-in type and a vehicle-mounted type according to a battery charging mode of the hybrid electric vehicle. The plug-in hybrid electric vehicle utilizes an external charger to charge the battery, and the driving mileage is short after one-time charging; the vehicle-mounted hybrid electric vehicle charges the battery by using the engine and the vehicle-mounted charger, and has large volume and high oil consumption. Patent [ CN201110268778.7] discloses a hybrid electric vehicle charging engine with compact structure and small volume, which alleviates the problems of short driving distance of the hybrid electric vehicle, large volume and heavy weight of the charging engine, but does not change the current situations of low engine efficiency and high emission, so the invention adopts the HCCI engine as the charging engine. And the exhaust gas of the engine of the hybrid electric vehicle has the characteristics of high temperature and large heat, and the exhaust gas is directly discharged to the atmosphere to cause great energy waste.
Disclosure of Invention
Aiming at the defects of low thermal efficiency and high emission of a hybrid power engine in the prior art, the invention designs a hybrid power automobile energy management method based on an HCCI engine, which utilizes the HCCI engine as a power source of a power battery of the hybrid power automobile and simultaneously utilizes the waste gas of the HCCI engine to drive a Stirling engine to charge the power battery, thereby realizing the energy management of the hybrid power automobile.
The technical scheme for solving the technical problems is that the hybrid electric vehicle based on the mean value charge compression ignition HCCI engine and the energy management system thereof are provided, the hybrid electric vehicle comprises the HCCI engine, a Stirling engine, a generator A, a generator B, a battery and a motor, waste gas of the HCCI engine is connected through a pipeline to drive the Stirling engine and is used for driving the engine B, the HCCI engine directly drives the generator A, the generator A, B is charged by a power battery, when the power battery needs to be charged, the HCCI engine drives the generator A to charge the power battery, meanwhile, the waste gas of the HCCI engine drives the Stirling engine, the Stirling engine drives the generator B to charge the power battery, the power battery provides energy for a power motor of the hybrid electric vehicle to drive the motor, and a transmission device is controlled to drive the vehicle to.
Wherein the exhaust gas driven stirling engine of the HCCI engine further comprises: the variable valve timing technology is used for controlling the timing of an intake valve and an exhaust valve to control the ignition timing of the HCCI engine, so that exhaust gas recompression is realized, and the mixed gas is subjected to compression ignition.
Still further, controlling ignition of the HCCI engine further comprises: the method comprises the steps that a three-layer BP neural network which takes intake and exhaust valve timing, intake manifold pressure, intake manifold temperature, engine speed and fuel injection quantity as input layers and takes HCCI engine ignition timing as output is established to predict the ignition timing of the HCCI engine, and input signals are firstly input into the input layers, then pass through the hidden layers and finally reach the output layers. The input layer comprises 8 neural nodes of the input vector, and the hidden layer comprises a hidden layer neuron activation function
Figure BDA0001357221770000031
(. h) hidden layer neuron threshold θjThe output layer comprises 1 neuron node for constructing an output vector y by an activation function psi (-) of an output layer neuron, wherein a threshold value theta of the output layer neuron, and a connection weight omega between the input layer neuron and the hidden layer neuron are set to be equal to or larger than a threshold value theta of the output layer neuronijConnection weight ω between hidden layer neurons to output layer neuronsj。
According to the formula
Figure BDA0001357221770000032
Calculating input signal net of j-th neuron node of hidden layerj(ii) a According to the formula
Figure BDA0001357221770000033
Calculating the output signal o of the jth neural node of the hidden layerj(ii) a According to the formula:
Figure BDA0001357221770000034
calculating an input signal net of a neuron node of an output layer; according to the formula
Figure BDA0001357221770000035
And calculating an output signal y of the neural node of the output layer, and realizing the regulation and control of the ignition timing of the HCCI engine through an output vector.
The invention also provides a hybrid electric vehicle based on the mean value inflation compression ignition HCCI engine and an energy management method thereof, and the hybrid electric vehicle comprises the steps that waste gas of the HCCI engine is connected through a pipeline to drive the Stirling engine and is used for driving the engine B, the HCCI engine directly drives the generator A, the generator A, B is charged by a power battery, when the power battery needs to be charged, the HCCI engine drives the generator A to charge the power battery, meanwhile, the waste gas of the HCCI engine drives the Stirling engine, the Stirling engine drives the generator B to charge the power battery, the power battery provides energy for a power motor of the hybrid electric vehicle to drive the motor, and a transmission device is controlled to drive the vehicle to move.
The power battery of the hybrid electric vehicle provides energy for the driving motor to drive the vehicle to move, when the electric quantity of the power battery is insufficient, the HCCI engine is used for driving the generator A to charge the power battery, meanwhile, the waste gas of the HCCI engine provides heat for the Stirling engine, and the generator B is driven to charge the power battery; the variable valve timing technology is used for controlling the timing of an intake valve and an exhaust valve, the ignition of the HCCI engine is effectively controlled, the exhaust gas recompression is realized, and therefore the mixed gas is compressed to ignite, and the HCCI engine drives a generator to charge a power battery.
Aiming at the current situations of low efficiency and high emission of the engine and the characteristics of high temperature and large heat of the exhaust gas of the engine of the hybrid electric vehicle, the exhaust gas is directly discharged to the atmosphere to cause great energy waste. The invention adopts the HCCI engine as a charging engine, and on the basis of charging the HCCI engine for the power battery, the waste gas of the HCCI engine is utilized to drive the Stirling engine to drive another generator to charge the power battery, thereby fully improving the utilization energy rate of the engine.
Drawings
FIG. 1 is a hybrid vehicle energy system based on an HCCI engine;
FIG. 2 is a hybrid vehicle energy management method based on an HCCI engine;
FIG. 3 is a BP neural network model for ignition timing prediction.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and examples.
FIG. 1 is a schematic diagram of a hybrid vehicle power system configuration based on a mean-value charge compression ignition HCCI engine. Aiming at the characteristics of high efficiency and low emission of the HCCI engine, the hybrid electric vehicle energy system based on the HCCI engine comprises the HCCI engine, the Stirling engine, a generator A, a generator B, a battery and a motor. The waste gas of the HCCI engine drives the Stirling engine through a pipeline connection and is used for driving an engine B; the HCCI engine drives a generator A through a mechanical connection; the generator A, B is connected to the battery and electrically connected to drive the motor, controlling the transmission.
The details of the invention are further explained below by means of fig. 2.
The HCCI engine is a power source of the whole hybrid electric vehicle, when a power battery needs to be charged, the HCCI engine is started, and the HCCI engine drives a generator A to charge the battery; considering that the ignition timing of the HCCI engine is not measurable, the method takes the opening and closing timing of an intake valve and an exhaust valve, the engine speed, the intake manifold pressure, the intake manifold temperature and the fuel injection quantity as input, builds a neural network, trains for multiple times, and predicts and outputs the ignition timing of the HCCI engine. Inputting the difference value of the ignition timing expected value and the predicted value into a proportional-integral-derivative (PID) controller, and controlling the ignition timing of the HCCI engine by controlling the fuel injection quantity and the opening and closing timing of an intake valve and an exhaust valve; the HCCI engine utilizes a variable valve technology, and part of exhaust gas is reserved to participate in the combustion process again, so that the high efficiency and low emission characteristics of the HCCI engine can be enhanced; after the HCCI engine operates, the Stirling engine collects energy in the tail gas of the HCCI engine and drives a generator B to be stored in a battery in the form of electric energy; when charging is complete, the HCCI engine is turned off.
Fig. 3 is a structural model of a BP neural network for ignition timing prediction, and an input signal is firstly input into an input layer, then passes through a hidden layer, and finally reaches an output layer. The input signal includes: intake valve opening timing u1And opening timing u of exhaust valve2Intake valve closing timing u3Exhaust valve closing timing u4Engine speed u5Intake manifold pressure u6Intake manifold temperature u7Oil injection quantity u8. Wherein, the input layer comprises 8 neural nodes (i ═ 1,2,3,4,5,6,7,8), and the input vector u ═ u (u ═ c)1，u2，u3，u4，u5，u6，u7，u8)T∈R8(ii) a (T is the transpose of the matrix, R is a real number) the hidden layer contains 4 neuron nodes (j ═ 1,2,3,4),
Figure BDA0001357221770000051
(. The) represents the activation function of hidden layer neurons, θjA threshold representing hidden layer neurons; the output layer contains 1 neuron node (k ═ 1), the output vector is y, ψ (·) represents the activation function of the output layer neurons, and θ represents the threshold of the output layer neurons. OmegaijRepresenting the connection weight between the input layer neuron and the hidden layer neuron; omegajRepresenting the connection weights between the hidden layer neurons to the output layer neurons.
And in the forward propagation process of the input signal, respectively calculating the input and output signals of the hidden layer and the output layer according to the threshold values of neurons of all layers and the connection weight values between all layers. And obtaining an output vector y of the output layer, and realizing the regulation and control of the ignition timing of the HCCI engine through the output vector.
Input signal net of j-th neuron node of hidden layerjIs composed of
Figure BDA0001357221770000061
In the formula, ω represents a connection weight between an input layer neuron and a hidden layer neuron, θ represents a threshold value of an output layer neuron, and u is an input vector.
Output signal o of jth neural node of hidden layerjIs composed of
Figure BDA0001357221770000062
In the formula (I), the compound is shown in the specification,
Figure BDA0001357221770000063
(. cndot.) represents the activation function of hidden layer neurons.
The input signal net of the output layer neuron node is
Figure BDA0001357221770000064
The output signal y of the neural node of the output layer is
Figure BDA0001357221770000065
In the formula, ψ (·) represents an activation function of neurons of the output layer.
In the error back propagation process, the error of each layer is calculated from the output layer in the back direction, and the connection value and the threshold value of each layer are updated according to the gradient descent algorithm, so that the actual output of the network is as close to the expected output as possible.
Assuming that the training sample set contains P training samples, for each training sample P (P1, 2.., P), the quadratic criterion function of the error is
Figure BDA0001357221770000066
The overall error function of the network on P training samples is
Figure BDA0001357221770000071
Wherein E(p)Represents a single sample error, E represents the cumulative sum of all sample errors; d(p)And y(p)Respectively representing the expected output and the actual output of the output layer neuron node when the input training sample is p.
Updating the connection weight and the threshold value of the network layer by layer according to a gradient descent algorithm, and correcting the connection weight from the hidden layer to the output layer by a correction quantity delta omegajOutput layer threshold correction amount delta theta, input layer to hidden layer connection weight correction amount delta omegaijOutput layer threshold correction amount Δ θj：
The formula for adjusting the connection weight from the hidden layer to the output layer is
Figure BDA0001357221770000072
The output layer threshold value is adjusted by the formula
Figure BDA0001357221770000073
The formula for adjusting the connection weight from the input layer to the hidden layer is
Figure BDA0001357221770000074
The hidden layer threshold adjustment formula is
Figure BDA0001357221770000075
And because of
Figure BDA0001357221770000076
Figure BDA0001357221770000077
Figure BDA0001357221770000078
Figure BDA0001357221770000079
Figure BDA0001357221770000081
Finally, the formula is obtained:
Figure BDA0001357221770000082
ωj(k+1)＝ηδoj+ωj(k) (17)
Figure BDA0001357221770000083
θ(k+1)＝ηδ+θ(k) (19)
Figure BDA0001357221770000084
ωij(k+1)＝ηδjui+ωij(k) (21)
Figure BDA0001357221770000085
θj(k+1)＝ηδj+θj(k) (23)
where η is the learning rate and k is the trainingDegree of, δ and δjRepresenting the error signals of the output layer and the hidden layer, respectively.
The invention considers that the ignition timing of the HCCI engine is not measurable, and according to the relationship between the ignition timing of the HCCI engine and the rotating speed, the pressure of an intake manifold, the fuel injection quantity and the like of the HCCI engine, the ignition timing of the engine is predicted by a three-layer BP neural network which takes the timing of an intake valve, the pressure of the intake manifold, the temperature of the intake manifold, the rotating speed of the engine and the fuel injection quantity as input and takes the ignition timing of the HCCI engine as output.

Claims (8)

1. A hybrid electric vehicle energy management system based on an average value charge compression ignition HCCI engine is characterized by comprising an HCCI engine, a Stirling engine, a generator A, a generator B, a battery and a motor, wherein exhaust gas of the HCCI engine is connected through a pipeline to drive the Stirling engine and is used for driving the engine B, the HCCI engine directly drives the generator A, the generator A, B is charged by a power battery, when the power battery needs to be charged, the HCCI engine is started, the HCCI engine drives the generator A to charge the power battery, timing of opening and closing of an intake and exhaust valve, engine speed, intake manifold pressure, intake manifold temperature and fuel injection quantity are used as input, a neural network is built and trained, ignition timing of the HCCI engine is predicted to be output, a difference value between a desired value and a predicted value of the ignition timing is input into a proportional-integral-derivative controller, the ignition timing of the HCCI engine is regulated and controlled by controlling the fuel injection quantity and the opening and closing timing of an intake valve and an exhaust valve; the HCCI engine reserves partial waste gas to participate in combustion again, after the HCCI engine operates, the Stirling engine collects energy in tail gas of the HCCI engine and drives a generator B to be stored in a power battery in the form of electric energy to charge the power battery, when charging is completed, the HCCI engine is turned off, the power battery supplies energy to a power motor of a hybrid electric vehicle to drive the motor, and a transmission device is controlled to drive the vehicle to move.
2. The system of claim 1, wherein controlling ignition of the HCCI engine further comprises: establishing a three-layer BP neural network comprising an input layer, a hidden layer and an output layer to predict the ignition timing of the HCCI engine, taking the timing of an intake valve and an exhaust valve, the pressure of an intake manifold, the temperature of the intake manifold, the rotating speed of the engine and the fuel injection quantity as the BP neural network input layer, respectively calculating input and output signals of the hidden layer and the output layer according to the threshold value of neurons of each layer and the connection weight value among the layers, obtaining an output vector y of the output layer, regulating and controlling the ignition timing of the HCCI engine through the output vector, and taking the ignition timing of the HCCI engine as output.
3. The system of claim 2, wherein the input layer comprises 8 neural nodes of the input vector and the hidden layer comprises a hidden layer neuron activation function
Figure FDA0002352975930000021
4 neuron nodes of a hidden layer neuron threshold theta j, 1 neuron node of an output vector y constructed by an activation function psi (-) of an output layer neuron is included in an output layer, the threshold theta of the output layer neuron, a connection weight omega ij between the input layer neuron and the hidden layer neuron and a connection weight omega j between the hidden layer neuron and the output layer neuron are updated layer by layer, and the output layer neuron is connected with the output layer neuron according to a formula delta omegaij＝ηδjuiCalculating the correction quantity delta omega of the connection weight from the input layer to the hidden layerijAccording to the formula Δ ωj＝ηδojCalculating the correction quantity delta omega of the connection weight from the hidden layer to the output layerjCalculating an output layer threshold correction amount according to a formula delta theta- η delta, wherein η is learning rate, ojFor the output signal of the j-th neural node of the hidden layer, uiFor the ith input vector, δ and δjRepresenting the error signals of the output layer and the hidden layer, respectively.
4. The system of claim 3, wherein the data is based on a formula
Figure FDA0002352975930000022
Calculating input signal net of j-th neuron node of hidden layerj(ii) a According to the formula
Figure FDA0002352975930000023
Calculating the output signal o of the jth neural node of the hidden layerj(ii) a According to the formula:
Figure FDA0002352975930000024
calculating an input signal net of a neuron node of an output layer; according to the formula
Figure FDA0002352975930000025
And calculating an output signal y of the neural node of the output layer.
5. A hybrid electric vehicle energy management method based on mean value charge compression ignition HCCI engine, it is characterized in that the waste gas of the HCCI engine drives the Stirling engine through the pipeline connection, used for driving an engine B, an HCCI engine directly drives a generator A, a generator A, B is charged by a power battery, when the power battery needs to be charged, the HCCI engine is started, the HCCI engine drives the generator A to charge the power battery, the opening and closing timing of an intake valve and an exhaust valve, the rotating speed of the engine, the pressure of an intake manifold, the temperature of the intake manifold and the fuel injection quantity are used as input, a neural network is built and trained for multiple times, the ignition timing of the HCCI engine is predicted and output, the difference between the desired value and the predicted value of the ignition timing is input to a proportional-integral-derivative controller, the ignition timing of the HCCI engine is regulated and controlled by controlling the fuel injection quantity and the opening and closing timing of an intake valve and an exhaust valve; the HCCI engine reserves partial waste gas to participate in combustion again, after the HCCI engine operates, the Stirling engine collects energy in tail gas of the HCCI engine and drives a generator B to be stored in a power battery in the form of electric energy to charge the power battery, when charging is completed, the HCCI engine is turned off, the power battery supplies energy to a power motor of a hybrid electric vehicle to drive the motor, and a transmission device is controlled to drive the vehicle to move.
6. The method of claim 5, wherein controlling ignition timing of the HCCI engine further comprises: establishing a three-layer BP neural network comprising an input layer, a hidden layer and an output layer to predict the ignition timing of the HCCI engine, taking the timing of an intake valve and an exhaust valve, the pressure of an intake manifold, the temperature of the intake manifold, the rotating speed of the engine and the fuel injection quantity as the input layer, respectively calculating the input and output signals of the hidden layer and the output layer according to the threshold value of neurons of each layer and the connection weight value between the layers, obtaining the output vector y of the output layer, regulating and controlling the ignition timing of the HCCI engine through the output vector, and taking the ignition timing of the HCCI engine as the output.
7. The method of claim 6, wherein the input layer comprises 8 neural nodes of the input vector, and wherein the hidden layer comprises a hidden layer neuron activation function
Figure FDA0002352975930000031
Hidden layer neuron threshold θjThe output layer comprises 1 neuron node for constructing an output vector y by an activation function psi (-) of the neuron in the output layer, the threshold value theta of the neuron in the output layer, the connection weight value omega ij between the neuron in the input layer and the neuron in the hidden layer and the connection weight value omega j between the neuron in the hidden layer and the neuron in the output layer are updated layer by layer, and the output layer comprises the 1 neuron node for constructing the output vector y by the activation function psi (-) of the neuron in the output layer, and the output layer is updated according to the formulaij＝ηδjuiCalculating the correction quantity delta omega of the connection weight from the input layer to the hidden layerijAccording to the formula Δ ωj＝ηδojCalculating the correction quantity delta omega of the connection weight from the hidden layer to the output layerjCalculating an output layer threshold correction amount according to a formula delta theta- η delta, wherein η is learning rate, ojFor the output signal of the j-th neural node of the hidden layer, uiFor the ith input vector, δ and δjRepresenting the error signals of the output layer and the hidden layer, respectively.
8. The method of claim 7, wherein the method is based on a formula
Figure FDA0002352975930000041
Calculating input signal net of j-th neuron node of hidden layerj(ii) a According to the formula
Figure FDA0002352975930000042
Calculating the output signal o of the jth neural node of the hidden layerj(ii) a According to the formula
Figure FDA0002352975930000043
Calculating an input signal net of a neuron node of an output layer; according to the formula
Figure FDA0002352975930000044
And calculating an output signal y of the neural node of the output layer, and realizing the regulation and control of the ignition timing of the HCCI engine through an output vector.
CN201710601345.6A 2017-07-21 2017-07-21 Hybrid electric vehicle energy management method based on HCCI engine Active CN107472038B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710601345.6A CN107472038B (en) 2017-07-21 2017-07-21 Hybrid electric vehicle energy management method based on HCCI engine

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710601345.6A CN107472038B (en) 2017-07-21 2017-07-21 Hybrid electric vehicle energy management method based on HCCI engine

Publications (2)

Publication Number Publication Date
CN107472038A CN107472038A (en) 2017-12-15
CN107472038B true CN107472038B (en) 2020-05-12

Family

ID=60595257

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710601345.6A Active CN107472038B (en) 2017-07-21 2017-07-21 Hybrid electric vehicle energy management method based on HCCI engine

Country Status (1)

Country Link
CN (1) CN107472038B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109466375B (en) * 2018-12-05 2020-12-25 北京车和家信息技术有限公司 Range extender control method and device, computer-readable storage medium and vehicle
US11459962B2 (en) * 2020-03-02 2022-10-04 Sparkcognitton, Inc. Electronic valve control
CN112963256B (en) * 2021-03-22 2022-09-30 重庆邮电大学 HCCI/SI combustion mode switching process control method
CN114030461B (en) * 2021-11-26 2023-07-07 深圳技术大学 Hybrid electric vehicle energy management method and device based on dual-mode engine

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7415347B2 (en) * 2006-08-25 2008-08-19 Michigan Technological University Combustion knock detection and control through statistical characterization of knock levels
CN101603473A (en) * 2009-07-16 2009-12-16 天津大学 The waste heat recovery plant of utilization Stirling engine
US8726661B2 (en) * 2010-08-09 2014-05-20 GM Global Technology Operations LLC Hybrid powertrain system including an internal combustion engine and a stirling engine
CN104656443B (en) * 2014-12-31 2017-05-24 重庆邮电大学 HCCI engine ignition timing self-adaptive PID control method based on BP neural network
CN105221282B (en) * 2015-08-25 2018-06-26 重庆邮电大学 Based on the hcci engine of linear model and sliding mode controller burning timing control method

Also Published As

Publication number Publication date
CN107472038A (en) 2017-12-15

Similar Documents

Publication Publication Date Title
CN107472038B (en) Hybrid electric vehicle energy management method based on HCCI engine
Yang et al. A stochastic predictive energy management strategy for plug-in hybrid electric vehicles based on fast rolling optimization
Zhao et al. Real-time energy management for diesel heavy duty hybrid electric vehicles
Arsie et al. Optimization of supervisory control strategy for parallel hybrid vehicle with provisional load estimate
Huang et al. Multi-objective genetic algorithm for hybrid electric vehicle parameter optimization
CN109552112A (en) A kind of battery energy storage control method for hybrid vehicle
CN110185548A (en) The control device and control method of internal combustion engine
Arsie et al. Evaluation of CO2 reduction in SI engines with Electric Turbo-Compound by dynamic powertrain modelling
CN109878499B (en) Hybrid vehicle power control method
CN116263128A (en) Method and system for a multi-fuel engine
CN110077389B (en) Energy management method for plug-in hybrid electric vehicle
He et al. Fuel economy optimization of diesel engine for plug-in hybrid electric vehicle based on equivalent operating points
Sun et al. Optimal control of transient processes in marine hybrid propulsion systems: Modeling, optimization and performance enhancement
US20230415791A1 (en) Method and system for multi-fuel engine
CN108563844A (en) A kind of supercharged diesel engine Optimum EGR rate determines method
CN211058916U (en) Engine air inlet control device of hybrid electric vehicle
Liu et al. Development of performance and combustion system of Atkinson cycle internal combustion engine
Giardiello et al. Comparative Analysis on Fuel Consumption Between Two Online Strategies for P2 Hybrid Electric Vehicles: Adaptive-RuleBased (A-RB) vs Adaptive-Equivalent Consumption Minimization Strategy (A-ECMS)
CN113997925B (en) Energy management method for plug-in hybrid power system
CN112963256B (en) HCCI/SI combustion mode switching process control method
Sheng et al. Fuzzy energy optimization strategy of parallel hybrid electric vehicle based on improved adaptive genetic algorithm
CN112046463A (en) Control method for power assembly of hybrid electric vehicle
Zhang et al. An optimal energy management strategy for parallel HEVs
Karystinos Nonlinear model predictive control of a hybrid diesel-electric marine propulsion plant
Norbakyah et al. Modeling, simulation and model optimization of internal combustion engine for PHERB powertrain

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