CN112660127B - CACC energy management method for queue hybrid truck based on deep migration learning - Google Patents
CACC energy management method for queue hybrid truck based on deep migration learning Download PDFInfo
- Publication number
- CN112660127B CN112660127B CN202011617231.9A CN202011617231A CN112660127B CN 112660127 B CN112660127 B CN 112660127B CN 202011617231 A CN202011617231 A CN 202011617231A CN 112660127 B CN112660127 B CN 112660127B
- Authority
- CN
- China
- Prior art keywords
- cacc
- energy management
- function
- optimal
- queue
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000007726 management method Methods 0.000 title claims abstract description 43
- 238000013508 migration Methods 0.000 title claims abstract description 37
- 230000005012 migration Effects 0.000 title claims abstract description 37
- 238000003062 neural network model Methods 0.000 claims abstract description 16
- 238000013528 artificial neural network Methods 0.000 claims abstract description 6
- 230000007704 transition Effects 0.000 claims abstract description 4
- 230000006870 function Effects 0.000 claims description 53
- 238000000034 method Methods 0.000 claims description 35
- 238000000605 extraction Methods 0.000 claims description 20
- 238000011176 pooling Methods 0.000 claims description 17
- 230000008569 process Effects 0.000 claims description 9
- 230000001133 acceleration Effects 0.000 claims description 6
- NRJAVPSFFCBXDT-HUESYALOSA-N 1,2-distearoyl-sn-glycero-3-phosphocholine Chemical compound CCCCCCCCCCCCCCCCCC(=O)OC[C@H](COP([O-])(=O)OCC[N+](C)(C)C)OC(=O)CCCCCCCCCCCCCCCCC NRJAVPSFFCBXDT-HUESYALOSA-N 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 230000001131 transforming effect Effects 0.000 claims description 3
- 230000008447 perception Effects 0.000 claims description 2
- 239000000446 fuel Substances 0.000 abstract description 18
- 230000003044 adaptive effect Effects 0.000 abstract description 7
- 238000012360 testing method Methods 0.000 description 5
- 238000005457 optimization Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000012634 fragment Substances 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- 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
-
- 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
- Y02T10/62—Hybrid vehicles
Landscapes
- Hybrid Electric Vehicles (AREA)
Abstract
The invention discloses a CACC energy management method of a queue hybrid truck based on deep migration learning, which comprises the following steps: constructing a deep migration learning neural network model based on the VGG network; respectively extracting characteristic parameters of the short-time running condition segment in the previous stage and characteristic data of the optimal SOC track corresponding to the short-time running condition segment in the next stage by using a neural network model; obtaining a reference SOC track based on the characteristic data of the optimal SOC track, and obtaining an optimal following vehicle speed track corresponding to a short-time running condition segment at the later stage of the train hybrid truck according to the state transition of the neural network; and controlling the actual SOC track to follow the optimal SOC track in the next-stage short-time running working condition according to the reference SOC track and the optimal follow-up vehicle speed track in real time in a prediction mode, and achieving CACC energy management of the queue hybrid truck. The invention can realize real-time energy management in a Cooperative Adaptive Cruise Control (CACC) system of the hybrid power trucks in the queue, thereby realizing the optimal equivalent fuel economy when the queue runs with the trucks.
Description
Technical Field
The invention relates to the technical field of energy management of a Coordinated Adaptive Cruise Control (CACC) system of a train truck, in particular to a CACC energy management method of a train hybrid truck based on deep migration learning.
Background
The 'queue following' of the hybrid power truck is used as the field of the automatic driving leading application, the distance between vehicles and the running state of a motorcade can be effectively controlled, the wind resistance in the running of the motorcade is reduced, and therefore the fuel consumption of the vehicles is reduced. The cooperative adaptive cruise system can well realize the queue running of the hybrid power trucks, but because the queue running can meet various working conditions, the existing rule-based or optimization-based energy management method cannot well realize the optimal equivalent fuel economy when the hybrid power trucks run in the queue and run, and how to creatively provide an energy management method based on learning, so that the hybrid power trucks can manage and control the energy in real time when running in the queue, the CACC of the hybrid power trucks can realize the optimal equivalent fuel economy, and the problem to be solved in the field is urgent.
The invention patent ZL201510896784.5 proposes an apparatus and method for controlling the speed of a CACC system, and the disadvantages of the invention patent are: only the method of controlling the speed of the vehicle using the collected information to reduce the width of deceleration and acceleration to improve fuel efficiency is provided, and a real-time management control of energy is not provided, so the improved fuel efficiency is low; the invention patent ZL201710832339.1 proposes a control device and method for improving fuel efficiency in CACC system, and the invention patent has the following disadvantages: only the use of an optimization cost to run the vehicle in consideration of the target vehicle speed, the current vehicle speed, the minimum running speed set in the vehicle, and the deceleration distance is provided to improve the fuel efficiency, and an advanced energy management method to achieve the optimum equivalent fuel economy is not provided.
Disclosure of Invention
This section is for the purpose of summarizing some aspects of embodiments of the invention and to briefly introduce some preferred embodiments. In this section, as well as in the abstract and the title of the invention of this application, simplifications or omissions may be made to avoid obscuring the purpose of the section, the abstract and the title, and such simplifications or omissions are not intended to limit the scope of the invention.
The present invention has been made in view of the above-mentioned conventional problems.
Therefore, the technical problem solved by the invention is as follows: the prior art cannot provide a method for managing and controlling energy in real time and an advanced energy management method to realize optimal equivalent fuel economy.
In order to solve the technical problems, the invention provides the following technical scheme: a queue hybrid truck CACC energy management method based on deep migration learning is characterized by comprising the following steps: constructing a deep migration learning neural network model based on the VGG network; respectively extracting characteristic parameters of the short-time running condition segment of the previous stage as input of the model and characteristic data of an optimal SOC track corresponding to the short-time running condition segment of the next stage as output of the model by using the neural network model; obtaining a reference SOC track based on the characteristic data of the optimal SOC track, and obtaining an optimal following vehicle speed track corresponding to a short-time running condition segment at the later stage of the train hybrid truck according to the state transition of a neural network; and controlling the actual SOC track to follow the optimal SOC track in the next-stage short-time running working condition according to the reference SOC track and the optimal following vehicle speed track in a real-time prediction mode, and achieving CACC energy management of the train-in-line hybrid truck.
As an optimal scheme of the deep migration learning-based CACC energy management method for the queue hybrid truck, the method comprises the following steps: the VGG network comprises 13 convolutional layers, 5 pooling layers, 3 full-connection layers and a Softmax layer.
As an optimal scheme of the deep migration learning-based CACC energy management method for the queue hybrid truck, the method comprises the following steps: the deep migration learning neural network model is composed of six-element groups and comprises two parallel networks, specifically comprising primary feature extraction and secondary feature extraction.
As an optimal scheme of the deep migration learning-based CACC energy management method for the queue hybrid truck, the method comprises the following steps: the first parallel network may include a first parallel network,
M＝(V1,V2,D,C,P,S)
wherein, V1、V2Representing a feature extraction function, D representing a full-join function, C representing a convolution function, P representing a pooling function, and S representing a Softmax classification function.
As an optimal scheme of the deep migration learning-based CACC energy management method for the queue hybrid truck, the method comprises the following steps: the feature extraction function includes a function of extracting a feature,
V1(I)＝R
wherein R is ∈ RE multiplied by F, V1(.) represents a feature space mapping, the V1(.) transforming the input operating condition segment characteristic parameter information I into a characteristic of E x F dimension.
As an optimal scheme of the deep migration learning-based CACC energy management method for the queue hybrid truck, the method comprises the following steps: the full-concatenation function includes a function of,
D(R)＝T
where D (.) converts the E × F dimensional data to the EF × 1 dimensional data.
As an optimal scheme of the deep migration learning-based CACC energy management method for the queue hybrid truck, the method comprises the following steps: the convolution function includes a function of a function including,
where i and j represent the row and column indices of the matrix, K represents the convolution kernel, and P represents the pooling function.
As an optimal scheme of the deep migration learning-based CACC energy management method for the queue hybrid truck, the method comprises the following steps: the pooling function may include a pooling function of,
as an optimal scheme of the deep migration learning-based CACC energy management method for the queue hybrid truck, the method comprises the following steps: the extraction method of the characteristic parameters of the short-time running condition segment in the previous stage comprises environment perception, DSPC, GIS and GPS.
As an optimal scheme of the deep migration learning-based CACC energy management method for the queue hybrid truck, the method comprises the following steps: inputting the optimal following vehicle speed track corresponding to the short-time running condition segment at the later stage of the train hybrid truck as a reference following vehicle speed track into model prediction control; and (3) optimizing the disturbance variable by adding a disturbance observer and a Smith predictor by taking the acceleration of the vehicle, the torque of the motor and the rotating speed of the engine as control variables.
The invention has the beneficial effects that: the actual SOC track of the hybrid power truck in the next stage short-time running working condition in the running process of the coordinated adaptive cruise and the vehicle follows the optimal SOC track, so that real-time energy management in a coordinated adaptive cruise system (CACC) of the hybrid power truck in the queue is realized, and the optimal equivalent fuel economy is realized when the hybrid power truck in the queue runs with the vehicle.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise. Wherein:
fig. 1 is a basic flowchart of a method for CACC energy management of a queue hybrid truck based on deep migration learning according to an embodiment of the present invention;
fig. 2 is another schematic flow chart of a method for CACC energy management of a queued hybrid truck based on deep migration learning according to an embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, specific embodiments accompanied with figures are described in detail below, and it is apparent that the described embodiments are a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making creative efforts based on the embodiments of the present invention, shall fall within the protection scope of the present invention.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways than those specifically described and will be readily apparent to those of ordinary skill in the art without departing from the spirit of the present invention, and therefore the present invention is not limited to the specific embodiments disclosed below.
Furthermore, reference herein to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one implementation of the invention. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments.
The present invention will be described in detail with reference to the drawings, wherein the cross-sectional views illustrating the structure of the device are not enlarged partially in general scale for convenience of illustration, and the drawings are only exemplary and should not be construed as limiting the scope of the present invention. In addition, the three-dimensional dimensions of length, width and depth should be included in the actual fabrication.
Meanwhile, in the description of the present invention, it should be noted that the terms "upper, lower, inner and outer" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of describing the present invention and simplifying the description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation and operate, and thus, cannot be construed as limiting the present invention. Furthermore, the terms first, second, or third are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
The terms "mounted, connected and connected" in the present invention are to be understood broadly, unless otherwise explicitly specified or limited, for example: can be fixedly connected, detachably connected or integrally connected; they may be mechanically, electrically, or directly connected, or indirectly connected through intervening media, or may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Example 1
Referring to fig. 1 to 2, an embodiment of the present invention provides a deep migration learning-based queue hybrid truck CACC energy management method, including:
s1: constructing a deep migration learning neural network model based on the VGG network;
it should be noted that the VGG network includes 13 convolutional layers, 5 pooling layers, 3 full-connectivity layers, and Softmax layer.
The deep migration learning neural network model consists of six elements and comprises two parallel networks, specifically comprising primary feature extraction and secondary feature extraction.
Further, the first parallel network includes,
M＝(V1,V2,D,C,P,S)
wherein, V1、V2Representing a feature extraction function, D representing a full-join function, C representing a convolution function, P representing a pooling function, and S representing a Softmax classification function.
Wherein the feature extraction function comprises:
V1(I)＝R
wherein R is ∈ RE multiplied by F, V1(.) represents a feature space map, V1(.) transforming the input operating condition segment characteristic parameter information I into a characteristic of E x F dimension.
Further, the full-join function includes:
D(R)＝T
where D (.) converts the E × F dimensional data to the EF × 1 dimensional data.
Further, the convolution function includes:
where i and j represent the row and column indices of the matrix, K represents the convolution kernel, and P represents the pooling function.
The pooling function includes:
specifically, in the neural network model construction, as shown in fig. 2, the deep migration learning network is composed of two parallel networks, and the network includes two parts, namely primary feature extraction and secondary feature extraction. In the operation process of the classifier, the features of different dimensions are interacted by any one parallel network, so that the features of different working condition fragments can be extracted, and therefore, the relation between the different features of the input working condition fragments can be captured simultaneously through the two parallel networks; the deep migration learning network model consists of six elements, taking the first parallel network as an example, and the functional form is as follows:
M＝(V1,V2,D,C,P,S)
wherein V1、V2Representing a feature extraction function, D representing a full-join function, C representing a convolution function, P representing a pooling function, and S representing a Softmax classification function.
Feature extraction function V1(.) is done by VGG model, V1(.) for feature space mapping, the data feature space is transformed:
V1(I)＝R
wherein R is ∈ RE × F, V1(.) converting the input working condition segment characteristic parameter information I into an E x F dimensional characteristic; and then performing dimension transformation on the output of the two VGG feature extraction functions through a full-connection function D:
D(R)＝T
d (·) converting the data of E multiplied by F dimensionality into data of EF multiplied by 1 dimensionality, C is a convolution function, and extracting the characteristics of the output data of the VGG again, wherein the convolution function is as follows:
wherein i and j represent row and column indexes of the matrix, K is a convolution kernel, P is a pooling function, the function of the pooling function is to converge the output of each convolution layer into a final reference SOC track characteristic data characteristic, the calculation amount is reduced, and the calculation formula of P is as follows:
s2: respectively extracting characteristic parameters of the short-time running condition segment of the previous stage as input of the model and characteristic data of an optimal SOC track corresponding to the short-time running condition segment of the next stage as output of the model by using a neural network model;
it should be noted that the extraction approach of the characteristic parameters of the short-time driving condition segment in the previous stage includes environment sensing, DSPC, GIS and GPS.
Specifically, in the neural network model training, the running condition is optimized to obtain a global optimal SOC track under the running condition through a DP (dynamic programming algorithm), feature parameters of a short-time running condition segment in a previous stage and feature data of the optimal SOC track corresponding to the short-time running condition segment in the next stage are respectively extracted, the feature parameters of the current working condition segment are used as model input, the feature data of the optimal SOC track of the working condition segment in the next stage are used as model output, and a mapping relation between the feature parameters of the short-time running condition segment in the previous stage and the feature data of the optimal SOC track of the short-time running condition segment in the next stage is established through a multi-input multi-output training model.
S3: obtaining a reference SOC track based on the characteristic data of the optimal SOC track, and obtaining an optimal following vehicle speed track corresponding to a short-time running condition segment at the later stage of the train hybrid truck according to the state transition of the neural network;
s4: and controlling the actual SOC track to follow the optimal SOC track in the next-stage short-time running working condition according to the reference SOC track and the optimal follow-up vehicle speed track in real time in a prediction mode, and achieving CACC energy management of the queue hybrid truck.
It should be noted that, the real-time prediction control process includes,
inputting the optimal following vehicle speed track corresponding to the short-time running condition segment at the later stage of the train hybrid truck as a reference following vehicle speed track into a model for prediction control;
and (3) optimizing the disturbance variable by adding a disturbance observer and a Smith predictor by taking the acceleration of the vehicle, the torque of the motor and the rotating speed of the engine as control variables.
Specifically, in CACC real-time prediction control, when a hybrid truck is in a queue for running, extracting characteristic parameters of a short-time running condition segment in the previous stage from environment sensing (radar, a camera), a DSPC, a GIS and a GPS, using the characteristic parameters of the short-time running condition segment in the previous stage as input of a deep migration neural network, outputting characteristic data of an optimal SOC track corresponding to the short-time running condition segment in the next stage, outputting the optimal following vehicle speed track corresponding to the short-time running condition segment in the next stage of the queue hybrid truck according to state migration of the neural network, using the optimal following vehicle speed track corresponding to the short-time running condition segment in the next stage as a reference SOC track, and inputting the optimal following vehicle speed track corresponding to the short-time running condition segment in the next stage of the queue hybrid truck into model prediction control as a reference following vehicle speed track, the method is characterized in that vehicle acceleration, motor torque and engine speed are used as control variables, a disturbance observer and a Smith predictor are added to optimize the disturbance variables, so that an actual SOC track of a train hybrid truck follows an optimal SOC track in a next-stage short-time running working condition in a coordinated self-adaptive cruise and vehicle following running process, real-time energy management in a coordinated self-adaptive cruise system (CACC) of the train hybrid truck is realized, and the optimal equivalent fuel economy during the train and vehicle following running is further realized.
The method comprises the steps of neural network model construction, neural network model training and CACC real-time prediction control, data information is collected by a radar, a camera, a DSPC, a GIS and a GPS, running state information of the hybrid truck during the queue running can be accurately obtained in real time, an optimal SOC track is obtained by a deep migration learning neural network model and is used as a reference SOC track, the method has the advantages that the vehicle acceleration, the motor torque and the engine rotating speed are used as control variables, the disturbance observer and the Smith predictor are added to optimize the disturbance variables, so that the actual SOC track of the hybrid power trucks in the train follows the optimal SOC track in the next short-time running working condition in the running process of the coordinated adaptive cruise and follow-up, the real-time energy management in the coordinated adaptive cruise system (CACC) of the hybrid power trucks in the train can be realized, and the optimal equivalent fuel economy in the running process of the train and follow-up is further realized.
Example 2
In order to verify the technical effects adopted in the method, the embodiment adopts the traditional technical scheme to carry out comparison test on the energy management method based on optimization and the method of the invention, and compares the test results by means of scientific demonstration to verify the real effect of the method.
The traditional technical scheme is as follows: due to the fact that the queue running can meet various working conditions, the existing rule-based or optimization-based energy management method cannot well achieve the optimal equivalent fuel economy when the hybrid truck is in the queue and runs with the train. In order to verify that the method has higher fuel economy compared with the traditional method, the CACC traditional energy management method and the method are adopted to respectively measure and compare the cruising speed overshoot, the fuel consumption and the battery charge state of the simulated vehicle in real time.
And (3) testing environment: selecting 6 mixed trucks of the same type to form 2 groups of trucks, performing comparative test simulation under the UDDS working condition, respectively simulating the two groups of mixed trucks under the UDDS working condition by using two energy management methods, testing the truck group by using the two methods, performing multiple experiments by taking economy as a judgment standard, and obtaining results shown in the following table:
table 1: the experimental results are shown in a comparison table.
As can be seen from the above table, in the process of cruising and vehicle-following driving of the hybrid truck, the method can control the vehicle speed to be accurately kept at the rational cruising speed (the overshoot is less than 3%), when the set cruising speed changes, the vehicle can be controlled to respond quickly, in the process of vehicle-following driving, the inter-vehicle distance error is small, the fuel consumption is reduced while the good vehicle-following precision is ensured, the SOC value is reduced more, and the fuel economy is obviously improved.
It should be noted that the above-mentioned embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, which should be covered by the claims of the present invention.
Claims (8)
1. A queue hybrid truck CACC energy management method based on deep migration learning is characterized by comprising the following steps:
constructing a deep migration learning neural network model based on the VGG network;
the deep migration learning neural network model consists of six elements and comprises two parallel networks, specifically comprising primary feature extraction and secondary feature extraction;
respectively extracting characteristic parameters of the short-time running condition segment of the previous stage as input of the model and characteristic data of an optimal SOC track corresponding to the short-time running condition segment of the next stage as output of the model by using the neural network model;
obtaining a reference SOC track based on the characteristic data of the optimal SOC track, and obtaining an optimal following vehicle speed track corresponding to a short-time running condition segment at the later stage of the train hybrid truck according to the state transition of a neural network;
according to the reference SOC track and the optimal following vehicle speed track, real-time prediction control is carried out on an actual SOC track to follow the optimal SOC track in the next-stage short-time driving working condition, and CACC energy management of the queue hybrid truck is achieved;
the real-time predictive control process includes,
inputting the optimal following vehicle speed track corresponding to the short-time running condition segment at the later stage of the train hybrid truck as a reference following vehicle speed track into model prediction control;
and (3) optimizing the disturbance variable by adding a disturbance observer and a Smith predictor by taking the acceleration of the vehicle, the torque of the motor and the rotating speed of the engine as control variables.
2. The deep migration learning based CACC energy management method for the train hybrid truck according to claim 1, characterized in that: the VGG network comprises 13 convolutional layers, 5 pooling layers, 3 full-connection layers and a Softmax layer.
3. The deep migration learning based CACC energy management method for the train hybrid truck according to claim 1, characterized in that: the first parallel network includes a first group of parallel networks,
M＝(V1,V2,D,C,P,S)
wherein, V1、V2Representing a feature extraction function, D representing a full-join function, C representing a convolution function, P representing a pooling function, and S representing a Softmax classification function.
4. The CACC energy management method of the train hybrid electric truck based on the deep migration learning of any one of claims 1 to 3, wherein: the feature extraction function includes a function of extracting a feature,
V1(I)＝R
wherein R is ∈ RE multiplied by F, V1(.) represents a feature space mapping, the V1(.) transforming the input operating condition segment characteristic parameter information I into a characteristic of E x F dimension.
5. The deep migration learning based CACC energy management method for the queue hybrid truck according to claim 3, characterized in that: the full-concatenation function includes a function of,
D(R)＝T
where D (.) converts the E × F dimensional data to the EF × 1 dimensional data.
6. The deep migration learning based CACC energy management method for the queue hybrid truck according to claim 3, characterized in that: the convolution function includes a function of a function including,
where i and j represent the row and column indices of the matrix, K represents the convolution kernel, and P represents the pooling function.
7. The deep migration learning based CACC energy management method for the queue hybrid truck according to claim 3, characterized in that: the pooling function may include a pooling function of,
8. the deep migration learning based train hybrid truck CACC energy management method according to claim 7, characterized in that: the extraction method of the characteristic parameters of the short-time running condition segment in the previous stage comprises environment perception, DSPC, GIS and GPS.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011617231.9A CN112660127B (en) | 2020-12-30 | 2020-12-30 | CACC energy management method for queue hybrid truck based on deep migration learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011617231.9A CN112660127B (en) | 2020-12-30 | 2020-12-30 | CACC energy management method for queue hybrid truck based on deep migration learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112660127A CN112660127A (en) | 2021-04-16 |
| CN112660127B true CN112660127B (en) | 2022-04-22 |
Family
ID=75411435
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011617231.9A Active CN112660127B (en) | 2020-12-30 | 2020-12-30 | CACC energy management method for queue hybrid truck based on deep migration learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112660127B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109895760A (en) * | 2019-04-03 | 2019-06-18 | 大连理工大学 | Plug-in hybrid-power automobile energy management method based on SOC reference locus |
| JP2020035182A (en) * | 2018-08-30 | 2020-03-05 | トヨタ自動車株式会社 | Controller and control method |
| CN112084700A (en) * | 2020-08-06 | 2020-12-15 | 南京航空航天大学 | Hybrid power system energy management method based on A3C algorithm |
| CN112101684A (en) * | 2020-11-10 | 2020-12-18 | 北京理工大学 | Plug-in hybrid electric vehicle real-time energy management method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101838512B1 (en) * | 2017-04-04 | 2018-03-14 | 현대자동차주식회사 | Hybrid vehicle and method of controlling charge mode |
-
2020
- 2020-12-30 CN CN202011617231.9A patent/CN112660127B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2020035182A (en) * | 2018-08-30 | 2020-03-05 | トヨタ自動車株式会社 | Controller and control method |
| CN109895760A (en) * | 2019-04-03 | 2019-06-18 | 大连理工大学 | Plug-in hybrid-power automobile energy management method based on SOC reference locus |
| CN112084700A (en) * | 2020-08-06 | 2020-12-15 | 南京航空航天大学 | Hybrid power system energy management method based on A3C algorithm |
| CN112101684A (en) * | 2020-11-10 | 2020-12-18 | 北京理工大学 | Plug-in hybrid electric vehicle real-time energy management method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112660127A (en) | 2021-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021103625A1 (en) | Short-term vehicle speed condition real-time prediction method based on interaction between vehicle ahead and current vehicle | |
| CN110936949B (en) | Energy control method, equipment, storage medium and device based on driving condition | |
| CN111267830B (en) | Hybrid power bus energy management method, device and storage medium | |
| CN106427589A (en) | Electric car driving range estimation method based on prediction of working condition and fuzzy energy consumption | |
| CN110717218B (en) | Electric drive vehicle distributed power drive system reconstruction control method and vehicle | |
| CN112668799A (en) | Intelligent energy management method and storage medium for PHEV (Power electric vehicle) based on big driving data | |
| CN108764298B (en) | Electric power image environment influence identification method based on single classifier | |
| CN113525396B (en) | Hybrid electric vehicle layered prediction energy management method integrating deep reinforcement learning | |
| CN112785845B (en) | Vehicle speed prediction method based on K-means clustering and RBF neural network | |
| CN116070783B (en) | Learning type energy management method of hybrid transmission system under commute section | |
| CN110490275A (en) | A kind of driving behavior prediction technique based on transfer learning | |
| CN112298155B (en) | Method for predicting energy consumption of hybrid power truck based on variable time domain model | |
| CN112949931A (en) | Method and device for predicting charging station data with hybrid data drive and model | |
| Lee et al. | Study on power management strategy of HEV using dynamic programming | |
| CN115107733A (en) | Energy management method and system for hybrid electric vehicle | |
| CN113479187B (en) | Layered different-step-length energy management method for plug-in hybrid electric vehicle | |
| CN113715629B (en) | Residual driving range prediction method based on improved symbolic regression and XGBoost algorithm | |
| CN115204041A (en) | Large aircraft approach and approach stage oil consumption prediction method | |
| CN112660127B (en) | CACC energy management method for queue hybrid truck based on deep migration learning | |
| Lin et al. | AER adaptive control strategy via energy prediction for PHEV | |
| Kang et al. | Logical control approach to fuel efficiency optimization for commuting vehicles | |
| CN111959491A (en) | Rule control parameter optimization method based on unified model of hybrid electric vehicle | |
| CN116901981A (en) | Online self-learning Markov vehicle speed prediction method | |
| CN113788007B (en) | Layered real-time energy management method and system | |
| CN112035536A (en) | Electric automobile energy consumption prediction method considering dynamic road network traffic flow |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240530 Address after: 230000 B-1015, wo Yuan Garden, 81 Ganquan Road, Shushan District, Hefei, Anhui. Patentee after: HEFEI MINGLONG ELECTRONIC TECHNOLOGY Co.,Ltd. Country or region after: China Address before: 250023 No. 5 Jiao Tong Road, Tianqiao District, Shandong, Ji'nan Patentee before: SHANDONG JIAOTONG University Country or region before: China |
|
| TR01 | Transfer of patent right |