CN115964959B - Domain centralized electronic and electric architecture modeling and multi-objective optimization method and system - Google Patents

Domain centralized electronic and electric architecture modeling and multi-objective optimization method and system Download PDF

Info

Publication number
CN115964959B
CN115964959B CN202310231043.XA CN202310231043A CN115964959B CN 115964959 B CN115964959 B CN 115964959B CN 202310231043 A CN202310231043 A CN 202310231043A CN 115964959 B CN115964959 B CN 115964959B
Authority
CN
China
Prior art keywords
ecu
swc
objective optimization
constraint
hardware
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
CN202310231043.XA
Other languages
Chinese (zh)
Other versions
CN115964959A (en
Inventor
田子昂
孔国杰
于会龙
席军强
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.)
Beijing Institute of Technology BIT
Original Assignee
Beijing Institute of Technology BIT
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 Beijing Institute of Technology BIT filed Critical Beijing Institute of Technology BIT
Priority to CN202310231043.XA priority Critical patent/CN115964959B/en
Publication of CN115964959A publication Critical patent/CN115964959A/en
Application granted granted Critical
Publication of CN115964959B publication Critical patent/CN115964959B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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

Landscapes

  • Supply And Distribution Of Alternating Current (AREA)
  • Small-Scale Networks (AREA)

Abstract

The invention discloses a domain centralized electronic and electric architecture modeling and multi-objective optimization method and system, which relate to the technical field of electronic and electric architecture modeling and optimization, and are used for respectively establishing a communication relation adjacency matrix of a software layer, a hardware structure adjacency matrix of a hardware layer and an SWC-ECU distribution relation matrix so as to model an electronic and electric architecture. And then, establishing a multi-objective optimization model, wherein the objective of the multi-objective optimization model is to minimize a cost value, a load rate and a time delay value, and solving the multi-objective optimization model by utilizing a multi-objective optimization algorithm to obtain an optimal solution of a hardware structure adjacent matrix and an SWC-ECU distribution relation matrix, so that the multi-objective optimization algorithm is adopted to carry out multi-objective optimization on the hardware structure based on cost, load and delay.

Description

Domain centralized electronic and electric architecture modeling and multi-objective optimization method and system
Technical Field
The invention relates to the technical field of modeling and optimizing of electronic and electric architecture, in particular to a domain centralized electronic and electric architecture modeling and multi-objective optimizing method and system of an unmanned special vehicle.
Background
At present, the important research and development of unmanned special vehicles are carried out at home and abroad. With the continuous development of unmanned, the number of sensors and actuators mounted on unmanned special vehicles is increasing, and the unmanned special vehicles are different from common civil vehicles, and more sensors and actuators than the common civil vehicles can be mounted on the unmanned special vehicles, so that more and more complex tasks besides automatic driving can be completed under more complex environments. Under such a background, the traditional distributed electronic and electric architecture can only deal with the increase of functions by continuously increasing the number of ECUs, which leads to the rapid increase of the number of wire harnesses in a vehicle, the arrangement of the wire harnesses is complex, the maintenance difficulty of the vehicle is increased while the performance of the vehicle is reduced, and the development of unmanned special vehicles is not facilitated.
According to the evolution roadmap of automobile EEA (Electrical Electronic Architecture, electronic and electric architecture) proposed by BOSCH in month 4 of 2018, the calculation power of the future automobile is more and more centralized, and the integration of multiple functions in the automobile into various domains is a current development trend. The unmanned special vehicle has multiple functions, a large number of sensors and actuators are installed, and the field centralized electronic and electric architecture is adopted to reduce the length of the wire harness, simplify the hardware connection mode and enhance the vehicle performance, so that the unmanned special vehicle is a necessary trend of future development. However, the current domain centralized electronic and electric architecture is often designed and optimized in a manual development mode, and in the aspect of design, the manual design is influenced by subjective factors, so that the whole vehicle design period is long and the universality of the architecture is poor; in the aspect of optimization, various indexes of the domain centralized electronic and electric architecture are mutually influenced, so that the optimization efficiency is low and the optimization effect is poor.
Based on this, there is a need for an automated domain-centric electronic electrical architecture modeling and multi-objective optimization technique.
Disclosure of Invention
The invention aims to provide a domain centralized electronic and electric architecture modeling and multi-objective optimization method and system, which utilize an adjacency matrix to carry out mathematical modeling on a hardware structure and a software architecture of a vehicle, and adopt a multi-objective optimization algorithm to carry out multi-objective optimization on the hardware structure based on cost, load and delay.
In order to achieve the above object, the present invention provides the following solutions:
a domain-centralized electronic electrical architecture modeling and multi-objective optimization method, comprising:
establishing a communication relation adjacency matrix of a software layer according to the signal transmission relation among the sensor, the SWC and the actuator; the elements of the ith row and the jth column of the communication relation adjacency matrix are used for representing the signal transmission relation of the ith first component and the jth first component, and the first components comprise a sensor, an SWC and an actuator;
establishing a hardware structure adjacency matrix of a hardware layer according to the hardware connection relation among the sensor, the public CAN node, the ECU and the actuator; establishing an SWC-ECU distribution relation matrix according to the connection relation between the SWC and the ECU; the hardware structure adjacent matrix ith row and jth column elements are used for representing hardware connection relations between an ith second component and a jth second component, and the second component comprises a sensor, a common CAN node, an ECU and an actuator;
establishing a multi-objective optimization model for optimizing the hardware structure adjacency matrix and the SWC-ECU distribution relation matrix; the multi-objective optimization model aims at minimizing cost values, load rates and delay values, and constraint conditions comprise delay value constraint, hardware requirement constraint and co-location constraint;
and solving the multi-objective optimization model by utilizing a multi-objective optimization algorithm to obtain an optimal solution of the hardware structure adjacency matrix and the SWC-ECU distribution relation matrix.
A domain-centralized electronic electrical architecture modeling and multi-objective optimization system, comprising:
the software layer modeling module is used for establishing a communication relation adjacency matrix of the software layer according to the signal transmission relation among the sensor, the SWC and the actuator; the elements of the ith row and the jth column of the communication relation adjacency matrix are used for representing the signal transmission relation of the ith first component and the jth first component, and the first components comprise a sensor, an SWC and an actuator;
the hardware layer modeling module is used for establishing a hardware structure adjacency matrix of the hardware layer according to the hardware connection relation among the sensor, the public CAN node, the ECU and the executor; establishing an SWC-ECU distribution relation matrix according to the connection relation between the SWC and the ECU; the hardware structure adjacent matrix ith row and jth column elements are used for representing hardware connection relations between an ith second component and a jth second component, and the second component comprises a sensor, a common CAN node, an ECU and an actuator;
the optimization model building module is used for building a multi-objective optimization model for optimizing the hardware structure adjacency matrix and the SWC-ECU distribution relation matrix; the multi-objective optimization model aims at minimizing cost values, load rates and delay values, and constraint conditions comprise delay value constraint, hardware requirement constraint and co-location constraint;
and the optimization solving module is used for solving the multi-objective optimization model by utilizing a multi-objective optimization algorithm to obtain an optimal solution of the hardware structure adjacency matrix and the SWC-ECU distribution relation matrix.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention is used for providing a domain centralized electronic and electric architecture modeling and multi-objective optimizing method and system, and establishing a communication relation adjacency matrix of a software layer according to a signal transmission relation among a sensor, an SWC and an actuator, establishing a hardware structure adjacency matrix of a hardware layer according to a hardware connection relation among the sensor, a public CAN node, an ECU and the actuator, and establishing an SWC-ECU distribution relation matrix according to the connection relation among the SWC and the ECU, thereby modeling the hardware structure and the software architecture of the electronic and electric architecture. And then, establishing a multi-objective optimization model, wherein the objective of the multi-objective optimization model is to minimize a cost value, a load rate and a time delay value, and solving the multi-objective optimization model by utilizing a multi-objective optimization algorithm to obtain an optimal solution of a hardware structure adjacent matrix and an SWC-ECU distribution relation matrix, so that the multi-objective optimization algorithm is adopted to carry out multi-objective optimization on the hardware structure based on cost, load and delay.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are needed in the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of the method according to embodiment 1 of the present invention;
FIG. 2 is a schematic block diagram of the method according to embodiment 1 of the present invention;
FIG. 3 is a schematic diagram of a chromosome according to example 1 of the present invention;
fig. 4 is a plan view of a mounting board harness provided in embodiment 1 of the present invention;
fig. 5 is a bottom view of a mounting plate bottom harness provided in embodiment 1 of the present invention;
fig. 6 is a cross-sectional view of an inner wire harness provided in embodiment 1 of the present invention;
fig. 7 is a system block diagram of a system according to embodiment 2 of the present invention.
Symbol description:
1-a first lidar; 2-a second lidar; 3-a first vehicle-mounted camera; 4-a second vehicle-mounted camera; 5-a third vehicle-mounted camera; 6-a fourth onboard camera; 7-GPS; 8-a mechanical arm camera; 9-a mechanical arm; 10-a front wiring hole of the mounting plate; 11-a rear wiring hole of the mounting plate; 12-a first vehicle ECU; 13-a second vehicle ECU.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention aims to provide a domain centralized electronic and electric architecture modeling and multi-objective optimization method and system, which utilize an adjacency matrix to carry out mathematical modeling on a hardware structure and a software architecture of a vehicle, and adopt a multi-objective optimization algorithm to carry out multi-objective optimization on the hardware structure based on cost, load and delay.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Example 1:
because of the problems of design and optimization of the domain centralized electronic and electric architecture by adopting a manual development mode, the design and optimization of the domain centralized electronic and electric architecture by adopting a multi-objective optimization algorithm is a scientific and efficient means, not only can a design scheme for reference be provided, the design period of a vehicle be shortened, but also various indexes can be comprehensively optimized, or a certain index can be subjected to key optimization by setting a weight coefficient. Under such a background, the embodiment provides an unmanned special vehicle domain centralized electronic and electric architecture modeling and multi-objective optimization method, which comprises a method for carrying out mathematical modeling on a software architecture and a hardware structure of a vehicle EEA and a vehicle EEA multi-objective optimization algorithm, so that modeling and optimization problems of the current vehicle EEA can be well solved.
Based on this, the present embodiment is used to provide a domain-centralized electronic electrical architecture modeling and multi-objective optimization method, as shown in fig. 1 and 2, including:
s1: establishing a communication relation adjacency matrix of a software layer according to the signal transmission relation among the sensor, the SWC and the actuator; the elements of the ith row and the jth column of the communication relation adjacency matrix are used for representing the signal transmission relation of the ith first component and the jth first component, and the first components comprise a sensor, an SWC and an actuator;
the embodiment abstracts the whole vehicle electronic and electric architecture into a demand layer, a task layer, a software layer and a hardware layer for mathematical modeling. Before mathematical modeling is carried out on a software layer and a hardware layer, modeling preprocessing is carried out on an electronic and electric architecture, and a modeling preprocessing process is the mathematical modeling process of a demand layer and a task layer.
(1) Mathematical modeling of demand layer:
the modeling of the electronic and electric architecture of the whole vehicle starts from a demand layer, and the type and the number of hardware components (ECU, sensor and actuator) and software components (SWC) contained in the whole vehicle need to be determined at the level, namely, the mathematical modeling process of the demand layer comprises the following steps: the type and number of each third component included in the vehicle is determined, and the third component includes a sensor, a SWC, an actuator, and an ECU. In this embodiment, the total number of ECUs is set to |e|, the total number of sensors is set to |k|, the total number of actuators is set to |a|, and the total number of SWCs is set to |c|.
(2) Mathematical modeling of task layer:
after the mathematical modeling of the demand layer is completed, the third components need to be divided into task groups in the mathematical modeling of the task layer, i.e. the third components which together complete a certain vehicle function are divided into the same set. The task layer needs to count how many functions the whole vehicle shares, i.e. determine the task group number F, then for eachThe third components are uniquely numbered to distinguish the third components, and finally the third components are divided according to the actual function condition of the whole vehicle to obtain task groups
Figure SMS_1
. The mathematical modeling of the task layer includes: and dividing all third assemblies which jointly complete a certain vehicle function into the same task group to obtain a plurality of task groups with the same number as the total vehicle function.
And carrying out mathematical modeling on the demand layer and the task layer to complete modeling pretreatment of the vehicle electronic and electric architecture.
After the modeling preprocessing is completed, an overall communication relation adjacency matrix needs to be established in a software layer, and a whole vehicle network topology relation is generated to be used as a basis for analyzing the performance of the vehicle electronic and electric architecture. Because the connection relation between the software layer and the hardware layer of the whole vehicle electronic and electric architecture is complex, in order to describe the whole vehicle architecture, adopting an adjacent matrix as the EEA representation of the whole vehicle is a feasible scheme. An adjacency matrix is a method of graphically representing a Boolean (0 or 1) matrix, where the directed graph or undirected graph is represented on a computer in the form of a square matrix, where the Boolean of the matrix represents whether a direct path exists between two vertices.
S1 may include:
(1) The signal transfer relationship of the sensor is input into the adjacency matrix.
And inputting the signal transmission relation related to the sensor into the adjacent matrix row by row according to the sequence of the serial numbers of the sensors determined in the task layer.
(2) The signaling relationship of the SWC is input into the adjacency matrix.
And inputting the signal transmission relation related to the SWCs into the adjacency matrix row by row according to the sequence of the SWC serial numbers determined in the task layer.
(3) The signal transfer relationship of the actuator is input into the adjacency matrix.
And inputting the signal transmission relation related to the actuator row by row into the adjacent matrix row by row according to the sequence of the serial numbers of the actuators determined in the task layer.
Through the above steps, an overall communication relation adjacency matrix S can be obtained, which is used for representing the specific transmission mode of the signals in the vehicle, and the calculation expression is as follows:
Figure SMS_2
；
s is a communication relation adjacency matrix of a software layer;
Figure SMS_3
element values for the j-th column of the i-th row of the communication relation adjacency matrix for characterizing the signaling relation of the i-th first component and the j-th first component, if the ith first component is signalled to the jth first component, then +.>
Figure SMS_4
1, otherwise->
Figure SMS_5
Is 0; />
Figure SMS_6
The method comprises the steps of carrying out a first treatment on the surface of the V is a first component comprising a sensor, an actuator and SWC; e is a collection of edges.
In this embodiment, the signaling relationship of the sensor and the first component corresponds to the front in the communication relationship adjacency matrix
Figure SMS_7
The position of the row, the signalling relationship of SWC and first component corresponds to the middle +_ in the communication relationship adjacency matrix>
Figure SMS_8
The position of the row, the signalling relationship of the actuator to the first component corresponds to the last +_ in the communication relationship adjacency matrix>
Figure SMS_9
The row positions, and the final communication relation adjacency matrix of the software layer are as follows: />
Figure SMS_10
Through the steps, mathematical modeling of a software layer of the vehicle electronic and electric architecture can be completed.
S2: establishing a hardware structure adjacency matrix of a hardware layer according to the hardware connection relation among the sensor, the public CAN node, the ECU and the actuator; establishing an SWC-ECU distribution relation matrix according to the connection relation between the SWC and the ECU; the hardware structure adjacent matrix ith row and jth column elements are used for representing hardware connection relations between an ith second component and a jth second component, and the second component comprises a sensor, a common CAN node, an ECU and an actuator;
the hardware layer is to complete the connection of the hardware components and distribute SWCs to the ECUs, the buses when the hardware components are connected CAN be selected to be CAN or Ethernet, and in order to ensure diversity, the components CAN be selected to be directly connected to the ECUs or connected to common CAN nodes (vehicle common CAN bus nodes), and the number of the common CAN nodes is set to be |N|. When connecting the assembly, provision is made to connect the bus to the ECU first and then to connect the assembly to the ECU or to a common CAN node.
The process of establishing the hardware structure adjacency matrix may include:
(1) The hardware connection relation of the sensor is input into the adjacency matrix.
And inputting the hardware connection relation related to the sensor into the adjacent matrix row by row according to the sequence of the serial numbers of the sensors determined in the task layer.
(2) And inputting the hardware connection relation of the common CAN nodes into an adjacent matrix.
And inputting the hardware connection relation related to the public CAN nodes into the adjacent matrix row by row according to the sequence number of the public CAN nodes currently designed by the vehicle.
(3) The hardware connection relation of the ECU is input into the adjacency matrix.
And inputting the hardware connection relation related to the ECU into the adjacent matrix row by row according to the sequence of the ECU serial numbers of the current design of the vehicle.
(4) The hardware connection relation of the actuator is input into the adjacency matrix.
And inputting the hardware connection relation related to the actuator row by row into the adjacency matrix row by row according to the sequence of the serial numbers of the actuators determined in the task layer.
Through the steps, the hardware structure adjacent matrix H of the whole vehicle can be obtained, and the calculation expression is as follows:
Figure SMS_11
；/>
wherein H is a hardware structure adjacency matrix of the hardware layer;
Figure SMS_12
for the element value of the ith row and the jth column of the hardware structure adjacency matrix, the element value is used for representing the hardware connection relation between the ith second component and the jth second component, if the ith second component is in hardware connection with the jth second component, < ->
Figure SMS_13
W is otherwise->
Figure SMS_14
Is 0;
Figure SMS_15
the method comprises the steps of carrying out a first treatment on the surface of the w represents a bus class, w=1, and is a CAN bus; w=2, ethernet bus; v is a second component, comprising a sensor, a common CAN node, an ECU and an actuator; e is a collection of edges.
The hardware connection of the sensor to the second component corresponds to the front in the hardware structure adjacency matrix
Figure SMS_16
The row position, the hardware connection of the common CAN node to the second component corresponds to the middle after the sensor in the hardware structure adjacency matrix ∈>
Figure SMS_17
The position of the row, the hardware connection of the ECU to the second component corresponds to the middle after the common CAN node in the hardware configuration adjacency matrix ∈>
Figure SMS_18
The position of the row, the hardware connection of the actuator to the second component corresponds to the last +_ in the hardware structure adjacency matrix>
Figure SMS_19
The row positions, the resulting hardware structure adjacency matrix H of the hardware layer representing the connection mode of the hardware structure in the vehicle is as follows:
Figure SMS_20
the SWC-ECU distribution relation matrix establishment process comprises the following steps:
and distributing the SWCs to the ECUs row by row according to the sequence number sequence of the SWCs in the task layer to obtain an SWC-ECU distribution relation matrix. The SWC-ECU distribution relation matrix has the following calculation expression:
Figure SMS_21
；
wherein U is SWC-ECU distribution relation matrix;
Figure SMS_22
assigning a relationship matrix of the element values of the ith row and the jth column to the SWC-ECU for characterizing whether the jth SWC is assigned to the ith ECU, if the jth SWC is assigned to the ith ECU->
Figure SMS_23
N is n, otherwise->
Figure SMS_24
Is 0; />
Figure SMS_25
The method comprises the steps of carrying out a first treatment on the surface of the n is the number of SWC, i.e. n=j; />
Figure SMS_26
Is the j-th SWC; />
Figure SMS_27
Is the ith ECU. />
Based on the above steps, the resulting SWC-ECU distribution relation matrix for characterizing the distribution relation of SWCs and ECUs in the vehicle is as follows:
Figure SMS_28
and the mathematical modeling of the hardware layer of the electronic and electric architecture of the vehicle can be completed by establishing a hardware structure adjacency matrix and an SWC-ECU distribution relation matrix of the whole vehicle.
It should be noted that the mathematical modeling process converts a known or artificially determined vehicle electronic and electric architecture into a mathematical model, so that the signal transmission relationship, the hardware connection relationship, and the distribution relationship between the SWC and the ECU are all known.
S3: establishing a multi-objective optimization model for optimizing the hardware structure adjacency matrix and the SWC-ECU distribution relation matrix; the multi-objective optimization model aims at minimizing cost values, load rates and delay values, and constraint conditions comprise delay value constraint, hardware requirement constraint and co-location constraint;
specifically, the wire harness cost of the whole vehicle is composed of a CAN wire and an Ethernet wire, and a first objective function of the multi-objective optimization model is as follows:
Figure SMS_29
；
wherein ,
Figure SMS_30
for individuals in the initial population->
Figure SMS_31
The cost of the whole wire harness; />
Figure SMS_32
Is the cost value of the CAN bus; />
Figure SMS_33
Cost value for Ethernet bus。
The second objective function of the multi-objective optimization model is:
Figure SMS_34
；
wherein ,
Figure SMS_35
is the bus load rate; b is a bus set, namely a bus set of the whole vehicle; />
Figure SMS_36
Is the weighted bus load of bus b.
The third objective function of the multi-objective optimization model is:
Figure SMS_37
；
wherein ,
Figure SMS_38
delay for bus signaling; />
Figure SMS_39
The transfer delay of the CAN bus; />
Figure SMS_40
Is the transfer delay of the Ethernet bus.
The constraint conditions include:
(1) Time delay value constraint:
for population individuals
Figure SMS_41
And (3) performing delay value constraint, wherein a constraint formula is as follows:
Figure SMS_42
；
wherein ,
Figure SMS_43
for time of arrivalValue-extending constraint; s is a signal set in a software architecture; s is the number of signals in the set S; />
Figure SMS_44
Is the worst-case propagation delay of signal s; />
Figure SMS_45
End-to-end time requirement for signal s; 1{P } represents a function that returns a 1 if P is true, and a 0 otherwise.
(2) Hardware demand constraint:
hardware requirement constraint is carried out on individuals in a population, and certain SWCs have specific allocation requirements on the ECU, wherein the constraint formula is as follows:
Figure SMS_46
；
wherein ,
Figure SMS_47
constraint for hardware requirements; c is the entire SWC set, i.e., the set consisting of all SWCs; the |C| is the number of SWCs in set C; />
Figure SMS_48
An ECU assigned to SWC; />
Figure SMS_49
ECU requiring specific allocation for SWC;
(3) Co-location constraint:
co-location demand constraint is carried out on population individuals, and certain SWCs have demands distributed on the same ECU, wherein the constraint formula is as follows:
Figure SMS_50
；
wherein ,
Figure SMS_51
is a co-location constraint; c is a set of all SWCs; the |C| is the number of SWCs in set C; />
Figure SMS_52
Is a collection of SWCs assigned to the ECU; />
Figure SMS_53
Is a set of SWCs required to be assigned to the same ECU.
S4: and solving the multi-objective optimization model by utilizing a multi-objective optimization algorithm to obtain an optimal solution of the hardware structure adjacency matrix and the SWC-ECU distribution relation matrix.
The multi-objective optimization algorithm of this embodiment is an NSGA-ii algorithm, that is, the NSGA-ii algorithm is adopted to optimize a hardware structure adjacency matrix and an SWC-ECU allocation relation matrix (that is, an SWC allocation scheme) of a hardware layer of an electronic and electric architecture of a whole vehicle, and S4 may include:
(1) Encoding the hardware structure adjacency matrix and the SWC-ECU distribution relation matrix into chromosomes; and carrying out random initialization operation on the chromosome to obtain an initial population.
In this embodiment, the hardware structure adjacency matrix of the hardware layer and the SWC-ECU allocation relation matrix are encoded into chromosomes, and the obtained chromosomes are shown in fig. 3. In fig. 3, ek, ea, ESWC shows the connection relationship between the sensor, the actuator, and the ECU, and the distribution relationship of the SWC to the ECU, respectively; hw represents the harness class of the bus; kz and Az represent the routing hole assignment of the sensor and actuator harnesses, respectively, on the mounting plate. In the chromosome
Figure SMS_54
Each index i corresponds to the number of a component (sensor, actuator, SWC) or harness, variable +.>
Figure SMS_55
The value of (2) corresponds to the connection relation, distribution relation or class of the wire harness of the component i. />
Figure SMS_56
The value of (c) can only be an integer and uniquely represents an allocation. Wherein Ek, ESWC, ea varies in the range of
Figure SMS_57
The variation range of Hw, kz, az is +.>
Figure SMS_58
. To reduce the search space, the co-located groups of SWCs are encoded by a single gene in the chromosome, and each variable +.>
Figure SMS_59
Are represented by one gene in the chromosome.
The acquisition process of the initial population is as follows: the chromosome is subjected to random initialization operation, and the gene values in the chromosome are given random integer values within a variation range so as to generate an initial population.
Through the steps, the hardware structure adjacent matrix to be optimized and the SWC-ECU distribution relation matrix can be encoded to generate a chromosome of the NSGA-II multi-objective optimization algorithm, and the initial population is further randomly generated according to the chromosome and input into the NSGA-II algorithm.
(2) Calculating a target value of each individual in the initial population under the constraint of the constraint condition by utilizing a multi-target optimization model; the target values include a cost value, a load factor, and a delay value.
The specific process of calculating the cost value, the load rate and the time delay value of each individual in the initial population is as follows:
(2.1) solving for cost values for the initial population:
the first objective function of the multi-objective optimization model provided in S3 may be used to calculate a cost value, where the hardware arrangement of the special vehicle is shown in fig. 4, 5 and 6, and the vehicle is composed of a chassis and a top-loading part, where the top-loading part includes all sensors and actuators, and the top-loading part is mounted on a mounting plate fixed on the chassis by fasteners. The ECU is installed inside the chassis, and wiring harness between the two is connected together through the wiring hole on the mounting panel. In the figure, 1 is a first laser radar, 2 is a second laser radar, 3 is a first vehicle-mounted camera, 4 is a second vehicle-mounted camera, 5 is a third vehicle-mounted camera, 6 is a fourth vehicle-mounted camera, 7 is a GPS,8 is a mechanical arm camera, 9 is a mechanical arm, 10 is a front wiring hole of a mounting plate, 11 is a rear wiring hole of the mounting plate, 12 is a first vehicle-mounted ECU,13 is a second vehicle-mounted ECU, a dotted line represents a CAN line, and a thin solid line represents an Ethernet line.
1) Calculating the harness length of a public CAN bus
Figure SMS_60
。
The modules may be connected to a common CAN bus, in which case the aggregate harness length under the board to which the modules correspond is
Figure SMS_61
The calculation formula is as follows:
Figure SMS_62
；
wherein ,
Figure SMS_63
is the vertical height of the mounting plate; />
Figure SMS_64
Z coordinate of the ECU; />
Figure SMS_65
Y coordinates of the CAN bus; />
Figure SMS_66
Is the Y coordinate of the routing hole.
2) Calculating the wire harness length of CAN bus b
Figure SMS_67
。
The wire harness is divided into an upper part, a lower part and a vehicle edge part during calculation, wherein the wire harness length of the CAN bus b
Figure SMS_68
The calculation formula is as follows:
Figure SMS_69
；
wherein ,
Figure SMS_70
is the length on the board; />
Figure SMS_71
Is the length of the lower part of the plate and the edge of the vehicle.
The calculation formula of the length on the board is as follows:
Figure SMS_72
；
wherein ,
Figure SMS_73
、/>
Figure SMS_74
for X, Y coordinates of the component, +.>
Figure SMS_75
、/>
Figure SMS_76
X, Y coordinates of the routing holes.
The calculation formulas of the plate lower and the edge length are as follows:
Figure SMS_77
；
wherein ,
Figure SMS_78
y coordinates of the public CAN bus; />
Figure SMS_79
Is the vertical height of the mounting plate; />
Figure SMS_80
、/>
Figure SMS_81
X, Z coordinates of the ECU.
3) Calculation of EthernetLength of bus b
Figure SMS_82
。
Length of Ethernet bus b
Figure SMS_83
Is calculated in relation to the harness length of CAN bus b>
Figure SMS_84
The same way of calculation. />
4) Calculating the total cost value of the CAN bus
Figure SMS_85
。
Calculated from 1)
Figure SMS_86
And 2) calculated ∈>
Figure SMS_87
CAN calculate the cost value of CAN bus
Figure SMS_88
The following is shown:
Figure SMS_89
；
wherein ,
Figure SMS_90
is the price per unit length (unit: meta/m) of the CAN bus.
5) Calculating cost value of Ethernet bus
Figure SMS_91
。
Calculated from 3)
Figure SMS_92
The cost value of the Ethernet bus can be calculated>
Figure SMS_93
The following is shown:
Figure SMS_94
；
wherein ,
Figure SMS_95
is the price per unit length of the Ethernet bus.
6) Whole vehicle wire harness cost for calculating individual x in population
Figure SMS_96
。
Calculated from 4)
Figure SMS_97
And 5) calculated +.>
Figure SMS_98
The cost of the whole wire harness of the individual x in the calculation population can be calculated>
Figure SMS_99
The following is shown:
Figure SMS_100
。
(2.2) solving the load factor of the initial population.
1) Calculating the load factor of bus b
Figure SMS_101
。
Figure SMS_102
For unbiased measurement of b bus occupied bus bandwidth, the calculation formula is as follows:
Figure SMS_103
；
wherein ,
Figure SMS_104
is a signal set transmitted through the CAN bus b; />
Figure SMS_105
Is an overhead factor, i.e. the ratio between the size of the entire frame and the payload size; />
Figure SMS_106
Is the bit size of the signal s; />
Figure SMS_107
Is the bit rate of the bus; />
Figure SMS_108
Is the cycle time of the signal on bus b.
2) Calculating the weighted load rate of the bus b
Figure SMS_109
。
Calculated from 1)
Figure SMS_110
The weighted bus load of bus b can be calculated>
Figure SMS_111
The calculation formula is as follows:
Figure SMS_112
；
wherein ,
Figure SMS_113
is the weight.
3) Calculating bus load rate of individuals x in population
Figure SMS_114
。/>
The weighted bus load calculated by 2)
Figure SMS_115
The bus load rate of the individuals x in the population can be calculated
Figure SMS_116
The following is shown:
Figure SMS_117
。
(2.3) solving the time delay value of the initial population.
1) Calculating the transmission time of the signal s on the CAN bus b
Figure SMS_118
。
Transmission time of signal s on CAN bus b
Figure SMS_119
The calculation formula of (2) is as follows:
Figure SMS_120
；
wherein ,
Figure SMS_121
for protocol overhead, by no data frame length +.>
Figure SMS_122
And bit number->
Figure SMS_123
Composition; />
Figure SMS_124
Is the signal length; />
Figure SMS_125
Is the CAN line transmission rate.
2) Calculating the latency of signal s on CAN bus b
Figure SMS_126
。
Signal s at CLatency of AN bus b
Figure SMS_127
The calculation formula of (2) is as follows:
Figure SMS_128
；
3) Calculating the transfer delay of a CAN bus
Figure SMS_129
。
From 1)
Figure SMS_130
And 2) the>
Figure SMS_131
Can calculate the transfer delay of CAN bus +.>
Figure SMS_132
The following is shown:
Figure SMS_133
；
wherein ,
Figure SMS_134
is a signal set transmitted through a CAN bus B, and B is a bus set of the whole vehicle.
4) Calculating the transmission time of the signal s on the Etherent bus b
Figure SMS_135
。
For the Ethernet bus, the present embodiment will wait for a time due to its high transmission efficiency
Figure SMS_136
Transmission time->
Figure SMS_137
The calculation formula of (2) is as follows:
Figure SMS_138
；
wherein ,
Figure SMS_139
representing the length of the data block; />
Figure SMS_140
Is the Ethernet line transmission rate.
5) Calculating transfer delay of Ethernet bus
Figure SMS_141
。
Calculated from 4)
Figure SMS_142
The transfer delay of the Ethernet bus can be determined>
Figure SMS_143
The following is shown:
Figure SMS_144
；/>
6) Calculating bus signaling delays for individuals x in a population
Figure SMS_145
。
Calculated from 3)
Figure SMS_146
And 5) calculated +.>
Figure SMS_147
The bus signaling delay of individual x in the population can be calculated>
Figure SMS_148
The following is shown:
Figure SMS_149
。
in the above formula, the known quantities include: vertical height of mounting board, wiring hole coordinates, position coordinates of each hardware component, signal length (bit size), unit length price of each bus, bit rate of each bus, cycle time of signal on bus, overhead factor,
Figure SMS_150
、S、Ds、C、/>
Figure SMS_151
、/>
Figure SMS_152
。
The target value of each individual in the initial population can be calculated by using the formula, and when the optimization is performed on three items of cost value, load rate and time delay value, the optimized constraint condition needs to be set so as to calculate the constraint violation condition according to the three constraint conditions.
(3) Screening the initial population by using a tournament selection method based on the target value of each individual in the initial population to obtain a screened population; carrying out two-point crossover mutation, exchange mutation and point mutation on the screened population to obtain a offspring population; combining the initial population and the offspring population to obtain a combined population; and non-dominant sorting is carried out on the combined population according to the target value, and the updated population is obtained.
In the embodiment, the cost value, the load rate and the time delay value are used as optimization targets of the population, and the set time delay value, the hardware requirement and the co-location are used as optimization constraint conditions for iterative optimization. Selecting an initial population by using a tournament selection method, carrying out two-point crossover mutation, exchange mutation and point mutation on chromosomes of winning individuals in the selected population to obtain a child population, combining the initial population and the child population, carrying out non-dominant sorting on the individuals in the combined population according to a target value, calculating the crowdedness of the obtained population, and selecting individuals to form a new parent according to the crowdedness, thus obtaining an updated population.
(4) Judging whether an iteration termination condition is reached;
the iteration termination condition in this embodiment may be the maximum iteration number, and determining whether the iteration termination condition is reached is equivalent to determining whether the current iteration number reaches the maximum iteration number.
(5) If yes, outputting the updated optimal individuals of the population, wherein the optimal individuals are the optimal solutions of the hardware structure adjacent matrix and the SWC-ECU distribution relation matrix.
According to the updated target values of the individuals in the population, the chromosomes of the proper individuals are selected and decoded into the optimal solution of the hardware structure of the vehicle and output, and when the unmanned special vehicle is designed, the optimal solution can be used as a reference model to optimize the electronic and electric architecture of the actual vehicle, so that the optimization speed of the electronic and electric architecture is increased, and the optimization effect is enhanced.
(6) If not, the updated population is used as the initial population of the next iteration, and the step of calculating the target value of each individual in the initial population under the constraint of the constraint condition by utilizing the multi-target optimization model is returned.
The embodiment provides a domain centralized electronic electric architecture modeling and multi-objective optimization method for an unmanned special vehicle, which models a hardware layer and a software layer of the vehicle electronic electric architecture by utilizing an adjacency matrix, adopts a multi-objective optimization algorithm based on NSGA-II to perform multi-objective optimization aiming at cost, load rate and time delay value on the unmanned special vehicle domain centralized electronic electric architecture, comprehensively improves various performance indexes of the electronic electric architecture, and can output a referential architecture scheme, so that the design and optimization of the architecture are more reasonable and efficient.
Example 2:
the present embodiment is used to provide a domain-centralized electronic electrical architecture modeling and multi-objective optimization system, as shown in fig. 7, including:
the software layer modeling module M1 is used for establishing a communication relation adjacency matrix of the software layer according to the signal transmission relation among the sensor, the SWC and the actuator; the elements of the ith row and the jth column of the communication relation adjacency matrix are used for representing the signal transmission relation of the ith first component and the jth first component, and the first components comprise a sensor, an SWC and an actuator;
the hardware layer modeling module M2 is used for establishing a hardware structure adjacency matrix of the hardware layer according to the hardware connection relation among the sensor, the public CAN node, the ECU and the executor; establishing an SWC-ECU distribution relation matrix according to the connection relation between the SWC and the ECU; the hardware structure adjacent matrix ith row and jth column elements are used for representing hardware connection relations between an ith second component and a jth second component, and the second component comprises a sensor, a common CAN node, an ECU and an actuator;
an optimization model building module M3, which is used for building a multi-objective optimization model for optimizing the hardware structure adjacency matrix and the SWC-ECU distribution relation matrix; the multi-objective optimization model aims at minimizing cost values, load rates and delay values, and constraint conditions comprise delay value constraint, hardware requirement constraint and co-location constraint;
and the optimization solving module M4 is used for solving the multi-objective optimization model by utilizing a multi-objective optimization algorithm to obtain an optimal solution of the hardware structure adjacency matrix and the SWC-ECU distribution relation matrix.
The same or similar parts between the various embodiments in this specification are referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The principles and embodiments of the present invention have been described herein with reference to specific examples, the description of which is intended only to assist in understanding the methods of the present invention and the core ideas thereof; also, it is within the scope of the present invention to be modified by those of ordinary skill in the art in light of the present teachings. In view of the foregoing, this description should not be construed as limiting the invention.

Claims (6)

1. A domain-centralized electronic electrical architecture modeling and multi-objective optimization method, comprising:
establishing a communication relation adjacency matrix of a software layer according to the signal transmission relation among the sensor, the SWC and the actuator; the communication relation adjacency matrix is the firstiLine 1jThe elements of the columns being used to characterize the firstiFirst component and second componentjA signaling relationship for a first component, the first component comprising a sensor, a SWC, and an actuator;
establishing a hardware structure adjacency matrix of a hardware layer according to the hardware connection relation among the sensor, the public CAN node, the ECU and the actuator; establishing an SWC-ECU distribution relation matrix according to the connection relation between the SWC and the ECU; the hardware structure adjacency matrix is the firstiLine 1jThe elements of the columns being used to characterize the firstiA second component and a secondjThe hardware connection relation of the second components comprises a sensor, a common CAN node, an ECU and an actuator;
establishing a multi-objective optimization model for optimizing the hardware structure adjacency matrix and the SWC-ECU distribution relation matrix; the multi-objective optimization model aims at minimizing cost values, load rates and delay values, and constraint conditions comprise delay value constraint, hardware requirement constraint and co-location constraint;
solving the multi-objective optimization model by utilizing a multi-objective optimization algorithm to obtain an optimal solution of the hardware structure adjacency matrix and the SWC-ECU distribution relation matrix;
the first objective function of the multi-objective optimization model is:
Figure QLYQS_1
；
wherein ,
Figure QLYQS_2
the cost of the whole wire harness is; />
Figure QLYQS_3
Is the cost value of the CAN bus; />
Figure QLYQS_4
The cost value of the Ethernet bus;
the second objective function of the multi-objective optimization model is:
Figure QLYQS_5
；
wherein ,
Figure QLYQS_6
is the bus load rate;Bis a bus set; />
Figure QLYQS_7
Is a busbIs a weighted bus load;
the third objective function of the multi-objective optimization model is:
Figure QLYQS_8
；
wherein ,
Figure QLYQS_9
delay for bus signaling; />
Figure QLYQS_10
The transfer delay of the CAN bus; />
Figure QLYQS_11
The transfer delay of the Ethernet bus;
the delay value constraint is as follows:
Figure QLYQS_12
；
wherein ,
Figure QLYQS_13
constraint is carried out on a time delay value;Sis a signal set in a software architecture; |SI is the signal setSThe number of medium signals;/>
Figure QLYQS_14
is a signalsIs a worst case transfer delay of (1); />
Figure QLYQS_15
Is a signalsEnd-to-end time requirements of (2);
the hardware requirement constraint is:
Figure QLYQS_16
；
wherein ,
Figure QLYQS_17
constraint for hardware requirements;Cis a set of all SWCs; |CI is the collectionCThe number of SWCs in (B); />
Figure QLYQS_18
An ECU assigned to SWC; />
Figure QLYQS_19
ECU requiring specific allocation for SWC; />
The co-location constraint is:
Figure QLYQS_20
；
wherein ,
Figure QLYQS_21
is a co-location constraint;Cis a set of all SWCs; |CI is the collectionCThe number of SWCs in (B); />
Figure QLYQS_22
Is a collection of SWCs assigned to the ECU; />
Figure QLYQS_23
Is a set of SWCs required to be assigned to the same ECU.
2. The domain-centralized electronic-electrical-architecture modeling and multi-objective optimization method of claim 1, further comprising, prior to establishing the communication-relationship adjacency matrix of the software layer based on the signaling relationships between the sensors, SWCs, and actuators:
determining the type and the number of each third component contained in the vehicle; the third component includes a sensor, a SWC, an actuator, and an ECU.
3. The domain centralized electronic electrical architecture modeling and multi-objective optimization method of claim 2, further comprising, after determining the type and number of each third component that the vehicle contains:
and dividing all the third assemblies which jointly complete a certain vehicle function into the same task group to obtain a plurality of task groups with the same number as the total vehicle function.
4. The domain-centralized electronic-electrical-architecture modeling and multi-objective optimization method of claim 1, wherein the multi-objective optimization algorithm is an NSGA-ii algorithm.
5. The method for modeling and optimizing multi-objective electronic and electrical architecture according to claim 4, wherein the solving the multi-objective optimization model by using a multi-objective optimization algorithm to obtain an optimal solution of the hardware structure adjacency matrix and the SWC-ECU allocation relation matrix specifically comprises:
encoding the hardware structure adjacency matrix and the SWC-ECU distribution relation matrix into chromosomes; carrying out random initialization operation on the chromosome to obtain an initial population;
calculating a target value of each individual in the initial population under the constraint of the constraint condition by utilizing the multi-target optimization model; the target values comprise cost values, load rates and delay values;
screening the initial population by using a tournament selection method based on the target value of each individual in the initial population to obtain a screened population; carrying out two-point crossover mutation, exchange mutation and point mutation on the screened population to obtain a offspring population; combining the initial population and the offspring population to obtain a combined population; non-dominant sorting is carried out on the combined population according to the target value, and an updated population is obtained;
judging whether an iteration termination condition is reached;
if yes, outputting the optimal individuals of the updated population, wherein the optimal individuals are the optimal solutions of the hardware structure adjacent matrix and the SWC-ECU distribution relation matrix;
if not, the updated population is used as an initial population of the next iteration, and the step of calculating the target value of each individual in the initial population under the constraint of the constraint condition by utilizing the multi-target optimization model is returned.
6. A domain-centralized electronic electrical architecture modeling and multi-objective optimization system, comprising:
the software layer modeling module is used for establishing a communication relation adjacency matrix of the software layer according to the signal transmission relation among the sensor, the SWC and the actuator; the communication relation adjacency matrix is the firstiLine 1jThe elements of the columns being used to characterize the firstiFirst component and second componentjA signaling relationship for a first component, the first component comprising a sensor, a SWC, and an actuator;
the hardware layer modeling module is used for establishing a hardware structure adjacency matrix of the hardware layer according to the hardware connection relation among the sensor, the public CAN node, the ECU and the executor; establishing an SWC-ECU distribution relation matrix according to the connection relation between the SWC and the ECU; the hardware structure adjacency matrix is the firstiLine 1jThe elements of the columns being used to characterize the firstiA second component and a secondjThe hardware connection relation of the second components comprises a sensor, a common CAN node, an ECU and an actuator;
the optimization model building module is used for building a multi-objective optimization model for optimizing the hardware structure adjacency matrix and the SWC-ECU distribution relation matrix; the multi-objective optimization model aims at minimizing cost values, load rates and delay values, and constraint conditions comprise delay value constraint, hardware requirement constraint and co-location constraint;
the optimization solving module is used for solving the multi-objective optimization model by utilizing a multi-objective optimization algorithm to obtain an optimal solution of the hardware structure adjacency matrix and the SWC-ECU distribution relation matrix;
the first objective function of the multi-objective optimization model is:
Figure QLYQS_24
；
wherein ,
Figure QLYQS_25
the cost of the whole wire harness is; />
Figure QLYQS_26
Is the cost value of the CAN bus; />
Figure QLYQS_27
The cost value of the Ethernet bus;
the second objective function of the multi-objective optimization model is:
Figure QLYQS_28
；
wherein ,
Figure QLYQS_29
is the bus load rate;Bis a bus set; />
Figure QLYQS_30
Is a busbIs a weighted bus load;
the third objective function of the multi-objective optimization model is:
Figure QLYQS_31
；
wherein ,
Figure QLYQS_32
delay for bus signaling; />
Figure QLYQS_33
The transfer delay of the CAN bus; />
Figure QLYQS_34
The transfer delay of the Ethernet bus;
the delay value constraint is as follows:
Figure QLYQS_35
；
wherein ,
Figure QLYQS_36
constraint is carried out on a time delay value;Sis a signal set in a software architecture; |SI is the signal setSThe number of medium signals; />
Figure QLYQS_37
Is a signalsIs a worst case transfer delay of (1); />
Figure QLYQS_38
Is a signalsEnd-to-end time requirements of (2);
the hardware requirement constraint is:
Figure QLYQS_39
；
wherein ,
Figure QLYQS_40
constraint for hardware requirements;Cis a set of all SWCs; |CI is the collectionCThe number of SWCs in (B); />
Figure QLYQS_41
An ECU assigned to SWC; />
Figure QLYQS_42
ECU requiring specific allocation for SWC;
the co-location constraint is:
Figure QLYQS_43
；
wherein ,
Figure QLYQS_44
is a co-location constraint;Cis a set of all SWCs; |CI is the collectionCThe number of SWCs in (B); />
Figure QLYQS_45
Is a collection of SWCs assigned to the ECU; />
Figure QLYQS_46
Is a set of SWCs required to be assigned to the same ECU. />
CN202310231043.XA 2023-03-13 2023-03-13 Domain centralized electronic and electric architecture modeling and multi-objective optimization method and system Active CN115964959B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310231043.XA CN115964959B (en) 2023-03-13 2023-03-13 Domain centralized electronic and electric architecture modeling and multi-objective optimization method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310231043.XA CN115964959B (en) 2023-03-13 2023-03-13 Domain centralized electronic and electric architecture modeling and multi-objective optimization method and system

Publications (2)

Publication Number Publication Date
CN115964959A CN115964959A (en) 2023-04-14
CN115964959B true CN115964959B (en) 2023-05-30

Family

ID=85889881

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310231043.XA Active CN115964959B (en) 2023-03-13 2023-03-13 Domain centralized electronic and electric architecture modeling and multi-objective optimization method and system

Country Status (1)

Country Link
CN (1) CN115964959B (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107813816A (en) * 2016-09-12 2018-03-20 法乐第(北京)网络科技有限公司 Energy hole track optimizing equipment, hybrid vehicle for hybrid vehicle

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2163735A1 (en) * 2008-08-21 2010-03-17 C.R.F. Società Consortile per Azioni System and method for multi-objective management of the electrical and thermal energy gernerated by a co/trigeneration energy system in a multi-source energy plant
CN110866332B (en) * 2019-10-29 2022-12-16 中国电子科技集团公司第三十八研究所 Complex cable assembly assembling method and system
CN111352650B (en) * 2020-02-25 2023-06-30 杭州电子科技大学 Software modularization multi-objective optimization method and system based on INSSGA-II
EP4030370A1 (en) * 2021-01-13 2022-07-20 Tata Consultancy Services Limited Method and system for fleet route optimization
CN112966805B (en) * 2021-03-15 2022-10-18 河海大学 Reservoir scheduling multi-objective optimization method based on graph convolution neural network and NSGA-II algorithm
CN113343349B (en) * 2021-05-13 2022-11-11 武汉理工大学 Multi-objective optimization method, equipment and storage medium for automotive electronic and electrical architecture

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107813816A (en) * 2016-09-12 2018-03-20 法乐第(北京)网络科技有限公司 Energy hole track optimizing equipment, hybrid vehicle for hybrid vehicle

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
无参分组大规模变量的多目标算法研究;朱登京;段倩倩;;计算机工程与科学(04);全文 *

Also Published As

Publication number Publication date
CN115964959A (en) 2023-04-14

Similar Documents

Publication Publication Date Title
Rosberg et al. Optimal control of service in tandem queues
CN112265546B (en) Networked automobile speed prediction method based on time-space sequence information
CN109190802B (en) Multi-microgrid game optimization method based on power generation prediction in cloud energy storage environment
CN112767682A (en) Multi-scale traffic flow prediction method based on graph convolution neural network
CN112766597B (en) Bus passenger flow prediction method and system
CN108632764B (en) Multi-sensor selective measurement data fusion estimation method
CN115473901B (en) Distributed computing power cluster intelligent scheduling method and device and computer equipment
CN110647398A (en) Intersection control task scheduling method facing edge calculation and based on task criticality and timeliness
CN113259469A (en) Edge server deployment method, system and storage medium in intelligent manufacturing
CN108764510B (en) Urban rail transit parallel simulation task decomposition method facing large-scale road network
CN115964959B (en) Domain centralized electronic and electric architecture modeling and multi-objective optimization method and system
CN109118412B (en) Urban rail transit network passenger flow online control system
CN113365229B (en) Network time delay optimization method of multi-union chain consensus algorithm
CN112785770B (en) Dynamic entity queuing model construction method based on time series
CN113422795A (en) Vehicle-mounted edge task centralized scheduling and resource allocation joint optimization method based on deep reinforcement learning
CN109889573B (en) NGSA multi-target-based copy placement method in hybrid cloud
CN113628435A (en) Information processing method and device
CN116301919A (en) Chip management method, device, equipment and storage medium in vehicle
CN109376954A (en) Concrete tank quantity optimization Model Calculating Method and computer-readable medium
CN111210088B (en) Traffic state index prediction method based on space-time factors
CN113420926A (en) Method for compiling multi-traffic mode transportation capacity collaborative optimization scheme between urban hubs
CN115660324B (en) Power grid multi-section out-of-limit regulation and control method and system based on graph reinforcement learning
CN117041040B (en) Command center distributed layout system based on intelligent scheduling algorithm
CN112883020B (en) Big data application-based analysis and management system
CN112418715B (en) Method and device for generating wind power sequence scene set and storage medium

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