CN115660024A - Vehicle-mounted network fault diagnosis method based on intelligent optimization algorithm and machine learning - Google Patents
Vehicle-mounted network fault diagnosis method based on intelligent optimization algorithm and machine learning Download PDFInfo
- Publication number
- CN115660024A CN115660024A CN202211088548.7A CN202211088548A CN115660024A CN 115660024 A CN115660024 A CN 115660024A CN 202211088548 A CN202211088548 A CN 202211088548A CN 115660024 A CN115660024 A CN 115660024A
- Authority
- CN
- China
- Prior art keywords
- gull
- optimal
- fault diagnosis
- vehicle
- network fault
- 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.)
- Granted
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 49
- 238000005457 optimization Methods 0.000 title claims abstract description 30
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000010801 machine learning Methods 0.000 title claims abstract description 17
- 241000272168 Laridae Species 0.000 claims abstract description 91
- 230000006870 function Effects 0.000 claims abstract description 32
- 238000013507 mapping Methods 0.000 claims abstract description 11
- 230000007246 mechanism Effects 0.000 claims abstract description 10
- 230000002457 bidirectional effect Effects 0.000 claims abstract description 7
- 239000013598 vector Substances 0.000 claims abstract description 5
- 230000006399 behavior Effects 0.000 claims description 17
- 239000003016 pheromone Substances 0.000 claims description 9
- 238000013508 migration Methods 0.000 claims description 6
- 230000005012 migration Effects 0.000 claims description 6
- 230000035772 mutation Effects 0.000 claims description 6
- 230000000739 chaotic effect Effects 0.000 claims description 4
- 241000272161 Charadriiformes Species 0.000 claims description 3
- YZCKVEUIGOORGS-IGMARMGPSA-N Protium Chemical compound [1H] YZCKVEUIGOORGS-IGMARMGPSA-N 0.000 claims description 3
- 238000002790 cross-validation Methods 0.000 claims description 3
- 229910052731 fluorine Inorganic materials 0.000 claims description 2
- 125000001153 fluoro group Chemical group F* 0.000 claims description 2
- 238000012360 testing method Methods 0.000 abstract description 5
- 230000007547 defect Effects 0.000 abstract description 3
- 230000000694 effects Effects 0.000 abstract description 3
- 238000012706 support-vector machine Methods 0.000 description 19
- 238000012549 training Methods 0.000 description 5
- 241000364019 Theisoa Species 0.000 description 3
- 239000000243 solution Substances 0.000 description 3
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 description 2
- 108700026140 MAC combination Proteins 0.000 description 2
- 229910052739 hydrogen Inorganic materials 0.000 description 2
- 239000001257 hydrogen Substances 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000019637 foraging behavior Effects 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
- 238000012795 verification Methods 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
Landscapes
- Feedback Control In General (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a vehicle-mounted network fault diagnosis method based on an intelligent optimization algorithm and machine learning, which comprises the following steps of: extracting fault features; determining an objective function; initializing a seagull population position by Gaussian mapping based on a seagull algorithm, namely initializing a penalty factor C and a kernel function parameter g of each seagull, and calculating an optimal fitness value and an optimal seagull position according to a target function; introducing a position updating mechanism of a black wife optimization algorithm, improving a position updating mode of an original gull algorithm, and updating the position; performing bidirectional sine variation on the optimal gull position; constructing a vehicle-mounted network fault diagnosis model by taking the optimal penalty factor C and the kernel function parameter g as SVM model parameters; and inputting the characteristic vector of the vehicle-mounted network fault diagnosis into the trained vehicle-mounted network fault diagnosis model to obtain a fault diagnosis result and test accuracy. The method overcomes the defects of the existing gull algorithm, and can remarkably improve the effect of vehicle-mounted network fault diagnosis.
Description
Technical Field
The invention relates to the technical field of vehicle-mounted network fault diagnosis, in particular to a vehicle-mounted network fault diagnosis method based on an intelligent optimization algorithm and machine learning.
Background
The in-vehicle network is a link connecting subsystems, devices, modules and components in a vehicle platform, and is a core supporting technology for realizing a vehicle integrated electronic system, which is considered as a central nerve of the vehicle platform. However, most of the existing vehicle-mounted network fault diagnosis still depends on manual experience for troubleshooting, and the fault reason is often difficult to diagnose effectively.
A Support Vector Machine (SVM) is a machine learning technique, and is widely applied to the field of network fault diagnosis and the like. For example, dongfeng et al proposed a support vector machine-based MAC protocol recognition method based on physical layer signals (dongfeng, quemoku, liaisajing, auobe, queghai. Support vector machine-based MAC protocol recognition method [ P ]. Jiangsu province: CN107231427B, 2020-04-07.).
In the SVM training process, the penalty factor C of the SVM and the selection quality of the RBF kernel function parameter g directly influence the final fault diagnosis result. The gull algorithm is a novel intelligent optimization algorithm for simulating gull foraging behavior, can effectively optimize SVM model parameters, and can also be applied to the problem of vehicle-mounted network fault diagnosis. However, there still exist some defects in the gull optimization algorithm, so that the algorithm is easy to fall into local optimum and has low convergence accuracy, and an ideal fault diagnosis effect is often not achieved when network fault diagnosis is performed.
Disclosure of Invention
In order to solve the problems, the invention provides the vehicle-mounted network fault diagnosis method based on the intelligent optimization algorithm and the machine learning, overcomes the defects of the existing gull algorithm, and can remarkably improve the effect of vehicle-mounted network fault diagnosis.
In order to achieve the above purpose, the present invention provides the following technical solutions.
A vehicle-mounted network fault diagnosis method based on an intelligent optimization algorithm and machine learning comprises the following steps:
extracting fault characteristics in original data of the vehicle-mounted network fault to be diagnosed;
taking the classification accuracy of the 5-fold cross validation SVM as a target function, and determining a penalty factor C of an SVM model and the upper and lower limits of an RBF kernel function parameter g;
initializing a gull population position through Gaussian mapping based on a gull algorithm, namely initializing a penalty factor C and a kernel function parameter g of each gull, and calculating an optimal fitness value and an optimal gull position according to a target function;
introducing a position updating mechanism of a black-wife optimization algorithm, improving a position updating mode of an original gull algorithm, updating the position, and obtaining an updated optimal fitness value and an optimal gull position;
performing bidirectional sine variation on the optimal gull position, and taking the gull position with the optimal fitness value before and after variation as the updated optimal gull position;
sequentially updating the optimal gull position according to the preset maximum iteration times, and determining the optimal gull position, namely obtaining a corresponding optimal penalty factor C and a kernel function parameter g;
constructing a vehicle-mounted network fault diagnosis model by taking the optimal penalty factor C and the kernel function parameter g as SVM model parameters;
and inputting the characteristic vector of the vehicle-mounted network fault diagnosis into the trained vehicle-mounted network fault diagnosis model to obtain a fault diagnosis result and test accuracy.
Preferably, the initializing the gull population position through gaussian mapping includes the following steps:
random numbers are generated by gaussian mapping:
using generated Gaussian random numbers x t Initializing seagull position P s (t) is:
P s (t)=(UB-LB)×x t +LB
in the formula, mod (·) is a complementation function, and LB is a gull optimizing lower boundary; UB optimizes the upper boundary for gull.
Preferably, the introducing the position updating mechanism of the black oligowomen optimization algorithm improves the position updating mode of the original gull algorithm, and performs position updating, and includes the following steps: updating the seagull position through the seagull migration behavior and the seagull global attack behavior.
Preferably, the gull migration behavior comprises the steps of:
the new position of the gull is calculated using the additional variable a to avoid collision with other gulls:
C s (t)=A×P s (t)
A=f c -(t×f c /Miter)
in the formula: c s (t) is a new position that does not conflict with other seagulls; p s (t) is the current position of the seagull; t is the current iteration number; miter is the maximum iteration number; a is the movement behavior of the seagull in a given search space; f. of c For controlling the coefficient, the value is reduced from 2 to 0;
moving towards the direction of the optimal position:
M s (t)=B×(P bs (t)-P s (t))
B=2×A 2 ×r d
in the formula: m s (t) is the direction in which the optimal position is located; p bs (t) is the optimal position; b is a random number responsible for balancing global and local search; r is a radical of hydrogen d Is [0,1 ]]A random number within a range;
arrival at the new location:
D s (t)=|C s (t)+M s (t)|
in the formula: d s (t) is the distance the gull has moved to the new position.
Preferably, the gull global attack behavior includes the following steps:
by changing the attack angle and speed continuously through the spiral motion, the spiral motion behavior is expressed as:
in the formula: r is the radius of each helix, θ is a random angle value in the range of [0,2 π ]; u and v are the correlation constants of the helical shape; e is the base number of the natural logarithm;
introducing a position updating mechanism of a black oligogyne optimization algorithm, wherein an improved gull position updating formula is as follows:
in the formula: p s (t + 1) is the gull position of the t +1 th iteration after updating; p bs (t) represents the optimal position for the t-th iteration; alpha is [0,1 ]]A random number in between; m is [0.4,0.9 ]]A random number in between; beta is [ -1,1]A random number within; r is 1 Is [1, N ]]A random integer between the number of the first and second integers,is the r th randomly selected 1 The position of the gull; p s (t) is the current gull position;
calculating the updated seagull pheromone value according to the following formula:
in the formula: pholomone (P) s (t + 1)) is the pheromone value of the current gull;fitness max And fitness min For worst and best fitness function values, fitness (P) s (t + 1)) is the fitness value of the current gull;
when the pheromone value of a seagull is equal to or less than 0.3, the seagull cannot be selected, and the formula of the update position of the seagull is as follows:
in the formula: gamma is [0,1 ]]A random number in between; r is a radical of hydrogen 1 And r 2 Is [1, N ]]A random integer between the number of the first and second integers,andrespectively, randomly selected r 1 And r 2 Position of gull, r 1 ≠r 2 (ii) a σ is a random binary number {0,1};
calculating a fitness value:
fitness(t)=F f (P s (t+1))
in the formula, F f (. Cndot.) is a fitness function in calculating the fitness value.
Preferably, the performing bidirectional sine variation on the optimal gull position includes the following steps:
calculating a sine chaotic value according to the current iteration times, and switching positive and negative directions at equal probability:
Sin Value=sin(πx 0 )
wherein rand is a random number from 0 to 1; x is the number of 0 Is an iterative sequence value;
carrying out variation disturbance on the optimal position:
P bs(j) (t+1)'=P bs(j) (t+1)+SinValue×P bs(j) (t+1)
in the formula: p is bs(j) (t + 1) represents the optimal position P for the t-th iteration bs The j-th dimension of (t + 1);
greedy update:
after mutation in each dimension, mutation was stopped.
The invention provides a vehicle-mounted network fault diagnosis method based on an intelligent optimization algorithm and machine learning, which has the following beneficial effects:
(1) By introducing Gaussian mapping to initialize the gull population position, the uniformity and diversity of population position distribution can be improved, and the stability of the algorithm is enhanced.
(2) The method improves the seagull position updating mode, introduces a position updating mechanism of a black oligowomen optimization algorithm to improve the seagull position updating mode, comprehensively considers factors such as different position updating modes, the position updating mode when the fitness value is too low and changed, the optimal position of the seagull in the iteration, other seagull positions in the population and the like to update the seagull position, realizes the increase of the algorithm search range, and enhances the adaptability of the algorithm.
(3) The capability of jumping out the local optimal solution in the later stage of the algorithm is realized by utilizing the bidirectional sine chaotic mapping variation on the optimal seagull.
Drawings
FIG. 1 is an embodiment of the present invention: a flow chart of a vehicle-mounted network fault diagnosis method based on an intelligent optimization algorithm and machine learning;
FIG. 2 is a vehicle network experimental platform of an embodiment of the present invention;
fig. 3 is a diagram of the results of the fault diagnosis in the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Example 1
The invention relates to a vehicle-mounted network fault diagnosis method based on an intelligent optimization algorithm and machine learning, which aims at several problems existing in SOA (service oriented architecture), provides an enhanced gull optimization algorithm (ISOA) optimized SVM and is used for fault diagnosis of a vehicle-mounted network, namely: the flow of the vehicle-mounted network fault diagnosis method based on the intelligent optimization algorithm and the machine learning is shown in fig. 1, and the specific steps are as follows:
s1: the method comprises the steps of extracting fault characteristics of original data of vehicle-mounted network faults to obtain characteristic vectors of vehicle-mounted network fault diagnosis, constructing a data set of the vehicle-mounted network fault diagnosis, and dividing the data set into a training data set and a testing data set.
S2: and establishing an objective function funtion of vehicle-mounted network fault diagnosis based on the enhanced gull algorithm optimization SVM. Because the classification accuracy of the SVM is calculated by using the training data set, the classification accuracy of the SVM cross validation by 5 folds of the training data set can be used as a target function, and the penalty factor C of the SVM model and the upper and lower limits of the RBF kernel function parameter g are determined.
S3: and setting parameters. The size of the population of seagulls (i.e., the number of individual seagulls) N; the maximum number of iterations (i.e., the condition under which the iteration stops) Miter; optimizing a lower boundary LB of the gull; seagull optimizing upper boundary UB
S4: based on a gull algorithm, initializing gull population positions through Gaussian mapping, namely initializing a penalty factor C and a kernel function parameter g of each gull, and calculating an optimal fitness value and an optimal gull position according to a target function:
mapping random number x by gaussians t Generating:
using generated Gaussian random numbers x t Initializing seagull position P s (t) is:
P s (t)=(UB-LB)×x t +LB
in the formula, mod (·) is a complementation function, and LB is a gull optimizing lower boundary; UB optimizes the upper boundary for gull.
S5: seagull migration behavior:
during migration, the algorithm simulates how gull clusters move from one location to another. Three behaviors are mainly involved in this phase: avoid collision, move to the optimal position direction and approach the optimal position.
The new position of the gull is calculated using the additional variable a to avoid collision with other gulls:
C s (t)=A×P s (t)
A=f c -(t×f c /Miter)
in the formula: c s (t) is a new position that does not conflict with other seagulls; p is s (t) is the current position of the seagull; t is the current iteration number; miter is the maximum iteration number; a is the movement behavior of the seagull in a given search space; f. of c For controlling the coefficient, the value is reduced from 2 to 0;
moving towards the direction of the optimal position:
M s (t)=B×(P bs (t)-P s (t))
B=2×A 2 ×r d
in the formula: m s (t) is the direction in which the optimal position is located; p is bs (t) is the optimal position; b is a random number responsible for balancing global and local search; r is a radical of hydrogen d Is [0,1 ]]A random number within a range;
arrival at the new location:
D s (t)=|C s (t)+M s (t)|
in the formula: d s (t) is the distance the gull has moved to the new position.
S6: gull attack behavior:
by changing the attack angle and speed continuously through the spiral motion, the spiral motion behavior is expressed as:
in the formula: r is the radius of each helix, θ is a random angle value in the range of [0,2 π ]; u and v are the correlation constants of the helical shape; e is the base of the natural logarithm.
In the original seagull algorithm, the optimal seagull position is only used for guiding to update the seagull position, in order to effectively improve the global search capability of the seagull, a position update mechanism of a black-wife optimization algorithm is introduced to improve a seagull position update mode, the seagull position is updated by comprehensively considering factors such as different position update modes, a time-varying update mode with too low fitness value, the optimal seagull position of the iteration, other seagull positions in the population and the like, the local optimal seagull position in each iteration is avoided, and the global search capability of the seagull algorithm is further improved.
Introducing a position updating mechanism of a black oligogyne optimization algorithm, wherein an improved gull position updating formula is as follows:
in the formula: p s (t + 1) is the gull position of the t +1 th iteration after updating; p bs (t) represents the optimal position for the t-th iteration; alpha is [0,1 ]]A random number in between; m is [0.4,0.9 ]]A random number in between; beta is [ -1,1 [ ]]A random number within; r is 1 Is [1, N ]]A random integer between the number of the first and second integers,is the r th randomly selected 1 The position of the gull; p is s (t) is the current gull position;
calculating the updated pheromone value of the seagull according to the following formula:
in the formula: pholomone (P) s (t + 1)) is that of the current gullAn pheromone value; fitness max And fitness min For the worst and optimal fitness function value, fitness (P) s (t + 1)) is the fitness value of the current seagull;
when the pheromone value of the gull is equal to or less than 0.3, the gull is not selected, and the formula of the update position is as follows:
in the formula: gamma is [0,1 ]]A random number in between; r is a radical of hydrogen 1 And r 2 Is [1, N ]]A random integer between the number of the first and second integers,andrespectively, randomly selected r 1 And r 2 Position of gull, r 1 ≠r 2 (ii) a σ is a random binary number {0,1};
calculating a fitness value:
fitness(t)=F f (P s (t+1))
in the formula, F f (. Cndot.) is a fitness function in calculating a fitness value.
S7: the optimal gull in the current iteration is recorded.
S8: and performing bidirectional sine variation on the optimal seagull position, and taking the seagull position with the optimal fitness value before and after variation as the updated optimal seagull position.
Calculating a sine chaotic value according to the current iteration times, and switching positive and negative directions at equal probability:
Sin Value=sin(πx 0 )
wherein rand is a random number from 0 to 1; x is a radical of a fluorine atom 0 Is an iterative sequence value;
carrying out variation disturbance on the optimal position:
P bs(j) (t+1)'=P bs(j) (t+1)+Sin Value×P bs(j) (t+1)
in the formula: p bs(j) (t + 1) represents the optimal position P for the t-th iteration bs The j-th dimension of (t + 1);
greedy update:
after mutation is performed for each dimension, the mutation is stopped.
S9: the optimal gull in the current iteration is recorded.
S10: repeatedly executing S6-S9, sequentially updating the optimal gull position according to the preset maximum iteration times, and determining the optimal gull position to obtain the optimal parameter C of the SVM best And g best 。
S11: constructing a vehicle-mounted network fault diagnosis model by taking the optimal penalty factor C and the kernel function parameter g as SVM model parameters; and inputting the characteristic vector of the vehicle-mounted network fault diagnosis into the trained vehicle-mounted network fault diagnosis model to obtain a fault diagnosis result and test accuracy.
In this embodiment:
as shown in fig. 2, the vehicle-mounted network experimental platform adopts a dual-bus topology structure, a terminator is connected to the end of a bus to prevent the emission phenomenon caused by impedance mismatching, and 1 vehicle-mounted network protocol analyzer, 1 vehicle-mounted network fault injection device, 6 network nodes, and the like are respectively attached to a network. In a verification experiment, the platform simulates 6 typical states of an on-board network, which are respectively: (1) the network is normal; (2) network disconnection; (3) a near end termination failure; (4) a distal termination failure; (5) double-sided termination failure; (6) intermittent disconnection.
2000 groups of samples of each state of the vehicle-mounted network are collected, 1600 groups of samples are taken as training samples at random, and the rest 400 groups of samples are taken as testing samples. And fault diagnosis is carried out on the vehicle-mounted network by adopting the SOA-SVM and the ISOA-SVM respectively. The parameters in the SOA algorithm are: n =50, maximum =200, the search ranges for c and g are both between 0-100, i.e. LB =0, ub =100; the parameters in the ISOA algorithm are: n =50, maximer =200, the search ranges for c and g are each between 0-100, i.e. LB =0, ub =100.
As shown in fig. 3, compared with the SOA-SVM, the ISOA-SVM has higher independent diagnosis accuracy for 6 states of the vehicle-mounted network; for the average diagnostic accuracy, the ISOA-SVM is improved by 6.57% compared with the SOA-SVM. That is, SVM parameters obtained by the ISOA search are better than SVM parameters obtained by the SOA search. Simulation results show that the ISOA algorithm has stronger searching capability than the SOA algorithm, the ISOA-SVM has higher diagnosis accuracy than the SOA-SVM, and the effectiveness of the method is verified.
The present invention is not limited to the above preferred embodiments, and any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. The vehicle-mounted network fault diagnosis method based on the intelligent optimization algorithm and the machine learning is characterized by comprising the following steps of:
extracting fault characteristics in original data of the vehicle-mounted network fault to be diagnosed;
taking the classification accuracy of the 5-fold cross validation SVM as a target function, and determining a penalty factor C of the SVM model and the upper and lower limits of an RBF kernel function parameter g;
initializing a gull population position through Gaussian mapping based on a gull algorithm, namely initializing a penalty factor C and a kernel function parameter g of each gull, and calculating an optimal fitness value and an optimal gull position according to a target function;
introducing a position updating mechanism of a black oligowomen optimization algorithm, improving a position updating mode of an original gull algorithm, updating the position, and obtaining an updated optimal fitness value and an optimal gull position;
performing bidirectional sine variation on the optimal gull position, and taking the gull position with the optimal fitness value before and after variation as the updated optimal gull position;
sequentially updating the optimal gull position according to the preset maximum iteration times, and determining the optimal gull position, namely obtaining a corresponding optimal penalty factor C and a kernel function parameter g;
constructing a vehicle-mounted network fault diagnosis model by taking the optimal penalty factor C and the kernel function parameter g as SVM model parameters;
and inputting the characteristic vector of the vehicle-mounted network fault diagnosis into the trained vehicle-mounted network fault diagnosis model to obtain a fault diagnosis result.
2. The vehicle network fault diagnosis method based on the intelligent optimization algorithm and the machine learning of claim 1, wherein the initializing gull population positions through Gaussian mapping comprises the following steps:
random numbers are generated by gaussian mapping:
using generated Gaussian random numbers x t Initializing seagull position P s (t) is:
P s (t)=(UB-LB)×x t +LB
in the formula, mod (·) is a complementation function, and LB is a gull optimizing lower boundary; UB optimizes the upper boundary for gull.
3. The vehicle network fault diagnosis method based on intelligent optimization algorithm and machine learning of claim 1, wherein the position update mechanism of the introduced black oligowomen optimization algorithm improves the position update mode of the raw gull algorithm and performs position update, and comprises the following steps: updating the gull position through the gull migration behavior and the gull global attack behavior.
4. The vehicle network fault diagnosis method based on intelligent optimization algorithm and machine learning of claim 3, wherein the gull migration behavior comprises the following steps:
the new position of the gull is calculated using the additional variable a to avoid collision with other gulls:
C s (t)=A×P s (t)
A=f c -(t×f c /Miter)
in the formula: c s (t) is a new position that does not conflict with other seagulls' positions; p s (t) is the current position of the seagull; t is the current iteration number; miter is the maximum iteration number; a is the movement behavior of the seagull in a given search space; f. of c For controlling the coefficient, the value is reduced from 2 to 0;
moving towards the direction of the optimal position:
M s (t)=B×(P bs (t)-P s (t))
B=2×A 2 ×r d
in the formula: m is a group of s (t) is the direction in which the optimal position is located; p bs (t) is the optimal position; b is a random number responsible for balancing global and local search; r is d Is [0,1 ]]A random number within a range;
arrival at the new location:
D s (t)=|C s (t)+M s (t)|
in the formula: d s (t) is the distance the gull has moved to the new position.
5. The vehicle-mounted network fault diagnosis method based on the intelligent optimization algorithm and the machine learning, as claimed in claim 3, wherein the gull global attack behavior comprises the following steps:
by changing the attack angle and speed continuously through the spiral motion, the spiral motion behavior is expressed as:
in the formula: r is the radius of each helix, θ is a random angle value in the range of [0,2 π ]; u and v are the correlation constants of the helical shape; e is the base number of the natural logarithm;
introducing a position updating mechanism of a black oligogyne optimization algorithm, wherein an improved gull position updating formula is as follows:
in the formula: p is s (t + 1) is the gull position of the t +1 th iteration after updating; p bs (t) represents the optimal position for the t-th iteration; alpha is [0,1 ]]A random number in between; m is [0.4,0.9 ]]A random number in between; beta is [ -1,1 [ ]]A random number within; r is 1 Is [1, N ]]A random integer in between, and a random integer,is the r th randomly selected 1 The position of the gull; p s (t) is the current gull position;
calculating the updated pheromone value of the seagull according to the following formula:
in the formula: pholomone (P) s (t + 1)) is the pheromone value of the current gull; fitness max And fitness min For worst and best fitness function values, fitness (P) s (t + 1)) is the fitness value of the current gull;
when the pheromone value of the gull is equal to or less than 0.3, the gull is not selected, and the formula of the update position is as follows:
in the formula: gamma is [0,1 ]]A random number in between; r is a radical of hydrogen 1 And r 2 Is [1, N ]]A random integer in between, and a random integer,andrespectively of randomly selected r-th 1 And r 2 Position of gull, r 1 ≠r 2 (ii) a σ is a random binary number {0,1};
calculating a fitness value:
fitness(t)=F f (P s (t+1))
in the formula, F f (. Cndot.) is a fitness function in calculating a fitness value.
6. The vehicle-mounted network fault diagnosis method based on the intelligent optimization algorithm and the machine learning of claim 1, wherein the bidirectional sine variation of the optimal gull position comprises the following steps:
calculating a sine chaotic value according to the current iteration times, and switching positive and negative directions at equal probability:
SinValue=sin(πx 0 )
wherein rand is a random number from 0 to 1; x is a radical of a fluorine atom 0 Is an iterative sequence value;
carrying out variation disturbance on the optimal position:
P bs(j) (t+1)'=P bs(j) (t+1)+SinValue×P bs(j) (t+1)
in the formula: p bs(j) (t + 1) represents the optimal position P for the t-th iteration bs The j-th dimension of (t + 1);
greedy update:
after mutation is performed for each dimension, the mutation is stopped.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211088548.7A CN115660024B (en) | 2022-09-07 | 2022-09-07 | Vehicle-mounted network fault diagnosis method based on intelligent optimization algorithm and machine learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211088548.7A CN115660024B (en) | 2022-09-07 | 2022-09-07 | Vehicle-mounted network fault diagnosis method based on intelligent optimization algorithm and machine learning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115660024A true CN115660024A (en) | 2023-01-31 |
CN115660024B CN115660024B (en) | 2024-01-19 |
Family
ID=84984230
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211088548.7A Active CN115660024B (en) | 2022-09-07 | 2022-09-07 | Vehicle-mounted network fault diagnosis method based on intelligent optimization algorithm and machine learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115660024B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116386215A (en) * | 2023-03-16 | 2023-07-04 | 淮阴工学院 | Intelligent charging method for mobile electric box based on people flow |
CN117521533A (en) * | 2024-01-08 | 2024-02-06 | 中国人民解放军陆军装甲兵学院 | Integrated matching design method for hub motor driving device |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113867368A (en) * | 2021-12-03 | 2021-12-31 | 中国人民解放军陆军装甲兵学院 | Robot path planning method based on improved gull algorithm |
CN113962161A (en) * | 2021-11-05 | 2022-01-21 | 重庆邮电大学 | Optimal QoS service selection method and device based on black and wife optimization algorithm |
CN114201913A (en) * | 2021-11-22 | 2022-03-18 | 西北工业大学 | Design method for flight parameters of tail end of sea-skimming aircraft |
CN114638170A (en) * | 2022-03-31 | 2022-06-17 | 河北工业大学 | Electric energy metering device fault diagnosis method and system based on data mining technology |
CN114662386A (en) * | 2022-03-14 | 2022-06-24 | 宁波大学 | Bearing fault diagnosis method and system |
CN114818809A (en) * | 2022-04-27 | 2022-07-29 | 合肥工业大学 | Fault diagnosis method for SCA-SVM motor rolling bearing based on cross wavelets |
CN114969995A (en) * | 2021-11-22 | 2022-08-30 | 昆明理工大学 | Rolling bearing early fault intelligent diagnosis method based on improved sparrow search and acoustic emission |
-
2022
- 2022-09-07 CN CN202211088548.7A patent/CN115660024B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113962161A (en) * | 2021-11-05 | 2022-01-21 | 重庆邮电大学 | Optimal QoS service selection method and device based on black and wife optimization algorithm |
CN114201913A (en) * | 2021-11-22 | 2022-03-18 | 西北工业大学 | Design method for flight parameters of tail end of sea-skimming aircraft |
CN114969995A (en) * | 2021-11-22 | 2022-08-30 | 昆明理工大学 | Rolling bearing early fault intelligent diagnosis method based on improved sparrow search and acoustic emission |
CN113867368A (en) * | 2021-12-03 | 2021-12-31 | 中国人民解放军陆军装甲兵学院 | Robot path planning method based on improved gull algorithm |
CN114662386A (en) * | 2022-03-14 | 2022-06-24 | 宁波大学 | Bearing fault diagnosis method and system |
CN114638170A (en) * | 2022-03-31 | 2022-06-17 | 河北工业大学 | Electric energy metering device fault diagnosis method and system based on data mining technology |
CN114818809A (en) * | 2022-04-27 | 2022-07-29 | 合肥工业大学 | Fault diagnosis method for SCA-SVM motor rolling bearing based on cross wavelets |
Non-Patent Citations (2)
Title |
---|
JOEL M. ET AL.: "Development of the Black Widow Micro Air Vehicle", 《AMERICAN INSTITUTE OF AERONAUTICS AND ASTRONAUTICS》 * |
孙培胜 等: "基于SOA-SVM 模型的光伏阵列故障诊断研究", 《电源学报》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116386215A (en) * | 2023-03-16 | 2023-07-04 | 淮阴工学院 | Intelligent charging method for mobile electric box based on people flow |
CN116386215B (en) * | 2023-03-16 | 2024-04-19 | 淮阴工学院 | Intelligent charging method for mobile electric box based on people flow |
CN117521533A (en) * | 2024-01-08 | 2024-02-06 | 中国人民解放军陆军装甲兵学院 | Integrated matching design method for hub motor driving device |
Also Published As
Publication number | Publication date |
---|---|
CN115660024B (en) | 2024-01-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN115660024A (en) | Vehicle-mounted network fault diagnosis method based on intelligent optimization algorithm and machine learning | |
CN115756925B (en) | Vehicle-mounted network fault diagnosis method based on intelligent optimization algorithm | |
US11544917B2 (en) | Power electronic circuit fault diagnosis method based on optimizing deep belief network | |
CN115700593B (en) | Vehicle-mounted network fault diagnosis method based on machine learning | |
CN115204316A (en) | Vehicle-mounted network fault diagnosis method based on support vector machine | |
CN109067773B (en) | Vehicle-mounted CAN network intrusion detection method and system based on neural network | |
CN108828436B (en) | Analog circuit fault diagnosis method based on chaotic cloud self-adaptive firefly algorithm | |
US20170103308A1 (en) | Acceleration of convolutional neural network training using stochastic perforation | |
CN111310915A (en) | Data anomaly detection and defense method for reinforcement learning | |
US20210057920A1 (en) | Systems and methods of applying artificial intelligence to battery technology | |
CN105388402B (en) | Crosslinked cable Partial Discharge Pattern Recognition Method based on parameter optimization SVM algorithm | |
CN107169557A (en) | A kind of method being improved to cuckoo optimized algorithm | |
CN115081316A (en) | DC/DC converter fault diagnosis method and system based on improved sparrow search algorithm | |
CN112465160A (en) | VR-based vehicle maintenance auxiliary system | |
CN114170789A (en) | Intelligent network connected vehicle lane change decision modeling method based on space-time diagram neural network | |
CN112801281A (en) | Countermeasure generation network construction method based on quantization generation model and neural network | |
CN113222035B (en) | Multi-class imbalance fault classification method based on reinforcement learning and knowledge distillation | |
Omar et al. | Fault classification on transmission line using LSTM network | |
CN111932396A (en) | Automatic identification method for power distribution network topology network | |
CN116087667A (en) | LSTM-based relay protection fault detection method for extra-high voltage direct current transmission line | |
CN115545120A (en) | Vehicle-mounted network fault diagnosis method and device, storage medium and terminal equipment | |
CN111814414A (en) | Coverage rate convergence method and system based on genetic algorithm | |
CN111412795A (en) | Test point setting scheme generation method and device | |
CN116031879A (en) | Hybrid intelligent feature selection method suitable for transient voltage stability evaluation of power system | |
CN113466630B (en) | Power distribution network fault cause classification method based on RSSPN model |
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 |