CN104536435B - A kind of line control system network inline diagnosis method - Google Patents
A kind of line control system network inline diagnosis method Download PDFInfo
- Publication number
- CN104536435B CN104536435B CN201410795268.9A CN201410795268A CN104536435B CN 104536435 B CN104536435 B CN 104536435B CN 201410795268 A CN201410795268 A CN 201410795268A CN 104536435 B CN104536435 B CN 104536435B
- Authority
- CN
- China
- Prior art keywords
- network
- fuzzy
- control system
- health degree
- line control
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0208—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the configuration of the monitoring system
- G05B23/0213—Modular or universal configuration of the monitoring system, e.g. monitoring system having modules that may be combined to build monitoring program; monitoring system that can be applied to legacy systems; adaptable monitoring system; using different communication protocols
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Small-Scale Networks (AREA)
- Maintenance And Management Of Digital Transmission (AREA)
Abstract
The present invention relates to a kind of line control system network inline diagnosis method, the method builds a model for including two Adaptive Neuro-fuzzy Inferences first, the line control system network state parameters obtained under using failure environment are input into as sample data, off-line training is carried out to two Adaptive Neuro-fuzzy Inferences using hybrid learning algorithm, fuzzy control rule is extracted respectively, then fuzzy control rule is changed into inquiry table, generation network health degree calculates function and network failure reason decision function, in the controller node of addition line control system, form the line control system inline diagnosis algorithm based on Adaptive Neuro-fuzzy Inference, inline diagnosis is carried out to line control system network health degree and failure cause.
Description
Technical field
The present invention relates to a kind of line control system network inline diagnosis method based on Adaptive Neuro-fuzzy Inference.
Background technology
With continuing to develop for modern electronic technology, the direction of automotive electronics forward direction intelligentized control method is developed, and using total
Various Vehicle Electronic Control modules are connected to become In-vehicle networking by line, it may not be necessary to traditional mechanical mechanism transmission control letter
Number, but driven a car using electronically, this electronically is exactly wire control technology.The continuous application of wire control technology is vapour
The communication of in-car portion offer convenience with it is intelligentized simultaneously, the also development to automotive electronics brings new challenge.Due to directly closing
Be tied to the security of automobile, the reliability of its each part for being included require it is high, especially including playing crucial biography
The bus communication technology of defeated effect, bus communication technology is more total using the FlexRay based on time triggered in wire control technology
Line.
The inline diagnosis of network is exactly that the ruuning situation of bus is diagnosed in real time from the angle of bus communication, for
For line control system network, a rational inline diagnosis strategy is for the holding normal communication process of line control system, and is
The reliability and security of system are all necessary.In-circuit diagnostic system (OBD) is integrated with vehicle at present, to inspection
The mistake for measuring is stored in the form of DTC (DTC), but only obviously network failure can just be detected
Come, such as bus short circuit is breaking, communicate situations such as stopping.This kind of situation may occur, even if wrong presence in bus,
But as long as these mistakes do not influence normal communication, number of errors is exactly traditional inline diagnosis system in tolerance interval
System cannot just find these implicit failures, and these problems are progressively accumulated, it is more likely that cause the system cannot proper communication.
For the insurmountable problem of traditional in-circuit diagnostic system, domestic and foreign scholars are from data fusion, network management and examine
The multiple angles such as disconnected agreement, diagnosis algorithm launch research, and such as document " with diagnosis assist by the network management of CAN_FlexRay bus systems
The research of view " (ginseng ginseng, Master's thesis, HeFei University of Technology) devises CAN-FlexRay gateway network management strategies, can be with
Node time-out, message time-out etc. are diagnosed, but the Diagnostic Strategy only rests on monitoring aspect, and fault data is not done
Further treatment, it is impossible to enough analyzing failure causes.Document " Diagnostic Fusion for Time-Triggered
Automotive Networks " (Utsav Drolia etc., Carnegie Mellon University, 2012IEEE 14th
International Symposium on High-Assurance Systems Engineering) in devise based on god
Through the FlexRay bus system diagnostic methods of network, classification diagnosis can be carried out to the error situation of FlexRay buses, but
Neural network algorithm designed by the method has only obtained network failure reason, it is impossible to provide one to current network running status
Qualitatively judge, such as current network has a potential danger, or current network is although wrong, but can still run, so as to be
Fault processing module provides foundation.Can be according to historical data and the currently monitored data for arriving, while simultaneously there is presently no seeing
The efficient inline diagnosis method for providing network operation state and network failure reason.
The content of the invention
The purpose of the present invention is directed to using the FlexRay line control systems of time triggered bus, proposes a kind of based on adaptive
The inline diagnosis method of neuro fuzzy systems, the real time execution situation of supervisory network are answered, and failure cause is judged.The party
Method can timely find potential problem in line control system network, line control system network operation state be judged, for improving line
The reliability for controlling the grid communication technology plays a positive role with real-time.
The present invention is carried out entirely in terms of network operation state judges and failure cause judges two to FlexRay network systems
Face real-time diagnosis.Introduce " network health degree " this concept to represent the running status of network, but " health " or " be not good for
Health " is a fuzzy concept, and it is also a fuzzy concept that failure cause judges in addition, because the display form of failure has
Similarity, no matter as " internal error " or " external disturbance " erroneous frame is likely to increase, determining " internal error " also
It is that " external disturbance " is also a fuzzy concept.It is just necessary if to finally give fuzzy concept from specific data
By fuzzy inference system.Fuzzy inference system is independent of accurate mathematical analysis, Heuristics and reasoning from human expert
Propose that If-Then rules set up fuzzy model, design Fuzzy control system based on this in journey.However, relying solely on people
Class knowledge, it is difficult to select the shape and number of appropriate membership function.If using Adaptive Neuro-fuzzy Inference,
Historical data is carried out offline training, then the fuzzy control rule that training is obtained is embedded into fuzzy control model, just
Training error can be to greatest extent reduced, the accuracy and reliability of system is improved.
Realizing the technical scheme of above-mentioned purpose is：
Line control system network inline diagnosis method based on Adaptive Neuro-fuzzy Inference, comprises the following steps：
Step one：The line control system network of FlexRay bus communications is built, the network operation shape under failure environment is extracted in
State parameter；
Step 2：Two Adaptive Neuro-fuzzy Inferences are built in Matlab/Simulink, with network operation shape
The sample data that state parameter is obtained is carried out using hybrid learning algorithm as input to two Adaptive Neuro-fuzzy Inferences
Off-line training, extracts the fuzzy control rule of network health degree and network failure reason respectively；
Step 3：Fuzzy control rule is converted into inquiry table, network inline diagnosis function, including network health degree is generated
Function and network failure reason decision function are calculated, is embedded into line control system Network Controller nodes, finally give line traffic control system
System network inline diagnosis result.
Described step one is further comprised the steps of：
Step A：Two communication modes of node of selection, its interior joint one implantation μ C/OS-II operating systems, in FlexRay
The treatment of multitask is carried out on the basis of bottom communication, node two carries out the statistics of state parameter using query method；
Step B：Random signal is added to disturb by FlexRay buses, electromagnetic interference manufactures the differences such as physical fault
Failure environment, extract FlexRay communication process in there is the state and network state parameters of each time slot.
Described step two is further comprised the steps of：
Step A：In the network state parameters for extracting, three indexs of inline diagnosis of selection, the cycle of signal in network,
The number of packet loss and the situation of mistake are the index of inline diagnosis.To these three selecting index multi-group datas, and to selected number
According to being standardized, its characteristic value is set to be mapped on [0,1] interval；
Step B：The input quantity of the fuzzy neural network of selected two Adaptive Neuro-fuzzy Inferences (ANFIS) construction
It is cycle, packet loss number and error situation, output quantity is respectively network health degree and network failure reason.Two adaptive neural networks
Fuzzy inference system input quantity dimension is 3, and the fuzzy partition number of input quantity is 5, and the membership function of each input component is Gauss
Function；
Step C：In two Adaptive Neuro-fuzzy Inferences, membership function is multiplied, calculates each rule
Relevance grade, relevance grade calculates the output of each fuzzy rule after normalizing computing, and obtains the output of system, respectively
Network health degree and network failure reason；
Step D：Using hybrid learning algorithm, carried out offline as input using the sample data that the state parameter of network is obtained
Training, determines consequent parameter, and two fuzzy control rules of Adaptive Neuro-fuzzy Inference are extracted respectively；
Step E：The network health degree of generation and network failure reason fuzzy control rule are directly embedded into Matlab/
In Fuzzy Logic controller control modules in Simulink, two fuzzy inference systems are obtained.
Described step three is further comprised the steps of：
Step A：Two fuzzy inference systems are regarded as a black box respectively, input-output mappings transformation is inquiry
Table, the automatic Code Generation in recycling Matlab/Simulink for Freescale series monolithics gives birth to inquiry table
Function and network failure reason decision function are calculated into network health degree.This mode is than directly carrying out certainly fuzzy inference system
Dynamic code building is in hgher efficiency, and more save resources；
Step B：The inline diagnosis function that will be generated includes that network health degree calculates function and network failure reason judges letter
Number, is directly embedded into line control system Network Controller nodes, by function call, it is possible to when abnormal conditions occurs in network
Obtain the result of line control system inline diagnosis.
Beneficial effects of the present invention：
Line control system network inline diagnosis algorithm of the present invention based on Adaptive Neuro-fuzzy Inference, by gathering net
Parameter during network malfunction carries out off-line training as sample using hybrid learning algorithm, obtains network inline diagnosis algorithm
Inline diagnosis is carried out in embedded network controller node again afterwards, can effectively according to the state parameter during the network operation
The health degree and failure cause of network are obtained, in real time the running status at monitoring networking, so that before ceased communication timely
It was found that potential problem in network, the effective reliability for improving line control system network service.
Brief description of the drawings
Fig. 1 is the main flow chart of line control system network inline diagnosis method of the present invention；
Fig. 2 is line control system network failure data sample citing of the present invention；
Fig. 3 is the flow chart of Adaptive Neuro-fuzzy Inference of the present invention；
Fig. 4 is Adaptive Neuro-fuzzy Inference structure chart of the present invention for network health degree；
Fig. 5 is Adaptive Neuro-fuzzy Inference off-line training error change figure of the present invention；
Fig. 6 is line control system network health degree variation diagram of the present invention；
Fig. 7 is line control system complete period network inline diagnosis figure of the present invention.
Specific embodiment
The present invention will be further described with reference to the accompanying drawings and detailed description.
Fig. 1 is line control system network inline diagnosis method of the present invention based on Adaptive Neuro-fuzzy Inference, first
Sample is used as by parameter when gathering FlexRay network failure states, two Adaptive Neural-fuzzy Inference moulds are built offline
Type, and generate corresponding network health degree and network failure reason fuzzy control rule.Further, fuzzy control rule is turned
Inquiry table is changed to, network inline diagnosis function is generated, including network health degree calculates function and network failure reason decision function,
Network inline diagnosis function is embedded into network controller, inline diagnosis just can be carried out, according to the shape during the network operation
The health degree and network failure reason of state gain of parameter network, the running status of monitor in real time network.
By taking wire-controlled steering system as an example, built comprising two line control system networks of node based on FlexRay buses, wherein
Node one is the steering wheel controller node of wire-controlled steering system, and node two is the steering actuating mechanism controls of wire-controlled steering system
Device node, is communicated between two nodes by FlexRay buses.In this embodiment, using Freescale
MC9S12XF512 series monolithics are used as FlexRay communication nodes.
Line control system network inline diagnosis method based on Adaptive Neuro-fuzzy Inference mainly includes three steps：
Step one：To in steering wheel controller node and steering actuator controller node in line traffic control grid
FlexRay protocol registers carry out reasonable disposition, extract network operation state by customized function under failure environment and join
Number, such as signal period, packet loss number, error situation, for Adaptive Neuro-fuzzy Inference off-line training provides sample.
Comprise the following steps that：
Step A：Two nodes of selection are communicated, the implantation μ C/OS-II operations of its steering wheel controller node of interior joint one
System, according to the file structure of μ C/OS-II, to OS_CPU.H, OS_CPU_A.ASM, OS_CPU_C.C file are modified,
On the basis of FlexRay bottom communications multi-task scheduling is realized using interrupt mode；Node two turns to actuator controller
Node reads the data of related status register using query method, and the state of each time slot, extracts letter in analysis communication process
Number cycle, the number of packet loss, the network state parameters such as the situation of mistake.
Step B：Random signal is added to disturb by FlexRay buses, electromagnetic interference manufactures the differences such as physical fault
Failure environment, slot status and network state parameters when extracting each in FlexRay communication process, is next as historical data
The off-line training of step provides sample.Fig. 2 illustrates for line control system network failure data sample, reflects message cycle with wrong feelings
Condition changes with time situation.
Step 2：As shown in figure 3, in the flow chart of Adaptive Neuro-fuzzy Inference, in Matlab/Simulink
Two Adaptive Neuro-fuzzy Inferences are built in environment, using hybrid learning algorithm, with network-like obtained by the first step
State parameter carries out off-line training as historical data input, and two Adaptive Neuro-fuzzy Inferences are individually trained,
Extract the fuzzy control rule of Adaptive Neuro-fuzzy Inference.
Comprise the following steps that：
Step A：In the network state parameters for extracting, the cycle of signal in selection network, number and the mistake of packet loss
Situation is three inline diagnosis indexs.These three achievement datas are standardized, its characteristic value is mapped to [0,1] area
Between on；
Step B：The input quantity of the fuzzy neural network of selected two Adaptive Neuro-fuzzy Inferences (ANFIS) construction
It is cycle, packet loss number and error situation, two Adaptive Neuro-fuzzy Inference input quantity dimensions are 3, the mould of input quantity
Paste segmentation number is 5.Output quantity is respectively network health degree and network failure reason, wherein the basic domain of network health degree for [0,
0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1], when network service is normal, network health degree is 1, works as network
When communication has stopped, network health degree is 0, during other network service abnormalities, network health degree correspondence in the range of 0-1
Value.The basic domain of network failure reason is [0,1,2,3], and when network service is normal, network failure reason is 0, works as network
Communication abnormality be because the internal error of controller software is caused, network failure reason be 1, when network service it is different normally due to
What external electromagnetic interference etc. was caused, network failure reason is 2, when network service is different normally due to what physical fault etc. was caused, net
Network failure cause is 3.It is Adaptive Neuro-fuzzy Inference structure chart of the present invention for network health degree shown in Fig. 4.
The membership function of each input component is Gaussian function：
Wherein, i=1,2,3；J=1,2,3,4,5；aijAnd cijWidth and center for membership function, xiIt is network
Given parameters.The Adaptive Neuro-fuzzy Inference includes 125 fuzzy control rules altogether, and its fuzzy control collection is as follows：
Rule 1：if(x1is A11)and(x2is A21)and(x3is A31)then f1=p10+p11x1+p12x2+p13x3
Rule 2：if(x1is A12)and(x2is A22)and(x3is A32)then f2=p20+p21x1+p22x2+p23x3
Rule 3：if(x1is A12)and(x2is A22)and(x3is A33)then f3=p30+p31x1+p32x2+p33x3
……
Rule 125：
if(x1is A15)and(x2is A25)and(x3is A35)then f125=p125,0+p125,1x1+p125,2x2+p125, 3x3
Wherein pkmIt is consequent parameter (k=1,2,3...125, m=0,1,2,3), fk(k=1,2,3...125) is correspondence
In every output of fuzzy rule.
Step C：In two Adaptive Neuro-fuzzy Inferences, membership function is multiplied, calculates each rule
Relevance grade wk, relevance grade through normalization computing after obtainFinal system is output as f：
wk=Ak1(x1)×Ak2(x2)×Ak3(x3)
Make x0=1, k=1,2,3...125, m=0,1,2,3.Given parameters aijAnd cij, study is needed in whole network
Parameter just only has consequent parameter pkm；
The output of two Adaptive Neuro-fuzzy Inferences is respectively network health degree and network failure reason.
Step D：Two Adaptive Neuro-fuzzy Inferences are individually trained using hybrid learning algorithm, to determine
Consequent parameter pkm, extract two fuzzy control rules of Adaptive Neuro-fuzzy Inference.Hybrid learning algorithm is comprising most
Small square law and back-propagation algorithm.The output of system can be expressed as first：
F=BX
Wherein, column vector X is conclusion vector, using least square method can obtain mean square error it is minimum (min | | BX-f |
|) the best estimate X of conclusion vector under meaning*：
X*=(BTB)-1BTf
Then, when there is N group input datas, if the error function of system is：
T=1 in formula, 2,3...N, ftIt is reality output, fdtIt is desired output.When conclusion vector is fixed, use
BP algorithm in feedforward neural network, input is reversely passed to by error by output end, with gradient descent method, updates premise parameter
cijAnd aij, so that change the shape of membership function, until training error value is less than threshold value set in advance.Training error such as Fig. 5
It is shown；
In formula, i=1,2,3, j=1,2,3,4,5, β>0 is learning efficiency.
Step E：The network health degree of generation and network failure reason fuzzy control rule are directly embedded into Matlab/
In Fuzzy Logic controller control modules in Simulink, two fuzzy inference systems are obtained.
Step 3：Two fuzzy inference systems are all trained after finishing, the fuzzy inference rule in system is converted to and is looked into
Table is ask, and inline diagnosis function is generated by automatic Code Generation, including network health degree calculates function and network failure
Reason decision function, the two functions can be directly embedded into line control system nodes controller, carry out line control system net
Network inline diagnosis.
Comprise the following steps that：
Step A：Two fuzzy inference systems are regarded as a black box respectively, input-output mappings transformation is inquiry
Table, the automatic Code Generation in recycling Matlab/Simulink for Freescale series monolithics gives birth to inquiry table
Function and network failure reason decision function are calculated into inline diagnosis function, including network health degree.This mode will be than directly will
It is in hgher efficiency that fuzzy inference system carries out real time workshop, and more save resources；
Step B：The inline diagnosis function that will be generated, including network health degree calculates function and network failure reason judges letter
Number, is directly embedded into line control system Network Controller nodes, by function call, it is possible to when abnormal conditions occurs in network
Obtain the result of line control system inline diagnosis.As shown in fig. 6, there is no additional any outside to do in line control system FlexRay networks
When disturbing influence, bus running status is normal, and the health degree of network is also maintained at 1, but when system adds external disturbance, network
Health degree also with declining.Understand, when the health degree of network is less than 1, network internal has just had some mistakes
Situation, the error number simply accumulated is insufficient to allow ceased communication.
Fig. 7 show one more fully based on FlexRay network complete period inline diagnosis figures, when there is message frame number
Reduce or increase, and wrong frame number it is more when, network inline diagnosis algorithm is by according to the cycle of current message, Network Packet Loss
Number, the wrong frame number of network carry out network health degree judgement, and when abnormal conditions occurs in network, network health degree will appear from
Decline, while it is internal error, external disturbance or physical fault that can determine current network abnormal cause.
The present invention is to carry out online diagnosis to the network of line control system, and bus communication is judged by the health degree of network
State, so as to timely find potential problem in network, analyze abnormal network cause, and carrying out correction makes network return to just
Normal state, effectively raises the reliability of line control system network service.
Claims (2)
1. a kind of line control system network inline diagnosis method, it is characterised in that described line control system network inline diagnosis method
Comprise the following steps：
Step one：The line control system network of FlexRay bus communications is built, random signal is added by FlexRay buses
Interference, electromagnetic interference manufactures physical fault, simulates different failure environments, is extracted in the FlexRay networks fortune under failure environment
Row network state parameters, including signal period, packet loss number and error situation；Step 2：Built in Matlab/Simulink
Two Adaptive Neuro-fuzzy Inferences, the FlexRay network operation state parameter sample datas obtained using step one as
Two Adaptive Neuro-fuzzy Inferences are carried out off-line training by input using hybrid learning algorithm, and network is extracted respectively
The fuzzy control rule of health degree and network failure reason, specifically includes following steps：
Step A：In the network state parameters for extracting, three indexs of inline diagnosis, the cycle of signal, packet loss in network are selected
Number and mistake situation for inline diagnosis index；To these three selecting index multi-group datas, and selected data is entered
Row standardization, makes its characteristic value be mapped on [0,1] interval；
Step B：The input quantity of the fuzzy neural network of selected two Adaptive Neuro-fuzzy Inferences (ANFIS) construction is week
Phase, packet loss number and error situation, output quantity are respectively network health degree and network failure reason；Two adaptive neural networks are obscured
Inference system input quantity dimension is 3, and the fuzzy partition number of input quantity is 5, and the membership function of each input component is Gaussian function；
Step C：In two Adaptive Neuro-fuzzy Inferences, membership function is multiplied, calculates being applicable for each rule
Degree, the relevance grade calculates the output of each fuzzy rule after normalizing computing, and obtains the output of system, respectively
Network health degree and network failure reason；
Step D：Using hybrid learning algorithm, instructed offline as input using the sample data that the state parameter of network is obtained
Practice, determine consequent parameter, two fuzzy control rules of Adaptive Neuro-fuzzy Inference are extracted respectively；
Step E：The network health degree of generation and network failure reason fuzzy control rule are directly embedded into Matlab/
In Fuzzy Logic controller control modules in Simulink, two fuzzy inference systems are obtained；
Step 3：Fuzzy control rule is converted into inquiry table, network inline diagnosis function is generated, including network health degree is calculated
Function and network failure reason decision function, are embedded into line control system Network Controller nodes, finally give line control system net
Network inline diagnosis result.
2. line control system network inline diagnosis method according to claim 1, it is characterised in that the step of described step two
Rapid B has following characteristics：
The input quantity of the fuzzy neural network of selected two Adaptive Neuro-fuzzy Inferences (ANFIS) construction is the cycle, is lost
Bag number and error situation, two Adaptive Neuro-fuzzy Inference input quantity dimensions are 3, and the fuzzy partition number of input quantity is
5；Output quantity is respectively network health degree and network failure reason, wherein the basic domain of network health degree for [0,0.1,0.2,
0.3,0.4,0.5,0.6,0.7,0.8,0.9,1], when network service is normal, network health degree is 1, when network service
During stopping, network health degree is 0, during other network service abnormalities, network health degree correspondence value in the range of 0-1；Net
The basic domain of network failure cause is [0,1,2,3], and when network service is normal, network failure reason is 0, when network service exception
It is that, because the internal error of controller software is caused, network failure reason is 1, when network service is different normally due to external electromagnetic
Interference is caused, and network failure reason is 2, and when network service is different normally due to what physical fault was caused, network failure reason is
3。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410795268.9A CN104536435B (en) | 2014-12-18 | 2014-12-18 | A kind of line control system network inline diagnosis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410795268.9A CN104536435B (en) | 2014-12-18 | 2014-12-18 | A kind of line control system network inline diagnosis method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104536435A CN104536435A (en) | 2015-04-22 |
| CN104536435B true CN104536435B (en) | 2017-06-30 |
Family
ID=52851977
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410795268.9A Active CN104536435B (en) | 2014-12-18 | 2014-12-18 | A kind of line control system network inline diagnosis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104536435B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11853863B2 (en) | 2019-08-12 | 2023-12-26 | Micron Technology, Inc. | Predictive maintenance of automotive tires |
Families Citing this family (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105721199B (en) * | 2016-01-18 | 2019-06-07 | 中国石油大学(华东) | A kind of real-time cloud service bottleneck detection method based on Density Estimator and fuzzy inference system |
| EP3258332B1 (en) * | 2016-06-13 | 2018-12-12 | ABB Schweiz AG | Method for assessing health conditions of industrial equipment |
| CN106597178B (en) * | 2017-01-18 | 2019-07-19 | 重庆水利电力职业技术学院 | A kind of pre- abnormal method of ANFIS number of test device for relay protection LPA |
| CN110149265B (en) * | 2018-03-14 | 2021-10-26 | 腾讯科技（深圳）有限公司 | Message display method and device and computer equipment |
| CN109088754A (en) * | 2018-07-31 | 2018-12-25 | 北京航空航天大学 | A kind of COMMUNICATION NETWORK PERFORMANCES failure reason and characteristic parameter association relationship analysis method |
| CN109101010B (en) * | 2018-09-30 | 2021-06-11 | 深圳市元征科技股份有限公司 | Automobile fault diagnosis method and related equipment |
| US11586943B2 (en) | 2019-08-12 | 2023-02-21 | Micron Technology, Inc. | Storage and access of neural network inputs in automotive predictive maintenance |
| US11775816B2 (en) | 2019-08-12 | 2023-10-03 | Micron Technology, Inc. | Storage and access of neural network outputs in automotive predictive maintenance |
| US11586194B2 (en) | 2019-08-12 | 2023-02-21 | Micron Technology, Inc. | Storage and access of neural network models of automotive predictive maintenance |
| US11635893B2 (en) | 2019-08-12 | 2023-04-25 | Micron Technology, Inc. | Communications between processors and storage devices in automotive predictive maintenance implemented via artificial neural networks |
| US11748626B2 (en) | 2019-08-12 | 2023-09-05 | Micron Technology, Inc. | Storage devices with neural network accelerators for automotive predictive maintenance |
| US11498388B2 (en) | 2019-08-21 | 2022-11-15 | Micron Technology, Inc. | Intelligent climate control in vehicles |
| US11702086B2 (en) | 2019-08-21 | 2023-07-18 | Micron Technology, Inc. | Intelligent recording of errant vehicle behaviors |
| US11650746B2 (en) | 2019-09-05 | 2023-05-16 | Micron Technology, Inc. | Intelligent write-amplification reduction for data storage devices configured on autonomous vehicles |
| US11693562B2 (en) | 2019-09-05 | 2023-07-04 | Micron Technology, Inc. | Bandwidth optimization for different types of operations scheduled in a data storage device |
| US11250648B2 (en) | 2019-12-18 | 2022-02-15 | Micron Technology, Inc. | Predictive maintenance of automotive transmission |
| US11531339B2 (en) * | 2020-02-14 | 2022-12-20 | Micron Technology, Inc. | Monitoring of drive by wire sensors in vehicles |
| US11709625B2 (en) | 2020-02-14 | 2023-07-25 | Micron Technology, Inc. | Optimization of power usage of data storage devices |
| CN113780592B (en) * | 2021-08-28 | 2022-05-03 | 特斯联科技集团有限公司 | Smart city air data link management system, method and computer program product |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007120960A (en) * | 2005-10-25 | 2007-05-17 | Honda Motor Co Ltd | Failure diagnosis device of sub-power supply |
| CN102868583A (en) * | 2012-06-06 | 2013-01-09 | 浙江吉利汽车研究院有限公司杭州分公司 | FlexRay bus joint with fault diagnosis function |
| CN104155574A (en) * | 2014-07-31 | 2014-11-19 | 国网湖北省电力公司武汉供电公司 | Power distribution network fault classification method based on adaptive neuro-fuzzy inference system |
| CN104202200A (en) * | 2014-09-15 | 2014-12-10 | 中国科学院电工研究所 | FlexRay bus-based online network diagnosis device |
-
2014
- 2014-12-18 CN CN201410795268.9A patent/CN104536435B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007120960A (en) * | 2005-10-25 | 2007-05-17 | Honda Motor Co Ltd | Failure diagnosis device of sub-power supply |
| CN102868583A (en) * | 2012-06-06 | 2013-01-09 | 浙江吉利汽车研究院有限公司杭州分公司 | FlexRay bus joint with fault diagnosis function |
| CN104155574A (en) * | 2014-07-31 | 2014-11-19 | 国网湖北省电力公司武汉供电公司 | Power distribution network fault classification method based on adaptive neuro-fuzzy inference system |
| CN104202200A (en) * | 2014-09-15 | 2014-12-10 | 中国科学院电工研究所 | FlexRay bus-based online network diagnosis device |
Non-Patent Citations (2)
| Title |
|---|
| 基于FlexRay总线的线控转向系统通信网络;段建民，朱亮，于涌川;《仪表技术与传感器》;20091231(第12期);第43-45页 * |
| 基于自适应模糊神经网络的多模型在线故障诊断;师黎;《自动化仪表》;20040731;第25卷(第7期);第10-15页 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11853863B2 (en) | 2019-08-12 | 2023-12-26 | Micron Technology, Inc. | Predictive maintenance of automotive tires |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104536435A (en) | 2015-04-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104536435B (en) | A kind of line control system network inline diagnosis method | |
| Cai et al. | Bayesian networks in fault diagnosis | |
| JP6830540B2 (en) | Cloud-based vehicle failure diagnostic methods, equipment and systems | |
| CN114585983B (en) | Method, device and system for detecting abnormal operation state of equipment | |
| CN109163913A (en) | A kind of Diagnosis method of automobile faults and relevant device | |
| CN104698839B (en) | A kind of multiple agent fault detect based on information interaction and compensating control method | |
| CN112249002B (en) | TD 3-based heuristic series-parallel hybrid power energy management method | |
| Suwatthikul et al. | In-vehicle network level fault diagnostics using fuzzy inference systems | |
| CN107367693A (en) | SOC detection system for power battery of electric vehicle | |
| WO2020121849A1 (en) | Determination device, determination program, determination method, and method for generating neural network model | |
| CN106960069A (en) | A kind of Bayesian network platform with self-learning function | |
| CN110188837A (en) | A kind of MVB network fault diagnosis method based on fuzzy neural | |
| Wang et al. | Research on interactive multi‐model fault diagnosis method of Li‐ion battery based on noise suppression | |
| CN109886439A (en) | Electronic-control vehicle remote diagnosis system and its diagnostic method | |
| Ebner et al. | Model-based dependability analysis of fail-operational electric drivetrains | |
| CN116080470A (en) | Power battery monitoring method and device for electric automobile, server and medium | |
| Wu et al. | Semigroup of fuzzy automata and its application for fast accurate fault diagnosis on machine and anti-fatigue control | |
| CN110018678B (en) | Fault diagnosis method for networked automobile control system | |
| Schwall et al. | Multi-modal diagnostics for vehicle fault detection | |
| Zhang | Model-Based Fault Diagnosis For Automotive Functional Safety | |
| Wotawa et al. | On classification and modeling issues in distributed model-based diagnosis | |
| Guo et al. | Detection and diagnosis of long-term cyber-attacks for predictive energy management system in hevs | |
| Bock et al. | Analytical test effort estimation for multisensor driver assistance systems | |
| Wang et al. | Research on Fault Diagnosis Based on Dynamic causality diagram and Fuzzy Reasoning Fusion Method | |
| Su et al. | Network level fault diagnosis of steer-by-wire system based on ANFIS |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |