CN105882649B - A kind of hybrid vehicle method for diagnosing faults - Google Patents
A kind of hybrid vehicle method for diagnosing faults Download PDFInfo
- Publication number
- CN105882649B CN105882649B CN201610322885.6A CN201610322885A CN105882649B CN 105882649 B CN105882649 B CN 105882649B CN 201610322885 A CN201610322885 A CN 201610322885A CN 105882649 B CN105882649 B CN 105882649B
- Authority
- CN
- China
- Prior art keywords
- signal
- fault
- identifier
- hcu
- suspect
- 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.)
- Expired - Fee Related
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W20/00—Control systems specially adapted for hybrid vehicles
- B60W20/50—Control strategies for responding to system failures, e.g. for fault diagnosis, failsafe operation or limp mode
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W50/02—Ensuring safety in case of control system failures, e.g. by diagnosing, circumventing or fixing failures
- B60W50/0205—Diagnosing or detecting failures; Failure detection models
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
Abstract
The invention discloses a kind of hybrid vehicle method for diagnosing faults, aim to solve the problem that existing hybrid vehicle method for diagnosing faults imperfection, the problem of hybrid vehicle is required short trouble, the diagnosis of closed-loop system intermittent fault can not be met, this method comprises the following steps：First, hybrid vehicle fault detect, fault detect is carried out to hybrid power system critical piece input signal, power source torque response signal and vehicle CAN communication signal, obtains characterizing the identifier of each suspect signal malfunction；2nd, hybrid vehicle fault recognition, each suspect signal malfunction identifier obtained according to step 1, hybrid power system fault recognition is further carried out.
Description
Technical field
The present invention relates to a kind of car fault diagnosis method, and more precisely, the present invention relates to a kind of hybrid vehicle
Method for diagnosing faults.
Background technology
Hybrid vehicle (Hybrid Electric Vehicle, HEV) is dirty with environment as current energy crisis is alleviated
One of important channel of the problems such as dye, significant progress has been obtained in recent years.Orthodox car is different from, hybrid vehicle is one
Individual complicated Mechanical ＆ Electrical Combination System, increased electrokinetic cell, battery management system, motor, electric machine controller etc. in system
Electric elements so that the reliability of hybrid power system is low compared with orthodox car, and the incipient fault that vehicle is likely to occur is also more.
On the one hand, due to the electromagnetic radiation and interference in hybrid power system between each part and equipment, and it is vehicle-mounted logical
Letter system and the constraint of in-vehicle network channel width, quantization error, network congestion and node competition etc. easily cause hybrid power system
System produces intermittent fault.Such failure is different from permanent fault and transient fault, and its duration is short, can occur repeatedly, without
Processing can die away again, be a kind of special impermanent failure, and detection is very difficult.On the other hand, hybrid power system is past
Toward the short trouble caused by signal noise spike be present, such trouble duration is very short, if being directly used in the appearance of system
Mistake control and failure code output certainly will bring detrimental effect to the stability of system, and non-essential code can be caused to deposit
Reserves.Simultaneously because the feedback compensation effect of vehicle closed-loop control system so that hybrid power system itself possesses certain appearance
Wrong ability, when failure is in early stage or smaller amplitude, influence that failure is brought may controlled variable cover, cause
Failure is difficult to detect.
The fault diagnosis technology of orthodox car is more for engine, speed changer and ABS system etc. at present, for above-mentioned mixed
Power vehicle system short trouble and special closed loop intermittent fault etc. are closed, the fault diagnosis technology of orthodox car obviously can not
Meet to require, for hybrid power system establish effective diagnostics architecture for hybrid vehicle security with it is reliable
Property is still very necessary.
Some existing patents, if China Patent Publication No. is CN 101364111A, publication date is on 2 11st, 2009,
Entitled " a kind of hybrid vehicle fault detect and processing method ", the invention is with hybrid vehicle control unit
Fault processing system is troubleshooting decision center, with motor control unit, battery management system, engine management system
Fault processing unit is that subsystem completes fault detect and processing jointly.China Patent Publication No. CN 101941439A, publication date
For on January 12nd, 2011, a kind of entitled " control system failure for avoiding parallel fault for hybrid vehicle
Diagnostic method ", the sequencing that the invention reaches according to fault message, can be real to report out of order ECU to be defined at first
Now to the accurate fault diagnosis functions for the hybrid vehicle electric-control system being made up of multiple ECUs.China Patent Publication No.
CN 102901639A, publication date are on January 30th, 2013, a kind of entitled " the accelerator pedal diagnosis of hybrid vehicle
Method ", the invention provides a kind of accelerator pedal diagnostic method of hybrid vehicle, after solving accelerator pedal long-time use
Sensor internal contact resistance needs to carry out the technical problem of high impedance inspection.China Patent Publication No. CN 104656026A, it is public
It is on May 27th, 2015 to open day, entitled " a kind of hybrid power automobile battery overcharges diagnostic method and system ", the invention
A kind of hybrid power automobile battery that whether can be overcharged, not influenceed by variation of ambient temperature with accurate judgement battery is provided to overcharge
Diagnostic method and system.
In summary, the fault diagnosis technology of orthodox car can not meet the needs of hybrid vehicle fault diagnosis；It is existing
Patent in terms of some hybrid vehicles is more for system fault diagnosis framework and each separate part of hybrid power system
Diagnosis and detection；And the short trouble for often occurring in hybrid power system, intermittent fault etc., not yet propose at present effective
Fault diagnosis solution.
The content of the invention
Present invention seek to address that existing hybrid vehicle method for diagnosing faults imperfection, hybrid vehicle can not be met
The problem of requiring short trouble, the diagnosis of closed-loop system intermittent fault, proposes a kind of reasonable, perfect hybrid vehicle failure
Diagnostic method.
In order to solve the above technical problems, the present invention adopts the following technical scheme that realization：
A kind of hybrid vehicle method for diagnosing faults, comprises the following steps：
Step 1: hybrid vehicle fault detect, to hybrid power system critical piece input signal, power source torque
Response signal and vehicle CAN communication signal carry out fault detect, obtain characterizing the identifier of each suspect signal malfunction, bag
Include following steps：
1) signal-obtaining carries out signal reading successively with processing, whole car controller of hybrid electric car HCU to each suspect signal
Take, signal A/D is changed with demarcation and signal filtering process, the suspect signal after being handled；
2) signal fault detect, whole car controller of hybrid electric car HCU will be by above-mentioned steps 1) handle after it is to be checked
Signal carries out logical relation between signal character detection and signal and detected, and obtains characterizing the identifier of signal fault state；
3) malfunction exports, and what whole car controller of hybrid electric car HCU output step 2) fault detects obtained respectively treats
Signal fault status identifier is examined, for hybrid vehicle fault recognition.
Step 2: hybrid vehicle fault recognition, each suspect signal malfunction identifier obtained according to step 1,
Further carry out hybrid power system fault recognition, including herein below：
1) the fault recognition method based on timing, whole car controller of hybrid electric car HCU foundations pass through above-mentioned steps one
Each suspect signal malfunction identifier that hybrid vehicle fault detect obtains, it is true to carry out failure by the way of timing
Recognize, with the short trouble of filtering appts；
2) the fault recognition method based on counting, whole car controller of hybrid electric car HCU foundations pass through above-mentioned steps one
Each suspect signal malfunction identifier that hybrid vehicle fault detect obtains, it is true to carry out failure by the way of counting
Recognize, realize that system intermittent fault confirms.
A kind of hybrid vehicle method for diagnosing faults of the present invention, wherein, step 1 hybrid vehicle failure
Signal-obtaining described in detection specifically includes following steps with processing：
(1) suspect signal is read and A/D conversions are with demarcating：Whole car controller of hybrid electric car reads each suspect signal,
A/D conversions are carried out to each suspect signal respectively, and the digital quantity obtained after being changed according to formula (1) to A/D is demarcated.
In formula, f --- signal post definite value
Vs--- the value after voltage signal A/D conversions
Vmin--- the lower limit of signal normal working voltage
Vmax--- the higher limit of signal normal working voltage
rh--- the higher limit of the actual span of signal
rl--- the lower limit of the actual span of signal
(2) signal filtering process, whole car controller of hybrid electric car by above-mentioned steps (1) A/D to changing and demarcating
Each suspect signal obtained afterwards, processing is filtered according to formula (2), to filter out the noise in signal, suppresses signal interference.
Y (k)=(1-Q) y (k-1)+Qx (k) (2)
In formula, y (k) --- current time filtering output value
Y (k-1) --- previous moment filtering output value
X (k) --- current instance sample value
Q --- filter factor, Q=dT/ (dT+ τ)
DT --- the sampling period
τ --- time constant
A kind of hybrid vehicle method for diagnosing faults of the present invention, wherein, step 1 hybrid vehicle failure
Signal fault detection described in detection specifically includes herein below：
(1) signal character detection, whole car controller of hybrid electric car HCU enter respectively to the signal after handling after filtering
The stuck failure of row signal, signal slope abnormal failure and the detection of signalc threshold fault signature.It is described in detail below：
A. the stuck detection of signal, hybrid power whole vehicle controller HCU enter at line delay to the signal after handling after filtering
Reason obtains previous moment suspect signal, and makes the difference extraction absolute value with current time suspect signal, when absolute value is higher than signal card
During dead detection threshold value, HCU judges that stuck failure is not present in suspect signal, and it is the to provide the stuck fault identifier of suspect signal
Two class identifiers；When absolute value is less than the stuck detection threshold value of signal, what HCU judged that suspect signal has a stuck failure can
Energy.The stuck fault signature of signal is usually the value for a certain signal long period maintaining a certain fixation, therefore HCU judges letter to be checked
Number exist and just to start timing after the possibility of stuck failure, when absolute value be less than the stuck detection threshold value of signal time exceed it is pre-
If during stuck fault time threshold value, it is the first class identifier that HCU, which provides the stuck fault identifier of suspect signal,；If stuck
Within fault time threshold value, signal difference absolute value recovers to higher than the dead detection threshold value of signal card, then HCU judges letter to be checked
Number stuck failure is not present, is the second class identifier by the identifier that timer reset and provided the stuck failure of suspect signal.
B. signal slope detects, and hybrid power whole vehicle controller HCU enters at line delay to the signal after handling after filtering
Reason obtains previous moment suspect signal, and calculates suspect signal slope according to formula (3).When suspect signal slope is oblique beyond signal
Rate detection threshold value, then HCU judges that suspect signal has slope abnormal failure, and provides signal slope abnormal failure identifier and be
First class identifier；When suspect signal slope is less than signal slope detection threshold value, then HCU judges that signal slope is normal, and
It is the second class identifier to provide signal slope abnormal failure identifier.
Dx (i)/dt=| x (i)-x (i-1) |/dT (3)
In formula, dx (i)/dt --- suspect signal slope
X (i) --- current time suspect signal
X (i-1) --- previous moment suspect signal
C. signalc threshold detects, hybrid power whole vehicle controller HCU by signal and the signal after handling after filtering just
Threshold value corresponding to normal scope is compared, and when suspect signal is higher than signalc threshold upper limit of detection value, HCU judges suspect signal mistake
Major break down, and the given excessive fault identifier of suspect signal is the first class identifier；When suspect signal detects less than signalc threshold
Lower limit, then HCU judges that suspect signal crosses glitch, and it is the first class identifier to provide the too small fault identifier of suspect signal；
When suspect signal is between signalc threshold upper limit of detection value and lower limit, HCU judges suspect signal without thresholding abnormal failure, and
Provide that suspect signal is excessive, too small fault identifier is the second class fault identifier.
(2) signal logic relation detect, whole car controller of hybrid electric car HCU by suspect signal and system therewith
The mutual verification of logical relation between associated signal, judge that suspect signal whether there is logical relation abnormal failure.When HCU is examined
Measure suspect signal and logical relation abnormal failure be present, then provide signal logic relation abnormality identifier and identified for the first kind
Symbol；When HCU detection suspect signals and correlation signal logical relation are correct, then providing signal logic relation abnormality identifier is
Second class identifier.
A kind of hybrid vehicle method for diagnosing faults of the present invention, wherein, step 2 hybrid vehicle failure
The fault recognition method based on timing described in confirmation specifically includes following rule：
(1) when suspect signal malfunction identifier zero setting, then hybrid power whole vehicle controller HCU confirm current demand signal without
Failure, timer are reset；
(2) when malfunction identifier is not zero, then hybrid power whole vehicle controller HCU calculates failure mark by timer
Know the reset time of symbol；
(3) the fault time thresholding of setting is exceeded when the suspect signal fail timer time, then HCU confirms suspect signal hair
Raw failure, and provide the identifier of fault recognition；
(4) when suspect signal zero setting in the fault time thresholding of setting, then hybrid power whole vehicle controller HCU confirmations are worked as
Front signal fault-free, fail timer are reset.
A kind of hybrid vehicle method for diagnosing faults of the present invention, wherein, step 2 hybrid vehicle failure
The fault recognition method based on counting described in confirmation specifically includes following rule：
(1) when signal fault status identifier zero setting, counter subtracts one；
(2) when malfunction identifier is not zero, counter adds one；
(3) when current time failure counter count value is between given count upper-limit and counting lower limit, counter is defeated
Go out actual count value；
(4) when current time failure counter count value exceedes count upper-limit, then counter exports count upper-limit value；
(5) current time failure counter count value, which is less than or equal to, counts lower limit, and current demand signal malfunction mark
When knowing symbol zero setting, then counter output counts lower limit；
(6) current time failure counter count value, which is less than or equal to, counts lower limit, and current demand signal malfunction mark
When knowledge symbol is not zero, then counter O reset.
(7) when failure counter counting output value reaches count upper-limit, HCU confirms the failure, and exports fault recognition
State.
Compared with prior art the beneficial effects of the invention are as follows：
1. removing the noise in signal and interference using signal filtering method in the fault detect stage, hybrid power system is improved
The precision of system fault detect；
2. the logical relation between the various features and unlike signal of comprehensive utilization signal carries out fault detect, improve mixed
Close the reliability of dynamical system fault detect；
3. hybrid vehicle failure is confirmed by the fault recognition method based on timing, can be by the duration
Shorter short trouble filters out, and excludes due to fault warning wrong caused by signal interference etc.；
4. hybrid vehicle failure is confirmed by the fault recognition method based on counting, it is possible to achieve mixing is dynamic
Force system intermittent fault diagnoses, and can prevent not long to the duration in fault detect but frequently intermittent fault occurs and leak
Report.
Brief description of the drawings
The present invention is further illustrated below in conjunction with the accompanying drawings：
Fig. 1 is hybrid vehicle method for diagnosing faults overall flow figure of the present invention；
Fig. 2 is hybrid vehicle fault detection method flow chart of the present invention；
Fig. 3 is the stuck overhaul flow chart of hybrid vehicle signal of the present invention；
Fig. 4 is hybrid vehicle signal slope overhaul flow chart of the present invention；
Fig. 5 is hybrid vehicle signalc threshold overhaul flow chart of the present invention；
Fig. 6 is hybrid vehicle signal logic relation overhaul flow chart of the present invention；
Fig. 7 is the hybrid vehicle fault recognition method flow diagram of the present invention based on timing；
Fig. 8 is the hybrid vehicle fault recognition method flow diagram of the present invention based on counting；
Fig. 9 is shape before E-Gas signal filtering in hybrid vehicle method for diagnosing faults embodiment of the present invention
State figure；
Figure 10 is in hybrid vehicle method for diagnosing faults embodiment of the present invention at E-Gas signal filtering
State diagram after reason；
Figure 11 is the stuck event of E-Gas signal in hybrid vehicle method for diagnosing faults embodiment of the present invention
Hinder state diagram；
Figure 12 is the stuck event of E-Gas signal in hybrid vehicle method for diagnosing faults embodiment of the present invention
Hinder testing result figure；
Figure 13 is E-Gas signal slope event in hybrid vehicle method for diagnosing faults embodiment of the present invention
Hinder state diagram；
Figure 14 is E-Gas signal slope event in hybrid vehicle method for diagnosing faults embodiment of the present invention
Hinder testing result figure；
Figure 15 is E-Gas signalc threshold event in hybrid vehicle method for diagnosing faults embodiment of the present invention
Hinder state diagram.
Figure 16 is E-Gas signalc threshold event in hybrid vehicle method for diagnosing faults embodiment of the present invention
Hinder testing result figure；
Figure 17 is that E-Gas signal logic closes in hybrid vehicle method for diagnosing faults embodiment of the present invention
System's detection E-Gas signal condition figure；
Figure 18 is that E-Gas signal logic closes in hybrid vehicle method for diagnosing faults embodiment of the present invention
System's detection IS Idle Switch signal condition figure；
Figure 19 is that E-Gas signal logic closes in hybrid vehicle method for diagnosing faults embodiment of the present invention
It is failure detection result figure；
Figure 20 is the malfunction based on timing in hybrid vehicle method for diagnosing faults embodiment of the present invention
Figure；
Figure 21 is the failure timing based on timing in hybrid vehicle method for diagnosing faults embodiment of the present invention
Time；
Figure 22 is the fault recognition based on timing in hybrid vehicle method for diagnosing faults embodiment of the present invention
Result figure；
Figure 23 is the malfunction based on counting in hybrid vehicle method for diagnosing faults embodiment of the present invention
Figure；
Figure 24 is the failure count based on counting in hybrid vehicle method for diagnosing faults embodiment of the present invention
Device count value；
Figure 25 is the fault recognition based on counting in hybrid vehicle method for diagnosing faults embodiment of the present invention
Result figure.
Embodiment
The present invention is explained in detail below in conjunction with the accompanying drawings：
Refering to Fig. 1, hybrid vehicle fault diagnosis flow scheme of the present invention is divided into two steps：Step 1 is mixing
Power vehicle fault detect, to hybrid power system critical piece input signal, power source torque response signal and vehicle CAN
Signal of communication carries out fault detect, including signal-obtaining exports with processing, signal fault detection and malfunction；Step 2 is
Hybrid vehicle fault recognition, including short trouble filters out and intermittent fault confirms that wherein short trouble is filtered out using base
In the fault recognition method of timing, intermittent fault confirms to use the fault recognition method based on counting.
First, by step 1 hybrid vehicle fault detect, the error condition identifier of all kinds of failures is obtained；Then
Foundation fault recognition method of the step 2 based on timing and the fault recognition method based on counting, to hybrid power system in short-term
Failure is confirmed with intermittent fault.
Hybrid vehicle method for diagnosing faults of the present invention comprises the following steps：
Step 1: hybrid vehicle fault detect, refering to Fig. 1, hybrid power system fault detect content includes：(1)
Critical piece input signal detection, detection are input to HEV controllers (Hybrid Electric Vehicle Control
Unit, HCU) each signal it is whether normal.For example, whether exceed the detection of its minimax threshold value to signal, to signal
The whether abnormal detection of slope and whether the logical relation between each signal is normally detected.(2) power source torque
Rationality checking, whether consistent with the desired torque that control strategy provides detect the actual torque response of each power source.Example
Such as, by detecting the expectation torque of the current actual torque of motor and motor, to ensure the accurate control to power source output torque
System, so as to ensure that HEV has excellent power performance and economy.(3) vehicle CAN communication detect, detect HEV in all parts it
Between CAN communication it is whether normal.
Refering to Fig. 2, hybrid vehicle fault detection method step of the present invention is as follows：
(1) signal-obtaining and processing
Hybrid power whole vehicle controller HCU reads each suspect signal, such as accelerator pedal signal, IS Idle Switch signal etc.,
And A/D conversions and demarcation and signal filtering process are carried out successively.
1) signal A/D conversion and demarcation
Hybrid vehicle HCU carries out A/D conversions respectively to each signal, and voltage signal is converted into digital quantity, Ran Hougen
Obtained data signal is demarcated according to formula (1).
In formula, f --- signal post definite value
Vs--- the value after voltage signal A/D conversions
Vmin--- the lower limit of signal normal working voltage
Vmax--- the higher limit of signal normal working voltage
rh--- the higher limit of the actual span of signal
rl--- the lower limit of the actual span of signal
2) signal filtering process
Before the fault detect of each suspect signal is carried out, the signal after HCU changes according to formula (2) to A/D is carried out
Filtering process.The purpose being filtered to signal is in the noise in signal is removed, when suppressing interference signal, while being filtered
Signal is undistorted before and after the guarantee that should try one's best filters.
Y (k)=(1-Q) y (k-1)+Qx (k) (2)
In formula, y (k) --- current time filtering output value
Y (k-1) --- previous moment filtering output value
X (k) --- current instance sample value
Q --- filter factor, Q=dT/ (dT+ τ)
DT --- the sampling period
τ --- time constant
(2) each signal fault detection
The suspect signal of processing step after filtering is subjected to signal fault detection, mainly include signal character detection and
Signal logic relation detects.Wherein, signal character detection includes the stuck detection of signal, signal slope detection and signalc threshold inspection
Survey.
1) signal character detection
Refering to Fig. 2, hybrid vehicle signal fault feature detection mainly includes the stuck detection of signal, signal slope detection
And signalc threshold detection.It is specifically described as follows：
1. the stuck detection of signal：The suspect signal x (i) after filtering process is subjected to delay process, obtained refering to Fig. 3, HCU
Extraction absolute value is made the difference to previous moment suspect signal x (i-1), and with current demand signal x (i), obtains signal difference △ x (i)；When
Suspect signal x (i) and previous moment suspect signal x (i-1) difference △ x (i) are higher than the stuck detection threshold value △ x of signalLIMWhen,
HCU judges that stuck failure is not present in suspect signal x (i), and provides the identifier that stuck failure is not present in suspect signal x (i)
Err01=0；Conversely, when suspect signal x (i) and previous moment suspect signal x (i-1) difference △ x (i) are less than the stuck inspection of signal
Survey threshold value △ xLIMWhen, then HCU judges that suspect signal x (i) has the possibility of stuck failure.The stuck fault signature of signal is general
The value for a certain signal long period maintaining a certain fixation, thus HCU judge that suspect signal x (i) has a stuck failure can
Timing can just be started afterwards, when suspect signal x (i) and previous moment suspect signal x (i-1) difference △ x (i) are stuck less than signal
Detection threshold value △ xLIMTime t (i) exceed stuck fault time thresholding tLIMWhen, card be present in HCU output suspect signal x (i)
The identifier Err01=1 of dead failure；If in stuck fault time thresholding tLIMWithin, suspect signal x (i) treats with previous moment
Inspection signal x (i-1) difference △ x (i) recover extremely to be higher than the dead detection threshold value △ x of signal cardLIM, then HCU judge suspect signal x (i)
In the absence of stuck failure, timer is reset and provides identifier Err01=0s of the suspect signal x (i) in the absence of stuck failure.
2. signal slope detects：The suspect signal x (i) after filtering process is subjected to delay process, obtained refering to Fig. 4, HCU
To previous moment suspect signal x (i-1).The slope of suspect signal can be calculated according to formula (3).When suspect signal slope
Beyond suspect signal slope detection threshold value dxLIMWhen, HCU directly exports the identifier of suspect signal x (i) slope abnormal failures
Err02=1；Conversely, when suspect signal slope is less than suspect signal slope detection threshold value dxLIMWhen, HCU judges suspect signal x
(i) slope is normal, and exports the identifier Err02=0 of suspect signal x (i) slope abnormal failures.
Dx (i)/dt=| x (i)-x (i-1) |/dT (3)
In formula, dx (i)/dt --- suspect signal slope
X (i) --- current time suspect signal
X (i-1) --- previous moment suspect signal
3. signalc threshold detects：Refering to Fig. 5, HCU is by the suspect signal x (i) after filtering process and the normal model of the signal
Threshold value corresponding to enclosing is compared, if suspect signal x (i) is higher than signalc threshold upper limit of detection value xmax, then HCU judge treat
Examine signal x (i) and cross major break down, and provide the excessive fault identifier Err03=1 of signal；If suspect signal x (i) is less than signal
Threshold detection lower limit xmin, then HCU judges that suspect signal x (i) crosses glitch, and provides the too small fault identifier Err04 of signal
=1；Otherwise, if suspect signal x (i) is in signalc threshold Monitoring lower-cut value xminWith signalc threshold upper limit of detection value xmaxIt
Between, then HCU judges suspect signal x (i) normally, and provides that signal is excessive, too small fault identifier is Err03=0, Err04=
0。
2) signal logic relation detects
Certain logical relation often be present between each signal of hybrid power system, hybrid vehicle failure can be used as
The foundation of detection.Refering to Fig. 6, HCU is by suspect signal x (i) and is further associated the phase of logical relation between signal y (i)
Mutually verification, judge that suspect signal x (i) whether there is logical relation abnormal failure.Logic pass between HCU detects signal be present
It is abnormal failure, then provides signal logic relation abnormal failure status identifier Err05=1；Conversely, when HCU detects each letter
Logical relation is normal between number, then provides signal logic relation abnormal failure status identifier Err05=0.
(3) malfunction is exported
After signal character detection and the detection of signal logic relation, HCU exports the malfunction mark of each signal
Symbol, and further judge that current demand signal whether there is failure by fault recognition.
Step 2: hybrid vehicle fault recognition
After completing fault detect to each signal of hybrid vehicle in step 1, the malfunction mark of each signal has been obtained
Know symbol.The fault recognition of hybrid vehicle system is carried out according to each signal fault status identifier.Wherein, by based on timing
Fault recognition method filtering appts short trouble, to exclude due to false alarm caused by signal interference, reduce inessential
Code amount of storage；And the confirmation of system intermittent fault is realized by the method based on counting, reduce the rate of failing to report of fault diagnosis.
(1) fault recognition based on timing
Refering to Fig. 7, all kinds of malfunction identifiers of HCU reading system failure detection outputs, malfunction identifier zero setting,
HCU confirms current demand signal fault-free, and exports fault recognition result；When malfunction identifier is not zero, i.e. upper strata event
Barrier testing result shows that suspect signal current time has failure, then HCU starts timing by timer.The principle of timing such as public affairs
Shown in formula (4), the time initial value that each timing starts is disposed as zero.In each sampling period, HCU is all in accordance with current signal
Malfunction identifier and the fault time threshold value of setting are judged.Setting fault time threshold value it
It is interior, if malfunction identifier does not have zero setting, timer just on the basis of previous moment adds up fault time T (i-1),
Accumulated samples cycle time dT again, the time of failure T (i) at current time is obtained, until failure cumulative time T (i) exceedes
The time threshold threshold value of setting, HCU confirms that current demand signal has failure, and exports fault recognition result；In the failure of setting
Between within threshold value, if malfunction identifier zero setting, HCU confirms current demand signal fault-free, and timer is reset.
T (i)=T (i-1)+dT (4)
In formula, T (i) --- the current sample period internal fault cumulative time
T (i-1) --- the previous sampling period internal fault cumulative time
(2) fault recognition based on counting
Refering to Fig. 8, HCU reads all kinds of fault status informations of upper strata fault detect output, and according to current failure state
Counted, the fault recognition state based on counting is finally worth to according to counting.
The counting rule of failure counter is described below：
1) when signal fault status identifier zero setting, counter subtracts one；
2) when malfunction identifier is not zero, counter adds one；
3) when current time failure counter count value is between given count upper-limit ThdUL and counting lower limit ThdLL,
Counter exports actual count value；
4) when current time failure counter count value exceedes count upper-limit ThdUL, then counter exports count upper-limit value
ThdUL；
5) current time failure counter count value, which is less than or equal to, counts lower limit ThdLL, and current demand signal failure shape
During state identifier zero setting, then counter output counts lower limit ThdLL；
6) current time failure counter count value, which is less than or equal to, counts lower limit ThdLL, and current demand signal failure shape
When state identifier is not zero, then counter O reset.
7) when failure counter count value reaches count upper-limit ThdUL, HCU confirms the failure, and it is true to export failure
Recognize state.
According to above-mentioned counting rule, hybrid power whole vehicle controller HCU is when carrying out fault recognition, when long before signal
Between in correct status and current time, there occurs during failure, confirm that the time needed for failure is longer；And short before signal
When occurring failure in the time and currently breaking down, confirm that the time of failure is shorter.It can thus be directed to what failure occurred
Time and the intermittent fault being spaced in more reasonably identifying system.
Embodiment
The present embodiment is based on hybrid vehicle E-Gas signal fault diagnosis, verifies fault diagnosis proposed by the present invention
The validity of method.
Refering to Fig. 9, Figure 10, filtered signal compares primary signal, and the time, the signal after filtering process was oblique about at 2s
Rate is there occurs slight change, but its change and size are basically identical；Begin signal error caused by the noise jamming of signal simultaneously
It is filtered out, normal scope is at three of signal value more than 100% after the filtered processing of spike between 4s-6s in the time
It is interior.
Refering to Figure 11, Figure 12, the time passes through after 40s, and E-Gas is stuck, and E-Gas opening amount signal maintains for a long time
In a fixed value.Now HCU detects the stuck malfunction of E-Gas after fault detect after a while,
And export malfunction identifier Err01=1.
Refering to Figure 13, Figure 14, time 30s-40s or so, E-Gas loose contact causes E-Gas opening amount signal to be sent out
Raw to be mutated, now HCU detects the abnormal malfunction of E-Gas slope after fault detect after a while, and
Export malfunction identifier Err02=1.
Refering to Figure 15, Figure 16, time 30s-40s or so, E-Gas opening amount signal exceedes the maximum of normal signal value
100%, now HCU detect the excessive malfunction of E-Gas signal after fault detect after a while, it is and defeated
Go out malfunction identifier Err03=1.
E-Gas opening amount signal and E-Gas IS Idle Switch signal have certain logical relation.E-Gas aperture
When signal is less than certain value, IS Idle Switch signal value is 1；When E-Gas opening amount signal is more than certain value, IS Idle Switch signal
It is changed into 0.Whether the logical relation that the detection of signal logic relation is detected between the two signals is normal, and logic between the signals
Firmware-error state signal is exported when relation occurs abnormal.Refering to Figure 17, Figure 18 and Figure 19, the time before the time is 45s
In section, IS Idle Switch signal is changed with E-Gas opening amount signal, and when E-Gas opening amount signal is 0, idling is opened
Off status is 1；When E-Gas opening amount signal is more than 0, IS Idle Switch state is 0.After about 45s, signal logic relation
Generation is abnormal, at this moment the logical relation between E-Gas opening amount signal and E-Gas IS Idle Switch signal no longer with above institute
State unanimously, now HCU detects the abnormal failure of E-Gas signal logic relation after fault detect after a while
State, and export malfunction identifier Err05=1.
Refering to Figure 20, Figure 21 and Figure 22, the fault recognition method proposed by the present invention based on timing, can be examined according to failure
The malfunction identifier record failure duration measured.When trouble duration exceedes given threshold value, timing
Malfunction after confirmation is changed into 1.And when failure duration is less than given threshold value, the malfunction after timing confirmation
For 0, that is, short trouble present in system is filtered out.
It is of the invention by the count upper-limit ThdUL values in the fault recognition method based on counting according to the standards of ISO 26262
For 127, it is -128 to count lower limit ThdLL values.Refering to Figure 23, Figure 24 and Figure 25, when malfunction identifier zero setting, failure
State counter count value reduces；When malfunction is not zero, malfunction counter becomes big；When count value reaches
During count upper-limit 127, HCU output fault recognition states.
Claims (5)
1. a kind of hybrid vehicle method for diagnosing faults, it is characterised in that comprise the following steps：
Step 1: hybrid vehicle fault detect, is responded to hybrid power system critical piece input signal, power source torque
Signal and vehicle CAN communication signal carry out fault detect, obtain characterizing the identifier of each suspect signal malfunction, including with
Lower step：
1) signal-obtaining with processing, whole car controller of hybrid electric car (HCU) each suspect signal is carried out successively signal-obtaining,
Signal A/D is changed with demarcation and signal filtering process, the suspect signal after being handled；
2) signal fault detect, whole car controller of hybrid electric car (HCU) will be by above-mentioned steps 1) handle after letter to be checked
Number carrying out logical relation between signal character detection and signal detects, and obtains characterizing the identifier of signal fault state；
3) malfunction exports, and whole car controller of hybrid electric car (HCU) output step 2) fault detect obtains each to be checked
Signal fault status identifier, for hybrid vehicle fault recognition；
Step 2: hybrid vehicle fault recognition, each suspect signal malfunction identifier obtained according to step 1, enter one
Step carries out hybrid power system fault recognition, including herein below：
1) the fault recognition method based on timing, whole car controller of hybrid electric car (HCU) is according to mixed by above-mentioned steps one
Each suspect signal malfunction identifier that power vehicle fault detect obtains is closed, fault recognition is carried out by the way of timing,
With the short trouble of filtering appts；
2) the fault recognition method based on counting, whole car controller of hybrid electric car (HCU) is according to mixed by above-mentioned steps one
Each suspect signal malfunction identifier that power vehicle fault detect obtains is closed, fault recognition is carried out by the way of counting,
Realize that system intermittent fault confirms.
2. according to a kind of hybrid vehicle method for diagnosing faults described in claim 1, it is characterised in that the step 1 is mixed
Close the signal-obtaining described in power vehicle fault detect and specifically include following steps with processing：
(1) suspect signal is read and A/D conversions are with demarcating：Whole car controller of hybrid electric car (HCU) reads each suspect signal,
A/D conversions are carried out to each suspect signal respectively, and the digital quantity obtained after being changed according to formula (1) to A/D is demarcated；
<mrow>
<mi>f</mi>
<mo>=</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>V</mi>
<mi>s</mi>
</msub>
<mo>-</mo>
<msub>
<mi>V</mi>
<mrow>
<mi>m</mi>
<mi>i</mi>
<mi>n</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>*</mo>
<mfrac>
<mrow>
<msub>
<mi>r</mi>
<mi>h</mi>
</msub>
<mo>-</mo>
<msub>
<mi>r</mi>
<mi>l</mi>
</msub>
</mrow>
<mrow>
<msub>
<mi>V</mi>
<mi>max</mi>
</msub>
<mo>-</mo>
<msub>
<mi>V</mi>
<mrow>
<mi>m</mi>
<mi>i</mi>
<mi>n</mi>
</mrow>
</msub>
</mrow>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula, f --- signal post definite value
Vs--- the value after voltage signal A/D conversions
Vmin--- the lower limit of signal normal working voltage
Vmax--- the higher limit of signal normal working voltage
rh--- the higher limit of the actual span of signal
rl--- the lower limit of the actual span of signal
(2) signal filtering process, whole car controller of hybrid electric car (HCU) by above-mentioned steps (1) A/D to changing and demarcating
Each suspect signal obtained afterwards, processing is filtered according to formula (2), to filter out the noise in signal, suppresses signal interference；
Y (k)=(1-Q) y (k-1)+Qx (k) (2)
In formula, y (k) --- current time filtering output value
Y (k-1) --- previous moment filtering output value
X (k) --- current instance sample value
Q --- filter factor, Q=dT/ (dT+ τ)
DT --- the sampling period
τ --- time constant.
3. according to a kind of hybrid vehicle method for diagnosing faults described in claim 1, it is characterised in that the step 1 is mixed
Close the detection of the signal fault described in power vehicle fault detect and specifically include herein below：
(1) signal character detection, whole car controller of hybrid electric car (HCU) are carried out respectively to the signal after handling after filtering
The stuck failure of signal, signal slope abnormal failure and the detection of signalc threshold fault signature, it is described in detail below：
A. the stuck detection of signal, whole car controller of hybrid electric car (HCU) enter line delay to the signal after handling after filtering
Processing obtains previous moment suspect signal, and makes the difference extraction absolute value with current time suspect signal, when absolute value is higher than signal
During stuck detection threshold value, HCU judges that stuck failure is not present in suspect signal, and provides the stuck fault identifier of suspect signal and be
Second class identifier；When absolute value is less than the stuck detection threshold value of signal, what HCU judged that suspect signal has a stuck failure can
Can, the stuck fault signature of signal is usually the value for a certain signal long period maintaining a certain fixation, therefore HCU judges letter to be checked
Number exist and just to start timing after the possibility of stuck failure, when absolute value be less than the stuck detection threshold value of signal time exceed it is pre-
If during stuck fault time threshold value, it is the first class identifier that HCU, which provides the stuck fault identifier of suspect signal,；If stuck
Within fault time threshold value, signal difference absolute value recovers to higher than the dead detection threshold value of signal card, then HCU judges letter to be checked
Number stuck failure is not present, is the second class identifier by the identifier that timer reset and provided the stuck failure of suspect signal；
B. signal slope detects, and whole car controller of hybrid electric car (HCU) enters line delay to the signal after handling after filtering
Processing obtains previous moment suspect signal, and calculates suspect signal slope according to formula (3), when suspect signal slope exceeds signal
Slope detection threshold value, then HCU judges that suspect signal has slope abnormal failure, and provides signal slope abnormal failure identifier
For the first class identifier；When suspect signal slope is less than signal slope detection threshold value, then HCU judges that signal slope is normal,
And it is the second class identifier to provide signal slope abnormal failure identifier；
Dx (i)/dt=| x (i)-x (i-1) |/dT (3)
In formula, dx (i)/dt --- suspect signal slope
X (i) --- current time suspect signal
X (i-1) --- previous moment suspect signal
C. signalc threshold detects, and whole car controller of hybrid electric car (HCU) is by signal and the signal after handling after filtering
Threshold value is compared corresponding to normal range (NR), and when suspect signal is higher than signalc threshold upper limit of detection value, HCU judges suspect signal
Major break down is crossed, and the given excessive fault identifier of suspect signal is the first class identifier；When suspect signal is examined less than signalc threshold
Lower limit is surveyed, then HCU judges that suspect signal crosses glitch, and provides the too small fault identifier of suspect signal and identified for the first kind
Symbol；When suspect signal is between signalc threshold upper limit of detection value and lower limit, HCU judges suspect signal without the abnormal event of thresholding
Barrier, and provide that suspect signal is excessive, too small fault identifier is the second class fault identifier；
(2) signal logic relation detects, and whole car controller of hybrid electric car (HCU) passes through phase therewith in suspect signal and system
The mutual verification of logical relation between correlation signal, judge that suspect signal whether there is logical relation abnormal failure, when HCU is detected
Go out suspect signal and logical relation abnormal failure be present, then provide signal logic relation abnormality identifier and identified for the first kind
Symbol；When HCU detection suspect signals and correlation signal logical relation are correct, then providing signal logic relation abnormality identifier is
Second class identifier.
4. according to a kind of hybrid vehicle method for diagnosing faults described in claim 1, it is characterised in that the step 2 is mixed
Close the fault recognition method based on timing described in power vehicle fault recognition and specifically include following rule：
(1) when suspect signal malfunction identifier zero setting, then whole car controller of hybrid electric car (HCU) confirms current demand signal
Fault-free, timer are reset；
(2) when malfunction identifier is not zero, then whole car controller of hybrid electric car (HCU) calculates failure by timer
The reset time of identifier；
(3) the fault time thresholding of setting is exceeded when the suspect signal fail timer time, then HCU confirms that event occurs for suspect signal
Barrier, and provide the identifier of fault recognition；
(4) when suspect signal zero setting in the fault time thresholding of setting, then whole car controller of hybrid electric car (HCU) confirmation
Current demand signal fault-free, fail timer are reset.
5. according to a kind of hybrid vehicle method for diagnosing faults described in claim 1, it is characterised in that the step 2 is mixed
Close the fault recognition method based on counting described in power vehicle fault recognition and specifically include following rule：
(1) when signal fault status identifier zero setting, counter subtracts one；
(2) when malfunction identifier is not zero, counter adds one；
(3) when current time failure counter count value is between given count upper-limit and counting lower limit, counter output is real
Border count value；
(4) when current time failure counter count value exceedes count upper-limit, then counter exports count upper-limit value；
(5) current time failure counter count value, which is less than or equal to, counts lower limit, and current demand signal malfunction identifier
During zero setting, then counter output counts lower limit；
(6) current time failure counter count value, which is less than or equal to, counts lower limit, and current demand signal malfunction identifier
When being not zero, then counter O reset；
(7) when failure counter counting output value reaches count upper-limit, HCU confirms the failure, and exports fault recognition state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610322885.6A CN105882649B (en) | 2016-05-16 | 2016-05-16 | A kind of hybrid vehicle method for diagnosing faults |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610322885.6A CN105882649B (en) | 2016-05-16 | 2016-05-16 | A kind of hybrid vehicle method for diagnosing faults |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105882649A CN105882649A (en) | 2016-08-24 |
| CN105882649B true CN105882649B (en) | 2018-01-26 |
Family
ID=56716234
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610322885.6A Expired - Fee Related CN105882649B (en) | 2016-05-16 | 2016-05-16 | A kind of hybrid vehicle method for diagnosing faults |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105882649B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107487230B (en) * | 2016-12-06 | 2020-06-02 | 宝沃汽车（中国）有限公司 | Fault control method of vehicle, motor controller and vehicle |
| CN107128312B (en) * | 2017-05-27 | 2019-03-12 | 奇瑞汽车股份有限公司 | Hybrid vehicle fault diagnosis management system and method |
| CN109547284A (en) * | 2018-11-05 | 2019-03-29 | 北京长城华冠汽车科技股份有限公司 | It is a kind of to detect the method, apparatus and electric car that controller LAN message goes offline |
| CN109484392B (en) * | 2018-11-26 | 2020-03-10 | 吉林大学 | Integrated fault diagnosis method for multi-wheel distributed hybrid power system |
| CN109920082B (en) * | 2019-03-11 | 2021-01-15 | 辽宁工业大学 | Fault diagnosis and early warning method for hybrid electric vehicle |
| CN111273635B (en) * | 2020-01-19 | 2021-02-12 | 浙江大学 | Unknown anomaly detection method for industrial control equipment |
| CN111381583A (en) * | 2020-03-23 | 2020-07-07 | 北京经纬恒润科技有限公司 | Control system fault monitoring method and device |
| CN112590615B (en) * | 2020-12-30 | 2022-06-07 | 大运汽车股份有限公司 | Pure electric power-on and power-off control method for new-energy commercial vehicle |
| CN116298866A (en) * | 2023-02-10 | 2023-06-23 | 江苏吉泰科电气有限责任公司 | Detection method for ground fault of motor controller output |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006031675A2 (en) * | 2004-09-10 | 2006-03-23 | General Motors Corporation | Fault tolerant control system |
| CN101610142B (en) * | 2009-06-30 | 2013-03-27 | 奇瑞汽车股份有限公司 | Detection method of CAN of hybrid motor vehicle |
| CN101879865A (en) * | 2010-07-02 | 2010-11-10 | 奇瑞汽车股份有限公司 | Vehicle breakdown treatment system |
| CN103552520B (en) * | 2013-11-11 | 2015-08-05 | 吉林大学 | Hybrid vehicle acceleration pedal signal processing method |
-
2016
- 2016-05-16 CN CN201610322885.6A patent/CN105882649B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN105882649A (en) | 2016-08-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105882649B (en) | A kind of hybrid vehicle method for diagnosing faults | |
| CN110018425B (en) | Power battery fault diagnosis method and system | |
| CN106353690B (en) | Utilize the method for Petri network diagnosis lithium battery failure | |
| CN114559819B (en) | Electric automobile battery safety early warning method based on signal processing | |
| CN102393733B (en) | Failure diagnosis method, fault diagnosis instrument and system thereof, new energy automobile | |
| CN113064939B (en) | New energy vehicle three-electric system safety feature database construction method | |
| CN109061495A (en) | A kind of hybrid energy-storing battery failure diagnostic method | |
| RU2256091C2 (en) | Method and device for internal combustion engine ignition | |
| CN111483350B (en) | Balance and micro-short circuit detection method for power battery pack | |
| CN112345955B (en) | Multi-fault online diagnosis method and system for power battery | |
| CN105259458B (en) | Verify battery bag management system insulating monitoring precision and earth leakage protective apparatus and method | |
| CN115327386B (en) | Multi-fault diagnosis method for battery pack based on electrothermal coupling model | |
| CN109615273B (en) | Method and system for evaluating state of electric vehicle charging facility | |
| CN112098850A (en) | Lithium ion battery voltage fault diagnosis method and system based on SDO algorithm | |
| CN109878373A (en) | Knee cloud platform interconnects charging method and system | |
| CN110341682A (en) | Method for diagnosing faults and device for electric vehicle brake switch | |
| CN113687234A (en) | Battery abnormality recognition method, apparatus, device, medium, and program product | |
| Chen et al. | Multifault diagnosis of Li-ion battery pack based on hybrid system | |
| CN103399569B (en) | For judging the miniature diagnostic device of automobile and the determination methods thereof of motoring condition | |
| CN111707943A (en) | Battery simulation-based electric vehicle charging fault early warning method and system | |
| Wang et al. | Research on interactive multi‐model fault diagnosis method of Li‐ion battery based on noise suppression | |
| CN112333007B (en) | Control method and test method for CAN network diagnosis | |
| CN108711643A (en) | Integrated form BMS+GSM all-in-one machines | |
| CN206430913U (en) | A kind of design verification test system of brake energy recovering system | |
| CN110275068A (en) | A kind of capacitor capacitance detection method and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Zeng Xiaohua Inventor after: Li Guanghan Inventor after: Song Meijie Inventor after: Peng Yujun Inventor after: Yang Nannan Inventor after: Feng Tao Inventor after: Chen Qinqin Inventor after: Wang Guangyi Inventor after: Zhang Junkai Inventor before: Zeng Xiaohua Inventor before: Li Guanghan Inventor before: Song Dafeng Inventor before: Peng Yujun Inventor before: Yang Nannan Inventor before: Feng Tao Inventor before: Chen Qinqin Inventor before: Wang Guangyi Inventor before: Zhang Junkai |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180126 Termination date: 20200516 |