Disclosure of Invention
The invention aims to overcome the defects and shortcomings of the prior art, and provides a single-tube open-circuit fault diagnosis method of an open-winding electric drive system based on predictive control.
In order to achieve the purpose, the technical scheme provided by the invention is as follows: the single-tube open-circuit fault diagnosis method of the open-winding electric drive system based on predictive control comprises a direct-current power supply VdcThe three-phase 3H bridge inverter (double parallel inverters) and the three-phase stator winding of the open winding motor; the DC power supply VdcIs connected in parallel with a three-phase 3H bridge inverter; in the three-phase 3H bridge inverter, each phase comprises two bridge arms, each bridge arm comprises two switching tubes with anti-parallel diodes, namely the three-phase 3H bridge inverter has 12 switching tubes, and the switching tubes are respectively marked as SnijWhere n represents the phase sequence, n ═ a, b, c, i represents the bridge arm serial number of each phase, i ═ 1,2, j represents the switching order of each bridge armThe number j is 1,2, the arms of the three-phase 3H bridge inverter are connected in parallel, the middle points of the A-phase arm are respectively marked as a1 and a2, the middle points of the B-phase arm are respectively marked as B1 and B2, and the middle points of the C-phase arm are respectively marked as C1 and C2; the three-phase stator winding of the open-winding motor is respectively connected with the midpoints of the bridge arms, namely two ends of the A-phase stator winding are respectively connected with the midpoint a1 and the midpoint a2 of the bridge arms, two ends of the B-phase stator winding are respectively connected with the midpoint B1 and the midpoint B2 of the bridge arms, and two ends of the C-phase stator winding are respectively connected with the midpoint C1 and the midpoint C2 of the bridge arms;
the single-tube open-circuit fault diagnosis method of the open-winding electric drive system based on the predictive control comprises the following steps of:
step 1: sampling phase voltages of three-phase stator windings of the open-winding motor in real time;
step 2: the estimated voltage is differed with the actually measured voltage at the corresponding moment to obtain the error voltage epsilon of three phasesa、εbAnd εcRespectively, with a given threshold h1And comparing, and preliminarily judging whether the fault occurs, wherein the following two conditions exist:
in the first case, | εa|≤h1,|εb|≤h1And | εc|≤h1Judging that no fault occurs, and returning to the step 1;
in the second case, when one of the following six states occurs, the fault is preliminarily determined, and the fault state variable F is set11, preliminarily judging the fault range of the switching tube:
state 1, epsilona>h1Switching tube Sa11Or Sa22An open circuit fault occurs;
state 2, epsilona<-h1Switching tube Sa12Or Sa21An open circuit fault occurs;
state 3, epsilonb>h1Switching tube Sb11Or Sb22An open circuit fault occurs;
state 4, epsilonb<-h1Switching tube Sb12Or Sb21Occurrence of open circuit failure
State 5,. epsilonc>h1Switching tube Sc11Or Sc22Occurrence of open circuit failure;
State 6, εc<-h1Switching tube Sc12Or Sc21An open circuit fault occurs;
and step 3: fault state variable F1After setting to 1, the respective states in the second case of step 2 are processed accordingly, i.e. the error voltage is integrated over a defined period of time and compared with a given threshold h2Comparing to further confirm whether the fault occurs; if the fault is judged not to occur, the fault state variable F is set1 Setting 0, and returning to the step 1; if the fault is judged to have occurred, the fault state variable F is set2 Setting 1 and executing the step 4;
and 4, step 4: fault state variable F2After setting 1, determining a faulty switch pair, and still further positioning a specific faulty switch tube; for each state in the step 2, constructing a corresponding diagnosis function J by using the relation between the fault phase bridge arm switch state and the error voltagekTo locate a specific faulty switching tube, where k represents each state in step 2, and k is 1,2,3,4,5, 6;
and 5: and finishing the fault diagnosis.
In step 2, a threshold h is given1And setting the voltage to be 50, and obtaining the estimated voltage of each phase from the predicted switching state by utilizing the characteristic that the switching state of the model predictive control is determined and unchanged in each control period.
In step 3, a threshold h is given2Set to 0.05, the corresponding processing for each state in the second case of step 2 is specifically as follows:
for
state 1 or
state 2, for |. epsilon
aL performing integration in time, when the integration time length T reaches 5 control periods, namely 5T
sAnd is and
less than a given threshold h
2If no fault occurs, the fault state variable F is set
1Resetting 0 and returning to the
step 1; when the integration time length T does not reach 5T
sWhen the temperature of the water is higher than the set temperature,
has been equal to or greater than a given threshold h
2If the fault is determined to have occurred, the fault state variable F is set
21, placing;
for
state 3 or
state 4, for |. epsilon
bL performing integration in time, when the integration time length T reaches 5 control periods, namely 5T
sAnd is and
less than a given threshold h
2If no fault occurs, the fault state variable F is set
1Resetting 0 and returning to the
step 1; when the integration time length T does not reach 5T
sWhen the temperature of the water is higher than the set temperature,
has been equal to or greater than a given threshold h
2If the fault is determined to have occurred, the fault state variable F is set
21, placing;
for state 5 or state 6, for |. epsilon
cL performing integration in time, when the integration time length T reaches 5 control periods, namely 5T
sAnd is and
less than a given threshold h
2If no fault occurs, the fault state variable F is set
1Resetting 0 and returning to the
step 1; when the integration time length T does not reach 5T
sWhen the temperature of the water is higher than the set temperature,
has been equal to or greater than a given threshold h
2If the fault is determined to have occurred, the fault state variable F is set
21, placing.
In step 4, a diagnostic function is constructed by using the relation between the switching states of the bridge arms of the fault phase and the error voltage, and the switching states of the two bridge arms of each phase are respectively set as sn1And sn2Wherein n represents a phase sequence, n is a, b, c, and a switching state of 1 indicates that the upper tube is on and the lower tube is off, and a switching state of 0 indicates that the upper tube is off and the lower tube is on;
the diagnostic function construction for each state in step 2, and the specific diagnostic method is as follows:
for
State 1, a diagnostic function is constructed
If J
1If it is less than 0, S is judged
a11A failure; if J
1If it is greater than 0, S is judged
a22A failure;
for
State 2, a diagnostic function is constructed
If J
2If it is less than 0, S is judged
a12A failure; if J
2If it is greater than 0, S is judged
a21A failure;
for
State 3, a diagnostic function is constructed
If J
3If it is less than 0, S is judged
b11A failure; if J
3If it is greater than 0, S is judged
b22A failure;
for
State 4, a diagnostic function is constructed
If J
4If it is less than 0, S is judged
b12A failure; if J
4If it is greater than 0, S is judged
b21A failure;
for State 5, a diagnostic function is constructed
If J
5If it is less than 0, S is judged
c11A failure; if J
5If it is greater than 0, S is judged
c22A failure;
for State 6, a diagnostic function is constructed
If J
6If it is less than 0, S is judged
c12A failure; if J
6If it is greater than 0, S is judged
c21A failure;
wherein, T is the integration duration, which indicates that the integration process continues until the system executes the fault-tolerant operation mode.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. the invention directly utilizes the characteristic that the switching state of the model predictive control is determined and unchanged in each control period, namely, the stator phase voltage of the open-winding motor is kept unchanged in one control period and can be predicted in advance, so that the method is simple and effective to realize.
2. The voltage quantity is the state variable which can most directly reflect the fault characteristics of the inverter, the invention directly extracts the voltage information as the basis of fault diagnosis and combines the excellent dynamic response capability of model predictive control, thereby realizing fault diagnosis in a shorter time.
3. When the system operation state is suddenly changed, the current detection diagnosis method has the advantages that the current transient state process is long in duration and irregular in waveform, misdiagnosis is easily caused, the voltage can more directly reflect the fault characteristics of the inverter, and the influence of system operation disturbance is not easily caused, so that the method has high reliability.
4. The invention overcomes the unique symmetry and redundancy of the double-inverter driving topology of the open-winding motor and can position a specific fault switch tube.
Detailed Description
The present invention will be further described with reference to the following specific examples.
As shown in FIG. 1, the open-winding electric drive system of the present embodiment includes a DC power supply VdcThe three-phase 3H bridge inverter (double parallel inverters) and the three-phase stator winding of the open winding motor; the DC power supply VdcIs connected in parallel with a three-phase 3H bridge inverter; in the three-phase 3H bridge inverter, each phase comprises two bridge arms, each bridge arm comprises two switching tubes with anti-parallel diodes, namely the three-phase 3H bridge inverter has 12 switching tubes, and the switching tubes are respectively marked as SnijThe neutral points of the bridge arms of the three-phase 3H bridge inverter are respectively marked as a1 and a2, the neutral points of the bridge arms of the phase A are respectively marked as B1 and B2, and the neutral points of the bridge arms of the phase C are respectively marked as C1 and C2; the three-phase stator winding of the open-winding motor is respectively connected with the middle points of the bridge arms, namely two ends of the A-phase stator winding are respectively connected with the middle points a1 and a2 of the bridge arms, two ends of the B-phase stator winding are respectively connected with the middle points B1 and B2 of the bridge arms, and two ends of the C-phase stator winding are respectively connected with the middle points C1 and C2 of the bridge arms.
As shown in fig. 2, the single-tube open-circuit fault diagnosis method for an open-winding electric drive system based on predictive control provided by the present embodiment includes the following steps:
step 1: and sampling phase voltages of three-phase stator windings of the open-winding motor in real time.
Step 2: the output switching signal is predicted and controlled by the model to obtain the estimated voltage of each phase, and the estimated voltage is differenced with the actually measured voltage at the corresponding moment to obtain the error voltage epsilon of three phasesa、εbAnd εcRespectively, with a given threshold h1For comparison, h1Is provided withAt 50, there are two cases:
in the first case, | εa|≤h1,|εb|≤h1And | εc|≤h1Judging that no fault occurs, and returning to the step 1;
in the second case, when one of the following six states occurs, the fault is preliminarily determined, and the fault state variable F is set11, preliminarily judging the fault range of the switching tube:
state 1, epsilona>h1Switching tube Sa11Or Sa22An open circuit fault occurs;
state 2, epsilona<-h1Switching tube Sa12Or Sa21An open circuit fault occurs;
state 3, epsilonb>h1Switching tube Sb11Or Sb22An open circuit fault occurs;
state 4, epsilonb<-h1Switching tube Sb12Or Sb21Occurrence of open circuit failure
State 5,. epsilonc>h1Switching tube Sc11Or Sc22An open circuit fault occurs;
state 6, εc<-h1Switching tube Sc12Or Sc21An open circuit fault occurs.
And step 3: fault state variable F1After setting to 1, the respective states in the second case of step 2 are processed accordingly, with a threshold h being specified2Set to 0.05:
for
state 1 or
state 2, for |. epsilon
aL performing integration in time, when the integration time length T reaches 5 control periods, namely 5T
sAnd is and
less than a given threshold h
2If no fault occurs, the fault state variable F is set
1 Resetting 0 and returning to the
step 1; when the integration time length T does not reach 5T
sWhen the temperature of the water is higher than the set temperature,
has been equal to or greater than a given threshold h
2If the fault is determined to have occurred, the fault state variable F is set
21, placing;
for
state 3 or
state 4, for |. epsilon
bL performing integration in time, when the integration time length T reaches 5 control periods, namely 5T
sAnd is and
less than a given threshold h
2If no fault occurs, the fault state variable F is set
1 Resetting 0 and returning to the
step 1; when the integration time length T does not reach 5T
sWhen the temperature of the water is higher than the set temperature,
has been equal to or greater than a given threshold h
2If the fault is determined to have occurred, the fault state variable F is set
21, placing;
for state 5 or state 6, for |. epsilon
cL performing integration in time, when the integration time length T reaches 5 control periods, namely 5T
sAnd is and
less than a given threshold h
2If no fault occurs, the fault state variable F is set
1 Resetting 0 and returning to the
step 1; when the integration time length T does not reach 5T
sWhen the temperature of the water is higher than the set temperature,
has been equal to or greater than a given threshold h
2If the fault is determined to have occurred, the fault state variable F is set
21, placing.
And 4, step 4: fault state variable F2After setting to 1, the faulty switch pair is determined, but the specific faulty switch tube still needs to be further positioned. Constructing a diagnosis function by using the relation between the switching states of the bridge arms of the fault phase and the error voltage, and setting the switching states of the two bridge arms of each phase as sn1And sn2Where n represents the phase sequence, n ═ a, b, and c, a switch state of 1 indicates top tube on and bottom tube off, and a switch state of 0 indicates top tube offThe tube is disconnected and connected. For each state in step 2, a corresponding diagnostic function J is constructedkWherein k represents each state in step 2, k is 1,2,3,4,5, 6; the specific diagnostic function construction and diagnostic method is as follows:
for
State 1, a diagnostic function is constructed
If J
1If it is less than 0, S is judged
a11A failure; if J
1If it is greater than 0, S is judged
a22A failure;
for
State 2, a diagnostic function is constructed
If J
2If it is less than 0, S is judged
a12A failure; if J
2If it is greater than 0, S is judged
a21A failure;
for
State 3, a diagnostic function is constructed
If J
3If it is less than 0, S is judged
b11A failure; if J
3If it is greater than 0, S is judged
b22A failure;
for
State 4, a diagnostic function is constructed
If J
4If it is less than 0, S is judged
b12A failure; if J
4If it is greater than 0, S is judged
b21A failure;
for State 5, a diagnostic function is constructed
If J
5If it is less than 0, S is judged
c11A failure; if J
5If it is greater than 0, S is judged
c22A failure;
for State 6, a diagnostic function is constructed
If J
6If it is less than 0, it is judgedS
c12A failure; if J
6If it is greater than 0, S is judged
c21And (4) failure.
Wherein, the integration duration T represents that the integration process continues until the system executes the fault-tolerant operation mode.
And 5: and finishing the fault diagnosis.
This embodiment is a pair of switch tubes S
a11Or S
a22Simulink simulation is carried out under the condition of open-circuit fault, and direct-current voltage V
dc300V, control period T
sWas 100. mu.s. FIG. 3 shows a switching tube S
a11Or S
a22Error voltage before and after open circuit fault, fault state variable F
1、
And a fault state variable F
2Waveform diagram of which the threshold h
1Set to 50, threshold h
2Set to 0.05.
FIG. 4 shows a switching tube Sa11Diagnostic function J at open-circuit fault occurrence1Waveform diagram and diagnosis duration diagram, FIG. 5 is a switch tube Sa22Diagnostic function J at open-circuit fault occurrence1Waveform diagram and diagnosis duration diagram, it can be seen that the diagnosis duration of the method of the present invention is 0.26ms, and rapid diagnosis can be realized.
FIG. 6a is a voltage waveform diagram of an error voltage before and after a load disturbance occurs when an open-winding electric drive system operates without a fault, and a rated load torque is suddenly changed from 2 N.m to 4 N.m; FIG. 6b is a voltage waveform diagram of an error voltage before and after a rotational speed disturbance occurs when the open-winding electric drive system operates without a fault, and the rated rotational speed is suddenly changed from 3000r/min to 2000 r/min. As can be seen from the graph, the error voltage is far smaller than the threshold value before and after the disturbance, and the misdiagnosis cannot be caused. Therefore, the diagnosis method is not greatly influenced by system operation disturbance, has good reliability and is worthy of popularization.
The above-mentioned embodiments are merely preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, so that the changes in the shape and principle of the present invention should be covered within the protection scope of the present invention.