Disclosure of Invention
The invention aims to provide a diagnosis method for considering both the inverter current sensor fault and the power tube open-circuit fault. The diagnosis method provided by the invention can realize the simultaneous diagnosis of the current sensor fault and the power tube open circuit fault, improves the comprehensiveness of fault diagnosis, and has the characteristics of reliability, rapidness and practicability.
The invention provides the following technical scheme:
the current sensor fault refers to the fault that the sensor completely fails.
A diagnosis method for considering both inverter current sensor faults and power tube open-circuit faults comprises the following steps:
(1) obtaining signals for diagnosis from a control system of the inverter, wherein the signals comprise direct current bus voltage at an input end of the inverter, duty ratio of a power switch, current of an output filter inductor and output voltage;
(2) calculating to obtain expected values of line voltages between bridge arms according to the direct-current bus voltage at the input end of the inverter and the duty ratio of the power switch, and calculating to obtain actual values of the line voltages between the bridge arms according to the inductive current and the output voltage of the inverter;
(3) calculating to obtain a deviation value of the line voltage between the bridge arms as a first diagnosis variable according to the expected value of the line voltage between the bridge arms and the actual value of the line voltage between the bridge arms;
(4) calculating a deviation value of the sum of the inductive currents as a second diagnosis variable according to the expected value of the sum of the inductive currents and the inductive current value obtained by sampling of the current sensor;
(5) setting a threshold value of a first diagnostic variable according to the sampling error, the parameter error, the dead zone and the delay time;
(6) setting a threshold value of a second diagnostic variable according to the sampling error of the inductive current;
(7) obtaining the polarity of the first diagnostic variable by comparing the first diagnostic variable obtained in step (3) with the threshold value of the first diagnostic variable obtained in step (5); obtaining the polarity of the second diagnostic variable by comparing the second diagnostic variable obtained in step (4) with the threshold value of the first diagnostic variable obtained in step (6); comprehensively diagnosing the faults of the inverter current sensor and the open-circuit faults of the power tube according to the fault diagnosis table;
(8) if the current sensor of a certain phase is diagnosed to have a fault, a reconstructed value of the current of the fault phase is obtained through the current calculation of the normal phase, and fault tolerance of the current sensor is carried out.
In the step (2), the expected value of the line voltage between the bridge arms
The calculation method comprises the following steps:
wherein [ n ]]Value, V, representing the nth sampling instantdcIs the inverse ofDC bus voltage at the input of the inverter, dkAnd diThe duty cycles of the power switches of the inverter k-phase and i-phase, respectively, k, i ≠ a, b, c, and k ≠ i, ki ═ ab, bc, ca.
In the step (2), the actual value v of the line voltage between the bridge armski[n]The calculation method comprises the following steps:
wherein L iskAnd LiOutput filter inductors i of the inverter phases k and i, respectivelykAnd iiThe inductive currents (i.e. output currents), v, of the k-phase and i-phase of the inverter, respectivelykAnd viOutput voltages of k-phase and i-phase of the inverter, TsIs the sampling period.
In step (3), said first diagnostic variable Δ vki[n]The calculation method comprises the following steps:
in step (4), the second diagnostic variable Δ is[n]The calculation method comprises the following steps:
wherein i
kIs the value of the inductance current sampled by the current sensor,
is the sum of the inductance current expected value.
In step (5), the threshold value v of the first diagnostic variableth,ki[n]The setting method comprises the following steps:
(5-1) calculating the maximum calculation error sigmav of the actual values of the line voltages between the arms caused by the sampling error and the parameter errorki[n]:
Wherein, σ vk、σviAnd σ ik、σiiAre respectively the output voltage vk、viAnd the inductor current ik、iiMaximum sampling error of σ Lk、σLiRespectively an output filter inductor Lk、LiThe maximum deviation of (d);
(5-2) calculating the maximum calculation error of the expected value of the line voltage between the bridge arms caused by the sampling error
Wherein, σ VdcFor dc bus voltage V at the input of the inverterdcMaximum sampling error of;
(5-3) calculating the error sigma of the expected value of the line voltage between the bridge arms caused by dead zone and delay timeddAnd σdl:
Wherein, tddIs the dead time, tdlIs the delay time;
(5-4) setting a threshold value v of the first diagnostic variable based on the above calculation errorth,ki[n]Is composed of
α ≧ 1, α is determined based on the sampling error of the system under normal operating conditions.
In step (6), the threshold value i of the second diagnostic variable isthThe setting method comprises the following steps:
ith=β·σis
wherein β is more than or equal to 1, sigma isIs the maximum calculated error of the actual value of the sum of the inductor currents due to the sampling error, β is determined based on the sampling error of the system under normal operating conditions.
In step (7), the polarity Δ v of the first diagnostic variableki_pol[n]The calculation method comprises the following steps:
in step (7), the polarity Δ i of the second diagnostic variables_pol[n]Is calculated by
In step (8), a reconstructed value i of the current of the failed phase is calculated from the current of the normal phasek *[n]The method comprises
Wherein ii[n]And ir[n]The inductor current sample values for normal phase i and phase r, respectively.
The diagnosis method provided by the invention realizes the function of simultaneously diagnosing the inverter current sensor fault and the power tube open circuit fault based on two diagnosis variables, and is a comprehensive diagnosis method. These two diagnostic variables are: the deviation value of the line voltage between the bridge arms (first diagnostic variable) and the deviation value of the sum of the inductor currents (second diagnostic variable). The polarity of the diagnosis variable is obtained by comparing the diagnosis variable with the threshold value of the diagnosis variable, and then the inverter current sensor fault and the power tube open-circuit fault are comprehensively diagnosed according to the fault diagnosis table.
Compared with the prior art, the invention has the following advantages:
1) the method can diagnose the fault of the current sensor of the inverter and the open-circuit fault of the power tube simultaneously, and integrate the fault diagnosis of different types of devices into a unified diagnosis method, thereby improving the comprehensiveness of the fault diagnosis, avoiding the problem of misdiagnosis of another type of device caused by the fault of one type of device, and improving the reliability of the fault diagnosis;
2) the diagnosis speed is high, the current sensor fault and the power tube open-circuit fault can be quickly diagnosed, and the diagnosis speed can reach one sampling period;
3) the diagnosis method has strong robustness, the diagnosis threshold value contains all diagnosis calculation errors, and misdiagnosis caused by noise interference, sampling errors, element parameter errors, load shedding, dead zones and delay time during normal work can be avoided;
4) the sampling frequency required by diagnosis is low, the calculation amount is small, the burden on software is small, and the diagnosis method has greater advantages in the high-frequency application field such as an inverter system based on SiC and GaN devices;
5) the diagnosis process is irrelevant to the control algorithm, and the diagnosis method is suitable for inverter systems adopting different control methods;
6) the diagnosis method is irrelevant to the working mode of the power converter, and the method is suitable for both an inversion mode and a rectification mode;
7) the signals required by diagnosis can be obtained from the control system of the inverter, an additional hardware circuit is not required to be built, the calculation is simple, the method can be easily applied to inverter systems with different control methods and inverters with energy bidirectional transmission, the cost and the system complexity are prevented from being further increased, and the method has stronger practicability.
Detailed Description
In order to more specifically describe the present invention, the diagnostic method of the present invention will be described in detail below with reference to the accompanying drawings and embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
According to the circuit schematic diagram of the diagnosis method in fig. 1, in the present embodiment, the diagnosis method for considering both the inverter current sensor fault and the power tube open circuit fault in the grid-connected three-phase inverter system is shown in fig. 2, and includes the following steps:
s1, obtaining a signal for diagnosis from a control system of the inverter.
In a three-phase grid-connected inverter, signals for fault diagnosis include: DC bus voltage V of inverter input enddcDuty cycle d of the power switchkThe current of the output filter inductor (i.e. the output line current) ikOutput voltage (i.e. grid side voltage) vk。
S2, calculating to obtain expected values of line voltages between bridge arms according to the direct-current bus voltage at the input end of the inverter and the duty ratio of the power switch, and calculating to obtain actual values of the line voltages between the bridge arms according to the inductive current (namely output line current) and the output voltage (namely network side phase voltage) of the inverter.
Actual value v of line voltage between bridge armski[n]The calculation method of (k, i ≠ a, b, c, and k ≠ i, as follows):
wherein L iskAnd LiThe output filter inductances of the inverter phases k and i, respectively (k, i ═ a, b, c, and k ≠ i, hereinafter the same), ikAnd iiThe inductive currents (i.e. output currents), v, of the k-phase and i-phase of the inverter, respectivelykAnd viOutput voltages of k-phase and i-phase of the inverter, TsFor a sampling period, can be etcIn the switching period of the power tube, [ n ]]The value at the nth sampling time is shown below.
For example, the actual value v of the line voltage between the arms of the two phases a and bab[n]The calculation method comprises the following steps:
expected value of line voltage between bridge arms
Comprises the following steps:
wherein, VdcIs the DC bus voltage at the inverter input, dkAnd diThe duty cycles of the power switches of the k-phase and i-phase of the inverter, respectively.
For example, expected line voltage values between bridge arms of a and b phases
Comprises the following steps:
and S3, calculating to obtain a deviation value of the line voltage between the bridge arms as a first diagnosis variable according to the expected value of the line voltage between the bridge arms and the actual value of the line voltage between the bridge arms.
Line voltage deviation value (i.e., first diagnostic variable) between arms Δ νki[n]Comprises the following steps:
for example, the line voltage deviation value (i.e., the first diagnostic variable) Δ ν between the arms for the two phases a, bab[n]Comprises the following steps:
and S4, calculating a deviation value of the sum of the inductive currents as a second diagnosis variable according to the expected value of the sum of the inductive currents and the inductive current value obtained by sampling of the current sensor.
Deviation value (i.e., second diagnostic variable) Δ i of the sum of inductor currentss[n]Is composed of
Wherein i
kIs the value of the inductance current sampled by the current sensor,
is the sum of the inductance current expected value. The inverter in the present embodiment is a three-phase three-wire inverter, so
Equal to 0.
And S5, setting a threshold value of a line voltage deviation value (namely, a first diagnosis variable) between bridge arms according to the sampling error, the parameter error, the dead zone and the delay time.
The specific process of setting the threshold of the inter-bridge-arm line voltage deviation value (i.e., the first diagnostic variable) is as follows:
s501, calculating the maximum calculation error sigma v of the actual value of the line voltage between the bridge arms caused by the sampling error and the parameter errorki[n]:
Wherein, σ vk、σviAnd σ ik、σiiAre respectively the output voltage vk、viAnd the inductor current ik、iiMaximum sampling error of σ Lk、σLiAre respectively a filter inductance Lk、LiThe maximum deviation of (c).
For example, the maximum calculation error σ v of the actual values of the line voltages between the arms of the two phases a and bab[n]Comprises the following steps:
s502, calculating the maximum calculation error of expected values of line voltages between bridge arms caused by sampling errors
Wherein, σ VdcFor dc bus voltage V at the input of the inverterdcThe maximum sampling error.
For example, the maximum calculation error of expected values of line voltages between arms of a and b phases
Is composed of
S503, calculating the error sigma of the expected value of the line voltage between the bridge arms caused by dead zone and delay timeddAnd σdl:
Wherein, tddIs the dead time, tdlIs the delay time.
S504, setting a threshold value v of a line voltage deviation value (first diagnosis variable) between bridge arms based on the calculated errorth,ki[n]Is composed of
α is more than or equal to 1, in the embodiment, α is 1.
For example, the threshold v of the line voltage deviation value (first diagnostic variable) between the arms of the two phases a and bth,ab[n]Is composed of
And S6, setting a threshold value of a deviation value (namely, a second diagnosis variable) of the sum of the inductive currents according to the sampling error of the inductive currents.
The specific process of setting the threshold value of the deviation value (i.e., the second diagnostic variable) of the sum of the inductor currents is:
s601, calculating the calculation error of the actual value of the sum of the inductive currents caused by the sampling error:
σis=σia+σib+σic
wherein, σ ia,σib,σicIs an inductive current ia,ib,icThe maximum sampling error.
S602, setting a threshold value i of a deviation value (i.e., a second diagnostic variable) of the sum of the inductor currents based on the above calculation errorthIs composed of
ith=β·σis
β is more than or equal to 1, in the embodiment, β is 7.
S7, obtaining the polarity of the first diagnostic variable by comparing the first diagnostic variable with the threshold value of the first diagnostic variable; obtaining a polarity of the second diagnostic variable by comparing the second diagnostic variable to a threshold of the second diagnostic variable; and comprehensively diagnosing the inverter current sensor fault and the power tube open-circuit fault according to the fault diagnosis table in the table 1.
Wherein the polarity av of the inter-leg line voltage deviation value (i.e. the first diagnostic variable) is obtainedki_pol[n]The method comprises
For example, the polarity Δ ν of the inter-arm line voltage deviation value (i.e., the first diagnostic variable) for the two phases a, b is obtainedab_pol[n]The method comprises the following steps: when Δ vab[n]≤-Δvth,ab[n]When, let Δ vab_pol[n]-1; when- Δ vth,ab[n]<Δvab[n]<Δvth,ab[n]When, let Δ vab_pol[n]0; when Δ vab[n]≥Δvth,ab[n]When, let Δ vab_pol[n]=1。
The polarity Δ i of the deviation value (i.e., the second diagnostic variable) of the sum of the inductor currents is obtaineds_pol[n]The method comprises
Based on the polarity of the diagnostic variable, a fault diagnosis table for realizing comprehensive diagnosis of both the inverter current sensor fault and the power tube open circuit fault is shown in table 1.
Table 1 fault diagnosis table for comprehensively diagnosing inverter current sensor fault and power tube open circuit fault
According to the fault diagnosis table, the current sensor fault and the power tube open-circuit fault of the three-phase inverter can be comprehensively diagnosed.
For example, if Δ v is calculated at a certain timeab_pol[n]=-1、Δvbc_pol[n]=0、Δvca_pol[n]=1、Δis_pol[n]The current sensor S of phase a is diagnosed as 1aFailure; if a certain time is calculated to have delta vab_pol[n]=-1、Δvbc_pol[n]=0、Δvca_pol[n]=1、Δis_pol[n]When the value is equal to 0, the upper arm power tube T of the a phase is diagnosedapAn open circuit fault occurs.
And S8, if the current sensor of a certain phase is diagnosed to have a fault, calculating a reconstruction value of the current of the fault phase through the current of the normal phase, and carrying out fault tolerance on the current sensor.
When a fault is diagnosed in the k-phase current sensor, a reconstructed value i of the current of the fault phase is calculated from the current of the normal phasek *[n]Is composed of
Wherein ii[n]And ir[n]The inductor current sample values for normal phase i and phase r, respectively.
For example, if a failure is diagnosed in the current sensor of the a-phase, the normal phase, i, will be selected from the currents i of the b and c phasesb[n]And ic[n]Calculating the current reconstruction value i of the fault phase aa *[n]:
To more clearly describe this example, fig. 3 and 4 show the experimental results under this example. The parameters used in this experiment are shown in table 2.
Table 2 examples experimental parameters
As shown in fig. 3, at t1The a-phase current sensor S is triggered at any momentaAnd (4) failure. Before the fault is triggered, the polarity of the diagnostic variable is maintained at Δ vab_pol[n]=0、Δvbc_pol[n]=0、Δvca_pol[n]=0、Δis_pol[n]The fault diagnosis flag is not triggered 0. Current sensor SaAfter the fault occurs, the polarity of the diagnostic variable changes rapidly. When Δ v is observedab_pol[n]=-1、Δvbc_pol[n]=0、Δvca_pol[n]=1、Δis_pol[n]When 1, the current sensor SaIs diagnosed as failing and the current sensor fault diagnosis flag is triggered. The diagnostic time is one sampling period (0.1 ms).
As shown in fig. 4, at t1A-phase upper bridge arm power tube T is triggered at any momentapOpen circuit failure. Before the fault is triggered, the polarity of the diagnostic variable is maintained at Δ vab_pol[n]=0、Δvbc_pol[n]=0、Δvca_pol[n]=0、Δis_pol[n]The fault diagnosis flag is not triggered 0. Power tube TapAfter the fault occurs, the polarity of the diagnostic variable changes rapidly. When Δ v is observedab_pol[n]=-1、Δvbc_pol[n]=0、Δvca_pol[n]=1、Δis_pol[n]When equal to 0, the power tube TapThe power tube fault diagnosis mark is triggered when the open-circuit fault is diagnosed. The diagnostic time is one sampling period (0.1 ms).
The results prove the comprehensiveness, accuracy and rapidity of the diagnosis method provided by the invention.
The embodiments described above are intended to facilitate one of ordinary skill in the art in understanding and using the invention. It will be readily apparent to those skilled in the art that various modifications to the above-described embodiments may be made, and the generic principles defined herein may be applied to other embodiments without the use of inventive faculty. Therefore, the present invention is not limited to the above embodiments, and those skilled in the art should make improvements and modifications to the present invention based on the disclosure of the present invention within the protection scope of the present invention.