CN118641992B - Open-circuit fault diagnosis method for permanent magnet synchronous motor inverter based on current bias - Google Patents

Abstract

The invention discloses a method for diagnosing open-circuit faults of a permanent magnet synchronous motor inverter based on current bias, which only needs basic signals (three-phase current and rotor position) of a closed-loop control system, considers the influence of feedback regulation action of a controller, adopts bias which does not depend on current amplitude, has low sensitivity to load change, judges whether faults occur by utilizing the sum of absolute values of instantaneous bias of three-phase stator currents, and takes the sum of absolute values as auxiliary variables for fault detection, thereby improving the anti-interference performance of the diagnosis method. Therefore, the method has low sensitivity to motor speed change and load change, does not depend on a complex model, has small calculated amount, is easy to realize, has good robustness under the mismatch of different control periods and inductance parameters, can be applied to different operation modes such as electric power generation/braking and the like, and can be applied to double-tube open circuit fault diagnosis in an expanding way.

Description

Open-circuit fault diagnosis method for permanent magnet synchronous motor inverter based on current bias

Technical Field

The invention belongs to the technical field of inverter fault diagnosis, and relates to a permanent magnet synchronous motor inverter open circuit fault diagnosis method based on current bias.

Background Art

Permanent magnet synchronous motors are widely used in high-end equipment such as new energy vehicles, rail transit and ship propulsion due to their high power density, good control performance and high working efficiency. High reliability is crucial for the continuous operation of industrial systems and is also one of the core requirements of motor drive systems. Rapid detection and accurate positioning of faults are key links in ensuring system reliability. Therefore, the development of online monitoring and fault diagnosis technology for the operating status of motor drive systems has very important theoretical and practical significance.

Fault diagnosis methods can be divided into two categories: data-based and signal-based. Data-based methods combine artificial intelligence algorithms to realize automatic classification and identification of fault characteristics. They do not require understanding of system models and working mechanisms, have strong anti-interference capabilities, and have obvious advantages in fault diagnosis of complex systems. Their disadvantages are that they rely on a large amount of historical data, have high requirements for hardware support such as digital processors, and their diagnostic speed is much slower than that of model-based diagnostic methods. Their application in industrial converters is still rarely heard of.

The research on signal-based methods is more extensive. According to the different characteristic values, signal-based methods can be divided into two categories: voltage signals and current signals. The voltage-based method is not affected by the electrical time constant and the system operation status. It can directly reflect the fault characteristics of the inverter and realize fault diagnosis within several switching cycles. It is less affected by factors such as load, speed and control strategy, and has strong rapidity. Its disadvantage is that the voltage method often needs to cooperate with the drive pulse signal to realize fault diagnosis. In the actual system, many factors such as measurement delay, switching delay, dead time, system noise, and non-ideal characteristics of the device may cause false fault signals to appear, and the requirements for circuit accuracy and noise resistance are extremely strict. In addition, the acquisition of voltage signals requires the installation of additional monitoring equipment, which requires space reservation and operation complexity of the system structure. This disadvantage is more obvious in multi-level inverters.

Depending on whether a driving signal is required, the current signal-based methods can be divided into two categories. The current signal method combined with the driving signal is almost comparable to the voltage signal-based method in terms of rapidity and robustness, and does not require additional sensors, but this comes at the cost of increasing the complexity of the algorithm to a certain extent; since the driving signal is required for fault diagnosis, this method is highly sensitive to dead time, switching delay, system noise and measurement error, and sufficient attention needs to be paid to ensure diagnostic accuracy. In addition, it should be noted that it is difficult to directly obtain the driving signal in some industrial application scenarios, which will limit the engineering application of this method.

The diagnostic speed of the current signal method without the drive signal is not as fast as that of the voltage signal-based method and the current signal method combined with the drive signal. However, this type of method only requires the three-phase current, rotor position and other signals collected by the controller, and has a wider application scenario. It is less sensitive to switch delays, dead zone effects, noise interference, etc.; this type of current method is more dependent on the familiarity with the research object, and how to select a suitable diagnostic threshold is another focus of this type of method. The current change under the closed-loop system condition is affected by the feedback regulation of the controller. This issue has received less attention. At present, a simplified model that ignores feedback regulation is often used to analyze fault currents. Although this reduces the amount of calculation, the simplified model will cause the loss of fault characteristics. In addition, there are differences in the current distortion amplitude under different control parameters. The fault characteristics obtained by the simplified model are likely to lead to insufficient generalization of the diagnostic algorithm under different control parameters, and its accuracy is easily affected by industrial scenarios.

Summary of the invention

In view of the above, the present invention provides a permanent magnet synchronous motor inverter open circuit fault diagnosis method based on current bias. The method has low sensitivity to motor speed changes and load changes, has good robustness under different control cycles and inductance parameter mismatches, can be applied to different operating modes such as electric/generating/braking, and can be extended to dual-tube open circuit fault diagnosis.

A method for diagnosing open-circuit faults of a permanent magnet synchronous motor inverter based on current bias comprises the following steps:

(1) Construct a stator current observation model based on the rotor position angle and q-axis reference current;

(2) using the stator current observation model to calculate the stator current residual after the inverter open circuit fault;

(3) quantifying the stator current residual using an error polarity function;

(4) calculating the stator current bias according to the error polarity function;

(5) Constructing an error discrimination function according to the stator current bias to diagnose the inverter open circuit fault.

Furthermore, the stator current observation model expression in step (1) is as follows:

Where: i _x represents the x-phase stator current of the permanent magnet synchronous motor, x = a, b or c, θ _e is the rotor position angle of the permanent magnet synchronous motor, i _qref is the q-axis reference current of the controller, Indicates the initial phase of the x-phase stator current I _m represents the current amplitude.

Furthermore, in step (2), the stator current residual after the inverter open circuit fault is calculated by the following formula:

Wherein: ε _xf represents the x-phase stator current residual after the inverter open-circuit fault, i _xf represents the x-phase stator current of the permanent magnet synchronous motor after the inverter open-circuit fault.

Furthermore, the error polarity function expression in step (3) is as follows:

Where: Λ _x represents the error polarity function of the x phase, and r ₁ is the given judgment threshold.

Furthermore, the specific implementation of step (4) is as follows: first, the error polarity function Λ _x is sampled multiple times in a given rotor position angle interval θ ₁ to θ ₂ , and its average value is defined as the stator current bias degree, which is used to characterize the distortion characteristics of the stator current. The specific expression is as follows:

Where: χ _x represents the x-phase stator current bias.

Furthermore, in step (5), the error discrimination function is constructed by the following relational expression:

F _ab =sgn(χ _a -χ _b )

F _bc =sgn(χ _b -χ _c )

_Fca ＝sgn( _χc - _χa )

Wherein: F _ab represents the error discrimination function between the stator current bias degrees of the two phases ab, F _bc represents the error discrimination function between the stator current bias degrees of the two phases bc, F _ca represents the error discrimination function between the stator current bias degrees of the two phases ca, and sgn() represents the sign function.

Furthermore, the standard for diagnosing the inverter open circuit fault in step (5) is as follows:

If F _ab = -1, F _bc = F _ca = 1, it is determined that the inverter a phase upper tube is open circuit fault;

If F _ab = 1, F _bc = F _ca = -1, it is determined that the inverter a phase lower tube is open circuit fault;

If F _bc = -1, F _ab = F _ca = 1, it is determined that the inverter b phase upper tube is open circuit fault;

If F _bc = 1, F _ab = F _ca = -1, it is determined that the inverter b phase lower tube is open circuit fault;

If _Fca = -1, _Fab = _Fbc = 1, it is determined that the upper tube of phase c of the inverter is open circuit fault;

If _Fca = 1, _Fab = _Fbc = -1, it is determined that the lower tube of phase c of the inverter is open circuit fault;

If _Fab = _Fbc = _Fca = 0, it is determined that there is no open circuit fault in the inverter.

Noise interference during normal operation will produce false fault signals. In order to solve the problem of false fault signal interference, preferably, the specific implementation method of diagnosing the inverter open circuit fault in step (5) is: first, the sum of the absolute values of the three-phase stator current residuals after the fault is used as the auxiliary variable ε _m for fault detection, that is, ε _m =|ε _af |+|ε _bf |+|ε _cf |; and then the fault detection trigger signal F _g is used to quantify the auxiliary variable ε _m , that is:

Where: tr ₀ is the given fault detection threshold;

Then, a unified fault location mark F is constructed according to the fault detection trigger signal _Fg and the error discrimination function, that is, F= _Fg ( _4Fab + _2Fbc + _Fca );

Finally, the inverter open circuit fault is diagnosed by the following criteria:

If F _g = 1, F = -1, it is determined that the inverter a phase upper tube is open circuit fault;

If F _g = 1, F = 1, it is determined that the inverter a phase lower tube is open circuit fault;

If F _g = 1, F = 3, it is determined that the inverter b phase upper tube is open circuit fault;

If F _g = 1, F = -3, it is determined that the inverter b phase lower tube is open circuit fault;

If F _g = 1, F = 5, it is determined that the inverter c phase upper tube is open circuit fault;

If F _g = 1, F = -5, it is determined that the inverter c phase lower tube is open circuit fault;

If F _g =F=0, it is determined that there is no open circuit fault in the inverter.

Based on the above technical solution, the present invention has the following beneficial technical effects:

1. The method of the present invention takes into account the influence of the feedback regulation of the controller and has strong generalization ability under different control parameters.

2. The method of the present invention only requires basic signals (three-phase current and rotor position) of the closed-loop control system, has strong versatility and a wider range of applications.

3. The method of the present invention does not rely on complex models, has a small amount of calculation and is easy to implement.

4. The method of the present invention uses the fault current bias as a characteristic quantity, is independent of the current amplitude, and has low sensitivity to load changes. In addition, the standardization process and the dual standard ensure the robustness of the present invention under load change conditions.

5. The method of the present invention uses the sum of the absolute values of the instantaneous bias of the three-phase stator current to determine whether a fault occurs, and uses it as an auxiliary variable for fault detection to improve the anti-interference ability of the diagnostic method.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a flowchart of a method for diagnosing an open-circuit fault of a permanent magnet synchronous motor inverter according to the present invention.

FIG2 is a schematic diagram of the inverter topology of a permanent magnet synchronous motor.

3 to 5 are schematic diagrams of experimental waveforms before and after an open-circuit fault occurs in the permanent magnet synchronous motor inverter switch device _Sa1 .

DETAILED DESCRIPTION

In order to describe the present invention more specifically, the technical solution of the present invention is described in detail below in conjunction with the accompanying drawings and specific implementation methods.

As shown in FIG1 , the current bias-based permanent magnet synchronous motor inverter open circuit fault diagnosis method of the present invention comprises the following steps:

Step 1: Construct the stator current observation model based on the rotor position angle and the q-axis reference current of the controller.

Where: i _x represents the x-phase stator current of the permanent magnet synchronous motor, x = a, b or c, θ _e is the rotor position angle of the permanent magnet synchronous motor, i _qref is the q-axis reference current of the controller, Indicates the initial phase of the x-phase stator current I _m represents the current amplitude.

Step 2: Obtain the stator current residual through stator current observation model calculation.

Wherein: ε _xf represents the x-phase stator current residual after the inverter open-circuit fault, i _xf represents the x-phase stator current of the permanent magnet synchronous motor after the inverter open-circuit fault.

Step 3: The fault characteristic lies in the difference in the current error polarity. The stator current residual is quantified using the error polarity function. Specifically:

Where: Λ _x represents the error polarity function of the x phase, and r ₁ is the given judgment threshold.

Step 4: Due to the existence of noise interference and measurement errors, it is difficult to ensure that Λ _x is strictly 0 in a healthy state. Therefore, Λ _x is sampled multiple times in the region (θ ₁ , θ ₂ ), and its average value χ _x is defined as the stator current bias degree, which is used to characterize the distortion characteristics of the fault current. The expression of χ _x is:

The inverter topology of the permanent magnet synchronous motor is shown in Figure 2, where the fault characteristics of different switch devices after opening the circuit (OC) (the core basis for locating the faulty device) are shown in Table 1:

Table 1

Step 5: According to the fault characteristics obtained in step 4, construct an error discrimination function based on the fault current bias degree:

F _ab =sgn(χ _a -χ _b )

F _bc =sgn(χ _b -χ _c )

_Fca ＝sgn( _χc - _χa )

Wherein: F _ab represents the error discrimination function between the stator current bias degrees of the two phases ab, F _bc represents the error discrimination function between the stator current bias degrees of the two phases bc, F _ca represents the error discrimination function between the stator current bias degrees of the two phases ca, and sgn() represents the sign function.

The _Fab , _Fbc and _Fca corresponding to different switching device faults are shown in Table 2:

Table 2

Step 6: Noise interference during normal operation will produce false fault signals. To solve the problem of false fault signal interference, the absolute value sum of the three-phase stator current residual ε _xf ε _m is used as the auxiliary variable for fault detection, that is:

ε _m =|ε _af |+|ε _bf |+|ε _cf |

The fault detection trigger signal _Fg is:

Where: tr ₀ is the given fault detection threshold.

Step 7: Construct a unified fault location function and set the fault location flag F to:

F＝ _Fg ( _4Fab + _2Fbc + _Fca )

The location marks F corresponding to different switch device faults are shown in Table 3:

Table 3

Verification example:

In order to verify the effectiveness and correctness of the permanent magnet synchronous motor inverter open circuit fault diagnosis method of the present invention, we conducted an experimental verification. The inverter open circuit fault test platform and surface mounted permanent magnet synchronous motor parameters used in the experiment are shown in Table 4:

Table 4

We take the open circuit fault of the power device S _a1 of the two-level converter as an example for diagnosis. Figure 3 shows the experimental results after S _a1 open circuit occurs during the acceleration process when the motor speed n increases from 500r/min to 800r/min. Figure 4 shows the experimental results after S _a1 open circuit occurs during the loading process when the load torque T _e increases from 2Nm to 6Nm. Figure 5 shows the experimental results after S _a1 open circuit fault occurs under different PI control parameters, where the PI control parameters corresponding to subgraph (a) are K _p = 21.15, _Ki = 2795; the PI control parameters corresponding to subgraph (b) are K _p = 28.2, _Ki = 3726.

The above experimental results show that the calculation results of the positioning mark F are consistent with Table 3, which verifies the correctness of the theoretical analysis; therefore, the permanent magnet synchronous motor inverter open circuit fault diagnosis method of the present invention has good diagnostic effect for faults under different control parameters within a wide speed range, has low sensitivity to load changes, and is suitable for different working modes.

The above description of the embodiments is to facilitate the understanding and application of the present invention by those skilled in the art. It is obvious that those skilled in the art can easily make various modifications to the above embodiments and apply the general principles described herein to other embodiments without creative work. Therefore, the present invention is not limited to the above embodiments. Improvements and modifications made by those skilled in the art to the present invention based on the disclosure of the present invention should be within the protection scope of the present invention.

Claims (2)

1. A permanent magnet synchronous motor inverter open circuit fault diagnosis method based on current bias degree comprises the following steps:

(1) A stator current observation model is constructed according to the rotor position angle and the q-axis reference current, and the expression is as follows:

Wherein: i _x denotes the x-phase stator current of the permanent magnet synchronous motor, x=a, b or c, θ _e is the rotor position angle of the permanent magnet synchronous motor, i _qref is the q-axis reference current of the controller, Representing the initial phase of the x-phase stator current, I _m representing the current magnitude;

(2) Calculating a stator current residual error after an open-circuit fault of the inverter by using the stator current observation model through the following formula;

Wherein: epsilon _xf represents the x-phase stator current residual error after the open-circuit fault of the inverter, and i _xf represents the x-phase stator current of the permanent magnet synchronous motor after the open-circuit fault of the inverter;

(3) And quantizing the stator current residual by adopting an error polarity function, wherein the error polarity function is expressed as follows:

Wherein: Λ _x represents the error polarity function of the x-phase, r ₁ is a given judgment threshold;

(4) Calculating the stator current bias according to the error polarity function, wherein the specific implementation mode is as follows: the error polarity function Λ _x is sampled multiple times in a given rotor position angle interval theta ₁～θ₂, and the average value is defined as a stator current offset degree and is used for representing the distortion characteristic of the stator current, wherein the specific expression is as follows:

Wherein: χ _x represents the x-phase stator current bias;

(5) Constructing an error discrimination function according to the stator current bias through the following relation to diagnose the open-circuit fault of the inverter;

F_ab＝sgn(χ_a-χ_b)

F_bc＝sgn(χ_b-χ_c)

F_ca＝sgn(χ_c-χ_a)

Wherein: f _ab represents an error discrimination function between ab two-phase stator current bias degrees, F _bc represents an error discrimination function between bc two-phase stator current bias degrees, F _ca represents an error discrimination function between ca two-phase stator current bias degrees, sgn () represents a sign function;

The criteria for diagnosing an open circuit fault of an inverter are as follows:

if F _ab＝-1,F_bc＝F_ca =1, judging that the open circuit fault of the upper tube of the phase a of the inverter exists;

if F _ab＝1,F_bc＝F_ca = -1, determining that the open circuit of the phase a down tube of the inverter is faulty;

If F _bc＝-1,F_ab＝F_ca =1, judging that the open circuit fault of the upper tube of the phase b of the inverter exists;

if F _bc＝1,F_ab＝F_ca = -1, determining that the open circuit fault of the phase b down tube of the inverter exists;

If F _ca＝-1,F_ab＝F_bc =1, judging that the open circuit fault of the upper tube of the phase c of the inverter exists;

If F _ca＝1,F_ab＝F_bc = -1, determining that the open circuit fault of the phase c down tube of the inverter;

If F _ab＝F_bc＝F_ca =0, it is determined that the inverter has no open circuit fault.

2. The method for diagnosing open-circuit faults of the inverter of the permanent magnet synchronous motor based on the current bias as claimed in claim 1, wherein the method comprises the following steps of: the specific implementation manner of diagnosing the open-circuit fault of the inverter in the step (5) is as follows: firstly, adopting the sum of absolute values of three-phase stator current residual errors after faults as an auxiliary variable epsilon _m for fault detection, namely epsilon _m＝|ε_af|+|ε_bf|+|ε_cf I; further, the auxiliary variable epsilon _m is quantized by using the fault detection trigger signal F _g, namely:

wherein: tr ₀ is a given fault detection threshold;

then, constructing a unified fault locating mark F according to the fault detection trigger signal F _g and the error discrimination function, namely F=F _g(4F_ab+2F_bc+F_ca;

finally, the inverter open circuit fault is diagnosed by the following criteria:

If F _g = 1 and F = -1, determining that the open circuit of the upper tube of the a phase of the inverter is faulty;

if F _g =1 and f=1, judging that the open circuit fault of the phase a down tube of the inverter exists;

if F _g =1 and f=3, judging that the open circuit fault of the upper tube of the phase b of the inverter exists;

If F _g = 1 and F = -3, determining that the open circuit fault exists in the phase b of the inverter;

If F _g =1 and f=5, judging that the open circuit fault of the upper tube of the phase c of the inverter exists;

If F _g =1 and f= -5, determining that the open circuit fault exists in the phase c of the inverter;

If F _g =f=0, it is determined that the inverter has no open circuit fault.