Non-intrusive detection method and system for dynamic eccentricity fault of brushless doubly-fed motor

Abstract

The invention provides a non-intrusive detection method and system for a dynamic eccentricity fault of a brushless doubly-fed motor, and the method comprises the steps: calculating the fault characteristic frequency of a control winding of a frequency converter connected with the motor according to the basic electromagnetic relation of the brushless doubly-fed motor when the dynamic eccentricity of a rotor occurs; detecting the current of the motor by using an FFT (Fast Fourier Transform) decomposition method, and if a fault characteristic frequency signal occurs, recording a time-varying current signal X (t) in a converter directly connected with a control winding in a certain running time period of the motor; performing dynamic signal decomposition on the current signal X (t) to obtain a natural mode function of the current signal; and fitting the natural mode function of the current signal, comparing the fitting curve with the curve in the non-eccentric working state, and if the fitting curve deviates from the curve in the non-eccentric working state, judging that a dynamic eccentric fault occurs. According to the method, the fault is detected on the premise of not adding an additional sensor, and the system reliability is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power equipment, and in particular to a non-intrusive detection method and system for a dynamic eccentricity fault of a brushless double-fed motor. Background Art

[0002] The brushless doubly-fed motor inherits the advantages of the traditional AC excitation motor's dual electrical ports, omitting brushes and slip rings, and has the advantages of simple structure and durability. It has broad application prospects in high-voltage explosion-proof drives, offshore wind power, pumped storage and other fields.

[0003] The wound rotor structure has good harmonic suppression capability and design flexibility, which can ensure the efficient and smooth operation of the brushless doubly-fed motor. It has become the most widely used brushless doubly-fed motor technology route in commercial applications. However, since the wound rotor structure needs to use unequal turns, unequal pitches and other means to adjust the rotor winding arrangement, it will lead to uneven distribution of empty slots or conductors in the slots, resulting in asymmetric internal mass of the motor, which can easily cause dynamic eccentricity of the motor in actual operation, and then cause serious damage to the motor. In order to effectively avoid the dynamic eccentricity fault of the brushless doubly-fed motor without increasing the system cost, a non-invasive diagnostic method for detecting the dynamic eccentricity fault of the brushless doubly-fed motor during operation is urgently needed.

[0004] Most of the existing technical documents only use finite element analysis to analyze the harmonic content of the motor air gap magnetic flux density, and need to rely on invasive means such as magnetic field detection coils to diagnose eccentricity faults. Even non-invasive detection methods are only applicable to static eccentricity faults of brushless doubly fed motors, and cannot be used to detect dynamic eccentricity faults of brushless doubly fed motors. Summary of the invention

[0005] To solve the above problems, the present invention provides a non-invasive detection method for the dynamic eccentricity fault of a brushless doubly-fed motor. Through the special electromagnetic coupling principle of the brushless doubly-fed motor, the dynamic eccentricity fault of the brushless doubly-fed motor is diagnosed using the converter signal without adding additional sensors, effectively coping with the potential faults of this type of motor due to the complex rotor topology.

[0006] To achieve the above object, the first aspect of the present invention discloses a method for diagnosing a short-circuit fault in a power winding of a brushless doubly-fed motor, comprising the following steps:

[0007] According to the basic electromagnetic relationship of the brushless double-fed motor, when the rotor dynamic eccentricity occurs, the fault characteristic frequency of the control winding of the motor connected to the inverter is calculated;

[0008] The motor current is detected by using the FFT decomposition method. If a fault characteristic frequency signal appears, the current signal X(t) that changes with time in the converter directly connected to the control winding during a certain period of motor operation is recorded;

[0009] Perform dynamic signal decomposition on the current signal X(t) to obtain the natural mode function of the current signal;

[0010] The natural mode function of the current signal is fitted and compared with the curve under the non-eccentric working state. If the fitting curve is offset compared with the curve under the non-eccentric working state, it is determined that a dynamic eccentricity fault has occurred.

[0011] Preferably, the calculation of the control winding fault characteristic frequency is specifically as follows:

[0012] When the brushless doubly-fed motor has dynamic eccentricity, calculate the air gap permeance;

[0013] Multiply the air gap permeability by the magnetomotive force of the brushless doubly-fed motor to obtain the harmonic magnetic field and calculate the magnetic induction intensity of the magnetic field;

[0014] The magnetic field harmonics caused by the dynamic eccentricity fault can be induced into the current frequency of the stator winding, and then the frequency of the dynamic eccentricity fault reaction to the time harmonics can be calculated.

[0015] Preferably, the air gap permeability can be expressed by the following formula:

[0016]

[0017] In the formula, is the air gap permeability, Λ a0 is the constant term when the motor is eccentric, Λ a is the coefficient of the non-constant term, a is the harmonic order, ω r is the rotor rotation angular frequency.

[0018] Preferably, the magnetic induction intensity of the magnetic field can be expressed by the following formula:

[0019]

[0020] In the formula, B p is the magnetic induction intensity of the power winding; B c To control the magnetic induction intensity of the winding; B 1p , B 1c B p and B c The amplitude of p p is the number of pole pairs of the power winding; p c To control the number of winding pole pairs; ω p is the angular frequency of the rotating magnetic field of the power winding; ω c To control the angular frequency of the winding rotating magnetic field; It is the initial phase angle of the synthetic magnetomotive force of the main wave of the power winding; To control the initial phase angle of the main wave of the winding, θ is the angle of the air gap circumferential space position, ω ε is the angular velocity of the eccentric air gap when the rotor rotates, and ε is the rotor eccentricity.

[0021] Preferably, the frequency of the dynamic eccentricity fault response to the time harmonic can be expressed by the following formula:

[0022]

[0023] In the formula, f pε is the distortion frequency of the power winding frequency after the eccentricity fault, f cε To control the distortion frequency of the winding frequency after an eccentricity fault, f p is the power winding current frequency, f c To control the winding current frequency, p p is the number of pole pairs of the power winding, p c To control the number of winding pole pairs, s p and c They are the slip rates of the power winding and the control winding respectively.

[0024] Preferably, the power winding slip rate s p and control winding slip s c It can be expressed by the following formula:

[0025]

[0026] In the formula, f p is the power winding current frequency, f c To control the winding current frequency, p p is the number of pole pairs of the power winding, p c To control the number of winding pole pairs, a negative sign is used when the phase sequence of the power winding and the control winding is the same, and a positive sign is used otherwise.

[0027] Preferably, the dynamic signal decomposition of the current signal X(t) is specifically performed as follows:

[0028] Identify all the maximum and minimum points of the original signal, fit the extreme points, and form the upper bound X of the signal max (t) and the lower bound X min (t), calculate the average value m1(t) of the upper and lower bounds;

[0029] Subtract the original current signal X(t) from the average value m1(t) of the upper and lower bounds to obtain the remaining signal d1(t);

[0030] The remaining signal d1(t) is regarded as the new original signal X(t) to be processed, and the above steps are processed repeatedly until the first-order mode component NF1(t) is obtained;

[0031] Subtract the original signal X(t) from the first-order mode component NF1(t) to obtain the first-order residual y1(t). Replace X(t) with y1(t) and iterate m times until the standard deviation meets the set stop criterion and the mth NF component is obtained. The mth-order mode function NF m (t) and the final residual quantity y that meets the requirements m (t);

[0032] The original signal X(t) is decomposed and represented by a natural mode function and a residual.

[0033] Preferably, the standard deviation can be expressed by the following formula:

[0034]

[0035] In the formula, SD is the standard deviation, d k-1 (t) and d k (t) are the time series of two continuous signals, and T represents the time span.

[0036] Preferably, the decomposed original signal X(t) can be expressed by the following formula:

[0037]

[0038] In the formula, NF m is the mth natural mode function, y M is the original signal X(t) minus NF m The remaining amount after.

[0039] The second aspect of the present invention discloses a non-intrusive detection system for a dynamic eccentricity fault of a brushless doubly-fed motor, and the non-intrusive detection method for a dynamic eccentricity fault of a brushless doubly-fed motor described above is executed, comprising:

[0040] A frequency calculation module, used to calculate the fault characteristic frequency of the control winding of the motor connected to the inverter;

[0041] A current detection module is used to detect the motor current by using the FFT decomposition method;

[0042] A signal decomposition module is used to dynamically decompose the current signal of the motor;

[0043] The fault judgment module is used to detect the eccentricity fault and judge the severity of the eccentricity fault.

[0044] Beneficial effects of the present invention:

[0045] The present invention proposes a non-intrusive detection method for dynamic eccentricity fault of a brushless doubly-fed motor. The method uses the converter current signal on the control winding side of the brushless doubly-fed motor for analysis. Without adding any additional sensors, the electromagnetic coupling relationship of the brushless doubly-fed motor can be used to calculate the fault characteristic frequency of the control winding current directly connected to the converter when a rotor eccentricity fault occurs. The fault degree can be identified through data processing methods such as EMD decomposition, thereby greatly improving the system reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flow chart of a non-intrusive detection method of a dynamic eccentricity fault of a brushless double-fed motor according to an embodiment of the present invention;

[0047] Figure 2 is a structural block diagram of a brushless doubly-fed motor system according to an embodiment of the present invention; DETAILED DESCRIPTION

[0048] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.

[0049] The present invention provides a non-intrusive detection method for a dynamic eccentricity fault of a brushless double-fed motor. The method uses a converter signal on a control winding side to diagnose the dynamic eccentricity fault of a brushless double-fed motor rotor without adding an additional sensor.

[0050] Figure 2 The brushless doubly-fed motor and its control system structure block diagram provided in the embodiment of the present invention include a brushless doubly-fed motor, an inverter, and a transformer; wherein the brushless doubly-fed motor includes a stator and a rotor; the stator of the brushless doubly-fed motor includes a power winding and a control winding, wherein the power winding is directly connected to the power grid, and the control winding is connected to the power grid in series through the inverter.

[0051] The first embodiment of the present invention provides a non-intrusive detection method for dynamic eccentricity fault of a brushless double-fed motor, such as Figure 1 As shown, the specific steps include:

[0052] Step 1: According to the basic electromagnetic relationship of the brushless doubly-fed motor, when the rotor dynamic eccentricity occurs, calculate the fault characteristic frequency of the control winding of the motor connected to the inverter. The specific calculation steps are as follows:

[0053] Step 1.1, when the motor has dynamic eccentricity, calculate the air gap permeance. The air gap permeance is a function of time and space and can be expressed by formula (1), as follows:

[0054]

[0055] In the formula, is the air gap permeability, Λ a0 is the constant term when the motor is eccentric, Λ a is the coefficient of the non-constant term, a is the harmonic order, ω r is the rotor rotation angular frequency.

[0056] Step 1.2, the synthetic air gap magnetomotive force of the brushless doubly fed motor is formed by the superposition of multiple magnetomotive forces. The air gap permeability expression obtained in step 1.1 is multiplied by the magnetomotive force to obtain the harmonic magnetic field. The magnetic induction intensity of the magnetic field can be expressed by equations (2) and (3), as follows:

[0057]

[0058] In the formula, B p is the magnetic induction intensity of the power winding; B c To control the magnetic induction intensity of the winding; B 1c B p and B c The amplitude of p p is the number of pole pairs of the power winding; p c To control the number of winding pole pairs; ω p is the angular frequency of the rotating magnetic field of the power winding; ω c To control the angular frequency of the winding rotating magnetic field; It is the initial phase angle of the synthetic magnetomotive force of the main wave of the power winding; To control the initial phase angle of the main wave of the winding, θ is the angle of the air gap circumferential space position, ω ε is the angular velocity of the eccentric air gap when the rotor rotates, and ε is the rotor eccentricity.

[0059] In step 1.3, the magnetic field harmonics caused by the dynamic eccentricity fault can be induced into the current frequency of the stator winding, and then the expression of the dynamic eccentricity fault response to the time harmonic frequency is obtained, that is, the coefficient of the time term of the magnetic induction intensity in step 1.2. This expression can be expressed by equations (4) and (5), as follows:

[0060]

[0061] In the formula, f pε is the distortion frequency of the power winding frequency after the eccentricity fault; f cε To control the distortion frequency of the winding frequency after an eccentricity fault, p pis the number of pole pairs of the power winding, p c To control the number of winding pole pairs, s p and c are the power winding and control winding slip rates, s p and c It can be expressed by formula (6) and formula (7), as follows:

[0062]

[0063] In the formula, f p is the power winding current frequency, f c To control the winding current frequency, p p is the number of pole pairs of the power winding, p c To control the number of winding pole pairs, a negative sign is used when the phase sequence of the power winding and the control winding is the same, and a positive sign is used otherwise.

[0064] For a brushless doubly-fed motor with 4 power winding poles and 8 control winding poles, when the motor speed is 700r / min, it can be calculated that the distortion frequency caused by dynamic eccentricity is 50Hz and 60Hz.

[0065] As a prominent substantive feature of the present invention and one of the significant advances it brings to the prior art, the present invention proposes a non-invasive diagnostic method for the dynamic eccentricity fault of a brushless doubly-fed motor. No additional sensors are required. A converter connected to the motor control winding is used to extract the control winding current signal, and data processing is performed to establish a mathematical model of the dynamic eccentricity fault of the brushless doubly-fed motor, and analyze the changing laws of the magnetic field and current.

[0066] Step 2: Use FFT decomposition method to detect the motor current. If the fault characteristic frequency signal f calculated in step 1 appears pε or cε , then record a current signal X(t) that changes with time in the converter directly connected to the control winding during a certain period of motor operation;

[0067] Step 3: Perform dynamic signal decomposition on the current signal X(t) calculated in step 2 to obtain a natural mode function of the current signal.

[0068] In a preferred but non-limiting embodiment of the present invention,

[0069] Step 3.1, identify all the maximum and minimum points of the original signal, fit these extreme points by curve interpolation method, and form the upper bound X of the signal max (t) and the lower bound X min (t), its average value m1(t) can be expressed by formula (7), as follows:

[0070]

[0071] Step 3.2, subtract the original signal X(t) from the average value m1(t) of the upper and lower bounds to obtain the remaining signal d1(t). Generally, for a stationary signal, it is the first mode function NF of the original signal X(t), but the signal is not stationary, it is not a monotonic function in a certain area, but there will be inflection points. If these inflection points that can reflect the specific characteristics of the original signal are not included, then its first mode function is not accurate, that is, the remaining signal does not meet the conditions of the natural mode function, so it is necessary to continue screening.

[0072] Step 3.3, treat the remaining signal d1(t) as the new original signal X(t) to be processed, and perform the above steps repeatedly until the first-order mode component NF1(t) that meets the conditions is obtained. Subtract the signal X(t) from the first-order mode component NF1(t) to obtain the first-order residual y1(t), and use y1(t) to replace X(t) and iterate m times until the standard deviation SD meets the set stop standard and the m-th NF component is obtained. The m-th order mode function NF m (t) and the final residual quantity y that meets the requirements m (t).

[0073] The stopping criterion can also be called the screening threshold, which is generally set between 0.2 and 0.3.

[0074] The standard deviation SD can be expressed by formula (8), as follows:

[0075]

[0076] In the formula, SD is the standard deviation, d k-1 (t) and d k (t) are the time series of two continuous signals, and T represents the time span.

[0077] Step 3.4, decompose the input signal into several natural mode functions and a residual component, which can be expressed by formula (9), as follows:

[0078]

[0079] Where X(t) is the original signal, NF m is the mth natural mode function, y M is the original signal minus NF m The remaining amount after.

[0080] In step 4, after iterative decomposition of the dynamic signal, its time domain characteristics can be reflected in the current amplitude. The natural mode function NF of the current signal obtained in step 3 is fitted and compared with the current curve under the normal non-eccentric working state. If the fitting curve is offset compared with the current curve under the normal non-eccentric working state, and there are large changes in most of the peaks, it is judged that a dynamic eccentricity fault has occurred. At the same time, the degree of offset can be used as a criterion for judging the severity of the dynamic eccentricity fault.

[0081] As a prominent substantive feature of the present invention and one of the significant advances it brings to the prior art, the present invention extracts the fault characteristics of dynamic eccentricity from the control winding current signal, and through multiple iterations of dynamic signal decomposition, fits and compares it with the curve under the normal non-eccentric working state, and uses the degree of offset as the basis for judging the severity of the eccentricity fault.

[0082] A second embodiment of the present invention provides a non-intrusive detection system for a dynamic eccentricity fault of a brushless doubly-fed motor, and runs a non-intrusive detection method for a dynamic eccentricity fault of a brushless doubly-fed motor as claimed in claim 1, comprising:

[0083] A frequency calculation module, used to calculate the fault characteristic frequency of the control winding of the motor connected to the inverter;

[0084] A current detection module is used to detect the motor current by using the FFT decomposition method;

[0085] A signal decomposition module is used to dynamically decompose the current signal of the motor;

[0086] The fault judgment module is used to detect the eccentricity fault and judge the severity of the eccentricity fault.

[0087] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0088] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0089] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A non-intrusive detection method for dynamic eccentricity fault of a brushless doubly-fed motor, characterized in that: include: According to the basic electromagnetic relationship of the brushless double-fed motor, when the rotor dynamic eccentricity occurs, the fault characteristic frequency of the control winding of the motor connected to the inverter is calculated; The motor current is detected by using FFT decomposition. If a fault characteristic frequency signal appears, the current signal X(t) that changes with time in the converter directly connected to the control winding during a certain period of motor operation is recorded; Perform dynamic signal decomposition on the current signal X(t) to obtain the natural mode function of the current signal; The natural mode function of the current signal is fitted and compared with the curve under the non-eccentric working state. If the fitting curve is offset compared with the curve under the non-eccentric working state, it is determined that a dynamic eccentricity fault has occurred.

2. A non-intrusive detection method for dynamic eccentricity fault of a brushless doubly-fed motor as claimed in claim 1, characterized in that: The specific calculation of the control winding fault characteristic frequency is: When the brushless doubly-fed motor has dynamic eccentricity, calculate the air gap permeance; Multiply the air gap permeability by the magnetomotive force of the brushless doubly-fed motor to obtain the harmonic magnetic field and calculate the magnetic induction intensity of the magnetic field; The magnetic field harmonics caused by the dynamic eccentricity fault can be induced into the current frequency of the stator winding, and then the frequency of the dynamic eccentricity fault reaction to the time harmonics can be calculated.

3. A non-intrusive detection method for dynamic eccentricity fault of a brushless doubly-fed motor as claimed in claim 2, characterized in that: The air gap permeability can be expressed by the following formula: In the formula, is the air gap permeability, Λ a0 is the constant term when the motor is eccentric, Λ a is the coefficient of the non-constant term, a is the harmonic order, ω r is the rotor rotation angular frequency.

4. A non-intrusive detection method for dynamic eccentricity fault of a brushless doubly-fed motor as claimed in claim 2, characterized in that: The magnetic induction intensity of the magnetic field can be expressed by the following formula: In the formula, B p is the magnetic induction intensity of the power winding; B c To control the magnetic induction intensity of the winding; B 1p , B 1c B p and B c The amplitude of p p is the number of pole pairs of the power winding; p c To control the number of winding pole pairs; ω p is the angular frequency of the rotating magnetic field of the power winding; ω c To control the angular frequency of the winding rotating magnetic field; It is the initial phase angle of the synthetic magnetomotive force of the main wave of the power winding; To control the initial phase angle of the main wave of the winding, θ is the angle of the air gap circumferential space position, ω ε is the angular velocity of the eccentric air gap when the rotor rotates, and ε is the rotor eccentricity.

5. A non-intrusive detection method for dynamic eccentricity fault of a brushless doubly-fed motor as claimed in claim 2, characterized in that: The frequency of the dynamic eccentricity fault response to time harmonics can be expressed by the following formula: In the formula, f pε is the distortion frequency of the power winding frequency after the eccentricity fault, f cε To control the distortion frequency of the winding frequency after an eccentricity fault, f p is the power winding current frequency, f c To control the winding current frequency, p p is the number of pole pairs of the power winding, p c To control the number of winding pole pairs, s p and c They are the slip rates of the power winding and the control winding respectively.

6. A non-intrusive detection method for dynamic eccentricity fault of a brushless doubly-fed motor as claimed in claim 5, characterized in that: Power winding slip rate s p and control winding slip s c It can be expressed by the following formula: In the formula, f p is the power winding current frequency, f c To control the winding current frequency, p p is the number of pole pairs of the power winding, p c To control the number of winding pole pairs, a negative sign is used when the phase sequence of the power winding and the control winding is the same, and a positive sign is used otherwise.

7. A non-intrusive detection method for dynamic eccentricity fault of a brushless doubly-fed motor as claimed in claim 1, characterized in that: The dynamic signal decomposition of the current signal X(t) is as follows: Identify all the maximum and minimum points of the original signal, fit the extreme points, and form the upper bound X of the signal max (t) and the lower bound X min (t), calculate the average value m1(t) of the upper and lower bounds; Subtract the original current signal X(t) from the average value m1(t) of the upper and lower bounds to obtain the remaining signal d1(t); The remaining signal d1(t) is regarded as the new original signal X(t) to be processed, and the above steps are processed repeatedly until the first-order mode component NF1(t) is obtained; Subtract the original signal X(t) from the first-order mode component NF1(t) to obtain the first-order residual y1(t). Replace X(t) with y1(t) and iterate m times until the standard deviation meets the set stop criterion and the mth NF component is obtained. The mth-order mode function NF m (t) and the final residual quantity y that meets the requirements m (t); The original signal X(t) is decomposed and represented by a natural mode function and a residual.

8. A non-intrusive detection method for dynamic eccentricity fault of a brushless doubly-fed motor as claimed in claim 7, characterized in that: The standard deviation can be expressed as follows: In the formula, SD is the standard deviation, d k-1 (t) and d k (t) are the time series of two continuous signals, and T represents the time span.

9. A non-intrusive detection method for dynamic eccentricity fault of a brushless doubly-fed motor as claimed in claim 7, characterized in that: The decomposed original signal X(t) can be expressed by the following formula: In the formula, NF m is the mth natural mode function, y M is the original signal X(t) minus NF m The remaining amount after.

10. A non-intrusive detection system for dynamic eccentricity fault of a brushless double-fed motor, running a non-intrusive detection method for dynamic eccentricity fault of a brushless double-fed motor as claimed in claims 1 to 9, characterized in that: include: A frequency calculation module, used to calculate the fault characteristic frequency of the control winding of the motor connected to the inverter; A current detection module is used to detect the motor current by using the FFT decomposition method; A signal decomposition module is used to dynamically decompose the current signal of the motor; The fault judgment module is used to detect the eccentricity fault and judge the severity of the eccentricity fault.

Images