Database-based permanent magnet motor short-circuit fault unbalanced magnetic pulling force correction method

Abstract

The invention relates to the technical field of permanent magnet motor operation and maintenance, in particular to a database-based permanent magnet motor short-circuit fault unbalanced magnetic pull correction method, which comprises the following steps of: arranging a two-phase back-wound correction winding on a permanent magnet generator rotor; obtaining unbalanced magnetic pulling force characteristic parameters of the permanent magnet generator rotor in typical working states, wherein the typical working states of the permanent magnet generator rotor comprise a normal working state and various typical fault states; and obtaining a correction strategy of the permanent magnet generator rotor in the typical working state. According to the embodiment of the invention, the current working state of the permanent magnet generator is quickly judged by establishing the permanent magnet generator fault state correction strategy database, and when the permanent magnet generator is in the fault state, the correction strategy in the database is quickly called; and a correction current is applied to a correction winding on a rotor of the permanent magnet generator based on a correction strategy, so that the unbalanced magnetic pulling force in the permanent magnet generator is rapidly counteracted.

Description

Technical Field

[0001] The present application relates to the technical field of permanent magnet motor operation and maintenance, and in particular to a database-based method for correcting unbalanced magnetic pull of a permanent magnet motor short-circuit fault. Background Art

[0002] Inter-turn short circuits often occur in permanent magnet motors. When short circuits occur in the permanent magnet motor components, unbalanced magnetic pull will be generated in the permanent magnet motor, significantly increasing the risk of damage to the permanent magnet motor. In the prior art, there is a lack of rapid processing methods for the above-mentioned phenomenon, making it difficult to curb the problem in a timely manner. Summary of the invention

[0003] In view of the shortcomings of the prior art, the present application provides a method for correcting the unbalanced magnetic pull of a permanent magnet motor short-circuit fault based on a database. The embodiment of the present application quickly judges the current working state of the permanent magnet generator by establishing a permanent magnet generator fault state correction strategy database. When the permanent magnet generator is in a fault state, the correction strategy in the database is quickly called, and a correction current is applied to the correction winding on the permanent magnet generator rotor based on the correction strategy, so that the unbalanced magnetic pull inside the permanent magnet generator is quickly offset.

[0004] The above application objectives of this application are achieved through the following technical solutions:

[0005] The method for correcting unbalanced magnetic pull of a permanent magnet motor short circuit fault based on a database comprises the following steps:

[0006] A two-phase back-wound correction winding is arranged on the rotor of the permanent magnet generator;

[0007] Acquire unbalanced magnetic pull characteristic parameters under a typical working state of a permanent magnet generator rotor, wherein the typical working state of the permanent magnet generator rotor includes a normal working state and various typical fault states;

[0008] Obtaining a correction strategy under a typical working state of a permanent magnet generator rotor, the correction strategy including a current phase and an amplitude of a correction current output to a correction winding;

[0009] Establish a data set based on the characteristic parameters and correction strategies corresponding to the typical working state of the permanent magnet generator rotor and its working state labels;

[0010] Obtain the current unbalanced magnetic pull characteristic parameters of the permanent magnet generator rotor;

[0011] Determine the current working status label of the permanent magnet generator rotor;

[0012] When the permanent magnet generator rotor is currently in a fault state, a correction strategy corresponding to the current fault state label is obtained;

[0013] According to the correction strategy corresponding to the current fault state label, the correction current is output to the two-phase correction winding.

[0014] In summary, this application has the following beneficial technical effects:

[0015] The embodiment of the present application establishes a permanent magnet generator fault state correction strategy database to quickly judge the current working state of the permanent magnet generator. When the permanent magnet generator is in a fault state, the correction strategy in the database is quickly called, and based on the correction strategy, a correction current is applied to the correction winding on the permanent magnet generator rotor, so that the unbalanced magnetic pull inside the permanent magnet generator is quickly offset. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the structure of a permanent magnet generator;

[0017] Figure 2 It is a schematic diagram of the spatial coordinates of the unbalanced magnetic pull force of the inter-turn short circuit;

[0018] Figure 3 It is a schematic diagram of the unbalanced magnetic pull correction principle of an embodiment of the present application;

[0019] Figure 4 It is a schematic diagram comparing the correction effects before and after correction of an embodiment of the present application. DETAILED DESCRIPTION

[0020] The present application is further described in detail below in conjunction with the accompanying drawings.

[0021] In order to more clearly understand the technical solution presented in the embodiments of the present application, a brief introduction to the existing permanent magnet generator and its failures is first given.

[0022] Permanent magnet generators have excellent power density characteristics and have shown broad application prospects in many fields such as wind power generation, electric vehicle drive, household appliance power supply, industrial drive and control, and have become an indispensable power supply unit in modern power systems. It consists of two core components: the stator and the rotor. The stator integrates the iron core and winding to carry the current conversion, while the rotor relies on the synergy of the shaft, permanent magnets and guard rings, especially the permanent magnets as the cornerstone of magnetic field generation, to achieve efficient conversion of mechanical energy to electrical energy.

[0023] Although permanent magnet generators are known for their high efficiency, high reliability, compact structure and significant energy-saving benefits, the potential fault of inter-turn short circuit frequently causes equipment damage, seriously disrupting the normal power supply order in power consumption sites. In addition, due to the lack of immediate and effective fault correction methods, the problem is often difficult to contain in time, further exacerbating economic losses and waste of resources.

[0024] In view of the above technical problems, the embodiment of the present application provides a method for correcting unbalanced magnetic pull of a permanent magnet motor short circuit fault based on a database, and implements a precise correction strategy for the inter-turn short circuit of the permanent magnet generator, thereby effectively offsetting the unbalanced magnetic pull and restoring the stable operation state of the generator, which includes the following steps:

[0025] S101. Arrange two-phase back-wound correction windings on the permanent magnet generator rotor. The two-phase back-wound correction windings have a phase difference of 120°, and each phase occupies 3 slots to inject correction current to balance the unbalanced magnetic pull caused by the inter-turn short circuit fault;

[0026] S102, obtaining unbalanced magnetic pull characteristic parameters of the permanent magnet generator rotor under a typical working state, wherein the typical working state of the permanent magnet generator rotor includes a normal working state and various typical fault states, and the operating characteristic parameters of the permanent magnet generator rotor can be obtained by a sensor disposed inside the permanent magnet generator;

[0027] S103, obtaining a correction strategy under a typical working state of the permanent magnet generator rotor, the correction strategy including a current phase and an amplitude of a correction current output to the correction winding;

[0028] S104, establishing a data set according to characteristic parameters and correction strategies corresponding to typical working states of permanent magnet generator rotors and working state labels;

[0029] S105, obtaining characteristic parameters of the current unbalanced magnetic pull of the permanent magnet generator rotor;

[0030] S106, determining the current working state label of the permanent magnet generator rotor;

[0031] S107, when the permanent magnet generator rotor is currently in a fault state, obtaining a correction strategy corresponding to the current fault state label;

[0032] S108. Output correction current to the two-phase correction winding according to the correction strategy corresponding to the current fault state label.

[0033] The following is a further introduction based on specific usage scenarios.

[0034] First, step S101 is performed: two-phase back-wound correction windings are arranged on the rotor of the permanent magnet generator, the two-phase back-wound correction windings are 120° apart from each other, and each phase occupies 3 slots, so as to inject correction current to balance the unbalanced magnetic pull caused by the inter-turn short circuit fault, and the correction winding and the working winding are independent of each other;

[0035] Then, step S102 is performed: obtaining unbalanced magnetic pull characteristic parameters of the permanent magnet generator rotor under a typical working state, wherein the characteristic parameters include a magnetic field intensity change rate, a current harmonic content, and a voltage imbalance, etc. The typical working state of the permanent magnet generator rotor includes a normal working state and various typical fault states. In the embodiment of the present application, the typical unbalanced magnetic pull characteristic parameters of the permanent magnet generator rotor can be obtained by the following steps:

[0036] First, the operation data of the permanent magnet generator is obtained, wherein the operation data includes magnetic field intensity, current magnitude and voltage value. The operation data of the permanent magnet generator can be captured by a sensor inside the permanent magnet generator;

[0037] Then the acquired operation data of the permanent magnet generator are preprocessed to remove obvious noise and outliers;

[0038] Finally, a data analysis algorithm is used to extract characteristic parameters related to the unbalanced magnetic pull characteristics of the faulty winding from the operating data. The characteristic parameters include the magnetic field intensity change rate, current harmonic content, and voltage imbalance.

[0039] By extracting features from the operating data of the permanent magnet generator, the amount of calculation of the operating data is reduced, thereby improving the correction efficiency of the unbalanced magnetic pull of the permanent magnet motor.

[0040] In step S103, a correction strategy under a typical working state of the permanent magnet generator rotor is obtained, the correction strategy including a current phase, an amplitude and a power-on time of a correction current output to the correction winding;

[0041] S104: establishing a data set according to characteristic parameters and correction strategies corresponding to typical working states of the permanent magnet generator rotor and its working state labels;

[0042] S105: Obtain the current unbalanced magnetic pull characteristic parameters of the permanent magnet generator rotor. In the embodiment of the present application, the current unbalanced magnetic pull characteristic parameters of the permanent magnet generator rotor can be obtained through the following steps:

[0043] First, the operation data of the permanent magnet generator is obtained, wherein the operation data includes magnetic field intensity, current magnitude and voltage value. The operation data of the permanent magnet generator can be captured by a sensor inside the permanent magnet generator;

[0044] Then the acquired operation data of the permanent magnet generator are preprocessed to remove obvious noise and outliers;

[0045] Then, a data analysis algorithm is used to extract characteristic parameters related to the unbalanced magnetic pull characteristics of the fault winding from the operating data, the characteristic parameters including the rate of change of magnetic field intensity, current harmonic content, and voltage unbalance, etc.;

[0046] Finally, the extracted characteristic parameters are quantitatively analyzed to determine the magnitude, direction and distribution of the unbalanced magnetic pull of the faulty winding;

[0047] The radial component of the unbalanced magnetic pull is:

[0048] where μ 0 is the vacuum permeability, A is a coefficient related to the stator and rotor structure of the motor. g is the eccentricity of the air gap length, and δ is the average air gap length;

[0049] By extracting features from the operating data of the permanent magnet generator, the amount of calculation of the operating data is reduced, thereby improving the correction efficiency of the unbalanced magnetic pull of the permanent magnet motor.

[0050] S106: Based on the data set, determine the current working state label of the permanent magnet generator rotor;

[0051] S107: When the permanent magnet generator rotor is currently in a fault state, obtaining a correction strategy corresponding to the current fault state label;

[0052] S108: Outputting correction current to the two-phase correction winding according to the correction strategy corresponding to the current fault state label.

[0053] In general, the embodiment of the present application quickly judges the current working state of the permanent magnet generator by establishing a permanent magnet generator fault state correction strategy database. When the permanent magnet generator is in a fault state, the correction strategy in the database is quickly called, and based on the correction strategy, a correction current is applied to the correction winding on the permanent magnet generator rotor, so that the unbalanced magnetic pull inside the permanent magnet generator is quickly offset.

[0054] In order to understand the above method more clearly, this application is based on Figure 1 Taking a three-phase permanent magnet generator as an example, high-precision sensors are installed at key positions inside the permanent magnet generator to continuously collect various data during the operation of the generator at an appropriate frequency, including magnetic field strength, current size, voltage value, etc.

[0055] The collected data is preliminarily sorted and preprocessed to remove obvious noise and outliers. The preprocessed data is deeply analyzed. Signal processing technology and data analysis algorithms are used to extract characteristic parameters related to the unbalanced magnetic pull characteristics of the faulty winding, such as the rate of change of magnetic field intensity, current harmonic content, voltage imbalance, etc. The extracted characteristic parameters are then quantitatively analyzed to determine the size, direction and distribution of the unbalanced magnetic pull of the faulty winding.

[0056] The radial component of the unbalanced magnetic pull is:

[0057] where μ0 is the vacuum permeability, A is a coefficient related to the stator and rotor structure of the motor, g is the eccentricity of the air gap length, and δ is the average air gap length.

[0058] Then, the permanent magnet generator is analyzed for faults in combination with the data set. The unbalanced magnetic pull characteristic data extracted under the current fault condition is matched with the records in the database to determine whether a fault has occurred. The database includes both normal operation data and fault data. The fault data includes fault data of turn-to-turn short circuit faults such as slot 1, slot 2, and slot 3.

[0059] By comparing with the database data, it is possible to preliminarily determine whether there is any abnormality in the stator slots inside the motor. If all the data are normal, no subsequent operations are performed and the permanent magnet generator is continuously monitored in real time.

[0060] If the diagnosis result is abnormal data, it is compared with the database data to diagnose that a turn-to-turn short circuit fault has occurred in stator slot 1. The diagnosis result also includes characteristic parameters such as fault degree and time. After that, a correction strategy matching the current fault condition is selected from the database.

[0061] Through the algorithm model, a comprehensive evaluation of similar correction strategies is performed to obtain the best two-phase correction strategies, namely, correction 1 and correction 2. The correction strategies of correction 1 and correction 2 include the current phase and amplitude that should be injected into the correction winding:

[0062] Exemplarily, the database of the correction strategy may include the data shown in Table 1:

[0063]

[0064] Table 1 Partial data of the database

[0065] The correction principle is as follows Figure 2 As shown:

[0066] In t 1 In the time period, two-phase correction currents, correction 1 and correction 2, are injected to generate a correction magnetic pull F 1 ;

[0067] The magnetic pull F will be corrected 1 Decompose the force on the x-axis to F x1 , the force on the y-axis is F y1 ;

[0068] At the same time, the rotor winding is 1 Unbalanced magnetic pull F generated during the time period 2 Decomposed to the x-axis as F x2 , decomposed to the y-axis as F y2 ;

[0069] Decomposition of the magnetic pull F by correctionx1 、F y1 Respectively decompose the unbalanced magnetic pull F x2 、F y2 interact with each other, thereby offsetting the unbalanced magnetic pull generated by the rotor winding.

[0070] Figure 3 The superposition effect of the unbalanced magnetic pull of the rotor winding and the magnetic pull generated by the correction winding alone is shown:

[0071] Among them, the first figure from the left shows the unbalanced magnetic pull of the rotor winding decomposed on the x-axis and y-axis, with high force amplitude corresponding to the y-axis and low amplitude corresponding to the x-axis;

[0072] The second figure from the left shows the magnetic pull force decomposed into the x-axis and y-axis when the two-phase correction windings Xiu 1 and Xiu 2 act separately;

[0073] At this time, the high force amplitude corresponds to the x-axis, and the low force amplitude corresponds to the y-axis;

[0074] By superimposing the two graphs, we can get Figure 3 The third picture from the left shows the effect of canceling out the unbalanced magnetic pull;

[0075] Finally, there is no unbalanced magnetic pull on either the x-axis or the y-axis.

[0076] The specific implementation steps of the repair 1 and repair 2 strategies are as follows:

[0077] First, the determined two-phase correction strategies 1 and 2 are converted into practical and operable instructions, including control signals, parameter settings, etc., wherein the instructions have clear formats and specifications, and can be accurately identified and executed by the control system of the correction winding. The generated instructions are transmitted to the control system of the two-phase correction winding through a reliable communication interface. According to the instructions, the control system forcibly injects the current phase of 60° and -60° of the two-phase correction winding back-wound in the corresponding slot 1, and the amplitude is 10A. In the process of executing the correction strategies 1 and 2, the operating status of the generator is continuously monitored, including the magnetic field strength, current size, voltage value, etc.;

[0078] Through the data fed back by the sensors, the effects of the combined effects of correction strategies 1 and 2 are evaluated in real time, such as Figure 4 As shown, the motor operates normally at 0-6ms, and the unbalanced magnetic pull generated is 0N;

[0079] At 6ms, a turn-to-turn short circuit fault occurs in the motor stator slot, generating an unbalanced magnetic pull on the rotor;

[0080] In the 6-12ms time period, the correction system quickly collects fault data and compares it with the database to determine the fault location and extent. At 12ms, it injects two-phase correction current to offset the unbalanced magnetic pull, thereby achieving the correction effect.

[0081] As a possible implementation method of the embodiment of the present application, obtaining a correction strategy under a typical working state of a permanent magnet generator rotor includes the following steps:

[0082] Step S201: Acquire historical corrected current data of the permanent magnet generator under various states. A large amount of historical corrected current data of different types of permanent magnet generators under various short-circuit fault conditions may be widely collected to establish a corrected current database. The data sources include actual generator failure cases, laboratory test results, and research literature from related fields.

[0083] Step S202: Classifying and processing various types of corrected current data according to the permanent magnet generator fault parameters, wherein the permanent magnet generator fault parameters include the short circuit fault location and the fault degree, and marking in detail according to factors such as the short circuit fault location and the fault degree, so as to quickly and accurately screen out records similar to the current short circuit fault situation in subsequent comparative analysis;

[0084] Step S203: Label the classified corrected current data, label the corresponding permanent magnet generator fault state type, and establish a corrected current database. Of course, as new fault cases continue to emerge and develop, the database needs to be updated and maintained regularly, new corrected current data should be added in time, and outdated or inaccurate data should be deleted to ensure the accuracy and timeliness of the database.

[0085] As a possible implementation method of the embodiment of the present application, the following steps are also included:

[0086] Step S301: Establishing a fault identification model;

[0087] Step S302: using the characteristic parameters in the data set and their corresponding state labels as input and supervision to train a fault recognition model;

[0088] Step S303: saving the trained fault identification model parameters, using the fault identification model to quickly identify and diagnose the current working state of the permanent magnet generator, and matching the relevant correction strategy, so as to realize the rapid diagnosis and correction of the unbalanced magnetic pull of the permanent magnet generator;

[0089] Step S304: inputting the current unbalanced magnetic pull characteristic parameters of the permanent magnet generator into the fault identification model;

[0090] Step S305: the fault identification model identifies the corresponding working state label according to the current unbalanced magnetic pull characteristic parameters of the permanent magnet generator, and when the permanent magnet generator rotor is currently in a normal working state, continuously obtains the unbalanced magnetic pull characteristic parameters of the permanent magnet generator rotor;

[0091] Step S306: After outputting the correction current to the two-phase correction winding, the corrected characteristic parameters of the permanent magnet generator are obtained;

[0092] Step S307: judging whether the corrected permanent magnet generator is in a normal working state by using a fault identification model according to the corrected permanent magnet generator characteristic parameters;

[0093] Step S308: When it is determined that the permanent magnet generator is still in a fault state, adjusting the fault identification model parameters;

[0094] Step S309: When it is determined that the permanent magnet generator is still in a fault state, the correction strategy adopted by it is marked, and the operator can delete or update the correction strategy according to the mark, so that the correction strategy database of the permanent magnet generator is dynamically updated to fit the use scenario of the permanent magnet generator.

[0095] In order to correct the unbalanced magnetic pull, the embodiment of the present application combines the dynamic physical model of the permanent magnet generator with the real-time operating parameters, and dynamically generates the optimal correction current control strategy by comparing the algorithm model with the database. It not only accurately specifies the phase and amplitude of the two-phase correction current, but also takes into account the real-time status and future trends of the generator operation, ensuring that the correction action is both accurate and efficient, and can offset the unbalanced magnetic pull to the maximum extent, restore the stable operation of the generator, and prevent secondary damage. The unbalanced magnetic pull may also aggravate the bearing wear and even cause the permanent magnet to contact the stator core, causing secondary damage. The implementation of the correction strategy can prevent this from happening and extend the service life of the motor.

[0096] The embodiment of the present application also incorporates a closed-loop control mechanism. By continuously monitoring the correction effect and adjusting the correction strategy in real time, a set of adaptive and self-optimizing correction processes are formed. The optimal correction strategy is determined in combination with the correction current database, thereby realizing intelligent correction of the inter-turn short-circuit fault of the permanent magnet generator; the correction effect is continuously monitored and optimized to form an adaptive and self-optimizing correction process, thereby improving the correction efficiency, improving the reliability, stability and service life of the generator, and reducing maintenance costs. This highly intelligent control method not only improves the accuracy and reliability of the correction, but also enhances the system's adaptability to complex operating environments, providing a strong guarantee for the long-term stable operation of the permanent magnet generator.

[0097] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A database-based method for correcting unbalanced magnetic pull of a permanent magnet motor short-circuit fault, characterized in that: include: A two-phase back-wound correction winding is arranged on the rotor of the permanent magnet generator; Acquire unbalanced magnetic pull characteristic parameters under a typical working state of a permanent magnet generator rotor, wherein the typical working state of the permanent magnet generator rotor includes a normal working state and various typical fault states; Obtaining a correction strategy under a typical working state of a permanent magnet generator rotor, the correction strategy including a current phase and an amplitude of a correction current output to a correction winding; Establish a data set based on the characteristic parameters and correction strategies corresponding to the typical working state of the permanent magnet generator rotor and its working state labels; Obtain the current unbalanced magnetic pull characteristic parameters of the permanent magnet generator rotor; Determine the current working status label of the permanent magnet generator rotor; When the permanent magnet generator rotor is currently in a fault state, a correction strategy corresponding to the current fault state label is obtained; According to the correction strategy corresponding to the current fault state label, the correction current is output to the two-phase correction winding.

2. According to the database-based method for correcting unbalanced magnetic pull of a permanent magnet motor short-circuit fault according to claim 1, it is characterized in that: The correction strategy for obtaining the typical working state of the permanent magnet generator rotor includes: Obtain historical corrected current data of the permanent magnet generator under various states; Classifying and processing various types of corrected current data according to permanent magnet generator fault parameters, wherein the permanent magnet generator fault parameters include short circuit fault location and fault degree; The classified corrected current data are labeled, the corresponding permanent magnet generator fault state types are labeled, and a corrected current database is established.

3. The method for correcting unbalanced magnetic pull of a permanent magnet motor short circuit fault based on a database according to claim 2 is characterized in that: Obtaining the current unbalanced magnetic pull characteristic parameters of the permanent magnet generator rotor includes: Acquiring operating data of the permanent magnet generator, wherein the operating data includes magnetic field strength, current magnitude and voltage value; Preprocess the running data to remove noise and outliers; Characteristic parameters related to the unbalanced magnetic pull characteristics of the faulty winding are extracted from the operating data, wherein the characteristic parameters include the magnetic field intensity change rate, the current harmonic content and the voltage unbalance degree.

4. A database-based method for correcting unbalanced magnetic pull of a permanent magnet motor short-circuit fault according to claim 3, characterized in that: It also includes: establishing a fault identification model; The fault recognition model is trained using the feature parameters in the data set and their corresponding state labels as input and supervision; Save the trained fault identification model parameters; Inputting the current unbalanced magnetic pull characteristic parameters of the permanent magnet generator into the fault identification model; The fault identification model identifies the corresponding working state label according to the current unbalanced magnetic pull characteristic parameters of the permanent magnet generator.

5. A database-based method for correcting unbalanced magnetic pull of a permanent magnet motor short-circuit fault according to claim 4, characterized in that: Also includes: When the permanent magnet generator rotor is currently in a normal working state, the unbalanced magnetic pull characteristic parameters of the permanent magnet generator rotor are continuously obtained.

6. A database-based method for correcting unbalanced magnetic pull of a permanent magnet motor short-circuit fault according to claim 5, characterized in that: Also includes: After outputting the correction current to the two-phase correction winding, the corrected characteristic parameters of the permanent magnet generator are obtained; According to the corrected permanent magnet generator characteristic parameters, a fault identification model is used to determine whether the corrected permanent magnet generator is in a normal working state; When it is determined that the permanent magnet generator is still in a fault state, the fault identification model parameters are adjusted.

7. A database-based method for correcting unbalanced magnetic pull of a permanent magnet motor short-circuit fault according to claim 6, characterized in that: Also includes: When it is determined that the permanent magnet generator is still in a fault state, the correction strategy adopted by it is marked.

Images