A short-circuit fault prediction method and device for a permanent magnet motor

Abstract

The application discloses a kind of permanent magnet motor short-circuit fault prediction method device, the method includes: in current traction system actual operating condition, the current variation trend of current fundamental component is obtained within set length;In response to the difference of first and second high-dimensional correlation coefficients meets preset condition, permanent magnet motor short-circuit fault prediction is carried out according to current variation trend;Wherein, the first high-dimensional correlation coefficient is calculated based on the joint distribution function of current and each influencing factor between inverter and isolation contactor, the second high-dimensional correlation coefficient is calculated based on the joint distribution function between each influencing factor related to current obtained by Pair-Copula method.The application carries out high-dimensional correlation analysis on short-circuit current influencing factor, improves fault diagnosis and prediction accuracy, solves the problem of induced back electromotive force and larger short-circuit torque when permanent magnet motor phase short-circuit, reduces the influence of permanent magnet motor short-circuit fault on train safe operation.

Description

Technical Field

[0001] This invention relates to the field of short-circuit fault detection technology for permanent magnet motors, and in particular to a method and apparatus for predicting short-circuit faults in permanent magnet motors. Background Technology

[0002] Short circuits in permanent magnet motors are mainly classified into two categories: inter-turn short circuits and phase-to-phase short circuits. If the insulation of coils on the same winding of a permanent magnet motor is poor, the coils stacked together will short-circuit, effectively rendering a portion of the coils ineffective. This leads to a reduction in the number of turns in that phase winding and an imbalance in the three-phase stator current; this phenomenon is called an inter-turn short circuit in a permanent magnet motor. Abnormal three-phase currents cause motor vibration, abnormal noise, and a significant decrease in the motor's average torque. If left uncontrolled, this can lead to other more serious short-circuit faults. When a two-phase or three-phase phase-to-phase short circuit suddenly occurs in the winding of a permanent magnet motor, a large transient inrush current will appear within the winding, with its peak value reaching several times the rated current. Because a phase-to-phase short circuit in a permanent magnet motor is an external short-circuit fault, and because permanent magnet motors have built-in permanent magnets and do not require excitation coils, a large short-circuit torque will still be generated after a phase-to-phase short circuit, which will have a certain impact on the safe operation of the permanent magnet motor and the train.

[0003] Currently, domestic and international experts and scholars have conducted in-depth research on inter-turn short-circuit faults in permanent magnet motors. However, most of this research is based on inter-turn short circuits in the stator windings of permanent magnet motors, and research on internal phase-to-phase short-circuit faults is still rare. See the patent document "A Method for Diagnosing Inter-turn Short-circuit Faults in Permanent Magnet Synchronous Motors" (CN201610327167.8), which provides a method for diagnosing inter-turn short-circuit faults in permanent magnet synchronous motors. This method provides real-time diagnosis of whether an inter-turn short-circuit fault has occurred in a permanent magnet synchronous motor, overcoming the shortcomings of existing technologies that cannot perform real-time diagnosis. Furthermore, it does not require studying the system's mathematical model or additional detection equipment. By using an extended Kalman filter to estimate the short-circuit current and the number of short-circuit turns, it can determine whether an inter-turn short-circuit fault has occurred and further identify the faulty phase, ensuring the real-time performance and reliability of the inter-turn short-circuit fault diagnosis for permanent magnet synchronous motors. However, this solution does not address phase-to-phase short circuits in permanent magnet motors. Based on simulation experiments and field operation, permanent magnet motors can still induce back electromotive force and generate a large short-circuit torque even when a phase-to-phase short-circuit fault occurs. Therefore, phase-to-phase short-circuit faults are more destructive than turn-to-turn short-circuit faults. Thus, it is necessary to propose a short-circuit fault prediction scheme for permanent magnet motors. Summary of the Invention

[0004] The technical problem to be solved by this invention is the problem of inducing back electromotive force and generating large short-circuit torque when a permanent magnet motor is short-circuited between phases, so as to further improve the accuracy of fault diagnosis and fault prediction and reduce the impact of short-circuit faults of permanent magnet motors on the safe operation of trains.

[0005] To address the aforementioned technical problems, this invention provides a method for predicting short-circuit faults in permanent magnet motors, comprising:

[0006] Under the current actual operating conditions of the traction system, obtain the current variation trend of the fundamental current component within a set time period;

[0007] If the difference between the first high-dimensional correlation coefficient and the second high-dimensional correlation coefficient meets the preset condition, then the short-circuit fault prediction of the permanent magnet motor is performed based on the current change trend; wherein, the first high-dimensional correlation coefficient is calculated based on the joint distribution function constructed by the current between the inverter and the isolation contactor and various influencing factors, and the second high-dimensional correlation coefficient is calculated based on the joint distribution function between various influencing factors related to the current obtained by the Pair-Copula method.

[0008] Optionally, the step of predicting short-circuit faults in the permanent magnet motor based on the current change trend includes:

[0009] If the current change trend is greater than a first set value, it is determined that the permanent magnet motor of the current traction system has a serious phase-to-phase short circuit fault.

[0010] Optionally, the step of predicting short-circuit faults in the permanent magnet motor based on the current change trend includes:

[0011] If the current change trend is less than the first set value and greater than the second set value, it is determined that the permanent magnet motor of the current traction system has experienced a minor phase-to-phase short circuit fault or is in a critical fault condition.

[0012] Optionally, the step of predicting short-circuit faults in the permanent magnet motor based on the current change trend includes:

[0013] If the current change trend is less than the specified value, then it is determined that the permanent magnet motor of the current traction system is operating normally.

[0014] Optionally, the preset conditions include:

[0015] The absolute value of the difference between the first high-dimensional correlation coefficient and the second high-dimensional correlation coefficient is less than the third set value.

[0016] To solve the above-mentioned technical problems, the present invention provides a short-circuit fault prediction device for permanent magnet motors, comprising:

[0017] The current change trend acquisition module is used to acquire the current change trend of the fundamental current component within a set time period under the current actual operating conditions of the traction system.

[0018] The short-circuit fault prediction module is used to predict short-circuit faults of permanent magnet motors based on the current change trend when the difference between the first high-dimensional correlation coefficient and the second high-dimensional correlation coefficient meets a preset condition. The first high-dimensional correlation coefficient is calculated based on the joint distribution function constructed from the current between the inverter and the isolation contactor and various influencing factors, and the second high-dimensional correlation coefficient is calculated based on the joint distribution function between various influencing factors related to current obtained by the Pair-Copula method.

[0019] Optionally, the short-circuit fault prediction module is specifically used to determine that a severe phase-to-phase short-circuit fault has occurred in the permanent magnet motor of the current traction system in response to the current change trend being greater than a first set value.

[0020] Optionally, the short-circuit fault prediction module is specifically used to determine that the permanent magnet motor of the current traction system has experienced a minor phase-to-phase short-circuit fault or is in a critical fault condition in response to the current change trend being less than the first set value and greater than the second set value.

[0021] Optionally, the short-circuit fault prediction module is specifically used to determine that the permanent magnet motor of the current traction system is operating normally in response to the current change trend being less than the specified value.

[0022] Optionally, the preset conditions include:

[0023] The absolute value of the difference between the first high-dimensional correlation coefficient and the second high-dimensional correlation coefficient is less than the third set value.

[0024] Compared with the prior art, one or more embodiments of the above solutions may have the following advantages or beneficial effects:

[0025] The permanent magnet motor short-circuit fault prediction method and device of the present invention acquires the current change trend of the fundamental current component within a set time period under the actual operating conditions of the current traction system; in response to the difference between the first high-dimensional correlation coefficient and the second high-dimensional correlation coefficient satisfying a preset condition, the permanent magnet motor short-circuit fault is predicted based on the current change trend; wherein, the first high-dimensional correlation coefficient is calculated based on the joint distribution function constructed from the current between the inverter and the isolation contactor and various influencing factors, and the second high-dimensional correlation coefficient is calculated based on the joint distribution function between various influencing factors related to the current obtained by the Pair-Copula method.

[0026] As can be seen, the present invention performs high-dimensional correlation analysis on the influencing factors of short-circuit current, providing data support for the formation of fault trees in fault diagnosis expert systems and laying the foundation for fault prediction using big data technology. Therefore, this invention overcomes the problem of traditional fault diagnosis and prediction methods that only consider the correlation between influencing factors without considering high-dimensional correlation analysis, thus further improving the accuracy of fault diagnosis and prediction. It also solves the problems of induced back electromotive force and large short-circuit torque during phase-to-phase short circuits in permanent magnet motors, reducing the impact of short-circuit faults in permanent magnet motors on the safe operation of trains. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 A main circuit diagram of a permanent magnet traction system provided in an embodiment of the present invention;

[0029] Figure 2 A flowchart of a short-circuit fault prediction method for permanent magnet motors provided in an embodiment of the present invention;

[0030] Figure 3 The Canonical vine structure diagram based on the Pair-Copula regression model provided by this invention;

[0031] Figure 4 This is a structural diagram of a short-circuit fault prediction device for permanent magnet motors provided in an embodiment of the present invention. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of the present invention clearer, the implementation method of the present invention will be described in detail below with reference to the accompanying drawings and embodiments, so that the process of how the present invention uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0033] In the prior art, permanent magnet motors have problems such as inducing back electromotive force and generating large short-circuit torque when the phases are short-circuited. Therefore, in order to further improve the accuracy of fault diagnosis and fault prediction and reduce the impact of short-circuit faults in permanent magnet motors on the safe operation of trains, this invention provides a method and device for predicting short-circuit faults in permanent magnet motors.

[0034] Please see Figure 1This is a main circuit diagram of a permanent magnet traction system provided in an embodiment of the present invention. Based on the main circuit of a subway vehicle traction system, it considers a two-phase or three-phase short circuit in the permanent magnet motor. In this case, the corresponding rotational speed of the permanent magnet motor is ω, and the motor torque is T. F The motor temperature is T. e The inverter output voltage is U, the current between the inverter and the isolation contactor is I, and the resistance values ​​at different locations of a phase-to-phase short circuit in the permanent magnet motor are Ω. A Pair-Copula autoregressive model is proposed to construct the relationship between the phase-to-phase short-circuit current I of the permanent magnet motor and related influencing factors ω and T. F T e A high-dimensional correlation model between U, Ω, etc. is established. Based on the obtained high-dimensional correlation model, big data technology is used to compare simulated sample data with actual traction system operating data to predict the possibility of phase-to-phase short circuits, thereby further reducing the impact of phase-to-phase short circuits of permanent magnet motors on train safety.

[0035] The following describes the short-circuit fault prediction method for permanent magnet motors provided by the present invention.

[0036] like Figure 2 The diagram shown is a flowchart of a short-circuit fault prediction method for permanent magnet motors provided in an embodiment of the present invention, which may include the following steps:

[0037] Step S101: Under the current actual operating conditions of the traction system, obtain the current change trend of the fundamental current component within a set time period.

[0038] Step S102: In response to the fact that the difference between the first high-dimensional correlation coefficient and the second high-dimensional correlation coefficient meets the preset condition, the short-circuit fault prediction of the permanent magnet motor is performed according to the current change trend; wherein, the first high-dimensional correlation coefficient is calculated based on the joint distribution function constructed by the current between the inverter and the isolation contactor and various influencing factors, and the second high-dimensional correlation coefficient is calculated based on the joint distribution function between various influencing factors related to the current obtained by the Pair-Copula method.

[0039] Please see Figure 3 An electrical simulation model of a permanent magnet traction system was built based on a hardware-in-the-loop simulation platform. Different phase-to-phase short-circuit locations, motor speeds, and inverter output voltages were then set to simulate various phase-to-phase short-circuit conditions. Based on a large number of simulation data samples, the kernel density estimation method was used to obtain the variables X (including I, ω, and T). F T e The marginal distribution function F(x, U, Ω) i (i.e., the first high-dimensional correlation coefficient).

[0040] It should be noted that although a single type of Copula function can directly yield the joint distribution function between multivariate random variables, the Copula function does not consider the cross-correlation between any two random variables and cannot fully account for the correlation between high-dimensional random variables, resulting in some discrepancies with reality. Therefore, this application introduces the Canonical vine-structured Pair-Copula method, employing a layer-by-layer merging approach to construct the joint distribution function of multivariate Copulas. Figure 3 The canonical structure diagram of the Pair-Copula method is given, where C represents the marginal distribution function F(x) corresponding to the random variable. i ) or conditional distribution function F(x) j |x i The copula function corresponding to the random variables is used to obtain the joint distribution of the random variables. This construction method combines random variables pairwise and introduces multiple binary copula functions (i.e., the second high-dimensional correlation coefficient), and can introduce multiple different types of copula distribution functions to enhance the practicality of the model.

[0041] In one implementation, the main steps of establishing a multivariate distribution function using the Pair-Copula method based on the Canonical vine structure are as follows:

[0042] Step (1): Determine the order of the random variables in the first layer of the Canonical vine structure based on the magnitude of the linear correlation coefficient between the random variables. Let the sorted order of the random variables be {x1, x2, ..., x...}. n}

[0043] Step (2): Using (n-1) binary Copula functions, combine the sorted random variables with the random variables corresponding to the first root node of the first layer to establish the first layer Pair-Copula sequence {c 12 (F(x1),F(x2)),...,c 1n (F(x1),F(x n )}.

[0044] Step (3): Take the (n-1) distribution functions obtained in step (2) as a new sequence of random variables, and repeat step (2) until all conditional Copula sequences {c} in the second layer are obtained. 23|1 (F(x2|x1),F(x3|x2)),...,c 2n | 1n (F(x2|x1),F(x n The calculation continues until |x1)} is completed.

[0045] Step (4): Repeat step (3) until only one binary conditional Copula function {c} remains. n-1,n|1,2,...,n-2 (F(x n-1 |x1,x2,...,x n-2 ),F(x n |x1,x2,...,x n-2 ))}until.

[0046] According to conditional probability theory, the joint probability density function of the n-ary random variables generated by the Pair-Copula method can be obtained:

[0047]

[0048] Among them, f k =f k (x k ) represents the variable x k The probability distribution function is obtained, and the kernel density estimation method is used to obtain c. j,j+i|1,...,j-1 =c j,j+i|1,2,...,j-1 (F(x j |x1,...,x j-1 ),F(x j+i |x1,...,x j-1 )).

[0049] Furthermore, the Pair-Copula method described above can overcome the limitation that only one type of Copula function can be used in obtaining the joint distribution function of multivariate random variables, and can consider the cross-correlation between any two random variables as well as the cross-correlation between high-dimensional random variables.

[0050] The main steps for predicting the likelihood of phase-to-phase short circuits by comparing sample data generated using the Pair-Copula method with actual traction system operating data are as follows:

[0051] Step (1) Calculate the random variables X1,...,X using the kernel density estimation method. n The marginal distribution function. Based on the Pair-Copula method described above, a Canonical vine structure diagram considering the correlation of random variables is constructed.

[0052] Step (2) estimates the parameters involved in each Copula function based on historical sample data, letting u1 = F1(x1),...,u n =F n (x n ).

[0053] Step 3: Generate n independent random numbers s1,...,s that follow a uniform distribution in the range [0,1]. n Let s1 = F1(x1), ..., s n =F(x) n |x1,...,x n ).

[0054] Step 4: Calculate the partial derivatives of the bivariate joint distribution function. Let u1 = s1 = F1(x1) and s2 = F(x2|x1), then we can obtain u2 = F2(x2).

[0055] Step 5, based on s3 = F(x3|x1,x2) and s2 = F(x2|x1),

[0056] The conditional distribution F(x3|x1) can be obtained. Based on s1=F1(x1), The value of s3 = F3(x3) can then be obtained.

[0057] Step 6, following the same procedure as step 5, use the generated random numbers s1,...,s i The current calculation results are u1,...,u i-1 and formula u was calculated in sequence. i =F i (x i ), i = 1, ..., n.

[0058] Step (7) uses kernel density estimation to determine the inverse function of the corresponding marginal distribution function and thus the corresponding random variables x1,...,x n The values ​​of are respectively

[0059] Optionally, the preset condition includes: the absolute value of the difference between the first high-dimensional correlation coefficient and the second high-dimensional correlation coefficient is less than a third preset value.

[0060] In one scenario, the prediction of a short-circuit fault in a permanent magnet motor based on the current change trend includes: in response to the current change trend being greater than a first set value, determining that a severe phase-to-phase short-circuit fault has occurred in the permanent magnet motor of the current traction system.

[0061] Specifically, the fundamental component of the current under actual operating conditions of the traction system shows a significant increasing trend within time Δt1, that is... Rapid changes Furthermore, the high-dimensional correlation coefficient ρ between current and other influencing factors g1 The high-dimensional correlation coefficient ρ with the generated sample data in 4.3 g2The difference satisfies Δσ=|ρ g1 -ρ g2 If |<ξ, then a serious phase-to-phase short-circuit fault is considered to have occurred.

[0062] In another scenario, the prediction of short-circuit faults of the permanent magnet motor based on the current change trend includes: in response to the current change trend being less than the first set value and greater than the second set value, determining that the permanent magnet motor of the current traction system has experienced a minor phase-to-phase short-circuit fault or is in a critical fault condition.

[0063] The fundamental component of the current under actual operating conditions of the traction system shows an increasing trend within time Δt1, but The change was not significant. Furthermore, the high-dimensional correlation coefficient ρ between current and other influencing factors g1 The high-dimensional correlation coefficient ρ with the generated sample data in 4.3 g2 The difference satisfies Δσ=|ρ g1 -ρ g2 If |<ξ, then it is considered that a minor phase-to-phase short circuit fault has occurred or that the fault is at its critical point, i.e., a minor fault.

[0064] In another scenario, the prediction of short-circuit faults in the permanent magnet motor based on the current change trend includes: in response to the current change trend being less than the specified value, determining that the permanent magnet motor of the current traction system is operating normally.

[0065] The fundamental component of the current under actual operating conditions of the traction system does not show an increasing trend within time Δt1, that is... Not much change Furthermore, the high-dimensional correlation coefficient ρ between current and other influencing factors g1 The high-dimensional correlation coefficient ρ with the generated sample data in 4.3 g2 The difference satisfies Δσ=|ρ g1 -ρ g2 If |<ξ, then this operating condition is considered to be a normal operating condition.

[0066] This invention performs high-dimensional correlation analysis on the influencing factors of short-circuit current, providing data support for the formation of fault trees in fault diagnosis expert systems and laying the foundation for fault prediction using big data technology. Therefore, this invention overcomes the problem of traditional fault diagnosis and prediction methods that only consider the correlation between influencing factors without considering high-dimensional correlation analysis. It can further improve the accuracy of fault diagnosis and prediction, solve the problem of induced back electromotive force and large short-circuit torque during phase-to-phase short circuits in permanent magnet motors, and reduce the impact of short-circuit faults in permanent magnet motors on the safe operation of trains.

[0067] Permanent magnet traction systems, due to their built-in permanent magnets and the absence of excitation coils, can still induce significant back electromotive force and generate substantial short-circuit torque after a phase-to-phase short circuit. This can negatively impact the safe operation of the permanent magnet motor and the train. This invention aims to propose a method for predicting and evaluating phase-to-phase short circuits in permanent magnet motors, enabling real-time diagnosis of short-circuit faults in permanent magnet traction systems. Its main innovations are as follows:

[0068] (1) Innovatively, the Pair-Copula method is proposed to consider the correlation between the phase-to-phase short-circuit current of the permanent magnet traction system and other influencing factors, so as to realize the comprehensive consideration of the nonlinear correlation between variables and the high-dimensional correlation between variables, making the correlation model more in line with the actual situation.

[0069] (2) For the first time, a method combining the variation trend of the fundamental current component with the high-dimensional correlation coefficient of the current influencing factors is proposed to predict the phase-to-phase short circuit of permanent magnet motors, and different fault types are defined to further improve the accuracy of phase-to-phase short circuit fault diagnosis.

[0070] (3) A high-dimensional correlation model of phase-to-phase short-circuit current is established, and a sample data generation method based on the Pair-Copula method is given. This method can quickly compare the differences between actual operating conditions and the corresponding operating conditions in the fault model library, and has good practicality.

[0071] Through literature review and retrieval, current research on short-circuit fault diagnosis methods and control systems in permanent magnet traction systems mainly focuses on inter-turn short-circuit fault diagnosis methods, but reports on phase-to-phase short-circuit fault diagnosis are still rare. Furthermore, inter-turn short-circuit fault diagnosis methods still primarily rely on traditional control methods and big data expert diagnostic systems, lacking research on key factors influencing short-circuit faults and their correlations. This results in reported fault diagnosis methods being applicable only to specific operating conditions and research models, and thus not suitable for large-scale application in real-world situations. This invention proposes a method for generating the joint probability distribution of short-circuit fault current in permanent magnet motors, representing an interdisciplinary and cross-professional application from the financial securities field to the industrial field. By employing the Pair-Copula method to construct a high-dimensional correlation model among the influencing factors of short-circuit fault current, considering both high-dimensional and nonlinear correlations among these factors, the model more closely approximates reality. Simultaneously, using the changing trend of the fundamental current component and the high-dimensional correlation coefficients corresponding to the current influencing factors as indicators, phase-to-phase short-circuit prediction of permanent magnet motors is performed, and different fault types are defined to improve the accuracy of fault diagnosis and prediction. This invention does not require additional hardware facilities. It only reserves sample storage space in the underlying control software for comparing actual operating condition data with simulated sample data to achieve fault identification and fault prediction. The method is simple, safe, and reliable.

[0072] The short-circuit fault prediction device for permanent magnet motors provided in the embodiments of the present invention will be described below.

[0073] like Figure 4 The diagram shown is a structural diagram of a short-circuit fault prediction device for a permanent magnet motor provided in an embodiment of the present invention, including: a current change trend acquisition module 210 and a short-circuit fault prediction module 220.

[0074] Among them, the current change trend acquisition module 210 is used to acquire the current change trend of the fundamental current component within a set time period under the current actual operating conditions of the traction system.

[0075] The short-circuit fault prediction module 220 is used to predict short-circuit faults of the permanent magnet motor based on the current change trend when the difference between the first high-dimensional correlation coefficient and the second high-dimensional correlation coefficient meets a preset condition. The first high-dimensional correlation coefficient is calculated based on the joint distribution function constructed from the current between the inverter and the isolation contactor and various influencing factors, and the second high-dimensional correlation coefficient is calculated based on the joint distribution function between various influencing factors related to the current obtained by the Pair-Copula method.

[0076] Optionally, the short-circuit fault prediction module 220 is specifically used to determine that a severe phase-to-phase short-circuit fault has occurred in the permanent magnet motor of the current traction system in response to the current change trend being greater than a first set value.

[0077] Optionally, the short-circuit fault prediction module 220 is specifically used to determine that the permanent magnet motor of the current traction system has experienced a minor phase-to-phase short-circuit fault or is in a critical fault condition in response to the current change trend being less than the first set value and greater than the second set value.

[0078] Optionally, the short-circuit fault prediction module 220 is specifically used to determine that the permanent magnet motor of the current traction system is operating normally in response to the current change trend being less than the specified value.

[0079] Optionally, the preset condition includes: the absolute value of the difference between the first high-dimensional correlation coefficient and the second high-dimensional correlation coefficient is less than a third preset value.

[0080] This invention performs high-dimensional correlation analysis on the influencing factors of short-circuit current, providing data support for the formation of fault trees in fault diagnosis expert systems and laying the foundation for fault prediction using big data technology. Therefore, this invention overcomes the problem of traditional fault diagnosis and prediction methods that only consider the correlation between influencing factors without considering high-dimensional correlation analysis. It can further improve the accuracy of fault diagnosis and prediction, solve the problem of induced back electromotive force and large short-circuit torque during phase-to-phase short circuits in permanent magnet motors, and reduce the impact of short-circuit faults in permanent magnet motors on the safe operation of trains.

[0081] Therefore, obtaining the joint probability distribution of short-circuit fault current in permanent magnet motors is the foundation and an important component of big data-based motor short-circuit fault diagnosis and early warning, and it is also a direction that permanent magnet traction systems currently need to break through.

[0082] It should be noted that the Pair-Copula method proposed in this invention is a correlation construction method based on a vine structure, which can consider high-dimensional and nonlinear correlations between variables. However, when there are few influencing factors or the model is simplified, constructing the correlation of variables using conventional Copula functions also has certain applicability and falls within the scope of protection of this invention.

[0083] Furthermore, although this invention is based on phase-to-phase short circuits in permanent magnet motors, the solutions of this invention are also applicable to inter-turn short circuit faults in motors, and are also within the scope of protection of this invention.

[0084] Furthermore, the short-circuit fault types defined in this invention are based on the degree of variation trend of the fundamental component of the actual operating current and the high-dimensional correlation coefficient of the current influencing factors. Similarly, performing Fast Fourier Transform or Wavelet Transform on each current influencing factor, and also considering high-dimensional correlation using the fundamental component, also falls within the scope of protection of this invention.

[0085] For system or device embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to in the description of the method embodiments.

[0086] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0087] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0088] It should be understood that the terminology used in this application specification is for the purpose of describing particular embodiments only and is not intended to limit the application. As used in this application specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0089] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0090] As used in this specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrases "if determined" or "if the described condition or event is detected" may be interpreted, depending on the context, as "once determined," "in response to determination," "once the described condition or event is detected," or "in response to the detection of the described condition or event."

[0091] While the embodiments disclosed in this invention are as described above, the content is merely for the purpose of facilitating understanding of the invention and is not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and changes in form and detail of the implementation without departing from the spirit and scope disclosed herein; however, the scope of protection of this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A method for predicting short-circuit faults in a permanent magnet motor, characterized in that, include: Under the current actual operating conditions of the traction system, the current variation trend of the fundamental current component within a set time period is obtained; wherein, the traction system includes an inverter, an isolation contactor and a permanent magnet motor connected in sequence, and the current is the current between the inverter and the isolation contactor; If the difference between the first high-dimensional correlation coefficient and the second high-dimensional correlation coefficient meets a preset condition, then a short-circuit fault prediction of the permanent magnet motor is performed based on the current change trend; wherein, the first high-dimensional correlation coefficient is calculated based on the joint distribution function constructed from the current between the inverter and the isolation contactor and various influencing factors, and the second high-dimensional correlation coefficient is calculated based on the joint distribution function between various influencing factors related to the current obtained by the Pair-Copula method; The preset conditions include: The absolute value of the difference between the first high-dimensional correlation coefficient and the second high-dimensional correlation coefficient is less than the third set value.

2. The method for predicting short-circuit faults in permanent magnet motors according to claim 1, characterized in that, The method of predicting short-circuit faults in permanent magnet motors based on the current change trend includes: If the current change trend is greater than a first set value, it is determined that the permanent magnet motor of the current traction system has a serious phase-to-phase short circuit fault.

3. The method for predicting short-circuit faults in permanent magnet motors according to claim 1, characterized in that, The method of predicting short-circuit faults in permanent magnet motors based on the current change trend includes: If the current change trend is less than a first set value and greater than a second set value, it is determined that the permanent magnet motor of the current traction system has experienced a minor phase-to-phase short circuit fault or is in a critical fault condition.

4. The method for predicting short-circuit faults in permanent magnet motors according to claim 1, characterized in that, The method of predicting short-circuit faults in permanent magnet motors based on the current change trend includes: If the current change trend is less than a second set value, then the permanent magnet motor of the current traction system is determined to be operating normally.

5. A short-circuit fault prediction device for a permanent magnet motor, characterized in that, include: The current change trend acquisition module is used to acquire the current change trend of the fundamental component of the current within a set time period under the current actual operating conditions of the traction system; wherein, the traction system includes an inverter, an isolation contactor and a permanent magnet motor connected in sequence, and the current is the current between the inverter and the isolation contactor; The short-circuit fault prediction module is used to predict short-circuit faults of permanent magnet motors based on the current change trend when the difference between the first high-dimensional correlation coefficient and the second high-dimensional correlation coefficient meets a preset condition. The first high-dimensional correlation coefficient is calculated based on the joint distribution function constructed from the current between the inverter and the isolation contactor and various influencing factors, and the second high-dimensional correlation coefficient is calculated based on the joint distribution function between various influencing factors related to current obtained by the Pair-Copula method. The preset conditions include: The absolute value of the difference between the first high-dimensional correlation coefficient and the second high-dimensional correlation coefficient is less than the third set value.

6. The short-circuit fault prediction device for permanent magnet motors according to claim 5, characterized in that, The short-circuit fault prediction module is specifically used to determine that a severe phase-to-phase short-circuit fault has occurred in the permanent magnet motor of the current traction system when the current change trend is greater than a first set value.

7. The short-circuit fault prediction device for permanent magnet motors according to claim 5, characterized in that, The short-circuit fault prediction module is specifically used to determine that the permanent magnet motor of the current traction system has experienced a minor phase-to-phase short-circuit fault or is in a critical fault condition when the current change trend is less than a first set value and greater than a second set value.

8. The short-circuit fault prediction device for permanent magnet motors according to claim 5, characterized in that, The short-circuit fault prediction module is specifically used to determine that the permanent magnet motor of the current traction system is operating normally in response to the current change trend being less than a second set value.

Images