Motor stator insulation fault on-line detection method, device, equipment and medium
By obtaining the time series of multiple state variables of the motor, establishing a co-integration model, and judging the motor health status, solving the problems of missing and false alarms caused by single signal analysis, and achieving accurate online detection of motor stator insulation faults, with high real-time performance and precision.
Patent Information
- Application Number
- CN202310538048.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-12
- Publication Date
- 2024-11-12
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing online detection methods for motor stator insulation faults rely on a single signal analysis, and cannot accurately detect faults, and are prone to missed or false alarms, and cannot detect motor stator insulation faults in time.
By obtaining the time series of multiple state variables of the motor, establishing a cointegration model, determining the new information vector, and judging the health status of the motor, avoiding false alarms and missed reports from single signal processing.
It realizes accurate online detection of motor stator insulation faults, avoids false alarms and missed alarms, has the advantages of high real-time performance, easy deployment and accurate detection, and is suitable for engineering applications.
Smart Images

Figure CN118937990A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of fault detection, and more specifically, to an online detection method, device, equipment and medium for motor stator insulation fault. Background Art
[0002] The motor is an important device for realizing the mutual conversion between electrical energy and mechanical energy. It is widely used in major industrial fields such as aviation, aerospace, rail transportation, petrochemicals, etc. Its importance is self-evident. However, with the comprehensive application of motors, the cost of operation and maintenance is also increasing. At the same time, motor failure will not only cause damage to equipment and threaten the safe production, but also the additional downtime caused by motor failure will also cause significant economic losses. Therefore, the development of motor fault diagnosis and monitoring technology to achieve fault warning has important application value. Common types of motor failures include stator failure, rotor failure and shaft failure. According to survey results, about 40% of motor failures are related to stator insulation.
[0003] The motor stator insulation fault detection technology is mainly divided into the motor offline insulation detection method and the online insulation detection method. However, the offline detection method cannot detect the motor stator insulation fault in the interval between two detections. Therefore, the online detection method of the motor stator insulation fault has important engineering significance and practical value. At present, the online motor stator insulation detection method mainly focuses on the signal processing-based method, which mainly takes a single signal as the object, such as: only analyzing and detecting through the current signal or voltage signal of the motor. This analysis method can contain limited object information. When the motor stator insulation system fails, the mutual correlation of various parts of the motor may cause the dynamic relationship between multiple system variables of the motor to change. In the early stage of the fault or under special working conditions, there is no fault feature in a single signal. At this time, there will be a fault omission. Or in the normal state of the motor, some special working conditions cause a single signal to have a fault feature, while other related signals have no abnormalities, which will cause a false alarm. Therefore, the monitoring method based on single signal analysis cannot accurately obtain the state information of the motor insulation and cannot detect the stator insulation fault of the motor in time. Summary of the invention
[0004] The purpose of the present invention is to provide an online detection method, device, equipment and medium for motor stator insulation fault, so as to detect the motor stator insulation fault in a timely and accurate manner.
[0005] To achieve the above object, the present invention provides an online detection method for motor stator insulation fault, comprising:
[0006] Acquire the time series of each state variable of the motor; the types of the state variables include at least two, each state variable is a first-order single integral non-stationary variable, and each state variable has a cointegration relationship;
[0007] Input the time series of each state variable into the cointegration model, and determine the innovation vector through the cointegration model; wherein the cointegration model is established based on the historical data of each state variable;
[0008] A health state of the motor is determined based on the innovation vector.
[0009] Optionally, before acquiring the time series of each state variable of the motor, the method further includes:
[0010] Obtain the historical time series of each initial state variable;
[0011] Perform a stationarity test on the historical time series of each initial state variable to obtain the target state variable; among which, the historical time series of each target state variable is a first-order single integral non-stationary series;
[0012] A cointegration test is performed on each target state variable to determine each state variable having a cointegration relationship and a cointegration relationship matrix of the cointegration model.
[0013] Optionally, the stationarity test of the historical time series of each initial state variable includes:
[0014] The ADF test method is used to test the stationarity of the historical time series of each initial state variable.
[0015] Optionally, the cointegration test for each target state variable includes:
[0016] The cointegration test of each target state variable is carried out using the Johansen test method.
[0017] Optionally, after determining each state variable having a cointegration relationship and the cointegration relationship matrix of the cointegration model, the method further includes:
[0018] To verify the cointegration model;
[0019] If the verification is successful, the step of obtaining the time series of each state variable of the motor is executed;
[0020] If the verification fails, the initial state variables are re-determined, and the step of obtaining the historical time series of each initial state variable is continued.
[0021] Optionally, obtaining the time series of each state variable of the motor includes:
[0022] Obtain a time series of at least two state variables of the motor, including the speed, voltage, current, torque, and flux.
[0023] Optionally, determining the health state of the motor according to the innovation vector includes:
[0024] Determine whether the row vector of the innovation vector is a stationary vector;
[0025] If not, it is determined that the motor has a stator insulation fault.
[0026] Optionally, if the motor has a stator insulation fault, the online detection method further includes:
[0027] Generate an alarm message indicating that the motor has a stator insulation fault.
[0028] Optionally, determining the innovation vector by using the cointegration model includes:
[0029] The innovation vector is determined by the cointegration model, and the cointegration model is Θ t =β T y; where Θ t is the new information vector, β is the cointegration relationship matrix, y is the time series of each state variable, and T represents the matrix transpose.
[0030] To achieve the above object, the present invention further provides an online detection device for motor stator insulation fault, comprising:
[0031] A time series acquisition module is used to acquire the time series of each state variable of the motor; the types of the state variables include at least two, each state variable is a first-order single integral non-stationary variable, and each state variable has a cointegration relationship;
[0032] An innovation vector determination module is used to input the time series of each state variable into a cointegration model and determine the innovation vector through the cointegration model; wherein the cointegration model is established based on the historical data of each state variable;
[0033] A state determination module is used to determine the health state of the motor according to the new information vector.
[0034] Optionally, the online detection device also includes
[0035] A historical time series acquisition module is used to obtain the historical time series of each initial state variable;
[0036] The stationarity test module is used to perform a stationarity test on the historical time series of each initial state variable to obtain the target state variable; wherein the historical time series of each target state variable is a first-order single integral non-stationary series;
[0037] The cointegration test module is used to perform cointegration test on each target state variable, determine each state variable with cointegration relationship, and the cointegration relationship matrix of the cointegration model.
[0038] Optionally, the online detection device further includes:
[0039] The model verification module is used to verify the cointegration model; if the verification is successful, the time series acquisition module is triggered; if the verification fails, the initial state variables are re-determined and the historical time series acquisition module is triggered.
[0040] Optionally, the time series acquisition module is specifically used to obtain the time series of at least two state variables of the motor: speed, voltage, current, torque, and flux.
[0041] To achieve the above object, the present invention further provides an electronic device, comprising:
[0042] Memory for storing computer programs;
[0043] A processor is used to implement the steps of the above-mentioned method for online detection of motor stator insulation fault when executing the computer program.
[0044] To achieve the above objectives, the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned online detection method of motor stator insulation fault are implemented.
[0045] It can be seen from the above scheme that the embodiment of the present invention provides an online detection method, device, equipment and medium for motor stator insulation fault; in this scheme, it is first necessary to obtain the time series of at least two state variables of the motor, each state variable is a first-order single integral non-stationary variable, and each state variable has a cointegration relationship; then the time series of each state variable is input into the cointegration model, and the new information vector is determined by the cointegration model; wherein the cointegration model is established according to the historical data of each state variable; the health state of the motor is determined according to the new information vector. It can be seen that this scheme can obtain the new information vector that can characterize the health state of the motor through the time series of at least two state variables and the cointegration model, realize the online detection of motor stator insulation fault, avoid the inaccurate fault detection caused by extracting the motor stator insulation fault characteristics through single signal processing, and avoid false alarms and missed alarms; and, when the scheme performs online fault diagnosis, no additional sensors are required, and no additional modifications are required to the train equipment. It has the advantages of easy deployment and implementation, high real-time performance, and accurate detection to avoid false alarms and missed alarms of stator insulation faults, and has practical engineering application value. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0047] Figure 1 A flow chart of an online detection method for motor stator insulation fault disclosed in an embodiment of the present invention;
[0048] Figure 2 A flow chart of motor stator insulation fault detection based on cointegration theory disclosed in an embodiment of the present invention;
[0049] Figure 3 A schematic structural diagram of an online detection device for motor stator insulation fault disclosed in an embodiment of the present invention;
[0050] Figure 4 The present invention is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present invention. DETAILED DESCRIPTION
[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0052] The embodiment of the invention discloses an online detection method, device, equipment and medium for motor stator insulation fault, so as to timely and accurately detect the motor stator insulation fault.
[0053] See also Figure 1 , a flow chart of an online detection method for motor stator insulation fault provided by an embodiment of the present invention includes:
[0054] S101, obtaining a time series of each state variable of the motor; the types of the state variables include at least two, each state variable is a first-order single integral non-stationary variable, and each state variable has a cointegration relationship;
[0055] Specifically, this scheme is mainly based on the cointegration theory to perform online detection of motor stator insulation faults. Cointegration is defined as: if the absolute values of two (or more) time series variables show instability (each variable has a unit root), but a certain linear combination of them shows stability, then there is a long-term stable relationship (cointegration relationship) between these variables. Therefore, the state variables selected in this scheme are first-order single-integrated non-stationary variables, and there is a cointegration relationship between the state variables; the first-order single-integration means: if a time series becomes stationary after a difference, the original series is called first-order single-integration.
[0056] In this solution, the verified state variables that meet the above requirements include five types: speed, voltage, current, torque, and flux. Therefore, in order to improve the detection accuracy and avoid false alarms and missed alarms during online detection, this solution selects at least two state variables for online detection, such as: selecting at least two from the motor speed, voltage, current, torque, and flux as the state variables of this solution. Correspondingly, when obtaining the time series of each state variable of the motor, this solution also obtains the time series of at least two state variables of the motor speed, voltage, current, torque, and flux.
[0057] S102, inputting the time series of each state variable into the cointegration model, and determining the new information vector through the cointegration model; wherein the cointegration model is established based on the historical data of each state variable;
[0058] In this embodiment, before detecting the health status of the motor, it is necessary to establish a cointegration model based on the historical data of each state variable to determine the cointegration relationship matrix in the cointegration model. Specifically, when determining the innovation vector through the cointegration model in this embodiment, the cointegration model used is Θ t =β T y; where Θ t is the new information vector, β is the cointegration relationship matrix, y is the time series of each state variable, and T represents the matrix transpose. If the state variables in this scheme include speed, voltage, current, torque, and flux, then y in this scheme = (y t ,y t-1 ,y t-2 ,y t-3 ,y t-4 ), where y t is the time series of each state variable at the current moment, y t-1 for y t The first-order lag term, y t-2 for y t The second-order lag term, y t-3 for y t The third-order lag term, y t-4 for y t The fourth-order lag term of .
[0059] S103. Determine the health status of the motor according to the new information vector.
[0060] Specifically, the new information vector obtained by inputting the time series of each state variable into the cointegration model can characterize the health status of the motor. When determining the health status of the motor based on the new information vector, this scheme can specifically determine whether the row vector of the new information vector is a stationary vector; if not, it is determined that the motor has a stator insulation fault, and if so, it is determined that the motor has no fault.
[0061] It should be noted that if the row vector of the new information vector is a non-stationary vector, it means that the new information vector contains fault characteristics, which means that the motor has a stator insulation fault; if the row vector of the new information vector is a stationary vector, it means that the new information vector does not contain fault characteristics, which means that the motor is a normal motor without any fault. In addition, after the scheme detects that the motor has a stator insulation fault, it can also generate an alarm message that the motor has a stator insulation fault. The alarm message includes the fault time, as well as the time series of each state variable, the new information vector, the cointegration model and other data. After the alarm message is generated, the staff can also be notified by means of light warnings, sound warnings, and sending the alarm message to the predetermined terminal, so as to conduct timely troubleshooting and repair the motor.
[0062] From the above, it can be seen that this scheme provides an online detection scheme for motor stator insulation fault based on cointegration theory. This scheme can be based on the cointegration model established by motor speed, voltage, current, torque, and flux, and can integrate all state parameters related to motor stator insulation to obtain a new information vector that can characterize the health state of the motor, so as to realize online detection of motor stator insulation fault and avoid the problem of inaccurate fault detection caused by extracting motor stator insulation fault characteristics through single signal processing. It can realize online monitoring of motor stator insulation fault and accurately detect the insulation state of motor stator at the same time to avoid false alarms and missed alarms. In addition, when this scheme performs online fault diagnosis, it does not require the addition of new sensors or additional modification of train equipment. It can perform online diagnosis of motor stator insulation fault based on existing motor control signals. It has the advantages of easy deployment and implementation, high real-time performance, and accurate detection to avoid false alarms and missed alarms of stator insulation faults, and has practical engineering application value.
[0063] Based on the contents described in the above embodiment, in this embodiment, before obtaining the time series of each state variable of the motor, it is also necessary to determine each state variable and the cointegration model through the following steps:
[0064] Obtain the historical time series of each initial state variable;
[0065] Perform a stationarity test on the historical time series of each initial state variable to obtain the target state variable; among which, the historical time series of each target state variable is a first-order single integral non-stationary series;
[0066] A cointegration test is performed on each target state variable to determine each state variable having a cointegration relationship and a cointegration relationship matrix of the cointegration model.
[0067] It should be noted that the primary prerequisite for using multiple state variables of the motor for cointegration modeling and realizing stator insulation fault detection is that there is a cointegration relationship between the state variables, so that a motor state monitoring model based on the cointegration relationship can be established. The cointegration relationship test first requires that the variables are non-stationary and first-order single integral, otherwise, there is no meaning of cointegration analysis between stable random variables. Even if there is cointegration between non-stationary variables with single integrals higher than the first order, stable new information variables cannot be generated. This is also meaningless for the application of the motor state monitoring model. Therefore, this embodiment first needs to determine the state variables of the input cointegration model.
[0068] Specifically, when determining the state variables of the input cointegration model, this embodiment can first select the initial state variables. The initial state variables are not necessarily first-order single integral non-stationary sequences. Therefore, this scheme can perform a stationarity test on the historical time series of each initial state variable through a unit root test, and screen out state variables that meet the stationarity test requirements. In this embodiment, the stationarity test can be performed on the historical time series of each initial state variable through the ADF (Augmented Dickey-Fuller test, unit root test) test method. For example: this scheme selects the initial state variables as the five state parameters of motor speed, voltage, current, torque, and flux. The five state parameters are interrelated. At the same time, in order to determine whether these five state parameters are first-order single integrals, it is necessary to perform a unit root test on the selected motor speed, voltage, current, torque, and flux time series. This method is a formal method for testing the stationarity of a time series. Here, the test is specifically described by taking the ADF test method as an example. Among them: the test equation of the ADF test is:
[0069] Δx t =γx t-1 +θ1Δx t-1 +θ2Δx t-2 +…+θ p-1 Δx t-p+1 +μ+e (1)
[0070] Where Δx t For the sequence x t The difference term, Δx t =x t -x t-1,γ,θ1,θ2…θ p-1 is a constant, x t-1 For the sequence x t The first-order lag term, Δx t-p+1 For the sequence x t The t-p+1 order difference lag term, μ is a constant variable, and e is a random error. In actual use, the least squares method is used to estimate formula (1), and the ADF test is to test whether x in formula (1) t-1 The coefficient γ is calculated by calculating the t statistic of γ. According to different significance levels, we test whether the t statistic of γ exceeds the ADF critical value of the significance level to judge whether x t Whether the sequence is stable. According to the ADF test method, the stability of the five time series of motor speed, voltage, current, torque and flux is tested respectively. At the same time, the ADF test is used to determine whether the five time series are first-order single integral sequences.
[0071] In this embodiment, the state variables that meet the requirements of the stationarity test screened out from the initial state variables are called target state variables, and the cointegration test is performed on each target state variable through the Johansen test method. Specifically, after the above-mentioned ADF test, it is confirmed that the five time series of the motor speed, voltage, current, torque, and flux are all first-order single integral time series. Therefore, the target state variables include: speed, voltage, current, torque, and flux, and the five target state variables are cointegrated. Test. In this embodiment, the above five motor state variables are used as input to establish a cointegration model, so it belongs to a multivariate cointegration test. The present invention uses the Johansen test to perform a multivariate cointegration test. When performing a multivariate cointegration test based on the Johansen test, the least squares method is first used to establish a VAR (Vector autoregressive model) model of motor speed, voltage, current, torque, and flux, specifically:
[0072] y t =A1y t-1 +…+A p y t-p +c+ξ t (2)
[0073] Among them, y t ,y t-1 , …, y t-p is the m×1 dimensional non-stationary first-order single integral sequence of motor speed, voltage, current, torque, and flux, y t There are p lags, p is less than or equal to 4; A1~A p is the m×m parameter matrix to be estimated, c is an (m×1) vector representing the constant term, ξt is the new information variable, and m is a constant. Using the difference relationship, subtract y from both sides of equation (2) t-1 We can get:
[0074] Distribution. There are two m×r dimensional matrices α and β such that:
[0075] Π=αβ T (4)
[0076] Where r is Γ i The rank of Π, in formula (4), the ranks of the two decomposition matrices α and β are both r. Substituting formula (4) into formula (3), we can get β T y t-1 Each row in the is a stationary sequence combination variable, that is, each row is a variable y0, y1, ..., y m-1 A linear combination with cointegration relationship. Therefore, the matrix β determines the number and form of cointegration relationship. Its rank r is the number of linearly independent cointegration vectors, and each column of it constitutes a cointegration vector. Through multivariate cointegration test, it is determined whether there is a cointegration relationship among the five time series of motor speed, voltage, current, torque, and flux linkage, and the cointegration relationship matrix β is obtained according to formula (4).
[0077] It should be noted that after the present solution determines the state variables with cointegration relationships and the cointegration relationship matrix of the cointegration model through the above process, it is also necessary to verify the cointegration model. If the verification is successful, the step of obtaining the time series of the motor state variables is executed; if the verification fails, the initial state variables are re-determined, and the step of obtaining the historical time series of each initial state variable is continued. Specifically, in this embodiment, the number of cointegration relationships calculated by the Johansen test method is 5, the cointegration relationship matrix is β, and there is a cointegration relationship between the motor speed, voltage, current, torque, and flux, so the motor speed, voltage, current, torque, and flux are all state variables. The cointegration model of these five state variables established in this embodiment is:
[0078] Θ t =β T y (5)
[0079] Where y=(y t ,y t-1 ,y t-2 ,y t-3 ,y t-4 ) are the state variables of the motor, Θ tis the new information vector obtained by the motor cointegration model. If the row vector of the matrix vector is a stationary vector, it means that the motor speed, voltage, current, torque, and flux variables meet the cointegration test, and then it can be determined that the relationship between the variables satisfies the long-term dynamic balance and has a coordinated long-term trend under normal working conditions. Therefore, this scheme selects the time series data of motor speed, voltage, current, torque, and flux in different time periods and substitutes them into formula (5) to verify the cointegration model.
[0080] If the verification fails, the process of selecting the initial state variables, stationarity test and cointegration test is re-executed. If the verification succeeds, the stator coil turn-to-turn short-circuit fault detection can be performed based on the motor cointegration model. During fault detection, if the motor turn-to-turn short-circuit obtained on-site or the speed, voltage, current, torque, and flux data obtained from the turn-to-turn short-circuit experiment are input into the motor cointegration model, the row vector of the new information vector is no longer a stationary vector, but a non-stationary vector containing the characteristics of the turn-to-turn short-circuit fault. If the speed, voltage, current, torque, and flux of the normal operation of the motor are input into the motor cointegration model, the row vector of the new information vector is a stationary vector without any fault characteristics, and it is judged to be a normal motor.
[0081] See also Figure 2 , is a flow chart of motor stator insulation fault detection based on cointegration theory disclosed in an embodiment of the present invention. It can be seen from the figure that after the scheme selects the motor speed, voltage, current, torque, and flux as the initial state variables for ADF test and Johansen test, the motor cointegration model can be constructed and tested. When performing fault detection, it is determined whether the row vector of the new information vector is a stationary vector; if it is a non-stationary vector, it is determined that the motor has a stator insulation fault; if it is a stationary vector, it is determined that the motor is normal. It should be noted that when the scheme performs stationarity detection on the motor speed, voltage, current, torque, and flux, it is not limited to the ADF test method, and the autocorrelation coefficient test method, segmented test method, run test method, etc. can also be used; when the scheme performs cointegration test on the motor speed, voltage, current, torque, and flux, it is not limited to the Johansen test, and the Engle-Granger test method, etc. can also be used.
[0082] In summary, it can be seen that the online detection technology of motor stator insulation based on cointegration theory proposed in this scheme can solve the problem of online detection of motor stator insulation faults in actual engineering, and at the same time can realize accurate detection of insulation faults to avoid false alarms and missed alarms.
[0083] The following is an introduction to the online detection device, equipment and storage medium provided in the embodiments of the present invention. The online detection device, equipment and storage medium described below can be referenced to the online detection method described above.
[0084] See also Figure 3 , a schematic structural diagram of an online detection device for motor stator insulation fault provided by an embodiment of the present invention, comprising:
[0085] A time series acquisition module 11 is used to acquire the time series of each state variable of the motor; the types of the state variables include at least two, each state variable is a first-order single integral non-stationary variable, and each state variable has a cointegration relationship;
[0086] An innovation vector determination module 12 is used to input the time series of each state variable into a cointegration model, and determine the innovation vector through the cointegration model; wherein the cointegration model is established based on the historical data of each state variable;
[0087] The state determination module 13 is used to determine the health state of the motor according to the innovation vector.
[0088] In another embodiment of the present invention, the online detection device further includes
[0089] A historical time series acquisition module is used to obtain the historical time series of each initial state variable;
[0090] The stationarity test module is used to perform a stationarity test on the historical time series of each initial state variable to obtain the target state variable; wherein the historical time series of each target state variable is a first-order single integral non-stationary series;
[0091] The cointegration test module is used to perform cointegration test on each target state variable, determine each state variable with cointegration relationship, and the cointegration relationship matrix of the cointegration model.
[0092] In another embodiment of the present invention, the stationarity test module is specifically used to: perform a stationarity test on the historical time series of each initial state variable by using an ADF test method.
[0093] In another embodiment of the present invention, the cointegration test module is specifically used to: perform cointegration test on each target state variable by using the Johansen test method.
[0094] In another embodiment of the present invention, the online detection device further includes:
[0095] The model verification module is used to verify the cointegration model; if the verification is successful, the time series acquisition module is triggered; if the verification fails, the initial state variables are re-determined and the historical time series acquisition module is triggered.
[0096] In another embodiment of the present invention, the time series acquisition module is specifically used to obtain the time series of at least two state variables of the motor: speed, voltage, current, torque, and flux.
[0097] In another embodiment of the present invention, the state determination module is specifically used to: determine whether the new information vector is a stable vector; if not, determine that the motor has a stator insulation fault.
[0098] In another embodiment of the present invention, the online detection device further includes:
[0099] The alarm information generating module is used to generate alarm information indicating that the motor has a stator insulation fault when the motor has a stator insulation fault.
[0100] In another embodiment of the present invention, the innovation vector determination module is specifically used to: determine the innovation vector through the cointegration model, and the cointegration model is Θ t =β T y; where Θ t is the new information vector, β is the cointegration relationship matrix, and y is the time series of each state variable.
[0101] See also Figure 4 , a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, the device includes:
[0102] A memory 21, used for storing computer programs;
[0103] The processor 22 is used to implement the steps of the online detection method of motor stator insulation fault described in any of the above method embodiments when executing the computer program.
[0104] In this embodiment, the device may be a PC (Personal Computer), or may be a terminal device such as a smart phone, a tablet computer, a PDA, or a portable computer.
[0105] The device may include a memory 21 , a processor 22 , and a bus 23 .
[0106] Among them, the memory 21 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. The memory 21 can be an internal storage unit of the device in some embodiments, such as a hard disk of the device. The memory 21 can also be an external storage device of the device in other embodiments, such as a plug-in hard disk equipped on the device, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Further, the memory 21 can also include both an internal storage unit of the device and an external storage device. The memory 21 can not only be used to store application software and various types of data installed in the device, such as program code for executing an online detection method, but also can be used to temporarily store data that has been output or is to be output.
[0107] In some embodiments, the processor 22 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor or other data processing chip, used to run program codes or process data stored in the memory 21, such as program codes for executing online detection methods.
[0108] The bus 23 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.
[0109] Furthermore, the device may also include a network interface 24, which may optionally include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the device and other electronic devices.
[0110] Optionally, the device may further include a user interface 25, which may include a display (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 25 may also include a standard wired interface and a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device, etc. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the device and to display a visual user interface.
[0111] Figure 4 Only the device with components 21-25 is shown, and it can be understood by those skilled in the art that Figure 4 The structure shown does not constitute a limitation of the device, and may include fewer or more components than shown, or combine certain components, or arrange the components differently.
[0112] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the online detection method for motor stator insulation fault described in any of the above method embodiments are implemented.
[0113] The storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.
[0114] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.
[0115] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. An online detection method for motor stator insulation fault, characterized in that: include: Acquire the time series of each state variable of the motor; the types of the state variables include at least two, each state variable is a first-order single integral non-stationary variable, and each state variable has a cointegration relationship; Input the time series of each state variable into the cointegration model, and determine the innovation vector through the cointegration model; wherein the cointegration model is established based on the historical data of each state variable; A health state of the motor is determined based on the innovation vector.
2. The online detection method according to claim 1, characterized in that: Before obtaining the time series of each state variable of the motor, the method further includes: Obtain the historical time series of each initial state variable; Perform a stationarity test on the historical time series of each initial state variable to obtain the target state variable; among which, the historical time series of each target state variable is a first-order single integral non-stationary series; A cointegration test is performed on each target state variable to determine each state variable having a cointegration relationship and a cointegration relationship matrix of the cointegration model.
3. The online detection method according to claim 2, characterized in that: The stationarity test of the historical time series of each initial state variable includes: The ADF test method is used to test the stationarity of the historical time series of each initial state variable.
4. The online detection method according to claim 2, characterized in that: The cointegration test for each target state variable includes: The cointegration test of each target state variable is carried out using the Johansen test method.
5. The online detection method according to claim 2, characterized in that: After determining each state variable having a cointegration relationship and the cointegration relationship matrix of the cointegration model, the method further includes: To verify the cointegration model; If the verification is successful, the step of obtaining the time series of each state variable of the motor is executed; If the verification fails, the initial state variables are re-determined, and the step of obtaining the historical time series of each initial state variable is continued.
6. The online detection method according to claim 1, characterized in that: The step of obtaining the time series of each state variable of the motor includes: Obtain a time series of at least two state variables of the motor, including the speed, voltage, current, torque, and flux.
7. The online detection method according to claim 1, characterized in that: Determining the health status of the motor according to the innovation vector comprises: Determine whether the row vector of the innovation vector is a stationary vector; If not, it is determined that the motor has a stator insulation fault.
8. The online detection method according to claim 1, characterized in that: If the motor has a stator insulation fault, the online detection method further includes: Generate an alarm message indicating that the motor has a stator insulation fault.
9. The online detection method according to any one of claims 1 to 8, characterized in that: Determining the innovation vector by the cointegration model includes: The innovation vector is determined by the cointegration model, and the cointegration model is Θ t =β T y; where Θ t is the new information vector, β is the cointegration relationship matrix, y is the time series of each state variable, and T represents the matrix transpose.
10. An online detection device for motor stator insulation fault, characterized in that: include: A time series acquisition module is used to acquire the time series of each state variable of the motor; the types of the state variables include at least two, each state variable is a first-order single integral non-stationary variable, and each state variable has a cointegration relationship; An innovation vector determination module is used to input the time series of each state variable into a cointegration model and determine the innovation vector through the cointegration model; wherein the cointegration model is established based on the historical data of each state variable; A state determination module is used to determine the health state of the motor according to the new information vector.
11. The online detection device according to claim 10, characterized in that: The online detection device also includes A historical time series acquisition module is used to obtain the historical time series of each initial state variable; The stationarity test module is used to perform a stationarity test on the historical time series of each initial state variable to obtain the target state variable; wherein the historical time series of each target state variable is a first-order single integral non-stationary series; The cointegration test module is used to perform cointegration test on each target state variable, determine each state variable with cointegration relationship, and the cointegration relationship matrix of the cointegration model.
12. The online detection device according to claim 10, characterized in that: The online detection device also includes: The model verification module is used to verify the cointegration model; if the verification is successful, the time series acquisition module is triggered; if the verification fails, the initial state variables are re-determined and the historical time series acquisition module is triggered.
13. The online detection method according to claim 10, characterized in that: The time series acquisition module is specifically used to obtain the time series of at least two state variables of the motor, including the speed, voltage, current, torque, and flux.
14. An electronic device, characterized in that: include: Memory for storing computer programs; A processor is used to implement the steps of the online detection method for motor stator insulation fault as claimed in any one of claims 1 to 9 when executing the computer program.
15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the online detection method for motor stator insulation fault according to any one of claims 1 to 9 are implemented.
Citation Information
Patent Citations
Cutter abrasion monitoring method based on co-integration modeling
CN102152172A
Online monitoring method and device for insulation of three-phase asynchronous alternating current motor stator
CN106841949A
Method and device for processing multivariate time series data
CN107577648A
Cointegration-analysis-based wind power generator gear case condition monitoring method
CN109356798A
