CN110907864B

CN110907864B - Fault detection method, device and equipment for motor stator winding and storage medium

Info

Publication number: CN110907864B
Application number: CN201911368875.6A
Authority: CN
Inventors: 李鲲鹏; 李雅婧
Original assignee: Beijing Haopeng Intelligent Technology Co ltd
Current assignee: Beijing Haopeng Intelligent Technology Co ltd
Priority date: 2019-12-26
Filing date: 2019-12-26
Publication date: 2021-12-28
Anticipated expiration: 2039-12-26
Also published as: CN110907864A

Abstract

The embodiment of the application provides a fault detection method, a fault detection device, fault detection equipment and a storage medium for a motor stator winding. The method comprises the following steps: acquiring three-phase current signals and three-phase voltage signals of a stator winding; determining characteristic parameters of a plurality of different characteristics of the stator winding based on the three-phase current signals and the three-phase voltage signals; inputting characteristic parameters of a plurality of different characteristics of the stator winding into a preset fault detection model so as to detect whether the motor is in fault or not through the preset fault detection model; the preset fault detection model is obtained by joint training based on feature samples of a plurality of different features of the stator winding. The fault detection accuracy of the motor stator winding can be improved.

Description

Fault detection method, device and equipment for motor stator winding and storage medium

Technical Field

The embodiment of the application relates to the technical field of mechanical fault detection, in particular to a fault detection method, a fault detection device, fault detection equipment and a storage medium for a motor stator winding.

Background

The motor is a main way for converting electric energy into mechanical energy, is one of the most important electromechanical devices at present, and is particularly important in various fields such as aerospace, transportation, industrial and agricultural production, robots, daily life and the like. With the wide use of motors, it is more and more important to ensure the safe and efficient operation of the motors. The motor is in a working environment with high voltage, high rotating speed, strong magnetic field, high temperature and high humidity for a long time, so that the motor is easy to break down.

Short circuit of stator coil winding of induction motor is a common electrical fault of motor, and if fault finding and processing are not timely, the motor is damaged, reliable operation of the whole electromechanical system is affected, and even safety accidents are caused. Therefore, early failure detection of the motor stator is a difficulty in motor failure diagnosis. At present, for the detection of the early faults of the stator of the motor, the detection is mainly focused on measuring and analyzing the fault characteristic conditions presented by single characteristic changes of the electrical parameters of the induction motor, and the comprehensive characteristics presented by a plurality of different characteristics or different signal spaces of the electrical parameters are not considered, so that the early detection of the faults of the stator winding is still not timely, and the accuracy of the detection result is not high.

Disclosure of Invention

The embodiment of the application provides a fault detection method, a fault detection device, equipment and a storage medium for a motor stator winding, so that the fault detection accuracy of the motor stator winding is improved.

In a first aspect, an embodiment of the present application provides a method for detecting a fault of a stator winding of an electric machine, including: acquiring three-phase current signals and three-phase voltage signals of the stator winding; determining characteristic parameters of a plurality of different characteristics of the stator winding based on the three-phase current signals and the three-phase voltage signals; inputting characteristic parameters of a plurality of different characteristics of the stator winding into a preset fault detection model so as to detect whether the motor is in fault or not through the preset fault detection model; the preset fault detection model is obtained by joint training based on feature samples of a plurality of different features of the stator winding.

In a second aspect, an embodiment of the present application provides a fault detection apparatus for a stator winding of an electric machine, including: the acquisition module is used for acquiring three-phase current signals and three-phase voltage signals of the stator winding; a determination module for determining characteristic parameters of a plurality of different characteristics of the stator winding based on the three-phase current signals and the three-phase voltage signals; the detection module is used for inputting the characteristic parameters of a plurality of different characteristics of the stator winding into a preset fault detection model so as to detect whether the motor is in fault or not through the preset fault detection model; the preset fault detection model is obtained by joint training based on feature samples of a plurality of different features of the stator winding.

In a third aspect, an embodiment of the present application provides a fault detection apparatus for a stator winding of an electric machine, including:

a memory;

a processor; and

a computer program;

wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method of the first aspect.

In a fourth aspect, embodiments of the present application provide a computer-readable storage medium on which a computer program is stored, the computer program being executed by a processor to implement the method of the first aspect.

According to the fault detection method, the fault detection device, the fault detection equipment and the fault detection storage medium for the motor stator winding, three-phase current signals and three-phase voltage signals of the motor stator winding are obtained; determining characteristic parameters of a plurality of different characteristics of the stator winding based on the three-phase current signals and the three-phase voltage signals; inputting characteristic parameters of a plurality of different characteristics of the stator winding into a preset fault detection model so as to detect whether the motor is in fault or not through the preset fault detection model; and the preset fault detection model is obtained by joint training based on the feature samples of a plurality of different features of the stator winding. Because a plurality of different characteristic parameters of the motor are extracted according to the three-phase current signals and the three-phase voltage signals of the motor stator winding, the extracted characteristic parameters of the motor stator winding are more diversified, so that a preset fault detection model can comprehensively detect whether the motor stator winding is in fault or not according to the diversified parameters, and the effect of improving the fault detection accuracy of the motor stator winding is achieved.

Drawings

Fig. 1 is a schematic structural diagram of a fault detection system for a stator winding of a motor according to an embodiment of the present disclosure;

fig. 2 is a schematic structural diagram of a data acquisition unit according to an embodiment of the present application;

fig. 3 is a flowchart of a method for detecting a fault of a stator winding of a motor according to an embodiment of the present disclosure;

fig. 4 is a flowchart of a method for detecting a fault in a stator winding of an electric machine according to another embodiment of the present application;

fig. 5 is a flowchart of a method for detecting a fault in a stator winding of an electric machine according to another embodiment of the present application;

fig. 6 is a flowchart of a method for detecting a fault in a stator winding of an electric machine according to another embodiment of the present application;

fig. 7 is a flowchart of a method for detecting a fault in a stator winding of an electric machine according to another embodiment of the present application;

fig. 8 is a flowchart of a method for detecting a fault in a stator winding of an electric machine according to another embodiment of the present application;

fig. 9 is a schematic structural diagram of a fault detection device for a stator winding of a motor according to an embodiment of the present application;

fig. 10 is a schematic structural diagram of a fault detection apparatus for a stator winding of a motor according to an embodiment of the present application.

With the foregoing drawings in mind, certain embodiments of the disclosure have been shown and described in more detail below. These drawings and written description are not intended to limit the scope of the disclosed concepts in any way, but rather to illustrate the concepts of the disclosure to those skilled in the art by reference to specific embodiments.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

Fig. 1 is a schematic structural diagram of a fault detection system for a stator winding of a motor according to an embodiment of the present disclosure. As shown in fig. 1, the fault detection system for stator windings of an electric machine comprises: the system comprises a plurality of voltage sensors 11, a plurality of current sensors 12, a plurality of data acquisition units 13, a wireless communication gateway 14, an edge calculation diagnosis unit 15, a data mining and model training unit 16 and a detection result presenting and man-machine interaction unit 17. Optionally, the number of the voltage sensors 11 is at least 3, and the number of the current sensors 12 is also at least 3.

In this embodiment, the motor may be various types of motors, taking an electrically driven compressor asynchronous driving motor as an example, 3 voltage sensors are installed at a three-phase output end of a motor power supply frequency converter, each voltage sensor 11 acquires a phase voltage of a motor stator winding, and the 3 voltage sensors 11 collectively acquire three-phase voltage signals of the motor stator winding. Similarly, 3 current sensors are respectively installed on three-phase power supply circuits of the motor stator, for example, one current sensor is installed on each power supply circuit, each current sensor 12 acquires one-phase current of the motor stator winding, and the 3 current sensors 12 collectively acquire three-phase current signals of the motor stator winding. Wherein, three-phase voltage signal and three-phase current signal need synchronous acquisition.

3 voltage sensor or 3 current sensor are connected to a data acquisition unit 13 respectively, and data acquisition unit 13 is used for storing and data preprocessing the three-phase voltage signal or the three-phase current signal that 3 voltage sensor or 3 current sensor gathered to and the clock control of 3 voltage sensor or 3 current sensor synchronous acquisition data. Wherein, 3 voltage sensor can detect the voltage signal of a plurality of motors, and 3 current sensor can detect the current signal of a motor.

It should be noted that the number of the data acquisition units in fig. 1 is only an exemplary illustration, and is not limited to this embodiment, and in practical applications, the number of the voltage sensors, the number of the current sensors, and the number of the data acquisition units may be set according to practical situations.

The present embodiment will be described in detail below by taking the structure shown in fig. 1 as an example:

the data acquisition unit 13 is connected to the edge calculation unit 15 through the wireless communication gateway 14, and is configured to send the preprocessed three-phase voltage signal and the preprocessed three-phase current signal to the edge calculation unit 15.

And the edge calculation unit 15 is used for determining a plurality of characteristic parameters of the stator winding according to the preprocessed three-phase voltage signals and the preprocessed three-phase current signals. The edge calculating unit 15 is connected to the data mining and model training unit 16, and is further configured to input a plurality of characteristic parameters into the data mining and model training unit 16, so as to identify whether a fault, such as a short circuit, occurs in the motor stator winding through the data mining and model training unit 16. The data mining and model training unit 16 needs to train through a large number of characteristic parameter samples, and can be used for accurately identifying the motor fault.

And the data mining and model training unit 16 is connected to the detection result presenting and human-computer interaction unit 17 and is used for sending the identification result to the detection result presenting and human-computer interaction unit 17 so as to present the identification result on the detection result presenting and human-computer interaction unit 17, and the detection result presenting and human-computer interaction unit 17 can also provide a human-computer interaction function.

Alternatively, the voltage sensor 11 may be a voltage transformer, and the current sensor 12 may be a hall current sensor.

Optionally, the process of extracting the feature parameters based on the three-phase voltage signals and the three-phase current signals may be completed in the edge calculation diagnosis unit 15, or may be completed in the data mining and model training unit 16, that is, the extraction of the feature parameters and the identification based on the feature parameters may be completed in the data mining and model training unit 16 at the same time, or may be completed in the edge calculation diagnosis unit 15 and the data mining and model training unit 16, respectively, which is not specifically limited in this embodiment.

Optionally, the data mining and model training unit 16 may be disposed in the cloud, or may not be disposed in the cloud, for example, locally, this embodiment is not specifically limited to this.

Fig. 2 is a schematic structural diagram of a data acquisition unit according to an embodiment of the present application. As shown in fig. 2, the data acquisition unit 12 includes: a plurality of sensors 20, a channel selection and signal conditioning module 21, an Analog-to-Digital converter (a/D) module 22, a collection clock generation module 23, a first Micro Control Unit (MCU) 24, a second Micro Control Unit (MCU) 25, a communication control module 26, and a wireless communication module 27; the path selection and signal conditioning module 21 may be connected to 6 sensors 20 (including 3 voltage sensors and 3 current sensors), and support voltage input and current input of 4-20 mA. The plurality of sensors 20 includes a voltage sensor and a current sensor.

The conditioning circuit in the channel selection and signal conditioning module 21 supports the functions of adaptive signal amplification and anti-aliasing filtering, the amplification factor is between 2 and 1000 times, the amplification factor is controlled by the MCU, the parameters of the filter can be controlled by the MCU, and the bandwidth of the filter can be selected from the interval of 5KHz to 100 KHz.

The A/D module 22 supports three single-channel devices with a sampling rate of 200Kbps at most, the sampling resolution reaches 16 bits/sampling point, and the sampling rate can be selected from 200Hz to 200KHz under the control of the MCU.

Optionally, three MCUs may be arranged in one data acquisition unit, wherein one of the MCUs serves as a main MCU, that is, a first micro control unit; the other two MCUs act as slave MCUs, i.e. second micro control units. Of course, the number of slave MCUs may not be 2, and those skilled in the art may arrange the number of slave MCUs according to actual requirements. Optionally, one a/D conversion module corresponds to one slave MCU, and one slave MCU may correspond to a plurality of a/D conversion modules, which is not specifically limited in this embodiment.

The following describes the technical solution of the present application and how to solve the above technical problem in a specific embodiment with reference to fig. 1 and fig. 2. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.

Fig. 3 is a flowchart of a method for detecting a fault of a stator winding of a motor according to an embodiment of the present disclosure. The embodiment of the application provides a fault detection method for a motor stator winding aiming at the above technical problems in the prior art, and the method comprises the following specific steps:

and 301, acquiring three-phase current signals and three-phase voltage signals of the stator winding.

Specifically, as shown in fig. 1, a data acquisition unit may acquire three-phase voltage signals from detection results of voltage sensors installed on a frequency converter of a power supply of the motor, and acquire three-phase current signals from detection results of current sensors installed on a three-phase power supply line of the motor.

Optionally, the data acquisition unit may acquire the detection result data of the voltage sensor and the current sensor synchronously with a sampling rate of 20Kbps and a sampling resolution of 16 bits/sampling point, so as to obtain three-phase current signals and three-phase voltage signals of the motor stator winding at multiple times, and obtain a time sequence of the three-phase current signals and a time sequence of the three-phase voltage signals. Three-phase current signals are respectively represented as i_a(n)，i_b(n)，i_c(N), N is 0, …, N-1, and the three-phase voltage signals are respectively represented as v_a(n)，v_b(N), v (N), N ═ 0, …, N-1, where N denotes the sampling time.

Step 302, determining characteristic parameters of a plurality of different characteristics of the stator winding based on the three-phase current signals and the three-phase voltage signals.

In this embodiment, the characteristic parameters of the plurality of different characteristics include: the park transformation vector mode signal comprises at least two items of frequency domain characteristics, voltage unbalance factors, negative sequence current change rate, impedance change rate and instantaneous positive sequence power. Accordingly, step 302 includes: and determining at least two items of frequency domain characteristics, voltage unbalance factors, negative sequence current change rates, impedance change rates and instantaneous positive sequence power of park transformation vector mode signals of the stator winding based on the three-phase current signals and the three-phase voltage signals.

The frequency domain characteristics of the park transformation vector mode signals are obtained by park transformation of the three-phase current signals to obtain park transformation vectors, and then the frequency domain characteristics of the three-phase current signals are extracted based on the park transformation vectors.

And 303, inputting the characteristic parameters of the plurality of different characteristics of the stator winding into a preset fault detection model so as to detect whether the motor is in fault or not through the preset fault detection model.

Optionally, step 303 includes: inputting the frequency domain characteristics, the voltage unbalance factor, the negative sequence current change rate, the impedance change rate and the instantaneous positive sequence power of the park transformation vector mode signal into a preset fault detection model so as to detect whether the motor is in fault or not through the fault detection model; the preset fault detection model is obtained by training based on a feature sample, a voltage unbalance factor sample, a negative sequence current change rate sample, an impedance change rate sample and an instantaneous positive sequence power sample of a park transform domain.

Specifically, whether the stator winding is short-circuited can be determined based on a characteristic parameter of one of a plurality of different characteristics. In some cases, the motor stator winding is not detected to have a fault according to the characteristic parameters of one characteristic, but the motor stator winding is comprehensively detected by combining other characteristics or the running state of the motor is indirectly reflected through other characteristics, so that whether the motor stator winding has the fault or not can be detected more accurately. In addition, the characteristics may affect each other, or there is a correlation between the characteristics, a change in one characteristic often causes a change in another characteristic, and the changes in the characteristics may have a chronological relationship, so that the motor may be indirectly reflected to be faulty through some characteristics, and a warning function is provided for motor fault detection, which is beneficial to early fault detection and improves accuracy and reliability of detection results.

The preset fault detection model is obtained by joint training based on feature samples of a plurality of different features of the stator winding. In this embodiment, the preset fault detection model performs fault detection according to a plurality of characteristic parameters of different characteristics, and in an optional implementation manner, the preset fault detection model sets a weight coefficient for the characteristic parameter of each characteristic, and when performing joint detection on the frequency domain characteristic, the voltage imbalance factor, the negative sequence current change rate, the impedance change rate, and the instantaneous positive sequence power based on the park transformation vector mode signal, each characteristic parameter is multiplied by a corresponding weight coefficient, and then each product is summed, that is, a plurality of different characteristic parameters are weighted and summed to obtain a comprehensive value, and whether a motor stator winding has a fault or not is identified according to the comprehensive value. For example, assume the frequency domain characteristics, voltage imbalance factor, negative sequence of park transform vector mode signalThe current change rate, the impedance change rate and the instantaneous positive sequence power are I, K and delta I, Z respectively_npP, the weight coefficient of each item is w1, w2, w3, w4 and w5, and the preset fault detection model transforms the frequency domain characteristic I of the vector mode signal, the voltage unbalance factor K, the negative sequence current change rate delta I and the impedance change rate Z according to the park, so that the fault detection model can detect the fault of the load_npAnd the instantaneous positive sequence power P output comprehensive value is as follows: i w1+ K w2+ Δ I w3+ Z_np*w4+P*w5。

According to the embodiment of the application, three-phase current signals and three-phase voltage signals of a stator winding of a motor are obtained; determining characteristic parameters of a plurality of different characteristics of the stator winding based on the three-phase current signals and the three-phase voltage signals; inputting characteristic parameters of a plurality of different characteristics of the stator winding into a preset fault detection model so as to detect whether the motor is in fault or not through the preset fault detection model; and the preset fault detection model is obtained by joint training based on the feature samples of a plurality of different features of the stator winding. Because a plurality of different characteristic parameters of the motor are extracted according to the three-phase current signals and the three-phase voltage signals of the motor stator winding, the extracted characteristic parameters of the motor stator winding are more diversified, so that a preset fault detection model can comprehensively detect whether the motor stator winding is in fault or not according to the diversified parameters, and the effect of improving the fault detection accuracy of the motor stator winding is achieved.

Optionally, the preset fault detection model is obtained by training a neural network; as shown in fig. 4, the training of the neural network includes the following steps:

step 401, obtaining three-phase voltage signal samples and three-phase current signal samples of a motor stator winding, and marking information.

The marking information is used for indicating the state of the motor, such as whether the stator winding of the motor is in failure.

Step 402, determining characteristic sample parameters of a plurality of different characteristics of the stator winding based on the three-phase voltage signal samples and the three-phase current signal samples.

Specifically, at least two items of a frequency domain characteristic sample, a voltage unbalance factor sample, a negative sequence current change rate sample, an impedance change rate sample and an instantaneous positive sequence power sample of park transformation vector mode signals of the motor electronic winding are obtained.

Step 402, training the neural network based on the characteristic sample parameters of a plurality of different characteristics of the stator winding and the marking information.

Optionally, in the process of training the neural network based on the characteristic sample parameters of the plurality of different characteristics of the stator winding and the label information, in addition to initializing the connection weights between the neural network layers, a weight coefficient value of each characteristic sample parameter may also be initialized, the weight coefficient value of each initial characteristic sample parameter may be set according to an empirical value, the initial weight coefficient value is selected from 0 to 1, and the initial weight coefficients of all the characteristic sample parameters are added to 1. The neural network comprises a plurality of neural network layers, the plurality of neural network layers are connected with one another, the connection weight between the neural network layers refers to the connection weight between the mutually connected neural network layers, for example, the neural network comprises an input layer, a hidden layer and an output layer, and the connection weight is arranged between the input layer and the hidden layer.

And 403, adjusting network parameters of the neural network based on the difference between the output result of the neural network and the labeling information.

Specifically, adjusting network parameters of the neural network based on a difference between an output result of the neural network and the labeling information includes: and adjusting the connection weight between the neural network layers and the weight coefficient of each characteristic parameter based on the difference between the output result of the neural network and the labeling information.

In this embodiment, a neural network including an input layer, q hidden layers, and an output layer may be established, then a training sample set formed by at least two of a frequency domain feature sample, a voltage imbalance factor sample, a negative sequence current change rate sample, an impedance change rate sample, and an instantaneous positive sequence power sample of a park transformation vector mode signal is used to train the neural network, and the neural network is tested by using a test sample, and network parameters of the neural network are adjusted to obtain appropriate network parameters, and finally the trained neural network is stored to obtain a fault detection model.

Specifically, the above process includes:

establishing a neural network of an input layer, q hidden layers and an output layer, and initializing the neural network.

Initializing the neural network means randomly selecting a value between 0 and 1 as a connection weight between the input layer and the hidden layer, and between the hidden layer and the input layer.

In this embodiment, the neural network may select an adaptive noise reduction encoder, which includes 1 input layer, 3 hidden layers, and 1 output layer; the input layer consists of 50 neurons, each hidden layer is provided with 100 nerve units, the hidden layers are in full connection with the hidden layers, the output layer is provided with 24 neurons, and the hidden layers are in full connection with the output layer.

And secondly, calculating the output of each processing unit of the hidden layer. The output of each processing unit of the hidden layer is calculated by adopting the following formula:

in the formula, mu_l，jThe output of the jth neuron representing the ith hidden layer, ω_l，j，mIs the weight value of the mth neuron of the l-1 input layer and the jth neuron of the l hidden layer; mu.s_l-1，mIs the output value of the mth neuron of the l-1 neural network layer (comprising an input layer and a hidden layer); n is a radical of_l-1Is the number of l-1 neural network layer neurons, and f represents the functional relationship of the hidden layer.

And thirdly, calculating the output of each neuron of the output layer.

Specifically, the output of each neuron of the output layer is calculated by the following formula:

in the formula, mu_L，jIndicating output layer L layerOutput of the jth neuron, ω_L，j，mThe weight value of j-th neuron in the L layer of the output layer is associated with m-th neuron in the L-1 layer of the hidden layer_l，mIs the output value of the mth neuron of the mth hidden layer; n is a radical of_lThe number of hidden layer neurons.

Fourthly, adjusting the weight among the hidden layer, the input layer and the output layer.

Specifically, the adjustment of the weight between the hidden layer and the input layer and the output layer includes:

in the formula, ω_l，j，i(k +1) is a weight after the connection weight between the jth neuron of the ith neural network layer and the ith neuron of the (l-1) th neural network layer is adjusted for the (k +1) th time; likewise, ω_l，j，i(k) The weight value after the connection weight value between the jth neuron of the ith neural network layer and the ith neuron of the (l-1) th neural network layer is adjusted for the kth time; u is a positive constant, called the learning rate, to adjust the learning step size.

Fifthly, repeating the steps from the second step to the fourth step until the error correction value is reached.

Whether the motor stator winding is short-circuited or not is detected through the fault detection model, and the accuracy and the reliability of the detection result are higher because the preset fault detection model is obtained by training a large amount of sample data based on a plurality of different characteristics of the stator winding.

Fig. 5 is a flowchart of a method for detecting a fault of a stator winding of an electric machine according to another embodiment of the present application. As shown in fig. 5, the frequency domain characteristic of the park transformation vector mode signal is obtained according to the following method steps, or the frequency domain characteristic of the park transformation vector mode signal of the stator winding is determined based on the three-phase current signal and the three-phase voltage signal, and the method includes:

and step 501, acquiring the power frequency of the stator winding.

Expressing the supply frequency as f_e。

Step 502, performing park transformation on the three-phase current signals to obtain park transformation vectors.

In this embodiment, the three-phase current signals are three-phase current signals in the abc coordinate system of the three-phase stationary coordinate system. Step 502 is to convert the three-phase current signals in the three-phase stationary coordinate system to the rotating coordinate system dq 0.

Alternatively, step 502 may be implemented by the following equations (1) and (2):

in the formula (1), the reaction mixture is,

in the formula (1), i_d(n) represents the coordinate values on the d-axis after the three-phase current signals at time n are converted to the rotating coordinate system dq 0; likewise, i_q(n) represents the coordinate value on the q-axis after the three-phase current signal at time n is converted to the dq0 coordinate system. γ represents the angular displacement between the rotating coordinate system dq0 and the stationary coordinate system α β 0, which varies with time, cos γ represents the cosine value of the angular displacement; sin γ denotes the sine value of the angular displacement.

In the formula (2), i_α(n) a coordinate value on an α axis after the three-phase current signal at the time n is converted into an α β 0 coordinate system; likewise, i_q(n) represents a coordinate value on the β axis after the three-phase current signal at the time n is converted into the α β 0 coordinate system. i.e. i_a(n) represents the phase-a current signal at n moments in a three-phase coordinate system abc; i.e. i_b(n) represents the phase b current signal at n moments in a three-phase coordinate system abc; i.e. i_cAnd (n) represents the c-phase current signal at n moments in a three-phase coordinate system abc. It is noted that N represents the sampling time, and N is a variable and takes a value from 0 to N-1.

Step 503, performing modulus and square on the park transformation vector to obtain a square value of the park transformation vector modulus.

Wherein the square value of the park transformation vector mode is i_s(n)，i_s(n) can be expressed as the following formula (3):

in the formula (3), j²＝-1，|i_d(n)+ji_q(n)|²Representing the modulo operation on the park transformation vector.

And step 504, calculating the average value of the square value of the park transformation vector modulus.

Wherein, the square value i of the park transformation vector mode_sThe average value of (n) can be realized by the following formula (4):

in the formula (4), the reaction mixture is,

the mean value of the square value of the vector modulus of the park transformation is calculated; n represents the number of samples, and is also the number of sampling instants.

And 505, calculating the amplitude of the power frequency of the preset multiple based on the result of Fourier transform on the average value.

In this step, the average value is calculated

Performing Fourier transform or Fast Fourier Transform (FFT) to obtain Fourier transform result I_s(k)。

Thereafter, use is made of I_s(k) And solving the amplitude of the preset multiple power supply frequency. Optionally, the preset multiple may be 2 times. For example, by means of I_s(k) Finding the amplitude of 2 times of power frequency, 2 times of power frequency 2f_eIs passed through by I_s(2f_e) I is calculated, specifically, firstly, Fourier transform result I is obtained_s(k) K is the variable inChange to 2f_eThen to I_s(2f_e) Performing modulo operation to obtain 2 times of power frequency 2f_eThe amplitude of (c).

Step 506, calculating the ratio of the amplitude of the power supply frequency of the preset multiple to the average value to obtain the frequency domain characteristic of the park transformation vector mode signal.

In this step, 2 times of power frequency 2f is calculated by taking 2 times of power frequency as an example_eAmplitude of | I_s(2f_e) Average of squared values of | and park transformation vector mode signals

Ratio of

Obtaining the frequency domain characteristics of park transformation vector mode signals

The amplitude of the component of the power supply fundamental frequency (power supply frequency) is 2 times, and whether the running state of the motor stator winding is normal or not can be determined according to the amplitude of the component of the power supply fundamental frequency 2 times. In general, the amplitude of the 2-fold component of the fundamental frequency of the power supply is smaller than the preset value. Thus, can be obtained by

Comparing with a preset value if

If the value is larger than or equal to the preset value, determining that the motor stator winding has a short-circuit fault; if it is

Less than the predetermined value, the motor stator winding is generally considered normal, and the embodiments of the present application may be in

And under the condition that the short circuit is smaller than the preset value, comprehensively determining whether the motor stator winding is short-circuited or not by combining other characteristic parameters of the motor stator winding.Can also be at

And when the short circuit is larger than or equal to the preset value, comprehensively determining whether the motor stator winding is short-circuited or not by combining other characteristic parameters of the motor stator winding.

In this embodiment, after the three-phase current signal obtained by the data acquisition unit is subjected to park transformation, the ratio of the amplitude of the power frequency to the average value of the preset multiple can form a park transformation vector diagram. Specifically, when the motor is normal, the park transformation vector diagram becomes a regular circle, and when the motor fails, such as a stator winding short circuit, the park transformation vector diagram becomes an irregular circle, such as an ellipse.

Fig. 6 is a flowchart of a method for detecting a fault of a stator winding of an electric machine according to another embodiment of the present application. As shown in fig. 6, the voltage imbalance factor is obtained according to the following method steps, or the voltage imbalance factor of the stator winding is determined based on the three-phase current signals and the three-phase voltage signals, and the method includes:

and 601, respectively determining a voltage positive sequence component and a voltage negative sequence component of the three-phase voltage signal.

Optionally, the step 601 of determining the voltage positive sequence component and the voltage negative sequence component of the three-phase voltage signal respectively includes:

step 601a, calculating three-phase equivalent power supply voltage based on the three-phase voltage signals.

Specifically, step 601a may calculate the three-phase equivalent power voltage by the following formula (5):

in the formula (5), v_a，v_b，v_cThe voltage signals of the three-phase voltage signals are respectively on a phase, a phase and a phase c phase; v. of_a(n)，v_b(n)，v_c(n) are voltage signals on three phases a, b and c at the moment n respectively; u. of_a、u_bAnd u_cRespectively representing equivalent voltage values of three-phase voltage signals on a phase a, a phase b and a phase c，u_a(n)，u_b(n)，u_cAnd (N) represents the three-phase equivalent power supply voltage at the time N, wherein N is 0, 1, … and N-1.

Step 601b, converting the three-phase equivalent power supply voltage to a synchronous rotating speed coordinate system.

And 601c, calculating a voltage positive sequence component and a voltage negative sequence component under the synchronous rotating speed coordinate system.

Steps 601b to 601c are to convert the equivalent power voltage in the abc coordinate system to the stationary α β coordinate system, and then calculate the voltage positive sequence component and the voltage negative sequence component in the stationary α β coordinate system. Specifically, the conversion can be performed by the following equation (6):

in the formula (6), H is Hilbert transform Hilbert operator, wherein

Respectively are voltage positive sequence components on an alpha axis and a beta axis under an alpha-beta coordinate system,

the negative sequence components of the voltage on the alpha axis and the beta axis under the alpha-beta coordinate system are respectively.

And 601d, calculating a voltage positive sequence component and a voltage negative sequence component under a three-phase coordinate system based on the voltage positive sequence component and the voltage negative sequence component under the synchronous rotating speed coordinate system.

The step is that

Calculating the positive sequence component of the equivalent power supply voltage in a three-phase coordinate system

Specifically, the following can be mentionedEquation (7) to calculate:

and, from

Calculating the negative sequence component of the equivalent power supply voltage in a three-phase coordinate system

Specifically, it can be calculated by the following formula (8):

step 602, determining a voltage imbalance factor based on the voltage negative sequence component and the voltage positive sequence component.

Optionally, step 602 determines a voltage imbalance factor based on the voltage negative sequence component and the voltage positive sequence component, and includes:

step 602a, calculating an average value of the effective values of the voltage positive sequence components of the three phases and an average value of the effective values of the voltage negative sequence components of the three phases.

The average value of the effective values of the voltage positive sequence components of the three phases and the average value of the effective values of the voltage negative sequence components of the three phases refer to the average values of the effective values of the voltage positive sequence components and the voltage negative sequence components of the three phases in the abc coordinate system of the three-phase stationary coordinate system. Specifically, effective values of a voltage positive sequence component and a voltage negative sequence component of each of the three phases a, b and c are calculated firstly, then the voltage positive sequence components of the three phases are averaged, and the voltage negative sequence components of the three phases are averaged.

The average value of the effective values of the voltage positive sequence components of the three phases and the average value of the effective values of the voltage negative sequence components of the three phases are respectively represented as V_pAnd V_n，V_pAnd V_nCan be calculated by the following formula (9) and formula (10)Obtaining:

step 602b, determining a voltage unbalance factor according to the ratio of the average value of the effective values of the voltage negative sequence components of the three phases and the average value of the effective values of the voltage positive sequence components of the three phases.

Specifically, the voltage imbalance factor is K,

the negative sequence current change rate is obtained according to the following method steps: converting the three-phase current signals to an inverse synchronous rotating speed coordinate system to obtain current signals in the inverse synchronous rotating speed coordinate system; determining the negative sequence current of the current signal under the inverse synchronous rotating speed coordinate system; determining a negative sequence current under a three-phase coordinate system based on the negative sequence current of the current signal under the inverse synchronous rotating speed coordinate system; and determining the change rate of the negative sequence current according to the negative sequence current in the three-phase coordinate system.

Fig. 7 is a flowchart of a method for detecting a fault in a stator winding of an electric machine according to another embodiment of the present application. As shown in fig. 7, the negative sequence current change rate is obtained according to the following method steps, or the negative sequence current change rate of the stator winding is determined based on the three-phase current signals and the three-phase voltage signals, and the method comprises the following steps:

step 701, converting the current signal in the three-phase coordinate system to a synchronous rotating speed coordinate system to obtain a current value in the synchronous rotating speed coordinate system.

Optionally, before step 701 is executed, low-pass filtering may be performed on the three-phase current signals to pass low-frequency signals, and high-frequency signals exceeding a cut-off frequency are blocked or weakened, so as to filter out high-frequency interference signals.

Wherein the rotation speeds are synchronizedThe coordinate system is an alpha-beta coordinate system, and the current values on the alpha axis and the beta axis in the alpha-beta coordinate system are respectively marked as i_qAnd i_dSpecifically, the current value i in the synchronous revolution speed coordinate system can be calculated by the following equation (11)_qAnd i_d：

In the formula (11), the reaction mixture is,

and

respectively representing voltage positive sequence components on abc phases under an abc coordinate system of a three-phase coordinate system; i.e. i_a、i_bAnd i_cRespectively representing current signals on abc phases under an abc coordinate system of a three-phase coordinate system; h denotes a hilbert transform operator,

presentation pair

Taking a negative number after the Hilbert transform is carried out,

and

can refer to

The explanation of (a) will not be described one by one here.

Step 702, converting the current signal in the three-phase coordinate system to the inverse synchronous rotating speed coordinate system to obtain a current value in the inverse synchronous rotating speed coordinate system.

Wherein, the inverse synchronous rotating speed coordinate system is an inverse coordinate system of an alpha beta coordinate system, and the current values of the d axis and the q axis under the inverse coordinate system are respectively recordedIs composed of

And

specifically, the current value in the inversely synchronous rotational speed coordinate system can be calculated by the following formula (12)

And

in the formula (12), the reaction mixture is,

and

respectively representing voltage positive sequence components on abc phases under an abc coordinate system of a three-phase coordinate system; i.e. i_a、i_bAnd i_cRespectively representing current signals on abc phases under an abc coordinate system of a three-phase coordinate system; h is the Hilbert transform (Hilbert) operator,

presentation pair

Taking a negative number after the Hilbert transform is carried out,

and

can refer to

The explanation of (a) will not be described one by one here.

And 703, determining the positive sequence current in the synchronous rotating speed coordinate system based on the current value in the synchronous rotating speed coordinate system.

Specifically, based on the current value i in the synchronous rotation speed coordinate system_qAnd i_dCalculating the positive sequence current

Wherein the positive sequence current can be calculated using the following equation (13)

The meaning of each character in formula (13) can be referred to in the description of the other parts, and the description is not repeated here.

And 704, determining the negative sequence current in the synchronous rotating speed coordinate system based on the current value in the inverse synchronous rotating speed coordinate system.

Specifically, the current value is based on the inverse synchronous rotating speed coordinate system

And

calculating negative sequence current under synchronous rotating speed coordinate system

Wherein the negative-sequence current can be calculated by the following equation (14)

The meaning of each character in the formula (14) can be referred to the description of the contents in other parts, and the description is not repeated here.

Step 705, determining the positive sequence current value in the three-phase coordinate system

And negative sequence current value

Wherein, the voltage positive sequence component of three-phase voltage signal under three-phase coordinate system and the positive sequence current under synchronous rotating speed coordinate system can be used

Calculating the positive sequence current value under a three-phase coordinate system

Specifically, the following equation (15) can be used to implement:

the meaning of each character in the formula (15) can be referred to the description of the contents of other parts, and the description is not repeated here.

Wherein, the voltage positive sequence component of three-phase voltage signal under three-phase coordinate system and the negative sequence current under synchronous rotating speed coordinate system can be used

Calculating the negative sequence current value under a three-phase coordinate system

Specifically, it can be calculated by the following formula (16)：

The meaning of each character in the formula (16) can be referred to the description of the contents of other parts, and the description is not repeated here.

And step 706, calculating the average value of the effective values of the three-phase negative sequence current values according to the negative sequence current values in the three-phase coordinate system.

In this step, first, the effective values of the negative sequence current values of the phases a, b, and c are calculated based on the negative sequence current values in the three-phase coordinate system, and then the average value of the effective values of the negative sequence current values of the phases a, b, and c is calculated. Specifically, the calculation can be performed by the following formula (17):

and step 707, calculating the average change rate of the negative sequence current effective values of the three phases.

Wherein the change rate of the average value of the effective values of the three-phase negative sequence currents is delta I,

in the formula I_n0For the preset value, for example, the three-phase average effective value of the negative sequence component of the stator current under the normal condition of the motor is taken as the preset value I_n0。

Fig. 8 is a flowchart of a method for detecting a fault in a stator winding of an electric machine according to another embodiment of the present application. As shown in fig. 8, the impedance change rate is obtained according to the following method steps, or the impedance change rate of the stator winding is determined based on the three-phase current signals and the three-phase voltage signals, and the method includes:

step 801, two groups of three-phase voltages and three-phase currents corresponding to the three-phase voltages are respectively obtained.

In this embodiment, the two sets of voltage and current data are voltage and current data collected at different times. The two groups of three-phase voltages and the three-phase currents corresponding to the three-phase voltages can be two groups of voltage and current data which are continuous in time sequence, and can also be two groups of voltage and current data which are separated in time sequence.

And step 802, calculating the negative sequence voltage of the three-phase voltage of each group respectively.

For the negative sequence voltage of the three-phase voltage of each group, reference may be made to the description of the embodiment shown in fig. 5, and details thereof are not repeated here.

And step 803, calculating the effective value of the negative sequence voltage of each group of three-phase voltages.

Wherein, the effective value of the negative sequence voltage of each group of three-phase voltages can be calculated by the following formula (18):

in formula (18), x is 1, 2. When x is 1, V_1nAn effective value representing a negative sequence voltage of the first set of three phase voltages; when x is 2, V_2nAn effective value of a negative sequence voltage representing a second set of three-phase voltages.

Representing the negative sequence voltage component of the x group of three-phase voltage signals in the phase a;

representing the negative sequence voltage component of the x group of three-phase voltage signals in the phase b;

and the negative sequence voltage component of the x-th group of three-phase voltage signals in the c phase is shown.

N in the three-phase voltage signal represents the sampling time of the three-phase voltage signal, and the subscript x represents the group number of the acquired three-phase voltage signal and takes the value of 1 or 2.

And step 804, determining whether the effective values of the negative sequence voltages of the two groups of three-phase voltages are equal.

Specifically, V is determined_1nAnd V_2nWhether or not equal.

And 805, if the effective values of the negative sequence voltages of the two groups of three-phase voltages are not equal, respectively calculating the effective value of the negative sequence current and the effective value of the positive sequence current of each group of three-phase currents.

In particular, if V_1n≠V_2nFirst, the negative sequence current and the positive sequence current of each group of three-phase currents are calculated, and for the calculation process of the negative sequence current and the positive sequence current of each group of three-phase currents, reference may be made to the specific implementation manner of the embodiment shown in fig. 7, which is not described herein again.

Wherein, the effective value of the ith group of positive sequence currents is realized by the following formula (19):

in the formula (19), I_xp、

The subscript x in the drawing represents the group number of the acquired three-phase current signals, and the value is 1 or 2. I is_xpRepresents the effective value of the positive sequence current of the x group;

respectively represent the positive sequence current of the phase a, the phase b and the phase c of the x group.

The effective value of the x-th group of negative-sequence currents is realized by the following formula (20):

in the formula (20), I_xn、

The subscript x in the drawing represents the group number of the acquired three-phase current signals, and the value is 1 or 2. I is_xnRepresents the effective value of the x-th group of negative-sequence currents;

respectively represent the negative sequence current of the phase a, b and c of the x-th group.

And 806, calculating an impedance value according to the effective value of the negative sequence current and the effective value of the positive sequence current of each group of three-phase currents and the effective value of the negative sequence voltage of each group of three-phase voltages.

Alternatively, the impedance value may be calculated according to the following equation (21):

in the formula (21), Z_npRepresenting the impedance value at time n; i is_1n，I_2nEffective values of negative sequence currents respectively representing the 1 st group of three-phase current signals and the 2 nd group of three-phase current signals; v_1n，V_2nRespectively representing the effective values of the negative sequence voltages of the two groups of three-phase voltages; i is_1pAnd I_2pRespectively representing the effective values of the positive sequence currents of the 1 st and 2 nd group three-phase current signals.

Alternatively, if V_1n＝V_2nAnd returning to continuously acquire the third group of three-phase voltage signals and three-phase current signals or returning to acquire the two groups of three-phase voltage signals and three-phase current signals again.

And step 807, calculating the impedance change rate of the stator according to the impedance value and the impedance preset value.

Optionally, the impedance change rate of the stator is calculated according to the impedance value and the impedance preset value, and may be calculated as follows:

in the formula (22), Z_np0For the impedance preset value, the impedance value of the motor in the normal state can be selected as the preset impedance value.

Optionally, the instantaneous positive sequence power is obtained according to the following method steps: determining positive sequence current components of the three-phase current signals; determining positive sequence voltage components of the three-phase voltage signals; based on the positive sequence voltage component and the positive sequence current component, an instantaneous positive sequence power is determined. Specifically, the positive sequence voltage component and the positive sequence current component of each phase are multiplied, and then summed to obtain the instantaneous positive sequence power. For example, let the instantaneous positive sequence power be P, then

For the

For the meaning of (c), reference is made to the description of the preceding examples, which is not repeated here.

Fig. 9 is a schematic structural diagram of a fault detection device for a stator winding of a motor according to an embodiment of the present application. The fault detection device for a motor stator winding provided in the embodiment of the present application may execute the processing procedure provided in the embodiment of the fault detection method for a motor stator winding, as shown in fig. 9, the fault detection device 90 for a motor stator winding includes: an acquisition module 91, a determination module 92 and a detection module 93; the acquiring module 91 is configured to acquire a three-phase current signal and a three-phase voltage signal of the stator winding; a determination module 92 for determining characteristic parameters of a plurality of different characteristics of the stator winding based on the three-phase current signals and the three-phase voltage signals; the detection module 93 is configured to input characteristic parameters of a plurality of different characteristics of the stator winding into a preset fault detection model, so as to detect whether the motor fails through the preset fault detection model; the preset fault detection model is obtained by joint training based on feature samples of a plurality of different features of the stator winding.

Optionally, the feature parameters of the plurality of different features include: at least two items of frequency domain characteristics, voltage unbalance factors, negative sequence current change rate, impedance change rate and instantaneous positive sequence power of park transformation vector mode signals; when the determining module 92 determines the characteristic parameters of the plurality of different characteristics of the stator winding based on the three-phase current signal and the three-phase voltage signal, the determining specifically includes: determining at least two items of frequency domain characteristics, voltage unbalance factors, negative sequence current change rates, impedance change rates and instantaneous positive sequence power of park transformation vector mode signals of the stator winding based on the three-phase current signals and the three-phase voltage signals; the detection module 93 inputs characteristic parameters of a plurality of different characteristics of the stator winding into a preset fault detection model, so as to detect whether the motor is in fault through the preset fault detection model, and the method specifically includes: inputting the frequency domain characteristics, the voltage unbalance factor, the negative sequence current change rate, the impedance change rate and the instantaneous positive sequence power of the park transformation vector mode signal into a preset fault detection model so as to detect whether the motor is in fault or not through the fault detection model; the preset fault detection model is obtained by jointly training a frequency domain characteristic sample, a voltage unbalance factor sample, a negative sequence current change rate sample, an impedance change rate sample and an instantaneous positive sequence power sample based on park transformation vector mode signals.

Optionally, the determining module 92 determines the frequency domain characteristic of the park transformation vector mode signal according to the following method steps: acquiring the power frequency of the stator winding; carrying out park transformation on the three-phase current signals to obtain park transformation vectors; performing modulus taking and square on the park transformation vector to obtain a square value of a park transformation vector modulus; calculating the average value of the square value of the park transformation vector mode; calculating the amplitude of the preset multiple of the power supply frequency based on the result of Fourier transform on the average value; and calculating the ratio of the amplitude of the power frequency rate of the preset multiple to the average value to obtain the frequency domain characteristic of the park transformation vector mode signal.

Optionally, the determining module 92 determines the voltage imbalance factor according to the following method steps: respectively determining a voltage positive sequence component and a voltage negative sequence component of the three-phase voltage signal; determining the voltage imbalance factor based on the voltage negative sequence component and the voltage positive sequence component.

Optionally, when the determining module 92 determines the voltage positive sequence component and the voltage negative sequence component of the three-phase voltage signal respectively, the method specifically includes: calculating a three-phase equivalent power supply voltage based on the three-phase voltage signal; converting the three-phase equivalent power supply voltage to a synchronous rotating speed coordinate system; calculating a voltage positive sequence component and a voltage negative sequence component under a synchronous rotating speed coordinate system; and calculating the voltage positive sequence component and the voltage negative sequence component under the three-phase coordinate system based on the voltage positive sequence component and the voltage negative sequence component under the synchronous rotating speed coordinate system.

Optionally, when the determining module 92 determines the voltage imbalance factor based on the voltage negative sequence component and the voltage positive sequence component, the determining specifically includes: calculating the average value of the effective values of the voltage positive sequence components of the three phases and the average value of the effective values of the voltage negative sequence components of the three phases; the voltage unbalance factor is determined from a ratio of an average of the effective values of the negative sequence of voltage components of the three phases and an average of the effective values of the positive sequence of voltage components of the three phases.

Optionally, the determining module 92 determines the negative-sequence current change rate according to the following method steps: converting the three-phase current signals to an inverse synchronous rotating speed coordinate system to obtain current signals in the inverse synchronous rotating speed coordinate system; determining the negative sequence current of the current signal under the inverse synchronous rotating speed coordinate system; determining negative sequence current under a three-phase coordinate system based on the negative sequence current of the current signal under the inverse synchronous rotating speed coordinate system and the positive sequence voltage of the three-phase voltage signal; and determining the change rate of the negative sequence current according to the negative sequence current in the three-phase coordinate system.

Optionally, the determining module 92 determines the impedance change rate according to the following method steps: respectively acquiring two groups of three-phase voltage signals and three-phase current signals corresponding to the three-phase voltage signals; respectively calculating the negative sequence voltage of each group of three-phase voltage signals; calculating the effective value of the negative sequence voltage of each group of three-phase voltage signals; under the condition that the effective values of the negative sequence voltages of the two groups of three-phase voltage signals are not equal, the effective value of the negative sequence current and the effective value of the positive sequence current of each group of three-phase current signals are respectively calculated.

Optionally, the determining module 92 determines the instantaneous positive sequence power according to the following method steps: determining positive sequence current components of the three-phase current signals at various moments; determining positive sequence voltage components of the three-phase voltage signals at each time; determining the instantaneous positive sequence power based on the positive sequence voltage component at each time instant and the positive sequence current component at the corresponding time instant.

Optionally, the three-phase voltage signal is acquired by a voltage sensor arranged at an output end of a power supply of the motor; the three-phase current signals are acquired through a current sensor arranged on a power supply line of the stator.

The fault detection apparatus for a stator winding of a motor in the embodiment shown in fig. 9 can be used to implement the technical solution of the above method embodiment, and the implementation principle and technical effect are similar, and are not described herein again.

Fig. 10 is a schematic structural diagram of a fault detection apparatus for a stator winding of a motor according to an embodiment of the present application. The fault detection device for a stator winding of a motor according to the embodiment of the present application may execute the processing procedure provided in the embodiment of the fault detection method for a stator winding of a motor, as shown in fig. 10, the fault detection device 100 for a stator winding of a motor includes: memory 101, processor 102, computer programs and communication interface 103; wherein the computer program is stored in the memory 101 and is configured to be executed by the processor 102 for the solution of the above method embodiment.

The fault detection device for the stator winding of the motor in the embodiment shown in fig. 10 can be used for implementing the technical solution of the above method embodiment, and the implementation principle and technical effect are similar, and are not described herein again.

In addition, the present application also provides a computer readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to implement the fault detection method for the stator winding of the motor described in the above embodiments.

In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.

The integrated unit implemented in the form of a software functional unit may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) or a processor (processor) to execute some steps of the methods according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

It is obvious to those skilled in the art that, for convenience and simplicity of description, the foregoing division of the functional modules is merely used as an example, and in practical applications, the above function distribution may be performed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to perform all or part of the above described functions. For the specific working process of the device described above, reference may be made to the corresponding process in the foregoing method embodiment, which is not described herein again.

Finally, it should be noted that: the above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method of fault detection of a stator winding of an electric machine, comprising:

acquiring three-phase current signals and three-phase voltage signals of the stator winding;

determining characteristic parameters of a plurality of different characteristics of the stator winding based on the three-phase current signals and the three-phase voltage signals, the characteristic parameters of the plurality of different characteristics including: at least two items of frequency domain characteristics, voltage unbalance factors, negative sequence current change rate, impedance change rate and instantaneous positive sequence power of park transformation vector mode signals;

inputting characteristic parameters of a plurality of different characteristics of the stator winding into a preset fault detection model so as to detect whether the motor is in fault or not through the preset fault detection model;

the inputting of characteristic parameters of a plurality of different characteristics of the stator winding into a preset fault detection model to detect whether the motor is faulty through the preset fault detection model includes:

inputting the frequency domain characteristics, the voltage unbalance factor, the negative sequence current change rate, the impedance change rate and the instantaneous positive sequence power of the park transformation vector mode signal into a preset fault detection model so as to detect whether the motor is in fault or not through the fault detection model;

the preset fault detection model is obtained by jointly training a frequency domain characteristic sample, a voltage unbalance factor sample, a negative sequence current change rate sample, an impedance change rate sample and an instantaneous positive sequence power sample based on park transformation vector mode signals.

2. The method of claim 1, wherein the frequency domain characteristics of the park transform vector mode signal are determined in accordance with the following method steps:

acquiring the power frequency of the stator winding;

carrying out park transformation on the three-phase current signals to obtain park transformation vectors;

performing modulus taking and square on the park transformation vector to obtain a square value of a park transformation vector modulus;

calculating the average value of the square value of the park transformation vector mode;

calculating the amplitude of the preset multiple of the power supply frequency based on the result of Fourier transform on the average value;

and calculating the ratio of the amplitude of the preset multiple of the power supply frequency to the average value to obtain the frequency domain characteristic of the park transformation vector mode signal.

3. The method according to claim 1, characterized in that the voltage unbalance factor is determined according to the following method steps:

respectively determining a voltage positive sequence component and a voltage negative sequence component of the three-phase voltage signal;

determining the voltage imbalance factor based on the voltage negative sequence component and the voltage positive sequence component.

4. The method of claim 3, wherein the separately determining a voltage positive sequence component and a voltage negative sequence component of the three-phase voltage signals comprises:

calculating a three-phase equivalent power supply voltage based on the three-phase voltage signal;

converting the three-phase equivalent power supply voltage to a synchronous rotating speed coordinate system;

calculating a voltage positive sequence component and a voltage negative sequence component under a synchronous rotating speed coordinate system;

and calculating the voltage positive sequence component and the voltage negative sequence component under the three-phase coordinate system based on the voltage positive sequence component and the voltage negative sequence component under the synchronous rotating speed coordinate system.

5. The method of claim 3 or 4, wherein determining the voltage imbalance factor based on the voltage negative sequence component and the voltage positive sequence component comprises:

calculating the average value of the effective values of the voltage positive sequence components of the three phases and the average value of the effective values of the voltage negative sequence components of the three phases;

the voltage unbalance factor is determined from a ratio of an average of the effective values of the negative sequence of voltage components of the three phases and an average of the effective values of the positive sequence of voltage components of the three phases.

6. The method according to claim 1, characterized in that the negative-sequence current rate of change is obtained according to the following method steps:

converting the three-phase current signals to an inverse synchronous rotating speed coordinate system to obtain current signals in the inverse synchronous rotating speed coordinate system;

determining the negative sequence current of the current signal under the inverse synchronous rotating speed coordinate system;

determining negative sequence current under a three-phase coordinate system based on the negative sequence current of the current signal under the inverse synchronous rotating speed coordinate system and the positive sequence voltage of the three-phase voltage signal;

and determining the change rate of the negative sequence current according to the negative sequence current in the three-phase coordinate system.

7. The method of claim 1, wherein the impedance rate of change is obtained according to the method steps of:

respectively acquiring two groups of three-phase voltage signals and three-phase current signals corresponding to the three-phase voltage signals;

respectively calculating the negative sequence voltage of each group of three-phase voltage signals;

calculating the effective value of the negative sequence voltage of each group of three-phase voltage signals;

under the condition that the effective values of the negative sequence voltages of the two groups of three-phase voltage signals are not equal, the effective value of the negative sequence current and the effective value of the positive sequence current of each group of three-phase current signals are respectively calculated.

8. The method of claim 1, wherein the instantaneous positive sequence power is obtained according to the following method steps:

determining positive sequence current components of the three-phase current signals at various moments;

determining positive sequence voltage components of the three-phase voltage signals at each time;

determining the instantaneous positive sequence power based on the positive sequence voltage component at each time instant and the positive sequence current component at the corresponding time instant.

9. The method according to any one of claims 1 to 4 and 6 to 8, characterized in that the three-phase voltage signals are acquired by a voltage sensor arranged at the output end of a power supply of the motor;

the three-phase current signals are acquired through a current sensor arranged on a power supply line of the stator.

10. A fault detection device for a stator winding of an electric machine, comprising:

the acquisition module is used for acquiring three-phase current signals and three-phase voltage signals of the stator winding;

a determination module configured to determine characteristic parameters of a plurality of different characteristics of the stator winding based on the three-phase current signals and the three-phase voltage signals, the characteristic parameters of the plurality of different characteristics including: at least two items of frequency domain characteristics, voltage unbalance factors, negative sequence current change rate, impedance change rate and instantaneous positive sequence power of park transformation vector mode signals;

the detection module is used for inputting the frequency domain characteristics, the voltage unbalance factor, the negative sequence current change rate, the impedance change rate and the instantaneous positive sequence power of the park transformation vector mode signal into a preset fault detection model so as to detect whether the motor is in fault or not through the fault detection model;

11. A fault detection device for a stator winding of an electrical machine, comprising:

a memory;

a processor; and

a computer program;

wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method of any one of claims 1-9.

12. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method according to any one of claims 1-9.