WO2020155147A1 - 电机运行状态的获取方法和装置 - Google Patents

电机运行状态的获取方法和装置 Download PDF

Info

Publication number
WO2020155147A1
WO2020155147A1 PCT/CN2019/074606 CN2019074606W WO2020155147A1 WO 2020155147 A1 WO2020155147 A1 WO 2020155147A1 CN 2019074606 W CN2019074606 W CN 2019074606W WO 2020155147 A1 WO2020155147 A1 WO 2020155147A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
motor
information
operating state
amplitude spectrum
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2019/074606
Other languages
English (en)
French (fr)
Inventor
李延召
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
SZ DJI Technology Co Ltd
Original Assignee
SZ DJI Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by SZ DJI Technology Co Ltd filed Critical SZ DJI Technology Co Ltd
Priority to PCT/CN2019/074606 priority Critical patent/WO2020155147A1/zh
Priority to CN201980005478.4A priority patent/CN111819452A/zh
Publication of WO2020155147A1 publication Critical patent/WO2020155147A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/34—Testing dynamo-electric machines
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P29/00—Arrangements for regulating or controlling electric motors, appropriate for both AC and DC motors

Definitions

  • This application relates to computer technology, and in particular to a method and device for obtaining the running state of a motor.
  • the current method of monitoring the running state of the motor is generally determined by the way of human ear listening, that is, the human ear judges the running state of the motor by listening to the noise during the running of the motor.
  • the user can stand near the motor and listen to the noise during the operation of the motor, or record the noise during the operation of the motor as an audio file for offline listening.
  • the embodiments of the present application provide a method and device for acquiring the running state of a motor, which can efficiently and accurately acquire the running state of the motor.
  • an embodiment of the present application provides a method for acquiring the operating state of a motor, including:
  • the characteristic information of the signal is acquired, and the operating state of the motor is acquired according to the characteristic information.
  • an embodiment of the present application provides an apparatus for obtaining a running state of a motor, including: a memory, a processor, and a communication bus, where the memory and the processor are connected through the communication bus;
  • Memory used to store computer programs
  • the processor is used to call the computer program to perform the following operations:
  • the characteristic information of the signal is acquired, and the operating state of the motor is acquired according to the characteristic information.
  • an embodiment of the present application provides a computer-readable storage medium, including a program or instruction.
  • the program or instruction runs on a computer, the methods described in the first aspect and various possible designs are executed.
  • an embodiment of the present application provides an electronic device including a motor and a device for acquiring the operating state of the motor, and the acquiring device is configured to acquire the operating state of the motor.
  • This application obtains the operating state of the motor by acquiring the characteristic information of the signal related to the motor operating state, and the operating state of the motor is judged without the need for human ears, and the characteristic information of the signal related to the motor operating state can be objectively reflected
  • the running state of the motor therefore, the method for obtaining the running state of the motor of the present application has high efficiency and accuracy.
  • FIG. 1 is a first flowchart of a method for acquiring a motor operating state according to an embodiment of the application
  • Fig. 2 is a noise signal diagram of a motor provided by an embodiment of the application
  • FIG. 3 is a second flowchart of the method for acquiring the operating state of a motor according to an embodiment of the application
  • FIG. 4 is a schematic diagram of triangular window downsampling provided by an embodiment of the application.
  • FIG. 5 is a third flowchart of a method for acquiring a motor running state according to an embodiment of the application
  • FIG. 6 is a schematic diagram of N segments of the signal in FIG. 2;
  • FIG. 7 is a first structural diagram of a device for acquiring a motor operating state provided by an embodiment of the application.
  • FIG. 8 is a second structural diagram of the apparatus for acquiring the operating state of a motor provided by an embodiment of the application.
  • At least one refers to one or more
  • multiple refers to two or more.
  • And/or describes the association relationship of the associated object, indicating that there can be three relationships, for example, A and/or B, which can mean: A alone exists, both A and B exist, and B exists alone, where A, B can be singular or plural.
  • the character “/” generally indicates that the associated objects are in an “or” relationship.
  • the following at least one item (a)” or similar expressions refers to any combination of these items, including any combination of a single item (a) or a plurality of items (a).
  • At least one item (a) of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple .
  • the terms “first” and “second” in this application are used to distinguish similar objects, and are not necessarily used to describe a specific sequence or sequence.
  • the "value between x and y" in this embodiment includes not only the value in the interval (x, y), but also the two endpoint values of x and y.
  • the embodiments of the present application provide an efficient and accurate method for obtaining the running state of a motor.
  • Fig. 1 is a first flowchart of a method for acquiring a motor running state according to an embodiment of the application.
  • the execution subject of this embodiment may be a device for acquiring the operating state of the motor, and the device for acquiring the operating state of the motor may be implemented by hardware or software.
  • the method of this embodiment includes:
  • Step S101 Obtain a signal related to the running state of the motor
  • the signal related to the running state of the motor is one or more of the following: the noise signal of the motor, the current signal of the motor, and the signal measured by the inertial measurement unit arranged on the outer wall of the motor.
  • the noise signal of the motor can be collected by a microphone, accelerometer, vibration sensor, piezoelectric crystal and/or barometer; the current signal of the motor can be measured by an ammeter; the signal of the inertial measurement unit can be, for example, a vibration signal or current signal.
  • the signal related to the running state of the motor may also be other signals that can reflect the running state of the motor, and this embodiment is only illustrative.
  • the motor in this embodiment can be a motor that is not mounted on the device, or a motor that has been mounted on the device.
  • the device equipped with a motor can be any electronic device that needs to be driven by a motor, such as radar, drone, The electric fan system, etc., among which the radar can be a mechanical scanning lidar, and the unmanned aerial vehicle can be an unmanned aerial vehicle, an unmanned vehicle, an unmanned ship, etc., which are not limited here.
  • Fig. 2 is a noise signal diagram of a motor provided by an embodiment of the application. As shown in Figure 2, the noise signal in Figure 2 is the original noise signal, that is, the noise signal in the time domain.
  • Step S102 Obtain characteristic information of the signal, and obtain the operating state of the motor according to the characteristic information.
  • the signal related to the running state of the motor can be processed to obtain the characteristic information of the signal related to the running state of the motor.
  • the signal related to the running state of the motor can be directly processed to obtain the characteristic information of the signal related to the running state of the motor. It can also be processed by extracting part of the signal from the signal related to the running state of the motor to obtain the Characteristic information of signals related to operating status.
  • the signal related to the running state of the motor can be a collected signal related to the running state of the motor, or it can be from the collected signal.
  • the signal is extracted from the signal related to the running state of the motor.
  • the signal related to the running state of the motor may be a collected signal related to the running state of the motor.
  • the characteristic information of the signal related to the operating state of the motor may be at least one of the following: intensity information and distribution information.
  • the characteristic information of the signal related to the running state of the motor may also be other information, such as power information, which is not limited here.
  • the intensity information can be obtained based on signals related to the running state of the motor in the time domain, and can also be obtained based on signals related to the running state of the motor in the frequency domain, and signals related to the running state of the motor in the frequency domain. It can be the amplitude spectrum corresponding to the signal related to the running state of the motor.
  • the distribution information may be obtained according to the distribution signal corresponding to the signal related to the running state of the motor, and the distributed signal may be a cepstrum signal corresponding to the signal related to the running state of the motor.
  • the distributed signal corresponding to the signal related to the operating state of the motor in the time domain may also be obtained according to other suitable algorithms.
  • preprocessing the signal related to the running state of the motor may include normalizing the signal related to the running state of the motor or the signal extracted from the signal related to the running state of the motor to obtain the normalized signal. One signal. At this time, the characteristic information of the signal related to the running state of the motor can be obtained according to the normalized signal.
  • the characteristic information of the signal related to the running state of the motor Since the characteristic information of the signal related to the running state of the motor is acquired, the characteristic information can reflect the running state of the motor. After acquiring the characteristic information, the running state of the motor can be acquired according to the characteristic information.
  • the motor may have at least one operating state, and the acquired characteristic information of the signal related to the operating state of the motor is different, and the operating state of the corresponding motor may be different.
  • the running state of the motor corresponding to the characteristic information can be obtained according to the correspondence between the preset characteristic information and the running state of the motor, and the running state of the motor corresponding to the characteristic information can also be obtained according to a machine learning algorithm.
  • acquiring characteristic information of a signal related to the running state of the motor, and acquiring the running state of the motor according to the characteristic information includes: acquiring characteristic information of the signal related to the running state of the motor, according to the The characteristic information obtains the operating state of the motor corresponding to the characteristic information according to a preset correspondence relationship, and the preset correspondence relationship includes a plurality of preset characteristic information and an operating state corresponding to each preset information.
  • obtaining the operating state of the motor corresponding to the feature information according to the preset correspondence relationship includes: determining the target preset feature information corresponding to the feature information; determining the preset correspondence relationship with the target
  • the running state corresponding to the preset characteristic information is the running state of the motor.
  • the motor may have the following but not limited to the following operating states:
  • the characteristic information of the signal related to the operating state of the motor corresponds to a smaller total intensity of the signal, for example, the total intensity of the signal is less than a predefined intensity value.
  • the characteristic information of the signal related to the operating state of the motor corresponds to the total strength of the signal.
  • the total strength of the signal is within the normal range, but the strength of the signal at some frequencies is relatively large, for example, the signal The intensity at some frequencies is outside the normal range.
  • the characteristic information of the signal related to the operating state of the motor corresponds to the total intensity of the signal.
  • the total intensity of the signal is within the normal range, and the intensity of the signal at each frequency is mostly or all within the normal range.
  • the characteristic information of the signal related to the operating state of the motor corresponds to the total strength of the signal.
  • the total strength of the signal is within the normal range, but the strength of the signal at multiple frequencies is slightly higher.
  • the strength of the signal at multiple frequencies is slightly outside the normal range.
  • the characteristic information of the signal related to the operating state of the motor corresponds to a larger total strength of the signal, for example, outside the normal range.
  • a machine learning method may also be used to obtain the operating state of the motor corresponding to the characteristic information according to the characteristic information.
  • the operating state of the motor obtained in this embodiment may also be any one of the states (1) to (5) in the above example.
  • obtaining the operating state of the motor corresponding to the feature information includes: using a machine learning algorithm to obtain the operating state of the motor corresponding to the feature information according to the feature information and a machine learning model.
  • the machine learning model is obtained by using the machine learning algorithm based on multiple training samples and the labels of the training samples.
  • the training samples include the characteristic information of the signal related to the operating state of the first motor.
  • the labels of the training samples are used for Indicates the operating status of the first motor.
  • the feature information obtained in this step is used as the input of the machine learning model, and after calculation by the machine learning algorithm, a target label is output, and the target label is used to indicate the running state of the motor.
  • the first motor may be the motor in this embodiment or other motors.
  • the feature type included in the feature information of the signal related to the operating state of the first motor in the training sample is the same as the feature type included in the feature information of the signal related to the operating state of the motor obtained in this step.
  • the label of the training sample may be a character string indicating the operating state of the first motor. Accordingly, the target label is a character string indicating the operating state of the motor in this implementation.
  • the machine learning model is a neural network model.
  • the neural network algorithm can be any of the following: Back Propagation (BP) neural network, Stacked AutoEncoder (SAE) neural network, Long Short Term Memory Network , LSTM) Neural Networks, Recurrent Neural Networks (RNN), Convolutional Neural Networks.
  • BP Back Propagation
  • SAE Stacked AutoEncoder
  • LSTM Long Short Term Memory Network
  • RNN Recurrent Neural Networks
  • Convolutional Neural Networks Convolutional Neural Networks.
  • the operating state of the motor is obtained by acquiring the characteristic information of the signal related to the motor operating state, and the operating state of the motor is determined without the need for human ears, and the characteristic information of the signal related to the motor operating state can be objective Reflects the running state of the motor. Therefore, the method of this embodiment has high efficiency and accuracy for obtaining the running state of the motor.
  • Fig. 3 is a second flowchart of the method for acquiring the running state of the motor provided by an embodiment of the application. Referring to Fig. 3, the method of this embodiment includes:
  • Step S201 Obtain a signal related to the running state of the motor.
  • step S101 in the previous embodiment, which will not be repeated here.
  • Step S202 Obtain signal intensity information and/or distribution information related to the running state of the motor.
  • acquiring signal strength information and/or distribution information related to the running state of the motor includes:
  • a Fast Fourier Transform (Fast Fourier Transform, FFT for short) may be performed on the signal related to the running state of the motor to obtain the amplitude spectrum corresponding to the signal related to the running state of the motor.
  • each amplitude value in the amplitude spectrum corresponding to the signal related to the running state of the motor may be divided by the amplitude value corresponding to the zero frequency to obtain the normalized amplitude spectrum corresponding to the signal related to the running state of the motor.
  • the amplitude value corresponding to the zero frequency can be called the DC component of the amplitude spectrum, that is, the normalization method in this step is to use the DC component in the amplitude spectrum to remove other components.
  • the following describes the acquisition of the intensity information and/or distribution information of the signal according to the normalized amplitude spectrum.
  • acquiring the intensity information of the signal according to the normalized amplitude spectrum corresponding to the signal includes: acquiring the intensity information according to the normalized amplitude spectrum corresponding to the signal and M preset thresholds, and the intensity information includes M Intensity value, M is an integer greater than or equal to 1.
  • obtaining the intensity information according to the normalized amplitude spectrum corresponding to the signal and M preset thresholds includes: for any one of the M preset thresholds, obtaining the normalized amplitude spectrum The sum of the difference between each amplitude value and the first preset threshold, the sum of the difference between each amplitude value in the normalized amplitude spectrum and the first preset threshold is the intensity information corresponding to the first preset threshold The intensity value. It can be understood that the number of intensity values included in the intensity information can be determined according to the number of preset thresholds.
  • each of the M preset thresholds may be any value from 0 to 1.
  • the intensity information includes 3 intensity values; further, 3 preset threshold values are preset to be 0, 0.1, 0.2.
  • the amplitude values in the normalized amplitude spectrum whose amplitude values are greater than 0 are added to obtain the intensity value E 11 corresponding to the preset threshold value 0 in the intensity information, and each amplitude value in the normalized amplitude spectrum is added to 0.1
  • the difference value of the intensity information is added to obtain the intensity value E 12 corresponding to the preset threshold value of 0.1, and each amplitude value in the normalized amplitude spectrum is added to the difference value of 0.2 to obtain the intensity information
  • E 11 , E 12 and E 13 are three intensity values in the intensity information obtained according to the normalized amplitude spectrum, that is, E 11 , E 12 and E 13 constitute the aforementioned intensity information.
  • obtaining the distribution information of the signal according to the normalized amplitude spectrum corresponding to the signal related to the operating state of the motor includes:
  • obtaining the distributed signal corresponding to the signal according to the normalized amplitude spectrum corresponding to the signal includes:
  • the inverse cosine transform is performed on the normalized amplitude spectrum to obtain the distributed signal corresponding to the signal.
  • Other types of transformations can also be performed on the normalized amplitude spectrum to obtain the distribution signal corresponding to the signal, which is not limited in this embodiment.
  • the distribution signal here may be a cepstrum signal corresponding to the signal related to the operating state of the motor.
  • the distribution signal corresponding to the signal related to the running state of the motor can characterize the composition of the signal related to the running state of the motor, and express the frequency distribution of the signal related to the running state of the motor.
  • obtaining the distribution signal corresponding to the signal includes:
  • Triangular window down-sampling can be performed on the normalized amplitude spectrum to obtain a signal after down-sampling the normalized amplitude spectrum.
  • triangular window down-sampling can be uniform triangular window down-sampling.
  • triangular window downsampling can refer to the schematic diagram of triangular window downsampling in FIG. 4.
  • the signal after down-sampling the normalized amplitude spectrum can be obtained by the following formula:
  • g(j) is the signal after the normalized amplitude spectrum plus triangular window downsampling
  • floor(*) is the rounding operation
  • R is the motor speed
  • fs is the sampling frequency of the signal (that is, the signal obtained in step S201 The sampling frequency of the signal related to the operating state of the motor)
  • N is the length of the first sub-signal to obtain the first normalized amplitude spectrum
  • G(*) is the first normalized amplitude spectrum
  • win(i) is the triangular window Function
  • j is the abscissa of the signal after down-sampling the normalized amplitude spectrum triangular window, where the abscissa can be the number of sampling points or frequency, which is not limited here;
  • the normalized amplitude can also be down-sampled by adding other types of windows to obtain the signal after the normalized amplitude spectrum is down-sampled, which is not limited in this embodiment.
  • the inverse cosine transform can be performed on the signal after the normalized amplitude spectrum is down-sampled to obtain the distributed signal. It is also possible to perform other types of transformations (such as inverse Fourier transform, etc.) on the signal after the normalized amplitude spectrum is down-sampled to obtain the distributed signal, which is not limited in this embodiment.
  • the inverse cosine transform is performed on the signal after the normalized amplitude spectrum is downsampled, and the distributed signal can be obtained by the following formula:
  • C k is the distributed signal
  • k is the abscissa of the distributed signal
  • the abscissa can be the reciprocal frequency
  • Step S203 Acquire the operating state of the motor according to the intensity information and/or distribution information of the signal.
  • obtaining the operating state of the motor corresponding to the energy information and/or the distribution information according to the intensity information and/or distribution information of the signal includes:
  • a machine learning algorithm is used to obtain the operating state of the motor corresponding to the intensity information and/or the distribution information.
  • the machine learning model is obtained by using the machine learning algorithm based on multiple training samples and the labels of the training samples.
  • the training samples include signal strength information and/or distribution information related to the operating state of the first motor.
  • the label of the sample is used to indicate the operating status of the first motor.
  • the intensity information and/or distribution information obtained in this step is used as the input of the machine learning model, and after calculation by the machine learning algorithm, a target label is output, and the target label is used to indicate the running state of the motor.
  • the training sample when the machine learning algorithm is used to obtain the operating state of the motor corresponding to the intensity information according to the signal strength information and the machine learning model, the training sample includes the signal strength information related to the operating state of the first motor .
  • the training sample includes the distribution information of the signal related to the running state of the first motor.
  • the training sample includes the intensity of the signal related to the operating status of the first motor Information and distribution information.
  • This method uses the machine learning method to obtain the operating state of the motor, and the obtained operating state of the motor has a relatively high accuracy.
  • the operating state of the motor corresponding to the intensity information and/or the distribution information is obtained, and there are three situations as follows:
  • obtaining the operating status of the motor corresponding to the intensity information according to the intensity information of the signal includes: obtaining the operating status of the motor corresponding to the intensity information according to the intensity information of the signal according to a preset correspondence relationship. It is assumed that the corresponding relationship includes multiple preset intensity information and the operating state corresponding to each preset intensity information. Specifically: determining the target preset intensity information corresponding to the intensity information; determining that the operating state corresponding to the target preset intensity information in the preset correspondence is the operating state of the motor.
  • the intensity information includes a first intensity value corresponding to a zero value, a second intensity value corresponding to the second preset threshold value, and a third intensity value corresponding to the first preset threshold value, the first preset threshold value Less than the second preset threshold, the five preset intensity information included in the preset correspondence relationship and the corresponding operating states of the five preset intensity information are as follows:
  • the first type of preset intensity information the first intensity value is less than the first preset intensity value; the operating state corresponding to the first type of preset intensity information is abnormal startup of the motor.
  • the first preset intensity value may be any value between 0.3 and 0.7, such as 0.5.
  • the first intensity value is the intensity value corresponding to the zero value
  • the first intensity value is the total intensity value of the signal related to the running state of the motor
  • the first intensity value can indicate The total intensity value of the signal related to the running state of the motor. Therefore, when the first intensity value is less than the first preset intensity value, it can be considered that the total intensity value of the signal related to the running state of the motor is too small. Therefore, at this time The corresponding running state can be abnormal starting of the motor.
  • the second preset intensity information the first intensity value is greater than or equal to the first preset intensity value and less than or equal to the second preset intensity value, and the second intensity value is greater than the third preset intensity value; the second preset The operation status corresponding to the intensity information is the risk of abnormal operation in the first period of time.
  • the second preset intensity value may be any value from 2.5 to 3.5, for example, 3.
  • the third preset intensity value may be any value from 0.02 to 0.07, such as 0.05.
  • the second intensity value is an intensity value corresponding to the second preset threshold value, and the second preset threshold value is greater than the first preset threshold value, according to the method of obtaining the intensity value in step S202, it can be known that the second intensity value can indicate the operation of the motor The total intensity of the signal segment with relatively high intensity in the state-related signal. Therefore, when the first intensity value is greater than or equal to the first preset intensity value and less than or equal to the second preset intensity value, it can be considered that the total intensity of the signal related to the operating state of the motor is moderate, and the second intensity value is greater than the third preset intensity value.
  • the signal segment with relatively high intensity in the signal related to the operating state of the motor has a large total intensity, and the intensity of certain frequency positions in the signal related to the operating state of the motor is relatively large. Therefore, at this time The corresponding operating state may be the risk of abnormal operation in the first period of time.
  • the third type of preset intensity information the first intensity value is greater than or equal to the first preset intensity value and less than or equal to the second preset intensity value, and the second intensity value is less than or equal to the third preset intensity value and the third intensity value The difference with the second intensity value is less than or equal to the fourth preset intensity value; the operating state corresponding to the third preset intensity information is that the motor is operating normally.
  • the fourth preset intensity value may be any value from 0.3 to 0.7, such as 0.5.
  • the third intensity value is an intensity value corresponding to the first preset threshold value, and the second preset threshold value is greater than the first preset threshold value, according to the method of obtaining the intensity value in step S202, it can be known that the third intensity value may indicate the operation of the motor
  • the state-related signal strength is relatively moderate to the total strength of the signal segment. Therefore, when the first strength value is greater than or equal to the first preset strength value and less than or equal to the second preset strength value, the signal related to the operating state of the motor can be considered The total intensity of is moderate, and the second intensity value is less than or equal to the third preset intensity value.
  • the signal segment with relatively large intensity in the signal related to the motor's operating state has a smaller total intensity, that is, the signal related to the motor's operating state
  • the signal does not exist or the intensity of the frequency position is relatively large, and the difference between the third intensity value and the second intensity value is less than the fourth preset intensity value, indicating that the signal segment of the signal related to the operating state of the motor is relatively moderate in intensity
  • the total intensity of is also small, that is, the signal related to the running state of the motor does not exist or the intensity of the few frequency positions is relatively moderate. Therefore, the corresponding running state at this time can be that the motor is running normally.
  • the fourth type of preset intensity information the first intensity value is greater than or equal to the first preset intensity value and less than or equal to the second preset intensity value, and the second intensity value is less than or equal to the third preset intensity value and the third intensity value The difference between the second intensity value and the fourth preset intensity value is greater than or equal to the fourth preset intensity value; the operation state corresponding to the fourth type of preset intensity information is the risk of abnormal operation during the second duration.
  • the first intensity value is greater than or equal to the first preset intensity value and less than or equal to the second preset intensity value, it can be considered that the total intensity of the signal related to the operating state of the motor is moderate, and the second intensity value is less than or equal to the third
  • the preset intensity value it can be considered that the signal segment with relatively high intensity in the signal related to the motor's operating state has a small total intensity, that is, the signal related to the operating state of the motor does not exist or has a relatively small frequency position.
  • the difference between the third intensity value and the second intensity value is greater than the fourth preset intensity value, it indicates that the signal segment with relatively moderate intensity in the signal related to the motor's operating state has a larger total intensity, that is, the signal related to the motor's operating state
  • the intensity of the more frequency positions in the signal is relatively moderate. Therefore, the corresponding operating state at this time may be the risk of abnormal operation during the second time period.
  • the second duration is greater than the first duration, that is, the intensity of certain frequency positions in the signal related to the operating state of the motor is relatively large, and the intensity of more frequency positions in the signal related to the operating state of the motor is relatively moderate. , The motor is more prone to abnormalities.
  • the fifth preset intensity information if the first intensity value is greater than the second preset intensity value.
  • the running state corresponding to the fourth preset intensity information is excessive noise of the motor.
  • the corresponding operating state at this time may be that the noise of the motor is too large.
  • the operating state of the motor corresponding to the intensity information is obtained according to the preset corresponding relationship.
  • the preset corresponding relationship includes multiple preset intensity information and corresponding to each preset intensity information
  • the first intensity value, the second intensity value and the third intensity value included in the intensity information of the signal it is determined which of the above-mentioned (1) to (5) preset intensity information corresponds to the intensity information, if it corresponds to the above-mentioned intensity information.
  • the first type of preset intensity information determines that the running state of the motor is the abnormal start of the motor corresponding to the first type of preset intensity information.
  • obtaining the operating status of the motor corresponding to the intensity information includes: obtaining the operating status of the motor corresponding to the distribution information according to the distribution information of the signal according to a preset correspondence relationship.
  • the corresponding relationship includes multiple types of preset distribution information and operating states corresponding to each type of preset distribution information. Specifically: determining the target preset distribution information corresponding to the distribution information; determining that the operating state corresponding to the target preset distribution information in the preset correspondence is the operating state of the motor.
  • obtaining the operating state of the motor corresponding to the distribution information and intensity information includes: obtaining the distribution information according to the preset corresponding relationship according to the distribution information and intensity information of the signal
  • the operating state of the motor corresponding to the intensity information, the preset correspondence relationship includes a plurality of preset characteristic information and the operating state corresponding to each preset characteristic information.
  • the multiple preset feature information are multiple combinations of each preset intensity information and each preset distribution information. Specifically, it is: determining the target preset characteristic information corresponding to the distribution information and intensity information; determining that the operating state corresponding to the target preset characteristic information in the preset correspondence is the operating state of the motor.
  • This implementation of obtaining the operating state of the motor does not require training of a machine learning model, nor does it require a complex machine learning algorithm.
  • the hardware requirements are not high and the efficiency of obtaining the operating state of the motor is higher than the previous method.
  • the operating state of the motor is obtained by obtaining the strength information and/or distribution information of the signal related to the operating state of the motor, and the operating state of the motor is determined without the need for human ears, and due to the signal related to the operating state of the motor
  • the intensity information and/or distribution information of, can objectively reflect the operating state of the motor. Therefore, the method of this embodiment has high efficiency and accuracy in obtaining the operating state of the motor.
  • FIG. 5 is the third flowchart of the method for acquiring the running state of the motor provided by the embodiment of the application. Referring to FIG. 5, the method of this embodiment includes:
  • Step S301 Obtain a signal related to the running state of the motor.
  • step S101 in the embodiment shown in FIG. 1, which will not be repeated here.
  • Step S302 Extract N segments of sub-signals in the signal related to the running state of the motor, where N is an integer greater than or equal to 1.
  • N may be an integer greater than or equal to 2.
  • the sub-signal in this embodiment is a segment of the sub-signal whose length is less than the length of the signal, and the length of each segment of the sub-signal is the same.
  • the number of sampling points of each sub-signal may be 20000.
  • N-segment sub-signal is the N-segment sub-signal of the signal in FIG. 2.
  • FIG. 6 is a schematic diagram of N segments of sub-signals of the signal in FIG. 2.
  • N-segment sub-signals include: signals between a and b and signals between c and d.
  • the signal between a and b is a sub-signal of the signal in FIG. 2
  • the signal between c and d is a sub-signal of the signal in FIG. 2.
  • Step S303 According to the N-segment sub-signals, obtain signal strength information and/or distribution information related to the running state of the motor.
  • obtaining signal strength information and/or distribution information related to the running state of the motor includes:
  • N normalized amplitude spectra corresponding to N-segment sub-signals, where the N-segment sub-signal corresponds to the N normalized amplitude spectra one-to-one.
  • the N normalized amplitude spectra are the normalized amplitude spectra corresponding to the signal.
  • acquiring the first normalized amplitude spectrum corresponding to the first sub-signal includes:
  • a Fast Fourier Transform (Fast Fourier Transform, FFT for short) may be performed on the first sub-signal to obtain the first amplitude spectrum corresponding to the first sub-signal.
  • each amplitude value in the first amplitude spectrum may be divided by the amplitude value corresponding to the zero frequency to obtain the first normalized amplitude spectrum.
  • the amplitude value corresponding to the zero frequency can be referred to as the DC component of the first amplitude spectrum, that is, the normalization method in this step is to use the DC component in the first amplitude spectrum to remove other components.
  • the following describes the acquisition of the intensity information and distribution information of the signal related to the operating state of the motor according to the N normalized amplitude spectra.
  • obtaining signal intensity information related to the operating state of the motor includes: obtaining N sets of intensity values according to N normalized amplitude spectra, where N normalized amplitudes There is a one-to-one correspondence between the spectrum and the N groups of intensity values. Further, each group of intensity values may include M intensity component values, and M is an integer greater than or equal to 1. In other words, a set of intensity values can be obtained according to each normalized amplitude spectrum, and each set of intensity values includes M intensity component values.
  • the N groups of intensity values obtained above are the signal intensity information related to the operating state of the motor.
  • the method for obtaining N groups of intensity values will be described below.
  • obtaining the first set of intensity values according to the first normalized amplitude spectrum includes: for any of the M preset thresholds A first preset threshold to obtain the sum of the difference between each amplitude value in the first normalized amplitude spectrum and the first preset threshold, each amplitude value in the first normalized amplitude spectrum and the first preset threshold The sum of the differences is the intensity component value corresponding to the first preset threshold in the first set of intensity information. It can be understood that the number of intensity component values in a set of intensity values can be determined according to the number of preset thresholds.
  • each preset threshold can be any value from 0 to 1.
  • the first group of intensity values includes 3 intensity component values; further, 3 preset threshold values are preset to 0, 0.1, 0.2.
  • the amplitude values in the first normalized amplitude spectrum whose intensity value is greater than 0 are added to obtain the intensity component value E 11 corresponding to the preset threshold 0 in the first group of intensity values, and the first normalized amplitude spectrum is The sum of the difference between each amplitude value and 0.1 is added to obtain the intensity component value E 12 corresponding to the preset threshold 0.1 in the first group of intensity values, and the difference between each amplitude value in the first normalized amplitude spectrum and 0.2 The sum is added to obtain the intensity component value E 13 corresponding to the preset threshold 0.2 in the first group of intensity values.
  • E 11 , E 12 and E 13 are the three intensity component values in the first group of intensity values obtained according to the first normalized amplitude spectrum, that is, E 11 , E 12 and E 13 form the aforementioned first group Strength value.
  • obtaining distribution information of signals related to the operating state of the motor includes:
  • N normalized amplitude spectra obtain N distributed signals corresponding to the signals related to the operating state of the motor, and the N normalized amplitudes correspond to the N distributed signals one-to-one.
  • the inverse cosine transform is performed on the first normalized amplitude spectrum to obtain the first distributed signal. It is also possible to perform other types of transformations (such as inverse Fourier transform, etc.) on the first normalized amplitude to obtain the first distributed signal, which is not limited in this embodiment.
  • the first normalized amplitude spectrum is obtained according to the first normalized amplitude spectrum, including :
  • down-sampling the first normalized amplitude spectrum to obtain the signal after down-sampling the first normalized amplitude spectrum includes: performing triangular window down-sampling on the first normalized amplitude spectrum to obtain the first normalized amplitude spectrum Unify the signal after the amplitude spectrum is downsampled.
  • triangular window down-sampling can be uniform triangular window down-sampling.
  • the first normalized amplitude may also be down-sampled by adding other types of windows to obtain a signal after down-sampling the first normalized amplitude spectrum, which is not limited in this embodiment.
  • the inverse cosine transform may be performed on the signal after the down-sampling of the first normalized amplitude spectrum to obtain the first distributed signal. It is also possible to perform other types of transformations (such as inverse Fourier transform, etc.) on the first normalized amplitude to obtain the first distributed signal, which is not limited in this embodiment.
  • each distribution signal a set of distribution values can be obtained, and each set of distribution values includes K distribution component values.
  • the N sets of distribution values are the distribution information of signals related to the running state of the motor.
  • obtaining the first set of distribution values according to the first distribution signal includes: determining the top K of the first distribution signals
  • the distribution value is the first set of distribution values.
  • Step S304 Obtain the operating state of the motor according to the intensity information and/or distribution information of the signal.
  • obtaining the operating state of the motor corresponding to the intensity information and/or the distribution information according to the intensity information and/or the distribution information of the signal includes:
  • a machine learning algorithm is used to obtain the operating state of the motor corresponding to the N sets of intensity values and/or N sets of distribution values.
  • the machine learning model is obtained by using the machine learning algorithm based on multiple training samples and the labels of the training samples.
  • the training samples include N sets of intensity values and N sets of distribution values of signals related to the operating state of the first motor.
  • the label of the training sample is used to indicate the running state of the first motor.
  • the N sets of intensity values and N sets of distribution values obtained in this step are used as the input of the machine learning model, and after calculation by the machine learning algorithm, a target label is output, and the target label is used to indicate the running state of the motor.
  • the training sample when the machine learning algorithm is used to obtain the operating state of the motor corresponding to the intensity information according to the N sets of intensity values of the signal and the machine learning model, the training sample includes the information of the signal related to the operating state of the first motor. N groups of intensity values.
  • the training sample includes N sets of distributions of signals related to the operating state of the first motor value.
  • the training sample includes the first motor N sets of intensity values and N sets of distribution values of signals related to the operating state.
  • This method uses the machine learning method to obtain the operating state of the motor, and the obtained operating state of the motor has a relatively high accuracy.
  • the operating state of the motor corresponding to the intensity information and/or the distribution information is obtained, and there are three situations as follows:
  • obtaining the operating state of the motor corresponding to the intensity information according to the intensity information of the signal includes: obtaining the operating state of the motor corresponding to the N groups of intensity values according to a preset correspondence relationship according to the N groups of intensity values of the signal State, the preset correspondence relationship includes a variety of preset intensity information and an operating state corresponding to each type of preset intensity information. Specifically: determining the target preset intensity information corresponding to the N groups of intensity values; determining that the operating state corresponding to the target preset intensity information in the preset correspondence is the operating state of the motor.
  • each of the N sets of intensity values includes an intensity component value corresponding to a zero value, an intensity component value corresponding to the second preset threshold value, and an intensity component value corresponding to the first preset threshold value
  • the first preset threshold is smaller than the second preset threshold.
  • the preset correspondence relationship is the same as in the corresponding example in the previous embodiment.
  • the N sets of intensity values included in the intensity information of the signal determine which of the above-mentioned (1) to (5) preset intensity information corresponds to the N sets of intensity values. If it corresponds to the first type of preset intensity information, then It is determined that the running state of the motor is the abnormal start of the motor corresponding to the first preset intensity information.
  • each of the N groups of intensity values includes an intensity component value corresponding to a zero value, an intensity component value corresponding to the second preset threshold value, and an intensity component value corresponding to the first preset threshold value, Then there are N intensity component values corresponding to the zero value, N intensity component values corresponding to the second preset threshold, and N intensity component values corresponding to the first preset threshold.
  • the average value of the N intensity component values corresponding to the zero value can be obtained, the first intensity value can be obtained, and the N intensity values corresponding to the second preset threshold can be obtained.
  • the average value of the component values is used to obtain the second intensity value, and the average value of the N intensity component values corresponding to the first preset threshold is obtained to obtain the third intensity value.
  • the median value or the mean square deviation of the N intensity component values corresponding to the zero value can also be obtained according to other suitable algorithms, such as median value, mean square error, etc., to obtain the first intensity value.
  • the median value or mean square deviation of the N intensity component values corresponding to the second preset threshold value is obtained to obtain the second intensity value, and the median value or mean square deviation of the N intensity component values corresponding to the first preset threshold value is obtained to obtain the first Three intensity values.
  • obtaining the operating state of the motor corresponding to the intensity information includes: obtaining the operating state of the motor corresponding to the N sets of distribution values according to the preset corresponding relationship according to the N sets of distribution values of the signal ,
  • the preset correspondence relationship includes multiple types of preset distribution information and operating states corresponding to each type of preset distribution information. Specifically: determining the target preset distribution information corresponding to the N sets of distribution values; determining that the operating state corresponding to the target preset distribution information in the preset correspondence is the operating state of the motor.
  • the distribution information and intensity information of the signal obtain the operating state of the motor corresponding to the distribution information and intensity information, including: obtaining and obtaining and corresponding to the N groups of intensity values and N groups of distribution values of the signal according to the preset corresponding relationship
  • the operating states of the motors corresponding to the N sets of intensity values and the N sets of distribution values, and the preset correspondence relationship includes multiple preset feature information and operating states corresponding to each preset feature information.
  • the multiple preset feature information are multiple combinations of each preset intensity information and each preset distribution information. Specifically: determining the target preset information corresponding to the N sets of intensity values and the N sets of distribution values; determining that the operating state corresponding to the target preset information in the preset correspondence is the operating state of the motor.
  • This implementation of obtaining the operating state of the motor does not require training of a machine learning model, nor does it require a complex machine learning algorithm.
  • the hardware requirements are not high and the efficiency of obtaining the operating state of the motor is higher than the previous method.
  • the operating state of the motor is obtained by obtaining the strength information and/or distribution information of the signal related to the operating state of the motor, and the operating state of the motor is determined without the need for human ears, and due to the signal related to the operating state of the motor
  • the intensity information and/or distribution information of, can objectively reflect the operating state of the motor. Therefore, the method of this embodiment has high efficiency and accuracy in obtaining the operating state of the motor.
  • the "acquiring a signal related to the operating state of the motor” in the foregoing embodiment includes: acquiring a plurality of signals related to the operating state of the motor within a preset period of time.
  • “according to the characteristic information of the signal related to the running state of the motor, obtaining the running state of the motor according to the characteristic information” includes: for a first signal among the signals related to the running state of the motor , Acquiring characteristic information of the first signal; acquiring the operating state of the motor according to the characteristic information corresponding to each of the multiple signals related to the operating state of the motor.
  • the multiple signals related to the operating state of the motor may be the same type of signal, or may be different types of signals.
  • the characteristic information of the signal is obtained by the method of the embodiment shown in Fig. 3 or Fig. 5, and the running state of the motor is obtained according to the characteristic information, that is, in the preset duration
  • the operating status of multiple motors is acquired in the internal, if the number of target operating statuses in the acquired operating status of multiple motors is greater than the preset value, the operating status of the motor is considered to be the target operating status; this can enhance the aforementioned operating status
  • the robustness of the acquisition method can be set to any value between 0.5 hour and 1 day, or it can be set to 3 to 5 days, or even longer. In other embodiments, more than one is related to the running state of the motor.
  • Related signals may have partially overlapping signals, and the preset duration can be set according to actual needs, which is not limited here.
  • 10 signals related to the running state of the motor are acquired within a preset time period, and the preset value is 5.
  • the method of the embodiment shown in FIG. 3 or FIG. 5 is used to obtain 10 motor running states based on 10 signals related to the motor running state, and 10 motor running states are obtained.
  • 6 of the 10 motor running states are that the motor noise is too large, and it is determined that the motor's running state is that the motor noise is too large.
  • FIG. 7 is a first structural diagram of an apparatus for acquiring a motor operating state according to an embodiment of the application; referring to FIG. 7, the apparatus 600 for acquiring a motor operating state in this embodiment includes a memory 61, a processor 62, and a communication bus 63. The memory 61 and the processor 62 are connected through the communication bus;
  • the memory 61 is used to store computer programs
  • the processor 62 is configured to call the computer program and perform the following operations:
  • the characteristic information of the signal is acquired, and the operating state of the motor is acquired according to the characteristic information.
  • the processor 62 may be a CPU, and the processor 62 may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices. , Discrete gates or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor may be a microprocessor or any conventional processor.
  • the device for acquiring the operating state of the motor in this embodiment can be used to implement the technical solutions in the foregoing method embodiments, and its implementation principles and technical effects are similar, and will not be repeated here.
  • FIG. 8 is a second structural diagram of the device for acquiring the operating state of a motor provided by an embodiment of the application; referring to FIG. 8, the device 600 for acquiring the operating state of a motor in this embodiment, on the basis of the device shown in FIG. 7, further includes: The communication bus 63 is connected to the signal collector 64 and the motor 65 of the processor;
  • the signal collector 64 is configured to collect signals related to the operating state of the motor 65;
  • the processor 62 is specifically configured to perform the following operations when acquiring a signal related to the running state of the motor:
  • the signal related to the operating state of the motor 65 is obtained from the signal collector 64.
  • the signal collector 64 may include at least one of the following: a microphone, an accelerometer, a vibration sensor, a piezoelectric crystal, a barometer, an ammeter, and an inertial measurement unit.
  • the device for acquiring the operating state of the motor in this embodiment can be used to implement the technical solutions in the foregoing method embodiments, and its implementation principles and technical effects are similar, and will not be repeated here.
  • the processor 62 is further configured to perform the following operations when acquiring characteristic information of the signal:
  • the signal is preprocessed.
  • the preprocessing includes normalizing the signal to obtain a normalized signal.
  • the characteristic information includes intensity information and/or distribution information of the signal, and when the processor 62 obtains the characteristic information of the signal, and obtains the operating state of the motor according to the characteristic information, specifically Used to perform the following operations:
  • the operating state of the motor is acquired.
  • the processor 62 acquires the intensity information and/or distribution information of the signal, it is specifically configured to perform the following operations:
  • the processor 62 is specifically configured to perform the following operations when acquiring the normalized amplitude spectrum corresponding to the signal:
  • the processor 62 is specifically configured to perform the following operations when acquiring the intensity information of the signal according to the normalized amplitude spectrum corresponding to the signal:
  • the intensity information is obtained according to the normalized amplitude spectrum corresponding to the signal and M preset thresholds, where the intensity information includes M intensity values, and M is an integer greater than or equal to 1.
  • the processor 62 is specifically configured to perform the following operations when acquiring the intensity information according to the normalized amplitude spectrum corresponding to the signal and M preset thresholds:
  • the sum of the difference between each amplitude value in the normalized amplitude spectrum and the first preset threshold is obtained, and each amplitude value is compared with the first preset threshold.
  • the sum of the difference of the threshold value be the intensity value corresponding to the first preset threshold in the intensity information.
  • the preset threshold is any value from 0 to 1.
  • the processor 62 is specifically configured to perform the following operations when acquiring the distribution information of the signal according to the normalized amplitude spectrum corresponding to the signal:
  • the first K distribution values in the distribution signal are distribution information of the signal, where K is an integer greater than or equal to 1.
  • the processor 62 is specifically configured to perform the following operations when obtaining the distributed signal corresponding to the signal according to the normalized amplitude spectrum corresponding to the signal:
  • the distributed signal corresponding to the signal is obtained.
  • the processor 62 is specifically configured to perform the following operations when obtaining the distributed signal corresponding to the signal according to the signal after down-sampling the normalized amplitude spectrum:
  • the processor 62 is specifically configured to perform the following operations when down-sampling the normalized amplitude spectrum to obtain a signal after down-sampling the normalized amplitude spectrum:
  • Triangular window down-sampling is performed on the normalized amplitude spectrum to obtain a signal after the down-sampled normalized amplitude spectrum.
  • the triangular window downsampling is uniform triangular window downsampling.
  • the motor includes at least one of the operating states
  • the characteristic information of the signal is acquired
  • the processor 62 is specifically configured to execute when acquiring the operating state of the motor according to the characteristic information Do as follows:
  • the characteristic information of the signal is acquired, and the operating state of the motor corresponding to the characteristic information is acquired according to the characteristic information according to a preset corresponding relationship.
  • the preset corresponding relationship includes multiple preset characteristic information and each The operating status corresponding to the preset feature information.
  • the processor 62 is specifically configured to perform the following operations when acquiring the operating state of the motor corresponding to the characteristic information according to a preset correspondence relationship according to the characteristic information:
  • Target preset feature information corresponding to the feature information, where the target preset feature information is feature information in the multiple preset feature information;
  • the operating state corresponding to the target preset information in the preset correspondence is the operating state of the motor.
  • the at least one of the operating states is at least one of the following:
  • the motor starts abnormally, the motor has a risk of abnormal operation after the first time period, the motor operates normally, the motor has a risk of abnormal operation after the second time period, and the noise during the operation of the motor is excessive ;
  • the first duration is less than the second duration.
  • the processor 62 is specifically configured to perform the following operations when acquiring the operating state of the motor corresponding to the characteristic information according to the characteristic information:
  • a machine learning algorithm is used to obtain the operating state of the motor corresponding to the characteristic information.
  • the machine learning algorithm is a neural network algorithm
  • the machine learning model is a neural network model
  • the signal related to the running state of the motor is one or more of the following:
  • the noise signal of the motor and the current signal of the motor are signals measured by an inertial measurement unit provided on the outer wall of the motor.
  • the noise signal of the motor is collected by a microphone, an accelerometer, a vibration sensor, a piezoelectric crystal, and/or a barometer.
  • the processor 62 is specifically configured to perform the following operations when acquiring a signal related to the running state of the motor:
  • the processor 62 is specifically configured to perform the following operations when acquiring characteristic information of the signal and acquiring the operating state of the motor according to the characteristic information:
  • the device for acquiring the operating state of the motor in this embodiment can be used to implement the technical solutions in the foregoing method embodiments, and its implementation principles and technical effects are similar, and will not be repeated here.
  • An embodiment of the present application also provides a computer-readable storage medium, including a program or instruction, and when the program or instruction runs on a computer, the method described in the foregoing method embodiment is executed.
  • An embodiment of the present application also provides an electronic device, including a motor and the above-mentioned obtaining device for obtaining the operating state of the motor.
  • the device for acquiring the operating state of the motor in this embodiment can be used to implement the technical solutions in the foregoing method embodiments, and its implementation principles and technical effects are similar, and will not be repeated here.
  • the electronic device can be any electronic device that needs to be driven by a motor, such as a radar, a drone, an electric fan system, etc., where the radar can be a mechanical scanning lidar, and the drone can be a Human aircraft, unmanned vehicles, unmanned ships, etc., are not limited here.
  • a motor such as a radar, a drone, an electric fan system, etc.
  • the radar can be a mechanical scanning lidar
  • the drone can be a Human aircraft, unmanned vehicles, unmanned ships, etc., are not limited here.
  • a person of ordinary skill in the art can understand that all or part of the steps in the foregoing method embodiments can be implemented by a program instructing relevant hardware.
  • the aforementioned program can be stored in a computer readable storage medium. When the program is executed, it executes the steps including the foregoing method embodiments; and the foregoing storage medium includes: ROM, RAM, magnetic disk, or optical disk and other media that can store program codes.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Control Of Electric Motors In General (AREA)

Abstract

一种电机运行状态的获取方法和装置，该方法包括：获取与电机的运行状态相关的信号（S101）；获取该信号的特征信息，根据该特征信息获取所述电机的运行状态（S102）。该电机运行状态的获取方法和装置可以高效准确的获取电机的运行状态。

Description

电机运行状态的获取方法和装置 技术领域
本申请涉及计算机技术，尤其涉及一种电机运行状态的获取方法和装置。
背景技术
对电机运行状态进行监测，是保证使用电机驱动的设备能够可靠稳定运行的重要因素。
目前对电机运行状态进行监测的方法，一般为人耳听音的方式来判断，即人耳通过听电机运行过程中的噪音来判断电机运行状态。其中，用户可以站在电机附近听电机运行过程中的噪音，也可将电机运行过程中的噪音录制成音频文件离线听。
但是采用人耳听音的方式判断电机的运行状态，判断不够准确且效率低。
发明内容
本申请实施例提供一种电机运行状态的获取方法和装置，可以高效准确的获取电机的运行状态。
第一方面，本申请实施例提供一种电机运行状态的获取方法，包括：
获取与电机的运行状态相关的信号；
获取所述信号的特征信息，根据所述特征信息获取所述电机的运行状态。
第二方面，本申请实施例提供一种电机运行状态的获取装置包括：存储器、处理器和通信总线，所述存储器和所述处理器通过所述通信总线连接；
存储器，用于存储计算机程序；
处理器，用于调用所述计算机程序，执行如下操作：
获取与电机的运行状态相关的信号；
获取所述信号的特征信息，根据所述特征信息获取所述电机的运行状 态。
第三方面，本申请实施例提供一种计算机可读存储介质，包括程序或指令，当所述程序或指令在计算机上运行时，第一方面以及各可能的设计中所述的方法被执行。
第四方面，本申请实施例提供一种电子装置，包括电机和电机运行状态的获取装置，所述获取装置用于获取所述电机的运行状态。
本申请通过获取与电机运行状态相关的信号的特征信息，来获取电机的运行状态，无需人耳听音的方式判断电机的运行状态，且由于电机运行状态相关的信号的特征信息可以客观的反映电机的运行状态，因此，本申请的获取电机运行状态的方法的效率和准确率均较高。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作一简单地介绍，显而易见地，下面描述中的附图是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1为本申请实施例的电机运行状态的获取方法的流程图一；
图2为本申请实施例提供的电机的噪声信号图；
图3为本申请实施例的电机运行状态的获取方法的流程图二；
图4为本申请实施例提供的三角窗降采样的示意图；
图5为本申请实施例的电机运行状态的获取方法的流程图三；
图6为图2中的信号的N段子信号示意图；
图7为本申请实施例提供的电机运行状态的获取装置的结构示意图一；
图8为本申请实施例提供的电机运行状态的获取装置的结构示意图二。
具体实施方式
为使本申请实施例的目的、技术方案和优点更加清楚，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有作出创造性劳动前提 下所获得的所有其他实施例，都属于本申请保护的范围。
具体地，本申请中，“至少一个”是指一个或者多个，“多个”是指两个或两个以上。“和/或”，描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B的情况，其中A,B可以是单数或者复数。字符“/”一般表示前后关联对象是一种“或”的关系。“以下至少一项(个)”或其类似表达，是指的这些项中的任意组合，包括单项(个)或复数项(个)的任意组合。例如，a,b,或c中的至少一项(个)，可以表示：a,b,c,a-b,a-c,b-c,或a-b-c，其中a,b,c可以是单个，也可以是多个。本申请中术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。本实施例中的“x～y之间的值”，除了包括位于(x，y)这一区间中的值，还包括x和y两个端点值。
本申请实施例提供了一种高效且准确的获取电机运行状态的方法。
下面采用具体的实施例对本申请实施例提供的电机运行状态的获取方法进行说明。
图1为本申请实施例的电机运行状态的获取方法的流程图一。本实施例的执行主体可为电机运行状态的获取装置，该电机运行状态的获取装置可通过硬件或软件实现。本实施例的方法包括：
步骤S101、获取与电机的运行状态相关的信号；
具体地，在电机运行过程中，获取与电机的运行状态相关的信号。与电机的运行状态相关的信号为如下中的一种或多种：电机的噪声信号，电机的电流信号，设置在电机外壁上的惯性测量单元测量得到的信号。其中，电机的噪声信号可通过在麦克风麦克风、加速度计、振动传感器、压电晶体和/或气压计采集；电机的电流信号可通过电流表测量得到；惯性测量单元的信号可以诸如是振动信号或电流信号。可以理解，与电机的运行状态相关的信号也可以是其他可以反映电机运行状态的信号，本实施例仅为示例性说明。
本实施例中的电机可为未装配至设备上的电机，也可为已经装配至设备上的电机，装配有电机的设备可为任意的需要电机驱动的电子设备，比如雷达、无人机、电风扇系统等，其中，雷达可以为机械扫描式激光雷达， 无人机可以为无人飞行器、无人车、无人船等，在此不作限定。
图2为本申请实施例提供的电机的噪声信号图。如图2所示，图2中的噪声信号为原始噪声信号，即为时域上的噪声信号。
步骤S102、获取该信号的特征信息，根据该特征信息获取该电机的运行状态。
具体地，在获取到与电机的运行状态相关的信号后，可对与电机的运行状态相关的信号进行处理，得到电机的运行状态相关的信号的特征信息。其中，可直接对与电机的运行状态相关的信号进行处理，得到电机的运行状态相关的信号的特征信息，还可以从与电机的运行状态相关的信号中的提取部分信号进行处理，得到电机的运行状态相关的信号的特征信息。
可以理解的是，在直接对与电机的运行状态相关的信号进行处理的情况下，该与电机的运行状态相关的信号可以是采集到的与电机的运行状态相关的信号，也可以是从采集到的与电机的运行状态相关的信号中提取出来的信号。在从与电机的运行状态相关的信号中的提取部分信号进行处理的情况下，该与电机的运行状态相关的信号可以是采集到的与电机的运行状态相关的信号。
其中，与电机的运行状态相关的信号的特征信息可为如下中的至少一种：强度信息、分布信息。在其他实施方式中，与电机的运行状态相关的信号的特征信息也可以为其他信息，例如功率信息等，在此不作限定。其中，强度信息可以是根据时域上的与电机的运行状态相关的信号获取，还可以根据频域上的与电机的运行状态相关的信号获取，频域上的与电机的运行状态相关的信号可为与电机的运行状态相关的信号对应的幅度谱。分布信息可根据与电机的运行状态相关的信号对应的分布信号获取，该分布信号可为与电机的运行状态相关的信号对应的倒谱信号。在其他实施方式中，也可以根据其他合适的算法获得时域上与电机的运行状态相关的信号对应的分布信号。
在一种方式中，对与电机的运行状态相关的信号进行预处理后，再获取与电机的运行状态相关的信号的特征信息。其中，对与电机的运行状态相关的信号进行预处理可包括对该与电机的运行状态相关的信号或者从该与电机的运行状态相关的信号中提取出的信号进行归一化处理，获取归 一化信号。此时，可根据归一化信号获取与电机的运行状态相关的信号的特征信息。
由于获取的是与电机的运行状态相关的信号的特征信息，因此该特征信息可以反映出电机的运行状态，那么在获取该特征信息后，便可根据该特征信息获取该电机的运行状态。
可以理解的是，电机可具有至少一种运行状态，获取到的与电机的运行状态相关的信号的特征信息不同，对应的电机的运行状态可不同。
其中，可以根据预设特征信息与电机运行状态的对应关系，获取与该特征信息对应的电机的运行状态，还可以根据机器学习算法，获取与该特征信息对应的电机的运行状态。
具体地，在一种实施方式中：获取与电机的运行状态相关的信号的特征信息，根据该特征信息获取电机的运行状态，包括：获取与电机的运行状态相关的信号的特征信息，根据该特征信息，按照预设对应关系获取与该特征信息对应的电机的运行状态，该预设对应关系为包括多种预设特征信息和每种预设信息对应的运行状态。
其中，根据该特征信息，按照预设对应关系获取与该特征信息对应的该电机的运行状态，包括：确定该特征信息所对应的目标预设特征信息；确定该预设对应关系中与该目标预设特征信息对应的运行状态为与电机的运行状态。
示例性地，电机可具有如下几种但不限于如下几种运行状态：
(1)电机启动异常。
此时，与电机的运行状态相关的信号的特征信息对应该信号的总强度较小，例如该信号的总强度小于预先定义的强度值。
(2)电机在第一时长后存在运行异常的风险。
此时，与电机的运行状态相关的信号的特征信息对应该信号的总强度适中，例如该信号的总强度处于正常范围之内，但该信号在一些频率上的强度相对较大，例如该信号在一些频率上的强度超出正常范围之外。
(3)电机运行正常；
此时，与电机运行状态相关的信号的特征信息对应该信号的总强度适中，例如该信号的总强度处于正常范围之内，同时该信号在各频率上的强 度大部分或全部处于正常范围之内。
(4)电机在第二时长后存在运行异常的风险；其中，第一时长小于第二时长；
此时，与电机的运行状态相关的信号的特征信息对应该信号的总强度适中，例如该信号的总强度处于正常范围之内，但该信号在多个频率上的强度稍微偏大，例如该信号在多个频率上的强度略超出正常范围。
(5)电机运行过程中的噪声过大。
此时，与电机的运行状态相关的信号的特征信息对应该信号的总强度较大，例如超出正常范围之外。
在另一种实施方式中，也可采用机器学习的方法，根据该特征信息来获取与该特征信息对应的电机的运行状态。该实施方式获取的电机的运行状态也可为如上示例中(1)～(5)中的任意一种状态。
具体的，在采用机器学习的方法，根据该特征信息来获取与该特征信息对应的电机的运行状态的情况下：
根据该特征信息，获取与该特征信息对应的电机的运行状态，包括：根据该特征信息和机器学习模型，采用机器学习算法获取与该特征信息对应的该电机的运行状态。其中，机器学习模型为采用该机器学习算法，基于多个训练样本和训练样本的标签训练得到的，训练样本包括与第一电机的运行状态相关的信号的特征信息，该训练样本的标签用于指示第一电机的运行状态。
具体地，将该步骤中得到的该特征信息作为机器学习模型的输入，采用机器学习算法计算后，输出一目标标签，该目标标签用于指示电机的运行状态。
可以理解的是，第一电机可为本实施中的电机，也可为其它的电机。训练样本中的与第一电机的运行状态相关的信号的特征信息包括的特征种类与该步骤中得到的与电机的运行状态相关的信号的特征信息包括的特征种类相同。训练样本的标签可为指示第一电机的运行状态的字符串，相应地，目标标签为指示本实施中的电机的运行状态的字符串。
在机器学习算法为神经网络算法时，机器学习模型为神经网络模型。其中，神经网络算法可为如下中的任意一种：反向传播(Back Propagation， 简称BP)神经网络、栈自动编码机(Stacked AutoEncoder，简称SAE)神经网络、长短时记忆(Long Short Term Memory Network,LSTM)神经网络、循环神经网络(Recurrent Neural Networks，简称RNN)、卷积神经网络。
本实施例中通过获取与电机运行状态相关的信号的特征信息，来获取电机的运行状态，无需人耳听音的方式判断电机的运行状态，且由于电机运行状态相关的信号的特征信息可以客观的反映电机的运行状态，因此，本实施例的方法获取电机运行状态的效率和准确率均较高。
下面对与电机的运行状态相关的信号的特征信息包括强度信息和/或分布信息时，对上一实施例的一种具体实现。图3为本申请实施例提供的电机运行状态的获取方法的流程图二。参见图3，本实施例的方法包括：
步骤S201、获取与电机的运行状态相关的信号。
具体地，该步骤的具体实现参见上一实施例中的步骤S101，此处不再赘述。
步骤S202、获取与电机的运行状态相关的信号的强度信息和/或分布信息。
具体地，获取与电机的运行状态相关的信号的强度信息和/或分布信息，包括：
a1、获取与电机的运行状态相关的信号对应的归一化幅度谱。
具体地，可对与电机的运行状态相关的信号进行快速傅里叶变换(Fast Fourier Transform，简称FFT)得到与电机的运行状态相关的信号对应的幅度谱。
接着，对与电机的运行状态相关的信号对应的幅度谱做归一化，得到与电机的运行状态相关的信号对应的归一化幅度谱。在一种方式中，可将与电机的运行状态相关的信号对应的幅度谱中各幅度值除以零频率对应的幅度值，得到与电机的运行状态相关的信号对应的归一化幅度谱。其中，零频率对应的幅度值可称为幅度谱的直流分量，即该步骤中的归一化方法为采用幅度谱中的直流分量去除其它分量。
a 2、根据与电机的运行状态相关的信号对应的归一化幅度谱，获取该信号的强度信息和/或分布信息。
下面分别对根据该归一化幅度谱获取该信号的强度信息和/或分布信息进行说明。
首先对根据该归一化幅度谱，获取该信号的强度信息的方法进行说明。
具体地，根据该信号对应的归一化幅度谱，获取该信号的强度信息，包括：根据该信号对应的归一化幅度谱以及M个预设阈值，获取该强度信息，该强度信息包括M个强度值，M为大于等于1的整数。
其中，根据信号对应的归一化幅度谱以及M个预设阈值，获取该强度信息，包括：对于M个预设阈值中的任意一个第一预设阈值，获取该归一化幅度谱中的各幅度值与第一预设阈值的差值之和，该归一化幅度谱中的各幅度值与第一预设阈值的差值之和为该强度信息中与该第一预设阈值对应的强度值。可以理解，强度信息中包括的强度值的个数可以根据预设阈值的个数确定。
可选地，M个预设阈值中的每个预设阈值可为0～1中的任意值。
示例性地，M＝3，此时强度信息包括3个强度值；进一步地，预先设置3个预设阈值为0，0.1，0.2。将该归一化幅度谱中幅度值大于0的各幅度值相加，得到该强度信息中与预设阈值0对应的强度值E 11，将该归一化幅度谱中的各幅度值与0.1的差值相加，得到该强度信息中与预设阈值0.1对应的强度值E 12，将该归一化幅度谱中的各幅度值与0.2的差值相加，得到该强度信息中与预设阈值0.2对应的强度值E 13。E 11、E 12和E 13为根据该归一化幅度谱得到的强度信息中的3个强度值，亦即，E 11、E 12和E 13组成了上述的强度信息。
其次，对根据与电机的运行状态相关的信号对应的归一化幅度谱，获取该信号的分布信息进行说明。
具体地，根据与电机的运行状态相关的信号对应的归一化幅度谱，获取该信号的分布信息，包括：
b1、根据该信号对应的归一化幅度谱，得到该信号对应的分布信号。
具体地，根据该信号对应的归一化幅度谱，得到该信号对应的分布信号，包括：
在一种方式中，对该归一化幅度谱进行反余弦变换，得到该信号对应的分布信号。还可以对该归一化幅度谱进行其它种类的变换(比如反傅里 叶变换等)得到该信号对应的分布信号，本实施例中并不限制。此处的分布信号可为于该与电机的运行状态相关的信号对应的倒谱信号。其中，与电机运行状态相关的信号对应的分布信号可以表征与电机运行状态相关的信号的构成，表现了与电机运行状态相关的信号的频率分布情况。
在另一种方式中，为了降低计算的复杂度，根据该信号对应的归一化幅度谱，得到该信号对应的分布信号，包括：
b11、对该归一化幅度谱进行降采样，得到对该归一化幅度谱降采样后的信号。
可对该归一化幅度谱进行三角窗降采样，得到对该归一化幅度谱降采样后的信号。其中，三角窗降采样可为三角窗均匀降采样。其中，三角窗降采样可参照图4中的加三角窗降采样的示意图。
在一种具体的实施例中，可通过如下的公式，得到对该归一化幅度谱降采样后的信号：
Figure PCTCN2019074606-appb-000001
win(i)＝|D-|i||            (2)
Figure PCTCN2019074606-appb-000002
Figure PCTCN2019074606-appb-000003
其中，g(j)为该归一化幅度谱加三角窗降采样后的信号，floor(*)为取整运算，R为电机转速，fs为该信号的采样频率(即为步骤S201中获取的与电机运行状态的信号的采样频率)，N为得到第一归一化幅度谱的第一子信号的长度，G(*)为第一归一化幅度谱，win(i)为三角窗函数；j为该归一化幅度谱三角窗降采样后的信号的横坐标，其中，该横坐标可为采样点数，也可为频率，在此不作限定；
可以理解的是，还可以对该归一化幅度进行加其它种类的窗的降采样，得到对该归一化幅度谱降采样后的信号，本实施例中并不限制。
b12、根据对该归一化幅度谱降采样后的信号，得到与电机的运行状态相关的信号对应的分布信号。
具体地，可对该归一化幅度谱降采样后的信号进行反余弦变换，得到该分布信号。还可以对该归一化幅度谱降采样后的信号进行其它种类的变换(比如反傅里叶变换等)得到该分布信号，本实施例中并不限制。
具体地，对该归一化幅度谱降采样后的信号进行反余弦变换，得到该分布信号可通过如下的公式实现：
Figure PCTCN2019074606-appb-000004
Figure PCTCN2019074606-appb-000005
其中，C k为该分布信号，k为该分布信号的横坐标，该横坐标可为倒频率。
b2、确定该分布信号中的前K个分布值为与电机的运行状态相关的信号的分布信息。
示例性地，若该分布信号的表达式如公式(5)所示，则将k＝1,…,K依次带入公式(5)，可得到该分布信号的前K个分布值：C 1,…,C k,…,C K，则C 1,…,C k,…,C K组成了该分布信号的分布信息。
步骤S203、根据该信号的强度信息和/或分布信息，获取电机的运行状态。
具体地，在一种实施方式中：根据该信号的强度信息和/或分布信息，获取与该能量信息和/或该分布信息对应的电机的运行状态，包括：
根据该信号的强度信息和/或分布信息，以及机器学习模型，采用机器学习算法获取与该强度信息和/或该分布信息对应的电机的运行状态。其中，机器学习模型为采用该机器学习算法，基于多个训练样本和训练样本的标签训练得到的，训练样本包括与第一电机的运行状态相关的信号的强度信息和/或分布信息，该训练样本的标签用于指示第一电机的运行状态。
具体地，将该步骤中得到的强度信息和/或分布信息作为机器学习模型的输入，采用机器学习算法计算后，输出一目标标签，该目标标签用于指示电机的运行状态。
可以理解的是，当根据该信号的强度信息以及机器学习模型，采用机器学习算法获取与强度信息对应的该电机的运行状态时，训练样本包括与第一电机的运行状态相关的信号的强度信息。当根据该信号的分布信息以 及机器学习模型，采用机器学习算法获取与分布信息对应的该电机的运行状态时，训练样本包括与第一电机的运行状态相关的信号的分布信息。当根据该信号的分布信息、强度信息以及机器学习模型，采用机器学习算法获取与分布信息和强度信息对应的该电机的运行状态时，训练样本包括与第一电机的运行状态相关的信号的强度信息和分布信息。
该种方式采用机器学习的方法获取的电机的运行状态，得到的电机的运行状态的准确率比较高。
在另一种实施方式中：根据该信号的强度信息和/或分布信息，获取与该强度信息和/或该分布信息对应的电机的运行状态，存在如下三种情况：
Ⅰ、根据该信号的强度信息，获取与该强度信息对应的电机的运行状态。
具体地，根据该信号的强度信息，获取与该强度信息对应的电机的运行状态，包括：根据该信号的强度信息，按照预设对应关系获取与该强度信息对应的电机的运行状态，该预设对应关系为包括多种预设强度信息和每种预设强度信息对应的运行状态。具体为：确定与该强度信息所对应的目标预设强度信息；确定该预设对应关系中与该目标预设强度信息对应的运行状态为电机的运行状态。
示例性地，若强度信息中包括与零值对应的第一强度值，与第二预设阈值对应的第二强度值和与第一预设阈值对应的第三强度值，第一预设阈值小于第二预设阈值，预设对应关系中包括的5种预设强度信息以及5种预设强度信息各自对应的运行状态如下：
(1)第一种预设强度信息：第一强度值小于第一预设强度值；第一种预设强度信息对应的运行状态为电机启动异常。
具体地，第一预设强度值可为0.3～0.7之间的任意一个值，比如为0.5。
由于第一强度值为与零值对应的强度值，根据步骤S202中获取强度值的方法可知，第一强度值为与电机的运行状态相关的信号的总强度值，第一强度值则可指示与电机的运行状态相关的信号的总强度值，因此，当第一强度值小于第一预设强度值时，可认为与电机的运行状态相关的信号的总强度值偏小，因此，此时对应的运行状态可为电机启动异常。
(2)第二种预设强度信息：第一强度值大于等于第一预设强度值且 小于等于第二预设强度值以及第二强度值大于第三预设强度值；第二种预设强度信息对应的运行状态为在第一时长内存在运行异常的风险。
具体地，第二预设强度值可为2.5～3.5中的任意一个值，比如为3。第三预设强度值可为0.02～0.07中的任意一个值，比如为0.05。
由于第二强度值为与第二预设阈值对应的强度值，第二预设阈值大于第一预设阈值，根据步骤S202中获取强度值的方法可知，第二强度值可指示与电机的运行状态相关的信号中强度相对较大的信号段的总强度。因此，当第一强度值大于等于第一预设强度值且小于等于第二预设强度值时，可认为与电机的运行状态相关的信号的总强度适中，第二强度值大于第三预设强度值，可认为与电机的运行状态相关的信号中强度相对较大的信号段的总强度较大，与电机的运行状态相关的信号中某些频率位置的强度相对较大，因此，此时对应的运行状态可为在第一时长内存在运行异常的风险。
(3)第三种预设强度信息：第一强度值大于等于第一预设强度值且小于等于第二预设强度值，第二强度值小于等于第三预设强度值以及第三强度值与第二强度值之差小于等于第四预设强度值；第三种预设强度信息对应的运行状态为电机运行正常。
具体地，第四预设强度值可为0.3～0.7中的任意一个值，比如为0.5。
由于第三强度值为与第一预设阈值对应的强度值，第二预设阈值大于第一预设阈值，根据步骤S202中获取强度值的方法可知，第三强度值可指示与电机的运行状态相关的信号强度相对适中的信号段的总强度，因此，当第一强度值大于等于第一预设强度值且小于等于第二预设强度值时，可认为与电机的运行状态相关的信号的总强度适中，第二强度值小于等于第三预设强度值，可认为与电机的运行状态相关的信号中强度相对较大的信号段的总强度较小，即与电机的运行状态相关的信号中不存在或者很少频率位置的强度相对较大，第三强度值与第二强度值之差小于第四预设强度值，说明了电机的运行状态相关的信号中强度相对适中的信号段的总强度也较小，即与电机的运行状态相关的信号中不存在或者很少频率位置的强度相对适中，因此，此时对应的运行状态可为电机运行正常。
(4)第四种预设强度信息：第一强度值大于等于第一预设强度值且 小于等于第二预设强度值，第二强度值小于等于第三预设强度值以及第三强度值与第二强度值之差大于等于第四预设强度值；第四种预设强度信息对应的运行状态为在第二时长内存在运行异常的风险。
具体地，当第一强度值大于等于第一预设强度值且小于等于第二预设强度值时，可认为与电机的运行状态相关的信号的总强度适中，第二强度值小于等于第三预设强度值，可认为与电机的运行状态相关的信号中强度相对较大的信号段的总强度较小，即与电机的运行状态相关的信号中不存在或者很少频率位置的强度相对较大，第三强度值与第二强度值之差大于第四预设强度值，说明与电机的运行状态相关的信号中强度相对适中的信号段总强度较大，即与电机的运行状态相关的信号中较多的频率位置的强度相对适中，因此，此时对应的运行状态可为在第二时长内存在运行异常的风险。其中，第二时长大于第一时长，即与电机的运行状态相关的信号中某些频率位置的强度相对较大，相对于与电机的运行状态相关的信号中较多的频率位置的强度相对适中，电机更容易出现异常。
(5)第五种预设强度信息：若第一强度值大于第二预设强度值。第四种预设强度信息对应的运行状态为电机的噪声过大。
具体地，当第一强度值大于第二预设强度值时，可认为与电机的运行状态相关的信号的总强度较大，因此，此时对应的运行状态可为电机的噪声过大。
在该示例下：根据该信号的强度信息，按照预设对应关系获取与该强度信息对应的电机的运行状态，该预设对应关系为包括多种预设强度信息和每种预设强度信息对应的运行状态，包括：
根据该信号的强度信息包括的第一强度值、第二强度值和第三强度值，确定与该强度信息对应上述(1)～(5)中的哪种预设强度信息，若对应上述的第一种预设强度信息，则确定电机的运行状态为第一种预设强度信息对应的电机启动异常。
Ⅱ、根据该信号的分布信息，获取与该强度信息对应的电机的运行状态，包括：根据该信号的分布信息，按照预设对应关系获取与该分布信息对应的电机的运行状态，该预设对应关系为包括多种预设分布信息和每种预设分布信息对应的运行状态。具体为：确定与该分布信息所对应的目标 预设分布信息；确定该预设对应关系中与该目标预设分布信息对应的运行状态为电机的运行状态。
Ⅲ、根据该信号的分布信息和强度信息，获取与该分布信息和强度信息对应的电机的运行状态，包括：根据该信号的分布信息和强度信息，按照预设对应关系，获取与该分布信息和强度信息对应的电机的运行状态，该预设对应关系为包括多种预设特征信息和每种预设特征信息对应的运行状态。其中，多种预设特征信息为各预设强度信息和各预设分布信息的多种组合。具体为：确定与该分布信息和强度信息所对应的目标预设特征信息；确定该预设对应关系中与该目标预设特征信息对应的运行状态为电机的运行状态。
该种获取电机的运行状态的实施方式不需要训练机器学习模型，也不需要进行复杂的机器学习算法，对硬件的要求不高且获取电机的运行状态的效率相对于上一种方式更高。
本实施例中通过获取与电机运行状态相关的信号的强度信息和/或分布信息，来获取电机的运行状态，无需人耳听音的方式判断电机的运行状态，且由于电机运行状态相关的信号的强度信息和/或分布信息可以客观的反映电机的运行状态，因此，本实施例的方法获取电机运行状态的效率且准确率均较高。
下面对与电机的运行状态相关的信号的特征信息包括强度信息和/或分布信息时，对图1所示的实施例的另一种具体实现进行说明。图5为本申请实施例提供的电机运行状态的获取方法的流程图三。参见图5，本实施例的方法包括：
步骤S301、获取与电机的运行状态相关的信号。
具体地，该步骤的具体实现，参见图1所示的实施例中步骤S101，此处不再赘述。
步骤S302、提取与电机的运行状态相关的信号中的N段子信号，其中，N为大于等于1的整数。
具体地，在一种方式中，为了使得得到的该信号的强度信息和/或分布信息比较准确，N可取大于等于2的整数。
本实施例中的子信号为该信号中长度小于该信号的长度的一段子信 号，每段子信号的长度相同。可选地，每段子信号的采样点的个数可为20000个。
示例性地，若该信号为图2中的信号，则N段子信号为图2中的信号的N段子信号。图6为图2中的信号的N段子信号的示意图。参见图6，N段子信号包括：a～b之间的信号、c～d之间的信号。a～b之间的信号为图2中的信号的一段子信号，c～d之间的信号为图2中的信号的一段子信号。
步骤S303、根据该N段子信号，获取与电机的运行状态相关的信号的强度信息和/或分布信息。
具体地，根据该N段子信号，获取与电机的运行状态相关的信号的强度信息和/或分布信息，包括：
c1、获取N段子信号对应的N个归一化幅度谱，其中，N段子信号与N个归一化幅度谱一一对应。该N个归一化幅度谱即为该信号对应的归一化幅度谱。
具体地，针对N段子信号中的任意一段第一子信号，获取第一子信号对应的第一归一化幅度谱，包括：
c11、获取第一子信号对应的第一幅度谱。
具体地，可对第一子信号进行快速傅里叶变换(Fast Fourier Transform，简称FFT)得到第一子信号对应的第一幅度谱。
c12、对该第一幅度谱进行归一化，得到第一归一化幅度谱。
具体地，可将该第一幅度谱中各幅度值除以零频率对应的幅度值，得到第一归一化幅度谱。其中，零频率对应的幅度值可称为第一幅度谱的直流分量，即该步骤中的归一化方法为采用第一幅度谱中的直流分量去除其它分量。
c2、根据N个归一化幅度谱，获取与电机的运行状态相关的信号的强度信息和/或分布信息。
下面分别对根据N个归一化幅度谱，获取与电机的运行状态相关的信号的强度信息和分布信息进行说明。
首先对根据N个归一化幅度谱，获取与电机的运行状态相关的信号的强度信息的方法进行说明。
具体地，根据N个归一化幅度谱，获取与电机的运行状态相关的信号的强度信息，包括：根据N个归一化幅度谱，获取N组强度值，其中，N个归一化幅度谱和N组强度值一一对应。进一步地，每组强度值可以包括M个强度分量值，M为大于等于1的整数。也就是说，根据每个归一化幅度谱均能得到一组强度值，每组强度值包括M个强度分量值。
上述得到的N组强度值即为与电机的运行状态相关的信号的强度信息。下面对N组强度值的获取方法进行说明。具体地，针对N个归一化幅度谱中的任意一个第一归一化幅度谱，根据第一归一化幅度谱，获取第一组强度值，包括：对于M个预设阈值中的任意一个第一预设阈值，获取第一归一化幅度谱中的各幅度值与第一预设阈值的差值之和，第一归一化幅度谱中的各幅度值与第一预设阈值的差值之和为第一组强度信息中与第一预设阈值对应的强度分量值。可以理解，一组强度值中强度分量值的个数可以根据预设阈值的个数确定。
其中，每个预设阈值可为0～1中的任意值。
示例性地，M＝3，此时第一组强度值包括3个强度分量值；进一步地，预先设置3个预设阈值为0，0.1，0.2。将第一归一化幅度谱中强度值大于0的各幅度值相加，得到第一组强度值中与预设阈值0对应的强度分量值E 11，将第一归一化幅度谱中的各幅度值与0.1的差值之和相加得到第一组强度值中与预设阈值0.1对应的强度分量值E 12，将第一归一化幅度谱中的各幅度值与0.2的差值相加得到第一组强度值中与预设阈值0.2对应的强度分量值E 13。E 11、E 12和E 13为根据第一归一化幅度谱得到的第一组强度值中的3个强度分量值，亦即，E 11、E 12和E 13组成了上述的第一组强度值。
其次，对根据N个归一化幅度谱，获取与电机的运行状态相关的信号的分布信息的方法进行说明。
具体地，根据N个归一化幅度谱，获取与电机的运行状态相关的信号的分布信息，包括：
d 1、根据N个归一化幅度谱，获取与电机的运行状态相关的信号对应的N个分布信号，N个归一化幅度和N个分布信号一一对应。
在一种方式中，针对N个归一化幅度谱中的任意一个第一归一化幅度 谱，对第一归一化幅度谱进行反余弦变换，得到第一分布信号。还可以对第一归一化幅度进行其它种类的变换(比如反傅里叶变换等)得到第一分布信号，本实施例中并不限制。
在另一种方式中，为了降低计算的复杂度，针对N个归一化幅度谱中的任意一个第一归一化幅度谱，根据第一归一化幅度谱，获取第一分布信号，包括：
d 11、对第一归一化幅度谱进行降采样，得到对第一归一化幅度谱降采样后的信号。
其中，对第一归一化幅度谱进行降采样，得到对第一归一化幅度谱降采样后的信号，包括：对第一归一化幅度谱进行三角窗降采样，得到对第一归一化幅度谱降采样后的信号。其中，三角窗降采样可为三角窗均匀降采样。
可以理解的是，还可以对第一归一化幅度进行加其它种类的窗的降采样，得到对第一归一化幅度谱降采样后的信号，本实施例中并不限制。
d 12、根据对第一归一化幅度谱降采样后的信号，得到第一分布信号。
具体地，可对第一归一化幅度谱降采样后的信号进行反余弦变换，得到第一分布信号。还可以对第一归一化幅度进行其它种类的变换(比如反傅里叶变换等)得到第一分布信号，本实施例中并不限制。
d 2、根据N个分布信号，获取N组分布值，每组分布值包括K个分布分量值，N个分布信号和N组分布值一一对应；K为大于等于1的整数，可选地，K＝5。
具体地，根据每个分布信号，均能得到一组分布值，每组分布值包括K个分布分量值。该N组分布值即为与电机运行状态相关的信号的分布信息。
下面对获取分布信息的方法进行说明：针对N个分布信号中的任意一个第一分布信号，根据第一分布信号，获取第一组分布值，包括：确定第一分布信号中的前K个分布值为第一组分布值。
步骤S304、根据该信号的强度信息和/或分布信息，获取电机的运行状态。
具体地，在一种实施方式中：根据该信号的强度信息和/或分布信息， 获取与该强度信息和/或该分布信息对应的电机的运行状态，包括：
根据该信号的N组强度值和/或N组分布值，以及机器学习模型，采用机器学习算法获取与该N组强度值和/或N组分布值对应的电机的运行状态。其中，机器学习模型为采用该机器学习算法，基于多个训练样本和训练样本的标签训练得到的，训练样本包括与第一电机的运行状态相关的信号的N组强度值和N组分布值，该训练样本的标签用于指示第一电机的运行状态。
具体地，将该步骤中得到的N组强度值和N组分布值作为机器学习模型的输入，采用机器学习算法计算后，输出一目标标签，该目标标签用于指示电机的运行状态。
可以理解的是，当根据该信号的N组强度值以及机器学习模型，采用机器学习算法获取与强度信息对应的该电机的运行状态时，训练样本包括与第一电机的运行状态相关的信号的N组强度值。当根据该信号的N组分布值以及机器学习模型，采用机器学习算法获取与N组分布值对应的该电机的运行状态时，训练样本包括与第一电机的运行状态相关的信号的N组分布值。当根据该信号的N组强度值和N组分布值以及机器学习模型，采用机器学习算法获取与N组强度值和N组分布值对应的该电机的运行状态时，训练样本包括与第一电机的运行状态相关的信号的N组强度值和N组分布值。
该种方式采用机器学习的方法获取的电机的运行状态，得到的电机的运行状态的准确率比较高。
在另一种实施方式中：根据该信号的强度信息和/或分布信息，获取与该强度信息和/或该分布信息对应的电机的运行状态，存在如下三种情况：
Ⅰ、根据该信号的强度信息，获取与该强度信息对应的电机的运行状态。
具体地，根据该信号的强度信息，获取与该强度信息对应的电机的运行状态，包括：根据该信号的N组强度值，按照预设对应关系获取与该N组强度值对应的电机的运行状态，该预设对应关系为包括多种预设强度信息和每种预设强度信息对应的运行状态。具体为：确定与该N组强度值所对应的目标预设强度信息；确定该预设对应关系中与该目标预设强度信息 对应的运行状态为电机的运行状态。
示例性地，若N组强度值中的每组强度值中包括与零值对应的强度分量值，与第二预设阈值对应的强度分量值和与第一预设阈值对应的强度分量值，第一预设阈值小于第二预设阈值。预设对应关系与上一实施例中相应示例中相同。
根据该信号的强度信息包括的N组强度值，确定与N组强度值对应上述(1)～(5)中的哪种预设强度信息，若对应上述的第一种预设强度信息，则确定电机的运行状态为第一种预设强度信息对应的电机启动异常。
具体地，由于N组强度值中的每组强度值中均包括与零值对应的强度分量值，与第二预设阈值对应的强度分量值和与第一预设阈值对应的强度分量值，则一共存在N个与零值对应的强度分量值、N个与第二预设阈值对应的强度分量值和N个与第一预设阈值对应的强度分量值。
因此，可根据该信号的强度信息包括的N组强度值，可获取N个与零值对应的强度分量值的平均值，得到第一强度值，获取N个与第二预设阈值对应的强度分量值的平均值，得到第二强度值，获取N个与第一预设阈值对应的强度分量值的平均值，得到第三强度值。接着，根据第一强度值、第二强度值和第三强度值，确定与N组强度值对应上述(1)～(5)中的哪种预设强度信息。
此外，在其他实施例中，也可以根据其他合适的算法，例如中位值、均方差等，获取N个与零值对应的强度分量值的中位值或均方差，得到第一强度值，N个与第二预设阈值对应的强度分量值的中位值或均方差，得到第二强度值，N个与第一预设阈值对应的强度分量值的中位值或均方差，得到第三强度值。
Ⅱ、根据该信号的分布信息，获取与该强度信息对应的电机的运行状态，包括：根据该信号的N组分布值，按照预设对应关系获取与该N组分布值对应的电机的运行状态，该预设对应关系为包括多种预设分布信息和每种预设分布信息对应的运行状态。具体为：确定与该N组分布值所对应的目标预设分布信息；确定该预设对应关系中与该目标预设分布信息对应的运行状态为电机的运行状态。
Ⅲ、根据该信号的分布信息和强度信息，获取与该分布信息和强度信 息对应的电机的运行状态，包括：根据该信号的N组强度值和N组分布值，按照预设对应关系获取与该N组强度值和N组分布值对应的电机的运行状态，该预设对应关系为包括多种预设特征信息和每种预设特征信息对应的运行状态。其中，多种预设特征信息为各预设强度信息和各预设分布信息的多种组合。具体为：确定与N组强度值和N组分布值所对应的目标预设信息；确定该预设对应关系中与该目标预设信息对应的运行状态为电机的运行状态。
该种获取电机的运行状态的实施方式不需要训练机器学习模型，也不需要进行复杂的机器学习算法，对硬件的要求不高且获取电机的运行状态的效率相对于上一种方式更高。
本实施例中通过获取与电机运行状态相关的信号的强度信息和/或分布信息，来获取电机的运行状态，无需人耳听音的方式判断电机的运行状态，且由于电机运行状态相关的信号的强度信息和/或分布信息可以客观的反映电机的运行状态，因此，本实施例的方法获取电机运行状态的效率且准确率均较高。
进一步地，为增强获取的电机的运行状态的准确性，上述实施例中的“获取与电机的运行状态相关的信号”包括：在预设时长内获取多个与电机的运行状态相关的信号。上述实施例中的“根据与电机的运行状态相关的信号的特征信息，根据所述特征信息获取电机的运行状态”，包括：对于多个与电机的运行状态相关的信号中的一个第一信号，获取第一信号的特征信息；根据多个与电机的运行状态相关的信号各自对应的特征信息，获取所述电机的运行状态。
具体的，多个与电机的运行状态相关的信号可为相同种类的信号，也可为不相同种类的信号。
对于每个获取到的与电机运行状态相关的信号，均采用图3或图5所示的实施例的方法得到该信号的特征信息，根据该特征信息获取电机的运行状态，即在预设时长内获取到多个电机的运行状态，若获取到的多个电机的运行状态中目标运行状态的个数大于预设值，则认为电机的运行状态即为目标运行状态；这样可以增强上述运行状态获取方法的鲁棒性。可以理解的是，预设时长可以设置为0.5小时～1日之间的任意一个值，也可以 设置为3～5日，甚至更长时间，在其他实施例中，多个与电机的运行状态相关的信号可存在相部分重叠的信号，可以根据实际需要对预设时长进行设置，在此不作限定。
示例性地，在预设时长内获取了10个与该电机运行状态相关的信号，预设值为5。采用图3或图5所示的实施例的方法根据10个与该电机运行状态相关的信号进行了10次电机运行状态的获取，得到了10个电机运行状态。其中，10个电机运行状态中有6个电机运行状态为电机噪声过大，则确定该电机的运行状态为电机噪声过大。
以上结合图1～图6为本申请实施例提供的电机运行状态的获取方法进行了说明，下面采用具体的实施例对电机运行状态的获取装置进行说明。
图7为本申请实施例提供的电机运行状态的获取装置的结构示意图一；参见图7，本实施例的电机运行状态的获取装置600包括：存储器61、处理器62和通信总线63，所述存储器61和所述处理器62通过所述通信总线连接；
存储器61，用于存储计算机程序；
处理器62，用于调用所述计算机程序，执行如下操作：
获取与电机的运行状态相关的信号；
获取所述信号的特征信息，根据所述特征信息获取所述电机的运行状态。
其中，处理器62可以是CPU，该处理器62还可以是其他通用处理器、数字信号处理器(DSP)、专用集成电路(ASIC)、现场可编程门阵列(FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者是任何常规的处理器等。
本实施例的电机运行状态的获取装置，可以用于执行上述各方法实施例中的技术方案，其实现原理和技术效果类似，此处不再赘述。
图8为本申请实施例提供的电机运行状态的获取装置的结构示意图二；参见图8，本实施例的电机运行状态的获取装置600在图7所示的装置的基础上，还包括：通过所述通信总线63与所述处理器连接的信号采集器64和电机65；
所述信号采集器64，用于采集与所述电机65的运行状态相关的信号；
所述处理器62在获取与电机的运行状态相关的信号时，具体用于执行如下操作：
从所述信号采集器64获取与所述电机65的运行状态相关的信号。
其中，信号采集器64可包括如下中的至少一项：麦克风、加速度计、振动传感器、压电晶体、气压计、电流表、惯性测量单元。
本实施例的电机运行状态的获取装置，可以用于执行上述各方法实施例中的技术方案，其实现原理和技术效果类似，此处不再赘述。
可选地，所述处理器62在获取所述信号的特征信息时，还用于执行如下操作：
对所述信号进行预处理。
可选地，所述预处理包括对所述信号进行归一化处理，获取归一化信号。
可选地，所述特征信息包括所述信号的强度信息和/或分布信息，所述处理器62在获取所述信号的特征信息，根据所述特征信息获取所述电机的运行状态时，具体用于执行如下操作：
获取所述信号的强度信息和/或分布信息；
根据所述信号的强度信息和/或分布信息，获取所述电机的运行状态。
可选地，所述处理器62获取所述信号的强度信息和/或分布信息时，具体用于执行如下操作：
获取所述信号对应的归一化幅度谱；
根据所述信号对应的归一化幅度谱，获取所述信号的强度信息和/或分布信息。
可选地，所述处理器62在获取所述信号对应的归一化幅度谱时，具体用于执行如下操作：
获取所述信号对应的幅度谱；
将所述幅度谱中的各幅度值除以零频率对应的幅度值，得到所述归一化幅度谱。
可选地，所述处理器62在根据所述信号对应的归一化幅度谱，获取所述信号的强度信息时，具体用于执行如下操作：
根据所述信号对应的归一化幅度谱以及M个预设阈值，获取所述强 度信息，其中，所述强度信息包括M个强度值，M为大于等于1的整数。
可选地，所述处理器62在根据所述信号对应的归一化幅度谱以及M个预设阈值，获取所述强度信息时，具体用于执行如下操作：
对于M个预设阈值中的任意一个第一预设阈值，获取所述归一化幅度谱中的各幅度值与第一预设阈值的差值之和，所述各幅度值与第一预设阈值的差值之和为所述强度信息中与所述第一预设阈值对应的强度值。
可选地，所述预设阈值为0～1中的任意值。
可选地，所述处理器62在根据所述信号对应的归一化幅度谱，获取所述信号的分布信息时，具体用于执行如下操作：
根据所述信号对应的归一化幅度谱，得到所述信号对应的分布信号；
确定所述分布信号中的前K个分布值为所述信号的分布信息，其中，K为大于等于1的整数。
可选地，所述处理器62在根据所述信号对应的归一化幅度谱，得到所述信号对应的分布信号时，具体用于执行如下操作：
对所述归一化幅度谱进行降采样，得到对所述归一化幅度谱降采样后的信号；
根据对所述归一化幅度谱降采样后的信号，得到所述信号对应的分布信号。
可选地，所述处理器62在根据对所述归一化幅度谱降采样后的信号，得到所述信号对应的分布信号时，具体用于执行如下操作：
对所述归一化幅度谱降采样后的信号进行反余弦变换，得到所述信号对应的分布信号。
可选地，所述处理器62在对所述归一化幅度谱进行降采样，得到对所述归一化幅度谱降采样后的信号时，具体用于执行如下操作：
对所述归一化幅度谱进行三角窗降采样，得到对所述归一化幅度谱降采样后的信号。
可选地，所述三角窗降采样为三角窗均匀降采样。
可选地，所述电机包括至少一种所述运行状态，所述获取所述信号的特征信息，所述处理器62在根据所述特征信息获取所述电机的运行状态时，具体用于执行如下操作：
获取所述信号的特征信息，根据所述特征信息，按照预设对应关系获取与所述特征信息对应的所述电机的运行状态，所述预设对应关系为包括多种预设特征信息和每种预设特征信息对应的运行状态。
可选地，所述处理器62在根据所述特征信息，按照预设对应关系获取与所述特征信息对应的所述电机的运行状态时，具体用于执行如下操作：
确定所述特征信息所对应的目标预设特征信息，所述目标预设特征信息为所述多种预设特征信息中的特征信息；
确定所述预设对应关系中与所述目标预设信息对应的运行状态为所述电机的运行状态。
可选地，所述至少一种所述运行状态如下中的至少一项：
所述电机启动异常、所述电机在第一时长后存在运行异常的风险、所述电机运行正常、所述电机在第二时长后存在运行异常的风险、所述电机运行过程中的噪声过大；
其中，所述第一时长小于所述第二时长。
可选地，所述处理器62在根据所述特征信息获取与所述特征信息对应的所述电机的所述运行状态时，具体用于执行如下操作：
根据所述特征信息和机器学习模型，采用机器学习算法获取与所述特征信息对应的所述电机的所述运行状态。
可选地，所述机器学习算法为神经网络算法，所述机器学习模型为神经网络模型。
可选地，与所述电机的运行状态相关的信号为如下中的一种或多种：
所述电机的噪声信号、所述电机的电流信号，设置在所述电机外壁上的惯性测量单元测量得到的信号。
可选地，所述电机的噪声信号由麦克风、加速度计、振动传感器、压电晶体和/或气压计采集。
可选地，所述处理器62在获取与电机的运行状态相关的信号时，具体用于执行如下操作：
在预设时长内获取多个与电机的运行状态相关的信号；
所述处理器62在获取所述信号的特征信息，根据所述特征信息获取所述电机的运行状态时，具体用于执行如下操作：
对于所述多个信号中的一个第一信号，获取所述第一信号的特征信息；
根据所述多个信号各自对应的特征信息，获取所述电机的运行状态。本实施例的电机运行状态的获取装置，可以用于执行上述各方法实施例中的技术方案，其实现原理和技术效果类似，此处不再赘述。
本申请实施例还提供一种计算机可读存储介质，包括程序或指令，当所述程序或指令在计算机上运行时，上述方法实施例所述的方法被执行。
本申请实施例还提供一种电子装置，包括电机和如上所述的用于获取电机的运行状态的获取装置。本实施例的电机运行状态的获取装置，可以用于执行上述各方法实施例中的技术方案，其实现原理和技术效果类似，此处不再赘述。
在一种实施方式中，该电子装置可以为任意的需要电机驱动的电子设备，比如雷达、无人机、电风扇系统等，其中，雷达可以为机械扫描式激光雷达，无人机可以为无人飞行器、无人车、无人船等，在此不作限定。
本领域普通技术人员可以理解：实现上述各方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成。前述的程序可以存储于一计算机可读取存储介质中。该程序在执行时，执行包括上述各方法实施例的步骤；而前述的存储介质包括：ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是：以上各实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述各实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的范围。

Claims (49)

  1. 一种电机运行状态的获取方法，其特征在于，包括：
    获取与电机的运行状态相关的信号；
    获取所述信号的特征信息，根据所述特征信息获取所述电机的运行状态。
  2. 根据权利要求1所述的方法，其特征在于，获取所述信号的特征信息，还包括：
    对所述信号进行预处理。
  3. 根据权利要求2所述的方法，其特征在于，所述预处理包括对所述信号进行归一化处理，获取归一化信号。
  4. 根据权利要求1所述的方法，其特征在于，所述特征信息包括所述信号的强度信息和/或分布信息，所述获取所述信号的特征信息，根据所述特征信息获取所述电机的运行状态包括：
    获取所述信号的强度信息和/或分布信息；
    根据所述信号的强度信息和/或分布信息，获取所述电机的运行状态。
  5. 根据权利要求4所述的方法，其特征在于，获取所述信号的强度信息和/或分布信息，包括：
    获取所述信号对应的归一化幅度谱；
    根据所述信号对应的归一化幅度谱，获取所述信号的强度信息和/或分布信息。
  6. 根据权利要求5所述的方法，其特征在于，获取所述信号对应的归一化幅度谱，包括：
    获取所述信号对应的幅度谱；
    将所述幅度谱中的各幅度值除以零频率对应的幅度值，得到所述归一化幅度谱。
  7. 根据权利要求5所述的方法，其特征在于，所述根据所述信号对应的归一化幅度谱，获取所述信号的强度信息，包括：
    根据所述信号对应的归一化幅度谱以及M个预设阈值，获取所述强度信息，其中，所述强度信息包括M个强度值，M为大于等于1的整数。
  8. 根据权利要求7所述的方法，其特征在于，所述根据所述信号对 应的归一化幅度谱以及M个预设阈值，获取所述强度信息，包括：
    对于M个预设阈值中的任意一个第一预设阈值，获取所述归一化幅度谱中的各幅度值与第一预设阈值的差值之和，所述各幅度值与第一预设阈值的差值之和为所述强度信息中与所述第一预设阈值对应的强度值。
  9. 根据权利要求8所述的方法，其特征在于，所述预设阈值为0～1中的任意值。
  10. 根据权利要求5所述的方法，其特征在于，所述根据所述信号对应的归一化幅度谱，获取所述信号的分布信息，包括：
    根据所述信号对应的归一化幅度谱，得到所述信号对应的分布信号；
    确定所述分布信号中的前K个分布值为所述信号的分布信息，其中，K为大于等于1的整数。
  11. 根据权利要求10所述的方法，其特征在于，所述根据所述信号对应的归一化幅度谱，得到所述信号对应的分布信号，包括：
    对所述归一化幅度谱进行降采样，得到对所述归一化幅度谱降采样后的信号；
    根据对所述归一化幅度谱降采样后的信号，得到所述信号对应的分布信号。
  12. 根据权利要求11所述的方法，其特征在于，所述根据对所述归一化幅度谱降采样后的信号，得到所述信号对应的分布信号，包括：
    对所述归一化幅度谱降采样后的信号进行反余弦变换，得到所述信号对应的分布信号。
  13. 根据权利要求11所述的方法，其特征在于，所述对所述归一化幅度谱进行降采样，得到对所述归一化幅度谱降采样后的信号，包括：
    对所述归一化幅度谱进行三角窗降采样，得到对所述归一化幅度谱降采样后的信号。
  14. 根据权利要求13所述的方法，其特征在于，所述三角窗降采样为三角窗均匀降采样。
  15. 根据权利要求1所述的方法，其特征在于，所述获取所述信号的特征信息，根据所述特征信息获取所述电机的运行状态，包括：
    提取所述信号的N段子信号；其中，N为大于等于1的整数；
    根据所述N段子信号，获取所述信号的特征信息，根据所述特征信息获取所述电机的运行状态。
  16. 根据权利要求1所述的方法，其特征在于，所述电机包括至少一种所述运行状态，所述获取所述信号的特征信息，根据所述特征信息获取所述电机的运行状态，包括：
    获取所述信号的特征信息，根据所述特征信息，按照预设对应关系获取与所述特征信息对应的所述电机的运行状态，所述预设对应关系为包括多种预设特征信息和每种预设特征信息对应的运行状态。
  17. 根据权利要求16所述的方法，其特征在于，根据所述特征信息，按照预设对应关系获取与所述特征信息对应的所述电机的运行状态，包括：
    确定所述特征信息所对应的目标预设特征信息，所述目标预设特征信息为所述多种预设特征信息中的特征信息；
    确定所述预设对应关系中与所述目标特征预设信息对应的运行状态为所述电机的运行状态。
  18. 根据权利要求16所述的方法，其特征在于，所述至少一种所述运行状态如下中的至少一项：
    所述电机启动异常、所述电机在第一时长后存在运行异常的风险、所述电机运行正常、所述电机在第二时长后存在运行异常的风险、所述电机运行过程中的噪声过大；
    其中，所述第一时长小于所述第二时长。
  19. 根据权利要求1所述的方法，其特征在于，所述根据所述特征信息获取与所述特征信息对应的所述电机的所述运行状态，包括：
    根据所述特征信息和机器学习模型，采用机器学习算法获取与所述特征信息对应的所述电机的所述运行状态。
  20. 根据权利要求19所述的方法，其特征在于，所述机器学习算法为神经网络算法，所述机器学习模型为神经网络模型。
  21. 根据权利要求1～20任一项所述的方法，其特征在于，与所述电机的运行状态相关的信号为如下中的一种或多种：
    所述电机的噪声信号、所述电机的电流信号，设置在所述电机外壁上的惯性测量单元测量得到的信号。
  22. 根据权利要求21所述的方法，其特征在于，所述电机的噪声信号由麦克风、加速度计、振动传感器、压电晶体和/或气压计采集。
  23. 根据权利要求21所述的方法，其特征在于，获取与电机的运行状态相关的信号，包括：
    在预设时长内获取多个与电机的运行状态相关的信号；
    所述获取所述信号的特征信息，根据所述特征信息获取所述电机的运行状态，包括：
    对于所述多个与电机的运行状态相关的信号中的一个第一信号，获取所述第一信号的特征信息；
    根据所述多个与电机的运行状态相关的信号各自对应的特征信息，获取所述电机的运行状态。
  24. 一种电机运行状态的获取装置，其特征在于，包括：存储器、处理器和通信总线，所述存储器和所述处理器通过所述通信总线连接；
    存储器，用于存储计算机程序；
    处理器，用于调用所述计算机程序，执行如下操作：
    获取与电机的运行状态相关的信号；
    获取所述信号的特征信息，根据所述特征信息获取所述电机的运行状态。
  25. 根据权利要求24所述的装置，其特征在于，还包括通过所述通信总线与所述处理器连接的信号采集器和电机
    信号采集器，用于采集与所述电机的运行状态相关的信号；
    所述处理器在获取与电机的运行状态相关的信号时，具体用于执行如下操作：
    从所述信号采集器获取与所述电机的运行状态相关的信号。
  26. 根据权利要求24所述的装置，其特征在于，所述处理器在获取所述信号的特征信息时，还用于执行如下操作：
    对所述信号进行预处理。
  27. 根据权利要求26所述的装置，其特征在于，所述预处理包括对所述信号进行归一化处理，获取归一化信号。
  28. 根据权利要求24所述的装置，其特征在于，所述特征信息包括 所述信号的强度信息和/或分布信息，所述处理器在获取所述信号的特征信息，根据所述特征信息获取所述电机的运行状态时，具体用于执行如下操作：
    获取所述信号的强度信息和/或分布信息；
    根据所述信号的强度信息和/或分布信息，获取所述电机的运行状态。
  29. 根据权利要求28所述的装置，其特征在于，所述处理器获取所述信号的强度信息和/或分布信息时，具体用于执行如下操作：
    获取所述信号对应的归一化幅度谱；
    根据所述信号对应的归一化幅度谱，获取所述信号的强度信息和/或分布信息。
  30. 根据权利要求29所述的装置，其特征在于，所述处理器在获取所述信号对应的归一化幅度谱时，具体用于执行如下操作：
    获取所述信号对应的幅度谱；
    将所述幅度谱中的各幅度值除以零频率对应的幅度值，得到所述归一化幅度谱。
  31. 根据权利要求29所述的装置，其特征在于，所述处理器在根据所述信号对应的归一化幅度谱，获取所述信号的强度信息时，具体用于执行如下操作：
    根据所述信号对应的归一化幅度谱以及M个预设阈值，获取所述强度信息，其中，所述强度信息包括M个强度值，M为大于等于1的整数。
  32. 根据权利要求31所述的装置，其特征在于，所述处理器在根据所述信号对应的归一化幅度谱以及M个预设阈值，获取所述强度信息时，具体用于执行如下操作：
    对于M个预设阈值中的任意一个第一预设阈值，获取所述归一化幅度谱中的各幅度值与第一预设阈值的差值之和，所述各幅度值与第一预设阈值的差值之和为所述强度信息中与所述第一预设阈值对应的强度值。
  33. 根据权利要求32所述的装置，其特征在于，所述预设阈值为0～1中的任意值。
  34. 根据权利要求29所述的装置，其特征在于，所述处理器在根据所述信号对应的归一化幅度谱，获取所述信号的分布信息时，具体用于执 行如下操作：
    根据所述信号对应的归一化幅度谱，得到所述信号对应的分布信号；
    确定所述分布信号中的前K个分布值为所述信号的分布信息，其中，K为大于等于1的整数。
  35. 根据权利要求34所述的装置，其特征在于，所述处理器在根据所述信号对应的归一化幅度谱，得到所述信号对应的分布信号时，具体用于执行如下操作：
    对所述归一化幅度谱进行降采样，得到对所述归一化幅度谱降采样后的信号；
    根据对所述归一化幅度谱降采样后的信号，得到所述信号对应的分布信号。
  36. 根据权利要求35所述的装置，其特征在于，所述处理器在根据对所述归一化幅度谱降采样后的信号，得到所述信号对应的分布信号时，具体用于执行如下操作：
    对所述归一化幅度谱降采样后的信号进行反余弦变换，得到所述信号对应的分布信号。
  37. 根据权利要求35所述的装置，其特征在于，所述处理器在对所述归一化幅度谱进行降采样，得到对所述归一化幅度谱降采样后的信号时，具体用于执行如下操作：
    对所述归一化幅度谱进行三角窗降采样，得到对所述归一化幅度谱降采样后的信号。
  38. 根据权利要求39所述的装置，其特征在于所述三角窗降采样为三角窗均匀降采样。
  39. 根据权利要求24所述的装置，其特征在于，所述电机包括至少一种所述运行状态，所述获取所述信号的特征信息，所述处理器在根据所述特征信息获取所述电机的运行状态时，具体用于执行如下操作：
    获取所述信号的特征信息，根据所述特征信息，按照预设对应关系获取与所述特征信息对应的所述电机的运行状态，所述预设对应关系为包括多种预设特征信息和每种预设特征信息对应的运行状态。
  40. 根据权利要求39所述的装置，其特征在于，所述处理器在根据 所述特征信息，按照预设对应关系获取与所述特征信息对应的所述电机的运行状态时，具体用于执行如下操作：
    确定所述特征信息所对应的目标预设特征信息，所述目标预设特征信息为所述多种预设特征信息中的特征信息；
    确定所述预设对应关系中与所述目标预设信息对应的运行状态为所述电机的运行状态。
  41. 根据权利要求39所述的装置，其特征在于，所述至少一种所述运行状态如下中的至少一项：
    所述电机启动异常、所述电机在第一时长后存在运行异常的风险、所述电机运行正常、所述电机在第二时长后存在运行异常的风险、所述电机运行过程中的噪声过大；
    其中，所述第一时长小于所述第二时长。
  42. 根据权利要求25所述的装置，其特征在于，所述处理器在根据所述特征信息获取与所述特征信息对应的所述电机的所述运行状态时，具体用于执行如下操作：
    根据所述特征信息和机器学习模型，采用机器学习算法获取与所述特征信息对应的所述电机的所述运行状态。
  43. 根据权利要求42所述的装置，其特征在于，所述机器学习算法为神经网络算法，所述机器学习模型为神经网络模型。
  44. 根据权利要求24～43任一项所述的装置，其特征在于，与所述电机的运行状态相关的信号为如下中的一种或多种：
    所述电机的噪声信号、所述电机的电流信号，设置在所述电机外壁上的惯性测量单元测量得到的信号。
  45. 根据权利要求44所述的装置，其特征在于，所述电机的噪声信号由麦克风、加速度计、振动传感器、压电晶体和/或气压计采集。
  46. 根据权利要求44所述的装置，其特征在于，所述处理器在获取与电机的运行状态相关的信号时，具体用于执行如下操作：
    在预设时长内获取多个与电机的运行状态相关的信号；
    所述处理器在获取所述信号的特征信息，根据所述特征信息获取所述电机的运行状态时，具体用于执行如下操作：
    对于所述多个信号中的一个第一信号，获取所述第一信号的特征信息；
    根据所述多个信号各自对应的特征信息，获取所述电机的运行状态。
  47. 一种计算机可读存储介质，包括程序或指令，当所述程序或指令在计算机上运行时，权利要求1～23任一所述的方法被执行。
  48. 一种电子装置，包括电机和权利要求24～46任一项所述的电机运行状态的获取装置，其特征在于，所述获取装置用于获取所述电机的运行状态。
  49. 如权利要求48所述的电子装置，其特征在于，所述电子装置为雷达或无人机。
PCT/CN2019/074606 2019-02-02 2019-02-02 电机运行状态的获取方法和装置 Ceased WO2020155147A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2019/074606 WO2020155147A1 (zh) 2019-02-02 2019-02-02 电机运行状态的获取方法和装置
CN201980005478.4A CN111819452A (zh) 2019-02-02 2019-02-02 电机运行状态的获取方法和装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/074606 WO2020155147A1 (zh) 2019-02-02 2019-02-02 电机运行状态的获取方法和装置

Publications (1)

Publication Number Publication Date
WO2020155147A1 true WO2020155147A1 (zh) 2020-08-06

Family

ID=71841343

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/074606 Ceased WO2020155147A1 (zh) 2019-02-02 2019-02-02 电机运行状态的获取方法和装置

Country Status (2)

Country Link
CN (1) CN111819452A (zh)
WO (1) WO2020155147A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030067277A1 (en) * 2001-07-27 2003-04-10 Texas A&M University System Method and system for early detection of incipient faults in electric motors
CN102437815A (zh) * 2011-10-13 2012-05-02 宁波杜亚机电技术有限公司 一种消除电机机械制造差异及安装使用中负载差异的方法
CN202798576U (zh) * 2012-08-31 2013-03-13 郑州春泉暖通节能设备有限公司 一种基于电压互感技术的简易电机档位识别装置
CN103684146A (zh) * 2012-09-11 2014-03-26 上海汽车集团股份有限公司 汽车座椅电机运行状态的监控
CN105337534A (zh) * 2014-08-08 2016-02-17 北京谊安医疗系统股份有限公司 一种麻醉机的无刷直流电机功耗与温度监测电路

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8994359B2 (en) * 2011-08-29 2015-03-31 General Electric Company Fault detection based on current signature analysis for a generator
US10359473B2 (en) * 2012-05-29 2019-07-23 Nutech Ventures Detecting faults in turbine generators
CN103698699B (zh) * 2013-12-06 2017-08-01 西安交通大学 一种基于模型的异步电动机故障监测诊断方法
CN106291352A (zh) * 2015-05-26 2017-01-04 江森自控空调冷冻设备(无锡)有限公司 电机故障检测方法和装置以及电机故障保护系统
KR101795384B1 (ko) * 2015-12-11 2017-11-09 현대자동차 주식회사 하이브리드 차량의 능동 진동 제어 방법 및 장치
FR3059784A1 (fr) * 2016-12-06 2018-06-08 Universite De Corse P Paoli Installation de diagnostic d'une machine electrique
CN106842023B (zh) * 2017-01-22 2019-05-21 浙江大学 旋转电机的故障诊断方法
US10928814B2 (en) * 2017-02-24 2021-02-23 General Electric Technology Gmbh Autonomous procedure for monitoring and diagnostics of machine based on electrical signature analysis
CN107831438A (zh) * 2017-10-25 2018-03-23 上海交通大学 一种电机故障的预测方法及预测系统
CN108490349B (zh) * 2018-01-23 2020-08-18 浙江大学山东工业技术研究院 基于Mel频率倒谱系数的电机异音检测方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030067277A1 (en) * 2001-07-27 2003-04-10 Texas A&M University System Method and system for early detection of incipient faults in electric motors
CN102437815A (zh) * 2011-10-13 2012-05-02 宁波杜亚机电技术有限公司 一种消除电机机械制造差异及安装使用中负载差异的方法
CN202798576U (zh) * 2012-08-31 2013-03-13 郑州春泉暖通节能设备有限公司 一种基于电压互感技术的简易电机档位识别装置
CN103684146A (zh) * 2012-09-11 2014-03-26 上海汽车集团股份有限公司 汽车座椅电机运行状态的监控
CN105337534A (zh) * 2014-08-08 2016-02-17 北京谊安医疗系统股份有限公司 一种麻醉机的无刷直流电机功耗与温度监测电路

Also Published As

Publication number Publication date
CN111819452A (zh) 2020-10-23

Similar Documents

Publication Publication Date Title
CN110634497B (zh) 降噪方法、装置、终端设备及存储介质
US6173074B1 (en) Acoustic signature recognition and identification
CN113608119B (zh) 电机运行状态监测方法、装置、设备和存储介质
JP6289400B2 (ja) 周期定常背景雑音を受ける信号内にイベントを検出する方法およびシステム
WO2023138337A1 (zh) 一种电机故障检测方法及装置
CN113125135A (zh) 旋转机械的故障诊断方法、存储介质及电子设备
EP2907121B1 (en) Real-time traffic detection
CN109643552A (zh) 用于可变噪声状况中语音增强的鲁棒噪声估计
WO2021000498A1 (zh) 复合语音识别方法、装置、设备及计算机可读存储介质
CN112697270B (zh) 故障检测方法、装置、无人设备及存储介质
CN113758709A (zh) 结合边缘计算和深度学习的滚动轴承故障诊断方法及系统
CN107644231A (zh) 一种发电机转子故障诊断方法及装置
CN119049511B (zh) 一种基于麦克风阵列的设备故障诊断与定位方法及系统
JP2016042613A (ja) 目的音声区間検出装置、目的音声区間検出方法、目的音声区間検出プログラム、音声信号処理装置及びサーバ
CN118882472A (zh) 基于gnss技术的桥梁结构动态变形监测方法、设备及介质
CN118899005A (zh) 一种音频信号处理方法、装置、计算机设备及存储介质
CN114639391A (zh) 机械故障提示方法、装置、电子设备及存储介质
CN113674763A (zh) 利用线谱特性的鸣笛声识别方法及系统、设备与存储介质
CN105830152A (zh) 集音装置、集音装置的输入信号校正方法以及移动设备信息系统
JP4871191B2 (ja) 目的信号区間推定装置、目的信号区間推定方法、目的信号区間推定プログラム及び記録媒体
CN114383848A (zh) 一种增压器近场气流噪声目标值的预测方法、装置、存储介质及设备
CN119068912A (zh) 基于声纹识别的工业诊断方法及相关设备
CN117854542A (zh) 一种基于奇异谱声纹的电力设备状态检测方法及装置
CN113362850B (zh) 音频信号采集设备的检测方法、设备和存储介质
Wu et al. Faults classification of a scooter engine platform using wavelet transform and artificial neural network

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19912457

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19912457

Country of ref document: EP

Kind code of ref document: A1