CN117477499B - Intelligent motor control protection system and method thereof - Google Patents
Intelligent motor control protection system and method thereof Download PDFInfo
- Publication number
- CN117477499B CN117477499B CN202311417497.2A CN202311417497A CN117477499B CN 117477499 B CN117477499 B CN 117477499B CN 202311417497 A CN202311417497 A CN 202311417497A CN 117477499 B CN117477499 B CN 117477499B
- Authority
- CN
- China
- Prior art keywords
- sound
- detection signal
- feature
- feature vector
- time sequence
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000001514 detection method Methods 0.000 claims abstract description 129
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000012098 association analyses Methods 0.000 claims abstract description 5
- 239000013598 vector Substances 0.000 claims description 145
- 238000009826 distribution Methods 0.000 claims description 44
- 238000012937 correction Methods 0.000 claims description 25
- 230000007613 environmental effect Effects 0.000 claims description 23
- 230000005236 sound signal Effects 0.000 claims description 18
- 238000013527 convolutional neural network Methods 0.000 claims description 12
- 238000003062 neural network model Methods 0.000 claims description 12
- 238000010219 correlation analysis Methods 0.000 claims description 11
- 238000005457 optimization Methods 0.000 claims description 10
- 238000005070 sampling Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 7
- 230000008859 change Effects 0.000 claims description 5
- 238000011176 pooling Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 2
- 238000003745 diagnosis Methods 0.000 abstract description 10
- 238000004422 calculation algorithm Methods 0.000 abstract description 8
- 238000010586 diagram Methods 0.000 description 14
- 238000004458 analytical method Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 230000002159 abnormal effect Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000001360 synchronised effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000013136 deep learning model Methods 0.000 description 2
- 238000007477 logistic regression Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 206010044565 Tremor Diseases 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000005406 washing Methods 0.000 description 1
- 238000004804 winding Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H7/00—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions
- H02H7/08—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions for dynamo-electric motors
- H02H7/0833—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions for dynamo-electric motors for electric motors with control arrangements
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H1/00—Details of emergency protective circuit arrangements
- H02H1/0092—Details of emergency protective circuit arrangements concerning the data processing means, e.g. expert systems, neural networks
-
- 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
- H02P29/02—Providing protection against overload without automatic interruption of supply
- H02P29/024—Detecting a fault condition, e.g. short circuit, locked rotor, open circuit or loss of load
Landscapes
- Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Power Engineering (AREA)
- Control Of Electric Motors In General (AREA)
Abstract
An intelligent motor control protection system and method thereof are disclosed. The method comprises the steps of firstly collecting operation sound detection signals of a monitored motor in a preset time period through a first sound sensor, collecting environment sound detection signals through a second sound sensor, then carrying out time sequence association analysis on the operation sound detection signals and the environment sound detection signals to obtain sound difference association characteristics, and then determining whether to start a motor protection strategy or not based on the sound difference association characteristics. In this way, the operation sound detection signal and the environment sound detection signal can be analyzed by adding a signal processing and analyzing algorithm at the rear end, so that the interference influence of the environment sound is filtered when the operation state of the motor is monitored, the operation state characteristic information of the motor is better captured, whether the monitored motor has a fault is more accurately judged, and the monitored motor is intelligently suspended based on a fault diagnosis result.
Description
Technical Field
The present disclosure relates to the field of motor control protection, and more particularly, to an intelligent motor control protection system and method thereof.
Background
Motors are power devices commonly used in industrial production to drive various mechanical devices. However, the motor may experience various faults during operation, and the faults may cause damage to the motor and even fire disaster. Therefore, it is important to perform control protection of the motor.
Conventional motor control protection systems rely primarily on sensors and preset protection parameters to monitor the operating state of the motor and to protect the motor from faults. However, different motors and working environments may require different parameter settings, and different protection parameters may have different degrees of influence on the control protection of the motors, so that the system cannot adapt to the characteristics of different motors and the changes of the working environments, and the adaptability and the accuracy of the motor control protection system are limited. Moreover, conventional motor control systems are typically only capable of detecting some significant faults, such as overloads, shorts, etc., while for some implicit faults, such as bearing wear, insulation degradation, etc., it is often difficult to discover and handle in time, which can lead to more serious faults and damage.
Accordingly, an optimized intelligent motor control protection system is desired.
Disclosure of Invention
In view of this, the disclosure provides an intelligent motor control protection system and a method thereof, which can analyze the operation sound detection signal and the environment sound detection signal by adding a signal processing and analyzing algorithm at the back end, so as to filter out the interference influence of the environment sound when the operation state of the motor is monitored, so as to better capture the operation state characteristic information of the motor, thereby more accurately judging whether the monitored motor has a fault, and intelligently suspending the monitored motor based on the fault diagnosis result.
According to an aspect of the present disclosure, there is provided an intelligent motor control protection system, including:
The sound signal acquisition module is used for acquiring an operation sound detection signal of the monitored motor in a preset time period through the first sound sensor and acquiring an environment sound detection signal through the second sound sensor;
The sound signal association analysis module is used for carrying out time sequence association analysis on the operation sound detection signal and the environment sound detection signal so as to obtain sound differential association characteristics; and
And the motor protection control module is used for determining whether to start a motor protection strategy based on the sound difference correlation characteristic.
According to another aspect of the present disclosure, there is provided an intelligent motor control protection method, including:
collecting an operation sound detection signal of a monitored motor in a preset time period through a first sound sensor, and collecting an environment sound detection signal through a second sound sensor;
performing time sequence correlation analysis on the operation sound detection signal and the environment sound detection signal to obtain sound difference correlation characteristics; and
And determining whether to start a motor protection strategy based on the sound difference correlation characteristic.
According to the embodiment of the disclosure, firstly, an operation sound detection signal of a monitored motor in a preset time period is collected through a first sound sensor, and an environment sound detection signal is collected through a second sound sensor, then, time sequence correlation analysis is conducted on the operation sound detection signal and the environment sound detection signal to obtain sound differential correlation characteristics, and then, whether a motor protection strategy is started or not is determined based on the sound differential correlation characteristics. In this way, the operation sound detection signal and the environment sound detection signal can be analyzed by adding a signal processing and analyzing algorithm at the rear end, so that the interference influence of the environment sound is filtered when the operation state of the motor is monitored, the operation state characteristic information of the motor is better captured, whether the monitored motor has a fault is more accurately judged, and the monitored motor is intelligently suspended based on a fault diagnosis result.
Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments, features and aspects of the present disclosure and together with the description, serve to explain the principles of the disclosure.
Fig. 1 illustrates a block diagram of an intelligent motor control protection system according to an embodiment of the present disclosure.
Fig. 2 shows a block diagram of the sound signal correlation analysis module in the intelligent motor control protection system according to an embodiment of the present disclosure.
Fig. 3 illustrates a block diagram of the sound detection signal timing variation feature extraction unit in the intelligent motor control protection system according to an embodiment of the present disclosure.
Fig. 4 shows a block diagram of the motor protection control module in the intelligent motor control protection system according to an embodiment of the present disclosure.
Fig. 5 shows a block diagram of the feature distribution optimizing unit in the intelligent motor control protection system according to an embodiment of the present disclosure.
Fig. 6 shows a flowchart of an intelligent motor control protection method according to an embodiment of the present disclosure.
Fig. 7 shows an architectural diagram of an intelligent motor control protection method according to an embodiment of the present disclosure.
Fig. 8 illustrates an application scenario diagram of an intelligent motor control protection system according to an embodiment of the present disclosure.
Detailed Description
The following description of the embodiments of the present disclosure will be made clearly and fully with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some, but not all embodiments of the disclosure. All other embodiments, which can be made by one of ordinary skill in the art without undue burden based on the embodiments of the present disclosure, are also within the scope of the present disclosure.
As used in this disclosure and in the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.
Various exemplary embodiments, features and aspects of the disclosure will be described in detail below with reference to the drawings. In the drawings, like reference numbers indicate identical or functionally similar elements. Although various aspects of the embodiments are illustrated in the accompanying drawings, the drawings are not necessarily drawn to scale unless specifically indicated.
In addition, numerous specific details are set forth in the following detailed description in order to provide a better understanding of the present disclosure. It will be understood by those skilled in the art that the present disclosure may be practiced without some of these specific details. In some instances, methods, means, elements, and circuits well known to those skilled in the art have not been described in detail in order not to obscure the present disclosure.
Motors include two broad categories, DC motors (direct current motors) and AC motors (alternating current motors). The DC motor is a motor for converting DC electric energy into mechanical energy, and is characterized by adjustable rotating speed, large starting torque, relatively simple control, and the DC motor is often used for applications requiring speed regulation and large starting torque, such as electric vehicles, electric tools and the like. The AC motor is a motor for converting AC electric energy into mechanical energy, and is characterized by simple structure, high reliability and low maintenance cost, and is commonly used in industrial and household appliances such as fans, washing machines, air conditioners and the like. In addition, the ac motor may be divided into two types, i.e., a synchronous motor and an asynchronous motor. The rotating speed of the synchronous motor is synchronous with the frequency of the power supply, and the synchronous motor is suitable for occasions of precise control and constant-speed operation; the rotating speed of the asynchronous motor is slightly lower than the power frequency, and the asynchronous motor is suitable for most industrial and household applications.
The motor produces a specific sound pattern during normal operation due to mechanical movements and electromagnetic vibrations within the motor, and different types of motors produce different sound characteristics during operation. When a motor fails, the sound pattern is often changed, because the failure may cause mechanical components inside the motor to be damaged or to operate abnormally, thereby affecting the generation of sound. Some common motor failure-induced sound variations include: 1. abnormal noise: abnormal noise of the motor may be caused by bearing wear, gear wear, unbalanced or loose parts, etc.; 2. tremor sound: the motor generates vibration sound or resonance sound, which may be caused by loosening, unbalance or damage of internal parts of the motor; 3. a whistle sound: the motor generates a whistling sound which is possibly caused by damage of a motor winding or a rotor, electromagnetic interference and the like; 4. noise increase: the noise level increases suddenly during normal operation of the motor, possibly due to internal faults or overloads of the motor. Through observation and analysis of the motor sound mode, whether the motor fails or not can be judged, and corresponding maintenance or replacement measures can be timely taken, so that further damage or potential safety hazards can be avoided.
Accordingly, it is considered that a specific sound pattern is generated when the motor is normally operated, and the sound pattern is changed when the motor is failed. By analyzing the motor operation sound, whether the motor is abnormal or not, such as noise increase, frequency change and the like, can be detected, so that whether the motor is in fault or not can be timely judged, and a corresponding protection strategy is adopted. However, in the process of collecting and analyzing the operation sound of the motor, the sound of the environment has a great influence on the analysis result and the fault judgment, and the sound signals of the environment may be different in different detection areas and different times.
In view of the above technical problems, the technical concept of the present disclosure is to add a processing and analysis algorithm of signals to the rear end to analyze an operation sound detection signal and an environmental sound detection signal after the operation sound detection signal and the environmental sound detection signal of a monitored motor are collected by a sound sensor, so as to filter out interference influence of environmental sound when monitoring the operation state of the motor, so as to better capture the operation state characteristic information of the motor, thereby more accurately judging whether the monitored motor has a fault, and intelligently suspending the monitored motor based on the fault diagnosis result.
Fig. 1 shows a block diagram schematic of an intelligent motor control protection system according to an embodiment of the present disclosure. As shown in fig. 1, an intelligent motor control protection system 100 according to an embodiment of the present disclosure includes: a sound signal acquisition module 110 for acquiring an operation sound detection signal of the monitored motor for a predetermined period of time through a first sound sensor, and acquiring an environmental sound detection signal through a second sound sensor; the sound signal correlation analysis module 120 is configured to perform time sequence correlation analysis on the operation sound detection signal and the environment sound detection signal to obtain a sound differential correlation characteristic; and a motor protection control module 130, configured to determine whether to start a motor protection policy based on the sound differential correlation characteristic.
Specifically, in the technical scheme of the present disclosure, first, an operation sound detection signal of a monitored motor for a predetermined period of time, which is collected by a first sound sensor, and an environmental sound detection signal, which is collected by a second sound sensor, are acquired. It will be appreciated that the ambient sound detection signal may provide information of background noise and ambient conditions, and that analysis of the ambient sound detection signal may provide a better understanding of the background environment of the motor's operating condition and take it into account for more accurate motor fault determination and protection decisions. It should be appreciated that the first and second sound sensors may be any suitable sound sensor, such as a microphone sensor or a sound sensor module, which may be used to collect the operating sound and ambient sound of the motor. For the first sound sensor, it should be placed close to the motor so as to accurately collect the operation sound of the motor, and the installation position of the sensor may be determined according to the specific application scenario and the characteristics of the motor. For the second sound sensor, it should be placed in a position away from the motor so that ambient sound can be collected, thus providing information of background noise and ambient conditions, helping to better understand the background environment of the motor operating state, and the sensor should be installed in a position away from the motor so as to avoid interference of the motor sound to the ambient sound collected by the sensor. When installing a sound sensor, the sensitivity and frequency response range of the sensor need to be considered to ensure that the desired sound signal can be accurately acquired. In addition, the sensor should be capable of interfacing with a data acquisition system or processor to transmit and process the acquired sound signals.
Then, the operation sound detection signal and the environment sound detection signal are subjected to discrete sampling to obtain a plurality of operation sound detection signal sample points and a plurality of environment sound detection signal sample points. It should be understood that the continuous sound detection signal can be converted into discrete sample points through discrete sampling, so that subsequent signal analysis and processing are facilitated, that is, the extracted characteristic information about the running state of the motor and the environment sound characteristic information are conveniently extracted subsequently, so as to judge the running state and the fault condition of the motor.
Further, considering that the plurality of operation sound detection signal sample points and the plurality of environment sound detection signal sample points are all discretely sampled in the time dimension, and the plurality of operation sound detection signal sample points and the plurality of environment sound detection signal sample points respectively have time-sequence association relations in the time dimension, the time-sequence association relations respectively represent the operation state time sequence association characteristic information of the monitored motor and the time sequence change characteristic information of the environment sound. Therefore, in order to more accurately monitor the operation state of the monitored motor, so as to control the motor protection strategy, in the technical scheme of the disclosure, the plurality of operation sound detection signal sample points and the plurality of environment sound detection signal sample points are further arranged as input vectors according to a time dimension, and then feature mining is performed in a time sequence feature extractor based on a one-dimensional convolutional neural network model, so as to extract time sequence implicit association feature distribution information of the sound detection signal of the monitored motor in the time dimension and time sequence implicit association feature distribution information of the environment sound detection signal in the time dimension, thereby obtaining an operation sound time sequence feature vector and an environment sound time sequence feature vector.
After the sound time sequence change characteristic information and the environment sound time sequence change characteristic information of the monitored motor are obtained, further calculating a difference characteristic vector between the operation sound time sequence characteristic vector and the environment sound time sequence characteristic vector, so that the time sequence hidden characteristic information of the environment sound is removed from the sound time sequence hidden characteristic information of the monitored motor to provide more time sequence characteristic information about the motor operation state, and the actual operation state monitoring of the monitored motor is facilitated to judge whether the operation state of the monitored motor is abnormal or not.
Accordingly, as shown in fig. 2, the sound signal correlation analysis module 120 includes: a sound detection signal discrete sampling unit 121, configured to perform discrete sampling on the operation sound detection signal and the environmental sound detection signal to obtain a plurality of operation sound detection signal sample points and a plurality of environmental sound detection signal sample points; a sound detection signal time sequence variation feature extraction unit 122, configured to arrange the plurality of operation sound detection signal sample points and the plurality of environmental sound detection signal sample points into input vectors according to a time dimension, and then pass through a time sequence feature extractor based on a deep neural network model to obtain operation sound time sequence feature vectors and environmental sound time sequence feature vectors; and a sound detection differentiating unit 123 for calculating a difference feature vector between the operation sound timing feature vector and the environmental sound timing feature vector as the sound difference correlation feature. It should be appreciated that by calculating the sound differential correlation characteristics, the type and extent of motor faults may be further analyzed, and the differential feature vectors may reflect differences between operating sounds and ambient sounds, thereby helping to determine the specific cause of the motor fault.
In the sound detection signal time sequence variation feature extraction unit 122, the deep neural network model is a one-dimensional convolutional neural network model. Further, as shown in fig. 3, the sound detection signal timing variation feature extraction unit 122 includes: a vectorization subunit 1221, configured to arrange the plurality of operation sound detection signal sample points and the plurality of environment sound detection signal sample points into input vectors according to a time dimension respectively to obtain an operation sound detection signal input vector and an environment sound detection signal input vector; and a convolution encoding subunit 1222 for performing one-dimensional convolution processing, pooling processing, and nonlinear activation on input data in forward transfer of layers using each layer of the depth neural network model-based time series feature extractor to output the operation sound time series feature vector and the environmental sound time series feature vector by a last layer of the depth neural network model-based time series feature extractor, respectively, wherein inputs of a first layer of the depth neural network model-based time series feature extractor are the operation sound detection signal input vector and the environmental sound detection signal input vector. It should be appreciated that the one-dimensional convolutional neural network model (1D CNN) is a deep learning model for processing sequence data. Unlike a conventional two-dimensional convolutional neural network model (2D CNN), a one-dimensional convolutional neural network model is mainly used to process one-dimensional sequence data, such as time-series data, audio signals, text data, and the like. The one-dimensional convolutional neural network model can capture local patterns and features in sequence data by performing convolutional operation in a time dimension, performs sliding convolutional operation on input data by using one-dimensional convolutional check, performs feature extraction by a nonlinear activation function, can reduce the dimension of the features by pooling operation (such as maximum pooling or average pooling), and finally inputs the extracted features into a fully connected layer for classification, regression or other tasks. The one-dimensional convolutional neural network model can extract time sequence characteristics of the sound signal, such as frequency spectrum characteristics, energy characteristics and the like, by carrying out one-dimensional convolutional operation on the sound signal of the motor, and the characteristics can be used for fault detection, diagnosis and classification to help judge the working state and the health condition of the motor. Meanwhile, the one-dimensional convolutional neural network model can also reduce noise and eliminate environmental interference on sound signals, and improves the quality and reliability of the signals. In other words, the one-dimensional convolutional neural network model is an effective deep learning model suitable for processing sequence data, including sound signals, which can be analyzed by extracting timing features for motor fault detection and diagnosis applications.
And then, the differential feature vector passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the running state of the monitored motor is normal or not. That is, the classification is performed by using the implicit characteristic information of the actual detection sound time sequence of the monitored motor after the characteristic of the environmental sound signal is filtered, so that the abnormal running state of the monitored motor is effectively detected in time. Further, based on the classification result, whether to start a motor protection strategy is determined. Therefore, the interference influence of the environmental sound can be filtered when the running state of the motor is monitored, so that whether the monitored motor has a fault or not can be judged more accurately, and the monitored motor is intelligently suspended based on the fault diagnosis result to achieve the purpose of control and protection.
Accordingly, as shown in fig. 4, the motor protection control module 130 includes: a feature distribution optimizing unit 131, configured to perform feature distribution optimization on the differential feature vector to obtain an optimized differential feature vector; a motor running state detecting unit 132, configured to pass the optimized differential feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether the running state of the monitored motor is normal; and a motor protection policy control unit 133 for determining whether to start the motor protection policy based on the classification result.
As shown in fig. 5, the feature distribution optimizing unit 131 includes: an optimization weighting factor calculation subunit 1311, configured to calculate feature distribution equalization correction feature vectors of the operation sound timing feature vector and the environmental sound timing feature vector; and a feature weighted optimization subunit 1312, configured to calculate a multiplication of the feature distribution balance correction feature vector and the difference feature vector by location points to obtain the optimized difference feature vector. It should be understood that the multiplication by location point (Element-wise Multiplication) refers to the operation of multiplying elements at corresponding locations of two vectors one by one. For vectors of the same length, a new vector is generated by position point multiplication, wherein each element is the result of multiplication of the elements at the corresponding positions of the original vector. In the feature distribution optimizing unit, the operation of multiplying by the position point is used to calculate the product between the feature distribution balance correction feature vector and the differential feature vector. This operation has several roles: 1. emphasis on important features: by multiplying the feature distribution balance correction feature vector and the differential feature vector by position points, important features in the feature distribution balance correction feature vector and relevant features in the differential feature vector can be multiplied, so that the important features are emphasized, information related to fault detection and diagnosis in the differential features is highlighted, and the distinguishing degree and the sensitivity of the features are improved; 2. removing redundant information: the differential feature vector generally contains the features of the environmental sound, the feature distribution balance correction feature vector is the feature for correcting the running sound obtained through feature distribution balance correction, and the feature distribution balance correction feature vector can be multiplied with redundant information in the differential feature vector through position point multiplication, so that the influence of the redundant information on the final feature is reduced, and the purity and reliability of the feature are improved. The position-based point multiplication is a commonly used operation in feature analysis, and can emphasize important features, remove redundant information and improve the distinguishing degree and reliability of the features.
In particular, in the technical solution of the present application, the operation sound time sequence feature vector and the environment sound time sequence feature vector respectively express local correlation features of discrete sampling signal parameter values of the operation sound detection signal and the environment sound detection signal under time sequence, and considering different fluctuation characteristics of discrete sampling values of the operation sound detection signal and the environment sound detection signal under time sequence distribution dimension and influence of noise, the operation sound time sequence feature vector and the environment sound time sequence feature vector may also have more significant distribution intensity difference, so that when position-by-position differential calculation is performed, target distribution information loss of one of the operation sound time sequence feature vector and the environment sound time sequence feature vector is caused, and feature expression effect of the differential feature vector is affected. Based on this, the applicant calculated the running sound timing feature vector, e.g. noted asAnd the ambient sound timing feature vector, e.g., denoted/>The feature distribution balance of (a) corrects feature vectors, e.g., denoted/>。
Accordingly, in a specific embodiment, the optimization weighting factor calculation subunit is configured to: calculating feature distribution balance correction feature vectors of the operation sound time sequence feature vector and the environment sound time sequence feature vector according to the following correction formula; wherein, the correction formula is: wherein/> Representing the operational sound timing feature vector,/>Representing the ambient sound timing feature vector,/>And/>Respectively represent the time sequence feature vector/>, of the operation soundAnd the ambient sound timing feature vector/>Inverse of global mean of (2), and/>Is a unit vector,/>Representing addition by position,/>Representing subtraction by location,/>Representing multiplication by location,/>Representing the feature distribution balance correction feature vector.
That is, if the running sound timing feature vector is taken into consideration in the case of imbalance in the feature distributionAnd the ambient sound timing feature vector/>One is considered as the feature distribution enhancement input of the other, then consider the running sound timing feature vector/>And the ambient sound timing feature vector/>The loss of target distribution information of the other target feature in the constraint space may result in constraint objective loss, so that by means of cross penalty to the outlier distribution (outlier distribution) of the feature distribution relative to each other, the self-supervision balance of feature enhancement and constraint robustness can be realized through feature interpolation fusion to promote the running sound time sequence feature vector/>And the ambient sound timing feature vector/>In such a way that the feature vector/>, is corrected again with the feature distribution balanceThe difference feature vector is subjected to dot multiplication, so that the feature expression effect of the difference feature vector can be improved, and the accuracy of a classification result obtained by the classifier is improved. Therefore, fault detection can be carried out based on the actual motor working state, and the monitored motor is intelligently suspended based on the fault diagnosis result, so that the control and protection purpose when the motor is in fault is achieved.
Further, the motor operation state detecting unit 132 is configured to: performing full-connection coding on the optimized differential feature vector by using a plurality of full-connection layers of the classifier to obtain a plurality of coding classification feature vectors; and inputting the coding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
That is, in the technical solution of the present disclosure, the label of the classifier includes a motor protection policy (first label) that is turned on, and a motor protection policy (second label) that is not turned on, wherein the classifier determines to which classification label the optimized differential feature vector belongs through a soft maximum function. It should be noted that the first tag p1 and the second tag p2 do not include a manually set concept, and in fact, during the training process, the computer model does not have a concept of "whether to turn on the motor protection strategy", which is just two kinds of classification tags, and the probability that the output feature is the sum of the two classification tags sign, that is, p1 and p2 is one. Therefore, the classification result of whether to start the motor protection strategy is actually converted into the classified probability distribution conforming to the natural rule through classifying the labels, and the physical meaning of the natural probability distribution of the labels is essentially used instead of the language text meaning of whether to start the motor protection strategy.
It should be appreciated that the role of the classifier is to learn the classification rules and classifier using a given class, known training data, and then classify (or predict) the unknown data. Logistic regression (logistics), SVM, etc. are commonly used to solve the classification problem, and for multi-classification problems (multi-class classification), logistic regression or SVM can be used as well, but multiple bi-classifications are required to compose multiple classifications, but this is error-prone and inefficient, and the commonly used multi-classification method is the Softmax classification function.
In summary, the intelligent motor control protection system 100 according to the embodiments of the present disclosure is illustrated, and may analyze the operation sound detection signal and the environment sound detection signal by adding a signal processing and analysis algorithm to the rear end, so as to filter out interference effects of the environment sound when monitoring the operation state of the motor, so as to better capture the operation state characteristic information of the motor, thereby more accurately judging whether the monitored motor has a fault, and intelligently suspending the monitored motor based on the fault diagnosis result.
As described above, the intelligent motor control protection system 100 according to the embodiment of the present disclosure may be implemented in various terminal devices, such as a server having an intelligent motor control protection algorithm, and the like. In one example, the intelligent motor control protection system 100 may be integrated into the terminal device as a software module and/or hardware module. For example, the intelligent motor control protection system 100 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the intelligent motor control and protection system 100 can also be one of a plurality of hardware modules of the terminal device.
Alternatively, in another example, the intelligent motor control and protection system 100 and the terminal device may be separate devices, and the intelligent motor control and protection system 100 may be connected to the terminal device through a wired and/or wireless network and transmit interactive information in a agreed data format.
Fig. 6 shows a flowchart of an intelligent motor control protection method according to an embodiment of the present disclosure. Fig. 7 shows a schematic diagram of a system architecture of an intelligent motor control protection method according to an embodiment of the present disclosure. As shown in fig. 6 and 7, an intelligent motor control protection method according to an embodiment of the present disclosure includes: s110, collecting an operation sound detection signal of a monitored motor in a preset time period through a first sound sensor, and collecting an environment sound detection signal through a second sound sensor; s120, performing time sequence correlation analysis on the operation sound detection signal and the environment sound detection signal to obtain sound difference correlation characteristics; and S130, determining whether to start a motor protection strategy based on the sound difference correlation characteristic.
In one possible implementation, performing a time-series correlation analysis on the operation sound detection signal and the environment sound detection signal to obtain a sound differential correlation feature includes: performing discrete sampling on the operation sound detection signal and the environment sound detection signal to obtain a plurality of operation sound detection signal sample points and a plurality of environment sound detection signal sample points; the plurality of operation sound detection signal sample points and the plurality of environment sound detection signal sample points are respectively arranged into input vectors according to time dimension and then respectively pass through a time sequence feature extractor based on a depth neural network model to obtain operation sound time sequence feature vectors and environment sound time sequence feature vectors; and calculating a difference feature vector between the operation sound timing feature vector and the environmental sound timing feature vector as the sound difference correlation feature.
Here, it will be understood by those skilled in the art that the specific operations of the respective steps in the above-described intelligent motor control protection method have been described in detail in the above description of the intelligent motor control protection system with reference to fig. 1 to 5, and thus, repetitive descriptions thereof will be omitted.
Fig. 8 illustrates an application scenario diagram of an intelligent motor control protection system according to an embodiment of the present disclosure. As shown in fig. 8, in this application scenario, first, an operation sound detection signal (e.g., D1 shown in fig. 8) of a monitored motor for a predetermined period of time is collected by a first sound sensor (e.g., N1 shown in fig. 8), and an environmental sound detection signal (e.g., D2 shown in fig. 8) is collected by a second sound sensor (e.g., N2 shown in fig. 8), and then the operation sound detection signal and the environmental sound detection signal are input to a server (e.g., S shown in fig. 8) where an intelligent motor control protection algorithm is deployed, wherein the server can process the operation sound detection signal and the environmental sound detection signal using the intelligent motor control protection algorithm to obtain a classification result for indicating whether the operation state of the monitored motor is normal.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (7)
1. An intelligent motor control protection system, comprising:
The sound signal acquisition module is used for acquiring an operation sound detection signal of the monitored motor in a preset time period through the first sound sensor and acquiring an environment sound detection signal through the second sound sensor;
The sound signal association analysis module is used for carrying out time sequence association analysis on the operation sound detection signal and the environment sound detection signal so as to obtain sound differential association characteristics; and
The motor protection control module is used for determining whether to start a motor protection strategy based on the sound difference correlation characteristics;
Wherein, the motor protection control module includes:
The feature distribution optimizing unit is used for carrying out feature distribution optimization on the differential feature vectors to obtain optimized differential feature vectors;
the motor running state detection unit is used for enabling the optimized differential feature vector to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether the running state of the monitored motor is normal or not; and
The motor protection strategy control unit is used for determining whether to start a motor protection strategy or not based on the classification result;
Wherein the feature distribution optimizing unit includes:
An optimization weighting factor calculating subunit, configured to calculate feature distribution equalization correction feature vectors of the operation sound time sequence feature vector and the environmental sound time sequence feature vector; and
The feature weighting optimization subunit is used for calculating the position-based point multiplication of the feature distribution balance correction feature vector and the difference feature vector to obtain the optimized difference feature vector;
Wherein the optimization weighting factor calculation subunit is configured to:
Calculating feature distribution balance correction feature vectors of the operation sound time sequence feature vector and the environment sound time sequence feature vector according to the following correction formula;
Wherein, the correction formula is:
,
wherein, Representing the operational sound timing feature vector,/>Representing the ambient sound timing feature vector,/>And/>Respectively represent the time sequence feature vector/>, of the operation soundAnd the ambient sound timing feature vector/>Inverse of global mean of (2), and/>Is a unit vector,/>Representing addition by position,/>Representing subtraction by location,/>Representing multiplication by location,/>Representing the feature distribution balance correction feature vector.
2. The intelligent motor control protection system of claim 1, wherein the sound signal correlation analysis module comprises:
The sound detection signal discrete sampling unit is used for performing discrete sampling on the operation sound detection signal and the environment sound detection signal to obtain a plurality of operation sound detection signal sample points and a plurality of environment sound detection signal sample points;
The sound detection signal time sequence change feature extraction unit is used for respectively arranging the plurality of operation sound detection signal sample points and the plurality of environment sound detection signal sample points into input vectors according to a time dimension and respectively passing through a time sequence feature extractor based on a deep neural network model to obtain operation sound time sequence feature vectors and environment sound time sequence feature vectors; and
And the sound detection difference unit is used for calculating a difference feature vector between the operation sound time sequence feature vector and the environment sound time sequence feature vector as the sound difference correlation feature.
3. The intelligent motor control protection system of claim 2, wherein the deep neural network model is a one-dimensional convolutional neural network model.
4. The intelligent motor control protection system according to claim 3, wherein the sound detection signal timing variation feature extraction unit includes:
The vectorization subunit is used for respectively arranging the plurality of operation sound detection signal sample points and the plurality of environment sound detection signal sample points into input vectors according to a time dimension to obtain operation sound detection signal input vectors and environment sound detection signal input vectors; and
And the convolution coding subunit is used for respectively carrying out one-dimensional convolution processing, pooling processing and nonlinear activation on input data in forward transfer of layers by using each layer of the time sequence feature extractor based on the depth neural network model so as to output the operation sound time sequence feature vector and the environment sound time sequence feature vector by the last layer of the time sequence feature extractor based on the depth neural network model, wherein the input of the first layer of the time sequence feature extractor based on the depth neural network model is the operation sound detection signal input vector and the environment sound detection signal input vector.
5. The intelligent motor control protection system of claim 4, wherein the motor operating state detection unit is configured to:
performing full-connection coding on the optimized differential feature vector by using a plurality of full-connection layers of the classifier to obtain a plurality of coding classification feature vectors; and
And inputting the coding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
6. The intelligent motor control protection method is characterized by comprising the following steps of:
collecting an operation sound detection signal of a monitored motor in a preset time period through a first sound sensor, and collecting an environment sound detection signal through a second sound sensor;
performing time sequence correlation analysis on the operation sound detection signal and the environment sound detection signal to obtain sound difference correlation characteristics; and
Determining whether to start a motor protection strategy based on the sound difference correlation characteristic;
wherein determining whether to turn on a motor protection strategy based on the sound differential correlation feature comprises:
Performing feature distribution optimization on the differential feature vectors to obtain optimized differential feature vectors;
the optimized differential feature vector passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the running state of the monitored motor is normal or not; and
Determining whether to start a motor protection strategy based on the classification result;
the feature distribution optimization of the differential feature vector is performed to obtain an optimized differential feature vector, which comprises the following steps:
Calculating characteristic distribution balance correction characteristic vectors of the operation sound time sequence characteristic vector and the environment sound time sequence characteristic vector; and
Calculating the position-based point multiplication of the feature distribution balance correction feature vector and the difference feature vector to obtain the optimized difference feature vector;
Wherein calculating the feature distribution balance correction feature vector of the operation sound time sequence feature vector and the environment sound time sequence feature vector comprises:
Calculating feature distribution balance correction feature vectors of the operation sound time sequence feature vector and the environment sound time sequence feature vector according to the following correction formula;
Wherein, the correction formula is:
,
wherein, Representing the operational sound timing feature vector,/>Representing the ambient sound timing feature vector,/>And/>Respectively represent the time sequence feature vector/>, of the operation soundAnd the ambient sound timing feature vector/>Inverse of global mean of (2), and/>Is a unit vector,/>Representing addition by position,/>Representing subtraction by location,/>Representing multiplication by location,/>Representing the feature distribution balance correction feature vector.
7. The intelligent motor control protection method according to claim 6, wherein performing a time-series correlation analysis on the operation sound detection signal and the environmental sound detection signal to obtain a sound difference correlation feature, comprises:
performing discrete sampling on the operation sound detection signal and the environment sound detection signal to obtain a plurality of operation sound detection signal sample points and a plurality of environment sound detection signal sample points;
the plurality of operation sound detection signal sample points and the plurality of environment sound detection signal sample points are respectively arranged into input vectors according to time dimension and then respectively pass through a time sequence feature extractor based on a depth neural network model to obtain operation sound time sequence feature vectors and environment sound time sequence feature vectors; and
And calculating a difference feature vector between the operation sound time sequence feature vector and the environment sound time sequence feature vector as the sound difference correlation feature.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311417497.2A CN117477499B (en) | 2023-10-30 | 2023-10-30 | Intelligent motor control protection system and method thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311417497.2A CN117477499B (en) | 2023-10-30 | 2023-10-30 | Intelligent motor control protection system and method thereof |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117477499A CN117477499A (en) | 2024-01-30 |
CN117477499B true CN117477499B (en) | 2024-04-26 |
Family
ID=89630518
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311417497.2A Active CN117477499B (en) | 2023-10-30 | 2023-10-30 | Intelligent motor control protection system and method thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117477499B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000039899A (en) * | 1998-07-23 | 2000-02-08 | Hitachi Ltd | Speech recognition apparatus |
CN214506722U (en) * | 2021-05-12 | 2021-10-26 | 唐山学院 | Intelligent control servo motor positioning and mounting device |
CN114993669A (en) * | 2022-04-20 | 2022-09-02 | 燕山大学 | Multi-sensor information fusion transmission system fault diagnosis method and system |
CN115950590A (en) * | 2023-03-15 | 2023-04-11 | 凯晟动力技术(嘉兴)有限公司 | Gas engine leakage early warning system |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10230322B2 (en) * | 2016-12-12 | 2019-03-12 | Stmicroelectronics, Inc. | Smart motor driver architecture with built-in MEMS sensor based early diagnosis of faults |
KR102608614B1 (en) * | 2018-09-21 | 2023-12-04 | 삼성전자주식회사 | Electronic device and method for controlling thereof |
-
2023
- 2023-10-30 CN CN202311417497.2A patent/CN117477499B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000039899A (en) * | 1998-07-23 | 2000-02-08 | Hitachi Ltd | Speech recognition apparatus |
CN214506722U (en) * | 2021-05-12 | 2021-10-26 | 唐山学院 | Intelligent control servo motor positioning and mounting device |
CN114993669A (en) * | 2022-04-20 | 2022-09-02 | 燕山大学 | Multi-sensor information fusion transmission system fault diagnosis method and system |
CN115950590A (en) * | 2023-03-15 | 2023-04-11 | 凯晟动力技术(嘉兴)有限公司 | Gas engine leakage early warning system |
Also Published As
Publication number | Publication date |
---|---|
CN117477499A (en) | 2024-01-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9483049B2 (en) | Anomaly detection and diagnosis/prognosis method, anomaly detection and diagnosis/prognosis system, and anomaly detection and diagnosis/prognosis program | |
CN107003663A (en) | The monitoring of device with movable part | |
Karaköse et al. | The intelligent fault diagnosis frameworks based on fuzzy integral | |
CN117375237B (en) | Substation operation and maintenance method and system based on digital twin technology | |
CN117113218A (en) | Visual data analysis system and method thereof | |
KR20210006832A (en) | Method and apparatus for machine fault diagnosis | |
CN117477499B (en) | Intelligent motor control protection system and method thereof | |
CN116502072B (en) | Robust fault diagnosis method for key components of wind generating set under complex variable working conditions | |
TWI780434B (en) | Abnormal diagnosis device and method | |
CN117435908A (en) | Multi-fault feature extraction method for rotary machine | |
CN117076869A (en) | Time-frequency domain fusion fault diagnosis method and system for rotary machine | |
CN110546657A (en) | Method and apparatus for assessing the lifecycle of a component | |
US20180087489A1 (en) | Method for windmill farm monitoring | |
CN115146675A (en) | Method for diagnosing migration of rotary machine under variable working condition of deep multi-feature dynamic countermeasure | |
Al-Haddad et al. | Application of AdaBoost for stator fault diagnosis in three-phase permanent magnet synchronous motors based on vibration–current data fusion analysis | |
Feng et al. | Multi-kernel learning based autonomous fault diagnosis for centrifugal pumps | |
Senanayaka et al. | Autoencoders and data fusion based hybrid health indicator for detecting bearing and stator winding faults in electric motors | |
Kotsiopoulos et al. | Fault Detection on Bearings and Rotating Machines based on Vibration Sensors Data | |
Łuczak | Machine Fault Diagnosis through Vibration Analysis: Time Series Conversion to Grayscale and RGB Images for Recognition via Convolutional Neural Networks | |
CN117672255B (en) | Abnormal equipment identification method and system based on artificial intelligence and equipment operation sound | |
CN117784710B (en) | Remote state monitoring system and method for numerical control machine tool | |
Li et al. | Fault pattern classification of turbine-generator set based on artificial neural network | |
Mandava | A novel lightweight customized convolution neural network for bearing fault classification | |
CN117520993B (en) | Electric drive axle test system and method thereof | |
CN117786385B (en) | Three-phase asynchronous motor fault monitoring method and system based on twin network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |