CN111059004B - Automatic state monitoring method and system suitable for fan series - Google Patents
Automatic state monitoring method and system suitable for fan series Download PDFInfo
- Publication number
- CN111059004B CN111059004B CN202010189549.5A CN202010189549A CN111059004B CN 111059004 B CN111059004 B CN 111059004B CN 202010189549 A CN202010189549 A CN 202010189549A CN 111059004 B CN111059004 B CN 111059004B
- Authority
- CN
- China
- Prior art keywords
- fan
- series
- standard
- time
- monitoring
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 104
- 238000000034 method Methods 0.000 title claims abstract description 27
- 230000009466 transformation Effects 0.000 claims abstract description 24
- 230000006835 compression Effects 0.000 claims abstract description 21
- 238000007906 compression Methods 0.000 claims abstract description 21
- 230000001360 synchronised effect Effects 0.000 claims abstract description 21
- 238000004364 calculation method Methods 0.000 claims description 8
- 239000013598 vector Substances 0.000 claims description 8
- 238000001228 spectrum Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 238000001816 cooling Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000009423 ventilation Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 239000000284 extract Substances 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 239000000779 smoke Substances 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000004378 air conditioning Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000001035 drying Methods 0.000 description 2
- 239000000428 dust Substances 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000012806 monitoring device Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
- F03D—WIND MOTORS
- F03D17/00—Monitoring or testing of wind motors, e.g. diagnostics
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B10/00—Integration of renewable energy sources in buildings
- Y02B10/30—Wind power
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Sustainable Development (AREA)
- Sustainable Energy (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a method and a system for automatically monitoring states of a fan series, wherein the method comprises the following steps: collecting fan series in a monitoring range; constructing a standard picture library for each series of fans; constructing standard time frequency spectrograms of each series of fans in different states; acquiring a real-time vibration signal and a picture of a monitoring fan, and performing synchronous compression wavelet transformation on the real-time vibration signal to obtain a real-time three-dimensional time-frequency spectrogram; comparing the picture of the monitoring fan with the standard picture library, and judging the series of the monitoring fan; calculating the correlation coefficient of the real-time three-dimensional time-frequency spectrogram and the standard-time spectrogram in different states corresponding to the series to which the fan belongs; and judging the state of the monitoring equipment based on the correlation coefficient. The invention judges the series of the fan, monitors different fan states according to different fan series, and improves the accuracy of fan monitoring.
Description
Technical Field
The invention relates to the field of fan monitoring, in particular to a method and a system for automatically monitoring states of fan series.
Background
When the fan works, gas axially enters the impeller from the air inlet, is pushed by blades on the impeller to increase the energy of the gas, and then flows into the guide vanes. The guide vane changes the deflection airflow into axial flow, and simultaneously introduces the gas into the diffuser pipe, further converts the kinetic energy of the gas into pressure energy, and finally introduces the pressure energy into a working pipeline. Monitoring of fan conditions is essential in terms of reducing maintenance time and maintenance costs.
The invention patent application with the publication number of CN 110296095A discloses an intelligent monitoring and diagnosing method for the operation state of an induced draft fan of a boiler of a power plant of a thermal power plant, wherein a vibration sensor for outputting a vibration signal is arranged on a bearing seat, and the process parameters of the rotating speed of the induced draft fan are obtained; collecting an original vibration signal Xi output by a vibration sensor, performing frequency spectrum analysis, and extracting a characteristic signal of a motor; respectively establishing vibration classification indexes caused by motor stator faults, extracting electromagnetic vibration fault characteristics caused by uneven motor air gaps, establishing electromagnetic vibration classification indexes caused by abnormal motor rotor guide bars, extracting mechanical vibration fault characteristics caused by unbalanced fan rotors, establishing mechanical vibration classification indexes caused by abnormal motor and fan bearing block rolling bearings and extracting mechanical vibration fault characteristics caused by poor installation; and accordingly giving out the fault early warning of the induced draft fan.
Although the application can carry out intelligent monitoring to the running state of fan, adopt the same standard vibration signal to monitor and judge to different fan series. Different series of fans have various structural forms, and corresponding vibration signals have different characteristics. Even if the fans of different series are under the same working condition, the vibration signals of the fans are different in characteristics. Therefore, the same standard is adopted to judge the states of the fans of different series with low accuracy. Therefore, how to adapt to the characteristics of different series of fans and improve the judgment accuracy of the fan state is a problem to be solved urgently in the field.
Disclosure of Invention
The invention aims to provide a method and a system for automatically monitoring the state of a fan series, aiming at the defects of the prior art. The invention judges the series of the fan, monitors different fan states according to different fan series, and improves the accuracy of fan monitoring.
In order to achieve the purpose, the invention adopts the following technical scheme:
a state automatic monitoring method suitable for a fan series comprises the following steps:
s1, acquiring a fan series in a monitoring range;
s2, constructing a standard picture library for each series of fans;
s3, constructing standard time spectrogram of each series of fans under different states;
s4, collecting real-time vibration signals and pictures of the monitoring fan, and performing synchronous compression wavelet transformation on the real-time vibration signals to obtain a real-time three-dimensional time-frequency spectrogram;
s5, comparing the picture of the monitoring fan with the standard picture library, and judging the series of the monitoring fan;
s6, calculating correlation coefficients of the real-time three-dimensional time-frequency spectrogram and the standard-time spectrogram in different states corresponding to the series to which the draught fan belongs;
and S7, judging the state of the monitoring equipment based on the correlation coefficient.
Further, the step S3 includes:
s31, setting at least one standard fan for each model in each series;
s32, collecting a first vibration signal when the standard fan runs in different states;
s33, calculating standard vibration signals of each series of fans in different states based on the first vibration signals;
and S34, performing synchronous compression wavelet transformation on the standard vibration signals to obtain standard time spectrogram of each series of fans in different states.
Further, the standard picture library of each series of fans comprises pictures of various angles of each model of fan in the series.
Further, the step S5 is specifically:
and extracting the characteristics of the picture of the monitored fan and all pictures in the standard picture library, constructing corresponding characteristic vectors, calculating the similarity between the pictures, and selecting the fan series corresponding to the picture with the maximum similarity with the picture of the monitored fan as the series to which the monitored fan belongs.
Further, the different states of the fan comprise a starting state, a violent vibration state, a normal state, a shutdown state, a real-time three-dimensional time-frequency spectrogram WT and a standard-time spectrogram WT of the starting state1Is related to the coefficient P1Comprises the following steps:
wherein,
are respectively WT and WT1The variance of (a) is determined,
is WT and WT1The covariance of (a).
The invention also provides an automatic state monitoring system suitable for the fan series, which comprises:
the first acquisition module is used for acquiring the fan series in the monitoring range;
the first building module is used for building a standard picture library for each series of fans;
the second construction module is used for constructing standard time spectrogram of each series of fans under different states;
the second transformation module is used for acquiring real-time vibration signals and pictures of the monitoring fan, and performing synchronous compression wavelet transformation on the real-time vibration signals to obtain a real-time three-dimensional time-frequency spectrogram;
the first judgment module is used for comparing the picture of the monitoring fan with the standard picture library and judging the series of the monitoring fan;
the second calculation module is used for calculating the correlation coefficient of the real-time three-dimensional time-frequency spectrogram and the standard-time spectrogram in different states corresponding to the series to which the draught fan belongs;
and the second judgment module is used for judging the state of the monitoring equipment based on the correlation coefficient.
Further, the second building module comprises:
the setting module is used for setting at least one standard fan for each model in each series;
the second acquisition module is used for acquiring a first vibration signal when the standard fan operates in different states;
the first calculation module is used for calculating standard vibration signals of each series of fans in different states based on the first vibration signals;
and the first transformation module is used for performing synchronous compression wavelet transformation on the standard vibration signals to obtain standard time spectrograms of each series of fans under different states.
Further, the standard picture library of each series of fans comprises pictures of various angles of each model of fan in the series.
Further, the first determining module includes:
and extracting the characteristics of the picture of the monitored fan and all pictures in the standard picture library, constructing corresponding characteristic vectors, calculating the similarity between the pictures, and selecting the fan series corresponding to the picture with the maximum similarity with the picture of the monitored fan as the series to which the monitored fan belongs.
Further, the different states of the fan comprise a starting state, a violent vibration state, a normal state, a shutdown state, a real-time three-dimensional time-frequency spectrogram WT and a standard-time spectrogram WT of the starting state1Is related to the coefficient P1Comprises the following steps:
wherein,
are respectively WT and WT1The variance of (a) is determined,
is WT and WT1The covariance of (a).
The invention provides a state automatic monitoring method and system suitable for fan series, aiming at the problems that the existing fan state monitoring scheme adopts the same standard vibration signal to monitor and judge different fan series and the fan state monitoring accuracy is low. According to the characteristics that the vibration signals of fans of the same series are similar and the vibration signals of different structures of different series are different, standard vibration signals are set for the fans of each series, the series to which the monitored fans belong is judged, and the standard vibration signals of the fans of the corresponding series are selected for comparison to judge the state of the fans. Therefore, the invention improves the accuracy of fan monitoring, and has high monitoring efficiency and low system processing complexity.
Drawings
FIG. 1 is a flow chart of a method for automatically monitoring a condition of a wind turbine system according to an embodiment;
FIGS. 2, 3 and 4 are schematic diagrams of three fan series structures respectively;
fig. 5 is a structural diagram of an automatic status monitoring system adapted to a fan series according to the second embodiment.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the drawings only show the components related to the present invention rather than the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation can be changed freely, and the layout of the components can be more complicated.
The invention is further described with reference to the following drawings and specific examples, which are not intended to be limiting.
Example one
As shown in fig. 1, the present embodiment provides an automatic monitoring method for a state of a fan series, including:
s1, acquiring a fan series in a monitoring range;
fans are widely used for ventilation, dust exhaust and cooling of factories, mines, tunnels, cooling towers, vehicles, ships and buildings, and for ventilation and draught of boilers and industrial furnaces and kilns; cooling and ventilation in air conditioning equipment and household appliances; drying and selecting grain, wind tunnel wind source and air cushion boat inflating and propelling. Therefore, in practical applications, a large number of fan states need to be automatically monitored. Because fan quantity, kind, model are various, if set up standard vibration signal to every fan alone to carry out fan state monitoring and judgement, handle complicacy, the treatment effeciency is low. If the same standard vibration signal is adopted for monitoring and judging different fan series, the accuracy rate of fan state monitoring is low. Therefore, the fan state monitoring method and the fan state monitoring device are based on the characteristic that the similarity of the vibration signals of the fans of the same series is higher than that of the vibration signals of the fans of different series, and the corresponding standards are set based on the fans of different series, so that the fan state monitoring of different fan series is realized.
In order to realize automatic monitoring of a large number of fans, the invention firstly collects fan series in a monitoring range so as to accurately monitor the fans in the monitoring range. The fan series comprises a DWT series roof fan, a SWF high-efficiency low-noise mixed flow fan, a CF series kitchen smoke exhaust pipeline fan, a GDF series centrifugal pipeline fan, 4-72(79) centrifugal fans and the like. For example, a DWT series rooftop fan may include a plurality of types of fans, and different types of fans may be selected according to different requirements of pressure, air volume and noise in use. For example, the DWT-I type axial flow roof fan is suitable for the use occasions of medium and low pressure and large flow. DWT-I series also includes 1200, 1400, 1500, 1600, 1800, 2000, 2200, and 2400mm fans. According to the invention, fans of different models are classified according to the series to which the fans belong, so that the data processing capacity of a large number of fans is reduced, and the monitoring efficiency is improved. The fan series may be obtained through a fan purchase list, and the like, which is not limited herein.
S2, constructing a standard picture library for each series of fans;
fig. 2, 3 and 4 respectively show three series of fans, namely a DWT series roof fan, a SWF high-efficiency low-noise mixed flow fan and a CF series kitchen smoke exhaust pipeline fan. The picture structures of different series of fans are completely different. Therefore, the invention constructs a standard picture library for each series of fans in turn. As fans in different series comprise fans of various models, the picture structures of the fans in the same series and different models are different. Therefore, the fan picture structure of each series of fan pictures is collected in sequence and stored in the standard picture library of the corresponding fan series.
The method comprises the steps of firstly collecting standard pictures for the fans of various models, wherein the standard pictures of the fans can be collected through official published pictures of the models, and field images can also be collected on the standard fans of the models. The collected fan pictures are named according to the models of the fan pictures, and the recognition efficiency of the fan is improved. The pictures corresponding to each type of fan are stored in the standard picture library corresponding to the series of fans, and specifically, the pictures of the fans in the same series can be stored in the folders corresponding to the series. For example, a fan picture for a DWT-I type axial flow rooftop fan is stored in a folder named DWT series rooftop fan.
In addition, in order to improve the identification precision of the fan series, for each model of fan, the invention collects pictures of each angle of the fan. The pictures of the same model fan at different angles can be distinguished by different serial numbers and put into a subfolder named by the model of the fan.
S3, constructing standard time spectrogram of each series of fans under different states;
the invention monitors the state of the fan by using the frequency domain characteristics of the fan vibration signal. Specifically, the real-time vibration signal of the fan is compared with a standard vibration signal to determine the state of the fan. Therefore, the specific steps of constructing the standard-time spectrogram of each series of fans in different states are as follows:
s31, setting at least one standard fan for each model in each series;
the invention fully considers the difference of vibration signals of fans of different models in the same series and sets at least one standard fan for each model in each series. Through integration and analysis of fans of different models, the standard vibration signals of the fans of each series can represent the characteristics of the fans of different models to the greatest extent. It should be noted that the present invention does not limit the number of each type of specific fans, and all of the fans may be set as one fan, or may be set according to the overall ratio of the monitoring fans. For example, when the number of DWT-I type axial-flow type roof fans is large in the monitoring fans, the number of standard fans in DWT-I type axial-flow type roof fans can be increased, and therefore automatic monitoring can be conducted on the fans better.
S32, collecting a first vibration signal when the standard fan runs in different states;
the standard fan runs in different states, and first vibration signals of the standard fan in all states are acquired. Specifically, a vibration sensor can be arranged on the fan bearing to acquire a vibration signal of the fan bearing. Sampling the vibration signal by adopting a specific sampling frequency, and acquiring the fan vibration signal within a period of time to obtain a time domain vibration signal.
The invention divides the state of the fan into a starting state, a violent vibration state, a normal state and a shutdown state. Specifically, the fan is respectively operated in a starting state, a violent vibration state, a normal state and a stopping state, and a vibration signal of the fan is collected through a vibration sensor.
S33, calculating standard vibration signals of each series of fans in different states based on the first vibration signals;
as mentioned above, fans of different series include a plurality of different first vibration signals, in order to calculate the standard vibration signals of each series of fans under different states, the invention stores the collected first vibration signals under the same series in the same folder, and the corresponding vibration signal waveform files are named jointly by model and fan state. Therefore, for the standard vibration signals of each series of fans under different states, the invention calls the vibration signal folders corresponding to the series, extracts the first vibration signals containing the same state, and calculates the average signal of the extracted first vibration signals to obtain the standard time domain signals of the series of fans under each working condition state. The invention sequentially processes all fan series to obtain standard vibration signals of all fan series in different states.
And S34, performing synchronous compression wavelet transformation on the standard vibration signals to obtain standard time spectrogram of each series of fans in different states.
In order to obtain the frequency domain information of the signal, the invention utilizes synchronous compression wavelet transform to process the standard vibration signal of each series of fans under different states, utilizes the characteristic that the phase in the frequency domain of the signal after the wavelet transform is not influenced by the scale transform to obtain the corresponding frequency under each scale, and then adds the scales under the same frequency, namely, the wavelet coefficient obtained by the wavelet transform is redistributed and compressed, thereby compressing the value near the same frequency into the frequency, improving the fuzzy phenomenon in the scale direction and improving the time-frequency resolution. The specific steps of the synchronous compression wavelet transform are not described in detail herein.
And obtaining a standard time spectrogram corresponding to the standard vibration signals of each series of fans in different states by synchronous compression wavelet transformation, wherein the abscissa of the graph is time b, the ordinate of the graph is frequency f, and the color of the graph represents signal energy distribution. The three-dimensional time-frequency spectrograms of the fans under different series of working conditions can be saved by using series names and fan states in a joint naming mode.
S4, collecting real-time vibration signals and pictures of the monitoring fan, and performing synchronous compression wavelet transformation on the real-time vibration signals to obtain a real-time three-dimensional time-frequency spectrogram;
in order to monitor the state of the fan, the invention collects the real-time vibration signal of the fan. The real-time vibration signal is collected as the standard signal, a vibration sensor such as a fan bearing is arranged at the position same as the standard signal, and the real-time vibration signal of the fan bearing is obtained. And after the real-time vibration signals are collected, performing synchronous compression wavelet transformation on the real-time vibration signals to obtain a real-time three-dimensional time-frequency spectrogram. The generation of the real-time three-dimensional time spectrum graph is consistent with the generation of the standard-time spectrum graph in step S34, and will not be described herein again.
In addition, the invention monitors different series of fans by adopting different standard signals so as to improve the accuracy of fan state monitoring. Therefore, for the fan to be monitored, the series to which the monitored fan belongs needs to be judged firstly. Therefore, the camera is arranged in the environment of the fan to collect the picture of the fan, and the camera and the vibration sensor are connected with the data processor to monitor the state of the fan by combining the collected picture and the real-time vibration signal.
It is worth noting that for monitoring the state of the fan, vibration signals of the fan need to be collected continuously so as to monitor the state of the fan in real time. And for the collection of fan pictures, only when the fan needs to be monitored for the first time, the fan images do not need to be collected continuously, the data transmission and processing amount is reduced, and the fan monitoring efficiency is improved.
S5, comparing the picture of the monitoring fan with the standard picture library, and judging the series of the monitoring fan;
specifically, the similarity between the acquired fan picture and the pictures in the standard picture library corresponding to each series of fans is calculated in sequence, and the greater the similarity is, the closer the monitoring fan is to the fan series corresponding to the picture is, that is, the monitoring fan is determined to be the fan series corresponding to the picture with the maximum similarity.
For the pictures of the monitoring fan and the pictures in the standard picture library, the invention respectively extracts the features in the pictures, and the extraction of the picture features can be extracted through a convolutional neural network model generated by training. Further, a feature vector of each picture is constructed for the picture based on the extracted features. And determining the similarity between the pictures by calculating the similarity between the feature vectors. The similarity can be calculated by the cosine of the included angle, the correlation coefficient, Dice, Jaccard, Ming's distance, Euclidean distance and the like.
As described above, the standard picture library corresponding to each series of fans stores standard pictures of fans of different types and at different angles, so that the present invention needs to compare with each picture in sequence. The similarity of the pictures is characterized by floating point numbers between [0,1], the larger the numerical value is, the more similar the two pictures are, and the smaller the numerical value is, the more dissimilar the two pictures are.
S6, calculating correlation coefficients of the real-time three-dimensional time-frequency spectrogram and the standard-time spectrogram in different states corresponding to the series to which the draught fan belongs;
after the series to which the fan belongs is judged, different states of the fan can be judged based on the standard time spectrogram of the series to which the fan belongs. Specifically, the invention relates to a real-time three-dimensional time-frequency spectrogram of a fan
And carrying out similarity calculation with the standard time spectrogram of the fan under each working condition state. Specifically, for the start-up state, the violent vibration state, the normal state and the shut-down stateState-corresponding standard time spectrogram WT1、WT2、WT3、WT4Calculating WT and WT separately1、WT2、WT3、WT4Is related to the coefficient P1、P2、P3、P4The method specifically comprises the following steps:
wherein,
are respectively WT and WT1The variance of (a) is determined,
is WT and WT1The covariance of (a). Similarly, P is calculated sequentially2、P3、P4。
And S7, judging the state of the monitoring equipment based on the correlation coefficient.
The larger the correlation coefficient of the time-frequency spectrograms of the two vibration signals is, the higher the similarity of the two vibration signals is, namely, the more likely the two vibration signals belong to the vibration signal of the fan in the same state. Thus, the present invention compares P1、P2、P3、P4And selecting the working condition of the standard signal corresponding to the maximum value as the real-time working condition of the fan, and realizing the automatic judgment of the state of the fan. For example, P1When the maximum, the fan is in a starting state; when P is present2When the vibration is maximum, the fan is in a violent vibration state; when P is present3When the maximum value is reached, the fan is in a normal state; when P is present4At maximum, the fan is in a shutdown state.
Example two
As shown in fig. 5, the present embodiment provides an automatic status monitoring system adapted to a fan series, including:
the first acquisition module is used for acquiring the fan series in the monitoring range;
fans are widely used for ventilation, dust exhaust and cooling of factories, mines, tunnels, cooling towers, vehicles, ships and buildings, and for ventilation and draught of boilers and industrial furnaces and kilns; cooling and ventilation in air conditioning equipment and household appliances; drying and selecting grain, wind tunnel wind source and air cushion boat inflating and propelling. Therefore, in practical applications, a large number of fan states need to be automatically monitored. Because fan quantity, kind, model are various, if set up standard vibration signal to every fan alone to carry out fan state monitoring and judgement, handle complicacy, the treatment effeciency is low. If the same standard vibration signal is adopted for monitoring and judging different fan series, the accuracy rate of fan state monitoring is low. Therefore, the fan state monitoring method and the fan state monitoring device are based on the characteristic that the similarity of the vibration signals of the fans of the same series is higher than that of the vibration signals of the fans of different series, and the corresponding standards are set based on the fans of different series, so that the fan state monitoring of different fan series is realized.
In order to realize automatic monitoring of a large number of fans, the invention firstly collects fan series in a monitoring range so as to accurately monitor the fans in the monitoring range. The fan series comprises a DWT series roof fan, a SWF high-efficiency low-noise mixed flow fan, a CF series kitchen smoke exhaust pipeline fan, a GDF series centrifugal pipeline fan, 4-72(79) centrifugal fans and the like. For example, a DWT series rooftop fan may include a plurality of types of fans, and different types of fans may be selected according to different requirements of pressure, air volume and noise in use. For example, the DWT-I type axial flow roof fan is suitable for the use occasions of medium and low pressure and large flow. DWT-I series also includes 1200, 1400, 1500, 1600, 1800, 2000, 2200, and 2400mm fans. According to the invention, fans of different models are classified according to the series to which the fans belong, so that the data processing capacity of a large number of fans is reduced, and the monitoring efficiency is improved. The fan series may be obtained through a fan purchase list, and the like, which is not limited herein.
The first building module is used for building a standard picture library for each series of fans;
the picture structures of different series of fans are completely different, for example, three series of fans: DWT series roof fan, SWF high-efficient low noise mixed flow fan, CF series kitchen exhaust pipe fan's structure is completely different. Therefore, the invention constructs a standard picture library for each series of fans in turn. As fans in different series comprise fans of various models, the picture structures of the fans in the same series and different models are different. Therefore, the fan picture structure of each series of fan pictures is collected in sequence and stored in the standard picture library of the corresponding fan series.
The method comprises the steps of firstly collecting standard pictures for the fans of various models, wherein the standard pictures of the fans can be collected through official published pictures of the models, and field images can also be collected on the standard fans of the models. The collected fan pictures are named according to the models of the fan pictures, and the recognition efficiency of the fan is improved. The pictures corresponding to each type of fan are stored in the standard picture library corresponding to the series of fans, and specifically, the pictures of the fans in the same series can be stored in the folders corresponding to the series. For example, a fan picture for a DWT-I type axial flow rooftop fan is stored in a folder named DWT series rooftop fan.
In addition, in order to improve the identification precision of the fan series, for each model of fan, the invention collects pictures of each angle of the fan. The pictures of the same model fan at different angles can be distinguished by different serial numbers and put into a subfolder named by the model of the fan.
The second construction module is used for constructing standard time spectrogram of each series of fans under different states;
the invention monitors the state of the fan by using the frequency domain characteristics of the fan vibration signal. Specifically, the real-time vibration signal of the fan is compared with a standard vibration signal to determine the state of the fan. Therefore, the constructing of the standard-time spectrogram of each series of fans in different states specifically includes:
the setting module is used for setting at least one standard fan for each model in each series;
the invention fully considers the difference of vibration signals of fans of different models in the same series and sets at least one standard fan for each model in each series. Through integration and analysis of fans of different models, the standard vibration signals of the fans of each series can represent the characteristics of the fans of different models to the greatest extent. It should be noted that the present invention does not limit the number of each type of specific fans, and all of the fans may be set as one fan, or may be set according to the overall ratio of the monitoring fans. For example, when the number of DWT-I type axial-flow type roof fans is large in the monitoring fans, the number of standard fans in DWT-I type axial-flow type roof fans can be increased, and therefore automatic monitoring can be conducted on the fans better.
The second acquisition module is used for acquiring a first vibration signal when the standard fan operates in different states;
the standard fan runs in different states, and first vibration signals of the standard fan in all states are acquired. Specifically, a vibration sensor can be arranged on the fan bearing to acquire a vibration signal of the fan bearing. Sampling the vibration signal by adopting a specific sampling frequency, and acquiring the fan vibration signal within a period of time to obtain a time domain vibration signal.
The invention divides the state of the fan into a starting state, a violent vibration state, a normal state and a shutdown state. Specifically, the fan is respectively operated in a starting state, a violent vibration state, a normal state and a stopping state, and a vibration signal of the fan is collected through a vibration sensor.
The first calculation module is used for calculating standard vibration signals of each series of fans in different states based on the first vibration signals;
as mentioned above, fans of different series include a plurality of different first vibration signals, in order to calculate the standard vibration signals of each series of fans under different states, the invention stores the collected first vibration signals under the same series in the same folder, and the corresponding vibration signal waveform files are named jointly by model and fan state. Therefore, for the standard vibration signals of each series of fans under different states, the invention calls the vibration signal folders corresponding to the series, extracts the first vibration signals containing the same state, and calculates the average signal of the extracted first vibration signals to obtain the standard time domain signals of the series of fans under each working condition state. The invention sequentially processes all fan series to obtain standard vibration signals of all fan series in different states.
And the first transformation module is used for performing synchronous compression wavelet transformation on the standard vibration signals to obtain standard time spectrograms of each series of fans under different states.
In order to obtain the frequency domain information of the signal, the invention utilizes synchronous compression wavelet transform to process the standard vibration signal of each series of fans under different states, utilizes the characteristic that the phase in the frequency domain of the signal after the wavelet transform is not influenced by the scale transform to obtain the corresponding frequency under each scale, and then adds the scales under the same frequency, namely, the wavelet coefficient obtained by the wavelet transform is redistributed and compressed, thereby compressing the value near the same frequency into the frequency, improving the fuzzy phenomenon in the scale direction and improving the time-frequency resolution. The specific steps of the synchronous compression wavelet transform are not described in detail herein.
And obtaining a standard time spectrogram corresponding to the standard vibration signals of each series of fans in different states by synchronous compression wavelet transformation, wherein the abscissa of the graph is time b, the ordinate of the graph is frequency f, and the color of the graph represents signal energy distribution. The three-dimensional time-frequency spectrograms of the fans under different series of working conditions can be saved by using series names and fan states in a joint naming mode.
The second transformation module is used for acquiring real-time vibration signals and pictures of the monitoring fan, and performing synchronous compression wavelet transformation on the real-time vibration signals to obtain a real-time three-dimensional time-frequency spectrogram;
in order to monitor the state of the fan, the invention collects the real-time vibration signal of the fan. The real-time vibration signal is collected as the standard signal, a vibration sensor such as a fan bearing is arranged at the position same as the standard signal, and the real-time vibration signal of the fan bearing is obtained. And after the real-time vibration signals are collected, performing synchronous compression wavelet transformation on the real-time vibration signals to obtain a real-time three-dimensional time-frequency spectrogram. The generation of the real-time three-dimensional time-frequency spectrogram is consistent with the generation of the standard-time spectrogram in the first transformation module, and details are not repeated herein.
In addition, the invention monitors different series of fans by adopting different standard signals so as to improve the accuracy of fan state monitoring. Therefore, for the fan to be monitored, the series to which the monitored fan belongs needs to be judged firstly. Therefore, the camera is arranged in the environment of the fan to collect the picture of the fan, and the camera and the vibration sensor are connected with the data processor to monitor the state of the fan by combining the collected picture and the real-time vibration signal.
It is worth noting that for monitoring the state of the fan, vibration signals of the fan need to be collected continuously so as to monitor the state of the fan in real time. And for the collection of fan pictures, only when the fan needs to be monitored for the first time, the fan images do not need to be collected continuously, the data transmission and processing amount is reduced, and the fan monitoring efficiency is improved.
The first judgment module is used for comparing the picture of the monitoring fan with the standard picture library and judging the series of the monitoring fan;
specifically, the similarity between the acquired fan picture and the pictures in the standard picture library corresponding to each series of fans is calculated in sequence, and the greater the similarity is, the closer the monitoring fan is to the fan series corresponding to the picture is, that is, the monitoring fan is determined to be the fan series corresponding to the picture with the maximum similarity.
For the pictures of the monitoring fan and the pictures in the standard picture library, the invention respectively extracts the features in the pictures, and the extraction of the picture features can be extracted through a convolutional neural network model generated by training. Further, a feature vector of each picture is constructed for the picture based on the extracted features. And determining the similarity between the pictures by calculating the similarity between the feature vectors. The similarity can be calculated by the cosine of the included angle, the correlation coefficient, Dice, Jaccard, Ming's distance, Euclidean distance and the like.
As described above, the standard picture library corresponding to each series of fans stores standard pictures of fans of different types and at different angles, so that the present invention needs to compare with each picture in sequence. The similarity of the pictures is characterized by floating point numbers between [0,1], the larger the numerical value is, the more similar the two pictures are, and the smaller the numerical value is, the more dissimilar the two pictures are.
The second calculation module is used for calculating the correlation coefficient of the real-time three-dimensional time-frequency spectrogram and the standard-time spectrogram in different states corresponding to the series to which the draught fan belongs;
after the series to which the fan belongs is judged, different states of the fan can be judged based on the standard time spectrogram of the series to which the fan belongs. Specifically, the invention relates to a real-time three-dimensional time-frequency spectrogram of a fan
And carrying out similarity calculation with the standard time spectrogram of the fan under each working condition state. Specifically, the standard time spectrogram WT corresponds to a start-up state, a severe-vibration state, a normal state, and a shut-down state1、WT2、WT3、WT4Calculating WT and WT separately1、WT2、WT3、WT4Is related to the coefficient P1、P2、P3、P4The method specifically comprises the following steps:
wherein,
are respectively WT and WT1The variance of (a) is determined,
is WT and WT1The covariance of (a). Similarly, P is calculated sequentially2、P3、P4。
And the second judgment module is used for judging the state of the monitoring equipment based on the correlation coefficient.
The larger the correlation coefficient of the time-frequency spectrograms of the two vibration signals is, the higher the similarity of the two vibration signals is, namely, the more likely the two vibration signals belong to the vibration signal of the fan in the same state. Thus, the present invention compares P1、P2、P3、P4And selecting the working condition of the standard signal corresponding to the maximum value as the real-time working condition of the fan, and realizing the automatic judgment of the state of the fan. For example, P1When the maximum, the fan is in a starting state; when P is present2When the vibration is maximum, the fan is in a violent vibration state; when P is present3When the maximum value is reached, the fan is in a normal state; when P is present4At maximum, the fan is in a shutdown state.
Therefore, the invention provides the automatic state monitoring method and system suitable for the fan series aiming at the problems that the existing fan state monitoring scheme adopts the same standard vibration signal to monitor and judge different fan series and the fan state monitoring accuracy is low. According to the characteristics that the vibration signals of fans of the same series are similar and the vibration signals of different structures of different series are different, standard vibration signals are set for the fans of each series, the series to which the monitored fans belong is judged, and the standard vibration signals of the fans of the corresponding series are selected for comparison to judge the state of the fans. Therefore, the invention improves the accuracy of fan monitoring, and has high monitoring efficiency and low system processing complexity.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (8)
1. A state automatic monitoring method suitable for a fan series is characterized by comprising the following steps:
s1, acquiring a fan series in a monitoring range;
s2, constructing a standard picture library for each series of fans;
s3, constructing standard time spectrogram of each series of fans under different states;
s4, collecting real-time vibration signals and pictures of the monitoring fan, and performing synchronous compression wavelet transformation on the real-time vibration signals to obtain a real-time three-dimensional time-frequency spectrogram;
s5, comparing the picture of the monitoring fan with the standard picture library, and judging the series of the monitoring fan;
s6, calculating correlation coefficients of the real-time three-dimensional time-frequency spectrogram and the standard-time spectrogram in different states corresponding to the series to which the draught fan belongs;
s7, judging the state of the monitoring equipment based on the correlation coefficient;
the different states of the fan comprise a starting state, a violent vibration state, a normal state, a shutdown state and a real-time three-dimensional time-frequency spectrogram
Normal time frequency spectrum chart of starting state
Correlation coefficient of
Comprises the following steps:
wherein,
are respectively as
And
the variance of (a) is determined,
is composed of
And
the covariance of (a).
2. The automatic status monitoring method according to claim 1, wherein the step S3 includes:
s31, setting at least one standard fan for each model in each series;
s32, collecting a first vibration signal when the standard fan runs in different states;
s33, calculating standard vibration signals of each series of fans in different states based on the first vibration signals;
and S34, performing synchronous compression wavelet transformation on the standard vibration signals to obtain standard time spectrogram of each series of fans in different states.
3. The method of claim 2, wherein the standard picture library of each fan type includes pictures of a plurality of fan angles of each fan type in the fan type.
4. The method for automatically monitoring the state according to claim 3, wherein the step S5 specifically comprises:
and extracting the characteristics of the picture of the monitored fan and all pictures in the standard picture library, constructing corresponding characteristic vectors, calculating the similarity between the pictures, and selecting the fan series corresponding to the picture with the maximum similarity with the picture of the monitored fan as the series to which the monitored fan belongs.
5. A condition automatic monitoring system adapted to a fan train, comprising:
the first acquisition module is used for acquiring the fan series in the monitoring range;
the first building module is used for building a standard picture library for each series of fans;
the second construction module is used for constructing standard time spectrogram of each series of fans under different states;
the second transformation module is used for acquiring real-time vibration signals and pictures of the monitoring fan, and performing synchronous compression wavelet transformation on the real-time vibration signals to obtain a real-time three-dimensional time-frequency spectrogram;
the first judgment module is used for comparing the picture of the monitoring fan with the standard picture library and judging the series of the monitoring fan;
the second calculation module is used for calculating the correlation coefficient of the real-time three-dimensional time-frequency spectrogram and the standard-time spectrogram in different states corresponding to the series to which the draught fan belongs;
the second judgment module is used for judging the state of the monitoring equipment based on the correlation coefficient;
the different states of the fan comprise a starting state, a violent vibration state, a normal state, a shutdown state and a real-time three-dimensional time-frequency spectrogram
Normal time frequency spectrum chart of starting state
Correlation coefficient of
Comprises the following steps:
wherein,
are respectively as
And
the variance of (a) is determined,
is composed of
And
the covariance of (a).
6. The automatic status monitoring system of claim 5, wherein the second building module comprises:
the setting module is used for setting at least one standard fan for each model in each series;
the second acquisition module is used for acquiring a first vibration signal when the standard fan operates in different states;
the first calculation module is used for calculating standard vibration signals of each series of fans in different states based on the first vibration signals;
and the first transformation module is used for performing synchronous compression wavelet transformation on the standard vibration signals to obtain standard time spectrograms of each series of fans under different states.
7. The system of claim 6, wherein the standard picture library for each fan type includes pictures of multiple angles of each fan type in the fan type.
8. The automatic status monitoring system according to claim 7, wherein the first determining module comprises:
and extracting the characteristics of the picture of the monitored fan and all pictures in the standard picture library, constructing corresponding characteristic vectors, calculating the similarity between the pictures, and selecting the fan series corresponding to the picture with the maximum similarity with the picture of the monitored fan as the series to which the monitored fan belongs.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010189549.5A CN111059004B (en) | 2020-03-18 | 2020-03-18 | Automatic state monitoring method and system suitable for fan series |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010189549.5A CN111059004B (en) | 2020-03-18 | 2020-03-18 | Automatic state monitoring method and system suitable for fan series |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111059004A CN111059004A (en) | 2020-04-24 |
| CN111059004B true CN111059004B (en) | 2020-06-09 |
Family
ID=70307964
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010189549.5A Active CN111059004B (en) | 2020-03-18 | 2020-03-18 | Automatic state monitoring method and system suitable for fan series |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111059004B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112211793A (en) * | 2020-08-25 | 2021-01-12 | 华北电力大学（保定） | Wind turbine generator fault automatic identification method based on video image analysis |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102778358A (en) * | 2012-06-04 | 2012-11-14 | 上海东锐风电技术有限公司 | Failure prediction model establishing method and system as well as fan monitoring pre-warning system and method |
| FR3016042B1 (en) * | 2013-12-30 | 2017-03-03 | Flakt Solyvent-Ventec | SYSTEM AND METHOD FOR ASSISTING THE DIAGNOSIS OF THE OPERATING STATE OF A ROTATING MACHINE |
| CN105547698B (en) * | 2015-12-31 | 2017-11-14 | 新疆金风科技股份有限公司 | The method for diagnosing faults and device of rolling bearing |
| CN107218999A (en) * | 2017-06-08 | 2017-09-29 | 攀钢集团西昌钢钒有限公司 | Continuous annealing unit fan vibration diagnostic device |
| CN107271170B (en) * | 2017-07-12 | 2019-05-31 | 西安因联信息科技有限公司 | A kind of automatic diagnosis method and system of mechanical equipment fault type |
| CN108154166A (en) * | 2017-11-30 | 2018-06-12 | 辽宁科技大学 | A kind of fan intelligent fault diagnosis system and method |
-
2020
- 2020-03-18 CN CN202010189549.5A patent/CN111059004B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN111059004A (en) | 2020-04-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Glowacz | Acoustic fault analysis of three commutator motors | |
| US7409854B2 (en) | Method and apparatus for determining an operating status of a turbine engine | |
| CN110608187A (en) | Axial flow compressor stall surge prediction device based on frequency characteristic change | |
| KR20110112427A (en) | Wind turbine for wind power generation | |
| CN102798529A (en) | Method and system for diagnosing bearing faults of large-size wind turbine bearing | |
| CN105756864B (en) | Blade imbalance fault diagnostic method based on double-fed wind power generator group stator current | |
| CN101629572A (en) | Stall and surge detection system and method | |
| CN109190166A (en) | A kind of blade pump cavitation determines and state evaluating method and its system | |
| CN109255333B (en) | Fault hybrid diagnosis method for rolling bearing of large wind turbine generator | |
| CN111059004B (en) | Automatic state monitoring method and system suitable for fan series | |
| CN108801641A (en) | The fault diagnosis of exhaust-driven turbo-charger exhaust-gas turbo charger and Reliability Prediction Method and its system | |
| CN112285554A (en) | Information fusion-based demagnetization fault diagnosis method and device for permanent magnet synchronous motor | |
| CN109630449B (en) | Three-proofing ventilation equipment fault prediction system based on RBF algorithm | |
| CN104888396A (en) | Method for detecting transient faults of fire pump in quick starting process | |
| CN110925233A (en) | Compressor surge fault diagnosis method based on acoustic signals | |
| CN114708885A (en) | Fan fault early warning method based on sound signals | |
| CN110953179A (en) | Fan state monitoring method and system based on inverse wavelet transform | |
| CN105829983A (en) | Apparatus and method for detecting the current damaged state of a machine | |
| CN110023621B (en) | Determining load on wind turbine | |
| US8682563B2 (en) | System and method for predicting turbine rub | |
| CN108131321B (en) | A kind of axial fan stall fault monitoring method | |
| US20210285457A1 (en) | Surging precursor detecting device, method of detecting surging precursor, and program | |
| CN110618984B (en) | Shutdown vibration data cleaning method | |
| CN109341780B (en) | Multi-hand low-cost fan fault monitoring method | |
| CN211288186U (en) | Axial flow compressor stall surge prediction device based on frequency characteristic change |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Li Xiaoming Inventor after: Chu Ning Inventor before: Xu Min Inventor before: Zhang Qiangsheng Inventor before: Xie Bo Inventor before: Yan Meiying |
|
| CB03 | Change of inventor or designer information | ||
| CP03 | Change of name, title or address |
Address after: 312300 No. 1818, Renmin West Road, Dongguan Street, Shangyu District, Shaoxing City, Zhejiang Province Patentee after: Zhejiang Shangfeng high tech special wind industry Co.,Ltd. Address before: 312300 No. 1818 Renmin West Road, Shangyu District, Zhejiang, Shaoxing Patentee before: ZHEJIANG SHANGFENG HI-TECH SPECIALIZED WIND INDUSTRIAL Co.,Ltd. |
|
| CP03 | Change of name, title or address |