CN117708748B - Operation monitoring system and method for centrifugal fan - Google Patents
Operation monitoring system and method for centrifugal fan Download PDFInfo
- Publication number
- CN117708748B CN117708748B CN202410160641.7A CN202410160641A CN117708748B CN 117708748 B CN117708748 B CN 117708748B CN 202410160641 A CN202410160641 A CN 202410160641A CN 117708748 B CN117708748 B CN 117708748B
- Authority
- CN
- China
- Prior art keywords
- operation monitoring
- dimension
- parameter
- fluctuation
- monitoring parameters
- 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 330
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000005070 sampling Methods 0.000 claims abstract description 113
- 230000002159 abnormal effect Effects 0.000 claims abstract description 109
- 230000035945 sensitivity Effects 0.000 claims abstract description 18
- 238000001514 detection method Methods 0.000 claims abstract description 12
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000010606 normalization Methods 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 8
- 230000005856 abnormality Effects 0.000 claims description 4
- 230000008569 process Effects 0.000 description 7
- 230000009467 reduction Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 238000009776 industrial production Methods 0.000 description 4
- 230000006399 behavior Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000009423 ventilation Methods 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Abstract
The invention relates to the technical field of abnormal data identification, in particular to an operation monitoring system and method for a centrifugal fan. According to the operation monitoring parameters of each dimension at each sampling time and the local fluctuation characteristics of the corresponding fitting operation curve, the abnormal fluctuation coefficient of each operation monitoring parameter is obtained, and the fluctuation habit characteristics of the operation monitoring parameters are further combined to obtain the noise factor of each operation monitoring parameter; and adjusting a preset K neighborhood parameter according to the noise factor, thereby acquiring an abnormal operation monitoring parameter and early warning. According to the invention, the fluctuation characteristics of the operation monitoring parameters are analyzed by combining the fluctuation characteristic difference of noise and abnormal operation of the centrifugal fan to obtain the noise factor of each operation monitoring data, so that the sensitivity of each operation monitoring parameter is adjusted during LOF detection, the proper K neighborhood parameter is determined, the false alarm rate of noise is reduced, and the accuracy and efficiency of operation monitoring are improved.
Description
Technical Field
The invention relates to the technical field of abnormal data identification, in particular to an operation monitoring system and method for a centrifugal fan.
Background
The centrifugal fan is widely applied to industrial manufacturing scenes such as ventilation, dust removal and filtration, cooling and pressurization, and the like, so the operation monitoring of the centrifugal fan is also an important ring for ensuring the safety of industrial production environment and improving the industrial production efficiency. When monitoring the operation parameters of the centrifugal fan, a local outlier factor (Local Outlier Factor, LOF) algorithm is generally adopted to perform outlier judgment on the monitored data, so that abnormal data in the operation process is pre-warned.
However, since the monitoring data of the centrifugal fan may be interfered in the process of collecting the monitoring data through the sensor, a large amount of noise data exists in the monitoring data, the noise data is similar to the abnormal fluctuation of the abnormal operation data, and the noise data is easy to be mistakenly used as the abnormal operation data to carry out false alarm so as to prompt related operators to process, thereby affecting the industrial production efficiency, and therefore, the selection of a proper K neighborhood parameter has a critical influence on the processing result of the LOF algorithm. However, a fixed K neighborhood parameter is usually set in the conventional algorithm, and a larger K neighborhood parameter may reduce the sensitivity of the algorithm to abnormal data, thereby causing abnormal data missing detection; the smaller K neighborhood parameter can also amplify the sensitivity to noise data, and false alarm is carried out by taking the noise data as abnormal operation data; unsuitable K-neighborhood parameters will affect the operation monitoring accuracy and efficiency of the centrifugal fan.
Disclosure of Invention
In order to solve the technical problems of low operation monitoring accuracy and efficiency caused by improper parameter setting of the existing LOF algorithm, the invention aims to provide an operation monitoring system and method for a centrifugal fan, and the adopted technical scheme is as follows:
the invention provides an operation monitoring method for a centrifugal fan, which comprises the following steps:
Acquiring operation monitoring parameters of at least two dimensions of the centrifugal fan at each sampling moment, and acquiring a fitting operation curve of the operation monitoring parameters of each dimension;
Obtaining abnormal fluctuation coefficients of the operation monitoring parameters of the corresponding dimensions at the corresponding sampling moments according to the local fluctuation characteristics of the operation monitoring parameters of the dimensions at each sampling moment and the local fluctuation characteristics of the corresponding fitting operation curves; under each sampling time, acquiring a noise factor of the operation monitoring parameter of the corresponding dimension under each sampling time according to the difference of the abnormal fluctuation coefficient of the operation monitoring parameter of each dimension and the operation monitoring parameters of all the other dimensions and the fluctuation habit characteristics of the operation monitoring parameter of the corresponding dimension;
and adjusting a preset K neighborhood parameter of the operation monitoring parameter of each dimension at each sampling time according to the noise factors, and performing anomaly detection on the operation monitoring parameter according to the adjusted preset K neighborhood parameter.
Further, the method for acquiring the abnormal fluctuation coefficient comprises the following steps:
Acquiring a local trend reference coefficient of an operation monitoring parameter of each dimension at each sampling time; acquiring fitting deviation between the operation monitoring parameters of each dimension and the fitting operation parameters of the corresponding sampling time on the fitting operation curve; according to the operation monitoring parameters and the local fluctuation characteristic difference of the fitting deviation, the abnormal fluctuation coefficient is obtained by combining the local trend reference coefficient; the calculation formula of the abnormal fluctuation coefficient is as follows:
; wherein/> For/>Fifth/th at each sampling instantAbnormal fluctuation coefficients of the operation monitoring parameters of the dimension; /(I)For/>Fifth/th at each sampling instantStandard deviation of all operation monitoring parameters in a preset neighborhood of the operation monitoring parameters of the dimension; /(I)For/>Fifth/th at each sampling instantThe standard deviation of fitting operation parameters under the sampling time corresponding to the fitting curve and all operation monitoring parameters in the preset neighborhood of the dimension operation monitoring parameters; /(I)For/>Fifth/th at each sampling instantLocal trend reference coefficients of the dimensional operation monitoring parameters; /(I)Is a normalization function; /(I)Is a preset first positive parameter.
Further, the calculation formula of the local trend reference coefficient is as follows:
; wherein/> For/>Fifth/th at each sampling instantLocal trend reference coefficients of the dimensional operation monitoring parameters; /(I)For/>Fifth/th at each sampling instantThe number of extreme points on the fitting curve in the preset neighborhood of the operation monitoring parameters of the dimension; /(I)For/>Fifth/th at each sampling instantFitting the extremely poor fitting operation parameters on the fitting curves in the preset neighborhood of the dimensional operation monitoring parameters; /(I)Is a normalization function; /(I)Is a natural constant.
Further, the method for obtaining the noise factor comprises the following steps:
Acquiring fluctuation sensitivity factors of the operation monitoring parameters of each dimension according to fluctuation characteristics of the operation monitoring parameters of each dimension in a preset time period; acquiring a noise factor according to a calculation formula of the noise factor; the noise factor is calculated by the following formula:
; wherein/> For/>Fifth/th at each sampling instantA noise factor of the dimensional operation monitoring parameter; /(I)For/>Fifth/th at each sampling instantA fluctuation sensitive factor of the operation monitoring parameter of the dimension; /(I)For/>Except for the/>, at the sample time instantFirst/>, in all dimensions outside the dimensionA fluctuation sensitive factor of the operation monitoring parameter of the dimension; /(I)To be except for the (th)Running the dimension serial numbers of the monitoring parameters in all dimensions outside the dimension; /(I)The total dimension number of the monitoring parameters is used for running; /(I)For/>Fifth/th at each sampling instantAbnormal fluctuation coefficients of the operation monitoring parameters of the dimension; /(I)For/>Fifth/th at each sampling instantAbnormal fluctuation coefficients of the operation monitoring parameters of the dimension; /(I)Is a standard normalization function; Is a preset second positive parameter.
Further, the calculation formula of the fluctuation sensitive factor includes:
; wherein/> For/>Fifth/th at each sampling instantA fluctuation sensitive factor of the operation monitoring parameter of the dimension; /(I)The sequence number of the operation monitoring parameter in the preset time period; /(I)For/>Fifth/th at each sampling instantThe total number of the operation monitoring parameters in the preset time period of the operation monitoring parameters of the dimension; /(I)For/>Fifth/th at each sampling instantThe first/>, within a preset period of time, of the operational monitoring parameters of the dimensionAmplitude standard deviations of all operation monitoring parameters in a preset neighborhood of each operation monitoring parameter; /(I)For/>Fifth/th at each sampling instantThe first/>, within a preset period of time, of the operational monitoring parameters of the dimensionThe average value of the amplitude values of all the operation monitoring parameters in the preset neighborhood of each operation monitoring parameter; For/> Fifth/th at each sampling instantThe first/>, within a preset period of time, of the operational monitoring parameters of the dimensionThe number of extreme points in the preset neighborhood of each operation monitoring parameter; /(I)Is a normalization function; /(I)Is a natural constant; /(I)Is an intermediate parameter; /(I)To positively adjust the parameter.
Further, the method for adjusting the preset K neighborhood parameters includes:
And adding a preset positive constant to the noise factor as an adjustment coefficient, multiplying the preset K neighborhood parameter by the adjustment coefficient, and rounding to obtain an adjusted K neighborhood parameter.
Further, the dimensions of the operation monitoring parameters at least comprise gas pressure, fan rotation speed, gas flow and gas temperature.
Further, the method for detecting the abnormality of the operation monitoring parameter is an LOF algorithm.
Further, the method for acquiring the preset neighborhood comprises the following steps:
And taking the operation monitoring parameters of each dimension at each sampling time as a center, and respectively acquiring a preset number of operation monitoring parameters along two directions of the acquisition time sequence to construct a preset neighborhood.
The invention also provides an operation monitoring system for the centrifugal fan, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize the steps of the operation monitoring method for the centrifugal fan.
The invention has the following beneficial effects:
Firstly, acquiring multi-dimensional operation monitoring parameters of a centrifugal fan at each sampling moment and all fitting operation curves of corresponding dimensions, and further acquiring abnormal fluctuation coefficients of the operation monitoring parameters of the corresponding dimensions at the corresponding sampling moment according to local fluctuation characteristics of the operation monitoring parameters of the corresponding dimensions at each sampling moment and local fluctuation characteristics of the corresponding fitting operation curves; under each sampling time, taking the characteristic that noise causes low possibility of abnormal fluctuation of all the dimension operation monitoring parameters into consideration, so that according to the difference between the abnormal fluctuation coefficients of the operation monitoring parameters of each dimension and the operation monitoring parameters of the other all the dimensions and the fluctuation habit characteristics of the operation monitoring parameters of the corresponding dimension, the noise factors of the operation monitoring parameters of the corresponding dimension under each sampling time are obtained, and the more the operation monitoring parameters tend to fluctuate, the influence of the operation monitoring parameters is weakened when the abnormal fluctuation is analyzed so as to accurately obtain the noise factors of each operation monitoring parameter; the sensitivity to noise during abnormal detection can be adjusted by adjusting the preset K neighborhood parameters of each operation monitoring parameter according to the noise factors, so that false detection is reduced, and the abnormal operation monitoring parameters are accurately acquired and early warned. According to the invention, by combining fluctuation characteristics caused by noise and fluctuation characteristics difference caused by abnormal operation of the centrifugal fan, fluctuation characteristics of operation monitoring parameters are analyzed to obtain noise factors of each operation monitoring data, so that sensitivity to the operation monitoring parameters with high noise possibility in abnormal detection is adjusted, K neighborhood parameters are adjusted, false alarm rate is reduced, and operation monitoring accuracy and efficiency are improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flow chart of a method for monitoring operation of a centrifugal fan according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following description refers to the specific implementation, structure, characteristics and effects of a system and a method for monitoring the operation of a centrifugal fan according to the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of an operation monitoring system and method for a centrifugal fan provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for monitoring operation of a centrifugal fan according to an embodiment of the present invention is shown, the method includes the following steps:
step S1, acquiring operation monitoring parameters of at least two dimensions of the centrifugal fan at each sampling moment, and acquiring a fitting operation curve of the operation monitoring parameters of each dimension.
In order to monitor abnormal operation of a centrifugal fan and avoid noise interference false alarm abnormality, the embodiment of the invention firstly analyzes fluctuation conditions of the operation monitoring parameters at each sampling moment by collecting the operation monitoring parameters at different dimensionalities at each sampling moment and further combining local fluctuation characteristics of the noise and the abnormal operation parameters to obtain noise factors of each operation monitoring parameter, thereby adjusting K neighborhood parameters of the existing LOF algorithm, reducing sensitivity to noise and accurately monitoring the abnormal operation while reducing false alarm to the noise.
In one embodiment of the invention, various sensors are arranged on the centrifugal fan, and the operation monitoring parameters in each dimension are collected in real time at the sampling frequency of 5s each time, wherein the dimension refers to the type of the operation monitoring parameters, and the operation monitoring parameters can comprise gas pressure, fan rotating speed, gas flow and gas temperature. And because the dimension and unit of the monitoring parameters of each dimension are different, the collected operation monitoring parameters are subjected to standardized pretreatment, so that the subsequent parameter fluctuation analysis is further facilitated.
It should be noted that, the standardized pretreatment of the data is a common technical means for those skilled in the art, and will not be described herein; the implementer can also obtain other monitoring parameter dimension types and quantity which can be used for evaluating the running state of the centrifugal fan according to the specific running scene and running management regulation of the centrifugal fan, and can also set time periods of other time spans for analysis and monitoring.
Because noise and abnormal operation of a centrifugal fan both can cause abnormal fluctuation of operation monitoring parameters, but the noise is often expressed as random disturbance small abnormal fluctuation in the operation monitoring parameters, and the influence of the noise on the overall fluctuation trend of the operation monitoring parameters in each dimension is low, after the operation monitoring parameters in each dimension are obtained, the embodiment of the invention further obtains a fitting operation curve of the operation monitoring parameters in each dimension, so that the possibility that each operation monitoring parameter is noise, namely the abnormal fluctuation coefficient, is obtained by judging the fitting fluctuation characteristic of the operation monitoring parameters and the fluctuation characteristic of the operation monitoring parameters which are actually collected.
Preferably, in one embodiment of the invention, the least square method is considered to be capable of adapting to various complex data change trends, has certain robustness to noise, can resist the influence of noise, and can accurately and efficiently fit a data curve; based on the above, a least square fitting method is adopted to obtain a fitting operation curve of operation monitoring parameters of each dimension.
It should be noted that, the least square method is a technical means well known to those skilled in the art, and will not be described herein again; in other embodiments of the present invention, the practitioner may also fit the running curve using polynomial fitting or spline interpolation.
Step S2, according to the local fluctuation characteristics of the operation monitoring parameters of each dimension at each sampling moment and the local fluctuation characteristics of the corresponding fitting operation curve, obtaining the abnormal fluctuation coefficients of the operation monitoring parameters of the corresponding dimension at the corresponding sampling moment; and under each sampling time, acquiring the noise factor of the operation monitoring parameter of the corresponding dimension under each sampling time according to the difference of the abnormal fluctuation coefficient of the operation monitoring parameter of each dimension and the operation monitoring parameters of all the other dimensions and the fluctuation habit characteristics of the operation monitoring parameter of the corresponding dimension.
Because the abnormal fluctuation of the random noise disorder has little influence on the fluctuation trend of the operation monitoring data, the abnormal fluctuation can not be completely fitted into the curve in the process of fitting the operation curve, and the abnormal fluctuation is further represented as a certain fitting deviation of the noise relative to the fitted curve; the abnormal operation of the centrifugal fan often causes that the operation parameters are shown to be fluctuated with a certain abnormal amplitude level after mutation due to equipment faults or abnormal load and the like, and the abnormal fluctuation has a certain influence on the fluctuation trend of the operation monitoring parameters and is usually fitted into the operation curve directly or approaches to a fitted curve. Because of certain difference between the fluctuation characteristics of the operation monitoring parameters caused by noise and abnormal operation, whether the abnormal fluctuation of the operation monitoring parameters is caused by noise or caused by abnormal operation of the centrifugal fan can be estimated according to the corresponding fluctuation characteristics, and therefore the embodiment of the invention obtains the abnormal fluctuation coefficient of the operation monitoring parameters of the corresponding dimension at the corresponding sampling moment according to the local fluctuation characteristics of the operation monitoring parameters of each dimension at each sampling moment and the local fluctuation characteristics of the corresponding fitting operation curve.
In one embodiment of the invention, firstly, taking the operation monitoring parameters of each dimension at each sampling time as a center, respectively acquiring a preset number of operation monitoring parameters along two directions of the acquisition time sequence to construct a preset neighborhood, and further analyzing the local fluctuation characteristics of each operation monitoring parameter in a local range to acquire the corresponding abnormal fluctuation coefficient. In the embodiment of the invention, the preset number is set to 10, that is, the time sequence length of the preset neighborhood is 21; in other embodiments of the present invention, other sized neighbors may be set for analysis of local wave characteristics, depending on the implementation.
Preferably, in one embodiment of the present invention, considering that noise and noise of abnormal operation parameters both present a certain severe fluctuation condition in a local range, but the noise deviates from a fitting curve more than the abnormal operation parameters, the fitting deviation value of the noise corresponding operation monitoring parameters is larger and more random than the fitting deviation value of the abnormal operation parameters, and the standard deviation can well reflect the fluctuation condition of data; in addition, the characteristic of distinguishing fluctuation of noise and abnormal operation parameters in a local range is considered, the fluctuation trend characteristics in the local range are reflected by the side surfaces of the local trend reference coefficients, and the larger the local trend reference coefficient is, the more monotonous local fluctuation parameters of an operation curve in the local range are explained, the more intense amplitude fluctuation is, and the more accords with the partial fluctuation trend characteristics when the centrifugal fan operates abnormally; conversely, the smaller and frequent the fluctuation, the more consistent the fluctuation trend characteristic of the noise; based on the abnormal fluctuation coefficient, the method comprises the steps of obtaining local monotone trend coefficients of operation monitoring parameters of each dimension at each sampling time; acquiring fitting deviation between the operation monitoring parameters of each dimension and fitting operation parameters of the fitting operation curve at corresponding sampling moments; according to the operation monitoring parameters and the local fluctuation characteristic differences of the fitting deviation, acquiring abnormal fluctuation coefficients by combining local monotone trend coefficients; the calculation formula of the abnormal fluctuation coefficient is as follows:
Wherein, For/>Fifth/th at each sampling instantAbnormal fluctuation coefficients of the operation monitoring parameters of the dimension; /(I)For/>Fifth/th at each sampling instantStandard deviation of all operation monitoring parameters in a preset neighborhood of the operation monitoring parameters of the dimension; /(I)Is the firstFifth/th at each sampling instantFitting deviation standard deviation of fitting operation parameters under sampling time corresponding to all operation monitoring parameters and fitting curves in preset neighborhood of the dimension operation monitoring parameters; /(I)For/>Fifth/th at each sampling instantLocal trend reference coefficients of the dimensional operation monitoring parameters; /(I)Is a normalization function; /(I)To preset the first positive parameter, in the embodiment of the present invention,/>Taking 0.1, ensuring that the denominator is not zero. It should be noted that, the obtaining of the standard deviation is a means well known to those skilled in the art, and is not described herein.
In a calculation formula of the abnormal fluctuation coefficient, the larger the standard deviation of all operation monitoring parameters in a preset neighborhood is, the larger the fluctuation degree of the parameters is, the larger the possibility of noise or abnormal operation parameters is, and the smaller the possibility of normal operation parameters is; the larger the standard deviation of the fitting deviation is, the more the fitting deviation characteristics corresponding to noise are met, the more the operation monitoring parameters are likely to be noise, and otherwise, the operation monitoring parameters are likely to be abnormal operation parameters or normal operation parameters; so is combined withAnd/>Judging the abnormal fluctuation coefficient of each operation monitoring parameter,/>The smaller the absolute value of the difference, the greater the likelihood that the corresponding operation monitoring parameter is noise, but when/>And/>When they are small, the higher the possibility of being a normal operation monitoring parameter, the more likely it will causeAlso show smaller, so pass/>Further constraint, such that/>The closer to 1, the decrease in it is the result of normal operation monitoring parameters/>A small influence; after normalizing the data, the corresponding normalization value is subtracted by 1 to adjust the corresponding logic relation; the local trend reference coefficient reflects the fluctuation characteristics in the preset neighborhood, the larger the value is, the more accords with the local fluctuation trend characteristics of abnormal operation of the equipment, the smaller the corresponding abnormal fluctuation coefficient is, the greater the possibility of abnormal fluctuation caused by noise is, and the smaller the possibility of abnormal operation parameters or normal operation parameters is.
Preferably, in one embodiment of the present invention, considering that the number and the range of the extreme points in the local range can reflect the fluctuation trend of the local range, the smaller the number and the smaller the range of the extreme points, the more random and small the fluctuation trend of the operation monitoring parameter in the local range, the more accords with the local fluctuation characteristic of the noise point; the calculation formula of the local trend reference coefficient is as follows:
Wherein, For/>Fifth/th at each sampling instantLocal trend reference coefficients of the dimensional operation monitoring parameters; /(I)For/>Fifth/th at each sampling instantThe number of extreme points on the fitting curve in the preset neighborhood of the operation monitoring parameters of the dimension; /(I)For/>Fifth/th at each sampling instantFitting the extremely poor fitting operation parameters on the fitting curves in the preset neighborhood of the dimensional operation monitoring parameters; Is a normalization function; /(I) Is a natural constant. It should be noted that, the extreme point and the extremely poor acquisition method are means well known to those skilled in the art, and are not described herein.
In a calculation formula of the local trend reference coefficient, the fewer the number of extreme points and the larger the extreme difference are, the more monotonous the local fluctuation parameters of the operation curve in the local range are, the more intense the amplitude fluctuation is, and the more accords with the partial fluctuation trend characteristics of the centrifugal fan during abnormal operation; whereas the smaller and frequent the fluctuations, the more consistent the wave trend characteristics of the noise.
Noise and abnormal operation can cause abnormal fluctuation of operation parameters, but as the centrifugal fan is usually abnormal when operating abnormally, a plurality of operation monitoring parameters are simultaneously abnormal, and random situations such as noise are often represented in the condition that abnormal fluctuation of one operation monitoring parameter occurs independently, the possibility of abnormal fluctuation of all dimension operation monitoring parameters caused by noise is extremely low; meanwhile, as the fluctuation condition of the operation monitoring parameters in each dimension under the normal operation condition is possibly inconsistent, such as gas flow, under different working conditions, the industrial production system is continuously adjusted according to the requirements due to different production requirements at regular intervals, so that the gas flow monitoring parameters show relatively frequent fluctuation relative to the operation monitoring parameters in other dimensions, and the fluctuation habit in the normal operation has a certain influence on abnormal fluctuation analysis; therefore, in the embodiment of the invention, under each sampling time, the noise factor of the operation monitoring parameter of the corresponding dimension under each sampling time is obtained according to the difference between the abnormal fluctuation coefficients of the operation monitoring parameter of each dimension and the operation monitoring parameters of all the other dimensions and the fluctuation habit characteristics of the operation monitoring parameters of the corresponding dimension.
Preferably, in one embodiment of the present invention, considering that noise is extremely low and abnormal fluctuation of operation monitoring parameters in all dimensions may be caused, and fluctuation habits of operation monitoring parameters in different dimensions are different, the more the operation monitoring parameters tend to fluctuate, the fluctuation performance of fluctuation generated by noise and abnormal operation of equipment in the overall fluctuation trend is relatively weaker, which is unfavorable for abnormal fluctuation detection; based on the fluctuation characteristics of the operation monitoring parameters of each dimension in a preset time period, the fluctuation sensitivity factors of the operation monitoring parameters of each dimension are obtained; and obtaining the noise factor according to a calculation formula of the noise factor.
It should be noted that, considering that the fluctuation habit of the operation monitoring parameter may change with time, and meanwhile, the fluctuation feature of the monitoring parameter may be smoothed out by a larger time span, the preset time period aimed at in one embodiment of the present invention is a time of pushing back forward for 1 week at each sampling time, and the fluctuation sensitivity factor of the operation monitoring parameter of each dimension in each sampling time period is obtained by analyzing the fluctuation situation of the multi-dimensional monitoring data collected in the preset time period, and the abnormality detection is performed on the operation monitoring parameter in the preset time period in combination with the corresponding fluctuation feature; in the process of determining the preset time period, a part of operation monitoring parameters may not be completely pushed back for one week, and then the operation monitoring parameters in the preset time period corresponding to the missing part of the operation monitoring parameters need to be fitted through a difference method to obtain a complete preset time period for fluctuation analysis, wherein the difference method is a prior art well known to those skilled in the art and is not described herein again; the practitioner may also obtain by other fitting means.
The noise factor is calculated by the following formula:
Wherein, For/>Fifth/th at each sampling instantA noise factor of the dimensional operation monitoring parameter; /(I)For/>Fifth/th at each sampling instantA fluctuation sensitive factor of the operation monitoring parameter of the dimension; /(I)For/>Except for the/>, at the sample time instantFirst/>, in all dimensions outside the dimensionA fluctuation sensitive factor of the operation monitoring parameter of the dimension; /(I)To be except for the (th)Running the dimension serial numbers of the monitoring parameters in all dimensions outside the dimension; /(I)The total dimension number of the monitoring parameters is used for running; /(I)For/>Fifth/th at each sampling instantAbnormal fluctuation coefficients of the operation monitoring parameters of the dimension; /(I)For/>Fifth/th at each sampling instantAbnormal fluctuation coefficients of the operation monitoring parameters of the dimension; /(I)Is a standard normalization function; /(I)To preset the second positive parameter, in the embodiment of the present invention,/>Taking 0.1, ensuring that the denominator is not zero.
In the calculation formula of the noise factor, the fluctuation sensitivity factor reflects the fluctuation tendency and habit of the operation monitoring parameter under the corresponding dimension, and the larger the fluctuation sensitivity factor is, the more the corresponding operation monitoring parameter tends to fluctuate, so the first time under the same sampling time is calculatedAbnormal fluctuation coefficients of dimension operation monitoring parameters and others such as the/>If the difference between abnormal fluctuation coefficients of the dimension operation monitoring parameters is the/>If the dimension operation monitoring parameters tend to fluctuate, the difference between the dimension operation monitoring parameters is weakened, and the fluctuation sensitivity factors corresponding to the operation monitoring parameters in other dimensions are subtracted by 1 to serve as weights of abnormal fluctuation coefficients corresponding to the operation monitoring parameters so as to weaken the difference; if/>Dimensional operation monitoring parameters tend to fluctuate by mapping the corresponding fluctuation sensitivity factors thereof into inverse values through negative correlation and taking the inverse values as the finally acquired/>The weight of the weighted difference between the abnormal fluctuation coefficient of the dimension operation monitoring parameter and the abnormal fluctuation coefficient of all other dimension operation monitoring parameters is used for weakening the influence caused by the fluctuation habit of the dimension operation monitoring parameter.
Preferably, in one embodiment of the present invention, considering that the standard deviation of the parameter in the preset neighborhood reflects the fluctuation condition of the data, but judging that the fluctuation condition of the parameter is susceptible to noise only according to the standard deviation, frequent fluctuation of the noise also causes the standard deviation of the data points in the preset neighborhood to be larger, so that the judgment is further performed by combining with the extreme point; meanwhile, the fluctuation condition of the data points is reflected by the difference of the amplitude mean value relative to the middle level in the neighborhood; based on the fluctuation characteristics of the operation monitoring parameters of each dimension in a preset time period, the fluctuation sensitivity factors of the operation monitoring parameters of each dimension can be obtained; the calculation formula of the fluctuation sensitive factor comprises:
Wherein, For/>Fifth/th at each sampling instantA fluctuation sensitive factor of the operation monitoring parameter of the dimension; /(I)The sequence number of the operation monitoring parameter in the preset time period; /(I)For/>Fifth/th at each sampling instantThe total number of the operation monitoring parameters in the preset time period of the operation monitoring parameters of the dimension; /(I)For/>Fifth/th at each sampling instantThe first/>, within a preset period of time, of the operational monitoring parameters of the dimensionAmplitude standard deviations of all operation monitoring parameters in a preset neighborhood of each operation monitoring parameter; /(I)For/>Fifth/th at each sampling instantThe first/>, within a preset period of time, of the operational monitoring parameters of the dimensionThe average value of the amplitude values of all the operation monitoring parameters in the preset neighborhood of each operation monitoring parameter; /(I)For/>Fifth/th at each sampling instantThe first/>, within a preset period of time, of the operational monitoring parameters of the dimensionThe number of extreme points in the preset neighborhood of each operation monitoring parameter; /(I)Is a normalization function; /(I)Is a natural constant; /(I)Is an intermediate parameter; /(I)To positively adjust the parameter. In the embodiment of the invention, the value range after the normalization of the amplitude mean value in the preset neighborhood is 0 to 1, so that the intermediate parameter/>Set to 0.5 as a reference while adjusting the non-zero adjustment parameter/>Setting to 2 amplifies the difference between the normalized value of the amplitude mean and the intermediate parameter, and in other embodiments of the present invention, other values may be set according to specific implementation situations. It should be noted that, the standard deviation, the mean value and the obtaining manner of the extreme point are means well known to those skilled in the art, and are not described herein.
In a calculation formula of the fluctuation sensitive factor, the difference between the average value of the amplitude of the operation monitoring parameter of each dimension and the intermediate parameter at each sampling time reflects the amplitude fluctuation condition of the data relative to the average level; the amplitude standard deviation of the parameter in the preset neighborhood reflects the fluctuation condition of the data point, and the larger the standard deviation is, the more frequent or severe the parameter fluctuation is; the more the number of extreme points reflects the fluctuation condition that the fluctuation degree of the parameter in the preset neighborhood is more frequent but the parameter tends to noise, and the parameter is mapped into an exponential function for normalization in a negative correlation mode, so that the interference of noise on a fluctuation sensitive factor is reduced; the fluctuation situation of each operation monitoring parameter can be estimated by combining the three, and the fluctuation sensitive factor of the operation monitoring parameter under each dimension can be obtained by integrating the fluctuation situations of all the operation parameters in a preset time period.
And step S3, adjusting preset K neighborhood parameters of the operation monitoring parameters of each dimension at each sampling time according to the noise factors, and performing anomaly detection on the operation monitoring parameters according to the adjusted preset K neighborhood parameters.
The noise factor of each operation monitoring parameter at each sampling time reflects the possibility of abnormal fluctuation of the operation monitoring parameter and also reflects the possibility of abnormal fluctuation caused by noise; the larger the noise factor, the greater the likelihood that the corresponding operation monitoring parameter is noise; in order to avoid that abnormal fluctuation caused by noise is mistakenly used as abnormal fluctuation caused by abnormal operation of the centrifugal fan, preset K neighborhood parameters of each operation monitoring data can be adjusted according to noise factors, in one embodiment of the invention, the LOF algorithm is specifically adopted for carrying out abnormal detection on the operation monitoring parameters, and the K neighborhood parameters of the LOF algorithm corresponding to the operation monitoring parameters with large noise factors are further reduced, so that the false alarm probability is reduced.
Preferably, in one embodiment of the present invention, the method for adjusting the preset K neighborhood parameter includes: and adding a preset positive constant to the noise factor as an adjustment coefficient, multiplying the preset K neighborhood parameter by the adjustment coefficient, and rounding to obtain an adjusted K neighborhood parameter. The adjustment formula of the preset K neighborhood parameters is as follows:
Wherein, For the adjusted K neighborhood parameters,/>Presetting K neighborhood parameters; /(I)Is a preset positive constant; /(I)For/>Fifth/th at each sampling instantA noise factor of the dimensional operation monitoring parameter; /(I)As a round-up function.
It should be noted that, in the embodiment of the present invention, the K neighborhood parameters are preset10, The implementer can set up by himself according to specific implementation conditions; considering that the noise factor has a value ranging from 0 to 1, a positive constant/>, will be presetTaking the intermediate value of 0.5 as a comparison reference, when the noise factor is greater than 0.5, considering that the probability of being noise is greater, the corresponding adjustment coefficient is greater than 1, and multiplying the preset K neighborhood parameter by the adjustment coefficient to increase so as to reduce the sensitivity of the LOF algorithm to the noise; when the noise factor is smaller than 0.5, the greater the possibility that the noise factor is the abnormal operation fluctuation or the normal operation monitoring parameter is, the corresponding adjustment coefficient is smaller than 1, and the preset K neighborhood parameter is multiplied by the adjustment coefficient to reduce so as to accurately detect the abnormal operation parameter.
The operator can use the adjusted preset K neighborhood parameters to carry out abnormal detection on the operation monitoring parameters of all dimensions at all sampling moments, obtain accurate abnormal operation monitoring data and carry out early warning while reducing noise influence, prompt related management personnel to evaluate or overhaul the operation state of the centrifugal fan according to the abnormal operation monitoring data, reduce false alarm rate and improve operation monitoring accuracy and efficiency.
In summary, the method includes the steps that firstly, multi-dimensional operation monitoring parameters of the centrifugal fan at each sampling moment and all fitting operation curves of corresponding dimensions are obtained, and then abnormal fluctuation coefficients of the operation monitoring parameters of the corresponding dimensions at the corresponding sampling moment are obtained according to local fluctuation characteristics of the operation monitoring parameters of the corresponding dimensions at each sampling moment and local fluctuation characteristics of the corresponding fitting operation curves; under each sampling time, acquiring noise factors of the operation monitoring parameters of the corresponding dimension under each sampling time according to differences between the operation monitoring parameters of each dimension and abnormal fluctuation coefficients of the operation monitoring parameters of all the other dimensions and fluctuation habit characteristics of the operation monitoring parameters of the corresponding dimension; and adjusting preset K neighborhood parameters of each operation monitoring parameter according to the noise factors, acquiring abnormal operation monitoring parameters and early warning. According to the invention, by combining fluctuation characteristics caused by noise and fluctuation characteristics difference caused by abnormal operation of the centrifugal fan, fluctuation characteristics of operation monitoring parameters are analyzed to obtain noise factors of each operation monitoring data, so that sensitivity of an LOF algorithm to the operation monitoring parameters with high noise possibility is adjusted, K neighborhood parameters are adjusted, false alarm rate is reduced, and operation monitoring accuracy and efficiency are improved.
The invention also provides an operation monitoring system for the centrifugal fan, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes any one of the steps of the operation monitoring method for the centrifugal fan when executing the computer program.
An embodiment of a noise reduction method for operation monitoring parameters of a centrifugal fan comprises the following steps:
Because monitoring data of the centrifugal fan is possibly interfered in the process of being collected by the sensor, a large amount of noise data exists in the monitoring data, the noise data is similar to abnormal fluctuation of abnormal operation data, the abnormal operation monitoring parameters are easily removed synchronously by the traditional data noise reduction method, and further the real-time operation monitoring of the centrifugal fan is not complete enough, so that the abnormal operation behavior of the centrifugal fan cannot be further evaluated and corresponding improvement strategies cannot be formulated. The invention provides a noise reduction method for operation monitoring parameters of a centrifugal fan, which comprises the following steps:
Step S1, acquiring multi-dimensional operation monitoring parameters of the centrifugal fan at each sampling time, and acquiring a fitting operation curve of the operation monitoring parameters of each dimension.
Step S2, according to the local fluctuation characteristics of the operation monitoring parameters of each dimension at each sampling moment and the local fluctuation characteristics of the corresponding fitting operation curve, obtaining the abnormal fluctuation coefficients of the operation monitoring parameters of the corresponding dimension at the corresponding sampling moment; and under each sampling time, acquiring the noise factor of the operation monitoring parameter of the corresponding dimension under each sampling time according to the difference between the abnormal fluctuation coefficients of the operation monitoring parameter of each dimension and the operation monitoring parameters of all the other dimensions and the fluctuation habit characteristics of the operation monitoring parameter of the corresponding dimension.
And S3, eliminating noise in all operation monitoring parameters according to the noise factor.
And taking the operation monitoring parameter with the noise factor larger than the preset threshold value as noise elimination, wherein in one embodiment of the invention, the preset threshold value is set to be 0.5, and the implementer can set other values according to specific implementation conditions.
The step S1 and the step S2 are already described in detail in the above embodiment of the operation monitoring method for a centrifugal fan, and are not described in detail.
A noise reduction system embodiment for operation monitoring parameters of a centrifugal fan:
The invention also provides a noise reduction system for the operation monitoring parameters of the centrifugal fan, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes any one of the steps of the noise reduction method for the operation monitoring parameters of the centrifugal fan when executing the computer program.
Firstly, acquiring multi-dimensional operation monitoring parameters of a centrifugal fan at each sampling moment and all fitting operation curves of corresponding dimensions, and further acquiring abnormal fluctuation coefficients of the operation monitoring parameters of the corresponding dimensions at the corresponding sampling moment according to local fluctuation characteristics of the operation monitoring parameters of the corresponding dimensions at each sampling moment and local fluctuation characteristics of the corresponding fitting operation curves; under each sampling time, taking the characteristic that noise causes low possibility of abnormal fluctuation of all the dimension operation monitoring parameters into consideration, so that according to the difference between the abnormal fluctuation coefficients of the operation monitoring parameters of each dimension and the operation monitoring parameters of the other all the dimensions and the fluctuation habit characteristics of the operation monitoring parameters of the corresponding dimension, the noise factors of the operation monitoring parameters of the corresponding dimension under each sampling time are obtained, and the more the operation monitoring parameters tend to fluctuate, the influence of the operation monitoring parameters is weakened when the abnormal fluctuation is analyzed so as to accurately obtain the noise factors of each operation monitoring parameter; and then eliminating noise in all operation monitoring parameters according to the noise factor. According to the invention, by combining fluctuation characteristics caused by noise and fluctuation characteristics difference caused by abnormal operation of the centrifugal fan, fluctuation characteristics of operation monitoring parameters are analyzed to obtain noise factors of each operation monitoring data, so that the operation monitoring parameters with high noise possibility are removed, noise is accurately removed while abnormal operation behaviors of the centrifugal fan are maintained, and accurate and complete operation monitoring data are provided for subsequent evaluation of operation conditions of the centrifugal fan.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (8)
1. An operation monitoring method for a centrifugal fan, the method comprising:
Acquiring operation monitoring parameters of at least two dimensions of the centrifugal fan at each sampling moment, and acquiring a fitting operation curve of the operation monitoring parameters of each dimension;
Obtaining abnormal fluctuation coefficients of the operation monitoring parameters of the corresponding dimensions at the corresponding sampling moments according to the local fluctuation characteristics of the operation monitoring parameters of the dimensions at each sampling moment and the local fluctuation characteristics of the corresponding fitting operation curves; under each sampling time, acquiring a noise factor of the operation monitoring parameter of the corresponding dimension under each sampling time according to the difference of the abnormal fluctuation coefficient of the operation monitoring parameter of each dimension and the operation monitoring parameters of all the other dimensions and the fluctuation habit characteristics of the operation monitoring parameter of the corresponding dimension;
adjusting a preset K neighborhood parameter of the operation monitoring parameter of each dimension at each sampling time according to the noise factors, and performing anomaly detection on the operation monitoring parameter according to the adjusted preset K neighborhood parameter;
The method for acquiring the abnormal fluctuation coefficient comprises the following steps:
Acquiring a local trend reference coefficient of an operation monitoring parameter of each dimension at each sampling time; acquiring fitting deviation between the operation monitoring parameters of each dimension and the fitting operation parameters of the corresponding sampling time on the fitting operation curve; according to the operation monitoring parameters and the local fluctuation characteristic difference of the fitting deviation, the abnormal fluctuation coefficient is obtained by combining the local trend reference coefficient; the calculation formula of the abnormal fluctuation coefficient is as follows:
; wherein/> For/>Fifth/th at each sampling instantAbnormal fluctuation coefficients of the operation monitoring parameters of the dimension; /(I)For/>Fifth/th at each sampling instantStandard deviation of all operation monitoring parameters in a preset neighborhood of the operation monitoring parameters of the dimension; /(I)For/>Fifth/th at each sampling instantThe standard deviation of fitting operation parameters under the sampling time corresponding to the fitting curve and all operation monitoring parameters in the preset neighborhood of the dimension operation monitoring parameters; /(I)For/>Fifth/th at each sampling instantLocal trend reference coefficients of the dimensional operation monitoring parameters; /(I)Is a normalization function; /(I)Presetting a first positive parameter;
the method for acquiring the noise factor comprises the following steps:
Acquiring fluctuation sensitivity factors of the operation monitoring parameters of each dimension according to fluctuation characteristics of the operation monitoring parameters of each dimension in a preset time period; acquiring a noise factor according to a calculation formula of the noise factor; the noise factor is calculated by the following formula:
; wherein/> For/>Fifth/th at each sampling instantA noise factor of the dimensional operation monitoring parameter; /(I)For/>Fifth/th at each sampling instantA fluctuation sensitive factor of the operation monitoring parameter of the dimension; /(I)For/>Except for the/>, at the sample time instantFirst/>, in all dimensions outside the dimensionA fluctuation sensitive factor of the operation monitoring parameter of the dimension; /(I)To be except for the (th)Running the dimension serial numbers of the monitoring parameters in all dimensions outside the dimension; /(I)The total dimension number of the monitoring parameters is used for running; /(I)For/>Fifth/th at each sampling instantAbnormal fluctuation coefficients of the operation monitoring parameters of the dimension; /(I)Is the firstFifth/th at each sampling instantAbnormal fluctuation coefficients of the operation monitoring parameters of the dimension; /(I)Is a standard normalization function; /(I)Is a preset second positive parameter.
2. The method for monitoring the operation of a centrifugal fan according to claim 1, wherein the calculation formula of the local trend reference coefficient is:
; wherein/> For/>Fifth/th at each sampling instantLocal trend reference coefficients of the dimensional operation monitoring parameters; /(I)For/>Fifth/th at each sampling instantThe number of extreme points on the fitting curve in the preset neighborhood of the operation monitoring parameters of the dimension; /(I)For/>Fifth/th at each sampling instantFitting the extremely poor fitting operation parameters on the fitting curves in the preset neighborhood of the dimensional operation monitoring parameters; /(I)Is a normalization function; /(I)Is a natural constant.
3. The operation monitoring method for a centrifugal fan according to claim 1, wherein the calculation formula of the surge sensitivity factor includes:
; wherein/> For/>Fifth/th at each sampling instantA fluctuation sensitive factor of the operation monitoring parameter of the dimension; /(I)The sequence number of the operation monitoring parameter in the preset time period; For/> Fifth/th at each sampling instantThe total number of the operation monitoring parameters in the preset time period of the operation monitoring parameters of the dimension; For/> Fifth/th at each sampling instantThe first/>, within a preset period of time, of the operational monitoring parameters of the dimensionAmplitude standard deviations of all operation monitoring parameters in a preset neighborhood of each operation monitoring parameter; /(I)For/>Fifth/th at each sampling instantThe first/>, within a preset period of time, of the operational monitoring parameters of the dimensionThe average value of the amplitude values of all the operation monitoring parameters in the preset neighborhood of each operation monitoring parameter; For/> Fifth/th at each sampling instantThe first/>, within a preset period of time, of the operational monitoring parameters of the dimensionThe number of extreme points in the preset neighborhood of each operation monitoring parameter; /(I)Is a normalization function; /(I)Is a natural constant; /(I)Is an intermediate parameter; /(I)To positively adjust the parameter.
4. The method for monitoring the operation of a centrifugal fan according to claim 1, wherein the method for adjusting the preset K neighborhood parameter comprises:
And adding a preset positive constant to the noise factor as an adjustment coefficient, multiplying the preset K neighborhood parameter by the adjustment coefficient, and rounding to obtain an adjusted K neighborhood parameter.
5. An operation monitoring method for a centrifugal fan according to claim 1, wherein the dimensions of the operation monitoring parameters include at least gas pressure, fan speed, gas flow and gas temperature.
6. The method for monitoring the operation of a centrifugal fan according to claim 1, wherein the method for detecting the abnormality of the operation monitoring parameter is an LOF algorithm.
7. A method for monitoring the operation of a centrifugal fan according to claim 1,2 or 3, wherein the method for obtaining the preset neighborhood comprises:
And taking the operation monitoring parameters of each dimension at each sampling time as a center, and respectively acquiring a preset number of operation monitoring parameters along two directions of the acquisition time sequence to construct a preset neighborhood.
8. An operation monitoring system for a centrifugal fan, characterized in that the system comprises a memory, a processor and a computer program stored in the memory and operable on the processor, the processor executing the computer program to carry out the steps of a method for operation monitoring of a centrifugal fan according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410160641.7A CN117708748B (en) | 2024-02-05 | 2024-02-05 | Operation monitoring system and method for centrifugal fan |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410160641.7A CN117708748B (en) | 2024-02-05 | 2024-02-05 | Operation monitoring system and method for centrifugal fan |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117708748A CN117708748A (en) | 2024-03-15 |
| CN117708748B true CN117708748B (en) | 2024-05-03 |
Family
ID=90161052
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410160641.7A Active CN117708748B (en) | 2024-02-05 | 2024-02-05 | Operation monitoring system and method for centrifugal fan |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117708748B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117909692A (en) * | 2024-03-18 | 2024-04-19 | 山东海纳智能装备科技股份有限公司 | Intelligent analysis method for operation data of double-disc motor bearing |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110865260A (en) * | 2019-11-29 | 2020-03-06 | 南京信息工程大学 | Method for monitoring and evaluating MOV actual state based on outlier detection |
| CN116292367A (en) * | 2023-03-22 | 2023-06-23 | 山东科技大学 | Power plant fan system abnormal condition detection method based on one-dimensional convolution |
| CN116756526A (en) * | 2023-08-17 | 2023-09-15 | 北京英沣特能源技术有限公司 | Full life cycle performance detection and analysis system of energy storage equipment |
| CN117093879A (en) * | 2023-10-19 | 2023-11-21 | 无锡尚航数据有限公司 | Intelligent operation management method and system for data center |
| CN117390562A (en) * | 2023-10-18 | 2024-01-12 | 国网山东省电力公司营销服务中心(计量中心) | Data transmission management method and system for distributed power generation system |
| CN117436353A (en) * | 2023-12-21 | 2024-01-23 | 泰安维创游乐设备有限公司 | Intelligent recreation facility fault prediction method based on big data |
-
2024
- 2024-02-05 CN CN202410160641.7A patent/CN117708748B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110865260A (en) * | 2019-11-29 | 2020-03-06 | 南京信息工程大学 | Method for monitoring and evaluating MOV actual state based on outlier detection |
| CN116292367A (en) * | 2023-03-22 | 2023-06-23 | 山东科技大学 | Power plant fan system abnormal condition detection method based on one-dimensional convolution |
| CN116756526A (en) * | 2023-08-17 | 2023-09-15 | 北京英沣特能源技术有限公司 | Full life cycle performance detection and analysis system of energy storage equipment |
| CN117390562A (en) * | 2023-10-18 | 2024-01-12 | 国网山东省电力公司营销服务中心(计量中心) | Data transmission management method and system for distributed power generation system |
| CN117093879A (en) * | 2023-10-19 | 2023-11-21 | 无锡尚航数据有限公司 | Intelligent operation management method and system for data center |
| CN117436353A (en) * | 2023-12-21 | 2024-01-23 | 泰安维创游乐设备有限公司 | Intelligent recreation facility fault prediction method based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117708748A (en) | 2024-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117708748B (en) | Operation monitoring system and method for centrifugal fan | |
| CA2634328C (en) | Method and system for trend detection and analysis | |
| CN112348237B (en) | Abnormal trend detection method for dynamic drilling data | |
| CN112284440B (en) | Sensor data deviation self-adaptive correction method | |
| CN110286656B (en) | False alarm filtering method and device for tolerance of error data | |
| CN110648480B (en) | Single variable alarm system and method based on change rate | |
| CN112629585A (en) | Equipment on-line monitoring method and device based on multi-dimensional parameter estimation | |
| US20100082295A1 (en) | Method and System of Wind Turbine Condition Monitoring | |
| CN117407700B (en) | Method for monitoring working environment in live working process | |
| WO2012128913A2 (en) | Degrading sensor detection implemented within a transmitter | |
| CN117196353B (en) | Environmental pollution assessment and monitoring method and system based on big data | |
| US20190163163A1 (en) | Numerical controller | |
| CN116873156B (en) | Intelligent monitoring method for power abnormality of natural gas ship based on big data | |
| CN112711850A (en) | Unit online monitoring method based on big data | |
| CN115220396A (en) | Intelligent monitoring method and system for numerical control machine tool | |
| JP2013104666A (en) | Stick-slip detector and detection method | |
| CN114738132A (en) | Supercharger speed sensor fault diagnosis method and system based on real ship data | |
| JPH08241121A (en) | Print abnormality detecting device | |
| CN116772285B (en) | Intelligent building heating load safety real-time monitoring method | |
| Wu et al. | Lightweight mixture faults detection method for gasoline engine using on-line trend analysis | |
| TWI745912B (en) | Blast furnace abnormality determination device, blast furnace abnormality determination method and blast furnace operation method | |
| JP2022084435A5 (en) | ||
| CN117804112B (en) | Cold and heat source system Ai energy efficiency management system | |
| CN117668470B (en) | Intelligent detection method and system for air inlet and outlet faults of engine | |
| CN116341993B (en) | State monitoring method and system for polystyrene production process |
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 |