CN116877452B - Non-positive-displacement water pump running state monitoring system based on Internet of things data - Google Patents
Non-positive-displacement water pump running state monitoring system based on Internet of things data Download PDFInfo
- Publication number
- CN116877452B CN116877452B CN202311147254.1A CN202311147254A CN116877452B CN 116877452 B CN116877452 B CN 116877452B CN 202311147254 A CN202311147254 A CN 202311147254A CN 116877452 B CN116877452 B CN 116877452B
- Authority
- CN
- China
- Prior art keywords
- time window
- time
- frequency
- sequence
- acquiring
- 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
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 80
- 238000012544 monitoring process Methods 0.000 title claims abstract description 38
- 238000006073 displacement reaction Methods 0.000 title claims abstract description 24
- 230000002159 abnormal effect Effects 0.000 claims abstract description 60
- 238000001228 spectrum Methods 0.000 claims abstract description 54
- 238000000034 method Methods 0.000 claims abstract description 51
- 230000005856 abnormality Effects 0.000 claims abstract description 50
- 238000009826 distribution Methods 0.000 claims abstract description 35
- 238000004364 calculation method Methods 0.000 claims description 25
- 239000006185 dispersion Substances 0.000 claims description 20
- 230000008569 process Effects 0.000 claims description 15
- 230000003595 spectral effect Effects 0.000 claims description 12
- 238000004458 analytical method Methods 0.000 claims description 11
- 238000001914 filtration Methods 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 5
- 238000009825 accumulation Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 230000001174 ascending effect Effects 0.000 claims description 2
- 238000013507 mapping Methods 0.000 claims description 2
- 230000009466 transformation Effects 0.000 claims description 2
- 238000012545 processing Methods 0.000 abstract description 5
- 238000001514 detection method Methods 0.000 description 7
- 230000002123 temporal effect Effects 0.000 description 4
- 230000000750 progressive effect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000005381 potential energy Methods 0.000 description 2
- 238000005299 abrasion Methods 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- VEMKTZHHVJILDY-UHFFFAOYSA-N resmethrin Chemical compound CC1(C)C(C=C(C)C)C1C(=O)OCC1=COC(CC=2C=CC=CC=2)=C1 VEMKTZHHVJILDY-UHFFFAOYSA-N 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
- F04D—NON-POSITIVE-DISPLACEMENT PUMPS
- F04D15/00—Control, e.g. regulation, of pumps, pumping installations or systems
- F04D15/0088—Testing machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2474—Sequence data queries, e.g. querying versioned data
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Probability & Statistics with Applications (AREA)
- General Engineering & Computer Science (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Mechanical Engineering (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention relates to the technical field of digital data processing, and provides a non-positive-displacement water pump running state monitoring system based on internet of things data, which comprises the following components: acquiring time sequence data of monitoring parameters; acquiring time domain jump heterogeneity coefficients of each time window according to fluctuation conditions and probability distribution conditions of data in each time window, and acquiring time-frequency energy anomaly coefficients of each time window according to the time domain jump heterogeneity coefficients of each time window and a frequency domain data distribution rule; acquiring a frequency spectrum energy attenuation distance sequence of each time window according to the frequency component energy attenuation rule in each time window, and acquiring a time-frequency hopping abnormality index and a joint time-frequency hopping abnormality index of each time window according to the frequency spectrum energy attenuation distance sequence of each time window; and identifying the abnormal condition of the water pump according to the combined time-frequency hopping abnormal index. The method effectively solves the problem that the slight abnormality of the water pump is difficult to judge, and improves the accuracy of detecting the abnormality of the water pump.
Description
Technical Field
The invention relates to the technical field of digital data processing, in particular to a non-positive-displacement water pump running state monitoring system based on Internet of things data.
Background
The water pump is used as a common engineering machine and has very wide application in the fields of municipal water supply, liquid transportation and the like. The non-positive displacement water pump is a water pump with fixed rotation speed, the flow and the lift of the water pump can not be adjusted along with the change of working conditions, and the water pump is driven by a motor with constant rotation speed, and mechanical energy is converted into potential energy of water through a mechanical device. Because the non-positive displacement water pump has the advantages of simple structure, strong adaptability and low cost, the market share is always kept high. However, when the water pump is abnormal due to environmental factors in operation, the problems of water supply interruption, equipment damage, energy waste and the like can be caused, and the daily life and industrial production of people can be seriously influenced, so that the abnormal detection of the non-positive-displacement water pump is very necessary. Abnormality detection may find a water pump problem in advance, and take corresponding measures to ensure normal operation of the water pump.
The traditional detection for the non-positive displacement water pump mainly comprises the following steps: the method based on the threshold value is used for judging by manually setting the threshold value and detecting the thermal signal of the water pump, but the method cannot accurately detect the abnormality of the water pump because the working environments are inconsistent and the thresholds are different. There are also statistical-based and manual-based methods that can detect the state of the water pump in a certain way, but cannot accurately judge slight abnormalities and prevent them in advance. The water pump abnormality detection based on the Internet of things mainly monitors the state of the water pump through the sensor, uploads the water pump abnormality detection to the Internet of things platform in real time, and can accurately judge the abnormality of the water pump through analyzing data, and timely take corresponding measures to reduce the influence on life of people.
Disclosure of Invention
The invention provides a non-positive-displacement water pump running state monitoring system based on internet of things data, which aims to solve the problem that slight abnormality of a water pump cannot be accurately judged, and adopts the following technical scheme:
the invention relates to a non-positive-displacement water pump running state monitoring system based on internet of things data, which comprises the following modules:
the data acquisition module acquires time sequence data of the monitoring parameters;
the anomaly analysis module is used for acquiring data in each time window with preset size according to the complexity of the time sequence data and acquiring the time domain distribution dispersion of each time window according to the fluctuation condition of the data in each time window; acquiring a time domain jump heterogeneity coefficient of each time window according to the time domain distribution dispersion of each time window; acquiring a time-frequency energy anomaly coefficient of each time window according to the time-domain jump heterogeneity coefficient of each time window and the frequency domain data distribution rule;
the abnormality extraction module is used for obtaining a spectrum energy attenuation distance sequence of each time window according to the energy attenuation rule of the frequency component in each time window; acquiring a spectrum energy difference sequence of each time window according to the spectrum energy attenuation distance sequence of each time window, and acquiring a time-frequency hopping abnormal index of each time window according to the spectrum energy difference sequence and the time-frequency energy abnormal coefficient of each time window; acquiring a combined time-frequency hopping abnormality index of each time window according to the time-frequency hopping abnormality index of each time window;
and the abnormality monitoring module is used for acquiring a state monitoring result in the running process of the water pump according to the combined time-frequency hopping abnormality index of each time window.
Preferably, the method for obtaining the time domain distribution dispersion of each time window according to the fluctuation condition of the data in each time window includes:
setting a time window with a preset size for any time sequence of monitoring parameters, and acquiring data in each time window in the time sequence;
for any time window, taking the accumulated sum of absolute values of differences between data average values of all data in the time window and each data in the time window as a first product factor;
taking the product of the first product factor and the variation coefficient of each time window as the time domain distribution dispersion of each time window.
Preferably, the method for obtaining the time domain hopping heterogeneity coefficient of each time window according to the time domain distribution dispersion of each time window comprises the following steps:
for any time window, counting the probability corresponding to each unequal data in the time window, acquiring the accumulated sum of the product between the probability corresponding to each unequal data and the logarithmic value of the corresponding probability in the time window, and taking the negative mapping result of the accumulated sum as a second product factor;
and taking a normalized value of the product of the second product factor of each time window and the time domain distribution dispersion as a time domain jump heterogeneity coefficient of each time window.
Preferably, the method for obtaining the time-frequency energy anomaly coefficient of each time window according to the time-domain jump heterogeneity coefficient and the frequency domain data distribution rule of each time window comprises the following steps:
for any time window, obtaining unilateral spectrum frequency domain data in the time window by utilizing Fourier transformation, taking frequency components with frequency domain energy higher than a preset screening threshold value of the frequency components as calculation objects, and taking output obtained by filtering the calculation objects by utilizing a Chebyshev band-pass filter as filtering components;
taking the frequency domain energy accumulation sum of the fundamental frequency component and each frequency component in the calculation object as a denominator, taking the frequency domain energy accumulation sum of each frequency component in the filtering component as a numerator, and taking the ratio of the numerator to the denominator as a third multiplication factor of a time window;
and taking the product of the third multiplication factor of each time window and the time domain jump heterogeneity coefficient as the time-frequency energy anomaly coefficient of each time window.
Preferably, the method for obtaining the spectrum energy attenuation distance sequence of each time window according to the energy attenuation rule of the frequency component in each time window comprises the following steps:
for any calculation object of the time window, acquiring a spectrum energy attenuation characteristic value of each frequency component in the calculation object of the time window according to the change of the frequency component energy of the calculation object of the time window;
and for any calculation object of the time window, taking the sequence of the spectral energy attenuation characteristic values of all frequency components in the calculation object as a sequence of the spectral energy attenuation distances of the time window according to the sequence of the frequency components from the base frequency component from near to far.
Preferably, the method for obtaining the spectral energy attenuation characteristic value of each frequency component in the calculation object of the time window according to the change of the frequency component energy of the calculation object of the time window comprises the following steps:
in the method, in the process of the invention,a spectral energy attenuation characteristic value representing an ith frequency component within the calculation object of the time window,i-th frequency component in the calculation object representing the time window,>fundamental frequency component in the computation object representing the time window, < >>Frequency domain energy of the ith frequency component in the computation object representing the time window, +.>The frequency domain energy of the fundamental frequency component within the computation object representing the time window.
Preferably, the method for obtaining the frequency spectrum energy difference sequence of each time window according to the frequency spectrum energy attenuation distance sequence of each time window and obtaining the time-frequency hopping anomaly index of each time window according to the frequency spectrum energy difference sequence and the time-frequency energy anomaly coefficient of each time window comprises the following steps:
for the spectrum energy attenuation distance sequence of any time window, taking the difference value of the last element and the previous element in the spectrum energy attenuation distance sequence as an adjacent difference value, and taking a sequence formed by all adjacent difference values according to time ascending sequence as a spectrum energy difference sequence of the time window;
acquiring a fourth product factor of each time window according to the spectrum energy difference sequence of each time window;
and taking the product of the fourth product factor of each time window and the time-frequency energy anomaly coefficient as the time-frequency hopping anomaly index of each time window.
Preferably, the method for obtaining the fourth product factor of each time window according to the spectrum energy difference sequence of each time window is as follows:
and for the spectrum energy differential sequence of any time window, obtaining the differential ratio of the last differential value to the previous differential value in the spectrum energy differential sequence, and taking the accumulated sum of the differential ratio on the spectrum energy differential sequence as a fourth product factor of the time window.
Preferably, the method for obtaining the combined time-frequency hopping abnormality index of each time window according to the time-frequency hopping abnormality index of each time window comprises the following steps:
and regarding the time-frequency hopping abnormal indexes of each time window of each monitoring parameter, taking the accumulated normalization result of the time-frequency hopping abnormal indexes of each corresponding time window of all the monitoring parameters as the combined time-frequency hopping abnormal index of each time window.
Preferably, the method for acquiring the state monitoring result in the running process of the water pump according to the combined time-frequency hopping abnormal index of each time window comprises the following steps:
and for the combined time-frequency hopping abnormal index of each time window, acquiring the abnormal degree of the water pump in the running process according to the comparison result of the combined time-frequency hopping abnormal index and the preset length interval, and monitoring the abnormal faults occurring in the running process of the water pump.
The beneficial effects of the invention are as follows: according to the invention, the noise signal is analyzed, so that the time domain distribution asymmetry coefficient of the noise is obtained in the time domain, and the time domain jump heterogeneity coefficient is obtained on the basis. Combining the time domain result on the frequency domain converted by the signal to obtain a time-frequency energy abnormal coefficient, constructing a frequency spectrum energy attenuation distance sequence, obtaining a time-frequency hopping abnormal index of a noise signal, obtaining the time-frequency hopping abnormal index of a vibration signal by the same method, and finally obtaining the combined time-frequency hopping abnormal index of the water pump. Compared with the traditional detection method, the accuracy of diagnosing the abnormal state of the water pump is improved by comprehensively analyzing the noise signal and the vibration signal of the non-variable-capacity water pump. The method effectively solves the problem that the slight abnormality of the water pump is difficult to judge, and improves the accuracy of detecting the abnormality of the water pump.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
Fig. 1 is a schematic flow chart of a non-positive displacement pump running state monitoring system based on internet of things data according to an embodiment of the present invention;
fig. 2 is a flowchart of an implementation of online monitoring of abnormal operation of a non-positive displacement pump according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a flowchart of a non-positive displacement pump running state monitoring system based on internet of things data according to an embodiment of the present invention is shown, where the system includes a data acquisition module, an anomaly analysis module, an anomaly extraction module, and an anomaly monitoring module.
The invention relates to a data acquisition module, which takes a non-variable-capacity water pump as an abnormal analysis object, and utilizes a data sensor to acquire data of monitoring data of the water pump during working in order to evaluate the working state of the water pump, wherein the monitoring data comprises a noise signal and a vibration signal of the water pump, and the data sensor comprises a sound sensor and a vibration sensor.
According to the scheme, the Internet of things platform is used for analyzing the monitoring data, the Internet of things technology is used for realizing the connection of equipment and equipment, equipment and a cloud control system through various data sensors, realizing the real-time monitoring of the equipment through the interconnection technology, analyzing according to the data and making corresponding decisions. According to the scheme, the Thingworx 8 IoT platform is adopted to realize abnormality detection of the non-positive-displacement water pump, data are collected through the sensor and uploaded to the Thingworx 8 IoT platform, data analysis is carried out, whether the water pump is abnormal or not is judged, a final result is sent back to the control end of the water pump, the state of the water pump can be adjusted in time, and loss is reduced.
When the non-positive-displacement water pump works, a sound sensor is arranged on the shell of the driving end of the water pump and is connected with the shell of the driving end into a whole in a patch mode, so that noise signals emitted by the water pump shaft end during operation are collected, and external noise is shielded to a certain extent. In addition, because the transmissibility of the vibration signal of the water pump impeller is strong, the position of the bearing seat is taken as an optimal vibration measuring point, and the vibration signal of the water pump impeller is measured by using the vibration sensor, so that the interference generated by other vibration signals is avoided as much as possible.
The data acquisition of the scheme is set to be carried out once every 5ms, and the noise signal is recorded asThe vibration signal is recorded as +.>Wherein->Indicating the acquisition time. For example, the noise signal of the first acquisition is +.>The noise signal of the second acquisition is +.>The noise signal of the third acquisition is +.>And so on. For the accuracy of the subsequent analysis, the noise signal and the vibration signal are subjected to data discretization, which is a well-known technique in the art, and redundant description is not made.
To this end, time-series data of the noise signal and the vibration signal are acquired.
The abnormal analysis module is used for converting mechanical energy of the motor into potential energy of water in normal operation of the water pump, but cavitation, impeller unbalance, bearing abrasion and other phenomena possibly occur in the water pump due to the influence of the working environment, so that abnormal conditions are generated in noise signals and vibration signals of the water pump, and the flow and efficiency of the water pump are directly influenced. Therefore, the present invention focuses on analyzing noise signals and vibration signals to generate abnormal conditions, and a flowchart of the implementation of the present invention is shown in fig. 2.
Based on the analysis, the noise signal after pretreatment is obtainedAnd vibration signal->Both are discrete time series data. Because the noise signal and the vibration signal can reflect the working state of the water pump, the noise signal and the vibration signal can acquire abnormal conditions through the same processing means, the noise signal is analyzed, and the vibration signal adopts the same processing.
Noise signal of theoretical non-positive displacement water pumpIs an infinitely long discrete time sequence, and only needs the time of signal acquisition to be long enough. However, data of an indefinite length cannot be processed in a computer system, and thus a pair +.>Slicing time segments, analyzing time segments of finite length/>Abnormal situation. And selecting a time window, analyzing the abnormality of the data in the time, and analyzing the abnormal state of the water pump by continuously sliding the time window. The size of the time window is selected according to a preset time interval, wherein the preset time interval is +.>The time window starts from the time sequence beginning of the noise signal, and slides back once for 10 seconds, and the preset time interval enforcer can select according to the actual situation. Such as a first time window of 0 to 10 seconds, a second time window of 10 to 20 seconds, a third time window of 20 to 30 seconds, and so on.
The noise signal in each time window is analyzed, and the noise signal is in a normal state for most of the time, so the signal fluctuation is near the mean value, and the variance is small. When the signal is abnormal, the value of the extremely individual signal has larger fluctuation, so that the dispersion degree of the data distribution is larger.
Here, the time domain distribution dispersion for each time window is calculated:
in the method, in the process of the invention,representing the temporal distribution dispersion of the noise signal within a time window,/->Representing the number of noise signals in a time window, +.>Represents the i-th noise signal in the time window, < >>Representing the mean value of the noise signal in the time window, +.>A coefficient of variation representing the amplitude of all noise signals of the time window. The coefficient of variation is a well known technique in the art and will not be described in detail.
Reflects the discrete condition of the noise signal in the time window in the time domain, when the noise signal in the time window is abnormal, the more discrete the distribution condition is, the difference between the i-th noise signal in the time window and the mean value of the noise signal in the time window is->The larger the first product factor +.>The greater the simultaneous coefficient of variation +.>The larger the temporal distribution dispersion of the time window +.>The larger. On the contrary, when the noise signals in the time window are normal and the noise signals are all near the mean value, the time domain distribution dispersion of the time window is +.>The smaller.
Further, the time domain distribution dispersion for each time windowReflecting the discrete situation in which the noise signal is distributed in amplitude. In addition, by combining the probability distribution characteristics in the time window, when the noise signal is abnormal, the probability distribution of the noise signal in the time window is more disordered, and the degree of heterogeneity of the noise signal is more.
Here, the time domain hopping heterogeneity coefficient for each time window is calculated:
in the method, in the process of the invention,time domain hopping heterogeneity coefficient representing a time window, < >>Representing the temporal distribution dispersion of the time window, < ->For normalization function->Representing the number of different noise signals in a time window, etc>Indicating the%>Different noise signals, < >>Indicating the%>A different noise signalProbability of occurrence.
When the non-positive-displacement water pump is working normally, its noise signal should fluctuate within a certain range, so that it is pretreatedShould belong to a fixed set of noise signals and have substantially uniform frequency of occurrence. However, when an abnormal noise signal occurs, its probability is small, resulting in calculation +.>Is larger, a second product factorThe larger the temporal distribution dispersion of the simultaneous time window +.>The larger the value is, the time-domain jump heterogeneity coefficient of the time window +.>The larger.
Therefore, the time domain jump heterogeneity coefficient reflects the abnormal situation of the jump of the noise signal on the time domain, and when the signal is abnormal, the time domain jump heterogeneity coefficientThe larger; when the signal is normal, the time domain jump heterogeneity coefficient +.>The smaller.
Further, for the time domain abnormal condition of the time sequence, only the numerical abnormality of the sequence is reflected, when the signal has abnormal high-frequency small-amplitude fluctuation, the signal is easy to submerge in the normal fluctuation of the signal, and at the moment, the abnormal condition is generally difficult to detect only through time domain analysis. Therefore, it is necessary to convert the time domain signal into the frequency domain for analysis.
Here, a discrete fourier transform is performed on the time domain noise signal in each time window. Because the signal is non-periodic discrete signal, after discrete Fourier transform, frequency spectrum leakage can occur, so that the frequency domain signal is continuous, and the value range of the independent variable is. Discrete fourier transforms are well known in the art and are not described in detail.
Specifically, the spectrogram is analyzed to obtain the fundamental frequency component asTo simplify the analysis, a further option is->The bilateral spectrum is converted into a unilateral spectrum for the center, and the value range is +.>. When the non-positive-displacement water pump works normally, the frequency spectrum capacity of the noise signal is mainly on the fundamental frequency component, and other frequency components occupy less energy. In addition, since spectrum leakage occurs, the frequency domain signal of noise is infinitely wide, i.e., the range of values +.>Number of medium frequency components +.>There are infinite ones, where the frequency domain energy is substantially 0 when the frequency component is far from the fundamental frequency, and where the computation is meaningless. The method selects the attenuation 10dB value of the frequency domain energy of the fundamental frequency signal as a screening threshold value, and only considers the frequency signals with the frequency domain energy of the frequency component higher than the screening threshold value.
After the above processing, the noise frequency signal to be considered is filtered, namely, the noise frequency signal is processed by the base frequencyAnd filtering the noise frequency signal to be considered by using a chebyshev band-pass filter to obtain a filtered frequency signal for the center frequency.
Here, the time-frequency energy anomaly coefficient for each time window is calculated:
in the method, in the process of the invention,time-frequency energy anomaly coefficient representing a time window, < ->Time domain hopping heterogeneity coefficient representing a time window, < >>Representing the number of frequency components in the frequency signal to be considered,/->Representing the frequency domain energy of the i-th frequency component after filtering the frequency signal to be considered,/or->Representing the frequency domain energy of the ith frequency component before filtering the frequency signal to be considered. It should be noted that->Representing fundamental frequency +.>Is a frequency domain energy of (a) a (b).
When the signal is normal, the frequency domain energy of the filtered frequency component is close to 0, and the third multiplication factorThe smaller the time-frequency energy anomaly coefficient of the time window +.>The smaller; on the contrary, when the signal has high frequency abnormal signal, besides the frequency domain energy of the fundamental frequency, other larger frequency domain energy exists, and the third multiplication factor is +.>The larger the time window is, the time-frequency energy anomaly coefficient +.>The value is larger. Time domain hopping heterogeneity coefficient of simultaneous time window +.>The larger the indication signal is abnormal, the time-frequency energy of the time windowCoefficient of volume abnormality->The greater the value.
The abnormality extraction module is used for extracting the time-frequency hopping abnormality index of each time window by taking the attenuation rule of the frequency component energy in each time window into consideration and taking the attenuation rule of the frequency component energy in normal and abnormal time into consideration, and extracting the combined time-frequency hopping abnormality index of each time window in a data fusion mode.
When the non-positive displacement water pump is operating normally, the further each frequency component is from the fundamental frequency, the more severely its energy decays. When the water pump is abnormal, an obvious energy gap exists on other frequency components, and a spectrum energy attenuation distance sequence of each time window is constructed:
in the method, in the process of the invention,a spectral energy attenuation characteristic value representing an i-th frequency component in the frequency signal to be considered,representing the i-th frequency component in the frequency signal to be considered,/or->Representing the fundamental frequency component +.>Frequency domain energy representing said i-th frequency component,/or->Frequency domain energy representing fundamental frequency component, +.>For a sequence of spectral energy decay distances of a time window, < >>And the characteristic value of the L-th spectral energy attenuation in the spectral energy attenuation distance sequence of the time window is represented.
Under the normal state, the spectrum energy attenuation distance sequence presents a gradually attenuated distribution state, but the distribution rule is broken when an abnormality occurs. Such as during normal operation… is stepwise decreasing but when abnormal, this distribution rule is broken.
Specifically, for the spectrum energy attenuation distance sequence of any time window, subtracting the previous spectrum energy attenuation characteristic value from the next spectrum energy attenuation characteristic value in the spectrum energy attenuation distance sequence to obtain the spectrum energy difference sequence of the time window.
Here, the time-frequency hopping abnormality index for each time window is calculated:
in the method, in the process of the invention,time-frequency hopping anomaly index indicating time window, < >>Length of spectrum energy difference sequence representing time window, +.>And->(k+1) th and kth spectrum energy difference values in spectrum energy difference sequences respectively representing time windows, ">Representing the time-frequency energy anomaly coefficient of the time window.
In normal, the spectrum energy attenuation distance sequence presents a progressive attenuation distribution state, and a fourth product factorThe smaller the time window of the time window is, the smaller the time-frequency energy abnormality coefficient of the time window is, the time-frequency hopping abnormality index +.>Smaller. Otherwise, when abnormal, the progressive attenuation distribution state of the spectrum energy attenuation distance sequence is broken, and a fourth product factor +.>The larger the time window of the simultaneous time window is, the larger the time-frequency energy anomaly coefficient of the time window is, the time-frequency hopping anomaly index +.>Larger.
Based on the analysis, the time-frequency hopping abnormality index of each time window of the noise signal is obtained, and the time-frequency hopping abnormality index of each time window in the vibration signal is obtained by adopting the same method. Here, the combined time-frequency hopping abnormality index for each time window is calculated:
in the method, in the process of the invention,joint time-frequency hopping anomaly index indicating time window, < >>Time-frequency hopping abnormality index indicating time window in noise signal,/->Time-frequency hopping abnormality index indicating time window in vibration signal,/->Is a normalization function.
So far, the combined time-frequency hopping abnormality index of each time window is obtained.
And the abnormality monitoring module is used for monitoring the abnormal condition generated when the water pump operates according to the combined time-frequency hopping abnormality index of each time window.
And according to the combined time-frequency hopping abnormal indexes of each time window, transmitting the processed and analyzed combined time-frequency hopping abnormal indexes to a water pump control end, and according to the size of the combined time-frequency hopping abnormal indexes, analyzing the working state of the water pump at the water pump control end to judge whether the abnormality exists. Frequency hopping anomaly index when combinedAt->When the water pump works normally, the water pump is judged to work normally; frequency hopping abnormality index when combined>At->When the water pump is judged to be abnormal; frequency hopping abnormality index when combined>Is greater than->When the water pump is judged to have abnormal faults, and +.>The larger the value, the more serious the fault. When the abnormal operation of the water pump is detected, the maintenance is needed in time, and further economic loss is reduced.
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. The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (5)
1. The non-positive-displacement water pump running state monitoring system based on the data of the Internet of things is characterized by comprising the following modules:
the data acquisition module acquires time sequence data of the monitoring parameters;
the anomaly analysis module is used for acquiring data in each time window with preset size according to the complexity of the time sequence data and acquiring the time domain distribution dispersion of each time window according to the fluctuation condition of the data in each time window; acquiring a time domain jump heterogeneity coefficient of each time window according to the time domain distribution dispersion of each time window; acquiring a time-frequency energy anomaly coefficient of each time window according to the time-domain jump heterogeneity coefficient of each time window and the frequency domain data distribution rule;
the abnormality extraction module is used for obtaining a spectrum energy attenuation distance sequence of each time window according to the energy attenuation rule of the frequency component in each time window; acquiring a spectrum energy difference sequence of each time window according to the spectrum energy attenuation distance sequence of each time window, and acquiring a time-frequency hopping abnormal index of each time window according to the spectrum energy difference sequence and the time-frequency energy abnormal coefficient of each time window; acquiring a combined time-frequency hopping abnormality index of each time window according to the time-frequency hopping abnormality index of each time window;
the abnormality monitoring module is used for acquiring a state monitoring result in the running process of the water pump according to the combined time-frequency hopping abnormality index of each time window;
the method for acquiring the time domain distribution dispersion of each time window according to the fluctuation condition of the data in each time window comprises the following steps:
setting a time window with a preset size for any time sequence of monitoring parameters, and acquiring data in each time window in the time sequence;
for any time window, taking the accumulated sum of absolute values of differences between data average values of all data in the time window and each data in the time window as a first product factor;
taking the product of the first product factor and the variation coefficient of each time window as the time domain distribution dispersion of each time window;
the method for acquiring the time domain jump heterogeneity coefficient of each time window according to the time domain distribution dispersion of each time window comprises the following steps:
for any time window, counting the probability corresponding to each unequal data in the time window, acquiring the accumulated sum of the product between the probability corresponding to each unequal data and the logarithmic value of the corresponding probability in the time window, and taking the negative mapping result of the accumulated sum as a second product factor;
taking a normalized value of the product of the second product factor of each time window and the time domain distribution dispersion as a time domain jump heterogeneity coefficient of each time window;
the method for acquiring the time-frequency energy anomaly coefficient of each time window according to the time-domain jump heterogeneity coefficient of each time window and the frequency domain data distribution rule comprises the following steps:
for any time window, obtaining unilateral spectrum frequency domain data in the time window by utilizing Fourier transformation, taking frequency components with frequency domain energy higher than a preset screening threshold value of the frequency components as calculation objects, and taking output obtained by filtering the calculation objects by utilizing a Chebyshev band-pass filter as filtering components;
taking the frequency domain energy accumulation sum of the fundamental frequency component and each frequency component in the calculation object as a denominator, taking the frequency domain energy accumulation sum of each frequency component in the filtering component as a numerator, and taking the ratio of the numerator to the denominator as a third multiplication factor of a time window;
taking the product of the third multiplication factor of each time window and the time domain jump heterogeneity coefficient as the time-frequency energy anomaly coefficient of each time window;
the method for acquiring the time-frequency hopping abnormality index of each time window according to the frequency spectrum energy difference sequence and the time-frequency energy abnormality coefficient of each time window comprises the following steps:
for the spectrum energy attenuation distance sequence of any time window, taking the difference value of the last element and the previous element in the spectrum energy attenuation distance sequence as an adjacent difference value, and taking a sequence formed by all adjacent difference values according to time ascending sequence as a spectrum energy difference sequence of the time window;
acquiring a fourth product factor of each time window according to the spectrum energy difference sequence of each time window;
taking the product of the fourth product factor of each time window and the time-frequency energy anomaly coefficient as the time-frequency hopping anomaly index of each time window;
the method for obtaining the fourth product factor of each time window according to the spectrum energy difference sequence of each time window comprises the following steps: and for the spectrum energy differential sequence of any time window, obtaining the differential ratio of the last differential value to the previous differential value in the spectrum energy differential sequence, and taking the accumulated sum of the differential ratio on the spectrum energy differential sequence as a fourth product factor of the time window.
2. The monitoring system of operation state of a non-positive-displacement pump based on internet of things data according to claim 1, wherein the method for obtaining the spectrum energy attenuation distance sequence of each time window according to the frequency component energy attenuation law in each time window is as follows:
for any calculation object of the time window, acquiring a spectrum energy attenuation characteristic value of each frequency component in the calculation object of the time window according to the change of the frequency component energy of the calculation object of the time window;
and for any calculation object of the time window, taking the sequence of the spectral energy attenuation characteristic values of all frequency components in the calculation object as a sequence of the spectral energy attenuation distances of the time window according to the sequence of the frequency components from the base frequency component from near to far.
3. The monitoring system of operation state of a non-positive-displacement pump based on internet of things data according to claim 2, wherein the method for obtaining the spectral energy attenuation characteristic value of each frequency component in the calculation object of the time window according to the change of the frequency component energy of the calculation object of the time window is as follows:
in the method, in the process of the invention,spectral energy attenuation characteristic value of ith frequency component in calculation object representing time window, < ->I-th frequency component in the calculation object representing the time window,>representing the fundamental frequency component within the computational object of the time window,frequency domain energy of the ith frequency component in the computation object representing the time window, +.>The frequency domain energy of the fundamental frequency component within the computation object representing the time window.
4. The monitoring system of operation state of a non-positive-displacement pump based on internet of things data according to claim 1, wherein the method for obtaining the combined time-frequency hopping abnormality index of each time window according to the time-frequency hopping abnormality index of each time window is as follows:
and regarding the time-frequency hopping abnormal indexes of each time window of each monitoring parameter, taking the accumulated normalization result of the time-frequency hopping abnormal indexes of each corresponding time window of all the monitoring parameters as the combined time-frequency hopping abnormal index of each time window.
5. The monitoring system of the running state of the non-positive-displacement water pump based on the internet of things data according to claim 1, wherein the method for acquiring the monitoring result of the running state of the water pump according to the combined time-frequency hopping abnormal index of each time window is as follows:
and for the combined time-frequency hopping abnormal index of each time window, acquiring the abnormal degree of the water pump in the running process according to the comparison result of the combined time-frequency hopping abnormal index and the preset length interval, and monitoring the abnormal faults occurring in the running process of the water pump.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311147254.1A CN116877452B (en) | 2023-09-07 | 2023-09-07 | Non-positive-displacement water pump running state monitoring system based on Internet of things data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311147254.1A CN116877452B (en) | 2023-09-07 | 2023-09-07 | Non-positive-displacement water pump running state monitoring system based on Internet of things data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116877452A CN116877452A (en) | 2023-10-13 |
| CN116877452B true CN116877452B (en) | 2023-12-08 |
Family
ID=88262607
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311147254.1A Active CN116877452B (en) | 2023-09-07 | 2023-09-07 | Non-positive-displacement water pump running state monitoring system based on Internet of things data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116877452B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117540235B (en) * | 2023-11-13 | 2024-03-29 | 中铁十四局集团建筑工程有限公司 | Real-time monitoring method for temporary electricity utilization abnormal data |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6115684A (en) * | 1996-07-30 | 2000-09-05 | Atr Human Information Processing Research Laboratories | Method of transforming periodic signal using smoothed spectrogram, method of transforming sound using phasing component and method of analyzing signal using optimum interpolation function |
| JP2003271241A (en) * | 2002-03-13 | 2003-09-26 | Mitsubishi Heavy Ind Ltd | Operation supervisory and controlling system |
| CN105134618A (en) * | 2015-09-10 | 2015-12-09 | 中国农业大学 | Centrifugal pump regulating method based on pressure pulsation time-frequency characteristic analysis |
| CN106015020A (en) * | 2016-05-17 | 2016-10-12 | 安徽理工大学 | Self speed regulation immersible pump remote scheduling system based on Internet of Things |
| CN108757502A (en) * | 2018-05-15 | 2018-11-06 | 江苏大学 | A kind of water pump assembly typical case's health status monitoring device and method based on Internet of Things |
| CN115326147A (en) * | 2022-10-17 | 2022-11-11 | 江苏煵笙重工有限公司 | Real-time monitoring method and system for desulfurization tower for ship |
| CN115562148A (en) * | 2022-11-11 | 2023-01-03 | 台州学院温岭研究院 | Intelligent networking detection system for fan air output |
| WO2023040575A1 (en) * | 2021-09-17 | 2023-03-23 | 中通服和信科技有限公司 | Internet-of-things-based abnormality early warning analysis system and method for special operation site |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7406398B2 (en) * | 2004-06-05 | 2008-07-29 | Schlumberger Technology Corporation | System and method for determining pump underperformance |
-
2023
- 2023-09-07 CN CN202311147254.1A patent/CN116877452B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6115684A (en) * | 1996-07-30 | 2000-09-05 | Atr Human Information Processing Research Laboratories | Method of transforming periodic signal using smoothed spectrogram, method of transforming sound using phasing component and method of analyzing signal using optimum interpolation function |
| JP2003271241A (en) * | 2002-03-13 | 2003-09-26 | Mitsubishi Heavy Ind Ltd | Operation supervisory and controlling system |
| CN105134618A (en) * | 2015-09-10 | 2015-12-09 | 中国农业大学 | Centrifugal pump regulating method based on pressure pulsation time-frequency characteristic analysis |
| CN106015020A (en) * | 2016-05-17 | 2016-10-12 | 安徽理工大学 | Self speed regulation immersible pump remote scheduling system based on Internet of Things |
| CN108757502A (en) * | 2018-05-15 | 2018-11-06 | 江苏大学 | A kind of water pump assembly typical case's health status monitoring device and method based on Internet of Things |
| WO2023040575A1 (en) * | 2021-09-17 | 2023-03-23 | 中通服和信科技有限公司 | Internet-of-things-based abnormality early warning analysis system and method for special operation site |
| CN115326147A (en) * | 2022-10-17 | 2022-11-11 | 江苏煵笙重工有限公司 | Real-time monitoring method and system for desulfurization tower for ship |
| CN115562148A (en) * | 2022-11-11 | 2023-01-03 | 台州学院温岭研究院 | Intelligent networking detection system for fan air output |
Non-Patent Citations (1)
| Title |
|---|
| 某船日用海水压力柜水泵振动异常问题分析;吴善跃;张振中;陈昕;;中国修船(第02期);第17-21页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116877452A (en) | 2023-10-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109855874B (en) | Random resonance filter for enhancing detection of weak signals in vibration assisted by sound | |
| CN116877452B (en) | Non-positive-displacement water pump running state monitoring system based on Internet of things data | |
| CN117805542B (en) | Mining flameproof intrinsically safe frequency converter operation monitoring system | |
| CN112881014B (en) | Offline NVH (noise, vibration and harshness) testing system and method of transmission assembly | |
| CN109063672A (en) | A kind of early stage bearing outer ring method for diagnosing faults based on adaptive M CKD | |
| JP2002090267A (en) | Failure diagnosis method | |
| CN116237817A (en) | Five-axis linkage numerical control machine intelligent monitoring system based on Internet of things | |
| Shao et al. | Gearbox vibration monitoring using extended Kalman filters and hypothesis tests | |
| Gao et al. | A wavelet packet and residual analysis based method for hydraulic pump health diagnosis | |
| CN118258454B (en) | Water flow detection system based on acoustic chromatography section detection | |
| CN114838924A (en) | Structural damping ratio identification method based on wind-induced vibration non-stationary response | |
| Yao et al. | A framework via impulses-oriented Gini index and extreme value distribution for rolling bearing dynamic fault alarm and identification | |
| CN115186707A (en) | Simple gear box fault detection method | |
| CN117668646A (en) | Speed reducer running state detection method, system, equipment and medium based on multi-feature fusion | |
| CN117807538A (en) | Fault detection method based on dynamic threshold and trend prediction | |
| CN113686577B (en) | Bearing fault diagnosis method based on rapid nonlinear sparse spectrum | |
| CN111980900B (en) | Water pump fault diagnosis method based on multi-source data fusion analysis | |
| CN113513304A (en) | Balance degree detection method based on electric diagram of pumping unit | |
| CN114412447A (en) | Fault detection method and device for screw pump well | |
| CN118520402B (en) | Fault diagnosis system and method for large-sized variable-speed unit | |
| CN114548156B (en) | Distributed optical fiber temperature measurement and noise reduction method based on downsampling and convolutional neural network | |
| CN205353084U (en) | Fluid on -line monitoring device with parallelly connected installation of working connection | |
| CN111043967B (en) | Method and device for monitoring optical fiber length of distributed optical fiber temperature sensing system in real time | |
| CN117433591B (en) | Automatic equipment monitoring system based on multi-sensor data | |
| CN118584391B (en) | Bidirectional direct-current power supply ripple monitoring system and method based on measured electric variable |
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 |