CN118391249B - Method and device for processing running state monitoring data of large water pump - Google Patents
Method and device for processing running state monitoring data of large water pump Download PDFInfo
- Publication number
- CN118391249B CN118391249B CN202410822683.2A CN202410822683A CN118391249B CN 118391249 B CN118391249 B CN 118391249B CN 202410822683 A CN202410822683 A CN 202410822683A CN 118391249 B CN118391249 B CN 118391249B
- Authority
- CN
- China
- Prior art keywords
- monitoring data
- sequence
- steady
- state
- water pump
- 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 314
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 192
- 238000000034 method Methods 0.000 title claims abstract description 64
- 238000012545 processing Methods 0.000 title claims abstract description 21
- 230000002159 abnormal effect Effects 0.000 claims abstract description 55
- 230000000875 corresponding effect Effects 0.000 claims description 43
- 230000011218 segmentation Effects 0.000 claims description 37
- 238000013507 mapping Methods 0.000 claims description 13
- 238000004422 calculation algorithm Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 6
- 230000002596 correlated effect Effects 0.000 claims description 6
- 230000035772 mutation Effects 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 2
- 238000003672 processing method Methods 0.000 claims 1
- 238000001514 detection method Methods 0.000 abstract description 7
- 238000006073 displacement reaction Methods 0.000 abstract description 2
- 238000010606 normalization Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 9
- 238000004458 analytical method Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000012423 maintenance Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 230000001360 synchronised effect Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000003646 Spearman's rank correlation coefficient Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001627 detrimental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
- F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
- F04B51/00—Testing machines, pumps, or pumping installations
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Control Of Positive-Displacement Pumps (AREA)
Abstract
The invention relates to the technical field of non-positive displacement pumps, in particular to a method and a device for processing running state monitoring data of a large water pump. Firstly, obtaining the running performance factor of a water pump; and further acquiring all steady-state time periods, and acquiring an electrical performance steady-state factor and a pipeline steady-state factor of the water pump in each steady-state time period, so as to acquire the steady-state performance factor of the water pump in each steady-state time period, thereby adjusting the abnormal score of the flow monitoring data and carrying out abnormal monitoring on the running state of the water pump. According to the invention, the running environment and performance of the water pump are firstly evaluated by using the current and the voltage, then the steady state condition of the voltage and the current and the steady state condition of the flow and the pressure at the water outlet pipe are analyzed in the running steady state period of the water pump, and the steady state running performance of the water pump in the steady state period, namely the fault possibility of the water pump is evaluated, so that the abnormal detection consideration weight of flow data in the running steady state is reduced, and the misjudgment of partial normal flow monitoring data caused by factors such as non-pump body faults is reduced.
Description
Technical Field
The invention relates to the technical field of non-positive displacement pumps, in particular to a method and a device for processing running state monitoring data of a large water pump.
Background
The steady operation of large-scale water pump is crucial to the production life of human, and the water outlet flow of water pump is one of the important indexes of measuring water pump performance, monitors water pump water outlet flow and can help know whether the pipeline of water pump is blockked up, whether the interior subassembly of pump wears out the trouble, and the unusual condition of water pump is convenient for in time discover, and then can make maintenance and repair plan to the water pump according to timely monitoring result, reduce reliance and trouble shut down risk to artifical experience or periodic maintenance, help optimizing equipment operating efficiency and management effect, more accord with modern industry to the requirement of safety, green, high-efficient production.
When the water pump outlet water flow is monitored, a local outlier factor (Local Outlier Factor, LOF) algorithm is generally adopted to perform outlier judgment on the flow monitoring data, so that early warning is performed on abnormal flow monitoring data in the running process; however, not all the outlier flow monitoring data are abnormal flow monitoring data caused by water pump faults, such as water pump outlet flow changes caused by external factors such as non-pump body faults such as unstable water pump power supply, and partial flow monitoring data deviation can be misjudged as abnormal flow monitoring data and early-warned, so that the accuracy of monitoring the running state of the water pump is reduced.
Disclosure of Invention
In order to solve the technical problems of inaccurate detection of flow monitoring data of a water pump and low accuracy of monitoring the running state of the water pump in the prior art, the invention aims to provide a method and a device for processing the running state monitoring data of a large water pump, and the adopted technical scheme is as follows:
the invention provides a method for processing running state monitoring data of a large water pump, which comprises the following steps:
Acquiring a monitoring data sequence of the water pump under each monitoring index, wherein the monitoring index at least comprises current and voltage of the water pump and flow and pressure at a water outlet pipe of the water pump;
Obtaining the running performance factor of the water pump according to the fluctuation similarity and the corresponding fluctuation characteristics between the voltage monitoring data sequence and the current monitoring data sequence; respectively segmenting the voltage monitoring data sequence and the current monitoring data sequence at least twice in different segmentation modes, and acquiring all steady-state time periods when the water pump operates according to segmentation results;
In each steady-state period, acquiring an electrical property steady-state factor of the water pump according to the fluctuation similarity of the voltage monitoring data sequence and the current monitoring data sequence, and acquiring a pipeline steady-state factor of the water pump pipeline according to the fluctuation characteristics between the flow monitoring data sequence and the pressure monitoring data sequence; acquiring a steady-state performance factor of the water pump in each steady-state period according to the running performance factor, the electrical performance steady-state factor and the pipeline steady-state factor;
Obtaining an abnormal score of each flow monitoring data in a flow monitoring data sequence, and adjusting the abnormal score of each flow monitoring data in each steady-state period according to the steady-state performance factor to obtain a corrected abnormal score; and carrying out abnormal monitoring on the running state of the water pump according to the corrected abnormal score or the abnormal score of each flow monitoring data in the flow monitoring data sequence.
Further, the method for acquiring the operation performance factor comprises the following steps:
Acquiring a coordination reference coefficient according to the sequence similarity condition between the voltage monitoring data sequence and the current monitoring data sequence;
Segmenting the voltage monitoring data sequence and the current monitoring data sequence respectively and sequencing labels in sequence; according to the fluctuation characteristics of the voltage sequence segments and the fluctuation characteristics of the current sequence segments under the same sequence number, obtaining fluctuation reference weights of the segments with the same sequence number in the corresponding time periods; according to the voltage monitoring data in the voltage sequence segment under the same sequence number and the current monitoring data in the current sequence segment, acquiring the performance parameters of the corresponding time period of the same sequence number segment; weighting and summing the performance parameters according to the fluctuation reference weight to obtain a performance reference coefficient;
Acquiring an operation performance factor according to the coordination reference coefficient and the performance reference coefficient; the coordination reference coefficient and the performance reference coefficient are both positively correlated with the running performance factor.
Further, the method for acquiring the steady-state period includes:
segmenting the voltage monitoring data sequence and the current monitoring data sequence according to the mutation of the monitoring data respectively to obtain all first voltage sequence segments and first current sequence segments;
Segmenting the voltage monitoring data sequence and the current monitoring data sequence according to the nuclear density of the monitoring data respectively to obtain all second voltage sequence segments and second current sequence segments;
and under all sequence segments with the same sequence number, taking the superposition time periods of the first voltage sequence segment, the first current sequence segment, the second voltage sequence segment and the second current sequence segment as steady-state time periods.
Further, the method for acquiring the first voltage sequence segment and the first current sequence segment comprises the following steps:
Taking the voltage monitoring data sequence and the current monitoring data sequence as a first monitoring sequence to be segmented;
When the differential value in the differential sequence is larger than a preset first threshold value, taking the sequence number of the differential value corresponding to the differential value in the differential sequence as the sequence number of a segmentation point of the first sequence to be segmented, segmenting the first sequence to be segmented according to the sequence number of the segmentation point, and obtaining all sequence segments of the first sequence to be segmented; all sequence segments of the voltage monitoring data sequence are all first voltage sequence segments, and all sequence segments of the current monitoring data sequence are all first current sequence segments.
Further, the method for acquiring the second voltage sequence segment and the second current sequence segment comprises the following steps:
Taking the voltage monitoring data sequence and the current monitoring data sequence as a second monitoring sequence to be segmented; constructing two-dimensional monitoring data according to each monitoring data and the corresponding acquisition time in the second monitoring sequence to be segmented, and clustering all the two-dimensional monitoring data based on a time point clustering algorithm to obtain all sequence segments of the second monitoring sequence to be segmented;
All sequence segments of the voltage monitoring data sequence are all second voltage sequence segments, and all sequence segments of the current monitoring data sequence are all second current sequence segments.
Further, the method for acquiring the electrical performance steady-state factor comprises the following steps:
Calculating the sum of autocorrelation coefficients of the voltage monitoring data in each steady-state period as a first parameter; calculating the sum of autocorrelation coefficients of the current monitoring data in each steady-state period as a second parameter; and taking the difference between the first parameter and the second parameter in the same steady-state period as an electrical performance steady-state factor in the steady-state period.
Further, the method for obtaining the pipeline steady-state factor comprises the following steps:
Respectively obtaining a flow variation coefficient of flow monitoring data in each steady-state period and a pressure variation coefficient of the pressure monitoring data in each steady-state period, and obtaining a first fluctuation characteristic parameter according to the flow variation coefficient and the pressure variation coefficient, wherein the flow variation coefficient and the pressure variation coefficient are both in autocorrelation with the first fluctuation characteristic parameter;
Respectively acquiring all flow peaks and flow valleys in the flow monitoring data and all pressure peaks and pressure valleys in the pressure monitoring data; obtaining peak value parameters according to all the flow peak values and all the pressure peak values, obtaining valley value parameters according to all the flow valley values and all the pressure valley values, and obtaining second fluctuation characteristic parameters according to the differences between the peak value parameters and the valley value parameters;
and acquiring a pipeline steady-state factor in the steady-state period according to the first fluctuation characteristic parameter and the second fluctuation characteristic parameter, wherein the first fluctuation characteristic parameter and the second fluctuation characteristic parameter are positively correlated with the pipeline steady-state factor.
Further, the method for obtaining the anomaly score comprises the following steps:
And obtaining an LOF value of each flow monitoring data in the flow monitoring data sequence by using an LOF algorithm, and taking the LOF value as an anomaly score.
Further, the method for obtaining the corrected anomaly score comprises the following steps:
Normalizing the negative correlation mapping of the steady-state performance factors to obtain an adjustment weight; and reducing the abnormal score corresponding to the flow monitoring data in the steady-state period according to the adjustment weight to obtain a corrected abnormal score.
The invention also provides a device for processing the monitoring data of the running state of the large water pump, which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein the processor realizes the steps of the method for processing the monitoring data of the running state of the large water pump when executing the computer program.
The invention has the following beneficial effects:
The method comprises the steps of acquiring a monitoring data sequence of the water pump under each monitoring index, wherein the monitoring indexes at least comprise current and voltage of the water pump and flow and pressure at a water outlet pipe of the water pump; according to the fluctuation similarity and the corresponding fluctuation characteristics between the voltage monitoring data and the current monitoring data, the running performance factor reflecting the stability and the performance quality of the water pump working environment is obtained, so that the performance state of the water pump is conveniently and subsequently evaluated, the abnormal consideration weight of the flow monitoring data in the normal running period is adjusted, and the misjudgment is reduced; then, the voltage monitoring data sequence and the current monitoring data sequence are segmented at least twice respectively in different segmentation modes, single-time division randomness is reduced, the accuracy of division of steady-state time periods is improved, and all steady-state time periods when the water pump operates are obtained according to segmentation results; in each steady-state period, acquiring an electrical performance steady-state factor of the water pump according to the fluctuation similarity of the voltage monitoring data and the current monitoring data; in each steady-state period, acquiring a pipeline steady-state factor of the water pump pipeline according to fluctuation characteristics between the flow monitoring data and the pressure monitoring data; acquiring a steady-state performance factor of the water pump in each steady-state period according to the running performance factor, the electrical performance steady-state factor and the pipeline steady-state factor, if the motor performance of the water pump is stable and the pipeline performance is stable, the possibility of failure of the water pump component is lower no matter what the working environment and the state of the water pump are, the water pump normally operates in a corresponding working state and generates steady-state response, but if the working state of the water pump is stable and the performance is good, the possibility of failure hidden danger of the water pump in the corresponding period is lower; then, obtaining the abnormal score of each flow monitoring data, and adjusting the abnormal score of each flow monitoring data in each steady-state period according to the steady-state performance factors to obtain a corrected abnormal score; and carrying out abnormal monitoring on the running state of the water pump according to the corrected abnormal score or abnormal score of each flow monitoring data in the flow monitoring data sequence. According to the invention, the running environment and performance of the water pump are firstly evaluated by using the current and the voltage, and then the steady state condition of the voltage and the current and the steady state condition of the flow and the pressure at the water outlet pipe are analyzed in the running steady state period of the water pump, so that the steady state running performance of the water pump in the steady state period is evaluated, the possibility of the failure of the water pump is evaluated, the abnormal detection consideration weight of the flow data in the running steady state is further reduced, the misjudgment of partial normal flow monitoring data caused by factors such as non-pump body failure is reduced, and the accuracy of the running state monitoring of the water pump is 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 processing monitoring data of the running state of a large water pump according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for obtaining a performance factor according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a steady-state period according to an embodiment of the present invention;
FIG. 4 is a flowchart of a method for obtaining a pipeline steady-state factor according to an embodiment of the present invention;
Fig. 5 is a schematic structural diagram of a large-scale water pump operation state monitoring data processing device 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 is a detailed description of a specific implementation, structure, characteristics and effects of the method and device for processing the monitoring data of the running state of the large water pump according to the invention, which are provided by the invention, with reference to the accompanying drawings and the preferred embodiment. 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 invention provides a method and a device for processing running state monitoring data of a large water pump, and the method and the device are specifically described below with reference to the accompanying drawings.
The invention is to monitor the flow of the water pump, firstly utilize electric current and voltage to evaluate the running performance stability of the water pump; and then obtaining an operation steady-state period of the water pump, and analyzing steady-state conditions of voltage and current and steady-state conditions of flow and pressure at the water outlet pipe in the steady-state period, so as to evaluate steady-state operation performance of the water pump in the steady-state period, further reduce abnormal detection consideration weight of flow data during stable operation, and reduce misjudgment of partial normal flow monitoring data caused by factors such as non-pump body faults.
Referring to fig. 1, a flowchart of a method for processing monitoring data of a running state of a large water pump according to an embodiment of the present invention specifically includes:
Step S1, a monitoring data sequence of the water pump under each monitoring index is obtained, wherein the monitoring index at least comprises current and voltage of the water pump and flow and pressure at a water outlet pipe of the water pump.
In one embodiment of the invention, a current transformer and a voltage transformer are arranged on a motor of a water pump, and voltage monitoring data and current monitoring data during the operation of the water pump are respectively acquired; meanwhile, a flowmeter and a pressure sensor are arranged at the water outlet pipeline of the water pump, and flow monitoring data and pressure monitoring data at the water outlet pipeline are respectively collected when the water pump operates; carrying out standardized treatment on all monitoring data under each monitoring index, and eliminating dimension for subsequent analysis; and constructing a monitoring data sequence according to the corresponding acquisition time by using all the standardized monitoring data under each monitoring index.
It should be noted that, sampling frequencies of all sensors or collecting devices are consistent, so that sequence lengths and time stamps of monitoring data sequences under each monitoring index are consistent, and feasibility of subsequent analysis is ensured; the implementer can set the sampling frequency by himself, and can obtain other kinds or quantity of water pump monitoring indexes according to specific implementation requirements; the manner of normalizing all the monitoring data in the monitoring data sequence is maximum value and minimum value normalization, which is already the prior art and is not described herein.
It should be noted that, because the current and the voltage are unstable when the water pump is started, there may be a certain deviation to the load analysis result, so all the monitoring data sequences are the data obtained by starting monitoring and collecting after the water pump operates for at least 1 minute.
Step S2, obtaining the running performance factor of the water pump according to the fluctuation similarity and the corresponding fluctuation characteristics between the voltage monitoring data and the current monitoring data; and respectively segmenting the voltage monitoring data sequence and the current monitoring data sequence at least twice in different segmentation modes, and acquiring all steady-state time periods when the water pump operates according to segmentation results.
When the water pump normally operates, the voltage and the current generally have certain variation balance and coordination, and if the coordination of the voltage monitoring data and the current monitoring data is poor, the working state and the performance of the water pump are abnormal; when the current and the voltage have the same fluctuation trend, the coordination between the current and the voltage can be considered to be better; meanwhile, when the fluctuation is smaller and the power is higher, the water pump can be considered to have good performance and stable working state; therefore, according to the fluctuation similarity and the corresponding fluctuation characteristics between the voltage monitoring data sequence and the current monitoring data sequence, the running performance factor of the water pump is obtained, the running performance factor reflects the stability and the performance quality of the working environment of the water pump, the subsequent evaluation of the performance state of the water pump is facilitated, the abnormal consideration weight of the flow monitoring data in the normal operation period is adjusted, and the misjudgment is reduced.
Preferably, in one embodiment of the present invention, considering that the similarity between sequences reflects the fluctuation similarity thereof, and considering that the fluctuation characteristics of the monitoring data in the sequences may be smoothed under the influence of long time sequences, the voltage monitoring data sequences and the current monitoring data sequences are respectively processed in a segmented manner, so as to analyze the local fluctuation characteristics and the power in the local time period, and evaluate the coordination of the current and the voltage and the water pump performance; based on this, the method for acquiring the running performance factor includes:
referring to fig. 2, a flowchart of a method for obtaining an operation performance factor according to an embodiment of the present invention specifically includes:
step S201, obtaining a coordination reference coefficient according to the sequence similarity condition between the voltage monitoring data sequence and the current monitoring data sequence.
The coordination reference coefficient preliminarily reflects the variation coordination of voltage and current, and the side surface reflects whether the water pump normally operates.
In one embodiment of the invention, the similarity between sequences can be evaluated by taking the pearson correlation coefficient into consideration, and the higher the pearson correlation coefficient is close to 1, the higher the synchronism and coordination between voltage and current are, the more stable the water pump running state is; therefore, the pearson correlation coefficient between the voltage monitoring data sequence and the current monitoring data sequence is normalized, and the pearson coefficient is specifically substituted as x intoThe normalization result is used as a coordination reference coefficient; in other embodiments of the present invention, the practitioner may also use a measure of sequence similarity such as a correlation coefficient, a spearman's rank correlation coefficient, or a dynamic time warping, which are well known to those skilled in the art and are not described herein.
Step S202, segmenting the voltage monitoring data sequence and the current monitoring data sequence respectively and sequencing labels in sequence; according to the fluctuation characteristics of the voltage sequence segments and the fluctuation characteristics of the current sequence segments under the same sequence number, obtaining fluctuation reference weights of the segments with the same sequence number in the corresponding time periods; according to the voltage monitoring data in the voltage sequence segment under the same sequence number and the current monitoring data in the current sequence segment, acquiring the performance parameters of the corresponding time period of the same sequence number segment; and carrying out weighted summation on the corresponding performance parameters according to the fluctuation reference weight to obtain the performance reference coefficient.
Since long timing may be detrimental to analysis of ripple characteristics, the voltage monitoring data sequence and the current monitoring data sequence are first segmented and numbered sequentially, respectively, so as to analyze the coordination between voltage and current in the same local timing range.
In one embodiment of the invention, a voltage differential sequence of a voltage monitoring data sequence is firstly obtained, when the differential value in the voltage differential sequence is larger than a preset first threshold value, a differential value serial number of a corresponding differential value in the voltage differential sequence is used as a voltage segmentation point serial number, the voltage monitoring data sequence is segmented according to the voltage segmentation point serial number, and all voltage sequence segments are obtained; for example: the voltage monitoring data sequence is {0.2,0.2,0.3,0.4,0.8,0.8,0.9}, the corresponding voltage differential sequence is {0,0.1,0.1,0.4,0,0.1}, the 4 th differential value in the voltage differential sequence is larger than a preset first threshold value, the sequence number of the voltage segmentation point is 4, and the voltage monitoring data sequence is segmented into {0.2,0.2,0.3,0.4} and {0.8,0.8,0.9}; and then mapping the sequence numbers of the voltage segmentation points into the current monitoring data sequences to obtain all current sequence segments, and ensuring that the segment lengths of the voltage sequences and the segment lengths of the current sequences with the same sequence numbers are consistent.
It should be noted that, the first threshold is preset to be 0.3, and the implementer can set the value by himself; the difference is already known in the art and is not described here in detail.
In another embodiment of the present invention, the practitioner may also fit a voltage monitoring data curve, obtain all the extreme points, and segment the voltage monitoring data sequence using the extreme points as segmentation data points; and then segments the current monitoring data sequence. In other embodiments of the present invention, the practitioner may segment the voltage monitoring data sequence and the current monitoring data sequence by other methods, such as uniform segmentation, but it is necessary to ensure that the segment lengths of the voltage sequences and the segment lengths of the current sequences with the same sequence number are consistent.
It should be noted that there may be some voltage sequence segments and current sequence segments that contain only one or a small amount of monitoring data, which is inconvenient for subsequent analysis of the fluctuation characteristics, and the corresponding monitoring data sequence segments may be combined into any adjacent segment.
Then, according to the fluctuation characteristics of the voltage sequence segments and the fluctuation characteristics of the current sequence segments under the same sequence number, obtaining the fluctuation reference weight of the corresponding time period of the same sequence number segments; the fluctuation reference weight reflects the fluctuation condition of voltage and current, and the current and the voltage simultaneously and stably fluctuate, so that the better the performance and the working state of the water pump are.
In one embodiment of the invention, the variance of the voltage monitoring data in the voltage sequence segment and the variance of the current monitoring data in the current sequence segment under the same serial number are added and combined to perform negative correlation mapping normalization to obtain the fluctuation reference weight, wherein the variance can reflect the fluctuation characteristic of the data. The variances are added and combined, and then the opposite numbers are taken, the opposite numbers are taken as indexes in an exponential function with a natural constant e, the value of the exponential function is the normalization result of the negative correlation mapping, an implementer can also adopt other negative correlation mapping means for normalization, and the fluctuation characteristics of the monitoring data in the corresponding sequence segments can be represented by standard deviation, peak point number and the like, which are all common technical means for those skilled in the art and are not repeated herein.
In the embodiment of the invention, the performance parameters of the segments with the same sequence number in the corresponding time period are further obtained according to the voltage monitoring data in the segments with the same sequence number and the current monitoring data in the segments with the current sequence; specifically, obtaining the average value of voltage monitoring data in a voltage sequence segment under the same sequence number and the product of the average value of current monitoring data in a current sequence segment, then normalizing the maximum value and the minimum value of the product, and evaluating the performance of the water pump through the average power level in a period corresponding to the segment with the same sequence number, wherein the higher the power is, the higher the performance of the water pump is; meanwhile, combining the fluctuation reference weight in the corresponding time period, and carrying out weighted summation on the performance parameters of all the segments to obtain the performance reference coefficient of the water pump; the larger the reference coefficient of performance, the more stable the voltage and current fluctuations and the higher the performance.
It should be noted that, in other embodiments of the present invention, an implementer may also obtain a product of the voltage monitoring data in the voltage sequence segment and the mode or median of the current monitoring data in the current sequence segment under the same sequence number to evaluate the performance reference coefficient; other normalization means may also be employed and are not described in detail herein.
Step S203, obtaining operation performance factors according to the coordination reference coefficients and the performance reference coefficients; the coordination reference coefficient and the performance reference coefficient are positively correlated with the running performance factor.
In one embodiment of the invention, the coordination reference coefficient is considered to be the coordination and balance between voltage and current through overall analysis so as to evaluate the working state and performance of the water pump; the performance reference coefficient is used for evaluating the working state and performance of the water pump by locally analyzing the fluctuation characteristics between the voltage and the current; the stability and the performance quality of the working state of the water pump can be accurately evaluated by combining the two, so that the two are multiplied and combined to obtain the running performance factor. In other embodiments of the present invention, the operator may also use basic mathematical operations such as addition or weighted summation or positive correlation mapping means to obtain the running performance factor, which are all well known in the art and are not described herein.
Considering that when the water pump is running, temporary equipment operation or external environment interference possibly exists, so that the working state of the water pump has certain fluctuation and change; however, after the water pump enters a stable running state, the working state of the water pump is relatively stable, no obvious change occurs, and the running state of the water pump is evaluated at the stage, so that the water pump performance and the running state can be evaluated more favorably. Therefore, the embodiment of the invention needs to acquire the steady-state period when the water pump is operated.
Certain randomness may exist in view of segmenting the monitoring data sequence to evaluate the steady-state period based only on the fluctuation of the voltage monitoring data or the current monitoring data; however, if the steady-state period is estimated according to different segmentation modes and according to the synchronous steady-state fluctuation characteristics of multiple parameters, the accuracy of dividing the steady-state period is greatly improved; therefore, the embodiment of the invention respectively segments the voltage monitoring data sequence and the current monitoring data sequence at least twice in different segmentation modes, and obtains all steady-state time periods when the water pump operates according to segmentation results.
Preferably, in one embodiment of the present invention, the method for acquiring a steady-state period includes:
Segmenting a voltage monitoring data sequence and a current monitoring data sequence according to the mutation of the monitoring data respectively to obtain all first voltage sequence segments and first current sequence segments;
Specifically, the method for acquiring the first voltage sequence segment and the first current sequence segment includes:
Taking the voltage monitoring data sequence and the current monitoring data sequence as a first to-be-segmented monitoring sequence;
Obtaining a differential sequence of a first sequence to be monitored by segmentation, when the differential value in the differential sequence is larger than a preset first threshold value, taking the sequence number of the differential value corresponding to the differential value in the differential sequence as the sequence number of a segmentation point of the first sequence to be monitored by segmentation, and segmenting the first sequence to be monitored by segmentation point according to the sequence number of the segmentation point to obtain all sequence segments of the first sequence to be monitored by segmentation; all sequence segments of the voltage monitoring data sequence are all first voltage sequence segments and all sequence segments of the current monitoring data sequence are all first current sequence segments.
It should be noted that, the segmentation method of the first voltage sequence segmentation has been described in step S201, and is not described herein; the first current sequence segmentation acquisition method is the same as the first voltage sequence segmentation acquisition method, and all the first current sequence segments are acquired according to the same segmentation thought.
Segmenting the voltage monitoring data sequence and the current monitoring data sequence according to the nuclear density of the monitoring data respectively to obtain all second voltage sequence segments and second current sequence segments;
specifically, the method for acquiring the second voltage sequence segment and the second current sequence segment includes:
Taking the voltage monitoring data sequence and the current monitoring data sequence as a second monitoring sequence to be segmented; in the second monitoring sequence to be segmented, constructing two-dimensional monitoring data according to each monitoring data and the corresponding acquisition time, and clustering all the two-dimensional monitoring data based on a time point clustering algorithm to obtain all sequence segments of the second monitoring sequence to be segmented; all sequence segments of the voltage monitoring data sequence are all second voltage sequence segments and all sequence segments of the current monitoring data sequence are all second current sequence segments.
As one example: in the voltage monitoring data sequence, constructing two-dimensional monitoring data according to each voltage monitoring data and the corresponding acquisition time, and clustering all the two-dimensional monitoring data based on a time point clustering algorithm to obtain all second voltage sequence segments; and similarly, acquiring all second current sequence segments.
It should be noted that, the segmentation of the time sequence by the time point clustering algorithm is well known in the art, and is not described herein.
And under all sequence segments with the same sequence number, the superposition time periods of the first voltage sequence segment, the first current sequence segment, the second voltage sequence segment and the second current sequence segment are used as steady-state time periods. Referring to FIG. 3, a schematic diagram of a steady-state period according to one embodiment of the present invention is shown; in fig. 3, two adjacent circles in each segmentation result correspond to one sequence segment, and the corresponding time period of the double arrow is the coincidence time period of the four sequence segments under the same sequence number, namely the steady-state time period.
It should be noted that, in other embodiments of the present invention, the practitioner may also acquire different segmentation results by using other segmentation methods such as peak analysis, slope analysis, etc., and evaluate the steady-state period according to the different segmentation results.
Step S3, in each steady-state period, acquiring an electrical performance steady-state factor of the water pump according to the fluctuation similarity of the voltage monitoring data and the current monitoring data; in each steady-state period, acquiring a pipeline steady-state factor of the water pump pipeline according to fluctuation characteristics between the flow monitoring data and the pressure monitoring data; and acquiring the steady-state performance factor of the water pump in each steady-state period according to the running performance factor, the electrical performance steady-state factor and the pipeline steady-state factor.
Because the water outlet flow of the water pump is also influenced by the electric signal of the water pump motor, when the electric performance is stable, the flow monitoring data should also be relatively stable; therefore, after the steady-state time period is acquired, the embodiment of the invention firstly acquires the electrical performance steady-state factor of the water pump according to the fluctuation similarity of the voltage monitoring data and the current monitoring data in each steady-state time period; and then, the steady-state performance factor of the water pump is estimated by combining the performance characteristics of the water pump flow and the pressure under steady-state response.
Preferably, in one embodiment of the present invention, considering that the autocorrelation coefficients can provide the period of the internal structure of the sequences and the trend correlation information, if the autocorrelation coefficients between the two sequences are different, the similarity between the two sequences is lower; based on the above, the method for obtaining the electrical performance steady-state factor includes:
Calculating the sum of autocorrelation coefficients of the voltage monitoring data in each steady-state period as a first parameter; calculating the sum of autocorrelation coefficients of the current monitoring data in each steady-state period as a second parameter; and taking the difference between the first parameter and the second parameter in the same steady-state period as an electrical performance steady-state factor in the corresponding steady-state period.
The calculation formula of the electrical property steady-state factor is as follows:
; in the method, in the process of the invention, Is the firstThe electrical performance steady-state factors of the water pump in the steady-state periods; Sequence number of steady state period; A delay amount of autocorrelation coefficients of the monitored data sequence for each steady-state period; monitoring the total amount of data for each steady state period; monitoring the sign of the data sequence for the voltage during the steady state period; Monitoring the sign of the data sequence for current during a steady state period; is the first Voltage monitoring data sequences during steady state periodsA step autocorrelation coefficient; is the first Current monitoring data sequences during steady state periodsA step autocorrelation coefficient; Is a first parameter; is a second parameter; Is an absolute value sign.
It should be noted that, the calculation of the autocorrelation coefficient is already known in the prior art, and is not described herein.
In another embodiment of the present invention, the method for obtaining the electrical performance steady-state factor is: in each steady-state period, calculating a dynamic time warping (DYNAMIC TIME WARPING, DTW) distance between a steady-state sequence segment corresponding to voltage monitoring data and a steady-state sequence segment corresponding to current monitoring data, evaluating an electrical performance steady-state factor in each steady-state period, and specifically carrying out negative correlation mapping normalization on DTW (digital television) to adjust logic, wherein the smaller the DTW distance is, the higher the coordination similarity between the two sequences is, and the better the steady-state response state of the water pump is; the specific negative correlation is mapped into an exponential function for normalization.
Considering that the flow rate and the pressure at the water outlet pipeline of the water pump have certain correlation, the larger the flow rate is, the smaller the pressure of the water at the water outlet pipeline is, and the larger the pressure is, and the flow rate and the pressure are characterized by synchronous stability; evaluating the performance state of the water pump pipeline by analyzing the fluctuation condition between the flow and the pressure; under the condition that the electrical performance of the water pump is stable, if the flow is synchronous with the fluctuation of the pressure, the water pump is in a stable and normal running state, and the possibility of fault hidden danger is low; therefore, after the electrical performance steady-state factors are obtained, the pipeline steady-state factors of the water pump pipeline are further obtained according to fluctuation characteristics between the flow monitoring data and the pressure monitoring data; and then the running performance factor and the electric performance steady-state factor of the water pump are combined, and the running performance of the water pump in the steady-state period is comprehensively estimated.
Preferably, in one embodiment of the present invention, the variance of the data in the sequence can be evaluated in consideration of the coefficient of variation, and the fluctuation characteristics of the data are reflected on the sides; also taking into account peaks or valleys in the sequence may help evaluate fluctuations in the data; based on the above, the method for acquiring the pipeline steady-state factor comprises the following steps:
Referring to fig. 4, a flowchart of a method for obtaining a pipeline steady-state factor according to an embodiment of the present invention is shown:
Step S401, respectively obtaining a flow variation coefficient of the flow monitoring data in each steady-state period and a pressure variation coefficient of the pressure monitoring data in each steady-state period, and obtaining a first fluctuation characteristic parameter according to the flow variation coefficient and the pressure variation coefficient.
Specifically, the smaller the variation coefficient is, the more stable the data is, and if the flow variation coefficient and the pressure variation coefficient are smaller at the same time, the fluctuation of the flow variation coefficient and the pressure variation coefficient is stable, the larger the first fluctuation characteristic parameter is; therefore, in one embodiment of the present invention, the flow coefficient of variation and the pressure coefficient of variation are added and combined, and then the negative correlation mapping is performed to obtain the first fluctuation characteristic parameter.
It should be noted that, the coefficient of variation is already known in the prior art by those skilled in the art, and is not described herein in detail; specifically, the flow variation coefficient and the pressure variation coefficient are added and combined, then the opposite number is taken as an index taking a natural constant e as an index function, and the function value of the index function is taken as a negative correlation mapping result.
Step S402, all flow peaks and flow valleys in the flow monitoring data and all pressure peaks and pressure valleys in the pressure monitoring data are respectively obtained; and obtaining peak parameters according to all flow peaks and all pressure peaks, obtaining valley parameters according to all flow valleys and all pressure valleys, and obtaining second fluctuation characteristic parameters according to the difference between the peak parameters and the valley parameters.
Specifically, the more the difference between the peak value and the valley value is in the steady-state period, the more the fluctuation of the data is considered; in addition, the corresponding preprocessing operation is carried out on the flow monitoring data and the pressure monitoring data in the process of acquisition, and the numerical difference performance of the flow monitoring data and the pressure monitoring data is relatively smaller; therefore, when the sum of all flow peaks and all pressure peaks is larger, and the sum of all flow valleys and all pressure valleys is smaller, it is indicated that at least one of the flow monitoring data and the pressure monitoring data fluctuates relatively severely, and the running state of the water pump may not be stable enough.
Therefore, the larger the difference between the sum of all flow peaks and all pressure peaks and the sum of all flow valleys and all pressure valleys is, the unstable running state of the water pump is, and based on the difference, the second fluctuation characteristic parameter is obtained through the adjustment logic and normalization of the negative correlation mapping. The calculation formula of the second fluctuation characteristic parameter is as follows:
; in the method, in the process of the invention, Is the firstA second fluctuation feature parameter within a steady-state period; to be with natural constant An exponential function that is a base; symbols of the flow monitoring data sequence during the steady state period; Monitoring the sign of the data sequence for pressure in the steady state period; is the first The sum of all flow peaks in the flow monitoring data sequence over the steady-state periods; is the first The sum of all pressure peaks in the flow monitoring data sequence over the steady-state periods; is the first The sum of all flow dips in the sequence of flow monitoring data over the steady state periods; is the first The sum of all pressure dips in the flow monitoring data sequence over the steady state period; is a peak parameter; is the valley parameter.
It should be noted that, the acquisition of the peaks and the valleys is well known in the art, and is not described herein.
In other embodiments of the present invention, the practitioner may also take the sum of the median or the mean of all flow peaks and the peak points in all pressure peaks as the peak parameter, or the total number of peak points as the peak parameter; and similarly, obtaining the valley value parameter.
Step S403, obtaining pipeline steady-state factors in the corresponding steady-state time periods according to the first fluctuation characteristic parameters and the second fluctuation characteristic parameters, wherein the first fluctuation characteristic parameters and the second fluctuation characteristic parameters are positively correlated with the pipeline steady-state factors.
Specifically, considering that the first fluctuation characteristic parameter and the second fluctuation characteristic parameter both reflect the overall fluctuation condition between the flow and the pressure, in one embodiment of the invention, the pipeline steady-state factor in the corresponding steady-state period is obtained by multiplying and combining the first fluctuation characteristic parameter and the second fluctuation characteristic parameter.
In other embodiments of the present invention, the practitioner may further combine the first fluctuation feature parameter and the second fluctuation feature parameter by adding or weighting and summing the basic mathematical means to obtain the performance steady-state parameter; which are all prior art and are not described in detail herein.
The running performance factor reflects the stability and the good and bad conditions of the running environment of the motor in the running process of the water pump, the electric performance steady-state factor reflects the performance steady state of the motor of the water pump, the pipeline steady-state factor reflects the performance steady state of the water outlet pipeline of the water pump, after the three factors are obtained, the steady-state performance factor of the water pump can be obtained according to the running performance factor, the electric performance steady-state factor and the pipeline steady-state factor, and the higher the steady-state performance factor is, the more stable the running state and the higher the performance of the water pump are, and the lower the possibility of abnormal fault hidden danger of the water pump is.
Preferably, in one embodiment of the present invention, the method for obtaining the steady-state performance factor includes:
And obtaining the steady-state performance factor of the water pump according to the operation performance factor, the electrical performance steady-state factor and the pipeline steady-state factor.
Specifically, the running performance factor, the electrical performance steady-state factor and the pipeline steady-state factor are multiplied and combined to obtain the steady-state performance factor, and in other embodiments of the present invention, the implementer may also use other basic mathematical operations, such as addition, weighted summation or other relevant mapping means, which are not described herein.
S4, obtaining the abnormal score of each flow monitoring data, and adjusting the abnormal score of each flow monitoring data in each steady-state period according to the steady-state performance factors to obtain a corrected abnormal score; and carrying out abnormal monitoring on the running state of the water pump according to the corrected abnormal score of each flow monitoring data.
The higher the steady-state performance factor is in the steady-state operation stage of the water pump, the lower the possibility of abnormal fault hidden danger of the water pump is; therefore, after the steady-state performance factors of the water pump in each steady-state period are obtained, the abnormal score of each flow monitoring data is adjusted according to the steady-state performance factors, so that misjudgment of partial normal flow monitoring data caused by external factors such as non-pump body faults is reduced.
Specifically, in one embodiment of the present invention, firstly, the LOF value, i.e., the anomaly score, of each flow monitoring data is obtained through a local anomaly factor algorithm (Local Outlier Factor, LOF), and then, the LOF value of the flow monitoring data in the corresponding steady-state period is adjusted according to the steady-state performance factor, so as to obtain a corrected anomaly score.
Preferably, in one embodiment of the present invention, the method for obtaining the corrected anomaly score includes:
Normalizing the negative correlation mapping of the steady-state performance factors to obtain an adjustment weight; and reducing the abnormal score of the corresponding flow monitoring data according to the adjustment weight to obtain a corrected abnormal score.
The calculation formula for correcting the anomaly score is as follows:
; in the method, in the process of the invention, Is the firstThe first time period of steady stateCorrection anomaly scores for the individual flow monitoring data; is the first Steady state performance factors of the water pump within a steady state period; Is a standard normalization function; is the first The first time period of steady stateAbnormal scores for the individual flow monitoring data.
In the calculation formula for correcting the anomaly score, the anomaly score is calculated byAs adjusting weight, the purpose of reducing the abnormal score is achieved; in other embodiments of the present invention, the implementer may map the negative correlation of the steady-state performance factor into the exponential function as the adjustment weight, and may also use other negative correlation normalization means, which are not described herein.
The abnormal score can be corrected to reduce the abnormal score of the flow monitoring data when the running state of the water pump is good, and increase the abnormal score when the running state is poor, so that false alarm can be reduced when the running state of the water pump is good, the sensitivity of abnormal detection is improved when the running state is poor, the abnormal state affecting the performance of the water pump can be more accurately identified, the accuracy of abnormal detection is improved, and further, the potential faults of the water pump can be timely found and responded, so that a maintenance plan is optimized, and the reliability of a water pump system is improved.
After the corrected anomaly score of each flow monitoring data in the steady-state period is obtained, the anomaly score initially obtained by the other flow monitoring data in the flow monitoring data sequence can be further combined to obtain the final anomaly score of each flow monitoring data, wherein the final anomaly score of the flow monitoring data in the steady-state period is the corresponding corrected anomaly score, and the final anomaly score initially obtained by the other flow monitoring data is the initial anomaly score; the implementer can further set a threshold value, and when the abnormal score of the monitoring data exceeds the set threshold value, the running state of the water pump is considered to be abnormal, potential fault hidden danger can exist in the pump body, and related personnel can be early warned in time to process and optimize the maintenance plan of the water pump.
In summary, the embodiment of the invention obtains the monitoring data sequence of the water pump under each monitoring index; then, according to the fluctuation similarity and the corresponding fluctuation characteristics between the voltage monitoring data and the current monitoring data, obtaining the running performance factor of the water pump; further acquiring all steady-state time periods when the water pump is running; and in each steady-state period, acquiring an electrical performance steady-state factor of the water pump and a pipeline steady-state factor of a water pump pipeline, and further acquiring the steady-state performance factor of the water pump in each steady-state period, so as to adjust the abnormal fraction of each flow monitoring data in each steady-state period according to the steady-state performance factor, and carrying out abnormal monitoring on the running state of the water pump. According to the invention, the running environment and performance of the water pump are firstly evaluated by using the current and the voltage, then the steady state condition of the voltage and the current and the steady state condition of the flow and the pressure at the water outlet pipe are analyzed in the running steady state period of the water pump, and the steady state running performance of the water pump in the steady state period, namely the fault possibility of the water pump is evaluated, so that the abnormal detection consideration weight of flow data in the running steady state is reduced, the misjudgment of partial normal flow monitoring data caused by factors such as non-pump body faults is reduced, and the accuracy of monitoring the running state of the water pump is improved.
Based on the same inventive concept, the present invention also provides a device for monitoring the operation state of a large water pump, referring to fig. 5, which shows a schematic structural diagram of a device for monitoring the operation state of a large water pump according to an embodiment of the present invention, where the device includes a memory 501, a processor 502, and a computer program 503 stored in the memory and capable of running on the processor, and the processor implements the steps of the method for processing the operation state of the large water pump when executing the computer program.
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 (6)
1. A method for processing monitoring data of the running state of a large water pump, which is characterized by comprising the following steps:
Acquiring a monitoring data sequence of the water pump under each monitoring index, wherein the monitoring index at least comprises current and voltage of the water pump and flow and pressure at a water outlet pipe of the water pump;
Obtaining the running performance factor of the water pump according to the fluctuation similarity and the corresponding fluctuation characteristics between the voltage monitoring data sequence and the current monitoring data sequence; respectively segmenting the voltage monitoring data sequence and the current monitoring data sequence at least twice in different segmentation modes, and acquiring all steady-state time periods when the water pump operates according to segmentation results;
In each steady-state period, acquiring an electrical property steady-state factor of the water pump according to the fluctuation similarity of the voltage monitoring data sequence and the current monitoring data sequence, and acquiring a pipeline steady-state factor of the water pump pipeline according to the fluctuation characteristics between the flow monitoring data sequence and the pressure monitoring data sequence; acquiring a steady-state performance factor of the water pump in each steady-state period according to the running performance factor, the electrical performance steady-state factor and the pipeline steady-state factor;
Obtaining an abnormal score of each flow monitoring data in a flow monitoring data sequence, and adjusting the abnormal score of each flow monitoring data in each steady-state period according to the steady-state performance factor to obtain a corrected abnormal score; performing abnormal monitoring on the running state of the water pump according to the corrected abnormal score or the abnormal score of each flow monitoring data in the flow monitoring data sequence;
The method for acquiring the operation performance factor comprises the following steps:
Acquiring a coordination reference coefficient according to the sequence similarity condition between the voltage monitoring data sequence and the current monitoring data sequence;
Segmenting the voltage monitoring data sequence and the current monitoring data sequence respectively and sequencing labels in sequence; according to the fluctuation characteristics of the voltage sequence segments and the fluctuation characteristics of the current sequence segments under the same sequence number, obtaining fluctuation reference weights of the segments with the same sequence number in the corresponding time periods; acquiring performance parameters of the same sequence number segment corresponding to the time period according to the voltage monitoring data in the voltage sequence segment under the same sequence number and the current monitoring data in the current sequence segment; weighting and summing the performance parameters according to the fluctuation reference weight to obtain a performance reference coefficient;
Acquiring an operation performance factor according to the coordination reference coefficient and the performance reference coefficient; the coordination reference coefficient and the performance reference coefficient are positively correlated with the running performance factor;
The steady state period acquisition method comprises the following steps:
segmenting the voltage monitoring data sequence and the current monitoring data sequence according to the mutation of the monitoring data respectively to obtain all first voltage sequence segments and first current sequence segments;
Segmenting the voltage monitoring data sequence and the current monitoring data sequence according to the nuclear density of the monitoring data respectively to obtain all second voltage sequence segments and second current sequence segments;
Under all sequence segments with the same sequence number, overlapping time periods of the first voltage sequence segment, the first current sequence segment, the second voltage sequence segment and the second current sequence segment are used as steady-state time periods;
the method for acquiring the electrical performance steady-state factor comprises the following steps:
Calculating the sum of autocorrelation coefficients of the voltage monitoring data in each steady-state period as a first parameter; calculating the sum of autocorrelation coefficients of the current monitoring data in each steady-state period as a second parameter; taking the difference between the first parameter and the second parameter in the same steady-state period as an electrical performance steady-state factor in the steady-state period;
the method for acquiring the pipeline steady-state factor comprises the following steps:
Respectively obtaining a flow variation coefficient of flow monitoring data in each steady-state period and a pressure variation coefficient of the pressure monitoring data in each steady-state period, and obtaining a first fluctuation characteristic parameter according to the flow variation coefficient and the pressure variation coefficient, wherein the flow variation coefficient and the pressure variation coefficient are both in autocorrelation with the first fluctuation characteristic parameter;
Respectively acquiring all flow peaks and flow valleys in the flow monitoring data and all pressure peaks and pressure valleys in the pressure monitoring data; obtaining peak value parameters according to all the flow peak values and all the pressure peak values, obtaining valley value parameters according to all the flow valley values and all the pressure valley values, and obtaining second fluctuation characteristic parameters according to the differences between the peak value parameters and the valley value parameters;
and acquiring a pipeline steady-state factor in the steady-state period according to the first fluctuation characteristic parameter and the second fluctuation characteristic parameter, wherein the first fluctuation characteristic parameter and the second fluctuation characteristic parameter are positively correlated with the pipeline steady-state factor.
2. The method for processing the monitoring data of the running state of the large water pump according to claim 1, wherein the method for acquiring the first voltage sequence segment and the first current sequence segment comprises the following steps:
Taking the voltage monitoring data sequence and the current monitoring data sequence as a first monitoring sequence to be segmented;
When the differential value in the differential sequence is larger than a preset first threshold value, taking the sequence number of the differential value corresponding to the differential value in the differential sequence as the sequence number of a segmentation point of the first sequence to be segmented, segmenting the first sequence to be segmented according to the sequence number of the segmentation point, and obtaining all sequence segments of the first sequence to be segmented; all sequence segments of the voltage monitoring data sequence are all first voltage sequence segments, and all sequence segments of the current monitoring data sequence are all first current sequence segments.
3. The method for processing the monitoring data of the running state of the large water pump according to claim 1, wherein the method for acquiring the second voltage sequence segment and the second current sequence segment comprises the following steps:
Taking the voltage monitoring data sequence and the current monitoring data sequence as a second monitoring sequence to be segmented; constructing two-dimensional monitoring data according to each monitoring data and the corresponding acquisition time in the second monitoring sequence to be segmented, and clustering all the two-dimensional monitoring data based on a time point clustering algorithm to obtain all sequence segments of the second monitoring sequence to be segmented;
All sequence segments of the voltage monitoring data sequence are all second voltage sequence segments, and all sequence segments of the current monitoring data sequence are all second current sequence segments.
4. The method for processing the monitoring data of the running state of the large water pump according to claim 1, wherein the method for obtaining the anomaly score is as follows:
And obtaining an LOF value of each flow monitoring data in the flow monitoring data sequence by using an LOF algorithm, and taking the LOF value as an anomaly score.
5. The method for processing the running state monitoring data of the large water pump according to claim 1, wherein the method for acquiring the correction anomaly score comprises the following steps:
Normalizing the negative correlation mapping of the steady-state performance factors to obtain an adjustment weight; and reducing the abnormal score corresponding to the flow monitoring data in the steady-state period according to the adjustment weight to obtain a corrected abnormal score.
6. A large water pump operation state monitoring data processing device, the device comprising a memory, a processor and a computer program stored in the memory and operable on the processor, wherein the processor implements the steps of a large water pump operation state monitoring data processing method according to any one of claims 1 to 5 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410822683.2A CN118391249B (en) | 2024-06-25 | 2024-06-25 | Method and device for processing running state monitoring data of large water pump |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410822683.2A CN118391249B (en) | 2024-06-25 | 2024-06-25 | Method and device for processing running state monitoring data of large water pump |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118391249A CN118391249A (en) | 2024-07-26 |
| CN118391249B true CN118391249B (en) | 2024-08-16 |
Family
ID=92007074
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410822683.2A Active CN118391249B (en) | 2024-06-25 | 2024-06-25 | Method and device for processing running state monitoring data of large water pump |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118391249B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111608899A (en) * | 2020-04-28 | 2020-09-01 | 茂盟（上海）工程技术股份有限公司 | Water pump running state abnormity discrimination method based on efficiency analysis |
| CN117074783A (en) * | 2023-10-12 | 2023-11-17 | 国网吉林省电力有限公司通化供电公司 | Real-time monitoring and early warning method for overheat state of power equipment |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB9811874D0 (en) * | 1998-06-02 | 1998-07-29 | Univ Brunel | Information management system |
| CN117345596B (en) * | 2023-09-04 | 2024-04-16 | 安徽峻平节能科技有限公司 | Efficient energy-saving circulating water utilization system |
| CN118170004B (en) * | 2024-05-14 | 2024-07-16 | 吉林烟草工业有限责任公司 | Control method and system based on Internet of things |
-
2024
- 2024-06-25 CN CN202410822683.2A patent/CN118391249B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111608899A (en) * | 2020-04-28 | 2020-09-01 | 茂盟（上海）工程技术股份有限公司 | Water pump running state abnormity discrimination method based on efficiency analysis |
| CN117074783A (en) * | 2023-10-12 | 2023-11-17 | 国网吉林省电力有限公司通化供电公司 | Real-time monitoring and early warning method for overheat state of power equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118391249A (en) | 2024-07-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117349781B (en) | Intelligent diagnosis method and system for faults of transformer | |
| CN116955895B (en) | Transformer operation abnormity monitoring method and system | |
| CN117312997B (en) | Intelligent diagnosis method and system for power management system | |
| CN116879662A (en) | Transformer fault detection method based on data analysis | |
| CN111881594B (en) | Non-stationary signal state monitoring method and system for nuclear power equipment | |
| CN114580666A (en) | Multi-mode fusion motor intelligent maintenance system | |
| CN112418687B (en) | User electricity utilization abnormity identification method and device based on electricity utilization characteristics and storage medium | |
| CN116070163A (en) | Indoor harmful gas concentration anomaly monitoring data processing method | |
| CN117092980B (en) | Electrical fault detection control system based on big data | |
| CN118070195B (en) | Mining alternating current frequency converter abnormal data state monitoring system | |
| CN117591964B (en) | Electric power intelligent analysis method based on artificial intelligence | |
| CN113757093A (en) | Fault diagnosis method for flash steam compressor unit | |
| CN116345700A (en) | Energy consumption monitoring method and monitoring system for energy storage power station | |
| CN112308391A (en) | Real-time monitoring and anomaly detection method for equipment state based on neural network | |
| CN114323642A (en) | Wind turbine generator vibration data processing system and data dilution method | |
| CN116070103A (en) | Rotating equipment health identification method and equipment based on multiple measuring points and multiple indexes | |
| CN114878172A (en) | Method and system for extracting temperature signal characteristics of marine main engine cylinder | |
| CN117150244B (en) | Intelligent power distribution cabinet state monitoring method and system based on electrical parameter analysis | |
| CN118391249B (en) | Method and device for processing running state monitoring data of large water pump | |
| CN114088389A (en) | Data processing method and related device for gearbox | |
| CN118443307A (en) | Identification method and device for bearing life degradation starting point | |
| CN116629843B (en) | Remote early warning and maintenance decision support system of intelligent diesel generator set | |
| CN114837902B (en) | Health degree evaluation method, system, equipment and medium for wind turbine generator | |
| US11339763B2 (en) | Method for windmill farm monitoring | |
| CN114412447A (en) | Fault detection method and device for screw pump well |
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 |