CN113077008A - Peristaltic pump fault prediction method and device and peristaltic pump - Google Patents
Peristaltic pump fault prediction method and device and peristaltic pump Download PDFInfo
- Publication number
- CN113077008A CN113077008A CN202110420318.5A CN202110420318A CN113077008A CN 113077008 A CN113077008 A CN 113077008A CN 202110420318 A CN202110420318 A CN 202110420318A CN 113077008 A CN113077008 A CN 113077008A
- Authority
- CN
- China
- Prior art keywords
- peristaltic pump
- fault
- data
- abnormal data
- time
- 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.)
- Granted
Links
- 230000002572 peristaltic effect Effects 0.000 title claims abstract description 171
- 238000000034 method Methods 0.000 title claims abstract description 55
- 230000002159 abnormal effect Effects 0.000 claims abstract description 151
- 239000007788 liquid Substances 0.000 claims description 37
- 238000012545 processing Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 7
- 239000012530 fluid Substances 0.000 claims description 6
- 230000005856 abnormality Effects 0.000 claims description 4
- 230000002265 prevention Effects 0.000 claims description 3
- 230000036592 analgesia Effects 0.000 description 3
- 238000001802 infusion Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000000202 analgesic effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000002512 chemotherapy Methods 0.000 description 1
- 230000035606 childbirth Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000002980 postoperative effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/211—Selection of the most significant subset of features
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
Abstract
The application provides a peristaltic pump fault prediction method, a peristaltic pump fault prediction device and a peristaltic pump, wherein the method comprises the following steps: in the operation process of the peristaltic pump, acquiring a plurality of feature data based on a plurality of fault features, and acquiring prediction data according to the feature data; respectively determining various abnormal data representing any fault of the peristaltic pump in the prediction data of various fault characteristics based on historical statistical results; respectively calculating the probability of the corresponding fault of the peristaltic pump represented by various abnormal data of various fault characteristics; calculating the probability sum of the corresponding faults represented by the abnormal data in the prediction data of each fault characteristic aiming at each abnormal data; and determining whether the probability sum of each type of abnormal data is greater than a first preset threshold, and if so, feeding back fault alarm information corresponding to the type of abnormal data. The method and the device determine whether the peristaltic pump breaks down or not based on multiple fault characteristics, can quickly and accurately locate the abnormal state of the peristaltic pump, and greatly reduce the possibility of false alarm of faults.
Description
Technical Field
The application belongs to the technical field of peristaltic pumps, and particularly relates to a peristaltic pump fault prediction method and device and a peristaltic pump.
Background
The peristaltic pump is mainly used for conveying and filling various liquids. The method is widely applied to the fields of industry, medical use and the like. For example, a medical peristaltic pump is a common analgesic infusion pump, and is used in postoperative analgesia, childbirth analgesia, chemotherapy analgesia, and other scenes. According to the requirements of medical apparatus and instruments laws and regulations, the medical peristaltic pump needs to give an alarm on abnormal infusion conditions so as to avoid personal safety of users.
In the prior art, a single sensor (e.g., a pressure sensor, a bubble sensor, a flow sensor, a photoelectric sensor, etc.) is usually used to detect an abnormality of the infusion state, and a corresponding alarm process is performed according to the detection result, such as the pressure sensor, the bubble sensor, the flow sensor, the photoelectric sensor, etc. Therefore, the abnormal state cannot be quickly and accurately positioned by adopting the algorithm or the threshold value of the single type of data for judgment, and the false alarm is easily caused.
Disclosure of Invention
The application provides a peristaltic pump fault prediction method and device and a peristaltic pump, and whether the peristaltic pump has a fault is determined based on multiple fault characteristics, so that the abnormal state of the peristaltic pump is rapidly and accurately positioned, and the possibility of false alarm of the fault is greatly reduced.
The embodiment of the first aspect of the application provides a peristaltic pump failure prediction method, which includes:
respectively acquiring a plurality of feature data based on a plurality of fault features of the peristaltic pump in the operation process of the peristaltic pump, and acquiring a plurality of prediction data according to the feature data;
respectively determining various abnormal data capable of representing any fault of the peristaltic pump in the prediction data of various fault characteristics based on historical statistical results;
respectively calculating the probability that each kind of abnormal data of each fault characteristic represents the corresponding fault of the peristaltic pump; calculating the probability sum of the abnormal data representing corresponding faults in the prediction data of each fault characteristic aiming at each abnormal data;
and determining whether the probability sum of each type of abnormal data is greater than a first preset threshold, and if so, feeding back fault alarm information corresponding to the type of abnormal data.
Optionally, during the operation of the peristaltic pump, respectively acquiring a plurality of feature data based on a plurality of fault features of the peristaltic pump, including:
in the operation process of the peristaltic pump, collecting characteristic data of various fault characteristics of the peristaltic pump in real time, and forming a queue with the length of N; the N is a natural number;
and updating the queues respectively in a sliding filtering mode.
Optionally, the obtaining a plurality of prediction data according to the feature data includes:
processing the current queue by adopting a pulse interference prevention filtering mode at different moments when a driving motor of the peristaltic pump rotates for one circle, and calculating a first arithmetic mean value;
and calculating a second arithmetic mean value and a variance of the first arithmetic mean values corresponding to a plurality of moments and a difference value of the first arithmetic mean values corresponding to every two moments, and taking the second arithmetic mean value, the variance and the difference value as the prediction data.
Optionally, the calculating the probability that the peristaltic pump has the corresponding fault according to the various types of abnormal data of the various fault characteristics includes:
based on historical statistical results, determining the probability that various faults occur to the peristaltic pump at various moments and corresponding abnormal data are generated in the prediction data of various fault characteristics;
determining the times of generating corresponding abnormal data in the acquired prediction data of each fault characteristic;
determining the probability of the corresponding fault of the peristaltic pump represented by various abnormal data in the prediction data of each fault characteristic at each moment according to the following formula:
wherein n represents the nth time, i represents the abnormal type, j represents the fault characteristic item, (kn)ijRepresenting the occurrence frequency of i-type abnormal data in j fault characteristics acquired at the nth moment; p (xn)ijAnd the probability of generating abnormal data in the j fault characteristics when the i-type fault occurs at the nth time is determined based on historical statistics.
Optionally, determining the probability that each type of fault occurs in the peristaltic pump at each time and corresponding abnormal data is generated in the prediction data of each fault characteristic, including:
based on historical statistical results, respectively calculating the probability of occurrence of various faults at each moment and the probability of generating abnormal data in the prediction data of various fault characteristics at each moment;
based on the probability of the occurrence of each kind of fault at each moment and the probability of generating abnormal data in the prediction data of each fault characteristic at each moment, determining the probability of the occurrence of each kind of fault of the peristaltic pump at each moment and the generation of corresponding abnormal data in the prediction data of each fault characteristic according to the following formula:
p(xn)ij＝p(xn)j*p(xn)i
wherein, p (xn)ijIndicating the probability of the occurrence of i-type faults and the occurrence of abnormal data in j fault characteristics at the nth time, p (xn)iIndicates the probability of occurrence of i-type failure at the nth time, p (xn)jAnd (4) representing the probability of generating abnormal data in the prediction data of j fault characteristics at the nth time.
Optionally, the different times at least include a first time, a second time, a third time, a fourth time, a fifth time and a sixth time;
the first moment is the initial moment of rotation of the peristaltic pump;
the second moment is the moment when the pressure in the pipeline of the peristaltic pump in the current rotating ring reaches an extreme value for the first time;
the third moment is the moment when the fluid in the pipeline in the current rotating ring completely flows out for the first time;
the fourth moment is the moment when the liquid outlet valve of the pipeline in the current rotating ring is completely closed;
the fifth moment is the moment when the liquid inlet valve of the pipeline in the current rotating ring is completely opened;
and the sixth moment is the moment when the liquid inlet valve of the pipeline in the current rotating ring is completely closed again.
Optionally, the second arithmetic mean is a mean of first arithmetic means corresponding to the third time, the fourth time and the fifth time;
the variance is the variance of a first arithmetic mean value corresponding to the third time, the fourth time and the fifth time;
the difference includes a first difference, a second difference, a third difference and a fourth difference, the first difference is a difference of first arithmetic averages corresponding to the first time and the second time, the second difference is a difference of first arithmetic averages corresponding to the second time and the third time, the third difference is a difference of first arithmetic averages corresponding to the first time and the third time, and the fourth difference is a difference of first arithmetic averages corresponding to the first time and the sixth time.
Optionally, after determining whether the probability sum of each type of abnormal data is greater than a first preset threshold, the method further includes:
if not, determining whether the probability sum of the abnormal data is smaller than a second preset threshold value, wherein the second preset threshold value is smaller than the first preset threshold value;
if so, eliminating the abnormal data, and respectively acquiring a plurality of data again based on a plurality of fault characteristics of the peristaltic pump; and if not, respectively acquiring a plurality of prediction data again according to the characteristic data directly based on each fault characteristic of the peristaltic pump.
Optionally, the fault characteristics include at least any two of a line pressure of a fluid conveyed within the peristaltic pump, a supply voltage of the peristaltic pump, and a drive motor current of the peristaltic pump.
Optionally, after the feedback of the abnormality alarm information corresponding to the type of abnormal data, the method further includes: and clearing the abnormal data.
Embodiments of a second aspect of the present application provide a peristaltic pump failure prediction device, the device comprising:
the prediction data acquisition module is used for respectively acquiring a plurality of prediction data based on various fault characteristics of the peristaltic pump in the operation process of the peristaltic pump;
the abnormal data determining module is used for respectively determining various abnormal data which can represent any fault of the peristaltic pump in the prediction data of each fault characteristic based on the historical statistical result;
the fault probability calculation module is used for respectively calculating the probability that the corresponding fault occurs on the peristaltic pump represented by various abnormal data; calculating the probability sum of the abnormal data representing corresponding faults in the prediction data of each fault characteristic aiming at each abnormal data;
and the fault prediction module is used for determining whether the probability sum of each type of abnormal data is greater than a first preset threshold value, and if so, feeding back fault alarm information corresponding to the type of abnormal data.
Embodiments of the third aspect of the present application provide a peristaltic pump, which includes a main body, and further includes the above-mentioned peristaltic pump failure prediction device, where the peristaltic pump failure prediction device is configured to perform failure prediction on the main body.
The technical scheme provided in the embodiment of the application at least has the following technical effects or advantages:
according to the peristaltic pump fault prediction method provided by the embodiment of the application, a plurality of prediction data are respectively obtained based on a plurality of fault characteristics, abnormal data in the prediction data are determined based on historical statistical results, the probability sum of corresponding faults of the peristaltic pump is represented according to various abnormal data of each fault characteristic, and when the probability sum exceeds a preset first preset threshold value, the fault of the peristaltic pump is determined (when the probability sum does not exceed the preset first preset threshold value, the peristaltic pump is determined not to be in fault), so that the abnormal state of the peristaltic pump is rapidly and accurately positioned through parameters of the plurality of fault characteristics, and the possibility of false alarm of the fault is greatly reduced compared with a single-threshold prediction mode of a single type.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the application. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 is a schematic flow chart illustrating a method for predicting a peristaltic pump failure provided by an embodiment of the present application;
fig. 2 shows a schematic structural diagram of a peristaltic pump failure prediction device provided in an embodiment of the present application.
Detailed Description
Exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
It is to be noted that, unless otherwise specified, technical or scientific terms used herein shall have the ordinary meaning as understood by those skilled in the art to which this application belongs.
The following describes a peristaltic pump failure prediction method, a peristaltic pump failure prediction device and a peristaltic pump according to an embodiment of the present application with reference to the drawings.
The embodiment of the present application provides a method for predicting a failure of a peristaltic pump, which can be applied to a device for predicting a failure of a peristaltic pump shown in fig. 1, where the device may be a processor or a processing unit having data acquisition and processing functions, and this embodiment is not particularly limited as long as the method can be used to implement the method for predicting a failure of a peristaltic pump. The method is based on multiple fault characteristics, multiple prediction data are respectively obtained, the probability of the corresponding fault of the peristaltic pump is represented according to various abnormal data in the prediction data, and whether the peristaltic pump has the fault is determined, so that the abnormal state of the peristaltic pump is rapidly and accurately positioned, and the possibility of false alarm of the fault is greatly reduced. As shown in fig. 1, the method may include the steps of:
and step S1, respectively acquiring a plurality of characteristic data based on a plurality of fault characteristics of the peristaltic pump in the operation process of the peristaltic pump, and acquiring a plurality of prediction data according to the characteristic data.
The fault characteristics may include a line pressure of a fluid (which may be liquid or gas as required) conveyed in the peristaltic pump, a power supply voltage of the peristaltic pump, and a driving motor current of the peristaltic pump, and the fault characteristics may include any two of the line pressure, the power supply voltage, and the driving motor current. The characteristic data can be understood as original parameters (such as a few Pa, a few volts, a few milliamperes and the like) such as pipeline pressure, power supply voltage, driving motor current and the like generated in the operation process of the peristaltic pump. The prediction data may be understood as data that can be used for failure prediction obtained by processing the raw data on the basis of the raw data (specifically, refer to the following process for obtaining prediction data).
When the peristaltic pump fault prediction device carries out the peristaltic pump fault prediction, a plurality of characteristic data are collected based on each fault characteristic, and a plurality of prediction data are obtained according to the characteristic data, so that the fault misjudgment probability caused by single data error is avoided.
In a specific embodiment of this embodiment, the collected data may be processed to improve the accuracy of the data, and accordingly, in the operation process of the peristaltic pump in step S1, the respectively collecting a plurality of feature data based on a plurality of fault features of the peristaltic pump may include the following processing: in the operation process of the peristaltic pump, collecting characteristic data of various fault characteristics of the peristaltic pump in real time, and forming a queue with the length of N; n is a natural number; and updating the queue by adopting a sliding filtering mode.
In this embodiment, the peristaltic pump fault prediction device can acquire characteristic data of the peristaltic pump in real time so as to monitor the condition of the peristaltic pump in real time, and thus, the abnormal condition of the peristaltic pump can be found in time. In order to ensure the accuracy of the collected characteristic data and prevent failure prediction errors caused by errors of the data, the embodiment can perform data processing on the collected characteristic data. Specifically, N pieces of feature data may be continuously obtained first to form a queue with a length of N. Then, the characteristic data is subjected to sliding filtering, that is, according to the first-in first-out principle, newly-added characteristic data is placed at the tail of the queue, and the first old characteristic data (queue with the length of N) is thrown away.
Further, acquiring a plurality of prediction data from the feature data may include: processing a current queue by adopting a pulse interference prevention filtering mode at different moments when a driving motor of the peristaltic pump rotates for one circle, and calculating a first arithmetic mean value; and calculating a second arithmetic mean value and a variance of the first arithmetic mean values corresponding to the multiple moments and a difference value of the first arithmetic mean values corresponding to every two moments, and taking the second arithmetic mean value, the variance and the difference value as prediction data.
Wherein the first arithmetic mean is the mean of the N-2 data after the maximum and minimum values are removed from the queue. The difference of the first arithmetic mean values corresponding to two time instants may include a difference (absolute value) between the first arithmetic mean values obtained every two time instants. The variance of the first arithmetic mean corresponding to a plurality of time instants can be understood as the total variance of the first arithmetic mean corresponding to the plurality of time instants, or the total variance of the first arithmetic mean corresponding to any number (the any number is less than the total number of sampling time instants and is more than 2) of the plurality of time instants. For example, if the predicted data is obtained at three different times for each rotation of the driving motor of the peristaltic pump, the difference between the first arithmetic mean values corresponding to two times may include the difference between the first arithmetic mean values corresponding to any two times (a total of three difference values). The variance of the first arithmetic mean corresponding to a plurality of time points may be the total variance of the first arithmetic mean corresponding to three time points, or may be the variance of the first arithmetic mean corresponding to any two time points among three time points.
It should be noted that the above processing of the feature data, the obtaining of the predicted data, and the selection of the feature time are only preferred embodiments of the present embodiment, and the present embodiment is not limited thereto, for example, the predicted data in the predicted data may be part of the feature data obtained in real time (the data belonging to a certain threshold range may be set as abnormal data), or may be obtained by directly averaging all the current feature data; after the first arithmetic mean is obtained, other calculations may be performed to obtain the prediction data (such as a ratio, a quotient product, etc.). It is sufficient that the failure prediction can be performed based on abnormal data in the prediction data as the prediction data.
Specifically, the predicted data can be obtained at six characteristic moments when a driving motor of the peristaltic pump rotates for each circle, wherein the six characteristic moments can be a first moment, a second moment, a third moment, a fourth moment, a fifth moment and a sixth moment; the first moment is the initial moment of the rotation of the peristaltic pump; the second moment is the moment when the pressure in the pipeline of the peristaltic pump in the current rotating ring reaches an extreme value for the first time; the third moment is the moment when the fluid in the pipeline in the current rotating ring completely flows out for the first time; the fourth moment is the moment when the liquid outlet valve of the pipeline in the current rotating ring is completely closed; the fifth moment is the moment when the liquid inlet valve of the pipeline in the current rotating ring is completely opened; and the sixth moment is the moment when the liquid inlet valve of the pipeline in the current rotating ring is completely closed again.
In this embodiment, the peristaltic pump generally includes a driving motor, a rotating roller and a flexible pipe for conveying liquid, the rotating roller is connected to the driving motor and is used for squeezing the liquid in the pipe to make the liquid in the pipe flow from one end to the other end, and the pipe is further provided with a liquid outlet valve (for controlling the liquid in the pipe to flow out) and a liquid inlet valve (for controlling the pipe to suck the liquid through a siphon). In the operation process of the peristaltic pump, liquid can be conveyed twice in each rotation, and the specific process is as follows: when the valve is just started (at the first moment), the two valves are both opened, and liquid flows into the pipeline; along with the rotation of the driving motor, the rotating roller continuously extrudes the liquid in the pipeline, so that the pressure in the pipeline reaches the maximum (the liquid outlet valve is about to be opened) at a certain moment (a second moment); then the driving motor continues to rotate, and the liquid outlet valve is completely opened at a certain moment (a third moment), so that the liquid in the pipeline is completely extruded; then the driving motor continues to rotate, and the liquid outlet valve is completely closed at a certain moment (fourth moment), namely the siphon liquid inlet valve is opened, and the pressure is minimum at the moment; then the driving motor continues to rotate, and the siphon valve is completely opened at a certain moment (the fifth moment) to suck liquid; then the driving motor continues to rotate, the siphon valve is completely closed at a certain moment (sixth moment), the siphon is stopped, the driving motor stops working, the liquid amount in the pipeline is recovered to be basically the same as the liquid amount at the first moment, and the siphon amount is basically the same as the liquid amount at the first moment under the general condition.
In the embodiment, the prediction data is obtained at the six characteristic moments, and under a normal condition, the prediction data generally has a relatively stable threshold range, and the accuracy of calculation can be further ensured by performing fault prediction calculation through the prediction data.
It should be noted that different peristaltic pumps have different operation processes, and the peristaltic pump may also deliver liquid once, three times or more per rotation, and accordingly, the selection of the characteristic time may be selected according to a specific operation process of the peristaltic pump, that is, the characteristic time may not be limited to six, and may also be less than or more than six, and the time of the specific characteristic time is not limited to the six times, and may be any number of different times in one rotation of the peristaltic pump driving motor, and preferably, a plurality of times with characteristics (for example, times with maximum and minimum line pressure, or other extreme values or specific values) may be selected along with the flow of liquid in the line.
Accordingly, the second arithmetic mean may be a mean of the first arithmetic mean corresponding to the third time, the fourth time, and the fifth time. The variance may be a variance of the first arithmetic mean corresponding to the third time, the fourth time, and the fifth time. The difference may include a first difference, a second difference, a third difference, and a fourth difference, where the first difference is a difference between first arithmetic averages corresponding to the first time and the second time, the second difference is a difference between first arithmetic averages corresponding to the second time and the third time, the third difference is a difference between first arithmetic averages corresponding to the first time and the third time, and the fourth difference is a difference between first arithmetic averages corresponding to the first time and the sixth time.
In this embodiment, as can be seen from the operation process of the peristaltic pump, if the peristaltic pump is not abnormal, the correlation among the first time, the second time, and the third time is large, the correlation among the third time, the fourth time, and the fifth time is also large, and the data of the sixth time is similar to the data of the first time. Therefore, the variance and the difference are selected according to the correlations, so that the calculation accuracy is further improved, and the fault rate of fault prediction is reduced. However, the present embodiment is not limited to this, for example, the variance may be a variance of any two or more data (first arithmetic mean), and the difference may be a difference of any two data (first arithmetic mean).
And step S2, respectively determining various abnormal data capable of representing any fault of the peristaltic pump in the prediction data of each fault characteristic based on the historical statistical result.
The abnormal data may be data exceeding a corresponding threshold range in a normal case (corresponding to the predicted data at each time point, where the threshold range is corresponding). Possible failures of peristaltic pumps include, but are not limited to, the presence of air bubbles, the absence of liquid, liquid seepage, and the like. The classification of the abnormal data can represent the type of the fault of the peristaltic pump based on the abnormal data, for example, when bubbles exist in a pipeline of the peristaltic pump, the pressure in the pipeline is lower than that in the full liquid state, and the predicted data of the six characteristic moments can be caused to generate specific changes; similarly, when no liquid exists in the pipeline of the peristaltic pump, another specific change of the predicted data of the six characteristic moments is caused, and if the detected abnormal data just accords with the another specific change, the abnormal data can be classified as no-liquid abnormal data.
In this embodiment, various abnormal data under various fault characteristics at each time and the fault types and times that can be characterized by the abnormal data can be recorded, and time sequences are respectively formed, so as to facilitate subsequent calculation and fault tracing.
It should be noted that, for the same abnormal data, the abnormal data may be multiple types of abnormal data (for example, the abnormal data may be bubble abnormal data or liquid-free abnormal data), and corresponding to each fault type, there may be a corresponding probability (the probability may be obtained according to statistics of historical data).
Step S3, respectively calculating the probability of the corresponding fault of the peristaltic pump represented by various abnormal data of each fault characteristic; and calculating the probability sum of the corresponding faults represented by the abnormal data in the prediction data of each fault characteristic aiming at each abnormal data.
The fault corresponding to each abnormal data can be one type or multiple types, namely, each type of abnormal data can represent multiple types of faults of the peristaltic pump. In this embodiment, based on the classification of the abnormal data, the probability that the corresponding fault occurs in the peristaltic pump is represented by each kind of abnormal data in each kind of fault feature data is calculated for the prediction data obtained at different times.
In a specific implementation manner of this embodiment, the calculating the probability that each kind of abnormal data represents the corresponding failure of the peristaltic pump in step S3 may include the following steps: based on historical statistical results, determining the probability that various faults occur to the peristaltic pump at various moments and corresponding abnormal data are generated in the prediction data of various fault characteristics; determining the times of generating corresponding abnormal data in the acquired prediction data of each fault characteristic; determining the probability of the corresponding fault of the peristaltic pump represented by various abnormal data in the prediction data of each fault characteristic according to the following formula (1):
wherein n representsAt the nth time, i represents the abnormal type, j represents the fault feature item, (kn)ijRepresenting the occurrence frequency of i-type abnormal data in j fault characteristics acquired at the nth moment; p (xn)ijAnd the probability of generating abnormal data in the j fault characteristics when the i-type fault occurs at the nth time is determined based on historical statistics.
Further, the determining, based on the historical statistical result, the probability of generating abnormal data in the prediction data of each fault characteristic when the peristaltic pump has various faults includes: based on historical statistical results, respectively calculating the probability of occurrence of various faults at each moment and the probability of generating abnormal data in the prediction data of various fault characteristics at each moment; based on the probability of various faults occurring at each moment and the probability of generating abnormal data in the prediction data of various fault characteristics at each moment, determining the probability of generating various faults in the peristaltic pump at each moment and generating corresponding abnormal data in the prediction data of various fault characteristics according to the following formula (2):
p(xn)ij＝p(xn)j*p(xn)i (2)
wherein, p (xn)ijIndicating the probability of the occurrence of i-type faults and the occurrence of abnormal data in j fault characteristics at the nth time, p (xn)iIndicates the probability of occurrence of i-type failure at the nth time, p (xn)jAnd (4) generating abnormal data in the prediction data of j fault characteristics representing the probability of the occurrence of the corresponding fault at the nth time.
In this embodiment, feature data of various fault features and fault data can be continuously collected during the operation of the peristaltic pump, and prediction data can be obtained. Then, the probability of occurrence of each type of fault at each moment (which can be called prior probability) and the probability of generating abnormal data in the prediction data of each fault characteristic at each moment (which can be called class condition probability) are counted according to a statistical method, and based on the probability, the probability that each type of fault occurs in the peristaltic pump at each moment and corresponding abnormal data is generated in the prediction data of each fault characteristic can be calculated according to the formula (2). Therefore, when fault prediction is carried out, according to the frequency of occurrence of various abnormal data at each moment and various faults of the peristaltic pump corresponding to the abnormal data, the probability of the corresponding abnormal data is generated in the prediction data of each fault characteristic. The probability (which may be referred to as likelihood probability or conditional probability) that each type of fault occurs can be characterized by each abnormal data of each fault feature at each time according to the above formula (1) in further combination with the occurrence number of each type of abnormal data.
Step S4, determining whether the probability sum of each type of abnormal data is larger than a first preset threshold, and if so, feeding back fault alarm information corresponding to the type of abnormal data.
The first preset threshold is also a probability value, can be determined through historical data statistics, and can be reduced by a little for improving the early warning effect; alternatively, to improve the operating efficiency of the peristaltic pump, the first preset threshold is slightly increased. The fault alarm information corresponding to the abnormal data can be one or more of sound, light and electricity, for example, if the abnormal data is the abnormal data corresponding to the bubble fault, the peristaltic pump can be prompted to have the bubble fault, and the bubble fault can be broadcasted through voice and/or displayed on a screen.
In addition, after the abnormal alarm information corresponding to the abnormal data is fed back, the abnormal data (namely the time sequence of the recorded abnormal data) can be cleared, so that the data can be prevented from continuously interfering with subsequent fault prediction, and the peristaltic pump fault prediction device can continuously give an alarm and the like.
In another specific implementation of this embodiment, after determining whether the sum of the probabilities of each type of abnormal data is greater than the first preset threshold, the following process may be further included: if the probability sum of a certain type of abnormal data is smaller than a first preset threshold, determining whether the probability sum of the certain type of abnormal data is smaller than a second preset threshold, wherein the second preset threshold is smaller than the first preset threshold; if so, the detected data is normal, and no abnormal data exists, so that the record of the abnormal data can be cleared (for example, the data is marked with an abnormal mark before, and the abnormal mark is not needed at present), and the data are respectively obtained again based on a plurality of fault characteristics of the peristaltic pump; if not, the detected data may be normal data or abnormal data, so that a plurality of predicted data can be obtained respectively according to the characteristic data again based on various fault characteristics of the peristaltic pump to make further determination.
The method for predicting the fault of the peristaltic pump provided by this embodiment includes obtaining a plurality of pieces of prediction data based on a plurality of fault features, determining abnormal data in the prediction data based on historical statistical results, characterizing the sum of probabilities of the peristaltic pump having corresponding faults according to various pieces of abnormal data of each fault feature, and determining that the peristaltic pump has a fault when the sum of the probabilities exceeds a preset first preset threshold (determining that the peristaltic pump has not failed when the sum of the probabilities does not exceed the preset first preset threshold).
Based on the same concept of the above peristaltic pump failure prediction method, this embodiment further provides a peristaltic pump failure prediction device, as shown in fig. 2, the device includes:
the prediction data acquisition module is used for respectively acquiring a plurality of prediction data based on various fault characteristics of the peristaltic pump in the operation process of the peristaltic pump;
the abnormal data determining module is used for respectively determining various abnormal data which can represent any fault of the peristaltic pump in the prediction data of each fault characteristic based on the historical statistical result;
the fault probability calculation module is used for respectively calculating the probability of the corresponding fault of the peristaltic pump represented by various abnormal data; calculating the probability sum of the abnormal data representing corresponding faults in the prediction data of each fault characteristic aiming at each abnormal data;
and the fault prediction module is used for determining whether the probability sum of each type of abnormal data is greater than a first preset threshold value, and if so, feeding back fault alarm information corresponding to the type of abnormal data.
The peristaltic pump failure prediction device provided by this embodiment can implement the peristaltic pump failure prediction method, and can obtain a plurality of prediction data based on a plurality of failure features, determine abnormal data in the prediction data based on historical statistics, then represent the probability sum of the corresponding failures of the peristaltic pump according to various abnormal data of each failure feature, and determine that the peristaltic pump fails when the probability sum exceeds a preset first preset threshold (determine that the peristaltic pump fails when the probability sum does not exceed the preset first preset threshold), so that the abnormal state of the peristaltic pump is quickly and accurately located through parameters of the plurality of failure features, and the possibility of false failure is greatly reduced compared with a single-type single-threshold prediction mode.
Based on the same concept of the peristaltic pump failure prediction method, the embodiment further provides a peristaltic pump, which comprises a main body and the peristaltic pump failure prediction device, wherein the peristaltic pump failure prediction device can be used for performing failure prediction on the main body.
The peristaltic pump provided by this embodiment includes the above-mentioned peristaltic pump failure prediction device, can realize the beneficial effect of above-mentioned peristaltic pump failure prediction device at least, and no longer repeated here.
It should be noted that the above-mentioned embodiments illustrate rather than limit the application, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The application may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.
The above description is only for the preferred embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (12)
1. A method of peristaltic pump failure prediction, the method comprising:
respectively acquiring a plurality of feature data based on a plurality of fault features of the peristaltic pump in the operation process of the peristaltic pump, and acquiring a plurality of prediction data according to the feature data;
respectively determining various abnormal data capable of representing any fault of the peristaltic pump in the prediction data of various fault characteristics based on historical statistical results;
respectively calculating the probability that each kind of abnormal data of each fault characteristic represents the corresponding fault of the peristaltic pump; calculating the probability sum of the abnormal data representing corresponding faults in the prediction data of each fault characteristic aiming at each abnormal data;
and determining whether the probability sum of each type of abnormal data is greater than a first preset threshold, and if so, feeding back fault alarm information corresponding to the type of abnormal data.
2. The method of claim 1, wherein separately collecting a plurality of characteristic data based on a plurality of fault characteristics of the peristaltic pump during operation of the peristaltic pump comprises:
in the operation process of the peristaltic pump, collecting characteristic data of various fault characteristics of the peristaltic pump in real time, and forming a queue with the length of N; the N is a natural number;
and updating the queues respectively in a sliding filtering mode.
3. The method of claim 2, wherein said obtaining a plurality of prediction data from said feature data comprises:
processing the current queue by adopting a pulse interference prevention filtering mode at different moments when a driving motor of the peristaltic pump rotates for one circle, and calculating a first arithmetic mean value;
and calculating a second arithmetic mean value and a variance of the first arithmetic mean values corresponding to a plurality of moments and a difference value of the first arithmetic mean values corresponding to every two moments, and taking the second arithmetic mean value, the variance and the difference value as the prediction data.
4. The method according to claim 3, wherein the calculating the respective types of abnormal data of the failure features respectively represents the probability of the corresponding failure of the peristaltic pump, and comprises:
based on historical statistical results, determining the probability that various faults occur to the peristaltic pump at various moments and corresponding abnormal data are generated in the prediction data of various fault characteristics;
determining the times of generating corresponding abnormal data in the acquired prediction data of each fault characteristic;
determining the probability of the corresponding fault of the peristaltic pump represented by various abnormal data in the prediction data of each fault characteristic at each moment according to the following formula:
wherein n represents the nth time, i represents the abnormal type, j represents the fault characteristic item, (kn)ijRepresenting the occurrence frequency of i-type abnormal data in j fault characteristics acquired at the nth moment; p (xn)ijAnd the probability of generating abnormal data in the j fault characteristics when the i-type fault occurs at the nth time is determined based on historical statistics.
5. The method according to claim 4, wherein the determining, based on the historical statistics, the probability that each type of fault occurs in the peristaltic pump at each time and corresponding abnormal data is generated in the prediction data of each fault characteristic comprises:
based on historical statistical results, respectively calculating the probability of occurrence of various faults at each moment and the probability of generating abnormal data in the prediction data of various fault characteristics at each moment;
based on the probability of the occurrence of each kind of fault at each moment and the probability of generating abnormal data in the prediction data of each fault characteristic at each moment, determining the probability of the occurrence of each kind of fault of the peristaltic pump at each moment and the generation of corresponding abnormal data in the prediction data of each fault characteristic according to the following formula:
p(xn)ij＝p(xn)j*p(xn)i
wherein, p (xn)ijIndicating the probability of the occurrence of i-type faults and the occurrence of abnormal data in j fault characteristics at the nth time, p (xn)iIndicates the probability of occurrence of i-type failure at the nth time, p (xn)jAnd (4) representing the probability of generating abnormal data in the prediction data of j fault characteristics at the nth time.
6. The method of claim 3, wherein the different time instants comprise at least a first time instant, a second time instant, a third time instant, a fourth time instant, a fifth time instant, and a sixth time instant;
the first moment is the initial moment of rotation of the peristaltic pump;
the second moment is the moment when the pressure in the pipeline of the peristaltic pump in the current rotating ring reaches an extreme value for the first time;
the third moment is the moment when the fluid in the pipeline in the current rotating ring completely flows out for the first time;
the fourth moment is the moment when the liquid outlet valve of the pipeline in the current rotating ring is completely closed;
the fifth moment is the moment when the liquid inlet valve of the pipeline in the current rotating ring is completely opened;
and the sixth moment is the moment when the liquid inlet valve of the pipeline in the current rotating ring is completely closed again.
7. The method according to claim 6, wherein the second arithmetic mean is a mean of the first arithmetic mean corresponding to the third time, the fourth time and the fifth time;
the variance is the variance of a first arithmetic mean value corresponding to the third time, the fourth time and the fifth time;
the difference includes a first difference, a second difference, a third difference and a fourth difference, the first difference is a difference of first arithmetic averages corresponding to the first time and the second time, the second difference is a difference of first arithmetic averages corresponding to the second time and the third time, the third difference is a difference of first arithmetic averages corresponding to the first time and the third time, and the fourth difference is a difference of first arithmetic averages corresponding to the first time and the sixth time.
8. The method according to claim 1, wherein after determining whether the sum of the probabilities of each type of abnormal data is greater than a first preset threshold, further comprising:
if not, determining whether the probability sum of the abnormal data is smaller than a second preset threshold value, wherein the second preset threshold value is smaller than the first preset threshold value;
if so, eliminating the abnormal data, and respectively acquiring a plurality of data again based on a plurality of fault characteristics of the peristaltic pump; and if not, respectively acquiring a plurality of prediction data again according to the characteristic data directly based on each fault characteristic of the peristaltic pump.
9. The method of any of claims 1-8, wherein the fault characteristic includes at least any two of a line pressure of a fluid conveyed within the peristaltic pump, a supply voltage of the peristaltic pump, and a drive motor current of the peristaltic pump.
10. The method according to claim 1, wherein after the feedback of the abnormality warning information corresponding to the abnormality data, the method further comprises: and clearing the abnormal data.
11. A peristaltic pump failure prediction device, the device comprising:
the prediction data acquisition module is used for respectively acquiring a plurality of prediction data based on various fault characteristics of the peristaltic pump in the operation process of the peristaltic pump;
the abnormal data determining module is used for respectively determining various abnormal data which can represent any fault of the peristaltic pump in the prediction data of each fault characteristic based on the historical statistical result;
the fault probability calculation module is used for respectively calculating the probability that the corresponding fault occurs on the peristaltic pump represented by various abnormal data; calculating the probability sum of the abnormal data representing corresponding faults in the prediction data of each fault characteristic aiming at each abnormal data;
and the fault prediction module is used for determining whether the probability sum of each type of abnormal data is greater than a first preset threshold value, and if so, feeding back fault alarm information corresponding to the type of abnormal data.
12. A peristaltic pump comprising a body, further comprising a peristaltic pump failure prediction device of claim 11, the peristaltic pump failure prediction device to failure predict the body.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110420318.5A CN113077008B (en) | 2021-04-19 | Peristaltic pump fault prediction method and device and peristaltic pump |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110420318.5A CN113077008B (en) | 2021-04-19 | Peristaltic pump fault prediction method and device and peristaltic pump |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113077008A true CN113077008A (en) | 2021-07-06 |
| CN113077008B CN113077008B (en) | 2024-05-14 |
Family
ID=
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114295936A (en) * | 2021-12-31 | 2022-04-08 | 合肥联信电源有限公司 | Power grid voltage detection system and method applied to static switch |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20080102640A (en) * | 2007-05-21 | 2008-11-26 | 삼성전자주식회사 | Pump fault prediction device and punp fault prediction method |
| CN105468917A (en) * | 2015-12-01 | 2016-04-06 | 北京无线电计量测试研究所 | Pipeline fault prediction method and apparatus |
| CN108537386A (en) * | 2018-04-13 | 2018-09-14 | 上海财经大学 | Maintenance forecast device based on history maintenance record |
| CN110623679A (en) * | 2019-09-20 | 2019-12-31 | 四川九八村信息科技有限公司 | System and method for predicting equipment failure of plasma apheresis machine |
| CN112395178A (en) * | 2020-11-18 | 2021-02-23 | 河南辉煌城轨科技有限公司 | Equipment fault prediction method |
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20080102640A (en) * | 2007-05-21 | 2008-11-26 | 삼성전자주식회사 | Pump fault prediction device and punp fault prediction method |
| CN105468917A (en) * | 2015-12-01 | 2016-04-06 | 北京无线电计量测试研究所 | Pipeline fault prediction method and apparatus |
| CN108537386A (en) * | 2018-04-13 | 2018-09-14 | 上海财经大学 | Maintenance forecast device based on history maintenance record |
| CN110623679A (en) * | 2019-09-20 | 2019-12-31 | 四川九八村信息科技有限公司 | System and method for predicting equipment failure of plasma apheresis machine |
| CN112395178A (en) * | 2020-11-18 | 2021-02-23 | 河南辉煌城轨科技有限公司 | Equipment fault prediction method |
Non-Patent Citations (2)
| Title |
|---|
| BOJIANG LIU: "Research on Fault Prediction and Health Management Technology Based on Machine Learning", 《2019 INTERNATIONAL CONFERENCE ON QUALITY, RELIABILITY, RISK, MAINTENANCE, AND SAFETY ENGINEERING (QR2MSE)》, pages 789 - 794 * |
| 马健 等: "输液泵的故障分析与风险控制", 《中国医学装备》, vol. 17, no. 12, pages 208 - 210 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114295936A (en) * | 2021-12-31 | 2022-04-08 | 合肥联信电源有限公司 | Power grid voltage detection system and method applied to static switch |
| CN114295936B (en) * | 2021-12-31 | 2023-08-22 | 合肥联信电源有限公司 | Power grid voltage detection system and method applied to static switch |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20190128288A1 (en) | Detection apparatus, detection method, and computer readable medium | |
| JP4762088B2 (en) | Process abnormality diagnosis device | |
| US20150219530A1 (en) | Systems and methods for event detection and diagnosis | |
| US9314742B2 (en) | Method and system for reverse osmosis predictive maintenance using normalization data | |
| EP1868145A1 (en) | Reverse failure analysis method and apparatus for diagnostic testing | |
| CN109469896B (en) | Industrial boiler fault diagnosis method and system based on time series analysis | |
| US10378332B2 (en) | Monitoring a component used in a well operation | |
| US7343267B1 (en) | Methods and systems for statistical based monitoring and power management | |
| JP2014237071A (en) | Plant failure warning sign detecting and recovering support system | |
| US20220212770A1 (en) | System and method for performing diagnostics of a water system on-board a watercraft | |
| JP2004240642A (en) | Maintenance support device for plant equipment | |
| CN111414955B (en) | Intelligent detection method and device for petroleum drilling lost circulation overflow and electronic equipment | |
| CN106678069A (en) | Detection method for preventing surge of centrifugal compressor from occurring | |
| CN111538723A (en) | Monitoring data processing method and device and electronic equipment | |
| CN112581719A (en) | Semiconductor packaging process early warning method and device based on time sequence generation countermeasure network | |
| CN112105939A (en) | Device for monitoring an actuator system, method for providing a device for monitoring an actuator system, and method for monitoring an actuator system | |
| CN117131110A (en) | Method and system for monitoring dielectric loss of capacitive equipment based on correlation analysis | |
| CN113077008A (en) | Peristaltic pump fault prediction method and device and peristaltic pump | |
| CN113077008B (en) | Peristaltic pump fault prediction method and device and peristaltic pump | |
| CN105466707A (en) | Method, system and computer program for fault detection in a machine | |
| Asadzadeh et al. | AFT regression-adjusted monitoring of reliability data in cascade processes | |
| JP2018193896A (en) | Hydraulic pump abnormality diagnosis device | |
| JP6937698B2 (en) | Processing tool monitor | |
| CN110165969B (en) | Control device for machine tool | |
| JP2008140109A (en) | Process operation support apparatus, and process operation support method |
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 |