CN117948269A - Water pump unit running state on-line monitoring method, medium and system - Google Patents
Water pump unit running state on-line monitoring method, medium and system Download PDFInfo
- Publication number
- CN117948269A CN117948269A CN202311775945.6A CN202311775945A CN117948269A CN 117948269 A CN117948269 A CN 117948269A CN 202311775945 A CN202311775945 A CN 202311775945A CN 117948269 A CN117948269 A CN 117948269A
- Authority
- CN
- China
- Prior art keywords
- water pump
- image
- state
- data
- abnormal
- 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.)
- Pending
Links
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 158
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000012544 monitoring process Methods 0.000 title claims abstract description 47
- 230000002159 abnormal effect Effects 0.000 claims abstract description 99
- 238000013507 mapping Methods 0.000 claims abstract description 56
- 230000001413 cellular effect Effects 0.000 claims abstract description 48
- 230000005856 abnormality Effects 0.000 claims abstract description 41
- 238000011161 development Methods 0.000 claims abstract description 25
- 238000012545 processing Methods 0.000 claims abstract description 21
- 238000012423 maintenance Methods 0.000 claims abstract description 14
- 239000003086 colorant Substances 0.000 claims abstract description 8
- 230000008859 change Effects 0.000 claims description 26
- 230000007704 transition Effects 0.000 claims description 15
- 238000009792 diffusion process Methods 0.000 claims description 11
- 230000002441 reversible effect Effects 0.000 claims description 7
- 238000013461 design Methods 0.000 claims description 5
- 230000008033 biological extinction Effects 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 4
- 238000007726 management method Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 14
- 238000005070 sampling Methods 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000000007 visual effect Effects 0.000 description 5
- 230000006399 behavior Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 4
- 230000010354 integration Effects 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011897 real-time detection Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 241001270131 Agaricus moelleri Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013506 data mapping Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention provides a water pump unit running state on-line monitoring method, medium and system, belonging to the technical field of operation and maintenance management, wherein the water pump unit running state on-line monitoring method comprises the following steps: collecting operation data and total water delivery of a water pump unit in real time; setting an abnormal operation threshold; establishing a plane grid; mapping the collected operation data into RGB colors, and filling the plane grids to form a monitoring mapping image; carrying out gray scale processing on the monitoring mapping image by adopting an image processing function to obtain a gray scale image; marking abnormal points and areas approaching to the abnormality on the gray level image according to the set operation abnormality threshold, namely, an alarm area; establishing a cellular automaton model on the gray level image, and calculating an alarm region development rule; the cellular automaton model predicts the gray level image variation to obtain a predicted image; the data is output to operation and maintenance personnel, and the running state of the water pump is intuitively and vividly presented through images, so that the abnormal development situation can be conveniently judged.
Description
Technical Field
The invention belongs to the technical field of operation and maintenance management, and particularly relates to an on-line monitoring method, medium and system for the running state of a water pump unit.
Background
The water pump unit is an important device widely used in various industrial and civil facilities, and the running state of the water pump unit directly influences the efficiency of water resource conveying and supplying. Therefore, the running process of the water pump needs to be monitored in real time, and the good equipment state is ensured. Traditional water pump state monitoring mostly adopts the manual mode, and the periodic inspection operation data judges whether unusual. The method relies on experience, has low efficiency and cannot reflect the state of the water pump in real time. Currently, sensors are commonly used for collecting operation parameters in real time, a threshold value is set for key parameters in a monitoring center, and an alarm is sent out when the parameters exceed the threshold value. Such methods can quantitatively detect problems, but are not comprehensive enough due to single signals. The water pump failure is progressive and cannot be suggested to trend toward abnormal subtle changes. Once a problem occurs, the loss is enormous.
The existing systems monitor the state of the water pump by adopting a multi-source information fusion mode, comprehensively analyze each operation parameter to judge abnormality, and have good effect. But the information expression mode is single, the intuitiveness is not strong, and the abnormal development situation is inconvenient to directly observe. Meanwhile, most of the prior art is based on real-time detection, and state changes in a future period of time cannot be accurately predicted.
Disclosure of Invention
In view of the above, the invention provides a method, medium and system for on-line monitoring of the running state of a water pump unit, which can effectively solve the problems of insufficient expressivity and insufficient predictability in the existing water pump state monitoring technology.
The invention is realized in the following way:
the first aspect of the invention provides an on-line monitoring method for the running state of a water pump unit, which comprises the following steps:
S10, collecting operation data and total water delivery quantity of a water pump unit in real time, wherein the operation data comprise vibration, temperature, pressure and water outlet flow rate;
s20, setting an operation abnormality threshold, wherein the operation abnormality threshold comprises a vibration abnormality threshold, a temperature abnormality threshold, a pressure abnormality threshold and a water outlet flow speed abnormality threshold, and judging that the operation is abnormal when any one of the operation data exceeds the abnormality threshold;
s30, establishing a plane grid by taking the acquisition time as a horizontal axis and the total water delivery as a vertical axis;
S40, mapping the collected operation data into RGB colors, and filling the plane grid to form a monitoring mapping image;
S50, carrying out gray scale processing on the monitoring mapping image by adopting an image processing function to obtain a gray scale image;
S60, marking abnormal points and areas approaching to the abnormal states on the gray level image according to a set operation abnormal threshold, namely an alarm area;
s70, establishing a cellular automaton model on the gray level image, and calculating rules of the cellular automaton as alarm region development rules according to variation rules of abnormal points and alarm regions on the gray level image;
S80, predicting gray level image variation in a specified time period according to the cellular automaton model and the estimated alarm region development rule to obtain a gray level image in the specified time period, and marking the gray level image as a predicted image;
And S90, outputting the abnormal points of the gray level image, the alarm region, the abnormal points in the predicted image, the corresponding moments of the alarm region and the operation data to operation and maintenance personnel.
On the basis of the technical scheme, the on-line monitoring method for the running state of the water pump unit can be improved as follows:
The specific steps of S10 are as follows: collecting operation data and total water delivery of a water pump unit in real time, wherein the operation data comprise vibration, temperature, pressure and water outlet flow rate; and setting corresponding sensors and connecting the acquisition system with the data processing device to finish data transmission.
The data processor refers to a device for classifying, merging, storing, retrieving, calculating and other operations on data, and comprises an accounting machine, a table making machine, a card processing machine and an automatic computer for storing programs. The data processor comprises a central processing unit, a main memory, an input-output interface and peripheral equipment.
Further, the specific step of S20 is as follows: setting an abnormal threshold value which is a parameter numerical range representing the normal working state of the water pump unit, and setting the threshold value after comprehensively considering the design parameters of the water pump, the normal working tolerance range, the data acquisition precision and the fault state parameter change condition.
Further, the specific steps of S30 are as follows:
Establishing a parameter two-dimensional grid, namely a plane grid, taking the acquisition time as a horizontal axis and the total water delivery as a vertical axis;
mapping state data related to the acquisition time and the total water delivery amount to the planar grid;
collecting the state data in real time, and mapping the state data to the planar grid;
And analyzing the change trend of the total water delivery and the relation with the acquisition time by observing the change of the state data on the plane grid.
Further, the specific step of S40 is as follows:
Defining a nonlinear mapping function, wherein the mapping function is used for mapping the state parameter values to corresponding RGB color values and determining the mapping relation between the state parameter values and the RGB color values;
for each data point, calculating a corresponding RGB color value through the mapping function according to the state parameter value of the data point;
the RGB color values for each of the data points are used as input to color the corresponding location of the image of the planar grid using image processing techniques.
Further, the specific step of S60 is as follows: and judging the data points exceeding the threshold value on the gray level image as abnormal points, and selecting the region with larger change gradient to judge the region as an abnormal trend region.
Further, the specific step of S70 is as follows:
Identifying and tracking abnormal points in the image;
Analyzing the diffusion mode and trend of the abnormal point by observing the diffusion condition of the abnormal point in the image;
Generating an extinction and diffusion behavior according to the analyzed abnormal points, and reversely pushing a state transition rule of the cellular automaton;
After the state transition rule of the cellular automaton is obtained through reverse thrust, a cellular automaton model is built according to the rule obtained through reverse thrust and the initial state.
Cellular automata is a discrete model whose state transition rules typically consist of a set of logical rules that describe the possible states of each cell in the next time step. By observing the behavior patterns of outliers, the state transition rules of cellular automata can be deduced.
Further, the specific steps of S80 are as follows: and (3) iteratively calculating cell state update based on a cell automaton rule, and advancing to a specified time period to obtain an abnormal development prediction image.
The invention provides a computer readable storage medium, wherein the computer readable storage medium stores program instructions, and the program instructions are used for the online monitoring method of the running state of a water pump unit when running.
The invention provides an on-line monitoring system for the running state of a water pump unit, which comprises the computer readable storage medium.
Compared with the prior art, the on-line monitoring method, medium and system for the running state of the water pump unit have the beneficial effects that: according to the method, a two-dimensional mapping model of time-water delivery is established, multi-source state information is mapped into images in real time in a color mapping mode, and compared with direct display data, the technology enables the running state of the water pump to be visually and vividly presented through the images, so that the state fine change is easier to detect, particularly, the color change can reflect the state evolution process along with the running time, and the abnormal development situation can be judged conveniently;
Meanwhile, the method realizes the development trend prediction of the water pump running abnormal region for a period of time by constructing the cellular automaton model and determining the state transition rule thereof, and the prediction function of the invention can give out early warning to the possible abnormal condition in advance, so that the maintenance of the water pump is more timely, the system fault can be discovered and treated early, and the loss caused by equipment abnormality can be greatly reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments of the present invention will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
As shown in fig. 1, a flowchart of a first embodiment of an on-line monitoring method for an operation state of a water pump unit according to a first aspect of the present invention is provided, in this embodiment, the method includes the following steps:
S10, collecting operation data and total water delivery quantity of a water pump unit in real time, wherein the operation data comprise vibration, temperature, pressure and water outlet flow rate;
The method comprises the following specific steps: collecting operation data and total water delivery of a water pump unit in real time, wherein the operation data comprise vibration, temperature, pressure and water outlet flow rate; and setting corresponding sensors and connecting the acquisition system with the data processing device to finish data transmission.
In this step, the operation data means various parameter signals capable of reflecting the operation state of the water pump unit. These parameter signals may include vibration, temperature, pressure, outlet flow rate, etc. Specifically:
(1) Vibration refers to vibration signals generated by the water pump unit and pipelines thereof in the running process. Vibration sensors can be arranged at the rotor part and the water inlet and outlet interfaces of the water pump unit to collect vibration signals.
(2) The temperature refers to the temperature signal of the water pump unit body and water inlet and outlet. And a temperature measuring device is arranged on the surface of the shell of the water pump unit and the water inlet and outlet to collect temperature signals.
(3) The pressure refers to the water inlet and outlet pressure of the water pump unit. Pressure signals are acquired by arranging pressure measuring devices at the water inlet end and the water outlet end of the water pump unit.
(4) The water outlet flow rate refers to the water flow rate at the output end of the water pump unit. A flow measurement device may be used to measure the water flow rate of the water pump output conduit.
The four signals can reflect the working state of the water pump unit from different side surfaces, and comprise vibration and temperature which are directly related to the working condition of the unit body, and pressure and flow which are related to the water supply state of the water pump. State parameter signals of different types are collected, so that the state of the water pump unit can be monitored more comprehensively.
In addition, the total water delivery of the water pump, namely the accumulated water output by the water pump in unit time, needs to be collected in real time. This can directly reflect the water supply efficiency of the water pump. The total water delivery can be calculated by adopting a flow integration method.
It should be noted that, in order to be able to detect the operation state of the water pump assembly in real time, the acquisition frequency of the state parameter needs to be set to a high value, for example, several times per second or several tens of times per second.
The state data and the water delivery data are collected by arranging corresponding sensors, such as a vibration sensor, a temperature sensor, a pressure sensor, a flow measuring device and the like, and are installed at corresponding positions of the water pump unit for collecting signals. And connecting the acquisition system with a subsequent data processing device to finish data transmission.
S20, setting an operation abnormality threshold, wherein the operation abnormality threshold comprises a vibration abnormality threshold, a temperature abnormality threshold, a pressure abnormality threshold and a water outlet flow speed abnormality threshold, and judging that the operation is abnormal when any one of the operation data exceeds the abnormality threshold;
the method comprises the following specific steps: setting an abnormal threshold value which is a parameter numerical range representing the normal working state of the water pump unit, and setting the threshold value after comprehensively considering the design parameters of the water pump, the normal working tolerance range, the data acquisition precision and the fault state parameter change condition.
In this step, a numerical range of each parameter indicating the normal operation state of the water pump unit, that is, an abnormal threshold value is set in advance. The anomaly threshold value may be set to a maximum value and a minimum value of the corresponding signals. And when the monitored real-time state parameter exceeds the threshold range, judging that the water pump unit has faults or anomalies.
The anomaly threshold should be set taking into account the following factors:
(1) Design parameters of the water pump unit and working parameter ranges under rated working conditions;
(2) Normal working tolerance range of the water pump unit;
(3) The acquisition accuracy and stability of the state parameters;
(4) And (3) parameter state change conditions during various faults in actual operation.
The abnormal threshold is set after the factors are comprehensively considered, so that fault detection can be more accurate and reliable. For example, a numerical range of ±10% of the rated operating condition parameter of the water pump is used as the threshold value for abnormality determination.
Further, in the above technical solution, the specific steps of S30 are:
Establishing a parameter two-dimensional grid, namely a plane grid, taking the acquisition time as a horizontal axis and the total water delivery as a vertical axis;
mapping state data related to the acquisition time and the total water delivery amount to a planar grid;
collecting state data in real time, and mapping the state data to a planar grid;
And analyzing the change trend of the total water delivery and the relation with the acquisition time by observing the change of the state data on the plane grid.
S30, establishing a plane grid by taking the acquisition time as a horizontal axis and the total water delivery as a vertical axis;
The purpose of this step is to map the state data of the water pump assembly to a two-dimensional graph so as to form a monitoring map image. For this purpose, a two-dimensional planar mesh model needs to be built first. Specifically, time data among the parameters of the state data may be selected as a horizontal axis variable of the mesh model, and the total water delivery amount as a vertical axis variable of the mesh model. The two are respectively used as two coordinates of a coordinate system, and a two-dimensional grid array is formed on the two coordinates. The data state and water output value at each instant are mapped as a point on the corresponding location of the grid array. As the system operates, a sufficient number of point-filled grid arrays can be formed to form a two-dimensional graph representing the change in state data over time. The time and the water delivery amount are selected as the coordinates, so that the two parameters can accurately represent the running condition of the water pump unit.
S40, mapping the collected operation data into RGB colors, and filling the plane grids to form a monitoring mapping image;
The method comprises the following specific steps:
Defining a nonlinear mapping function, wherein the mapping function is used for mapping the state parameter values to corresponding RGB color values, and determining the mapping relation between the state parameter values and the RGB color values;
For each data point, calculating a corresponding RGB color value through a mapping function according to the state parameter value of the data point;
the RGB color values for each data point are used as input to color the corresponding location of the image of the planar grid using image processing techniques.
And mapping each piece of operation data information, including vibration, temperature, pressure and water outlet flow rate data, into color information on the two-dimensional plane grid model. A common color representation method is to use an RGB three-primary color representation method, and use a numerical combination of three color channels of R (Red), G (Green), and B (Blue) to represent the corresponding color. The key of this step is to establish a mapping relationship between the state parameter values and the RGB color values. The state parameters first need to be normalized appropriately to map their values to between 0 and 255. The nonlinear mapping relation is set, and different mapping relations can be set for different parameter types so as to highlight the change of important parameters. And calculating the RGB value corresponding to each data point according to the set mapping relation. Then, color filling is performed at the corresponding position of the two-dimensional grid pattern with the RGB values. The process is repeated to obtain color filled patterns of data states at different times. The finally formed time-water delivery two-dimensional graph with the data mapping color is the monitoring mapping image to be established.
S50, carrying out gray scale processing on the monitoring mapping image by adopting an image processing function to obtain a gray scale image;
And carrying out graying treatment on the obtained monitoring image with the color mapping. Image graying is a common simplified image representation. The aim of the graying treatment adopted in the step is to simplify the image data, which is beneficial to keeping the main characteristic condition. Specific graying methods are various. One method that is relatively common is to average the RGB values of the original color image as the gray scale of the output image. Repeating the operation to convert all pixel gray scales. Other image processing functions and tool boxes can also be used to directly realize the graying of the image. The output of this step is a monitoring of the gray scale of the mapped image, i.e. the gray scale image.
S60, marking abnormal points and areas approaching to the abnormal states on the gray level image according to the set operation abnormal threshold, namely, an alarm area;
The method comprises the following specific steps: and judging the data points exceeding the threshold value on the gray level image as abnormal points, and selecting the region with larger change gradient to judge the region as an abnormal trend region.
Since the anomaly threshold value for the state data has been defined, the anomaly data points can be depicted and identified directly on the grayscale image. The judgment basis is to look at RGB mapping values of data points at each moment on the image, and if the RGB mapping values are lower than or higher than a set threshold value, abnormal points are judged. And the area around the abnormal point can be selected to be a certain range to be judged as an abnormal trend area, and the abnormal trend area is used as the abnormal development situation judgment. Compared with directly observing the data graph, the abnormal condition is identified on the image more intuitively and vividly. When the abnormal trend range is selected, the physical rule of abnormal diffusion can be considered, and the region with larger abnormal parameter change gradient can be selected.
S70, establishing a cellular automaton model on the gray level image, and calculating rules of the cellular automaton as alarm region development rules according to the abnormal points on the gray level image and the change rules of the alarm region;
The method comprises the following specific steps:
Identifying and tracking abnormal points in the image;
Analyzing the diffusion mode and trend of the abnormal points by observing the diffusion condition of the abnormal points in the image;
Generating an extinction and diffusion behavior according to the analyzed abnormal points, and reversely pushing a state transition rule of the cellular automaton;
After the state transition rule of the cellular automaton is obtained through reverse thrust, a cellular automaton model is built according to the rule obtained through reverse thrust and the initial state.
And (3) establishing a cellular automaton model based on the gray level image obtained in the step S60. Cellular automata is a dynamic system model in which a large number of cell cells update values according to a certain evolution rule. The key to establishing cellular automata is to determine its state transition rules, i.e. the basis for cell renewal. The method of the step is to reverse the state transition rule of cellular automaton by analyzing the generation, death, evolution and diffusion behaviors of abnormal points on the image. The method has the advantages that the evolution rule is designed directly according to the state change rule of the target system, and a well-matched cellular automaton model can be obtained. And defining an initial state after determining the cellular automaton rule, and then establishing a cellular automaton model.
S80, predicting gray level image variation in a specified time period according to a cellular automaton model and a calculated alarm region development rule to obtain a gray level image in the specified time period, and marking the gray level image as a predicted image;
the method comprises the following specific steps: and (3) iteratively calculating cell state update based on a cell automaton rule, and advancing to a specified time period to obtain an abnormal development prediction image.
On the basis of the cellular automaton model and the evolution rule thereof, the model can be utilized to simulate the image state evolution process in a given time period, so that the prediction of abnormal development situation is realized. The specific method is to define the initial state in cellular automaton in advance, considering the state of the image currently obtained. And then, according to the defined automaton evolution rule, advancing the cell state, and performing iterative calculation. The number of time steps required for the calculation matches the actual time range. And repeatedly updating the cell state, and obtaining an evolution rule from the current state to the future state as an alarm region development rule.
S90, outputting abnormal points of the gray level image, the alarm region, the abnormal points of the predicted image, the corresponding time of the alarm region and the operation data to operation and maintenance personnel.
And S90, outputting the acquisition time, the operation data, the identified abnormal points, the alarm region and the evolution prediction result to operation and maintenance personnel. Specifically, the method comprises the following sub-steps:
(1) And constructing a data display platform, such as a development monitoring software system, and building a Web page or an App application program for receiving data input and realizing data visual display.
(2) And (3) displaying the original operation data acquired in the step (S10) in time sequence to form a data graph, and displaying the data graph together with the abnormal threshold range predefined in the step (S20) so as to be convenient for directly observing whether the data is abnormal or not.
(3) The data graph is marked with the abnormal data points identified in step S60. When an outlier data point appears, it is marked with a distinct icon. And displaying information such as the abnormal type, abnormal data value, abnormal time and the like.
(4) And displaying the abnormal trend area determined in the step S60 in a mode of monitoring the mapping image or the gray level image, and visually displaying the abnormal development range.
(5) And displaying an abnormal development situation result image obtained based on cellular automaton prediction in the step S80. The emphasis marks the points and areas of anomalies that may occur over a period of time in the future.
(6) The human-computer interaction function is provided, and the operation and maintenance personnel can select and display historical data or predicted result data of different time periods, enlarge and fine-view the interested image area and scroll and browse results at different moments.
(7) When the system automatically recognizes or predicts the occurrence of the abnormality, a visual and audible alarm signal is sent to prompt operation and maintenance personnel.
Through the monitoring and display of the step, operation and maintenance personnel can clearly know the real-time state and the future operation trend of the water pump unit, find abnormality and quickly make judgment, guide the adjustment and maintenance of equipment, and realize the on-line monitoring and prediction of the state of the water pump unit.
The first aspect of the present invention provides a second embodiment of a method for online monitoring an operation state of a water pump unit, in this embodiment, the method includes the following steps:
S10, collecting operation data and total water delivery quantity of a water pump unit in real time, wherein the operation data comprise vibration, temperature, pressure and water outlet flow rate;
in this step, various operation data of the water pump unit and the total water delivery amount need to be collected in real time. The operation data are mainly acquired through sensors, and the total water delivery is calculated through flow integration.
1. Operation data:
The operation data mainly includes:
Vibration signal: x (t) represents a vibration value at time t;
temperature signal: t (T) represents the temperature at time T;
Pressure signal: p (t) represents the pressure at time t;
Flow rate of effluent: v (t) represents the water outlet rate at time t;
the entire operational data may be represented as one vector:
X(t)＝[x(t)，T(t)，p(t)，v(t)]；
It should be noted that the above signal requires a sampling frequency acquisition high enough to sensitively capture the state change. The sampling frequency is typically set to f s, which may range in value from tens of hertz to kilohertz.
Thus, the operational data may be represented as a time series:
X(t1),X(t2),...,X(tN)；
Wherein, Is a discrete sampling instant.
2. Total water delivery:
let the water delivery flow rate in the unit time of the water pump be Q (t), then the calculation formula of the total water delivery Q (t) is:
I.e. integrating the flow signal, τ represents the amount of change between 0 and t.
In the above formula, q (t) is not a continuous signal in practice, but is acquired in a discrete manner. The total water delivery can thus be estimated by numerical integration:
Wherein, Δt represents the variation value of t. The total water delivery at different moments can be obtained by recursive calculation.
S20, setting an operation abnormality threshold, wherein the operation abnormality threshold comprises a vibration abnormality threshold, a temperature abnormality threshold, a pressure abnormality threshold and a water outlet flow speed abnormality threshold, and judging that the operation is abnormal when any one of the operation data exceeds the abnormality threshold;
in this step, an abnormality threshold value of a parameter indicating the state of the water pump unit needs to be set in advance to determine whether an abnormality exists.
The anomaly threshold value is generally set as a boundary value of the normal operating range of the parameter. And thus can be expressed as:
Xmin＝[xmin,Tmin,pmin,vmin]；
Xmax＝[xmax,Tmax,pmax,vmax]；
If the signal value of a certain dimension of X (t) is lower than X min or higher than X max, the abnormality is determined.
The threshold value needs to be set by comprehensively considering factors such as design parameters of the water pump, normal working precision and the like. The conservation value can be generally selected to ensure detection sensitivity.
S30, establishing a plane grid by taking the acquisition time as a horizontal axis and the total water delivery as a vertical axis;
this step requires the creation of a two-dimensional planar grid, wherein:
The horizontal axis is time t;
the vertical axis is the total water delivery quantity Q (t);
a two-dimensional coordinate system of time-water delivery quantity is formed.
On the coordinate system, the data at each sampling time t n is mapped to one point (t n,Q(tn)). As time progresses, the space is populated with a large number of data points for subsequent state mapping.
S40, mapping the collected operation data into RGB colors, and filling the plane grids to form a monitoring mapping image;
the goal of this step is to map the multi-dimensional state data X (t) to color information in the image. The commonly used color model is the RGB model:
RGB＝[R(t)，G(t)，B(t)]；
Wherein the value range of each channel variable is 0-255.
It is necessary to establish a mapping relationship between the state data X (t) and RGB values:
RGB＝f(X)；
The mapping function f may be non-linear and designed for different state quantities to emphasize information importance.
The data X (t n) at each time t n may be converted to corresponding RGB values. Each point in the two-dimensional grid (t n,Q(tn)) is mapped to one RGB color. The final visualization is a color mapped time-water delivery graph.
S50, carrying out gray scale processing on the monitoring mapping image by adopting an image processing function to obtain a gray scale image;
The goal of this step is to simplify the image representation, converting the color image into a gray scale image. The usual graying method is to calculate the average value of the RGB channels as the output gray Y:
By repeating this operation, a gray scale time-water delivery amount image can be output. The image simplifies the information expression and is convenient for subsequent processing and analysis.
S60, marking abnormal points and areas approaching to the abnormal states on the gray level image according to the set operation abnormal threshold, namely, an alarm area;
After the gradation mapping image is obtained, it is possible to detect whether the data at each time is abnormal.
That is, for time t=t n, it is determined whether X (t n) satisfies the threshold condition:
Xmin＜X(tn)＜Xmax；
If not, the point is determined to be an abnormal point. And determines an abnormal region based on the position space information.
Finally, the abnormal points and the areas thereof can be directly marked on the gray level image, thereby realizing visual abnormal detection.
S70, establishing a cellular automaton model on the gray level image, and calculating rules of the cellular automaton as alarm region development rules according to the abnormal points on the gray level image and the change rules of the alarm region;
the method aims at simulating the evolution of the abnormal situation through cellular automaton modeling. The cellular automaton model can be expressed as:
Z＝{S，N，f}；
Wherein:
S: a set of cell states;
n: cell neighborhood geometry;
f: a state transition rule;
The key to establishing cellular automata is to determine f, the state transition rule. This step determines f by analyzing the outlier variation on the image and generating extinction.
After f is determined, a cellular automaton model can be built.
S80, predicting gray level image variation in a specified time period according to a cellular automaton model and a calculated alarm region development rule to obtain a gray level image in the specified time period, and marking the gray level image as a predicted image;
After the cellular automaton model is built, predictions can be made.
The cell state is iteratively updated according to the state transition rule:
S(t+1)＝f[S(t)，N(t)]；
through multiple pushing simulation calculation, the image situation at the future moment can be predicted, and the prediction of the abnormal development trend is realized.
S90, outputting abnormal points of the gray level image, the alarm region, the abnormal points of the predicted image, the corresponding time of the alarm region and the operation data to operation and maintenance personnel.
The goal of this step is to output the results of the monitoring and prediction to the operation and maintenance personnel. Mainly comprises the following sub-steps:
(1) And constructing a monitoring and displaying platform. The receiving and displaying of data and images may be implemented in the form of a software system or Web pages.
(2) Displaying an original operation data curve:
X(t)＝[x(t)，T(t)，p(t)，v(t)]；
And the defined abnormal threshold range is marked, so that whether the data are normal can be intuitively judged.
(3) Detected outliers are marked on the data curve, such as by specific symbols, and details of anomaly type, value, time, etc. are given.
(4) And displaying the gray mapping image, and labeling an abnormal region by using geometric figures such as a rectangle, a circle and the like, wherein R a is represented.
(5) And displaying an abnormal evolution image of the cellular automaton predicted future time.
(6) Providing interactive functions, the user may choose to display data for different time periods, zoom in on the area, etc.
(7) When an abnormality occurs, a visual or audible alarm is issued, prompting the user to deal with in time.
The variables involved in this embodiment are explained as follows:
x (t): a vibration signal indicating a vibration value at time t;
t (T): a temperature signal representing a temperature value at time t;
p (t): a pressure signal representing a pressure value at time t;
v (t): the water flow rate represents the water flow rate at the time t;
X (t): an operation data vector including the plurality of signals;
f s: sampling frequency of the state data;
t n: an nth sampling time;
q (t): the water delivery flow rate of the water pump in unit time;
Q (t): the total water delivery amount of the water pump;
x min: a lower threshold of operational data;
x max: an upper threshold of operational data;
r (t): red channels in RGB colors;
G (t): a green channel in RGB colors;
b (t): blue channels in RGB colors;
Y (t): gray value of the image;
S: a set of cell states;
n: a cell neighborhood structure;
f: cell state transition rules;
R a: an abnormal region;
Description of the formula:
running data vector: x (t);
sampling time: t n＝n/fs;
calculating the total water delivery:
Total water delivery quantity numerical integral estimation:
lower anomaly threshold: x min＝[xmin,Tmin,pmin,vmin ];
Upper threshold anomaly limit: x max＝[xmax,Tmax,pmax,vmax ];
RGB mapping relationship: rgb=f (X);
Gray level conversion:
cellular automaton model: z= { S, N, f };
The state update equation: s (t+1) =f [ S (t), N (t) ].
The second aspect of the present invention provides a computer readable storage medium, where a program instruction is stored, where the program instruction is used to execute the above-mentioned on-line monitoring method for an operation state of a water pump unit when the program instruction is running.
The invention provides an on-line monitoring system for the running state of a water pump unit, which comprises the computer readable storage medium.
Specifically, the principle of the invention is as follows:
1. By establishing a time-water delivery two-dimensional view and adopting a color mapping imaging expression mode, the dynamic change of the multi-source state information can be comprehensively reflected to form an intuitive abnormality detection basis;
2. Applying a cellular automaton theory, reversely solving a cellular transfer rule according to the characteristic of abnormal situation change on an image domain, and predicting an abnormal development trend by means of the evolution simulation capability of the cellular transfer rule; the two-dimensional view construction provides an intuitive state judgment mode, and the color mapping enables multi-source information to be visually expressed, so that the sensitivity to state change and abnormal recognition is improved. The method lays a foundation for subsequent abnormality judgment and trend prediction. The cellular automaton model has strong evolution simulation capability, and the automaton rule is deduced through learning image abnormal mode characteristics, so that abnormal future development situation can be effectively extrapolated, and the prediction and evaluation of the water pump health degree can be realized;
The two-dimensional visual expression is combined with cellular automaton prediction, so that the water pump state monitoring system has two capacities of sensitive real-time detection and accurate future prediction.
Claims (10)
1. The on-line monitoring method for the running state of the water pump unit is characterized by comprising the following steps of:
S10, collecting operation data and total water delivery quantity of a water pump unit in real time, wherein the operation data comprise vibration, temperature, pressure and water outlet flow rate;
s20, setting an operation abnormality threshold, wherein the operation abnormality threshold comprises a vibration abnormality threshold, a temperature abnormality threshold, a pressure abnormality threshold and a water outlet flow speed abnormality threshold, and judging that the operation is abnormal when any one of the operation data exceeds the abnormality threshold;
s30, establishing a plane grid by taking the acquisition time as a horizontal axis and the total water delivery as a vertical axis;
S40, mapping the collected operation data into RGB colors, and filling the plane grid to form a monitoring mapping image;
S50, carrying out gray scale processing on the monitoring mapping image by adopting an image processing function to obtain a gray scale image;
S60, marking abnormal points and areas approaching to the abnormal states on the gray level image according to a set operation abnormal threshold, namely an alarm area;
s70, establishing a cellular automaton model on the gray level image, and calculating rules of the cellular automaton as alarm region development rules according to variation rules of abnormal points and alarm regions on the gray level image;
S80, predicting gray level image variation in a specified time period according to the cellular automaton model and the estimated alarm region development rule to obtain a gray level image in the specified time period, and marking the gray level image as a predicted image;
And S90, outputting the abnormal points of the gray level image, the alarm region, the abnormal points in the predicted image, the corresponding moments of the alarm region and the operation data to operation and maintenance personnel.
2. The method for on-line monitoring the running state of a water pump unit according to claim 1, wherein the specific steps of S10 are as follows: collecting operation data and total water delivery of a water pump unit in real time, wherein the operation data comprise vibration, temperature, pressure and water outlet flow rate; and setting corresponding sensors and connecting the acquisition system with the data processing device to finish data transmission.
3. The method for on-line monitoring the operation state of a water pump unit according to claim 2, wherein the specific step of S20 is as follows: setting an abnormal threshold value which is a parameter numerical range representing the normal working state of the water pump unit, and setting the threshold value after comprehensively considering the design parameters of the water pump, the normal working tolerance range, the data acquisition precision and the fault state parameter change condition.
4. The on-line monitoring method for the running state of the water pump unit according to claim 3, wherein the specific step of S30 is as follows:
Establishing a parameter two-dimensional grid, namely a plane grid, taking the acquisition time as a horizontal axis and the total water delivery as a vertical axis;
mapping state data related to the acquisition time and the total water delivery amount to the planar grid;
collecting the state data in real time, and mapping the state data to the planar grid;
And analyzing the change trend of the total water delivery and the relation with the acquisition time by observing the change of the state data on the plane grid.
5. The method for on-line monitoring the operation state of a water pump unit according to claim 4, wherein the specific step of S40 is as follows:
Defining a nonlinear mapping function, wherein the mapping function is used for mapping the state parameter values to corresponding RGB color values and determining the mapping relation between the state parameter values and the RGB color values;
for each data point, calculating a corresponding RGB color value through the mapping function according to the state parameter value of the data point;
the RGB color values for each of the data points are used as input to color the corresponding location of the image of the planar grid using image processing techniques.
6. The method for on-line monitoring the operation state of a water pump unit according to claim 5, wherein the specific step of S60 is as follows: and judging the data points exceeding the threshold value on the gray level image as abnormal points, and selecting the region with larger change gradient to judge the region as an abnormal trend region.
7. The method for on-line monitoring the operation state of a water pump unit according to claim 6, wherein the specific step of S70 is:
Identifying and tracking abnormal points in the image;
Analyzing the diffusion mode and trend of the abnormal point by observing the diffusion condition of the abnormal point in the image;
Generating an extinction and diffusion behavior according to the analyzed abnormal points, and reversely pushing a state transition rule of the cellular automaton;
After the state transition rule of the cellular automaton is obtained through reverse thrust, a cellular automaton model is built according to the rule obtained through reverse thrust and the initial state.
8. The method for on-line monitoring the operation state of a water pump unit according to claim 7, wherein the specific step of S80 is: and (3) iteratively calculating cell state update based on a cell automaton rule, and advancing to a specified time period to obtain an abnormal development prediction image.
9. A computer readable storage medium, wherein program instructions are stored in the computer readable storage medium, and when the program instructions are executed, the program instructions are used for executing an on-line monitoring method for the running state of a water pump unit according to any one of claims 1 to 8.
10. An on-line monitoring system for the operation state of a water pump assembly, comprising the computer-readable storage medium of claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311775945.6A CN117948269A (en) | 2023-12-21 | 2023-12-21 | Water pump unit running state on-line monitoring method, medium and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311775945.6A CN117948269A (en) | 2023-12-21 | 2023-12-21 | Water pump unit running state on-line monitoring method, medium and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117948269A true CN117948269A (en) | 2024-04-30 |
Family
ID=90797500
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311775945.6A Pending CN117948269A (en) | 2023-12-21 | 2023-12-21 | Water pump unit running state on-line monitoring method, medium and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117948269A (en) |
-
2023
- 2023-12-21 CN CN202311775945.6A patent/CN117948269A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200081054A1 (en) | Power line issue diagnostic methods and apparatus using distributed analytics | |
| US20180224831A1 (en) | Event analyzing device, event analyzing system, event analyzing method, and non-transitory computer readable storage medium | |
| JP2007249997A (en) | Method and system for monitoring industrial process | |
| KR102343752B1 (en) | Computer-implemented method and system for automatically monitoring and determining the status of entire process segments in a process unit | |
| US20220341996A1 (en) | Method for predicting faults in power pack of complex equipment based on a hybrid prediction model | |
| CN110334816A (en) | A kind of industrial equipment detection method, device, equipment and readable storage medium storing program for executing | |
| CN116757097A (en) | Digital twin hydraulic engineering operation and maintenance monitoring system and method | |
| CN115640860B (en) | Electromechanical equipment remote maintenance method and system for industrial cloud service | |
| CN117196159A (en) | Intelligent water service partition metering system based on Internet big data analysis | |
| CN115563683A (en) | Hydraulic engineering automatic safety monitoring management system based on digital twins | |
| CN113850475A (en) | Production equipment performance evaluation system based on big data | |
| US11900223B2 (en) | Device and method for monitoring a system | |
| CN117151649A (en) | Construction method management and control system and method based on big data analysis | |
| CN116894747A (en) | Monitoring system for investigation of groundwater environment of polluted site | |
| CN117556366B (en) | Data abnormality detection system and method based on data screening | |
| CN112070400A (en) | Method and device for evaluating security situation of power grid database | |
| CN112926656A (en) | Method, system and equipment for predicting state of circulating water pump of nuclear power plant | |
| CN117072520A (en) | Hydraulic element detecting system based on PLC controller | |
| CN117948269A (en) | Water pump unit running state on-line monitoring method, medium and system | |
| CN114417741B (en) | One-machine one-file equipment data management and processing system | |
| CN115660513A (en) | Monitoring method and system based on aqueduct deformation of hydraulic engineering | |
| RU2735296C1 (en) | Method for multi-parameter monitoring of complex electrotechnical objects state | |
| CN103854075A (en) | Failure predicting method and device aiming at electronic single chip under hot standby environment | |
| KR102573254B1 (en) | System for predicting and analyzing trouble of mechanical equipment using federated learning | |
| WO2021181585A1 (en) | Analysis device, analysis method, and analysis program |
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 |