Intelligent testing system and method for water pump performance

Abstract

The present application relates to the technical field of water pump performance testing, and specifically discloses an intelligent testing system and method for water pump performance, comprising the following steps: step S1: setting a monitoring point on the pump shell to collect vibration signals under the condition of silt-containing water quality, and pre-processing to extract high-frequency conduction vibration signals; step S2: collecting sample data, including conduction vibration signals and wear intensity within a running cycle, and obtaining an amplitude distribution curve and calculating an amplitude wear coefficient accordingly; and step S3: setting a wear intensity threshold, combining the amplitude distribution curve and the amplitude wear coefficient, and calculating a wear rate of the water pump under the current silt-containing water quality environment; the present application not only improves the accuracy of wear monitoring, but also realizes the transformation from passive maintenance to active prediction, significantly improving the operation reliability and service life of the water pump system.

Description

Technical Field

[0001] This invention relates to the field of water pump performance testing technology, and specifically to an intelligent testing system and method for water pump performance. Background Technology

[0002] In the field of water pump performance testing, wear rate testing is a crucial step, as it directly relates to the accurate assessment of the overall operating condition and lifespan of the water pump.

[0003] However, existing testing technologies have significant shortcomings. Currently, when testing the wear rate of water pumps, the complex factors in actual operating conditions are generally not fully considered, especially when the pump operates in a sandy water environment. In real-world applications, sand carried in the water exerts a continuous and intense abrasive effect on the pump. However, existing technologies often ignore this crucial factor, conducting tests only under normal conditions. This results in wear rate test results that are significantly lower than the true values, failing to accurately reflect the wear condition of the pump during actual use. Such inaccurate test results not only affect the correct assessment of pump performance but also mislead subsequent maintenance planning and equipment replacement decisions, leaving related management work without a scientific basis and making it difficult to effectively ensure the stable and reliable operation of the pump system. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent testing system and method for water pump performance, and to solve the following technical problems.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] A smart testing method for water pump performance includes the following steps:

[0007] Step S1: Install vibration monitoring points on the pump casing of the water pump. The vibration monitoring points are used to collect vibration signals of the water pump in real time when it is running in a sandy water environment. The vibration signals are preprocessed to obtain high-frequency conducted vibration signals.

[0008] Step S2: Collect sample data, which includes the conducted vibration signals and wear intensity of all water pump samples during their operating cycle in a sediment-containing water environment; based on the sample data, obtain the amplitude distribution curve of all water pump samples, and obtain the amplitude wear coefficient based on the wear intensity corresponding to each amplitude distribution curve;

[0009] Step S3: Set the wear intensity threshold, obtain the amplitude distribution curve of the water pump, and obtain the wear rate of the water pump when it is running in the current sandy water quality environment based on the amplitude wear coefficient and the wear intensity threshold.

[0010] Based on the wear rate, the timestamp when the current wear intensity reaches the wear intensity threshold is predicted and recorded as the predicted shutdown time. A performance evaluation report of the water pump is generated based on the predicted shutdown time.

[0011] As a further aspect of the present invention, the preprocessing of the vibration signal includes:

[0012] Set a passband frequency range, and set a high-frequency bandpass filter according to the passband frequency range; perform Fourier transform on all vibration signals to convert all vibration signals into frequency domain signals, filter all frequency domain signals through the high-frequency bandpass filter, remove vibration signals whose frequencies do not belong to the passband frequency range, and record the remaining vibration signals as conducted vibration signals.

[0013] As a further aspect of the present invention: the operating cycle is the time period between the time when the water pump sample is put into use and the time when it is put out of use, wherein the time when it is put into use is the timestamp corresponding to the time when the water pump sample is put into use, and the time when it is put out of use is the timestamp corresponding to the time when the water pump sample is put out of use.

[0014] As a further aspect of the present invention: the process of collecting the wear intensity includes:

[0015] The pump casing of the water pump that has not experienced wear is denoted as the unworn pump casing. An image of the unworn pump casing is obtained and denoted as the initial image. The initial image is divided into several pixels, each pixel is numbered, and the pixel value of each pixel is obtained. Each numbered pixel is associated with its corresponding pixel value to obtain the set of pixel values ​​of the initial image.

[0016] Obtain an image of the pump casing of the water pump sample at the end of use, and denote it as a wear image; divide the wear image into the same number of pixels, denote them as wear pixels, assign the same number to each wear pixel, and obtain the pixel value of each wear pixel; associate the corresponding pixel values ​​of each numbered wear pixel to obtain the wear pixel value set of the wear image;

[0017] The wear intensity is obtained based on the set of pixel values ​​and the set of worn pixel values. Where n is the total number of pixels in the pixel value set, RGB i Represents the i-th pixel in the set of pixel values, RGB i ´ represents the i-th pixel in the set of worn pixel values, where i∈[1,n] and i is a positive integer.

[0018] As a further aspect of the present invention: the process of obtaining the amplitude distribution curves of all water pump samples includes:

[0019] For any water pump sample, acquire all conducted vibration signals of the water pump sample and obtain the maximum amplitude value among all conducted vibration signals; establish a coordinate system with amplitude as the abscissa and quantity as the ordinate; in the coordinate system, for any abscissa value, acquire the quantity of conducted vibration signals with the maximum amplitude value of that abscissa value, and record it as the ordinate value corresponding to that abscissa value; obtain a data point in the coordinate system based on the abscissa value and the ordinate value corresponding to the abscissa value; then, based on each abscissa value and its corresponding ordinate value, obtain several data points in the coordinate system, and connect each data point sequentially with a smooth curve, and record the curve as the amplitude distribution curve of the water pump sample.

[0020] As a further aspect of the present invention: the process of obtaining the amplitude wear coefficient includes:

[0021] Setting a quantity threshold N, for any amplitude distribution curve, a horizontal dividing line y=N is obtained in the coordinate system according to the quantity threshold, where y represents the ordinate in the coordinate system; the entire area enclosed by the horizontal dividing line and the amplitude distribution curve above the horizontal dividing line is then obtained, thus yielding the impact index. Where K is the transformation coefficient and K > 0, m is the total number of regions, and A x1 >A x2 A x1 A represents the first intersection point of the horizontal boundary line and the amplitude distribution curve in the x-th region. x2 Dis(A) represents the second intersection point of the horizontal boundary line and the amplitude distribution curve in the x-th region, and A represents the horizontal axis.

[0022] An initial regression model is established based on a linear function of one variable with a direct proportion. The impact index of each transmitted vibration signal and its corresponding wear intensity are recorded as the training set. The training set is input into the initial regression model to train the initial regression model and obtain the final regression model. The slope of the regression model is obtained and recorded as the amplitude wear coefficient.

[0023] As a further aspect of the present invention: the process of obtaining the wear rate of the water pump under the current sediment-laden water quality environment includes:

[0024] From the moment the water pump is put into use, the vibration signal of the pump casing is monitored in real time and recorded as the current vibration signal. The current transmitted vibration signal is obtained by preprocessing the current vibration signal. Based on the current transmitted vibration signal at each monitoring time, the impact index at each monitoring time is obtained and recorded as the current impact index. Based on the amplitude wear coefficient, the wear intensity at each monitoring time is obtained.

[0025] A new coordinate system is established with time as the horizontal axis and wear intensity as the vertical axis. The timestamp of each monitoring is obtained, and the timestamp of each monitoring and its corresponding wear intensity are fitted in the new coordinate system to obtain a wear intensity variation curve.

[0026] Several monitoring points were selected on the wear intensity variation curve, and the corresponding timestamps and wear intensity at each monitoring point were obtained to obtain the wear rate of the water pump under the current sediment-laden water quality environment. Where M is the total number of monitoring points, and Ws k ´ represents the current wear intensity of the kth monitoring point, k∈[1,M] and k is a positive integer, and t is the time interval between two monitoring points.

[0027] An intelligent testing system for water pump performance includes:

[0028] Signal acquisition unit: Vibration monitoring points are set on the pump casing of the water pump. The vibration monitoring points are used to collect the vibration signals of the water pump in real time when it is running in a sandy water environment; the vibration signals are preprocessed to obtain high-frequency conducted vibration signals.

[0029] Wear monitoring unit: Collects sample data, including the conducted vibration signals and wear intensity of all water pump samples during their operating cycle in a sandy water environment; Based on the sample data, obtains the amplitude distribution curves of all water pump samples, and obtains the amplitude wear coefficient based on the wear intensity corresponding to each amplitude distribution curve;

[0030] Performance evaluation unit: Set the wear intensity threshold, obtain the amplitude distribution curve of the water pump, and obtain the wear rate of the water pump when it is running in the current sandy water quality environment based on the amplitude wear coefficient and the wear intensity threshold;

[0031] Based on the wear rate, the timestamp when the current wear intensity reaches the wear intensity threshold is predicted and recorded as the predicted shutdown time. A performance evaluation report of the water pump is generated based on the predicted shutdown time.

[0032] The beneficial effects of this invention are:

[0033] This invention effectively isolates low-frequency vibration interference caused by mechanical motion such as impeller rotation by collecting high-frequency conducted vibration signals of water pumps operating in sandy water environments. It accurately captures the characteristic vibration signals generated by sand impacting the pump body, solving the technical problem that existing technologies cannot accurately reflect wear conditions under sandy conditions. Furthermore, by establishing a mapping model between amplitude distribution curves and wear intensity, it innovatively proposes the concept of amplitude wear coefficient, realizing a quantitative assessment of wear intensity based on vibration signal characteristics. This invention can accurately calculate the real-time wear rate of water pumps in specific sandy environments and predict the time to reach the critical wear threshold, providing a precise scientific basis for water pump maintenance decisions. Compared with traditional methods, this invention not only improves the accuracy of wear monitoring but also achieves a shift from passive maintenance to proactive prediction, significantly improving the operational reliability and service life of water pump systems, and has significant engineering application value. Attached Figure Description

[0034] The invention will now be further described with reference to the accompanying drawings.

[0035] Figure 1 This is a schematic diagram illustrating the steps of an intelligent testing method for water pump performance according to the present invention. Detailed Implementation

[0036] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] Please see Figure 1 As shown, this invention is an intelligent testing method for water pump performance, comprising the following steps:

[0038] Step S1: Install vibration monitoring points on the pump casing of the water pump. The vibration monitoring points are used to collect vibration signals of the water pump in real time when it is running in a sandy water environment. The vibration signals are preprocessed to obtain high-frequency conducted vibration signals.

[0039] It should be noted that the vibration monitoring point is based on a vibration sensor, preferably an integrated circuit piezoelectric (IEPE) accelerometer. Its metal shell is permanently connected to a mounting base fixed to the water pump housing using high-strength adhesive or welding to ensure a smooth and high-fidelity vibration transmission path. The surface of the mounting base should be smooth and parallel to the water flow direction to reduce direct impact from sand particles in the external flow field. The sensor connection cable should be properly secured to prevent it from swinging in the water flow and generating additional vibration noise.

[0040] Specifically, in a sandy water environment, the vibration signal of a water pump during operation is essentially composed of discrete vibration pulses generated by each instantaneous collision between sand particles in the water and the pump casing, impeller, and other flow-through components. Due to the randomness of sand particle movement, the collisions with pump components do not occur continuously but exist as independent, dispersed single events. Each collision of a sand particle will excite a brief vibration signal unit, and these signal units are discretely distributed on the time axis, collectively forming the overall vibration signal set. The characteristics of each vibration signal are fundamentally different from the smooth and continuous vibration during normal pump operation. Its core feature is that it is composed of the superposition of instantaneous signals corresponding to countless independent collision events. Therefore, the high-frequency conducted vibration signal obtained after preprocessing actually retains the characteristic information of each single collision event.

[0041] In a preferred embodiment of the present invention, the preprocessing of the vibration signal includes:

[0042] Set a passband frequency range, and set a high-frequency bandpass filter according to the passband frequency range; perform Fourier transform on all vibration signals to convert all vibration signals into frequency domain signals, filter all frequency domain signals through the high-frequency bandpass filter, remove vibration signals whose frequencies do not belong to the passband frequency range, and record the remaining vibration signals as conducted vibration signals;

[0043] Specifically, the transmitted vibration signal collected by this invention is the stress wave and vibration generated when the impeller, volute and other components inside the water pump are hit by sand particles in the water flow. This vibration is transmitted to the sensor installation point through the pump casing and is a type of structurally conducted sound. In the preprocessing process, the vibration generated by the sand particles in the water flow outside the water pump directly hitting the sensor housing is removed. This is a type of airborne sound or direct impact.

[0044] The high-speed impact of sand particles on the metal impeller generates enormous energy, thus the frequency of the transmitted vibration signal is very high, and the stress wave has a wide frequency band, typically ranging from several thousand hertz to tens of thousands of hertz. Vibration signals outside the aforementioned passband frequency range are generated by sand particles splashing out of the water flow or directly impacting the sensor's outer protective shell. These vibration signals have relatively low frequencies, weak energy, and their frequency bands are significantly different from those of internal impacts.

[0045] Step S2: Collect sample data, which includes the conducted vibration signals and wear intensity of all water pump samples during their operating cycle in a sediment-containing water environment; based on the sample data, obtain the amplitude distribution curve of all water pump samples, and obtain the amplitude wear coefficient based on the wear intensity corresponding to each amplitude distribution curve;

[0046] In a preferred embodiment of the present invention, the operating cycle is the time period between the time when the water pump sample is put into use and the time when it is put out of use, wherein the time when the water pump sample is put into use is the timestamp corresponding to the time when the water pump sample is put out of use, and the time when it is put out of use is the timestamp corresponding to the time when the water pump sample is put out of use.

[0047] In a preferred embodiment of the present invention, the process of collecting the wear intensity includes:

[0048] The pump casing of the water pump that has not experienced wear is denoted as the unworn pump casing. An image of the unworn pump casing is obtained and denoted as the initial image. The initial image is divided into several pixels, each pixel is numbered, and the pixel value of each pixel is obtained. Each numbered pixel is associated with its corresponding pixel value to obtain the set of pixel values ​​of the initial image.

[0049] Obtain an image of the pump casing of the water pump sample at the end of use, and denote it as a wear image; divide the wear image into the same number of pixels, denote them as wear pixels, assign the same number to each wear pixel, and obtain the pixel value of each wear pixel; associate the corresponding pixel values ​​of each numbered wear pixel to obtain the wear pixel value set of the wear image;

[0050] The wear intensity is obtained based on the set of pixel values ​​and the set of worn pixel values. Where n is the total number of pixels in the pixel value set, RGB i Represents the i-th pixel in the set of pixel values, RGB i ´ represents the i-th pixel in the set of worn pixel values, where i∈[1,n] and i is a positive integer;

[0051] Specifically, high-resolution industrial cameras are used to acquire surface images of the pump casing before and after wear, and image registration technology is used to ensure that the two acquisition positions are completely consistent. The images are converted to the HSV color space, and the feature values ​​of the saturation (S) and lightness (V) channels are extracted, because these two channels are more sensitive to changes in surface texture and material loss. By calculating the Euclidean distance difference between each pixel in the saturation (S) and lightness (V) channels, and depending on the actual situation, different pixel area weight coefficients can also be combined to obtain the wear intensity.

[0052] In a preferred embodiment of the present invention, the process of obtaining the amplitude distribution curves of all water pump samples includes:

[0053] For any water pump sample, acquire all conducted vibration signals of the water pump sample, and acquire the maximum amplitude value among all conducted vibration signals; establish a coordinate system with amplitude as the abscissa and quantity as the ordinate; in the coordinate system, for any abscissa value, acquire the quantity of conducted vibration signals with the maximum amplitude value of that abscissa value, and record it as the ordinate value corresponding to that abscissa value; based on the abscissa value and the ordinate value corresponding to the abscissa value, obtain a data point in the coordinate system; then based on each abscissa value and its corresponding ordinate value, obtain several data points in the coordinate system, and connect each data point sequentially with a smooth curve, and record the curve as the amplitude distribution curve of the water pump sample;

[0054] In a preferred embodiment of the present invention, the process of obtaining the amplitude wear coefficient includes:

[0055] Setting a quantity threshold N, for any amplitude distribution curve, a horizontal dividing line y=N is obtained in the coordinate system according to the quantity threshold, where y represents the ordinate in the coordinate system; the entire area enclosed by the horizontal dividing line and the amplitude distribution curve above the horizontal dividing line is then obtained, thus yielding the impact index. Where K is the transformation coefficient and K > 0, m is the total number of regions, and A x1 >A x2 A x1 A represents the first intersection point of the horizontal boundary line and the amplitude distribution curve in the x-th region. x2 Dis(A) represents the second intersection point of the horizontal boundary line and the amplitude distribution curve in the x-th region, and A represents the horizontal axis.

[0056] An initial regression model is established based on a linear function of one variable with a direct proportion. The impact index of each transmitted vibration signal and its corresponding wear intensity are recorded as the training set. The training set is input into the initial regression model to train the initial regression model and obtain the final regression model. The slope of the regression model is obtained and recorded as the amplitude wear coefficient.

[0057] In setting the quantity threshold N, a dynamic statistical analysis method is adopted based on the actual operating conditions of the water pump and the signal acquisition characteristics. Specifically:

[0058] Vibration signal samples of water pumps were collected under rated operating conditions and without sediment in water quality to obtain amplitude distribution curve samples. Through laboratory calibration tests, multiple sets of vibration signal samples were obtained under different sediment content conditions. The aggregation characteristics of the transmitted vibration signal samples in the amplitude distribution curve samples were statistically analyzed. The quantity value corresponding to the 95th percentile of the amplitude distribution curve sample was obtained as the basic threshold, and then multiplied by the operating condition safety factor. The value range of the safety factor is [1.2, 1.5].

[0059] The conversion coefficient setting process is based on a combination of physical modeling and experimental calibration using the energy equivalence principle. A dynamic model of sand impacting the pump casing material is established through finite element analysis, and the theoretical vibration energy values ​​generated by sand particles of different sizes impacting the impeller at different pumping velocities are calculated. On a laboratory calibration platform, controlled impact tests are conducted using standard sand samples under different operating conditions, and the actual vibration energy values ​​are accurately measured using a laser Doppler vibration meter. The conversion coefficient K is obtained by fitting the ratio between the theoretical and measured values ​​using the least squares method, and is typically set to a value in the range of [0.6, 1.2].

[0060] Step S3: Set the wear intensity threshold, obtain the amplitude distribution curve of the water pump, and obtain the wear rate of the water pump when it is running in the current sandy water quality environment based on the amplitude wear coefficient and the wear intensity threshold.

[0061] Based on the wear rate, predict the timestamp when the current wear intensity reaches the wear intensity threshold, and record it as the predicted shutdown time. Generate a performance evaluation report for the water pump based on the predicted shutdown time.

[0062] In a preferred embodiment of the present invention, the process of setting the wear intensity threshold includes:

[0063] Based on the sample data, the wear intensity of each water pump sample at the end of use is obtained, and the average value of all wear intensities is obtained and recorded as the wear intensity threshold.

[0064] In a preferred embodiment of the present invention, the process of obtaining the wear rate of the water pump under the current sediment-laden water quality environment includes:

[0065] From the moment the water pump is put into use, the vibration signal of the pump casing is monitored in real time and recorded as the current vibration signal. The current transmitted vibration signal is obtained by preprocessing the current vibration signal. Based on the current transmitted vibration signal at each monitoring time, the impact index at each monitoring time is obtained and recorded as the current impact index. Based on the amplitude wear coefficient, the wear intensity at each monitoring time is obtained.

[0066] A new coordinate system is established with time as the horizontal axis and wear intensity as the vertical axis. The timestamp of each monitoring is obtained, and the timestamp of each monitoring and its corresponding wear intensity are fitted in the new coordinate system to obtain a wear intensity variation curve.

[0067] Several monitoring points were selected on the wear intensity variation curve, and the corresponding timestamps and wear intensity at each monitoring point were obtained to obtain the wear rate of the water pump under the current sediment-laden water quality environment. Where M is the total number of monitoring points, and Ws k ´ represents the current wear intensity of the kth monitoring point, k∈[1,M] and k is a positive integer, and t is the time interval between two monitoring points.

[0068] It is worth noting that the intelligent testing method proposed in this invention is particularly suitable for monitoring and evaluating the wear performance of water pumps under constant pumping rate conditions. In practical applications, such as large-scale water diversion projects, agricultural irrigation systems, and industrial circulating water facilities, water pumps are often required to operate stably at their rated operating point for a long time, maintaining constant flow and head output. Under these operating conditions with a constant pumping rate, the sand content in the water flow and its impact wear on the pump body become the main variables affecting the equipment's lifespan. This invention characterizes the sand particle impact intensity by monitoring high-frequency vibration signals, effectively eliminating vibration interference caused by flow fluctuations, making the wear assessment results more accurate and reliable. When the water pump is running at a constant speed, its internal flow field is relatively stable, and the movement trajectory and impact characteristics of sand particles exhibit certain regularity. This provides a stable physical basis for the mapping relationship between the amplitude distribution curve and wear intensity established in this invention, significantly improving the accuracy and practicality of wear rate prediction.

[0069] When the water pump is operating under conditions of changing pumping rate, this invention introduces a dynamic flow compensation mechanism to ensure the accuracy of wear rate calculation. Specifically:

[0070] Under operating conditions where the pumping rate varies, real-time flow data of the water pump is collected synchronously each time the current vibration signal is monitored. The real-time flow value is normalized to eliminate the direct impact of flow fluctuations on the amplitude of the vibration signal. Based on a regression model established in advance using sample data, combined with the normalized current transmitted vibration signal, the equivalent wear intensity under the current operating condition is obtained. By analyzing the trend of the equivalent wear intensity change during each monitoring, the dynamic wear rate of the water pump under varying operating conditions is obtained.

[0071] An intelligent testing system for water pump performance includes:

[0072] Signal acquisition unit: Vibration monitoring points are set on the pump casing of the water pump. The vibration monitoring points are used to collect the vibration signals of the water pump in real time when it is running in a sandy water environment; the vibration signals are preprocessed to obtain high-frequency conducted vibration signals.

[0073] Wear monitoring unit: Collects sample data, including the conducted vibration signals and wear intensity of all water pump samples during their operating cycle in a sandy water environment; Based on the sample data, obtains the amplitude distribution curves of all water pump samples, and obtains the amplitude wear coefficient based on the wear intensity corresponding to each amplitude distribution curve;

[0074] Performance evaluation unit: Set the wear intensity threshold, obtain the amplitude distribution curve of the water pump, and obtain the wear rate of the water pump when it is running in the current sandy water quality environment based on the amplitude wear coefficient and the wear intensity threshold;

[0075] Based on the wear rate, the timestamp when the current wear intensity reaches the wear intensity threshold is predicted and recorded as the predicted shutdown time. A performance evaluation report of the water pump is generated based on the predicted shutdown time.

[0076] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the present invention should still fall within the scope of the invention.

Claims

1. A smart testing method for water pump performance, characterized in that, Includes the following steps: Step S1: Install vibration monitoring points on the pump casing of the water pump. The vibration monitoring points are used to collect vibration signals of the water pump in real time when it is running in a sandy water environment. The vibration signals are preprocessed to obtain high-frequency conducted vibration signals. Step S2: Collect sample data, which includes the conducted vibration signals and wear intensity of all water pump samples during their operating cycle in a sediment-containing water environment; based on the sample data, obtain the amplitude distribution curve of all water pump samples, and obtain the amplitude wear coefficient based on the wear intensity corresponding to each amplitude distribution curve; Step S3: Set the wear intensity threshold, obtain the amplitude distribution curve of the water pump, and obtain the wear rate of the water pump when it is running in the current sandy water quality environment based on the amplitude wear coefficient and the wear intensity threshold. Based on the wear rate, predict the timestamp when the current wear intensity reaches the wear intensity threshold, and record it as the predicted shutdown time. Generate a performance evaluation report for the water pump based on the predicted shutdown time. In step S2, the process of obtaining the amplitude wear coefficient includes: Setting a threshold value N, for any amplitude distribution curve, a horizontal boundary line y=N is obtained in the coordinate system based on the threshold value, where y represents the ordinate in the coordinate system; the entire region enclosed by the horizontal boundary line and the amplitude distribution curve above the horizontal boundary line is then obtained, yielding the impact index. Where K is the transformation coefficient and K > 0, m is the total number of regions, and A x1 >A x2 A x1 A represents the first intersection point of the horizontal boundary line and the amplitude distribution curve in the x-th region. x2 Dis(A) represents the second intersection point of the horizontal boundary line and the amplitude distribution curve in the x-th region, and A represents the horizontal axis. An initial regression model is established based on a linear function of one variable with a direct proportion. The impact index of each transmitted vibration signal and its corresponding wear intensity are recorded as the training set. The training set is input into the initial regression model to train the initial regression model and obtain the final regression model. The slope of the regression model is obtained and recorded as the amplitude wear coefficient. In step S3, the process of obtaining the wear rate of the water pump under the current sediment-laden water quality environment includes: From the moment the water pump is put into use, the vibration signal of the pump casing is monitored in real time and recorded as the current vibration signal. The current transmitted vibration signal is obtained by preprocessing the current vibration signal. Based on the current transmitted vibration signal at each monitoring time, the impact index at each monitoring time is obtained and recorded as the current impact index. Based on the amplitude wear coefficient, the wear intensity at each monitoring time is obtained. A new coordinate system is established with time as the horizontal axis and wear intensity as the vertical axis. The timestamp of each monitoring is obtained, and the timestamp of each monitoring and its corresponding wear intensity are fitted in the new coordinate system to obtain a wear intensity variation curve. Several monitoring points were selected on the wear intensity variation curve, and the corresponding timestamps and wear intensity at each monitoring point were obtained to obtain the wear rate of the water pump under the current sediment-laden water quality environment. Where M is the total number of monitoring points, and Ws k ´ represents the current wear intensity of the kth monitoring point, k∈[1,M] and k is a positive integer, and t is the time interval between two monitoring points.

2. The intelligent testing method for water pump performance according to claim 1, characterized in that, In step S1, the preprocessing of the vibration signal includes: Set a passband frequency range, and set a high-frequency bandpass filter according to the passband frequency range; perform Fourier transform on all vibration signals to convert all vibration signals into frequency domain signals, filter all frequency domain signals through the high-frequency bandpass filter, remove vibration signals whose frequencies do not belong to the passband frequency range, and record the remaining vibration signals as conducted vibration signals.

3. The intelligent testing method for water pump performance according to claim 1, characterized in that, In step S2, the operating cycle is the time period between the time when the water pump sample is put into use and the time when it is put out of use. The time when the water pump sample is put into use is the timestamp corresponding to the time when the water pump sample is put out of use, and the time when it is put out of use is the timestamp corresponding to the time when the water pump sample is put out of use.

4. The intelligent testing method for water pump performance according to claim 3, characterized in that, In step S2, the process of collecting the wear intensity includes: The pump casing of the water pump that has not experienced wear is denoted as the unworn pump casing. An image of the unworn pump casing is obtained and denoted as the initial image. The initial image is divided into several pixels, each pixel is numbered, and the pixel value of each pixel is obtained. Each numbered pixel is associated with its corresponding pixel value to obtain the set of pixel values ​​of the initial image. Obtain an image of the pump casing of the water pump sample at the end of use, and denote it as a wear image; divide the wear image into the same number of pixels, denote them as wear pixels, assign the same number to each wear pixel, and obtain the pixel value of each wear pixel; associate the corresponding pixel values ​​of each numbered wear pixel to obtain the wear pixel value set of the wear image; The wear intensity is obtained based on the set of pixel values ​​and the set of worn pixel values. Where n is the total number of pixels in the pixel value set, RGB i Represents the i-th pixel in the set of pixel values, RGB i ´ represents the i-th pixel in the set of worn pixel values, where i∈[1,n] and i is a positive integer.

5. The intelligent testing method for water pump performance according to claim 1, characterized in that, In step S2, the process of obtaining the amplitude distribution curves of all water pump samples includes: For any water pump sample, acquire all conducted vibration signals of the water pump sample and obtain the maximum amplitude value among all conducted vibration signals; establish a coordinate system with amplitude as the abscissa and quantity as the ordinate; in the coordinate system, for any abscissa value, acquire the quantity of conducted vibration signals with the maximum amplitude value of that abscissa value, and record it as the ordinate value corresponding to that abscissa value; obtain a data point in the coordinate system based on the abscissa value and the ordinate value corresponding to the abscissa value; then, based on each abscissa value and its corresponding ordinate value, obtain several data points in the coordinate system, and connect each data point sequentially with a smooth curve, and record the curve as the amplitude distribution curve of the water pump sample.

6. An intelligent testing system for water pump performance, characterized in that, include: Signal acquisition unit: Vibration monitoring points are set on the pump casing of the water pump. The vibration monitoring points are used to collect the vibration signals of the water pump in real time when it is running in a sandy water environment; the vibration signals are preprocessed to obtain high-frequency conducted vibration signals. Wear monitoring unit: Collects sample data, including the conducted vibration signals and wear intensity of all water pump samples during their operating cycle in a sandy water environment; Based on the sample data, obtains the amplitude distribution curves of all water pump samples, and obtains the amplitude wear coefficient based on the wear intensity corresponding to each amplitude distribution curve; Performance evaluation unit: Set the wear intensity threshold, obtain the amplitude distribution curve of the water pump, and obtain the wear rate of the water pump when it is running in the current sandy water quality environment based on the amplitude wear coefficient and the wear intensity threshold; Based on the wear rate, predict the timestamp when the current wear intensity reaches the wear intensity threshold, and record it as the predicted shutdown time. Generate a performance evaluation report for the water pump based on the predicted shutdown time. The process of obtaining the amplitude wear coefficient includes: Setting a threshold value N, for any amplitude distribution curve, a horizontal boundary line y=N is obtained in the coordinate system based on the threshold value, where y represents the ordinate in the coordinate system; the entire region enclosed by the horizontal boundary line and the amplitude distribution curve above the horizontal boundary line is then obtained, yielding the impact index. Where K is the transformation coefficient and K > 0, m is the total number of regions, and A x1 >A x2 A x1 A represents the first intersection point of the horizontal boundary line and the amplitude distribution curve in the x-th region. x2 Dis(A) represents the second intersection point of the horizontal boundary line and the amplitude distribution curve in the x-th region, and A represents the horizontal axis. An initial regression model is established based on a linear function of one variable with a direct proportion. The impact index of each transmitted vibration signal and its corresponding wear intensity are recorded as the training set. The training set is input into the initial regression model to train the initial regression model and obtain the final regression model. The slope of the regression model is obtained and recorded as the amplitude wear coefficient. The process of obtaining the wear rate of the water pump under the current sediment-laden water quality environment includes: From the moment the water pump is put into use, the vibration signal of the pump casing is monitored in real time and recorded as the current vibration signal. The current transmitted vibration signal is obtained by preprocessing the current vibration signal. Based on the current transmitted vibration signal at each monitoring time, the impact index at each monitoring time is obtained and recorded as the current impact index. Based on the amplitude wear coefficient, the wear intensity at each monitoring time is obtained. A new coordinate system is established with time as the horizontal axis and wear intensity as the vertical axis. The timestamp of each monitoring is obtained, and the timestamp of each monitoring and its corresponding wear intensity are fitted in the new coordinate system to obtain a wear intensity variation curve. Several monitoring points were selected on the wear intensity variation curve, and the corresponding timestamps and wear intensity at each monitoring point were obtained to obtain the wear rate of the water pump under the current sediment-laden water quality environment. Where M is the total number of monitoring points, and Ws k ´ represents the current wear intensity of the kth monitoring point, k∈[1,M] and k is a positive integer, and t is the time interval between two monitoring points.

Images