A photovoltaic pump group diagnosis method and system based on human-computer interaction

Abstract

The present invention relates to the technical field of centrifugal pump fault diagnosis, and specifically to a photovoltaic pump group diagnosis method and system based on human-computer interaction. The method includes: determining the battery working state of the photovoltaic pump group according to the output power of the photovoltaic device and the motor input power of the photovoltaic pump group; determining the reason for the reduction of the photovoltaic pump group inlet parameters according to the inlet parameters of the photovoltaic pump group and the sediment deposition rate; taking the liquid parameters obtained by the sensor and the maximum particle size calculated according to the filter mesh number as input, and the blade weight loss of the photovoltaic pump group as output to establish a blade loss model; taking the blade weight loss and liquid pressure as input, and the vibration parameters of the impeller as output to establish an impeller vibration model; calculating the vibration intensity according to the vibration parameters; setting an early warning level for the vibration intensity, and presetting disposal measures according to the early warning level. Automatic diagnosis of the vibration state of the photovoltaic pump group is achieved, reducing the number of times the impeller is disassembled to check the impeller fault, which is conducive to improving the use efficiency of the photovoltaic pump group.

Description

Technical Field

[0001] The present invention relates to the technical field of centrifugal pump fault diagnosis, and in particular to a photovoltaic pump group diagnosis method and system based on human-computer interaction. Background Art

[0002] The main structure of the photovoltaic pump group is a centrifugal water pump. The working principle is based on the action of centrifugal force. The water is thrown out through the impeller and enters the discharge pipe through the flow channel of the pump casing. A low-pressure area is formed at the center of the impeller. The water is sucked into the center of the impeller under the action of the atmospheric pressure difference, thus completing the pumping process. Since the water flow rate is very fast, if there are impurities in the water, such as sandy water flow, it will have a greater impact on the working efficiency of the centrifugal pump. When the sediment with a certain particle size and hardness passes through the centrifugal water pump at high speed, it will generate a considerable impact and abrasion force, which will grind the fine particles on the metal surface layer by layer and cause damage to the surface of the impeller blades, resulting in a reduction in the flow rate of the centrifugal pump, a reduction in head, a reduction in efficiency, and an increase in power, which will affect the life of the water pump. In severe cases, sediment siltation may even burn the motor.

[0003] However, due to the installation method of the water pump, impurities with uneven density in the water, and the distance between the impeller and the water suction port, it is difficult to achieve a consistent degree of wear on the impeller surface, resulting in changes in the axial force and radial force on the impeller. The balance of the impeller is destroyed, which reduces the working efficiency of the photovoltaic pump set. Accurate diagnosis of the photovoltaic pump set helps to determine reasonable disposal measures and maintain the normal working condition of the photovoltaic pump set. Traditional diagnostic methods rely on operators, but are limited by the operator's experience, and require the pump to be shut down and disassembled to check the wear of the impeller, which consumes a lot of time and labor costs. Therefore, there is an urgent need for a method and system for quickly diagnosing the operating status of a photovoltaic pump set. Summary of the invention

[0004] The purpose of the present invention is to provide a photovoltaic pump group diagnosis method and system based on human-computer interaction to reduce manual diagnosis and improve the diagnosis efficiency of photovoltaic pump groups, in view of the problem that the current manual diagnosis process of photovoltaic pump groups consumes a lot of time and is difficult to diagnose during operation.

[0005] In order to achieve the above-mentioned invention object, the present invention provides a photovoltaic pump group diagnosis method based on human-computer interaction, comprising:

[0006] Collecting the voltage and current of the photovoltaic pump group to calculate the output power of the photovoltaic device; calculating the motor input power according to the lift of the photovoltaic pump group; comparing the output power with the motor input power, and determining the battery working state of the photovoltaic pump group according to the comparison result;

[0007] Collecting inlet parameters of the photovoltaic pump group, the inlet parameters including inlet pressure and inlet flow; obtaining the sediment deposition rate of the photovoltaic pump group; if the inlet parameters continue to decrease, diagnosing the photovoltaic pump group according to the comparison result and the sediment deposition rate, and determining the reason for the decrease in the inlet parameters;

[0008] The liquid parameters and the vibration parameters of the photovoltaic pump group are obtained by sensors, wherein the liquid parameters include liquid pressure and liquid density; the maximum particle size of impurities in the liquid sucked by the photovoltaic pump group is calculated according to the mesh size of the filter; a blade loss model is established with the liquid parameters and the maximum particle size as input and the blade weight loss of the photovoltaic pump group as output;

[0009] An impeller vibration model is established with the blade weight loss and the liquid pressure as input and the vibration parameters as output; the vibration parameters of the photovoltaic pump group during operation are output through the impeller vibration model; a time window is preset, and when the time window is reached, the vibration intensity is calculated according to the vibration parameters; an early warning level is set for the vibration intensity, and disposal measures are preset according to the early warning level.

[0010] As a preferred technical solution of the present application, after obtaining the liquid parameters and the vibration parameters of the photovoltaic pump group through the sensor, the vibration amplitude of each blade is obtained, and when the time window is reached, the kurtosis is calculated according to the vibration amplitude; and whether the blade has block wear is determined according to whether the kurtosis increases.

[0011] As a preferred technical solution of the present application, it also includes: after the photovoltaic pump group is shut down, obtaining an image of the blade, extracting contour features of the block wear through image recognition, calculating the area and depth of the block wear according to the contour features, and calculating the weight of the block wear according to the area and the depth; and inputting the weight of the block wear into the impeller vibration model.

[0012] As a preferred technical solution of the present application, after extracting the contour features of the block wear through image recognition, the first center of mass of the block wear is calculated according to the pixel coordinates of the contour features; the third distance between the first center of mass and the impeller axis is obtained, and the ratio of the third distance to the impeller radius is used as a discount coefficient for the weight of the block wear.

[0013] As a preferred technical solution of the present application, after extracting the contour features of the block wear through image recognition, obtain the first pixel located at the edge part of the contour features and the corresponding depth; obtain the second pixel with the largest depth in the contour features; calculate the depth difference between the first pixel and the second pixel, calculate the horizontal distance according to the number of pixels between the first pixel and the second pixel, and calculate the slope according to the depth difference and the horizontal distance; take the standard deviation of all slopes as the wear degree of the block wear, and determine the disposal measures for the impeller according to the wear degree.

[0014] As a preferred technical solution of the present application, when the photovoltaic pump group is installed horizontally, the method for obtaining the liquid pressure also includes: dividing the circular area formed by the rotation of the impeller into a first area and a second area by a horizontal plane passing through the axial centerline of the impeller; the first area is the circular area above the horizontal plane, and the second area is the circular area below the horizontal plane; obtaining a first average value of the liquid pressure in the first area, obtaining a second average value of the liquid pressure in the second area, and taking the corresponding first average value and second average value as a group of liquid pressures.

[0015] As a preferred technical solution of the present application, when the impeller of the photovoltaic pump group is a multi-stage impeller, the vibration intensity of each stage of the impeller is obtained; and different first weights are set for the vibration intensity of each stage of the impeller according to the order in which the liquid flows through the impeller.

[0016] As a preferred technical solution of the present application, before obtaining the vibration intensity of each stage of impeller, the axial force balance point of the multi-stage impeller is obtained; the offset distance of the balance point is recorded, and the coefficient of variation of the multiple offset distances is used as the second weight; new vibration parameters are calculated based on the vibration parameters of each stage of impeller and the second weight; the new vibration parameters are superimposed to obtain overall vibration parameters, and the vibration intensity of the multi-stage impeller is calculated based on the overall vibration parameters.

[0017] As a preferred technical solution of the present application, before obtaining the vibration intensity of each stage of impeller, the axial force balance point of the multi-stage impeller is obtained; the third weight of the vibration parameter is set according to the fourth distance from the axis of each stage of impeller to the axial force balance point; new vibration parameters are calculated according to the vibration parameters of each stage of impeller and the third weight; the new vibration parameters are superimposed to obtain overall vibration parameters, and the vibration intensity of the multi-stage impeller is calculated according to the overall vibration parameters.

[0018] The present application also provides a photovoltaic pump group diagnostic system based on human-computer interaction, including the following modules:

[0019] A power module is used to collect the voltage and current of the photovoltaic pump group and calculate the output power of the photovoltaic device; receive the lift of the photovoltaic pump group from the human-computer interaction terminal and calculate the motor input power; compare the output power with the motor input power, and output the battery working status of the photovoltaic pump group according to the comparison result;

[0020] A flow fault judgment module is used to collect the inlet parameters of the photovoltaic pump group, the inlet parameters include inlet pressure and inlet flow; collect the sediment deposition rate of the photovoltaic pump group; if the inlet parameters continue to decrease, diagnose the photovoltaic pump group according to the comparison result and the sediment deposition rate, and determine the reason for the decrease in the inlet parameters;

[0021] The blade weight loss module is used to obtain liquid parameters and vibration parameters of the photovoltaic pump group through sensors, wherein the liquid parameters include liquid pressure and liquid density; receive the filter mesh number from the human-computer interaction terminal to calculate the maximum particle size of impurities in the liquid sucked by the photovoltaic pump group; and establish a blade loss model with the liquid parameters and the maximum particle size as input and the blade weight loss of the photovoltaic pump group as output;

[0022] The impeller vibration diagnosis module is used to establish an impeller vibration model based on the blade weight loss and the liquid pressure as input and the vibration parameters as output; output the vibration parameters of the photovoltaic pump group during the operating time through the impeller vibration model; receive a preset time window from the human-computer interaction terminal, and when the time window is reached, calculate the vibration intensity based on the vibration parameters; set an early warning level for the vibration intensity, and preset disposal measures based on the early warning level.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] 1. When the photovoltaic pump unit is pumping sandy liquid, the impeller is worn due to the impact of gravel on the impeller, which increases the vibration intensity of the impeller and affects the normal operation and service life of the photovoltaic pump unit. The weight loss of the impeller due to wear is predicted by the blade loss model, and the vibration parameters of the photovoltaic pump unit are predicted by the impeller vibration model according to the weight loss and liquid pressure. The vibration state of the photovoltaic pump unit is graded according to the vibration intensity calculated by the vibration parameters, which realizes automatic diagnosis of the vibration state of the photovoltaic pump unit and reduces the number of times the machine is disassembled to check for impeller failures. By quantitatively calculating the vibration intensity of the impeller, accurate classification of the vibration state is achieved, which is conducive to improving the use efficiency of the photovoltaic pump unit.

[0025] 2. Cavitation can cause block wear on the blade surface, and a pressure difference will be formed at the edge of the block wear, further increasing the probability of cavitation. By calculating the kurtosis of the blade vibration amplitude to determine whether block wear occurs on the blade surface, it is helpful to refine the factors affecting the diagnosis of the photovoltaic pump set. At the same time, it can also reduce the number of times the photovoltaic pump set is disassembled for diagnosis, which is helpful to extend the service life of the photovoltaic pump set.

[0026] 3. Block wear can cause a sudden change in blade weight, which in turn causes a sudden change in the vibration parameters of the impeller, resulting in a change in the classification of vibration severity. By predicting the change in vibration parameters caused by block wear during shutdown, it is helpful to diagnose the operating status of the photovoltaic pump set in advance, and take corresponding disposal measures in advance based on the diagnosis results, which is helpful to increase the service life of the photovoltaic pump set.

[0027] 4. The larger the third distance, that is, the larger the lever arm, the larger the torque generated by the block wear, and the greater the impact on the vibration parameters. Conversely, the closer the block wear is to the axis, the smaller the impact on the vibration parameters. Using the ratio of the lever arm length of the block wear to the impeller radius as a discount coefficient is conducive to improving the authenticity and accuracy of the training set of the impeller vibration model, thereby increasing the prediction accuracy of the impeller vibration model.

[0028] 5. The smoothness of block wear will affect the probability of cavitation of sand-containing liquid, further affect the vibration parameters of the impeller, and affect the operating efficiency of the photovoltaic pump set. The slope from the edge pixel to the pixel with the largest depth of the block wear image is calculated by the inverse tangent function, and the data discreteness of the slope value is used to reflect the smoothness of block wear, which is conducive to accurately assessing the wear state of the impeller, taking corresponding measures in time, and extending the service life of the impeller.

[0029] 6. Solid impurities in sandy liquids are affected by gravity or turbulence inside the liquid, and the impact forces on the impeller are different. In this application, the different impact forces measured are divided into two regions according to the influence of gravity, and the average values ​​in the two regions at the same time are used as a set of training data to train the blade loss model. Not only does it make the blade loss model closer to the actual impeller operating state, but it also increases the training data set, which can improve the accuracy of the blade loss model and further improve the accuracy of the impeller vibration model in predicting vibration parameters.

[0030] 7. Different impellers in a multi-stage impeller are subjected to different pressures, resulting in different effects of each impeller on the overall vibration intensity of the multi-stage impeller. Different weights are set for the vibration intensity of each impeller based on the different pressures on each stage of the impeller, making the calculation of the overall vibration intensity of the multi-stage impeller more accurate, which is conducive to accurate diagnosis of the photovoltaic pump set.

[0031] 8. The axial force balance of multi-stage impellers is destroyed due to the inconsistent liquid pressure and wear of each stage of the impeller, which leads to the intensification of the vibration of the multi-stage impeller. By recording the offset distance of the axial force balance point and using the coefficient of variation to indicate the offset degree of the axial force balance point, the technical features of the photovoltaic pump group for diagnosis are increased; the vibration caused by the offset of the axial force balance point and the original vibration parameters are superimposed, which is conducive to improving the accuracy of the vibration intensity of the multi-stage impeller.

[0032] 9. Due to the different fourth distances from each stage of the impeller to the axial force balance point and the components of the radial liquid pressure, the torque generated in the radial direction of the multi-stage impeller is different, which leads to the aggravation of the vibration of the multi-stage impeller. The fourth distance is the force arm of the impeller. The ratio of the fourth distance to the length of the impeller shaft represents the degree of influence of the impeller torque change on the vibration parameters, which increases the technical features of the photovoltaic pump group for diagnosis; superimposing the vibration caused by the torque change and the original vibration parameters is conducive to improving the accuracy of the vibration intensity of the multi-stage impeller. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is the structural diagram of the photovoltaic pump group.

[0034] Figure 2 The schematic diagram of the return pipe structure is as follows:

[0035] Figure 3 The following is a flow chart of diagnosing the photovoltaic pump group.

[0036] Figure 4 This is the structural diagram of the photovoltaic pump group diagnostic system.

[0037] Markings in the figure: 1- high-pressure hydraulic control valve, 2- pressure signal collection pipe, 3- high-pressure electric ball valve, 4- high-pressure steel expansion joint, 5- guide cover, 6- multi-stage impeller, 7- sand collection device, 8- water replenishment device, 9- motor, 10- electromagnetic flowmeter, 11- flexible joint, 12- gate valve, 13- return pipe. DETAILED DESCRIPTION

[0038] To make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be described clearly and completely in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them.

[0039] Therefore, the following detailed description of the embodiments of the present invention is not intended to limit the scope of the invention claimed for protection, but merely represents some embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features and technical solutions in the embodiments may be combined with each other.

[0041] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0042] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "upper", "lower", etc. is based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the invention product is usually placed when in use, or the orientation or positional relationship commonly understood by those skilled in the art. Such terms are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0043] Example 1: See Figure 1-Figure 3 As shown, this embodiment provides a photovoltaic pump group diagnosis method based on human-computer interaction, including:

[0044] The voltage and current of the photovoltaic pump group are collected to calculate the output power of the photovoltaic device; the motor input power is calculated according to the lift of the photovoltaic pump group; the output power and the motor input power are compared, and the battery working state of the photovoltaic pump group is determined according to the comparison result.

[0045] A photovoltaic pump group provided in this embodiment is as follows Figure 1 and Figure 2 As shown, the photovoltaic pump group includes a high-pressure hydraulic control valve 1, a pressure signal collection pipe 2, a high-pressure electric ball valve 3, a high-pressure steel expansion joint 4, a guide cover 5, a multi-stage impeller 6, a sand collection device 7, a water replenishment device 8, a motor 9, an electromagnetic flowmeter 10, a flexible joint 11, a gate valve 12 and a return pipe 13. It should be noted that although Figure 1 There is no indication therein, and those skilled in the art should know that the photovoltaic pump group also requires corresponding photovoltaic devices and energy storage devices for the energy supply of the photovoltaic pump group.

[0046] Furthermore, water inlets and outlets are provided at both ends of the photovoltaic pump group, so that vertical installation in water or horizontal installation on the ground can be realized. The multi-stage impeller 6 is wrapped by the guide cover 5, and rolling bearings are used between the multi-stage impeller 6 and the transmission shaft. When installed horizontally, the friction of the photovoltaic pump group's deadweight on the bearing can be reduced, wear can be reduced, and the maintenance workload of the photovoltaic pump group can be reduced. The multi-stage impeller is installed in a back-to-back structure, and the axial forces generated when rotating around the transmission shaft offset each other to achieve self-balancing. Since axial balancing measures such as balancing discs or balancing drums are eliminated, faults such as wear of the balancing disc and the balancing disc seat or seizure of the rotor are solved, and the service life of the pump is extended. The multi-stage impeller 6 is optimized in hydraulic performance and structural design, has the best guide vane matching design, reasonable matching clearance and a wider axial throttling design, ensuring that the water pump can still maintain a high degree of stability and high operating efficiency after long-term operation.

[0047] Furthermore, the deflector 5 is also used to assist the motor 9 in dissipating heat and reducing the rotor temperature. The water inlet of the deflector 5 adopts a 120-mesh filter screen, which can effectively prevent sand and stones from entering the mechanical structure inside the photovoltaic pump group and reduce the failure rate. The water replenishment device 8 is arranged on the outside of the deflector 5 and is connected to the inner cavity of the motor 9 through two pipelines. When the sensor detects that the silt in the sand collecting device 7 is deposited to the bottom height of the deflector 5, the high-pressure electric ball valve at the water outlet is automatically closed, and the electric ball valve on the return water pipeline 13 and the electric ball valve in the sand collecting device 7 are opened to automatically flush the silt deposited inside the pump casing and on the multi-stage impeller.

[0048] The voltage and current are acquired in real time by the voltage or current sensor installed in the photovoltaic device to calculate the output power of the photovoltaic device. The motor input power is calculated according to the lift of the photovoltaic pump group. The calculation formula is as follows:

[0049] ;

[0050] Wherein, P is the motor input power, m is the margin coefficient of the motor safety power, ranging from 1.1 to 1.3, a is the liquid density, Q is the flow rate of the photovoltaic pump group, g is the acceleration of gravity, H is the head of the photovoltaic pump group, and n is the efficiency of the photovoltaic pump group. The head is the height required for the photovoltaic water pump to lift water from one location to another.

[0051] Furthermore, the output power of the photovoltaic device is compared with the input power of the motor. If the output power is greater than the input power of the motor, it means that the power supply capacity of the photovoltaic pump group is sufficient, and the power generation of the photovoltaic device can not only meet the normal operation of the photovoltaic pump group, but also charge the battery of the photovoltaic pump group. If the output power is equal to the input power of the motor, it means that the power supply capacity of the photovoltaic pump group and the power consumption of the motor are equal, the photovoltaic pump group operates normally, and the battery does not participate in the operation of the photovoltaic pump group. If the output power is less than the input power of the motor, it means that the power supply capacity of the photovoltaic pump group is insufficient, and the photovoltaic pump group is difficult to operate normally alone, and the battery needs to be discharged to maintain the normal operation of the photovoltaic pump group.

[0052] By adjusting the battery working state of the photovoltaic pump group, the effective operating time can be increased and the operating efficiency of the photovoltaic pump group can be improved. Combined with the multi-stage impeller 6 after the hydraulic performance is optimized, efficient water pumping can be achieved under weak light conditions.

[0053] Collect inlet parameters of the photovoltaic pump group, the inlet parameters including inlet pressure and inlet flow; obtain the sedimentation rate of the photovoltaic pump group; if the inlet parameters continue to decrease, diagnose the photovoltaic pump group according to the comparison result and the sedimentation rate to determine the reason for the decrease in the inlet parameters.

[0054] The inlet pressure and the inlet flow are collected by a pressure sensor and a liquid sensor arranged at the impeller inlet. The sediment deposition speed is the ratio of the sediment height deposited in the sand collecting device 7 to the time. The sediment height is the difference between the first distance and the second distance measured by the ultrasonic sensor, the first distance being the distance from the bottom of the sand collecting device 7 measured by the ultrasonic sensor, and the second distance being the distance from the upper surface of the sediment in the sand collecting device 7 measured by the ultrasonic sensor.

[0055] When the inlet parameter collected by the liquid sensor decreases, if the motor input power is less than the output power, the human-computer interaction terminal prompts the motor to be insufficiently powered and needs to turn on the battery for energy supply; if the motor input power is greater than the output power and the sediment deposition rate is also reduced, the inlet of the photovoltaic pump group may be blocked or the inlet liquid level is too low, and the human-computer interaction terminal prompts the operator to perform a visual inspection. If the motor input power is greater than the output power and the inlet of the photovoltaic pump group is not blocked, the vibration signal of the impeller is further obtained to diagnose the photovoltaic pump group.

[0056] Liquid parameters and vibration parameters of the photovoltaic pump group are obtained through sensors, and the liquid parameters include liquid pressure and liquid density; the maximum particle size of impurities in the liquid sucked by the photovoltaic pump group is calculated according to the filter mesh size; a blade loss model is established with the liquid parameters and the maximum particle size as input and the blade weight loss of the photovoltaic pump group as output.

[0057] The applicable scenarios of the photovoltaic pump group provided in this embodiment include at least one heavy sediment liquid transportation scenario. The sand-containing liquid has a great influence on the characteristics of the photovoltaic pump group. When sediment with a certain particle size and hardness passes through the photovoltaic pump group at high speed, it will generate considerable impact and grinding force on the inner surface of the pump casing, the multi-stage impeller and other flow-through parts, and grind the fine particles on the metal surface layer by layer, resulting in wear on the surface of the flow-through parts, thereby causing the photovoltaic pump group to reduce flow, reduce head, and reduce efficiency. At the same time, impeller wear will lead to weight loss. The weight loss of the same impeller will be different due to the manufacturing material, processing quality, different liquid pressure changes and different sand density. The weight loss and wear position are also different: for a single-stage impeller, the balance in the radial direction will be broken; for a multi-stage impeller, the balance in the axial direction will also be broken. The impeller will vibrate when it is under liquid pressure and rotates around the axis. Impeller balance failure will lead to increased vibration.

[0058] Therefore, in this embodiment, the blade loss model is established to fit the process of blade weight loss, and then the impeller vibration model is used to fit the process of the influence of weight loss on impeller vibration. The photovoltaic pump group measures the liquid parameters and the vibration parameters in real time during operation, uses the pre-trained blade loss model and impeller vibration model to predict the vibration intensity of the impeller, and performs early warning classification and recommends corresponding disposal measures based on the prediction results. At the same time, in this embodiment, the impeller vibration model can also predict the vibration intensity of the next operation based on the recorded liquid parameters and vibration parameters after the photovoltaic pump group stops running, and determine the disposal measures taken in advance by the photovoltaic pump group based on the classification of the prediction results. Predicting the real-time vibration parameters based on the real-time measured liquid parameters can, on the one hand, accurately classify the early warning and take accurate disposal measures based on the real-time vibration parameters, and on the other hand, verify the vibration intensity predicted by the impeller vibration model, and perform online training to improve the accuracy of the model.

[0059] Furthermore, the liquid pressure and the liquid density are measured by a liquid sensor and a density sensor disposed at the impeller inlet. The filter mesh number is provided by the filter manufacturer, and the formula for calculating the maximum particle size d is as follows:

[0060] ;

[0061] Wherein, N is the number of filter meshes, C is a constant, and the value is usually 15000 in existing calculations. The photovoltaic pump group is installed on a bracket provided with a weight sensor. The photovoltaic pump group extracts the sand-containing liquid, and after the data value collected by the weight sensor reaches stability, the training of the blade loss model is started. The weight lost by the blade after wear is obtained by the difference of the data collected by the weight sensor twice.

[0062] In this embodiment, in order to reduce training interference, the sand-containing liquid used in training is a mixed liquid of impurities with a particle size of d and water. During training, each round of training time is set according to the first operating time of the photovoltaic pump group. The first operating time is the time required for the motor of the photovoltaic pump group to reach the maximum temperature when it operates normally. In each round of training, the liquid parameters collected in the first 2 / 3 of each round of training time are used as the training set, and the liquid parameters collected in the remaining 1 / 3 of the time are used as the verification set. The blade loss model uses MBE as the loss function for data prediction, and the MBE represents the difference between the real-time weight loss and the expected weight loss.

[0063] An impeller vibration model is established with the blade weight loss and the liquid pressure as input and the vibration parameters as output; the vibration parameters of the photovoltaic pump group during operation are output through the impeller vibration model; a time window is preset, and when the time window is reached, the vibration intensity is calculated according to the vibration parameters; an early warning level is set for the vibration intensity, and disposal measures are preset according to the early warning level.

[0064] The weight loss of blades will lead to the failure of impeller balance, causing the impeller to deflect and vibrate during operation. At the same time, the flow of liquid to the impeller will also generate pressure to cause the impeller to vibrate. Therefore, in this embodiment, the influence of the weight loss of blades and the liquid pressure on the vibration of the impeller is modeled to predict the vibration parameters of the impeller during the entire operation. The vibration parameters are measured by vibration sensors installed on the pump casing, impeller inlet and blades, and the vibration parameters include but are not limited to vibration speed, vibration amplitude and vibration frequency. The vibration intensity indicates the intensity of vibration, which is usually expressed by parameters that characterize the vibration level, such as the maximum value, average value or root mean square value of displacement, velocity and acceleration. In this embodiment, the root mean square value of the vibration speed is used. The time window is set by the operator of the photovoltaic pump group based on actual experience. When the time window is reached, all vibration parameters before the time window are obtained to calculate the vibration intensity, which is used to reflect the degree of vibration change of the photovoltaic pump group when it runs to the time window. According to the photovoltaic pump group, the warning level of the vibration intensity less than 1.8mm / s is recorded as good, and no treatment measures are required; the warning level of the vibration intensity greater than or equal to 1.8mm / s and less than 4.5mm / s is recorded as satisfactory, and no treatment measures are required; the warning level of the vibration intensity greater than or equal to 4.5mm / s and less than 11.2mm / s is recorded as qualified, and the speed needs to be reduced; the vibration intensity greater than or equal to 11.2mm / s is recorded as unqualified, and shutdown inspection is required.

[0065] When the photovoltaic pump unit extracts sandy liquid, the impeller is worn due to the impact of gravel on the impeller, which increases the vibration intensity of the impeller and affects the normal operation and service life of the photovoltaic pump unit. The weight loss of the impeller due to wear is predicted by the blade loss model, and the vibration parameters of the photovoltaic pump unit are predicted by the impeller vibration model based on the weight loss and liquid pressure. The vibration state of the photovoltaic pump unit is graded according to the vibration intensity calculated from the vibration parameters, thereby realizing automatic diagnosis of the vibration state of the photovoltaic pump unit and reducing the number of times the machine is disassembled to check for impeller failures. By quantitatively calculating the vibration intensity of the impeller, accurate classification of the vibration state is achieved, which is conducive to improving the use efficiency of the photovoltaic pump unit.

[0066] As a preferred technical solution of the present application, after obtaining the liquid parameters and the vibration parameters of the photovoltaic pump group through the sensor, the vibration amplitude of each blade is obtained, and when the time window is reached, the kurtosis is calculated according to the vibration amplitude; and whether the blade has block wear is determined according to whether the kurtosis increases.

[0067] During the use of the photovoltaic pump set, the pressure distribution in the pump is uneven due to the pump body structure, impeller design and complex liquid flow. The lowest point of pressure in the pump is near the inlet of the impeller blades. When the pressure of the sand-containing liquid at the impeller inlet is less than the saturated vapor pressure at this temperature, cavitation will form. The blade surface is repeatedly affected by the impact of cavitation, which will cause corrosion on the blade surface. When the pressure on the surface exceeds the metal fatigue strength of the impeller manufacturing material, block wear will occur on the surface.

[0068] The vibration amplitude is collected by the vibration sensor arranged on the blade. The kurtosis is used to measure the degree to which the discrete vibration amplitude deviates from the overall vibration amplitude data set. The farther the discrete vibration amplitude is from the mean of the vibration amplitude data set, the greater the influence on the calculation of the kurtosis. The greater the kurtosis, the more extreme values ​​there are in the vibration amplitude data set. In mechanical fault diagnosis, kurtosis is sensitive to impact signals and is suitable for the diagnosis of early surface damage faults. Therefore, in this embodiment, the block wear on the blade surface is detected by calculating the kurtosis of the vibration amplitude. The calculation formula of the kurtosis K is as follows:

[0069] ;

[0070] Among them, x i is the i-th vibration amplitude, x0 is the average amplitude of the vibration amplitude data set, and n is the size of the vibration amplitude data set. The block wear on the blade surface will cause a sudden change in the vibration parameters and a corresponding increase in the kurtosis value.

[0071] Cavitation can cause block wear on the blade surface, and a pressure difference will form at the edge of the block wear, further increasing the probability of cavitation. By calculating the kurtosis of the blade vibration amplitude to determine whether block wear occurs on the blade surface, it is helpful to refine the factors affecting the diagnosis of the photovoltaic pump set. At the same time, it can also reduce the number of times the photovoltaic pump set is disassembled for diagnosis, which is helpful to extend the service life of the photovoltaic pump set.

[0072] As a preferred technical solution of the present application, it also includes: after the photovoltaic pump group is shut down, if the blade has block wear, obtaining an image of the blade, extracting contour features of the block wear through image recognition, calculating the area and depth of the block wear according to the contour features, and calculating the weight of the block wear according to the area and the depth; and inputting the weight of the block wear into the impeller vibration model.

[0073] Since the edge of block wear easily increases the probability of cavitation, the wear on the blade surface is further increased, affecting the long-term use of the photovoltaic pump group. Therefore, after the photovoltaic pump group is shut down, an endoscope camera is used to capture an image of the blade surface, and the contour features of the block wear are extracted using an edge detection algorithm or a binary method. The area of ​​the block wear is the sum of the areas of the pixels surrounded by the contour features. The block wear is usually irregular in shape, and the depths of different pixel areas are also different. In this embodiment, the depth acquisition method includes but is not limited to stereo vision and monocular depth estimation. The block wear is divided into blocks in pixels, the depth of the center point of all pixel blocks is obtained, and the area of ​​the block wear is integrated with the depth of all center points to obtain the volume of the block area. Combined with the density of the impeller manufacturing material pre-set in the human-computer interaction terminal, the weight of the block wear is obtained, and the impeller vibration model is input to obtain the vibration intensity classification of the impeller during future operation, and corresponding disposal measures are taken according to the classification.

[0074] Block wear can cause a sudden change in blade weight, which in turn causes a sudden change in the vibration parameters of the impeller, resulting in a change in the classification of vibration severity. By predicting the change in vibration parameters caused by block wear during shutdown, it is helpful to diagnose the operating status of the photovoltaic pump set in advance, and take corresponding disposal measures in advance based on the diagnosis results, which is helpful to increase the service life of the photovoltaic pump set.

[0075] As a preferred technical solution of the present application, after extracting the contour features of the block wear through image recognition, the contour features of the block wear are extracted, and the first center of mass of the block wear is calculated according to the pixel coordinates of the contour features; the third distance between the first center of mass and the impeller axis is obtained, and the ratio of the third distance to the impeller radius is used as a discount coefficient for the weight of the block wear.

[0076] The vibration parameters are also affected by the third distance between the block wear and the impeller axis. The weight loss of the block wear will cause a change in the blade torque, and the third distance is the lever arm of the block wear. In this embodiment, the center of mass of the contour feature is used as the first center of mass. The product of the center point coordinates of each pixel in the contour feature and the area of ​​the corresponding pixel is calculated, and the ratio of the product result and the area of ​​the contour feature is the coordinates of the first center of mass. The axis is mapped to the coordinates of the contour feature, and then the third distance is calculated using the point-to-line distance formula. The impeller radius is pre-set by the operator at the human-computer interaction terminal.

[0077] The larger the third distance, that is, the larger the lever arm, the larger the torque generated by the block wear, and the greater the impact on the vibration parameters. Conversely, the closer the block wear is to the axis, the smaller the impact on the vibration parameters. Using the ratio of the lever arm length of the block wear to the impeller radius as a discount factor is conducive to improving the authenticity and accuracy of the training set of the impeller vibration model, thereby increasing the prediction accuracy of the impeller vibration model.

[0078] As a preferred technical solution of the present application, after extracting the contour features of the block wear through image recognition, obtain the first pixel located at the edge part of the contour features and the corresponding depth; obtain the second pixel with the largest depth in the contour features; calculate the depth difference between the first pixel and the second pixel, calculate the horizontal distance according to the number of pixels between the first pixel and the second pixel, and calculate the slope according to the depth difference and the horizontal distance; take the standard deviation of all slopes as the smoothness of the block wear, and determine the disposal measures for the impeller according to the smoothness.

[0079] The flow of sand-containing liquid through the block wear will cause changes in liquid parameters. The greater the degree of change in liquid parameters, the greater the probability of cavitation, and the more likely it is to affect the vibration parameters of the impeller. The change in liquid parameters is affected by the slope from the edge of the block wear to the deepest point. The gentler the slope, the more linear the change in liquid parameters, and the lower the probability of cavitation. In this embodiment, since the geometric parameters of the pixels are very small, the horizontal distance is estimated by multiplying the number of pixels by the length of each pixel. The slope from each first pixel to the second pixel of the edge of the contour feature is calculated using the inverse tangent function according to the depth difference and the horizontal distance. The degree of gentleness of the block wear is represented by the standard deviation of all slopes, and the disposal measures of the impeller are determined by the degree of gentleness. For example, in this embodiment, the degree of gentleness is graded according to the repair cost of the block wear. The grading value is set by the operator according to the actual benefits of the photovoltaic pump group and the price of the modified materials. The degree of gentleness is divided into three levels: satisfactory, acceptable and unacceptable. The corresponding disposal measures for the impeller are continued use, use at a reduced output height, and repair or replacement.

[0080] The smoothness of block wear will affect the probability of cavitation of sand-containing liquid, further affect the vibration parameters of the impeller, and affect the operating efficiency of the photovoltaic pump set. The slope from the edge pixel to the pixel with the largest depth of the block wear image is calculated by the inverse tangent function, and the data discreteness of the slope value is used to reflect the smoothness of block wear, which is conducive to accurately assessing the wear state of the impeller, taking corresponding measures in time, and extending the service life of the impeller.

[0081] Embodiment 2: The difference between this embodiment and embodiment 1 is that:

[0082] Furthermore, when the photovoltaic pump group is installed horizontally, the method for obtaining the liquid pressure also includes: dividing the circular area formed by the rotation of the impeller into a first area and a second area with a horizontal plane passing through the center line of the impeller; the first area is the circular area above the horizontal plane, and the second area is the circular area below the horizontal plane; obtaining a first average value of the liquid pressure in the first area, obtaining a second average value of the liquid pressure in the second area, and taking the corresponding first average value and second average value as a group of liquid pressures.

[0083] When the sand-containing liquid moves horizontally, the solid impurities such as gravel in the liquid will settle downward under the action of gravity. The more solid impurities the sand-containing liquid has, the greater the impact and abrasion force on the impeller surface, causing the impeller vibration parameters to change. At the same time, the impeller is subjected to impact and abrasion forces of different magnitudes, which will generate torque with the axis as the midpoint. As the impeller rotates, the torque changes continuously, which also causes the impeller vibration parameters to change. A rectangular coordinate system is established based on the same plane where the pressure sensor is located, with the straight line at the intersection of the horizontal plane and the plane where the pressure sensor is located as the x-axis, and the y-axis is perpendicular to the x-axis and intersects with the impeller axis. When the pressure sensor is located above the x-axis, the measured liquid pressure is marked as the first area; when the pressure sensor is located below the x-axis, the measured liquid pressure is marked as the second area. Calculate the first average value of the liquid pressure of the first area measured at the same time, calculate the second average value of the liquid pressure of the second area measured at the same time, and use the first average value and the second average value at the same time as a set of training data for the blade loss model.

[0084] Solid impurities in sandy liquids are affected by gravity or turbulence inside the liquid, and the impact forces on the impeller are different. In this application, the different impact forces measured are divided into two regions according to the influence of gravity, and the average values ​​of the two regions at the same time are used as a set of training data to train the blade loss model. Not only does it make the blade loss model closer to the actual impeller operating state, but it also increases the training data set, which can improve the accuracy of the blade loss model and further improve the accuracy of the impeller vibration model in predicting vibration parameters.

[0085] As a preferred technical solution of the present application, when the impeller of the photovoltaic pump group is a multi-stage impeller, the vibration intensity of each stage of the impeller is obtained; and different first weights are set for the vibration intensity of each stage of the impeller according to the order in which the liquid flows through the impeller.

[0086] When a photovoltaic pump unit uses a multi-stage impeller, the flow direction of the sand-containing liquid before being sucked in by the first-stage impeller is unstable, resulting in a large change in the liquid flow rate. The probability of cavitation is higher than that of other impellers, so the wear is the greatest. The sand-containing liquid is accelerated by multiple rotations of the multi-stage impeller, and the sand-containing liquid is continuously compressed, thereby converting energy into the flow rate and pressure of the sand-containing liquid. The last-stage impeller is most affected by the flow rate and pressure, so it is also more susceptible to wear. For impellers other than the first-stage impeller and the last-stage impeller, since the sand-containing liquid sucked in has been rectified by the previous-stage impeller, the wear is less than that of the first-stage impeller and the last-stage impeller.

[0087] Due to the different degrees of influence and wear of the liquid pressure on the impellers at each stage, the vibration parameters are also different, resulting in different degrees of influence on the overall vibration intensity of the multi-stage impeller. Therefore, in this embodiment, a different first weight is set for each stage of the impeller. The method for setting the weight includes but is not limited to: the operator manually sets it through the human-computer interaction terminal based on expert experience, so that the first weight of the first-stage impeller and the last-stage impeller is greater than the first weight of other impellers; the average liquid pressure of each stage of the impeller is collected over a period of time, and the average liquid pressure of each stage of the impeller is subjected to data normalization processing, and the absolute value of the obtained normalized value is used as the first weight, and the data normalization processing includes but is not limited to minimum-maximum normalization, Z-score normalization and decimal calibration normalization.

[0088] Different impellers in a multi-stage impeller are subjected to different pressures, resulting in different effects of each impeller on the overall vibration intensity of the multi-stage impeller. Different weights are set for the vibration intensity of each impeller based on the different pressures on each stage of the impeller, making the calculation of the overall vibration intensity of the multi-stage impeller more accurate, which is conducive to accurate diagnosis of the photovoltaic pump set.

[0089] As a preferred technical solution of the present application, before obtaining the vibration intensity of each stage of impeller, the axial force balance point of the multi-stage impeller is obtained; the offset distance of the balance point is recorded, and the coefficient of variation of the multiple offset distances is used as the second weight; new vibration parameters are calculated based on the vibration parameters of each stage of impeller and the second weight; the new vibration parameters are superimposed to obtain overall vibration parameters, and the vibration intensity of the multi-stage impeller is calculated based on the overall vibration parameters.

[0090] The reasons why the impeller generates axial force include but are not limited to: the areas of the front and rear cover plates that bear the liquid pressure are not equal, the magnitude and direction of the outflow speed of the sand-containing liquid change, and components are generated in both the axial and radial directions. In the axial direction, balancing measures need to be taken to offset the axial force, including but not limited to balancing holes, balancing plates or balancing drums. For multi-stage impellers, the balancing measure can also be to install the impellers symmetrically in the photovoltaic pump unit in a back-to-back manner. However, due to the different pressures and wear on each stage of the impeller, the balance of the axial force is broken, and the entire rotor part where the impeller is located experiences relative axial sliding, resulting in bearing wear or bending and deformation of the impeller shaft, further affecting the overall vibration intensity of the multi-stage impeller.

[0091] In this embodiment, the position of the axial force balance point is obtained by the pump group manufacturer through multiple experiments and measurements. When the photovoltaic pump group is running, the offset distance of the axial force balance point in the axial direction is measured by a distance sensor, and the coefficient of variation of multiple offset distances is calculated to obtain the second weight. The second weight is multiplied by the vibration parameter of each stage of the impeller, and the product result is superimposed on the vibration parameter to obtain a new vibration parameter. Multiple new vibration parameters are superimposed to obtain the overall vibration parameter of the multi-stage impeller, and then the vibration intensity of the multi-stage impeller is calculated.

[0092] The axial force balance of multi-stage impellers is destroyed due to the inconsistent liquid pressure and wear of each stage of impellers, which leads to the aggravation of the vibration of multi-stage impellers. By recording the offset distance of the axial force balance point and using the coefficient of variation to indicate the offset degree of the axial force balance point, the technical features of the photovoltaic pump group for diagnosis are increased; the vibration caused by the offset of the axial force balance point and the original vibration parameters are superimposed, which is conducive to improving the accuracy of the vibration intensity of the multi-stage impeller.

[0093] As a preferred technical solution of the present application, before obtaining the vibration intensity of each stage of impeller, the axial force balance point of the multi-stage impeller is obtained; the third weight of the vibration parameter is set according to the fourth distance from the axis of each stage of impeller to the axial force balance point; new vibration parameters are calculated according to the vibration parameters of each stage of impeller and the third weight; the new vibration parameters are superimposed to obtain overall vibration parameters, and the vibration intensity of the multi-stage impeller is calculated according to the overall vibration parameters.

[0094] In the radial direction, the multi-stage impeller has different components in the radial direction of the impeller due to the different pressures and wear on each stage of the impeller. At the same time, the fourth distance between each stage of the impeller and the axial force balance point is also different, resulting in different moments generated by each stage of the impeller relative to the axial force balance point, further affecting the overall vibration intensity of the multi-stage impeller.

[0095] In this embodiment, the fourth distance is obtained by the pump assembly manufacturer after multiple experiments and measurements. The fourth distance of each stage of the impeller is subjected to data standardization, and the absolute value of the standardized result is taken as the third weight; or the third weight is obtained by calculating the ratio of the fourth distance to the length of the impeller shaft. The third weight of each stage of the impeller is multiplied by the vibration parameter, and the product result is superimposed on the vibration parameter to obtain a new vibration parameter, and multiple new vibration parameters are superimposed to obtain the overall vibration parameter of the multi-stage impeller, and then the vibration intensity of the multi-stage impeller is calculated.

[0096] The vibration of multi-stage impellers is aggravated by the different fourth distances from each stage of impeller to the axial force balance point and the components of the radial liquid pressure, which generate different moments in the radial direction. The fourth distance is the force arm of the impeller. The ratio of the fourth distance to the length of the impeller shaft represents the degree of influence of the impeller moment change on the vibration parameters, which increases the technical features of the photovoltaic pump group for diagnosis; superimposing the vibration caused by the moment change and the original vibration parameters is conducive to improving the accuracy of the vibration intensity of the multi-stage impeller.

[0097] Example 3: See Figure 4 As shown, the difference between this embodiment and the first embodiment is that a photovoltaic pump group diagnostic system based on human-computer interaction is provided, including the following modules:

[0098] A power module is used to collect the voltage and current of the photovoltaic pump group and calculate the output power of the photovoltaic device; receive the lift of the photovoltaic pump group from the human-computer interaction terminal and calculate the motor input power; compare the output power with the motor input power, and output the battery working status of the photovoltaic pump group according to the comparison result;

[0099] A flow fault judgment module is used to collect the inlet parameters of the photovoltaic pump group, the inlet parameters include inlet pressure and inlet flow; collect the sediment deposition rate of the photovoltaic pump group; if the inlet parameters continue to decrease, diagnose the photovoltaic pump group according to the comparison result and the sediment deposition rate, and determine the reason for the decrease in the inlet parameters;

[0100] The blade weight loss module is used to obtain liquid parameters and vibration parameters of the photovoltaic pump group through sensors, wherein the liquid parameters include liquid pressure and liquid density; receive the filter mesh number from the human-computer interaction terminal to calculate the maximum particle size of impurities in the liquid sucked by the photovoltaic pump group; and establish a blade loss model with the liquid parameters and the maximum particle size as input and the blade weight loss of the photovoltaic pump group as output;

[0101] An impeller vibration diagnosis module is used to establish an impeller vibration model based on the blade weight loss and the liquid pressure as input and the vibration parameters as output; output the vibration parameters of the photovoltaic pump group during the operation time through the impeller vibration model; receive a preset time window from a human-computer interaction terminal, and when the time window is reached, calculate the vibration severity based on the vibration parameters; set an early warning level for the vibration severity, and preset disposal measures based on the early warning level;

[0102] The human-computer interaction module is used to connect with the above modules, collect data from the above modules and display it; provide the operator with a preset parameter setting window; display the diagnostic information and warning level of the photovoltaic pump group and prompt the preset disposal measures.

[0103] This embodiment also provides a storage medium and device for storing and executing the above method and system. It should be noted that the computer-readable storage medium of this embodiment can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0104] In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus, or device. In this embodiment, the computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable storage medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0105] The computer readable storage medium may be written in one or more programming languages ​​or a combination thereof to execute the computer program of the present embodiment, and the programming language includes an object-oriented programming language, such as Java, Python, C++, and a conventional procedural programming language, such as C or a similar programming language. The program may be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect through the Internet).

[0106] The above embodiments are only used to illustrate the present invention and are not intended to limit the technical solutions described in the present invention. Although the present invention has been described in detail with reference to the above embodiments, the present invention is not limited to the above specific implementation methods. Therefore, any modification or equivalent replacement of the present invention; and all technical solutions and improvements thereof that do not depart from the spirit and scope of the invention are included in the scope of the claims of the present invention.

Claims

1. A photovoltaic pump group diagnosis method based on human-computer interaction, characterized in that: include: Collecting the voltage and current of the photovoltaic pump group to calculate the output power of the photovoltaic device; calculating the motor input power according to the lift of the photovoltaic pump group; comparing the output power and the motor input power, and determining the battery working state of the photovoltaic pump group according to the comparison result; Collecting inlet parameters of the photovoltaic pump group, the inlet parameters including inlet pressure and inlet flow; obtaining the sediment deposition rate of the photovoltaic pump group; if the inlet parameters continue to decrease, diagnosing the photovoltaic pump group according to the comparison result and the sediment deposition rate, and determining the reason for the decrease in the inlet parameters; Acquire liquid parameters and vibration parameters of the photovoltaic pump group through sensors, wherein the liquid parameters include liquid pressure and liquid density; calculate the maximum particle size of impurities in the liquid sucked by the photovoltaic pump group according to the number of filter meshes; A blade loss model is established with the liquid parameters and the maximum particle size as input and the blade weight loss of the photovoltaic pump group as output; An impeller vibration model is established with the blade weight loss and the liquid pressure as input and the vibration parameters as output; the vibration parameters of the photovoltaic pump group during operation are output through the impeller vibration model; a time window is preset, and when the time window is reached, the vibration intensity is calculated according to the vibration parameters; an early warning level is set for the vibration intensity, and disposal measures are preset according to the early warning level.

2. A photovoltaic pump group diagnosis method based on human-computer interaction according to claim 1, characterized in that: After obtaining the liquid parameters and the vibration parameters of the photovoltaic pump group through the sensor, the vibration amplitude of each blade is obtained, and when the time window is reached, the kurtosis is calculated according to the vibration amplitude; and whether the blade has block wear is determined according to whether the kurtosis increases.

3. A photovoltaic pump group diagnosis method based on human-computer interaction according to claim 2, characterized in that: Also includes: After the photovoltaic pump group is shut down, an image of the blade is obtained, and contour features of the block wear are extracted through image recognition. The area and depth of the block wear are calculated based on the contour features, and the weight of the block wear is calculated based on the area and the depth; the weight of the block wear is input into the impeller vibration model.

4. A photovoltaic pump group diagnosis method based on human-computer interaction according to claim 3, characterized in that: After extracting the contour features of the block wear through image recognition, the first center of mass of the block wear is calculated according to the pixel coordinates of the contour features; the third distance between the first center of mass and the impeller axis is obtained, and the ratio of the third distance to the impeller radius is used as a discount coefficient for the weight of the block wear.

5. A photovoltaic pump group diagnosis method based on human-computer interaction according to claim 3, characterized in that: After extracting the contour feature of the block wear through image recognition, obtaining a first pixel located at the edge portion of the contour feature and the corresponding depth; obtaining a second pixel with the largest depth in the contour feature; calculating the depth difference between the first pixel and the second pixel, calculating the horizontal distance according to the number of pixels between the first pixel and the second pixel, and calculating the slope according to the depth difference and the horizontal distance; The standard deviation of all slopes is taken as the wear degree of the block wear, and the treatment measures for the impeller are determined according to the wear degree.

6. A photovoltaic pump group diagnosis method based on human-computer interaction according to claim 1, characterized in that: When the photovoltaic pump group is installed horizontally, the method for obtaining the liquid pressure also includes: dividing the circular area formed by the rotation of the impeller into a first area and a second area by a horizontal plane passing through the axial center line of the impeller; the first area is the circular area above the horizontal plane, and the second area is the circular area below the horizontal plane; obtaining a first average value of the liquid pressure in the first area, obtaining a second average value of the liquid pressure in the second area, and taking the corresponding first average value and second average value as a group of liquid pressures.

7. A photovoltaic pump group diagnosis method based on human-computer interaction according to claim 1, characterized in that: When the impeller of the photovoltaic pump group is a multi-stage impeller, the vibration intensity of each stage of the impeller is obtained; and different first weights are set for the vibration parameters of each stage of the impeller according to the order in which the liquid flows through the impeller.

8. A photovoltaic pump group diagnosis method based on human-computer interaction according to claim 7, characterized in that: Before obtaining the vibration severity of each stage of impeller, the axial force balance point of the multi-stage impeller is obtained; the offset distance of the balance point is recorded, and the coefficient of variation of the multiple offset distances is used as the second weight; new vibration parameters are calculated according to the vibration parameters of each stage of impeller and the second weight; the new vibration parameters are superimposed to obtain overall vibration parameters, and the vibration severity of the multi-stage impeller is calculated according to the overall vibration parameters.

9. A photovoltaic pump group diagnosis method based on human-computer interaction according to claim 7, characterized in that: Before obtaining the vibration severity of each stage of impeller, the axial force balance point of the multi-stage impeller is obtained; the third weight of the vibration parameter is set according to the fourth distance from the axis of each stage of impeller to the axial force balance point; new vibration parameters are calculated according to the vibration parameters of each stage of impeller and the third weight; the new vibration parameters are superimposed to obtain overall vibration parameters, and the vibration severity of the multi-stage impeller is calculated according to the overall vibration parameters.

10. A photovoltaic pump group diagnostic system based on human-computer interaction, characterized in that: Includes the following modules: A power module is used to collect the voltage and current of the photovoltaic pump group and calculate the output power of the photovoltaic device; receive the lift of the photovoltaic pump group from the human-computer interaction terminal and calculate the motor input power; compare the output power with the motor input power, and output the battery working status of the photovoltaic pump group according to the comparison result; A flow fault judgment module is used to collect the inlet parameters of the photovoltaic pump group, the inlet parameters include inlet pressure and inlet flow; collect the sedimentation speed of the photovoltaic pump group; if the inlet parameters continue to decrease, diagnose the photovoltaic pump group according to the comparison result and the sedimentation speed, and determine the reason for the decrease in the inlet parameters; The blade weight loss module is used to obtain liquid parameters and vibration parameters of the photovoltaic pump group through sensors, wherein the liquid parameters include liquid pressure and liquid density; receive the filter mesh number from the human-computer interaction terminal to calculate the maximum particle size of impurities in the liquid sucked by the photovoltaic pump group; A blade loss model is established with the liquid parameters and the maximum particle size as input and the blade weight loss of the photovoltaic pump group as output; The impeller vibration diagnosis module is used to establish an impeller vibration model based on the blade weight loss and the liquid pressure as input and the vibration parameters as output; output the vibration parameters of the photovoltaic pump group during the operating time through the impeller vibration model; receive a preset time window from the human-computer interaction terminal, and when the time window is reached, calculate the vibration intensity based on the vibration parameters; set an early warning level for the vibration intensity, and preset disposal measures based on the early warning level.

Images