Intelligent sniffing and positioning system for faults of water pump unit

Abstract

The invention relates to the technical field of data analysis, in particular to a water pump unit fault intelligent sniffing and positioning system which comprises an acquisition layer, an evaluation layer and a positioning layer. An input end water pumping image and an output end water drainage image of the water pump unit are collected through the collection layer, the collection layer synchronously senses operation vibration information of all working parts of the water pump unit and stores the collected images and vibration information, and the evaluation layer receives the images and vibration information stored in the collection layer and stores the images and vibration information in the evaluation layer. According to the method, the pump water image of the input end of the water pump unit and the drainage image of the output end of the water pump unit are collected, so that the existing technology for detecting the fault of the water pump unit by utilizing vibration sensing of the water pump unit is further optimized, and the fault detection predictability of the water pump unit is better; and the degree of dependence on sensing equipment during fault diagnosis of the water pump unit is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and particularly to an intelligent fault sniffing and positioning system for a water pump unit. Background Art

[0002] A water pump unit is a combination of equipment that converts mechanical energy into the potential energy or kinetic energy of water to achieve water transportation and lifting. It mainly consists of a water pump, a motor, and related auxiliary equipment. It is widely used in fields such as agricultural irrigation, industrial production, urban water supply and drainage, etc., and plays an indispensable role in ensuring the production and domestic water use, flood control and drainage, and the operation of industrial circulating water systems.

[0003] The invention patent application with the application number 202310984029.7 discloses an operation fault detection system for a water pump unit based on the Internet of Things, including: a fluid pressure sensor for monitoring the fluid pressure at the liquid phase inlet and outlet of the water pump; a flow velocity sensor for monitoring the fluid flow velocity at the liquid phase inlet and outlet of the water pump; an oil temperature sensor for monitoring the temperature of the lubricating oil entering the cooling device for cooling and the temperature of the lubricating oil discharged from the cooling device; a lubricating oil flow velocity sensor for monitoring the flow rate of the lubricating oil entering the cooling device; a memory for storing the relevant information collected by each sensor; a controller for judging whether there is a fault problem in the working water pump unit; the method for the controller to judge whether there is a fault problem in the working water pump unit includes the following steps: the first step is to monitor the real-time output power Ps of the water pump; the second step is to monitor the effective power Px of the cooling device; for a water pump unit, a number of sampling points are obtained, the abscissa of the sampling point is the real-time output power Ps when the corresponding water pump unit is working, the ordinate is the effective power Px of the corresponding cooling device, and after fitting a number of sampling points on the plane rectangular coordinate system, the corresponding fitting curve H is obtained.

[0004] This application aims to solve the problem of "when a fault occurs, it cannot be detected in time, which may lead to the further expansion of the fault impact when the water pump unit is abnormal, and the corresponding work cannot be completed, and in severe cases, it will affect the execution of relevant emergency work".

[0005] However, the existing technology has a high dependence on sensing devices for the fault detection of water pump units. When there are sensing errors in the sensing devices themselves, there must be errors in the fault detection results of the water pump units.

[0006] Therefore, we propose an intelligent fault sniffing and positioning system for a water pump unit. Summary of the Invention

[0007] In view of the above-mentioned disadvantages of the prior art, the present invention provides an intelligent sniffing and positioning system for pump unit faults, which solves the technical problems raised in the above-mentioned background art.

[0008] To achieve the above objectives, the present invention is realized through the following technical solutions:

[0009] An intelligent sniffing and positioning system for pump unit faults includes: an acquisition layer, an evaluation layer, and a positioning layer;

[0010] The input pump water image and the output drainage image of the pump unit are collected by the acquisition layer. The acquisition layer synchronously senses the operation vibration information of each working component of the pump unit and stores the collected images and vibration information. The evaluation layer receives the images and vibration information stored in the acquisition layer, identifies the pump unit fault risk factors based on the images, and evaluates whether the pump unit currently has a fault by combining the pump unit fault risk factors with the vibration information. The positioning layer receives the evaluation result of whether the pump unit currently has a fault in the evaluation layer and decides whether to determine the working components with faults on the pump unit based on the evaluation result;

[0011] The evaluation layer includes a receiving module, an identification module, and an evaluation module. The receiving module is used to receive the images and vibration information stored in the acquisition module. The identification module is used to obtain the images received by the receiving module and identify the pump unit fault risk factors based on the images. The evaluation module is used to obtain the vibration information in the receiving module, obtain the pump unit fault risk factors in the identification module, and evaluate whether the pump unit currently has a fault based on the pump unit fault risk factors and the vibration information;

[0012] The evaluation logic for whether the pump unit currently has a fault in the evaluation module is as follows:

[0013]

[0014] In the formula: G is the pump unit fault determination value; n is the total number of working components of the pump unit; K i is the vibration influence factor of the i-th pump unit operating component; A max , f max , ΔA max are the maximum values of the vibration amplitudes, vibration frequencies, and vibration amplitude change rates of all components; A i , f i , ΔA i are the vibration amplitude, vibration frequency, and vibration amplitude change rate of the i-th component; W A , W f , W ΔA are the weight coefficients; E max is the maximum value in the equilibrium index corresponding to each vibration information; a is the pump unit fault risk factor;

[0015] Among them, W A 、W f 、W ΔA are all positive numbers and their sum is 1. The larger the pump unit fault determination value G is, the healthier the pump unit is. On the contrary, it means that the pump unit has a higher fault risk. In the evaluation module, the pump unit fault determination threshold is user-defined and edited by the system end user. Based on the comparison between the pump unit fault determination threshold and G, it is determined whether the current pump unit is safe;

[0016] The identification logic of the pump unit fault risk factor in the identification module is as follows:

[0017]

[0018] In the formula: k1 and k2 are the measurement ratios corresponding to the pump water images collected in two directions; L0 and W0 are the length and width of the input end of the pump unit; L1 and L2 are the spans from the opening direction of the input end to the water area edge in the water area where the input end of the pump unit is located; the span from the surface of the input end to the bottom of the water area; M is the amount of water pumped into the input end of the pump unit per second; S is the opening area of the input end of the pump unit;

[0019]

[0020] In the formula: ε is the reference value used when calculating the pump unit fault risk factor; P1 and P2 are the total areas of the impurity images determined based on the rendering results in the pump water images collected in two directions; P OUT is the total area of the impurity image determined based on the rendering result in the drainage image corresponding to the output end of the pump unit; χ is the normalization factor;

[0021] Based on the time sequence, ε corresponding to each pump water image and drainage image is calculated, denoted as ε1, ε2, ε3,...;

[0022] a = ε1 × ε2 × ε2 ×...;

[0023] In the formula: a is the pump unit fault risk factor;

[0024] Among them, the normalization factor χ > 0. The larger the pump unit fault risk factor a is, the higher the blockage risk of the pump unit and the worse the transmission capacity. On the contrary, it means that the blockage risk of the pump unit is smaller and the transmission capacity is better.

[0025] Furthermore, the acquisition layer includes a camera module, a sensing module and a storage module. The camera module is used to collect the pump water image at the input end and the drainage image at the output end of the pump unit. The sensing module is used to sense the operation vibration information of each working component of the pump unit. The storage module is used to receive the images collected by the operation of the camera module and the vibration information sensed by the operation of the sensing module, and collect the images and vibration information.

[0026] Among them, the water pumping images at the input end and the water drainage images at the output end of the water pump unit collected by the camera module, that is, the water images of the first water pumped into the input end and the water images of the first water drained out of the output end each time the water pump unit starts. There are several sensing modules, and several sensing modules are respectively deployed on the surfaces of the working components of the water pump unit. When the storage module stores the water pumping images and the water drainage images, the images are marked with the acquisition timestamps of the images synchronously. When the storage module stores the vibration information, the vibration information is marked with the time domain of vibration perception and the name of the source working component synchronously.

[0027] Furthermore, there are three camera modules. One camera module is deployed at the output end of the water pump unit, and the remaining two camera modules are deployed at the input end of the water pump unit. The image acquisition directions of the two camera modules deployed at the input end of the water pump unit are perpendicular to each other in the same plane;

[0028] Among them, the parameter settings of the internal water flow transmission pipeline, the input end, and the output end of the water pump unit include:

[0029] The amount of water pumped in and out per second at the input end and the output end of the water pump unit is equal;

[0030] The water flow transmission pipeline is a multi-node bent pipeline, and the water storage amount of the water flow transmission pipeline section connected to the output end of the water pump unit is equal to the amount of water pumped into the input end per second.

[0031] Furthermore, after the storage module stores the water pumping images and the water drainage images, the water pumping images and the water drainage images are processed:

[0032] Both the water pumping images and the water drainage images are converted into grayscale images. A grayscale value determination interval for impurities in water is set, and based on the grayscale value determination interval for impurities in water, the impurity images in the water pumping images and the water drainage images converted into grayscale images are identified. The impurity images included in each image are rendered with the specified grayscale values that are not within the grayscale value interval to which the grayscale image belongs;

[0033] The water pumping images and the water drainage images converted into grayscale images for which rendering is performed are used to iterate the original water pumping images and water drainage images stored in the storage module, and the converted water pumping images and water drainage images for which rendering is performed are marked with the original corresponding image marking content synchronously;

[0034] Among them, the grayscale values applied to the rendering of the impurity images in the water pumping images and the water drainage images are the same, and the impurity images represent flocs and particulate matters in the images.

[0035] Furthermore, during the operation stage of the evaluation layer, the images stored in the acquisition layer correspond to at least the last three completed operation tasks of the water pump unit;

[0036] When the recognition module runs to recognize the fault risk factors of the water pump unit, the images applied are the pump water images and drainage images stored in the storage module, which are converted into grayscale images and completed rendering.

[0037] Further, the A max , f max , ΔA max are determined based on all the vibration information stored in the storage module, and the A i , f i , ΔA i are from the latest vibration information stored in the storage module;

[0038] The more critical the vibration influence factor corresponding to the operating components of the water pump unit during the operation of the water pump unit, the greater its value; conversely, the smaller its value.

[0039] The balance index of the vibration information is calculated by the following formula:

[0040]

[0041] In the formula: E is the balance index of the vibration information; u is the amplitude sequence of the vibration information; x i is the i-th amplitude in the vibration information; is the mean value;

[0042] Among them,

[0043] Further, the positioning layer includes a queue module, an inspection module, and an output module. The queue module is used to obtain the evaluation result of whether there is a fault in the water pump unit in the evaluation layer. When the evaluation result is yes, it

[0044] is sorted in descending order, simply denoted as g i , g i+1 , g i+2 ,... The inspection module is used to obtain g i , g i+1 , g i+2 ,... and sniff for the working components of the water pump unit with faults among the working components of the water pump unit corresponding to g i , g i+1 , g i+2 ,... The output module is used to output the working components with faults sniffed by the inspection module;

[0045] Among them, the output module is connected to the mobile computer device held by the system-end user through a wireless network, and outputs the working components with faults sniffed by the inspection module by their marked names.

[0046] Further, after the inspection module obtains g i 、g i+1 、g i+2 、... and then selects the middle item of g i 、g i+1 、g i+2 、... to perform offline detection on the corresponding working component, and determines whether there is a fault problem, and synchronously transmits the determination result back to the inspection module:

[0047] If the determination result is no, among g i 、g i+1 、g i+2 、..., the working component corresponding to the adjacent lower-level parameter of the previous selected parameter based on the descending queue is used as the offline detection target, and the detection operation is performed again. If the detection result is no, the working component is selected for detection again based on the above logic, and so on, until the detection result is yes. The working component and all the working components corresponding to the lower-level parameters of the parameter corresponding to the working component in the descending queue are recorded as faulty working components;

[0048] If the determination result is yes, among g i 、g i+1 、g i+2 、..., the working component corresponding to the adjacent upper-level parameter of the previous selected parameter based on the descending queue is used as the offline detection target, and the detection operation is performed again. If the detection result is yes, the working component is selected for detection again based on the above logic, and so on, until the detection result is no. All the working components corresponding to the lower-level parameters of the parameter corresponding to the working component in the descending queue are recorded as faulty working components.

[0049] Further, the receiving module is wirelessly interconnected with a storage module, the storage module is wirelessly interconnected with a sensing module and a camera module, the receiving module is wirelessly interconnected with an identification module and an evaluation module, the evaluation module is wirelessly interconnected with a queue module, and the queue module is wirelessly interconnected with an inspection module and an output module.

[0050] Adopting the technical solution provided by the present invention, compared with the known public technology, it has the following beneficial effects:

[0051] The present invention provides an intelligent sniffing and positioning system for pump unit failures. During operation, this system optimizes the existing technology of detecting pump unit failures by sensing the vibration of the pump unit by collecting the pump water images at the input end and the drainage images at the output end of the pump unit, making the predictability of pump unit failure detection better and reducing the dependence on sensing devices during pump unit failure diagnosis. Thus, it ensures that during the operation of the pump unit, more reliable failure diagnosis and sniffing can be obtained, and at the same time, the faulty working components on the pump unit are centrally located, thereby achieving a full-coverage failure maintenance effect for the working components of the pump unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings according to these drawings without creative efforts.

[0053] Figure 1 is a schematic structural diagram of an intelligent sniffing and positioning system for pump unit failures;

[0054] Figure 2 is a schematic diagram of the deployment position of the camera module in the system acquisition layer of the present invention;

[0055] The reference numerals in the figures represent respectively: 1, water flow transmission pipeline; 2, output end; 3, input end; 4, water suction volume per second at the input end; 5, drainage volume per second at the output end; 6, pump water image acquisition direction; 7, drainage image acquisition direction. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] In order to make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0057] The following further describes the present invention with reference to the embodiments.

[0058] Embodiment:

[0059] An intelligent sniffing and positioning system for pump unit failures in this embodiment, as Figure 1 shown, includes: an acquisition layer, an evaluation layer, and a positioning layer;

[0060] The water pumping images at the input end and the water drainage images at the output end of the water pump unit are collected by the acquisition layer. The acquisition layer synchronously senses the operating vibration information of each working component of the water pump unit and stores the collected images and vibration information. The evaluation layer receives the images and vibration information stored in the acquisition layer, identifies the fault risk factors of the water pump unit based on the images, and evaluates whether there is a current fault in the water pump unit by combining the fault risk factors of the water pump unit with the vibration information. The positioning layer receives the evaluation result of whether there is a current fault in the water pump unit in the evaluation layer and decides whether to determine the working components with faults on the water pump unit based on the evaluation result;

[0061] The acquisition layer includes a camera module, a sensing module, and a storage module. The camera module is used to collect the water pumping images at the input end and the water drainage images at the output end of the water pump unit. The sensing module is used to sense the operating vibration information of each working component of the water pump unit. The storage module is used to receive the images collected by the operation of the camera module and the vibration information sensed by the operation of the sensing module, and collect the images and vibration information;

[0062] Among them, the water pumping images at the input end and the water drainage images at the output end of the water pump unit collected by the camera module are the water images of the first water pumped into the input end and the first water discharged from the output end each time the water pump unit starts. There are several sensing modules, and several sensing modules are respectively deployed on the surfaces of each working component of the water pump unit. When the storage module stores the water pumping images and the water drainage images, it synchronously marks the images with the acquisition timestamps of the images. When the storage module stores the vibration information, it synchronously marks the vibration information with the time domain of vibration information perception and the name of the source working component;

[0063] After the storage module stores the water pumping images and the water drainage images, it processes the water pumping images and the water drainage images:

[0064] Convert both the water pumping images and the water drainage images into grayscale images, set the determination interval of the gray value of impurities in water, and identify the impurity images in the water pumping images and the water drainage images converted into grayscale images based on the determination interval of the gray value of impurities in water, and render the impurity images included in each image with the specified gray values that are not in the gray value interval to which the grayscale image belongs;

[0065] Iterate the water pumping images and the water drainage images converted into grayscale images that have been rendered in the storage module for the originally stored water pumping images and water drainage images, and synchronously mark the water pumping images and the water drainage images converted into grayscale images that have been rendered with the original marking content of the corresponding images;

[0066] Among them, the gray values applied to the rendering of the impurity images in the water pumping images and the water drainage images are the same, and the impurity images represent flocs and particulate matters in the images;

[0067] The evaluation layer includes a receiving module, an identification module, and an evaluation module. The receiving module is used to receive the images and vibration information stored in the acquisition module. The identification module is used to obtain the images received by the receiving module and identify the pump unit fault risk factors based on the images. The evaluation module is used to obtain the vibration information in the receiving module, obtain the pump unit fault risk factors in the identification module, and evaluate whether there is a current fault in the pump unit based on the pump unit fault risk factors and the vibration information;

[0068] The evaluation logic for whether there is a current fault in the pump unit in the evaluation module is as follows:

[0069]

[0070] In the formula: G is the pump unit fault determination value; n is the total number of working components of the pump unit; K i is the vibration influence factor of the i-th operating component of the pump unit; A max , f max , ΔA max are the maximum values of the vibration amplitudes, vibration frequencies, and vibration amplitude change rates of all components; A i , f i , ΔA i are the vibration amplitude, vibration frequency, and vibration amplitude change rate of the i-th component; W A , W f , W ΔA are the weight coefficients; E max is the maximum value among the equilibrium indicators corresponding to each vibration information; a is the pump unit fault risk factor;

[0071] Among them, W A , W f , W ΔA are all positive numbers and their sum is 1. The larger the pump unit fault determination value G, the healthier the pump unit. On the contrary, it means the higher the pump unit fault risk. In the evaluation module, the pump unit fault determination threshold is user-defined and edited by the system end. Based on the comparison between the pump unit fault determination threshold and G, it is determined whether the current pump unit is safe;

[0072] Through the above logical formula calculation, the calculation method of the pump unit fault determination value is further defined, ensuring the stable output of the pump unit fault determination value in the system and providing data support for the further operation of the positioning layer.

[0073] The identification logic of the pump unit fault risk factor in the identification module is as follows:

[0074]

[0075] Where: k1 and k2 are the measurement ratios corresponding to the pump water images collected in two directions; L0 and W0 are the length and width of the input end of the pump unit; L1 and L2 are the spans from the opening direction of the input end to the water area edge and from the surface of the input end to the bottom of the water area within the water area where the input end of the pump unit is located; M is the amount of water pumped into the input end of the pump unit per second; S is the opening area of the input end of the pump unit;

[0076]

[0077] Where: ε is the reference value used when obtaining the fault risk factor of the pump unit; P1 and P2 are the total areas of the impurity images determined based on the rendering results in the pump water images collected in two directions; P OUT is the total area of the impurity image determined based on the rendering result in the corresponding drainage image at the output end of the pump unit; χ is the normalization factor;

[0078] Based on the time series, ε corresponding to each pump water image and drainage image is obtained, denoted as ε1, ε2, ε3,...;

[0079] a = ε1 × ε2 × ε2 ×...;

[0080] Where: a is the fault risk factor of the pump unit;

[0081] Among them, the normalization factor χ > 0. The larger the fault risk factor a of the pump unit, the higher the blockage risk and the worse the transmission capacity of the pump unit. On the contrary, it indicates that the blockage risk of the pump unit is smaller and the transmission capacity is better;

[0082] A max , f max , ΔA max are determined based on all the vibration information stored in the storage module. A i , f i , ΔA i are from the latest vibration information stored in the storage module;

[0083] The more critical the vibration influence factor corresponding to the operating components of the pump unit during the operation of the pump unit, the larger its value. On the contrary, its value is smaller;

[0084] The balance index of the vibration information is calculated by the following formula:

[0085]

[0086] Where: E is the balance index of the vibration information; u is the amplitude sequence of the vibration information; x i is the i-th amplitude in the vibration information; is the mean value;

[0087] Among them,

[0088] Through the above logical formula calculation, the parameters applied in obtaining the fault determination value G of the water pump unit are obtained, further ensuring the completion of the calculation of the fault determination value G of the water pump unit.

[0089] The positioning layer includes a queue module, an inspection module, and an output module. The queue module is used to obtain the evaluation result of whether there is a fault in the water pump unit in the evaluation layer. When the evaluation result is yes,

[0090] perform a descending order sorting, simply denoted as g i 、g i+1 、g i+2 、... The inspection module is used to obtain g i 、g i+1 、g i+2 、... In g i 、g i+1 、g i+2 、... sniff for faulty working components of the water pump unit corresponding to the working components, and the output module is used to output the faulty working components sniffed by the inspection module;

[0091] Among them, the output module is connected to the mobile computer device held by the system-side user through a wireless network, and outputs the faulty working components sniffed by the inspection module by their marked names;

[0092] After the inspection module obtains g i 、g i+1 、g i+2 、... it selects the working component corresponding to the middle item of g i 、g i+1 、g i+2 、... for offline detection, and determines whether there is a fault problem, and synchronously transmits the determination result back to the inspection module:

[0093] If the determination result is no, in g i 、g i+1 、g i+2 、... the working component corresponding to the lower-level parameter adjacent to the previous selected parameter based on the descending queue is used as the offline detection target, and the detection operation is performed again. If the detection result is no, the working component is selected again based on the above logic for detection, and so on, until the detection result is yes. The working component and all the working components corresponding to the lower-level parameters of the parameter corresponding to the working component in the descending queue are all recorded as faulty working components;

[0094] If the determination result is yes, in g i 、g i+1 、g i+2, among others, the working component corresponding to the superior parameter adjacent to the previously selected parameter based on the descending queue is used as the offline detection target, and the detection operation is performed again. If the detection result is yes, the working component is selected for detection again based on the above logic, and so on, until the detection result is no. The working component corresponding to the parameter and all the working components corresponding to the subordinate parameters in the descending queue are recorded as faulty working components;

[0095] The receiving module is connected to the storage module through wireless network interaction. The storage module is connected to the sensing module and the camera module through wireless network interaction. The receiving module is connected to the recognition module and the evaluation module through wireless network interaction. The evaluation module is connected to the queue module through wireless network interaction. The queue module is connected to the inspection module and the output module through wireless network interaction.

[0096] In this embodiment, the camera module operates to collect the pump water image at the input end and the drainage image at the output end of the water pump unit. The sensing module synchronously senses the operating vibration information of each working component of the water pump unit. The storage module further receives the images collected by the operation of the camera module and the vibration information sensed by the operation of the sensing module, and collects the images and vibration information. The receiving module runs later to receive the images and vibration information stored in the collection module. Then the recognition module obtains the images received by the receiving module, and based on the images, identifies the fault risk factors of the water pump unit. The evaluation module operates to obtain the vibration information in the receiving module and obtain the fault risk factors of the water pump unit in the recognition module. Based on the fault risk factors of the water pump unit and the vibration information, it evaluates whether the water pump unit has a fault at present. Finally, the queue module obtains the evaluation result on whether the water pump unit has a fault in the evaluation layer. When the evaluation result is yes,

[0097] Perform a descending order sorting, simply denoted as g i 、g i+1 、g i+2 、..., the inspection module obtains g i 、g i+1 、g i+2 、..., among the working components of the water pump unit corresponding to g i 、g i+1 、g i+2 、..., sniff out the faulty working components of the water pump unit, and the output module outputs the faulty working components sniffed out by the inspection module.

[0098] Through the operation of the system in the above embodiment, a precise, predictive, and less sensor device-dependent water pump unit fault detection scheme is brought, ensuring that during the operation of the water pump unit, fault problems can be predicted in a timely and accurate manner, so as to use this as support for the daily maintenance of the water pump unit and ensure that the water pump unit can operate more stably and for a longer time.

[0099] As shown Figure 1 in the figure, there are three camera modules. One camera module is deployed at the output end of the water pump unit, and the remaining two camera modules are deployed at the input end of the water pump unit. The image acquisition directions of the two camera modules deployed at the input end of the water pump unit are perpendicular to each other in the same plane;

[0100] Among them, the parameter settings of the internal water flow transmission pipeline, input end, and output end of the water pump unit include:

[0101] The amount of water pumped in and out per second at the input end and output end of the water pump unit is equal;

[0102] The water flow transmission pipeline is a multi-node bent pipeline, and the water storage volume of the pipeline section connected to the output end of the water pump unit is equal to the amount of water pumped in per second at the input end.

[0103] Refer to Figure 2 the figure. This figure further shows the deployment positions of the camera modules in the system acquisition layer, which is intended to represent the acquisition directions of the water pumping image and the drainage image of the water pump unit during acquisition;

[0104] It should be noted that in the actual application scenario of the technical solution in this embodiment, the condition that the water suction volume 4 per second at the input end is equal to the water discharge volume 5 per second at the output end in this figure should be met.

[0105] As shown Figure 1 in the figure, during the operation stage of the evaluation layer, the images stored in the acquisition layer correspond to at least the last three completed operation tasks of the water pump unit;

[0106] When the recognition module runs to recognize the fault risk factors of the water pump unit, the images used are the pumped water images and drainage images stored in the storage module that have been converted into grayscale images and completed rendering.

[0107] Through the above settings, the specific data targets processed during the operation of the evaluation layer of the system in this embodiment are further defined, making the system operation more logical and targeted.

[0108] In summary, during the operation of the system in the above embodiment, by collecting the water pumping image at the input end and the drainage image at the output end of the water pump unit, it provides further optimization for the existing technology of using the vibration perception of the water pump unit to detect faults, making the fault detection predictability of the water pump unit better, and reducing the dependence on the perception device during the fault diagnosis of the water pump unit. Thus, it ensures that during the operation of the water pump unit, a more reliable fault diagnosis sniffing can be obtained, and at the same time, the faulty working parts on the water pump unit are centrally located, so as to achieve the full-coverage fault maintenance effect of the working parts of the water pump unit.

[0109] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than limiting it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent sniffing and positioning system for pump unit faults, characterized in that, It includes: An acquisition layer, an evaluation layer, and a positioning layer; The input pump water image and the output drain water image of the water pump unit are acquired by the acquisition layer. The acquisition layer synchronously senses the operation vibration information of each working component of the water pump unit and stores the acquired images and vibration information. The evaluation layer receives the images and vibration information stored in the acquisition layer, identifies the fault risk factors of the water pump unit based on the images, and evaluates whether there is a current fault in the water pump unit by combining the fault risk factors of the water pump unit with the vibration information. The positioning layer receives the evaluation result of whether there is a current fault in the water pump unit in the evaluation layer and decides whether to determine the working components with faults on the water pump unit based on the evaluation result; The evaluation layer includes a receiving module, an identification module, and an evaluation module. The receiving module is used to receive the images and vibration information stored in the acquisition module. The identification module is used to obtain the images received by the receiving module and identify the fault risk factors of the water pump unit based on the images. The evaluation module is used to obtain the vibration information in the receiving module and the fault risk factors of the water pump unit in the identification module, and evaluate whether there is a current fault in the water pump unit based on the fault risk factors of the water pump unit and the vibration information; The evaluation logic for whether there is a current fault in the water pump unit in the evaluation module is: Where: G is the failure determination value of the water pump unit; n is the total quantity of the working components of the water pump unit; K i is the vibration influence factor of the i-th operating component of the water pump unit; A max , f max , ΔA max are the maximum values of the vibration amplitude, vibration frequency and vibration amplitude change rate of all components; A i , f i , ΔA i are the vibration amplitude, vibration frequency and vibration amplitude change rate of the i-th component; W A , W f , W ΔA are the weight coefficients; E max is the maximum value in the balance index corresponding to each vibration information; a is the failure risk factor of the water pump unit; Among them, W A , W f , W ΔA are all positive numbers and their sum is 1. The larger the pump unit fault determination value G is, the healthier the pump unit is. On the contrary, it indicates a higher fault risk of the pump unit. In the evaluation module, the pump unit fault determination threshold is user-defined and edited by the system end user. Based on the comparison between the pump unit fault determination threshold and G, it is determined whether the current pump unit is safe.

2. The intelligent sniffing and positioning system for pump unit faults according to claim 1, characterized in that, The acquisition layer includes a camera module, a sensing module, and a storage module. The camera module is used to acquire the input pump water image and the output drain water image of the water pump unit. The sensing module is used to sense the operation vibration information of each working component of the water pump unit. The storage module is used to receive the images acquired by the operation of the camera module and the vibration information sensed by the operation of the sensing module, and collect the images and vibration information; Among them, the input pump water image and the output drain water image of the water pump unit acquired by the camera module are the water images of the first pump-in water at the input end and the first drained water at the output end each time the water pump unit starts. There are several sensing modules, and the several sensing modules are respectively deployed on the surfaces of each working component of the water pump unit. When the storage module stores the pump water image and the drain water image, it synchronously marks the images with the acquisition timestamp of the images. When the storage module stores the vibration information, it synchronously marks the vibration information with the perception time domain and the name of the source working component; 3. The intelligent fault sniffing and positioning system for a water pump unit according to claim 2, characterized in that, There are three camera modules. One camera module is deployed at the output end of the water pump unit, and the remaining two camera modules are deployed at the input end of the water pump unit. The image acquisition directions of the two camera modules deployed at the input end of the water pump unit are perpendicular to each other in the same plane; Among them, the parameter settings of the internal water flow transmission pipeline, the input end, and the output end of the water pump unit include: The amount of water pumped in and out per second at the input end and the output end of the water pump unit is equal; The water flow transmission pipeline is a multi-node bent pipeline, and the water storage volume of the pipeline section connected to the output end of the water pump unit is equal to the amount of water pumped in per second at the input end.

4. The intelligent fault sniffing and positioning system for a water pump unit according to claim 1, characterized in that After the storage module stores the pump water image and the drain water image, it processes the pump water image and the drain water image: Convert both the pump water image and the drainage image into grayscale images, set the gray value determination interval for impurities in water, and identify the impurity images in the pump water image and the drainage image converted into grayscale images based on the gray value determination interval for impurities in water. Render the impurity images included in each image with the gray values that are not within the gray value interval to which the grayscale image belongs; Iteratively store the original pump water image and the drainage image in the storage module with the pump water image and the drainage image converted into grayscale images after rendering, and synchronously mark the pump water image and the drainage image converted into grayscale images after rendering with the marking content of the original corresponding images; Among them, the gray values applied to the rendering of the impurity images in the pump water image and the drainage image are the same, and the impurity images represent flocs and particulate matters in the images.

5. An intelligent sniffing and positioning system for pump unit faults according to claim 1, characterized in that, During the operation stage of the evaluation layer, the images stored in the acquisition layer correspond to at least the last three completed operation tasks of the water pump unit; When the recognition module runs to recognize the fault risk factors of the water pump unit, the images used are the pump water image and the drainage image converted into grayscale images and completed with rendering stored in the storage module.

6. The intelligent sniffing and positioning system for pump unit faults according to claim 5, wherein The recognition logic of the fault risk factors of the water pump unit in the recognition module is as follows: In the formula: k1 and k2 are the measurement ratios corresponding to the pump water images collected in two directions; L0 and W0 are the length and width of the input end of the water pump unit; L1 and L2 are the spans from the opening direction of the input end to the water area edge and from the surface of the input end to the bottom surface of the water area within the water area where the input end of the water pump unit is located; M is the water volume pumped into the input end of the water pump unit per second; S is the opening area of the input end of the water pump unit; where: ε is the reference value used when obtaining the fault risk factor of the water pump unit; P1 and P2 are the total areas of the impurity images determined based on the rendering results in the pump water images collected in two directions; P OUT is the total area of the impurity image determined based on the rendering result in the drainage image corresponding to the output end of the water pump unit; χ is the normalization factor; Based on the time sequence, calculate ε corresponding to each pump water image and drainage image, denoted as ε1, ε2, ε3,...; a = ε1 × ε2 × ε2 ×...; In the formula: a is the fault risk factor of the water pump unit; Among them, the normalization factor χ > 0. The larger the fault risk factor a of the water pump unit, the higher the blockage risk and the worse the transmission capacity of the water pump unit. On the contrary, it means that the blockage risk of the water pump unit is smaller and the transmission capacity is better.

7. An intelligent sniffing and positioning system for pump unit faults according to claim 1, characterized in that, The said A max , f max , ΔA max is determined based on all the vibration information stored in the storage module, and the said A i , f i , ΔA i is derived from the latest vibration information stored in the storage module; The vibration influence factor corresponds to the operating components of the water pump unit. During the operation of the water pump unit, the more critical it is, the larger its value. On the contrary, its value is smaller; The balance index of the vibration information is calculated by the following formula: Where: E is the equilibrium index of vibration information; u is the amplitude sequence of vibration information; xi i is the i-th amplitude in the vibration information; is the mean value; Among them, 8. An intelligent fault sniffing and positioning system for a water pump unit according to claim 1, characterized in that, The positioning layer includes a queue module, an inspection module, and an output module. The queue module is used to obtain the evaluation result in the evaluation layer regarding whether there is a fault in the water pump unit. When the evaluation result is yes, for Sort in descending order, simply denoted as g i , g i+1 , g i+2 ,..., the inspection module is used to obtain g i , g i+1 , g i+2 ,..., among the working components of the pump unit corresponding to g i , g i+1 , g i+2 ,..., sniff for the working components of the pump unit with faults in the working components of the corresponding pump unit, and the output module is used to output the working components with faults sniffed in the inspection module; Among them, the output module is connected to the mobile computer device held by the system-side user through a wireless network, and outputs the working components with faults detected by the inspection module by their marked names.

9. An intelligent fault sniffing and positioning system for a water pump unit according to claim 8, characterized in that, The inspection module, after obtaining g i , g i+1 , g i+2 ,... selects the middle item of g i , g i+1 , g i+2 ,... to perform offline detection on the corresponding working component, and determines whether there is a fault problem, and synchronously transmits the determination result back to the inspection module: The determination result is no, at g i , g i+1 , g i+2 ,... Among them, based on the parameter selected last time, the working parts corresponding to the adjacent lower-level parameters in the descending queue are used as the offline detection targets, and the detection operation is performed again. If the detection result is no, the working parts are selected for detection again based on the above logic, and so on, until the detection result is yes. The working part and the working parts corresponding to all lower-level parameters of the parameter corresponding to the working part in the descending queue are all recorded as faulty working parts; The determination result is yes, at g i , g i+1 , g i+2 ,... Among them, based on the adjacent superior parameters of the descending queue with the previously selected parameter, the corresponding working component is used as the offline detection target, and the detection operation is performed again. If the detection result is yes, the working component is selected for detection again based on the above logic, and so on. Until the detection result is no, the working components corresponding to the parameters of this working component and all the subordinate parameters in the descending queue are recorded as faulty working components.

10. The intelligent fault sniffing and positioning system for a water pump unit according to claim 1, characterized in that, The receiving module is connected to the storage module through wireless network interaction. The storage module is connected to the sensing module and the camera module through wireless network interaction. The receiving module is connected to the recognition module and the evaluation module through wireless network interaction. The evaluation module is connected to the queue module through wireless network interaction. The queue module is connected to the inspection module and the output module through wireless network interaction.

Images