Water pump wear ring wear diagnosis and wear degree prediction method and device and storage medium

Abstract

The invention relates to the technical field of mechanical fault diagnosis and computer artificial intelligence, in particular to a water pump wear ring wear diagnosis and wear degree prediction method and device and a storage medium. The method comprises: acquiring operation data of a water pump; judging whether the water pump has wear ring wear faults or not based on the operation characteristics calculated by the operation data; when a wear ring wear fault occurs, carrying out dimensionality reduction on the operation characteristics by adopting principal component analysis to obtain wear trend characteristics; and processing the wear trend characteristics based on a pre-constructed wear ring wear degree prediction model to obtain the wear ring wear degree. According to the embodiment of the invention, the wear ring fault diagnosis is carried out according to the operation characteristics extracted from the data during the wear ring operation, and when the existence of the fault is determined, the principal component analysis is adopted to carry out dimension reduction processing on the operation characteristics to obtain the wear trend characteristics, so that the pre-constructed wear ring wear degree prediction model is adopted to carry out wear ring wear prediction. The wear degree of the wear ring is accurately predicted, and a data basis is provided for the replacement time period of the wear ring.

Description

Technical Field

[0001] The invention relates to the field of mechanical fault diagnosis and computer artificial intelligence technology, and in particular to a method, device and storage medium for diagnosing wear of a water pump mouth ring and predicting the degree of wear. Background Art

[0002] Pumps are currently the most commonly used hydraulic machinery. They are mechanical equipment that converts energy by rotating blades to work on fluids. They are widely used in energy and power, aviation, marine areas, water conservancy and hydropower, environment, agricultural water and soil, petroleum, chemical industry and mining industries. With the continuous improvement of the technical level of the application fields related to centrifugal pumps, in addition to requiring centrifugal pumps to have basic hydraulic performance and anti-cavitation capabilities, higher requirements are also put forward for their high-efficiency area range and hydraulic dynamic characteristics.

[0003] The sealing ring is a common component of the centrifugal pump impeller seal. It is divided into flat ring type, right angle type and labyrinth type. Its main function is to limit the leakage of high-pressure liquid in the front and rear pump chambers to the low-pressure area of ​​the impeller inlet. Its size and shape have a great influence on the performance of the pump. In the long-term operation of the pump, the ring will inevitably wear a certain amount after being washed by the fluid for a long time. Once wear occurs, the overall hydraulic performance of the pump will drop sharply and it will gradually deviate from the operating requirements as the wear of the ring deepens, resulting in unnecessary economic losses.

[0004] At present, the industry mainly relies on manual labor to identify the wear fault of the mouth ring. This method is difficult to identify by manpower in the early stage of the fault. Often when the fault is serious, the operating personnel will pay attention to the abnormal reaction and deal with it. Moreover, the degree of wear of the mouth ring is difficult to judge by manpower, which makes it difficult to predict the time cycle for replacing the mouth ring when a fault occurs. In the existing intelligent technology, the commonly used rotating machinery fault diagnosis method collects signal data and pre-processes the data by empirical mode decomposition, local mean decomposition, wavelet packet decomposition and other methods. After pre-processing, the original data is subjected to feature extraction and fault identification. Its methods are diverse and not very applicable. Summary of the invention

[0005] In view of this, the present invention provides a method, device and storage medium for diagnosing water pump mouth ring wear and predicting the degree of wear, so as to solve the problem that the water pump mouth ring wear judgment method in the prior art is difficult to predict the degree of wear of the mouth ring.

[0006] In a first aspect, the present invention provides a method for diagnosing water pump mouth ring wear and predicting the degree of wear, the method comprising: acquiring operating data of the water pump; judging whether a mouth ring wear failure occurs in the water pump based on operating characteristics calculated from the operating data; when a mouth ring wear failure occurs, using principal component analysis to reduce the dimension of the operating characteristics to obtain wear trend characteristics; processing the wear trend characteristics based on a pre-constructed mouth ring wear degree prediction model to obtain the degree of mouth ring wear.

[0007] The method for diagnosing the wear of a water pump mouth ring and predicting the degree of wear provided by an embodiment of the present invention first performs a mouth ring fault diagnosis based on the operating characteristics extracted from the data when the mouth ring is in operation. When it is determined that a fault exists, principal component analysis is used to reduce the dimension of the operating characteristics to obtain wear trend characteristics. A pre-constructed mouth ring wear degree prediction model is then used in combination with the wear trend characteristics to predict the wear of the mouth ring, thereby achieving accurate prediction of the degree of wear of the mouth ring and providing a data basis for the time period for replacing the mouth ring.

[0008] In an optional embodiment, determining whether a ring wear failure occurs in a water pump based on operating characteristics calculated from operating data includes: constructing multiple performance curves based on operating characteristics calculated from operating data; and determining that a ring wear failure occurs in the water pump if multiple performance curves all decline compared to corresponding preset performance curves.

[0009] In an optional embodiment, the operating characteristics include head, efficiency and radial force, and the multiple performance curves include a flow-head curve, a flow-efficiency curve, and a flow-radial force curve.

[0010] In this embodiment, the operating characteristics calculated by the operating data include head, efficiency and radial force, thereby achieving accurate determination of the fault characteristics of the positioning ring wear fault. At the same time, the performance curve of the water pump is constructed according to the operating characteristics, and the ring wear fault is determined to have occurred when multiple performance curves decrease. This embodiment performs wear fault diagnosis through multiple performance curves, thereby achieving accurate judgment of the ring wear fault.

[0011] In an optional embodiment, the operating characteristics include head, efficiency and radial force, and the head, efficiency and radial force are respectively expressed by the following formulas:

[0012]

[0013] In the formula, H is the head, η is the efficiency, F is the radial force, p2, v2 is the outlet pressure and speed, p1, v1 is the inlet pressure and speed, n is the speed, M is the torque, D2 is the impeller outer diameter, B2 is the impeller outlet width including the cover plate, Q N The flow rate at the highest efficiency point.

[0014] In this embodiment, by using the above formula to calculate the operating characteristics such as head, efficiency and radial force, it is achieved to calculate the operating characteristics by theoretical calculation, thereby ensuring the accuracy of the calculation of the operating characteristics.

[0015] In an optional embodiment, when a mouth ring wear failure occurs, principal component analysis is used to reduce the dimension of the operating characteristics to obtain wear trend characteristics, including: when a mouth ring wear failure occurs, obtaining operating data collected at preset time intervals; performing mean processing on the collected operating data, and calculating interval operating characteristics; principal component analysis is used to reduce the dimension of the interval operating characteristics to obtain wear trend characteristics.

[0016] In this embodiment, since the degree of wear of the mouth ring changes relatively slowly, the operation data collected at preset time intervals is obtained, the collected operation data is averaged, and the interval operation characteristics are calculated; the interval operation characteristics are reduced in dimension using principal component analysis to obtain the wear trend characteristics, thereby reducing the amount of prediction calculations.

[0017] In an optional implementation, the buccal ring wear degree prediction model is obtained by training with a long short-term memory network, and the buccal ring wear degree prediction model updates parameters at preset time intervals.

[0018] In this embodiment, a long short-term memory network that is sensitive to time series is used as a trend quantity training model for the degree of wear of the oral ring. It can sensitively record characteristic information that changes over time and can achieve better results in predicting the degree of wear of the oral ring.

[0019] In a second aspect, the present invention provides a device for diagnosing the wear of a water pump mouth ring and predicting the degree of wear, the device comprising: a data acquisition module for acquiring the operating data of the water pump; a wear diagnosis module for judging whether a mouth ring wear failure occurs in the water pump based on the operating characteristics calculated from the operating data; a dimension reduction module for reducing the dimension of the operating characteristics by principal component analysis when a mouth ring wear failure occurs, so as to obtain wear trend characteristics; and a wear degree prediction module for processing the wear trend characteristics based on a pre-built mouth ring wear degree prediction model, so as to obtain the degree of mouth ring wear.

[0020] In a third aspect, the present invention provides a water pump mouth ring wear diagnosis and wear degree prediction system, comprising: a sensor for acquiring the operating data of the water pump; an acquisition card for converting the analog signal in the operating data into a digital signal; a processor for determining whether the water pump has a mouth ring wear failure based on the operating characteristics calculated from the operating data containing the digital signal; when a mouth ring wear failure occurs, principal component analysis is used to reduce the dimension of the operating characteristics to obtain wear trend characteristics; the wear trend characteristics are processed based on a pre-built mouth ring wear degree prediction model to obtain the mouth ring wear degree.

[0021] In a fourth aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the water pump mouth ring wear diagnosis and wear degree prediction method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0022] In a fifth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method for diagnosing water pump mouth ring wear and predicting the degree of wear according to the first aspect or any corresponding embodiment thereof.

[0023] In a sixth aspect, the present invention provides a computer program product, comprising computer instructions for causing a computer to execute the method for diagnosing water pump mouth ring wear and predicting the degree of wear according to the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0025] Figure 1 is a flow chart of a method for diagnosing wear of a water pump mouth ring and predicting the degree of wear according to an embodiment of the present invention;

[0026] Figure 2 is a flow chart of another method for diagnosing wear of a water pump mouth ring and predicting the degree of wear according to an embodiment of the present invention;

[0027] Figure 3 is a diagram showing the effect of predicting the degree of wear of the mouth ring according to an embodiment of the present invention;

[0028] Figure 4 is a structural block diagram of a device for diagnosing wear of a water pump mouth ring and predicting wear degree according to an embodiment of the present invention;

[0029] Figure 5 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order 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 clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0031] According to an embodiment of the present invention, a method embodiment of a water pump mouth ring wear diagnosis and wear degree prediction method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0032] In this embodiment, a method for diagnosing the wear of a water pump mouth ring and predicting the wear degree is provided, which can be used in electronic devices such as computers, mobile phones, tablet computers, etc. Figure 1 FIG. 1 is a flow chart of a method for diagnosing wear of a water pump mouth ring and predicting the degree of wear according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0033] Step S101, obtaining the operation data of the water pump. Specifically, the operation data of the water pump may be parameter data of the water pump obtained during the operation of the water pump. For example, in order to ensure the comprehensiveness and accuracy of the obtained data, the parameter data of the water pump under different operating conditions may be obtained, such as rated operating conditions, overload conditions, low load conditions, idling conditions, parallel operation conditions, and series operation conditions, etc.

[0034] The acquired operation data includes the flow rate, inlet pressure, outlet pressure, torque, speed, etc. of the water pump. Different sensors can be used to collect different operation data. For example, a flow meter is installed at the outlet of the water pump to collect the flow rate of the water pump, a pressure sensor is installed at the inlet and outlet of the impeller to collect the inlet pressure and outlet pressure, and a torque and speed sensor is installed at the connection end between the motor and the pump to collect the torque and speed, etc. Each sensor is connected to the acquisition card, and the acquisition card is connected to the host computer.

[0035] In this embodiment, the installation of each sensor must be accurately installed in accordance with the standards of each industry in order to obtain accurate data. The acquisition card connects the sensor through a preset channel. In this test application, channel one is the flow meter data interface, channel two is the water pump inlet pressure sensor data interface, channel three is the water pump outlet pressure sensor data interface, channel four is the speed sensor data interface, and channel five is the torque sensor data interface. A total of 5 channel data are converted into digital quantities by the acquisition card and input into the host computer. The host computer performs ring wear diagnosis and wear degree prediction.

[0036] Step S102, judging whether the water pump has a ring wear fault based on the operation characteristics calculated from the operation data. Specifically, the operation characteristics may be characteristics that characterize the operation performance of the water pump, and the operation characteristics may be calculated by feature extraction, and then the ring wear fault may be diagnosed based on the calculated operation characteristics.

[0037] Step S103, when a ring wear fault occurs, the operating characteristics are reduced in dimension by principal component analysis to obtain a wear trend characteristic. Specifically, when a wear fault is diagnosed based on the operating characteristics, the wear degree of the ring can be predicted, for example, the wear degree can be predicted by predicting the amount of ring wear. Before predicting the wear degree, the calculated operating characteristics can be reduced in dimension, for example, principal component analysis is used to reduce the dimension of the operating characteristics and the first principal component is extracted as the wear trend characteristic.

[0038] Step S104, the wear trend characteristics are processed based on the pre-constructed mouth ring wear degree prediction model to obtain the mouth ring wear degree. Specifically, the mouth ring wear degree prediction model can be constructed before the wear degree prediction is performed, so that after the wear trend characteristics are obtained, they can be input into the mouth ring wear degree prediction model to predict the mouth ring wear degree.

[0039] The method for diagnosing the wear of a water pump mouth ring and predicting the degree of wear provided by an embodiment of the present invention first performs a mouth ring fault diagnosis based on the operating characteristics extracted from the data when the mouth ring is in operation. When it is determined that a fault exists, principal component analysis is used to reduce the dimension of the operating characteristics to obtain wear trend characteristics. A pre-constructed mouth ring wear degree prediction model is then used in combination with the wear trend characteristics to predict the wear of the mouth ring, thereby achieving accurate prediction of the degree of wear of the mouth ring and providing a data basis for the time period for replacing the mouth ring.

[0040] In this embodiment, a method for diagnosing wear of a water pump mouth ring and predicting the degree of wear is provided, and the process includes the following steps:

[0041] Step S201, obtain the operation data of the water pump. Figure 1 Step S104 of the illustrated embodiment will not be described in detail here.

[0042] Step S202, judging whether the pump has a ring wear fault based on the operation characteristics calculated from the operation data. In this embodiment, the calculated operation characteristics specifically include head, efficiency and radial force, and the head, efficiency and radial force are respectively expressed by the following formulas:

[0043]

[0044] In the formula, H is the head, η is the efficiency, F is the radial force, p2, v2 is the outlet pressure and speed, p1, v1 is the inlet pressure and speed, n is the speed, M is the torque, D2 is the impeller outer diameter, B2 is the impeller outlet width including the cover plate, Q N The flow rate at the highest efficiency point.

[0045] In this embodiment, the head, efficiency and radial force are calculated by the above formula, which ensures the accuracy of the calculated operating characteristics.

[0046] Specifically, the above step S202 includes:

[0047] Step S2021, construct multiple performance curves based on the operation characteristics calculated from the operation data; the multiple performance curves include a flow-head curve, a flow-efficiency curve, and a flow-radial force curve. Specifically, after the head, efficiency, and radial force are calculated using the above formula, the flow data in the operation data can be combined, that is, the flow data is used as the horizontal coordinate, and the head, efficiency, and radial force are used as the vertical coordinate to construct the flow-head curve, the flow-efficiency curve, and the flow-radial force curve.

[0048] Step S2022, if multiple performance curves show a decrease compared to the corresponding preset performance curves, it is determined that the water pump has a mouth ring wear fault. Specifically, when judging the mouth ring wear fault, the constructed performance curve can be compared with the preset performance difference, wherein the preset performance curve can be the performance curve of the water pump provided in the parameter manual and other documents when the water pump leaves the factory. When comparing, the flow-head curve, flow-efficiency curve, and flow-radial force curve can be compared with the corresponding performance curves in the factory state, such as comparing the flow-head curve with the flow-head curve obtained according to the relevant data in the factory state. If the comparison result is that the constructed flow-head curve, flow-efficiency curve, and flow-radial force curve all have a downward trend compared to the preset performance curve, it is determined that a mouth ring wear fault has occurred. If only one or two performance curves decrease, it is determined that no mouth ring wear fault has occurred, and other faults may have occurred at this time.

[0049] Step S203: When a ring wear failure occurs, principal component analysis is used to reduce the dimension of the operation characteristics to obtain wear trend characteristics; see Figure 1 Step S104 of the illustrated embodiment will not be described in detail here.

[0050] Step S204: Process the wear trend characteristics based on the pre-built ring wear degree prediction model to obtain the ring wear degree. Figure 1 Step S104 of the illustrated embodiment will not be described in detail here.

[0051] In this embodiment, a method for diagnosing wear of a water pump mouth ring and predicting the degree of wear is provided, and the method comprises the following steps:

[0052] Step S301, obtain the operation data of the water pump; see Figure 1 Step S101 of the illustrated embodiment will not be described in detail here.

[0053] Step S302: Determine whether the pump has a ring wear fault based on the operation characteristics calculated from the operation data; see Figure 1 Step S102 of the illustrated embodiment will not be described in detail here.

[0054] Step S303: When a ring wear failure occurs, principal component analysis is used to reduce the dimension of the operation characteristics to obtain wear trend characteristics.

[0055] Specifically, the above step S303 includes:

[0056] Step S3031, when a ring wear fault occurs, obtain the operation data collected at a preset time interval; specifically, since the wear of the ring changes slowly when it is worn, in order to reduce the amount of calculated data, the wear degree of the ring can be predicted once at a preset interval. When the acquisition card is connected to the sensor for data acquisition, it is usually based on a certain sampling rate, that is, the acquisition card collects operation data at a preset time interval. In this embodiment, since the sampled signal is a low-frequency signal, the sampling rate is set to 10kHz.

[0057] Step S3032, average processing is performed on the collected operation data, and interval operation characteristics are calculated; specifically, the preset time for predicting the wear degree of the mouth ring includes multiple preset time intervals, that is, after collecting multiple operation data according to the preset time interval, a wear degree prediction can be performed. Therefore, the operation data collected multiple times can be averaged, that is, the flow rate, inlet pressure, outlet pressure, torque, speed and other data collected multiple times are averaged, and then the averaged data are substituted into the above-mentioned head, efficiency and radial force calculation formula to calculate the interval operation characteristics, that is, head, efficiency and radial force.

[0058] Step S3033: principal component analysis is used to reduce the dimension of the interval operation characteristics to obtain the wear trend characteristics. The specific process of principal component analysis includes:

[0059] 1. Remove the mean value and subtract the mean value of each feature from each dimension. Each dimension represents head, efficiency and radial force. By removing the mean value, it is possible to ensure that the covariance between features is calculated independently, so that each feature is distributed around the origin, that is, the mean value of each feature becomes 0.

[0060] 2. Calculate the covariance matrix.

[0061] Let x, y, and z represent the dimensions of lift, efficiency, and radial force, respectively. Then the covariance matrix C is as follows:

[0062]

[0063] The larger the absolute value of the covariance, the greater the impact on each other. Specifically, the covariance matrix reflects the covariance between each feature, that is, the correlation of their changing trends. If the covariance value of two features is large (positive), it means that they tend to increase or decrease together; if it is negative, when one increases, the other tends to decrease.

[0064] 3. Calculate the eigenvalues ​​and eigenvectors of the covariance matrix. Specifically, by solving the eigenvalue problem of the covariance matrix, a series of eigenvalues ​​and corresponding eigenvectors can be obtained. The eigenvalue reflects the variance of the data in the direction of the corresponding eigenvector, while the eigenvector is the basic direction of the covariance matrix transformation.

[0065] 4. Sort the eigenvalues ​​from large to small. Specifically, sort the calculated eigenvalues ​​from large to small. This is because larger eigenvalues ​​represent more variance information in the data, which is the part that is most desired to be retained during dimensionality reduction.

[0066] 5. Select the largest k eigenvalues ​​and their corresponding eigenvectors. Specifically, according to the required dimension reduction k, select the first k largest eigenvalues ​​and their corresponding eigenvectors. These eigenvectors form a new coordinate system.

[0067] 6. Convert the data to a new space constructed by k eigenvectors. Specifically, project the original data into the new space constructed by these k eigenvectors. This means that the original high-dimensional data is mapped to a low-dimensional space through a linear transformation, while keeping the variation information of the original data as much as possible.

[0068] The specific steps are:

[0069] Assume that the number of samples is m, the number of features is n, the sample matrix after subtracting the mean is DataAdjust(m*n), the covariance matrix is ​​n*n, and the matrix composed of the selected k eigenvectors is EigrnVectors(n*k). Then the projected data FinalData is

[0070] FinalData(m*k)=DataAdjust(m*n)×EigrnVectors(n*k)

[0071] 7. Take the first principal component data as the wear trend feature for predicting the wear degree of the mouth ring.

[0072] Step S304, based on the pre-built mouth ring wear degree prediction model, the wear trend characteristics are processed to obtain the mouth ring wear degree. Specifically, the mouth ring wear degree prediction model is trained using a long short-term memory network (Long Short-Term Memory, LSTM), and the mouth ring wear degree prediction model updates parameters at preset intervals.

[0073] Among them, the long short-term memory network includes five gates, namely input, output, forget gate, and two memory gates. Their respective calculation formulas are as follows:

[0074] Input Gate:

[0075] i t =σ(W i ·[h t-1 , x t ]+b i )

[0076] Forget Gate:

[0077] f t =σ(W f ·[h t-1 , x t ]+b f )

[0078]

[0079] Output Gate:

[0080] O t =σ(W o [h t-1 , x t ]+b o )

[0081] Long Memory:

[0082]

[0083] Short Memory:

[0084] h t =O t *tanh(C t )

[0085] It should be noted that the parameters in the above calculation formula can be determined by referring to the principle of long short-term memory network, which will not be repeated here.

[0086] In an optional embodiment, the parameters of the constructed ring wear prediction model can be updated at preset intervals. Specifically, at each interval, the data acquired within the preset time can be used as a training set to re-adjust and update the parameters of the model to ensure the accuracy of the model prediction.

[0087] In this embodiment, a water pump mouth ring wear diagnosis and wear degree prediction system is also provided, including:

[0088] Sensors, used to obtain operating data of the water pump;

[0089] An acquisition card is used to convert analog signals in the operating data into digital signals;

[0090] The processor is used to determine whether a ring wear failure occurs in a water pump based on an operation characteristic calculated from operation data containing a digital signal; when a ring wear failure occurs, the operation characteristic is reduced in dimension using principal component analysis to obtain a wear trend characteristic; the wear trend characteristic is processed based on a pre-built ring wear degree prediction model to obtain the ring wear degree.

[0091] As a specific application example of the embodiment of the present invention, Figure 2 As shown in the figure, the prediction system uses the following process to realize wear diagnosis and wear degree prediction:

[0092] S1) Install a flow meter at the outlet of the water pump, a pressure sensor at the inlet and outlet of the impeller, and a torque and speed sensor at the connection end between the motor and the pump, and connect each sensor to the acquisition card. The collected original signals such as flow Q, torque H, inlet pressure P1, outlet pressure P2 and speed n are uploaded to the acquisition card. The acquisition card converts the collected signals into digital quantities and inputs them into the host computer.

[0093] S2) In the host computer, a logical framework is written through LabVIEW to perform real-time data collection, storage, processing, diagnosis and prediction.

[0094] Specifically, the logical framework in the above step S2) is implemented by the following steps:

[0095] S21) storing the flow rate, inlet pressure, outlet pressure, torque, and speed of the water pump under actual operating conditions to obtain operating data including different states such as different working conditions.

[0096] S22) Analyze the operation data and draw the performance curve of the water pump. If the head, efficiency and radial force of each operating point decrease, it is judged that the water pump has a ring wear failure.

[0097] S23) Since the degree of wear of the mouth ring changes slowly, 10 minutes is set as the interval, and the operation data obtained multiple times within 10 minutes are averaged, and the characteristics of head, efficiency and radial force are extracted from the processed signal average.

[0098] S24) The proposed features are reduced in dimension through principal component analysis and the first principal component is taken to construct a trend-based unified quantity of the outlet ring wear degree, namely, the wear trend feature.

[0099] S25) The Long Short Term Memory (LSTM) network is used for sample training to obtain a prediction model for the wear degree of the mouth ring, and the current wear trend characteristics are input into the model to output the prediction result at the current moment. In the experiment, the data of the previous 4 days are set as the training set, and the prediction results are as follows: Figure 3 shown.

[0100] S3) Set the model update time, retrain the historical data, and update the model.

[0101] In this embodiment, a water pump mouth ring wear diagnosis and wear degree prediction device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0102] This embodiment provides a water pump mouth ring wear diagnosis and wear degree prediction device, such as Figure 4 As shown, including:

[0103] A data acquisition module 41 is used to acquire the operation data of the water pump;

[0104] A wear diagnosis module 42, used to determine whether a ring wear fault occurs in the water pump based on the operation characteristics calculated from the operation data;

[0105] A dimension reduction module 43 is used to reduce the dimension of the operation characteristics by using principal component analysis to obtain the wear trend characteristics when the ring wear failure occurs;

[0106] The wear degree prediction module 44 is used to process the wear trend characteristics based on the pre-built mouth ring wear degree prediction model to obtain the mouth ring wear degree.

[0107] The further functional description of each of the above modules is the same as that of the above corresponding embodiments and will not be repeated here.

[0108] The embodiment of the present invention also provides a computer device having the above Figure 4 The water pump mouth ring wear diagnosis and wear degree prediction device shown.

[0109] See also Figure 5 , Figure 5 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 5 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 5 A processor 10 is taken as an example.

[0110] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0111] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0112] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the use of a computer device based on the presentation of a small program landing page, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0113] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0114] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0115] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0116] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0117] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for diagnosing wear of a water pump mouth ring and predicting the degree of wear, characterized in that: The method comprises: Get the operation data of the water pump; Determining whether a ring wear failure occurs in the water pump based on the operation characteristics calculated from the operation data; When a ring wear failure occurs, principal component analysis is used to reduce the dimension of the operation characteristics to obtain wear trend characteristics; The wear trend characteristics are processed based on a pre-constructed mouth ring wear degree prediction model to obtain the mouth ring wear degree.

2. The method according to claim 1, characterized in that Determining whether a ring wear failure occurs in the water pump based on the operation characteristics calculated from the operation data includes: constructing a plurality of performance curves based on the operating characteristics calculated from the operating data; If multiple performance curves all show a decrease compared to the corresponding preset performance curves, it is determined that the water pump has a ring wear failure.

3. The method according to claim 2, characterized in that The operation characteristics include head, efficiency and radial force, and the multiple performance curves include flow-head curve, flow-efficiency curve, and flow-radial force curve.

4. The method according to claim 1, characterized in that The operation characteristics include head, efficiency and radial force, and the head, efficiency and radial force are respectively expressed by the following formulas: In the formula, H is the head, η is the efficiency, F is the radial force, p2, v2 is the outlet pressure and speed, p1, v1 is the inlet pressure and speed, n is the speed, M is the torque, D2 is the impeller outer diameter, B2 is the impeller outlet width including the cover plate, Q N The flow rate at the highest efficiency point.

5. The method according to claim 1, characterized in that When a ring wear failure occurs, principal component analysis is used to reduce the dimension of the operating characteristics to obtain wear trend characteristics, including: When a ring wear failure occurs, obtain the operating data collected at preset time intervals; The collected operation data is processed by averaging, and the interval operation characteristics are calculated; The principal component analysis is used to reduce the dimension of the interval operation characteristics to obtain the wear trend characteristics.

6. The method according to claim 1, characterized in that The mouth ring wear degree prediction model is obtained by long short-term memory network training, and the mouth ring wear degree prediction model updates parameters at preset time intervals.

7. A water pump mouth ring wear diagnosis and wear degree prediction device, characterized in that: The device comprises: A data acquisition module, used to acquire the operation data of the water pump; A wear diagnosis module, used for judging whether a ring wear fault occurs in the water pump based on the operation characteristics calculated from the operation data; A dimension reduction module is used to reduce the dimension of the operation characteristics by principal component analysis to obtain wear trend characteristics when a ring wear failure occurs; The wear degree prediction module is used to process the wear trend characteristics based on the pre-built mouth ring wear degree prediction model to obtain the mouth ring wear degree.

8. A water pump mouth ring wear diagnosis and wear degree prediction system, characterized in that: include: Sensors, used to obtain operating data of the water pump; An acquisition card, used for converting analog signals in the operation data into digital signals; A processor, for determining whether a ring wear fault occurs in the water pump based on an operation feature calculated from operation data including a digital signal; when a ring wear fault occurs, principal component analysis is used to reduce the dimension of the operation feature to obtain a wear trend feature; The wear trend characteristics are processed based on a pre-constructed mouth ring wear degree prediction model to obtain the mouth ring wear degree.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the water pump mouth ring wear diagnosis and wear degree prediction method according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the water pump mouth ring wear diagnosis and wear degree prediction method according to any one of claims 1 to 7.

Images