Fault identification method, device and medium for shielded pump of closed structure reactor

Abstract

The present invention discloses a method, device, and medium for identifying faults in a closed-structure reactor shielded pump, comprising: establishing a guide bearing model for the reactor shielded pump; obtaining the bearing capacity at a specific position and speed based on the guide bearing model; establishing a rotor dynamics model, dynamically coupling the rotor dynamics model with the guide bearing model to obtain a rotor-bearing coupled dynamics model, and obtaining first nonlinear transient dynamic response data of the system; loading a fault excitation into the rotor-bearing coupled dynamics model to obtain second nonlinear transient dynamic response data; establishing a pump casing transient dynamics model, loading the second nonlinear transient dynamic response data into the pump casing transient dynamics model to calculate third nonlinear transient dynamic response data; and further employing an artificial intelligence algorithm for fault pattern identification. The present invention can obtain effective fault data and improves the accuracy and efficiency of fault identification.

Description

Technical Field

[0001] The present invention relates to the technical field of reactor shielded pump fault identification, and in particular to a closed-structure reactor shielded pump fault identification method, device and medium. Background Art

[0002] Rotating machinery, such as shielded pumps, is essential for the safe and efficient operation of reactor systems. For example, the primary pump in the reactor's primary circuit drives the coolant circulation to cool the core and transfer heat. Its operational stability is crucial to the safe and efficient operation of the entire reactor system.

[0003] Abnormal vibration in rotating machinery, such as canned motor pumps, can be caused by a variety of factors, including rotor mass eccentricity and bearing wear. These factors can result in non-stable rotor motion, posing a safety hazard to reactor operation. However, due to the enclosed nature of rotating machinery like canned motor pumps, it is impossible to directly detect faults in internal components. The only way to analyze the type and severity of internal faults is through the vibration response signal from the pump casing foundation. This makes it difficult to obtain valid fault data, and fault information is easily lost in a multitude of interfering signals.

[0004] At present, the research on fault identification of rotating mechanical equipment such as reactor system shield pumps is mainly based on fault simulation experiments, which requires the construction of a large-scale experimental platform, consumes a lot of manpower, material and financial resources, is costly, and effective fault data is difficult to obtain, which leads to low fault identification accuracy and efficiency. Summary of the Invention

[0005] The technical problem to be solved by the present invention is that the existing fault identification for rotating mechanical equipment such as reactor system shielded pumps is mainly based on test data, and effective fault data is difficult to obtain, which leads to the problem of low accuracy and efficiency of fault identification. The purpose of the present invention is to provide a method, device and medium for fault identification of a closed-structure reactor shielded pump. By establishing a multi-module coupled dynamic model of the reactor shielded pump and simulating the response excitation of internal faults on the basis of the pump casing, more effective fault dynamic response data of the reactor shielded pump are obtained; based on this, the pump fault response characteristics are analyzed, and fault pattern identification is further performed based on artificial intelligence algorithms, laying the foundation for digital operation and maintenance of shielded pump-type rotating machinery. The present invention can obtain effective fault data and improve the accuracy and efficiency of fault identification.

[0006] The present invention is achieved through the following technical solutions:

[0007] In a first aspect, the present invention provides a method for identifying a fault of a canned pump in a closed structure reactor, the method comprising:

[0008] Based on fluid dynamics theory, a guide bearing model for a reactor canned pump is established. Based on the guide bearing model, the bearing load capacity at a certain position and speed is obtained.

[0009] The rotor dynamics model of the reactor canned motor pump is established, and the rotor dynamics model is dynamically coupled with the guide bearing model to obtain the rotor-bearing coupled dynamics model of the reactor canned motor pump.

[0010] Based on the rotor-bearing coupled dynamics model, obtaining the first nonlinear transient dynamic response data of the system; the system refers to the rotor-bearing coupled dynamics system;

[0011] The fault excitation is loaded into the rotor-bearing coupled dynamic model to obtain the second nonlinear transient dynamic response data; and a transient dynamic model of the reactor canned pump casing is established;

[0012] The second nonlinear transient dynamic response data is loaded into the pump casing transient dynamic model to calculate the third nonlinear transient dynamic response data of the pump casing of the reactor canned pump under internal fault excitation;

[0013] Based on the third nonlinear transient dynamic response data, the fault response characteristics are analyzed; and artificial intelligence algorithms are used to identify fault patterns.

[0014] Based on mechanism analysis and physical modeling, this invention establishes a system dynamics coupling model for rotating machinery such as canned motor pumps, develops a dynamic response calculation method, and obtains fault dynamic response data (third nonlinear transient dynamic response data) and fault response characteristics. Furthermore, artificial intelligence technology is used to identify equipment fault patterns, laying the foundation for digital operation and maintenance of canned motor pumps. By obtaining mechanism simulation data through the system dynamics coupling model, this invention can obtain effective fault data, improving the accuracy and efficiency of fault identification.

[0015] Furthermore, in the guide bearing model, the lubricating material of the guide bearing is water, which has low viscosity but static pressure; the inlet and outlet of the guide bearing is shaft end inlet and outlet, with one shaft end inlet and the other shaft end outlet.

[0016] Furthermore, the rotor dynamics model is dynamically coupled with the guide bearing model to obtain the rotor-bearing coupled dynamics model of the reactor canned pump, including:

[0017] The rotor axis position and speed at the current moment are calculated through the rotor dynamics model, and the rotor axis position and speed at the current moment are input into the guide bearing model to calculate the bearing capacity at the next time step;

[0018] The bearing capacity at the next time step is used as the input of the rotor dynamics model at the next time step. The rotor axis position and speed at the next time step are calculated based on the rotor dynamics model.

[0019] According to the above steps, it is iterated cyclically with time steps to realize the rotor-bearing coupled dynamics analysis and obtain the rotor-bearing coupled dynamics model of the reactor canned pump.

[0020] Furthermore, the first nonlinear transient dynamic response data includes rotor displacement, rotor speed, rotor acceleration and bearing load.

[0021] Furthermore, the fault excitation is loaded into the rotor-bearing coupled dynamic model to obtain the second nonlinear transient dynamic response data, including:

[0022] The fault excitation is loaded into the rotor-bearing coupled dynamic model to simulate the nonlinear dynamic response of the rotor and the nonlinear change of the bearing load under the fault excitation as the second nonlinear transient dynamic response data.

[0023] Furthermore, the method uses MATLAB software to establish a guide bearing model, and uses ABAQUS, COMSOL or ANSYS software to establish a rotor dynamics model and a pump casing transient dynamics model; then uses PYTHON or C++ language to perform secondary development on the above software to achieve dynamic coupling between the guide bearing model, rotor dynamics model and pump casing transient dynamics model under multi-physical fields, and obtain a nonlinear dynamic coupling model of the rotor-bearing-pump casing system of a closed-structure reactor shield pump.

[0024] Furthermore, based on the third nonlinear transient dynamic response data, an artificial intelligence algorithm is used to perform fault mode identification, including:

[0025] Different pump faults are injected into the nonlinear dynamic coupling model of the rotor-bearing-pump casing system to directly simulate the dynamic response of the fault excitation on the pump casing and obtain key characteristic parameters. Key characteristic parameters are characteristic parameters that can characterize different faults of reactor canned pump rotors, such as eccentricity, misalignment, and bearing wear.

[0026] Based on key characteristic parameters, using simulation calculation data and combining artificial intelligence methods, a fault pattern recognition model for the reactor shield pump rotating mechanical equipment was established;

[0027] The fault pattern recognition model is used to perform online fault identification of the rotating mechanical equipment of the reactor shield pump.

[0028] Furthermore, the third nonlinear transient dynamic response data includes multiple groups of curves; each group of curves is a curve of the rotor displacement at any position of the pump casing changing with time, a curve of the rotor speed changing with time, a curve of the rotor acceleration changing with time, and a curve of the bearing load changing with time.

[0029] In a second aspect, the present invention further provides a closed-structure reactor shielded pump fault identification device, the device comprising:

[0030] The guide bearing model establishment unit is used to establish the guide bearing model of the reactor canned pump based on fluid dynamics theory; based on the guide bearing model, the bearing load capacity at a certain position and speed is obtained;

[0031] A rotor-bearing coupling model building unit is used to build a rotor dynamics model of a reactor canned motor pump and dynamically couple the rotor dynamics model with the guide bearing model to obtain a rotor-bearing coupling dynamics model of the reactor canned motor pump;

[0032] A first data acquisition unit is used to obtain first nonlinear transient dynamic response data of the system based on the rotor-bearing coupling dynamic model;

[0033] a second data acquisition unit, configured to load the fault excitation into the rotor-bearing coupled dynamics model to obtain second nonlinear transient dynamic response data;

[0034] The third data acquisition unit is used to establish a transient dynamic model of the pump casing of the reactor canned pump; load the second nonlinear transient dynamic response data into the pump casing transient dynamic model, and calculate and obtain third nonlinear transient dynamic response data of the pump casing of the reactor canned pump under internal fault excitation;

[0035] The pump fault identification unit is used to identify fault modes based on the third nonlinear transient dynamic response data using an artificial intelligence algorithm.

[0036] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above-mentioned closed-structure reactor shielded pump fault identification method is implemented.

[0037] In a fourth aspect, the present invention further provides a computer-readable storage medium storing a computer program, which implements the above-mentioned closed-structure reactor shielded pump fault identification method when executed by a processor.

[0038] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0039] The present invention provides a method, device, and medium for identifying faults in a closed-structure reactor canned pump. Based on mechanism analysis and physical modeling, the present invention establishes a system dynamics coupling model for rotating machinery such as canned pumps, obtains a dynamic response calculation method, and obtains fault dynamic response data (third nonlinear transient dynamic response data) and fault response characteristics. Furthermore, artificial intelligence technology is used to identify equipment fault patterns, laying the foundation for the digital operation and maintenance of canned pump-type rotating machinery. Compared to existing technologies, the present invention obtains mechanism simulation data through a system dynamics coupling model, which can obtain effective fault data and improve the accuracy and efficiency of fault identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:

[0041] Figure 1 This is a flow chart of a method for identifying a fault of a canned pump in a closed structure reactor according to the present invention;

[0042] Figure 2 This is a schematic diagram of the flow chart for dynamically coupling the rotor dynamics model with the guide bearing model in the present invention;

[0043] Figure 3 This is a schematic diagram of the model architecture of the present invention;

[0044] Figure 4 This is a structural block diagram of the closed-structure reactor shielded pump fault identification device of the present invention. DETAILED DESCRIPTION

[0045] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with examples and drawings. The exemplary embodiments of the present invention and their descriptions are only used to explain the present invention and are not intended to limit the present invention.

[0046] The existing fault identification for rotating machinery and equipment such as reactor system shielded pumps is mainly based on test data, which is costly and difficult to obtain effective fault data, leading to problems of low fault identification accuracy and efficiency. The present invention designs a closed-structure reactor shielded pump fault identification method, device and medium based on mechanism analysis and physical modeling. By establishing a multi-module (rotor-bearing-pump casing) coupled dynamic model of the reactor shielded pump and simulating the response excitation of internal faults on the pump casing, more effective fault dynamic response data of the reactor shielded pump are obtained; based on this analysis of the pump fault response characteristics, fault pattern identification is further performed based on artificial intelligence algorithms (support vector machines, artificial neural networks and other machines, deep learning methods, etc.), fault diagnosis and health management are achieved, and the foundation is laid for the digital operation and maintenance of shielded pump type rotating machinery. The present invention can obtain effective fault data and improve the accuracy and efficiency of fault identification.

[0047] The specific design ideas are as follows:

[0048] (1) Guide bearing model of reactor shielded pump

[0049] The radial sliding bearings (i.e., guide bearings) used in reactor shield pumps are water-guided bearings with an unconventional structure and lubrication and cooling method. Their load-bearing capacity is relatively limited compared to oil-lubricated hydrodynamic bearings. Therefore, instability can occur in the event of a fault, causing nonlinear motion of the rotor axis. Conventional bearing modeling methods, based on the assumption that the rotor only undergoes small displacements, simplify the bearing into a spring-damped support system. This results in significant errors when the rotor axis moves significantly, and the nonlinear characteristics of the bearing are ignored, making it impossible to accurately capture the dynamic characteristics of reactor shield pump failures. Therefore, it is necessary to establish a separate guide bearing model for the reactor shield pump (i.e., a guide bearing hydrodynamic calculation model) based on fluid dynamics theory to determine the bearing hydrodynamic water film pressure distribution and bearing load capacity at a specific position and speed.

[0050] (2) Rotor-bearing coupling dynamic model of reactor shielded pump

[0051] When modeling the rotor-bearing system of a reactor canned motor pump under the influence of a fault, the bearing cannot be simplified into a general spring-damper system. Therefore, when performing the dynamic analysis of the rotor-bearing system, it is necessary to perform dynamic coupling between the rotor dynamics model and the bearing model to obtain a rotor-bearing coupled dynamics model.

[0052] (3) Fault response data simulation method for reactor shielded pump

[0053] Since the shielded pump is a closed structure, the location and status of the internal fault of the pump cannot be directly detected. Fault diagnosis can only be performed through the vibration response of the pump casing foundation. Therefore, in order to effectively simulate the vibration response of the pump casing foundation under the excitation of the internal fault of the pump casing, the fault excitation is first loaded into the rotor-bearing coupling dynamic model to simulate the nonlinear dynamic response of the rotor and the nonlinear change of the bearing capacity under the fault excitation. Then, a transient dynamic model of the pump casing foundation (i.e., the pump casing transient dynamic model) is established. The nonlinear response data of the bearing capacity after the fault excitation is loaded into the pump casing transient dynamic model. The nonlinear dynamic response of the pump casing foundation of the reactor shielded pump under the internal fault excitation is calculated, the fault response characteristics are analyzed, and the fault pattern recognition is supported by the artificial intelligence algorithm combined with the fault test data.

[0054] Example 1

[0055] like Figure 1 As shown, the closed structure reactor shielded pump fault identification method of the present invention comprises:

[0056] Step 1: Based on fluid dynamics theory, a guide bearing model of a reactor canned pump is established; based on the guide bearing model, the bearing load capacity at a certain position and speed is obtained;

[0057] In this embodiment, to simulate the dynamic response of a canned motor pump rotating machine under fault conditions, it is necessary to perform dynamic modeling of the canned motor pump rotor based on rotor dynamics. When the rotor rotates at high speed, a hydrodynamic water film is generated between the guide bearing and the rotor, providing radial support for the rotor, bearing all radial loads, and constraining radial displacement. Therefore, a coupled model of the rotor-bearing system is required. However, the lubrication method of the guide bearing differs from that of a typical hydrodynamic sliding bearing and cannot be simplified to a general spring-damper system. Therefore, a mechanistic model of the guide bearing (and a bearing model) is required to understand the bearing's load-bearing effect on the rotor.

[0058] First, we simplify the water-lubricated guide bearing into a thin film and model it. According to the fluid continuity theorem, we can get:

[0059]

[0060] Where ρ is the fluid density, v x 、v y 、v z are the velocities of the fluid element along the x, y, and z directions respectively. According to the law of conservation of fluid momentum, we can obtain:

[0061]

[0062] Where, f x 、f y 、fz are the mass forces of the fluid element along the x, y, and z directions, σ xx , σ yy , σ zz are the normal stresses of the fluid element in the x, y, and z directions, σ xx , σ yy , σ zz are the normal stresses of the fluid element in the x, y, and z directions, τ xy , τ yx , τ xz , τ zx , τ yz , τ zy are the shear stresses of the fluid element on the surface. For Newtonian fluids, the above equation can be further transformed to obtain the NS equation:

[0063]

[0064] Where p is pressure, μ is fluid dynamic viscosity, is the velocity vector, is a vector differential operator. For thin film lubrication fluid, simplifying the above equation and combining it with the fluid continuity equation yields the Reynolds equation for thin film lubrication:

[0065]

[0066] Where h is the film thickness, U1 and U2 are the velocities of the two solid surfaces along the x-direction, and W1 and W2 are the velocities of the two solid surfaces along the y-direction. For thin film fluid lubrication, the unsteady Reynolds equation for thin film lubrication of sliding bearings can be obtained by changing the above equation according to the bearing fluid lubrication method:

[0067]

[0068] Where R is the rotor radius, θ is the circumferential angle, and Ω is the rotor speed.

[0069] The water film pressure distribution can be obtained by numerically iteratively solving the above equation.

[0070] The present invention studies guide bearings with unconventional structures and lubrication methods. These bearings are non-enclosed, lubricated with pressurized water as the lubricant. The fluid has low viscosity but exhibits static pressure, and the lubricant enters and exits through the shaft end. This significantly impacts the water film's load-bearing characteristics and pressure distribution, necessitating rational modeling of its boundary conditions. Numerical solutions (such as the finite difference method, finite element method, and finite volume method) are then used to iteratively solve the aforementioned partial differential equations, yielding the bearing water film pressure distribution. This is then integrated to determine the water film's load-bearing capacity (i.e., the bearing's load-bearing capacity).

[0071] Step 2: Establish a rotor dynamics model of the reactor canned motor pump, and dynamically couple the rotor dynamics model with the guide bearing model to obtain a rotor-bearing coupled dynamics model of the reactor canned motor pump;

[0072] In this embodiment, Figure 2 As shown in the figure, the rotor dynamics model is dynamically coupled with the guide bearing model to obtain the rotor-bearing coupled dynamics model of the reactor canned pump, including:

[0073] The rotor axis position and speed at the current moment are calculated through the rotor dynamics model, and the rotor axis position and speed at the current moment are input into the guide bearing model to calculate the bearing capacity at the next time step;

[0074] The bearing capacity at the next time step is used as the input of the rotor dynamics model at the next time step. The rotor axis position and speed at the next time step are calculated based on the rotor dynamics model.

[0075] According to the above steps, it is iterated cyclically with time steps to realize the rotor-bearing coupled dynamics analysis and obtain the rotor-bearing coupled dynamics model of the reactor canned pump.

[0076] In this embodiment, the shielded pump is mainly composed of a pump casing foundation, an impeller, an upper flywheel, an upper radial sliding bearing, a pump casing foundation shell, a pump casing foundation core, a rotor core, a lower radial sliding bearing, a bidirectional thrust bearing, and a lower flywheel. The pump rotor structure is divided into different shaft segments, wherein the flywheel and impeller are modeled as disk units with equal mass and moment of inertia, and the rotor core is modeled as a solid unit with equal diameter. A rotor transient dynamics calculation model (i.e., a rotor dynamics model) is established based on existing large-scale finite element analysis and calculation software such as ANSYS. A reactor shielded pump radial sliding bearing model (i.e., a bearing model) is written in languages ​​such as MATLAB, Python, or C++ and packaged as an exe file. The rotor dynamics model and the bearing model are coupled for analysis and calculation using secondary development technology to obtain the nonlinear dynamic response of the rotor-bearing system (i.e., a rotor-bearing coupled dynamics model). In the subsequent step 4, a basic dynamic calculation model of the reactor shielded pump casing (i.e., the pump casing transient dynamic model) is established using large-scale finite element commercial software such as ABAQUS, COMSOL, and ANSYS. The rotor-bearing system is coupled with the pump casing transient dynamic model using secondary development technology to obtain a nonlinear dynamic response calculation model of the closed-structure reactor shielded pump rotor-bearing-pump casing basic system (i.e., the rotor-bearing-pump casing system nonlinear dynamic coupling model). Figure 3 As shown, Figure 3 A schematic diagram of the model architecture.

[0077] Step 3: obtaining first nonlinear transient dynamic response data of the system based on the rotor-bearing coupled dynamic model; the system refers to the rotor-bearing coupled dynamic system;

[0078] In this embodiment, the first nonlinear transient dynamic response data includes time domain response data such as rotor acceleration, rotor speed, rotor displacement, and time domain variation of bearing load.

[0079] Step 4: Load the fault excitation into the rotor-bearing coupled dynamic model to obtain second nonlinear transient dynamic response data; and establish a pump casing transient dynamic model of the reactor canned pump;

[0080] In this embodiment, the fault excitation is loaded into the rotor-bearing coupled dynamic model to obtain the second nonlinear transient dynamic response data, including:

[0081] The fault excitation is loaded into the rotor-bearing coupled dynamic model to simulate the nonlinear dynamic response of the rotor and the nonlinear change of the bearing load under the fault excitation as the second nonlinear transient dynamic response data.

[0082] Step 5: Load the second nonlinear transient dynamic response data into the pump casing transient dynamic model to calculate and obtain third nonlinear transient dynamic response data of the pump casing of the reactor canned pump under internal fault excitation;

[0083] In this embodiment, the third nonlinear transient dynamic response data includes multiple groups of curves; each group of curves is a curve of the rotor displacement at any position of the pump casing changing with time, a curve of the rotor speed changing with time, a curve of the rotor acceleration changing with time, and a curve of the bearing load changing with time.

[0084] Step 6: Analyze the fault response characteristics based on the third nonlinear transient dynamic response data; and use an artificial intelligence algorithm to perform fault mode recognition.

[0085] As a further implementation, this method uses MATLAB software to establish a guide bearing model, and uses ABAQUS, COMSOL or ANSYS software to establish a rotor dynamics model and a pump casing transient dynamics model; then uses PYTHON or C++ language to perform secondary development on the above software to achieve dynamic coupling between the guide bearing model, rotor dynamics model and pump casing transient dynamics model under multi-physical fields, and obtain a nonlinear dynamic coupling model of the rotor-bearing-pump casing system of a closed-structure reactor shield pump.

[0086] As a further implementation, an artificial intelligence algorithm is used to perform fault mode identification based on the third nonlinear transient dynamic response data, including:

[0087] Different pump faults are injected into the nonlinear dynamic coupling model of the rotor-bearing-pump casing system to directly simulate the dynamic response of the fault excitation on the pump casing and obtain key characteristic parameters. Key characteristic parameters are characteristic parameters that can characterize different faults of reactor canned pump rotors, such as eccentricity, misalignment, and bearing wear.

[0088] Based on key characteristic parameters, using simulation calculation data and combining artificial intelligence methods, a fault pattern recognition model for the reactor shield pump rotating mechanical equipment was established;

[0089] The fault pattern recognition model is used to perform online fault identification of the rotating mechanical equipment of the reactor shield pump.

[0090] In the above technical solution, the nonlinear dynamic response calculation of the reactor shielded pump rotor-bearing system is established to obtain the pump dynamic response and its changes under normal conditions, such as the time-varying relationship between displacement, velocity, acceleration, bearing capacity, and other dynamic responses. Furthermore, combined with the mechanism characterization model of different faults such as pump rotor mass eccentricity, misalignment, and bearing wear, the dynamic response and its changes of the pump rotating mechanical equipment under the action of faults are obtained. The time and frequency domain characteristics of the equipment under normal and fault conditions are further analyzed to identify the differences in the equipment's responses under normal and fault conditions, and to obtain key characteristic parameters that can characterize different faults such as reactor shielded pump rotor eccentricity, misalignment, and bearing wear. Finally, a fault pattern recognition model for shielded pump rotating mechanical equipment is established through machine and deep learning methods such as support vector machines and artificial neural networks, providing a theoretical basis for online fault analysis and diagnosis of shielded pump rotating machinery and supporting the digitalization of equipment operation and maintenance.

[0091] Based on mechanism analysis and physical modeling, this invention establishes a system dynamics coupling model for rotating machinery such as canned motor pumps, develops a dynamic response calculation method, and obtains fault dynamic response data (third nonlinear transient dynamic response data) and fault response characteristics. Furthermore, artificial intelligence technology is used to identify equipment fault patterns, laying the foundation for digital operation and maintenance of canned motor pumps. By obtaining mechanism simulation data through the system dynamics coupling model, this invention can obtain effective fault data, improving the accuracy and efficiency of fault identification.

[0092] Example 2

[0093] like Figure 4 As shown, the difference between this embodiment and embodiment 1 is that this embodiment provides a closed structure reactor canned pump fault identification device, the device comprising:

[0094] The guide bearing model establishment unit is used to establish the guide bearing model of the reactor canned pump based on fluid dynamics theory; based on the guide bearing model, the bearing load capacity at a certain position and speed is obtained;

[0095] A rotor-bearing coupling model building unit is used to build a rotor dynamics model of a reactor canned motor pump and dynamically couple the rotor dynamics model with the guide bearing model to obtain a rotor-bearing coupling dynamics model of the reactor canned motor pump;

[0096] A first data acquisition unit is used to obtain first nonlinear transient dynamic response data of the system based on the rotor-bearing coupling dynamic model;

[0097] a second data acquisition unit, configured to load the fault excitation into the rotor-bearing coupled dynamics model to obtain second nonlinear transient dynamic response data;

[0098] The third data acquisition unit is used to establish a transient dynamic model of the pump casing of the reactor canned pump; load the second nonlinear transient dynamic response data into the pump casing transient dynamic model, and calculate and obtain third nonlinear transient dynamic response data of the pump casing of the reactor canned pump under internal fault excitation;

[0099] The pump fault identification unit is used to identify fault modes based on the third nonlinear transient dynamic response data using an artificial intelligence algorithm.

[0100] The execution process of each unit can be performed according to the process steps of the closed structure reactor canned pump fault identification method in Example 1, and will not be described in detail in this embodiment.

[0101] At the same time, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, the above-mentioned closed structure reactor shielded pump fault identification method is implemented.

[0102] At the same time, the present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned closed structure reactor shielded pump fault identification method is implemented.

[0103] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0104] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0105] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0106] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0107] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for identifying a fault of a shielded pump in a closed reactor structure, characterized in that: The method includes: Based on fluid dynamics theory, a guide bearing model of a reactor canned pump is established; based on the guide bearing model, the bearing load capacity at a certain position and speed is obtained; Establishing a rotor dynamics model of a reactor canned motor pump, and dynamically coupling the rotor dynamics model with the guide bearing model to obtain a rotor-bearing coupled dynamics model of the reactor canned motor pump; Based on the rotor-bearing coupling dynamics model, obtaining first nonlinear transient dynamic response data of the system; Loading the fault excitation into the rotor-bearing coupled dynamic model to obtain second nonlinear transient dynamic response data; and establishing a pump casing transient dynamic model of the reactor canned pump; Loading the second nonlinear transient dynamic response data into the pump casing transient dynamic model to calculate and obtain third nonlinear transient dynamic response data of the pump casing of the reactor canned pump under internal fault excitation; Based on the third nonlinear transient dynamic response data, an artificial intelligence algorithm is used to perform fault mode identification.

2. The closed structure reactor canned pump fault identification method according to claim 1, characterized in that: The lubricating material of the guide bearing in the guide bearing model is water, which has low fluid viscosity but static pressure; the inlet and outlet of the guide bearing is shaft end inlet and outlet, with one shaft end inlet and the other shaft end outlet.

3. The closed structure reactor canned pump fault identification method according to claim 1, characterized in that: The rotor dynamics model is dynamically coupled with the guide bearing model to obtain a rotor-bearing coupled dynamics model of a reactor canned pump, including: Calculating the rotor axis position and speed at the current moment through the rotor dynamics model, and inputting the rotor axis position and speed at the current moment into the guide bearing model to calculate the bearing capacity at the next time step; Using the bearing capacity of the next time step as the input of the rotor dynamics model for the next time step, and calculating based on the rotor dynamics model to obtain the rotor axis position and speed for the next time step; According to the above steps, it is iterated cyclically with time steps to realize the rotor-bearing coupled dynamics analysis and obtain the rotor-bearing coupled dynamics model of the reactor canned pump.

4. The closed structure reactor canned pump fault identification method according to claim 1, characterized in that: The first nonlinear transient dynamic response data includes rotor displacement, rotor speed, rotor acceleration and bearing load.

5. The closed structure reactor canned pump fault identification method according to claim 1, characterized in that: Loading the fault excitation into the rotor-bearing coupled dynamic model to obtain second nonlinear transient dynamic response data includes: The fault excitation is loaded into the rotor-bearing coupled dynamic model to simulate the rotor dynamic nonlinear response and the nonlinear change of the bearing load under the fault excitation as the second nonlinear transient dynamic response data.

6. The closed structure reactor canned pump fault identification method according to claim 1, characterized in that: Abstract: In this method, the guide bearing model is established by MATLAB software, and the rotor dynamics model and pump casing transient dynamics model are established by ABAQUS, COMSOL or ANSYS software. The above software is then secondary developed using PYTHON or C++ language to realize the dynamic coupling among the guide bearing model, rotor dynamics model and pump casing transient dynamics model under multi-physics field, and obtain the nonlinear dynamic coupling model of the rotor-bearing-pump casing system of the closed structure reactor shield pump.

7. The closed structure reactor canned pump fault identification method according to claim 6, characterized in that: Based on the third nonlinear transient dynamic response data, an artificial intelligence algorithm is used to perform fault mode identification, including: Different pump faults are injected into the nonlinear dynamic coupling model of the rotor-bearing-pump casing system to simulate the dynamic response of the fault excitation on the pump casing to obtain key characteristic parameters; the key characteristic parameters are characteristic parameters that can characterize different faults of the reactor canned pump rotor eccentricity, misalignment, and bearing wear; Based on the key characteristic parameters, a fault mode recognition model for the rotating mechanical equipment of the reactor shield pump is established by using simulation calculation data in combination with artificial intelligence methods; The fault pattern recognition model is used to perform online fault identification of the rotating mechanical equipment of the reactor shield pump.

8. The closed structure reactor canned pump fault identification method according to claim 4, characterized in that: The third nonlinear transient dynamic response data includes multiple groups of curves; each group of curves is a curve showing the rotor displacement at any position of the pump casing changing with time, a curve showing the rotor speed changing with time, a curve showing the rotor acceleration changing with time, and a curve showing the bearing capacity changing with time.

9. A closed-structure reactor shielded pump fault identification device, characterized in that: The device includes: A guide bearing model establishment unit is used to establish a guide bearing model of a reactor canned pump based on fluid dynamics theory; based on the guide bearing model, obtain the bearing capacity at a certain position and speed; a rotor-bearing coupling model establishing unit, configured to establish a rotor dynamics model of the reactor canned motor pump, and dynamically couple the rotor dynamics model with the guide bearing model to obtain a rotor-bearing coupling dynamics model of the reactor canned motor pump; A first data acquisition unit is configured to acquire first nonlinear transient dynamic response data of the system based on the rotor-bearing coupling dynamic model; a second data acquisition unit, configured to load a fault excitation into the rotor-bearing coupled dynamics model to obtain second nonlinear transient dynamic response data; a third data acquisition unit, configured to establish a transient dynamic model of a pump casing of a reactor canned pump; load the second nonlinear transient dynamic response data into the pump casing transient dynamic model, and calculate and obtain third nonlinear transient dynamic response data of the pump casing of the reactor canned pump under internal fault excitation; The pump fault identification unit is used to perform fault mode identification based on the third nonlinear transient dynamic response data using an artificial intelligence algorithm.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the closed structure reactor canned pump fault identification method according to any one of claims 1 to 8 is implemented.

11. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for identifying a fault of a canned pump in a closed structure reactor according to any one of claims 1 to 8 is implemented.

Images