Method, device and apparatus for monitoring structure of frame of pumped-storage power unit

Abstract

To provide a frame structure monitoring method, device, apparatus, and medium for comprehensively grasping the state of stress and deformation of a pumped-storage power generation unit frame.SOLUTION: The method includes constructing a frame structure FE model based on structural parameters and operational parameters, performing simulation analysis on a simulation parameter set based on the FE model to obtain first stress and deformation simulation values, obtaining an optimized proxy model based on the parameter set and the first stress and deformation simulation values, inputting a test parameter set into the FE model and the optimized proxy model to obtain second stress and deformation simulation values and stress and deformation prediction values, and monitoring stress and deformation of the frame using the stress and deformation prediction values when the second stress and deformation simulation values and the stress and deformation prediction values meet a preset accuracy condition.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to the technical field of integrating machine learning with energy equipment, and in particular to a method, apparatus and equipment for monitoring the structure of a pumped storage power generation unit frame. [Background technology]

[0002] Pumped storage power generation units play a role in regulating peak loads in the power grid, filling valleys, and providing backup for faults. The frame of a pumped storage power generation unit is one of the unit's important force-bearing components. During operation, it is not only subjected to the gravity of rotating components such as the pump, turbine, and generating motor, but is also affected by unbalanced forces from the rotor mass, unbalanced magnetic attraction forces from the motor, nonlinear oil film forces in the bearings, nonlinear sealing forces from the fluid, and unstable exciting forces from the fluid. The safety and reliability of the structure play a crucial role in the safe, reliable, and stable operation of the pumped storage power generation unit. Therefore, it is necessary to monitor the frame's stress and deformation in real time.

[0003] However, existing methods that use only static analysis during the design phase and monitor local stress and deformation using condition monitoring technology during the operation and maintenance phase have problems in that they are unable to comprehensively grasp the stress and deformation conditions at each position of the frame and are unable to intuitively grasp the stress and deformation conditions of the frame in real time. Summary of the Invention [Problem to be solved by the invention]

[0004] In view of this, the present invention provides a method, device and equipment for monitoring the structure of a pumped storage power generation unit frame, so as to solve the problem of being unable to comprehensively grasp the stress and deformation conditions of the frame. [Means for solving the problem]

[0005] In a first aspect, the present invention provides a method for producing a composition comprising: obtaining structural parameters of a pumped storage power generation unit frame and operating parameters for a plurality of operating conditions; constructing a frame structure finite element model according to the structural parameters and the operating parameters of a plurality of operating conditions, extracting characteristic parameters of a plurality of operating conditions according to the operating parameters of the plurality of operating conditions, and dividing the characteristic parameters of the plurality of operating conditions into a test parameter set of a plurality of operating conditions and a simulation parameter set of a plurality of operating conditions; Under a preset operating condition, performing a simulation analysis on a set of simulation parameters for the corresponding operating condition based on the frame structure finite element model to obtain a first stress and deformation simulation value for the corresponding operating condition of the pumped storage power generation unit frame; optimizing the preset initial proxy model according to the simulation parameter set of the corresponding operating condition and the first stress and deformation simulation value of the corresponding operating condition to obtain an optimized proxy model; In a preset operating condition, inputting a test parameter set of the corresponding operating condition into the frame structure finite element model and the optimized proxy model, respectively, to obtain a second stress and deformation simulation value of the corresponding operating condition and a stress and deformation prediction value of the corresponding operating condition, respectively; and monitoring the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation values ​​for the corresponding operating conditions when the second stress and deformation simulation values ​​for the corresponding operating conditions and the predicted stress and deformation values ​​for the corresponding operating conditions satisfy a predetermined accuracy condition.

[0006] The method for monitoring the structure of a pumped storage power generation unit frame according to the present invention includes: performing a simulation analysis on a simulation parameter set for a corresponding operating condition using a frame structure finite element model; obtaining first simulated stress and deformation values ​​for the corresponding operating condition of the pumped storage power generation unit frame; obtaining an optimized proxy model based on the first simulated stress and deformation values ​​and the extracted simulation parameter set for the corresponding operating condition; inputting a test parameter set for the corresponding operating condition into the frame structure finite element model and the optimized proxy model, respectively, to obtain second simulated stress and deformation values ​​for the corresponding operating condition and predicted stress and deformation values ​​for the corresponding operating condition; and if the second simulated stress and deformation values ​​for the corresponding operating condition and the predicted stress and deformation values ​​for the corresponding operating condition meet the predetermined accuracy requirements, using the predicted stress and deformation values ​​for the corresponding operating condition to monitor the stress and deformation of the pumped storage power generation unit frame, thereby achieving the objective of comprehensively and real-timely monitoring the stress and deformation conditions at each position of the pumped storage power generation unit frame under various operating conditions and solving the problem of not being able to comprehensively grasp the stress and deformation conditions of the frame.

[0007] In one alternative embodiment, the structural parameters of the pumped storage power generation unit frame include geometric dimensions, material properties, and structural connection relationships of the pumped storage power generation unit frame, and the operational parameters of the pumped storage power generation unit frame under a plurality of operating conditions include working loads under a plurality of operating conditions; The step of constructing a frame structure finite element model based on the structural parameters and the operating parameters of a plurality of operating conditions includes: The method includes using a parametric programming language to construct a frame structure finite element model under a plurality of operating conditions based on the geometric dimensions, material attributes, structural connection relationships, and operating loads under a plurality of operating conditions of the pumped storage power generation unit frame.

[0008] The structural monitoring method for a pumped storage power generation unit frame of the present invention uses a parametric programming language to construct a frame structure finite element model under multiple operating conditions based on the geometric dimensions, material attributes, structural connection relationships, and operating loads under multiple operating conditions of the pumped storage power generation unit frame, thereby achieving the goal of constructing a refined frame structure finite element model, and providing a model basis for subsequently obtaining first stress and strain simulation values.

[0009] In one alternative embodiment, the step of performing a simulation analysis on a simulation parameter set of a corresponding operating condition based on the frame structure finite element model under a preset operating condition to obtain a first simulated stress and deformation value of the corresponding operating condition of the pumped storage power generation unit frame includes: At a preset operating condition, sampling a simulation parameter set of the corresponding operating condition using a Latin hypercube sampling algorithm to obtain input data of the corresponding operating condition; and inputting the input data of the corresponding operating conditions into the frame structure finite element model to perform static simulation analysis, and obtaining first stress and deformation simulation values ​​of the corresponding operating conditions of the pumped storage power generation unit frame.

[0010] The structure monitoring method for a pumped storage power generation unit frame according to the present invention involves sampling a set of simulation parameters for corresponding operating conditions using a Latin hypercube sampling algorithm under preset operating conditions to obtain input data for the corresponding operating conditions, inputting the input data for the corresponding operating conditions into a frame structure finite element model to perform static simulation analysis, and obtaining first simulated stress and deformation values ​​for the corresponding operating conditions of the pumped storage power generation unit frame. The frame structure finite element model is used to realize static simulation analysis under various operating conditions, and further obtaining the first simulated stress and deformation values ​​under various operating conditions. The first simulated stress and deformation values ​​under various operating conditions can be used to determine weak points of the pumped storage power generation unit frame, and provide a data basis for subsequent optimization of the initial proxy model.

[0011] In one alternative embodiment, the step of optimizing the preset initial proxy model based on the simulation parameter set of the corresponding operating condition and the first stress and deformation simulation value of the corresponding operating condition to obtain the optimized proxy model includes: In a preset operating condition, sampling characteristic parameters of the simulation parameter set of the corresponding operating condition using a Latin hypercube sampling algorithm to obtain input data of the corresponding operating condition; inputting the input data of the corresponding operating conditions and the first stress and deformation simulation values ​​of the corresponding operating conditions as sample data points into a preset initial proxy model to obtain related parameters of the initial proxy model; calculating an optimal solution for the relevant parameters using the Pelican optimization algorithm; optimizing the initial proxy model using the optimal solution to obtain an optimized proxy model.

[0012] The structural monitoring method for a pumped storage power generation unit frame according to the present invention includes, under a preset operating condition, using a Latin hypercube sampling algorithm to sample characteristic parameters of a simulation parameter set for the corresponding operating condition to obtain input data for the corresponding operating condition, inputting the input data for the corresponding operating condition and the first stress and deformation simulation values ​​for the corresponding operating condition as sample data points into a preset initial proxy model to obtain relevant parameters of the initial proxy model, calculating an optimal solution for the relevant parameters using a Pelican optimization algorithm, optimizing the initial proxy model using the optimal solution, and obtaining an optimized proxy model, which provides a model basis for subsequent prediction of the stress and deformation conditions of the pumped storage power generation unit frame, thereby improving the prediction accuracy of the stress and deformation conditions.

[0013] In one alternative embodiment, when the second stress and deformation simulation value for the corresponding operating condition and the predicted stress and deformation value for the corresponding operating condition satisfy a predetermined accuracy condition, the step of monitoring the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation value for the corresponding operating condition includes: calculating a first relative error between a second simulated stress and deformation value for the corresponding operating condition and a predicted stress and deformation value for the corresponding operating condition; determining whether the first relative error satisfies a preset accuracy condition; and if the first relative error satisfies the preset accuracy condition, monitoring the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation values ​​of the corresponding operating condition.

[0014] The structural monitoring method for a pumped storage power generation unit frame of the present invention realizes accuracy verification of the first relative error by determining whether the first relative error between the second stress and deformation simulation value for the corresponding operating condition and the stress and deformation predicted value for the corresponding operating condition meets the predetermined accuracy condition, thereby improving the accuracy of the stress and deformation predicted value for the corresponding operating condition, and if the first relative error meets the predetermined accuracy condition, monitors the stress and deformation of the pumped storage power generation unit frame using the stress and deformation predicted value for the corresponding operating condition, thereby achieving the purpose of monitoring the stress and deformation of the pumped storage power generation unit frame in real time under various operating conditions.

[0015] In one alternative embodiment, when the first relative error satisfies the preset accuracy condition, the step of monitoring the stress and deformation of the pumped storage power generation unit frame using the stress and deformation predicted value of the corresponding operating condition includes: If the first relative error satisfies the predetermined accuracy condition, collecting actual measured values ​​of stress and deformation of the pumped storage power generation unit frame under the corresponding operating condition; Calculating a second relative error between the actual stress and deformation values ​​for the corresponding operating conditions and the predicted stress and deformation values ​​for the corresponding operating conditions; and if the second relative error satisfies the preset accuracy condition, monitoring the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation values ​​of the corresponding operating condition.

[0016] The structural monitoring method for a pumped storage power generation unit frame of the present invention continues to verify the accuracy of a second relative error between the actual measured stress and deformation values ​​for the corresponding operating conditions and the predicted stress and deformation values ​​for the corresponding operating conditions when the first relative error satisfies the predetermined accuracy condition, and monitors the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation values ​​for the corresponding operating conditions when the second relative error satisfies the predetermined accuracy condition, thereby improving the accuracy of verifying the predicted stress and deformation values ​​for the corresponding operating conditions and making the monitoring of the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation values ​​for the corresponding operating conditions more accurate.

[0017] In one alternative embodiment, a method for monitoring the structure of a pumped storage power generation unit frame includes: if the first relative error does not satisfy the preset accuracy condition, adding the second stress and deformation simulation values ​​as additional data to the sample data points to obtain additional sample data points, and updating the optimized proxy model based on the additional sample data points; Or, If the second relative error does not satisfy the predetermined accuracy condition, the method further includes updating the frame structure finite element model using the predetermined correction coefficient, adding the stress and deformation actual measurement values ​​as additional data to the sample data points to obtain additional sample data points, and updating the optimized proxy model based on the additional sample data points.

[0018] The method for monitoring the structure of a pumped storage power generation unit frame according to the present invention includes the steps of: if the first relative error does not satisfy the predetermined accuracy condition, adding simulated stress and deformation values ​​to the sample data points as additional data to obtain additional sample data points, and updating the optimized proxy model based on the additional sample data points; or if the second relative error does not satisfy the predetermined accuracy condition, updating the frame structure finite element model using a predetermined correction coefficient, adding actual measured stress and deformation values ​​to the sample data points as additional data to obtain additional sample data points, and updating the optimized proxy model based on the additional sample data points, and updating the frame structure finite element model using the predetermined correction coefficient, thereby further optimizing the frame structure finite element model and improving the simulation accuracy of the finite element model; further optimizing the optimized proxy model using the additional sample data points, and further improving the prediction accuracy of the optimized proxy model.

[0019] In a second aspect, the present invention provides a method for producing a pharmaceutical composition comprising: an acquisition module for acquiring structural parameters of the pumped storage power generation unit frame and operating parameters of a plurality of operating conditions; a construction module for constructing a frame structure finite element model according to the structural parameters and operating parameters of a plurality of operating conditions, extracting characteristic parameters of a plurality of operating conditions according to the operating parameters of the plurality of operating conditions, and dividing the characteristic parameters of the plurality of operating conditions into a test parameter set of a plurality of operating conditions and a simulation parameter set of a plurality of operating conditions; a simulation module for performing a simulation analysis on a simulation parameter set of a corresponding operating condition based on the frame structure finite element model under a predetermined operating condition, to obtain a first stress and deformation simulation value of the corresponding operating condition of the pumped storage power generation unit frame; an optimization module for optimizing the preset initial proxy model according to the characteristic parameters of the simulation parameter set for the corresponding operating condition and the first stress and deformation simulation values ​​for the corresponding operating condition to obtain an optimized proxy model; a test module for inputting a test parameter set of a corresponding operating condition into the frame structure finite element model and the optimized proxy model, respectively, under a preset operating condition, to obtain a second stress and deformation simulation value of the corresponding operating condition and a stress and deformation prediction value of the corresponding operating condition, respectively; and a monitoring module for monitoring the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation values ​​for the corresponding operating conditions when the second stress and deformation simulation values ​​for the corresponding operating conditions and the predicted stress and deformation values ​​for the corresponding operating conditions satisfy a predetermined accuracy condition.

[0020] In a third aspect, the present invention provides a computer apparatus, comprising: a memory; and a processor, the memory and the processor being communicatively connected; computer instructions stored in the memory; and the processor executing the computer instructions to perform the method for structural monitoring of a pumped storage power generation unit frame according to the first aspect above or any of its corresponding embodiments.

[0021] In a fourth aspect, the present invention provides a computer-readable storage medium having stored thereon computer instructions for causing a computer to execute the method for structural monitoring of a pumped storage power generation unit frame according to the first aspect or any of its corresponding embodiments.

[0022] In order to more clearly describe the specific embodiments of the present invention or the technical solutions of the prior art, the following will briefly describe the drawings that need to be used to describe the specific embodiments or the prior art. It is obvious that the drawings described below are some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative work. [Brief explanation of the drawings]

[0023] [Figure 1] 1 is a schematic flowchart of a method for monitoring the structure of a pumped storage power generation unit frame according to an embodiment of the present invention. [Figure 2] 10 is a schematic flowchart of another method for monitoring the structure of a pumped storage power generation unit frame according to an embodiment of the present invention. [Figure 3] 10 is a schematic flowchart of yet another method for monitoring the structure of a pumped storage power generation unit frame according to an embodiment of the present invention. [Figure 4] 10 is a schematic flowchart of yet another method for monitoring the structure of a pumped storage power generation unit frame according to an embodiment of the present invention. [Figure 5] 10 is a schematic flowchart of yet another method for monitoring the structure of a pumped storage power generation unit frame according to an embodiment of the present invention. [Figure 6] 10 is a schematic flowchart of yet another method for monitoring the structure of a pumped storage power generation unit frame according to an embodiment of the present invention. [Figure 7] 1 is a structural block diagram of a structural monitoring device for a pumped storage power generation unit frame according to an embodiment of the present invention; [Figure 8] FIG. 2 is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention; DETAILED DESCRIPTION OF THE INVENTION

[0024] In order to clarify the objectives, technical solutions and advantages of the embodiments of the present invention, the technical solutions of the embodiments of the present invention will be described clearly and completely below with reference to the drawings of the embodiments of the present invention, and it is obvious that the described embodiments are only some of the embodiments of the present invention, and not all of the embodiments. All other embodiments that a person skilled in the art can obtain based on the embodiments of the present invention without any creative work fall within the scope of protection of the present invention.

[0025] According to an embodiment of the present invention, an embodiment of a method for structural monitoring of a pumped storage power generation unit frame is provided, and the steps shown in the flowchart of the drawings can be performed by a computer system such as a computer-executable instruction set, and although a logical order is shown in the flowchart, it should be noted that in some cases the steps shown or described can be performed in an order different from that shown here.

[0026] This embodiment provides a structural monitoring method for a pumped storage power generation unit frame that can be applied to a pumped storage power generation unit frame. Figure 1 is a flowchart of the structural monitoring method for a pumped storage power generation unit frame according to an embodiment of the present invention. As shown in Figure 1, the process includes the following steps S101 to S106.

[0027] Step S101: Obtain structural parameters of a pumped storage power generation unit frame and operating parameters for a plurality of operating conditions.

[0028] Specifically, the structural parameters of the pumped storage power generation unit frame may include the geometric dimensions, material attributes, and structural connection relationships of the pumped storage power generation unit frame, and the multiple operating conditions of the pumped storage power generation unit in actual operation include two steady-state operating conditions, namely, pump operating condition and turbine operating condition, and three transient operating conditions, namely, turbine braking operating condition and reverse pump operating condition. The operating parameters include operating characteristic parameters of the pumped storage power generation unit frame under each operating condition, working load, and other parameters. The structural parameters of the pumped storage power generation unit frame can be obtained from the operation manual of the pumped storage power generation unit frame, and the frame operating parameters can be collected when the pumped storage power generation unit is operating under various operating conditions.

[0029] Step S102: Construct a frame structure finite element model based on the structural parameters and the operating parameters of multiple operating conditions, extract characteristic parameters for multiple operating conditions based on the operating parameters of the multiple operating conditions, and divide the characteristic parameters for the multiple operating conditions into a test parameter set for multiple operating conditions and a simulation parameter set for multiple operating conditions.

[0030] Specifically, the finite element model is a model created using a finite element analysis method, and is a collection of combinations of units connected by nodes, transmitting forces through the nodes, and being constrained by the nodes. The frame structure finite element model is a model constructed based on the geometric characteristics of the pumped storage power generation unit frame, and represents abstract mechanical relationships. Based on the operating parameters of multiple operating conditions, characteristic parameters of multiple operating conditions are extracted, including but not limited to parameters such as the rotation speed, flow rate, and head of the pumped storage power generation unit.

[0031] The characteristic parameters of the multiple operating conditions are divided into a test parameter set for multiple operating conditions and a simulation parameter set for multiple operating conditions, so as to subsequently perform a simulation analysis of the frame structure finite element model, optimize an initial proxy model, and simulate and predict stress and deformation conditions using the frame structure finite element model and the optimized initial proxy model. The simulation parameter set for the multiple operating conditions is used for the simulation analysis of the finite element model and the optimization of the initial proxy model. The test parameter set for the multiple operating conditions is used for simulating and predicting stress and deformation conditions using the frame structure finite element model and the optimized initial proxy model.

[0032] Step S103: Under a preset operating condition, a simulation analysis is performed on a simulation parameter set of the corresponding operating condition based on the frame structure finite element model, and a first stress and deformation simulation value of the corresponding operating condition of the pumped storage power generation unit frame is obtained.

[0033] Specifically, the preset operating conditions are one of two steady-state operating conditions, namely, pump operating conditions and turbine operating conditions, and three transient operating conditions, namely, pump braking operating conditions, turbine braking operating conditions, and reverse pump operating conditions.

[0034] Under predetermined operating conditions, a simulation analysis is performed on a simulation parameter set for the corresponding operating conditions based on the frame structure finite element model, and a first stress and deformation simulation value for the corresponding operating condition of the pumped storage power generation unit frame is obtained, and the weak points of the pumped storage power generation unit frame are initially determined based on the first stress and deformation simulation value.

[0035] Step S104: Optimizing the preset initial proxy model based on the simulation parameter set of the corresponding operating condition and the first stress and deformation simulation value of the corresponding operating condition to obtain an optimized proxy model.

[0036] Specifically, the initial proxy model can use the Kriging model, which is an unbiased estimation model that minimizes the estimation variance and has the characteristics of local estimation.

[0037] A simulation parameter set of the corresponding operating conditions and a first stress and deformation simulation value of the corresponding operating conditions are input into a preset initial proxy model, and the preset initial proxy model is optimized to obtain an optimized proxy model.

[0038] Step S105: Under a preset operating condition, the test parameter set of the corresponding operating condition is input into the frame structure finite element model and the optimized proxy model, respectively, to obtain the second stress and deformation simulation value of the corresponding operating condition and the stress and deformation prediction value of the corresponding operating condition, respectively.

[0039] Specifically, under a preset operating condition, a test parameter set for the corresponding operating condition is input into the frame structure finite element model to obtain a second stress and deformation simulation value for the corresponding operating condition, and the test parameter set for the corresponding operating condition is input into the optimized proxy model to obtain a stress and deformation prediction value for the corresponding operating condition.

[0040] Step S106: If the second stress and deformation simulation value for the corresponding operating condition and the stress and deformation predicted value for the corresponding operating condition satisfy the preset accuracy condition, the stress and deformation of the pumped storage power generation unit frame is monitored using the stress and deformation predicted value for the corresponding operating condition.

[0041] Specifically, the predetermined accuracy condition is set according to the actual situation and may be within a predetermined accuracy value range, but is not specifically limited herein. Accuracy verification is performed on the second stress and deformation simulation values ​​for the corresponding operating conditions and the predicted stress and deformation values ​​for the corresponding operating conditions according to the predetermined accuracy condition. If the second stress and deformation simulation values ​​for the corresponding operating conditions and the predicted stress and deformation values ​​for the corresponding operating conditions satisfy the predetermined accuracy condition, it means that the predicted stress and deformation values ​​for the corresponding operating conditions predicted using the optimized proxy model are closest to the second stress and deformation simulation values, and the predicted stress and deformation values ​​for the corresponding operating conditions can be used to monitor the stress and deformation of the pumped storage power generation unit frame.

[0042] The structural monitoring method for the pumped storage power generation unit frame of this embodiment includes: performing a simulation analysis on a simulation parameter set for a corresponding operating condition using a frame structure finite element model; obtaining a first simulated stress and deformation value for the corresponding operating condition of the pumped storage power generation unit frame; obtaining an optimized proxy model based on the first simulated stress and deformation value and the extracted simulation parameter set for the corresponding operating condition; inputting a test parameter set for the corresponding operating condition into the frame structure finite element model and the optimized proxy model, respectively, under a preset operating condition; obtaining a second simulated stress and deformation value for the corresponding operating condition and a predicted stress and deformation value for the corresponding operating condition; and if the second simulated stress and deformation value for the corresponding operating condition and the predicted stress and deformation value for the corresponding operating condition meet the preset accuracy condition, monitoring the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation value for the corresponding operating condition, thereby achieving the purpose of comprehensively and real-time monitoring of the stress and deformation conditions at each position of the pumped storage power generation unit frame under various operating conditions and solving the problem of not being able to comprehensively grasp the stress and deformation conditions of the frame.

[0043] This embodiment provides a structural monitoring method for a pumped storage power generation unit frame that can be applied to a pumped storage power generation unit. Figure 2 is a flowchart of the structural monitoring method for a pumped storage power generation unit frame according to an embodiment of the present invention. As shown in Figure 2, the process includes the following steps S201 to S206.

[0044] Step S201: Obtain structural parameters of the pumped storage power generation unit frame and operating parameters of multiple operating conditions. For details, please refer to step S101 in the embodiment shown in Figure 1, and the description will not be repeated here.

[0045] Step S202: construct a frame structure finite element model based on the structural parameters and the operating parameters of multiple operating conditions; extract characteristic parameters for multiple operating conditions based on the operating parameters of the multiple operating conditions; and divide the characteristic parameters for the multiple operating conditions into a test parameter set for multiple operating conditions and a simulation parameter set for multiple operating conditions.

[0046] Specifically, the structural parameters of the pumped storage power generation unit frame include the geometric dimensions, material attributes and structural connection relationships of the pumped storage power generation unit frame, and the operating parameters of the pumped storage power generation unit frame under multiple operating conditions include the operating loads under the multiple operating conditions. In the above step S20, constructing a frame structure finite element model based on the structural parameters and the operating parameters under the multiple operating conditions includes: The method includes using a parametric programming language to construct a frame structure finite element model under a plurality of operating conditions based on the geometric dimensions, material attributes, structural connection relationships, and operating loads under a plurality of operating conditions of the pumped storage power generation unit frame.

[0047] Specifically, based on the design data of the pumped storage power generation unit or three-dimensional scanning technology, a three-dimensional geometric solid model of the pumped storage power generation unit frame is created using CAD software, and the material attributes of the pumped storage power generation unit frame, including density, elastic modulus, and Poisson's ratio, are set. The pumped storage power generation unit frame is discretized using the SOLID187 unit to obtain a finite element mesh model. The SOLID187 unit is a high-order three-dimensional 10-node solid structural unit.

[0048] After analyzing the load conditions during actual operation of the pumped storage power generation unit frame, loads are applied to the frame. The frame is mainly subjected to the gravity of its own structure and auxiliary equipment, and the axial water thrust. Gravity includes the weight of the frame itself, the weight of rotating parts such as runners, and the gravity of accessories such as thrust bearings and oil tanks. The axial water thrust includes the combined force of the water impact force acting on the runners and the water buoyancy. A rough estimate can be obtained based on an empirical formula for the axial water thrust of a Francis turbine, or an accurate value can be obtained through numerical simulation using CFD (Computational Fluid Dynamics).

[0049] The empirical formula is JPEG2026031418000002.jpg15170, where λ is a correction coefficient. According to experimental data in the related art, it is found that under normal operating conditions, λ increases with the increase in the output and flow rate of the pumped storage power generation unit. The value ranges from 0.7 to 0.8. It is recommended that the value be 0.75±0.01 at rated output, 0.77±0.01 when the rated output is exceeded, and a fixed value of 1±0.01 under load dump operating conditions. ω is the rotation speed of the pumped storage power generation unit in rpm, and Q is the flow rate of the unit in m 3 / s, H is the working head of the extraction unit frame in m, and D1 is the nominal diameter of the runner in m.

[0050] The parametric programming language is also called APDL language (ANSYS Parametric Design Language), and is used to analyze the geometric dimensions of the pumped storage power generation unit frame, material attributes, structural connection relationships divided into finite element mesh models, and operating loads under multiple operating conditions, and to create a refined frame structure finite element model.

[0051] Step S203: Under a preset operating condition, a simulation analysis is performed on a simulation parameter set of the corresponding operating condition based on the frame structure finite element model, and a first stress and deformation simulation value of the corresponding operating condition of the pumped storage power generation unit frame is obtained.

[0052] Specifically, the above step S203 includes the following steps S2031 and S2032.

[0053] Step S2031: Under a preset operating condition, the simulation parameter set of the corresponding operating condition is sampled using a Latin hypercube sampling algorithm to obtain input data of the corresponding operating condition.

[0054] Specifically, the Latin Hypercube Sampling algorithm (abbreviated as LHS) is an algorithm for approximately randomly sampling from a multivariate parameter distribution. Under a predetermined operating condition, the Latin Hypercube Sampling algorithm is used to sample a simulation parameter set for the corresponding operating condition to obtain input data for the corresponding operating condition. That is, the obtained input data for the corresponding operating condition is part of the simulation parameter set for the corresponding operating condition. Specifically, the type of input data obtained by sampling is preset according to the actual situation and is not specifically limited here.

[0055] In step S2032, the input data of the corresponding operating conditions is input into the frame structure finite element model to perform static simulation analysis, and obtain the first stress and deformation simulation values ​​of the corresponding operating conditions of the pumped storage power generation unit frame.

[0056] Specifically, as shown in Figure 4, the static simulation analysis is a static simulation analysis of the structure using a frame finite element model for input data of corresponding operating conditions. The pumped storage power generation unit frame is deformed due to the action of pre-set operating conditions and operating loads, and the internal material of the pumped storage power generation unit frame is in a complex force-bearing state. Through the static simulation analysis, the relationship between stress and deformation of the pumped storage power generation unit frame is explained, and finally, the first stress and deformation simulation value of the corresponding operating condition of the pumped storage power generation unit frame is obtained. Using the first stress and deformation simulation value, a stress and deformation cloud map of the entire frame is generated by finite element software, and the weak points of the frame can be identified from the stress and deformation cloud map.

[0057] The structural monitoring method for the pumped storage power generation unit frame in this embodiment involves using a Latin hypercube sampling algorithm to sample a simulation parameter set for a corresponding operating condition under a preset operating condition to obtain input data for the corresponding operating condition, inputting the input data for the corresponding operating condition into a frame structure finite element model to perform static simulation analysis, and obtaining first stress and deformation simulation values ​​for the corresponding operating condition of the pumped storage power generation unit frame. The frame structure finite element model can then be used to realize static simulation analysis under various operating conditions, and the first stress and deformation simulation values ​​under various operating conditions can be obtained. The first stress and deformation simulation values ​​under various operating conditions can be used to determine the weak points of the pumped storage power generation unit frame, and provide a data basis for subsequent optimization of the initial proxy model.

[0058] Step S204: Optimizing the preset initial proxy model based on the simulation parameter set of the corresponding operating condition and the first stress and deformation simulation value of the corresponding operating condition to obtain an optimized proxy model.

[0059] Specifically, the above step S204 includes the following steps S2041 to S2044.

[0060] Step S2041: In a preset operating condition, the characteristic parameters of the simulation parameter set of the corresponding operating condition are sampled using a Latin hypercube sampling algorithm to obtain input data of the corresponding operating condition.

[0061] Specifically, this step is the same as the process of step S2031, and under the preset operating conditions, the characteristic parameters of the simulation parameter set for the corresponding operating conditions are sampled using the Latin hypercube sampling algorithm to obtain input data for the corresponding operating conditions, and the obtained input data for the corresponding operating conditions is the same as the input data for the corresponding operating conditions obtained in step S2031.

[0062] In step S2042, the input data of the corresponding operating conditions and the first stress and deformation simulation values ​​of the corresponding operating conditions are input as sample data points into the preset initial proxy model, and relevant parameters of the initial proxy model are obtained.

[0063] Specifically, the preset initial proxy model uses a Kriging model, and input data of the corresponding operating conditions and the first stress and deformation simulation values ​​of the corresponding operating conditions are input as sample data points into the preset initial proxy model.

[0064] The response values ​​and independent variables in the Kriging model satisfy the relationship in equation (2), JPEG2026031418000003.jpg14170where x=(x1,x2,…,x n ) T is the input data for any corresponding operating condition, i.e., the sample data points, n is the dimension of the sample data points, y(x) is the predicted response value of the corresponding sample data points, and β i (i=1,2,…,n) are the estimated regression coefficients, and f i (x)(i=1,2,…,n) is the basis function vector of the polynomial, and z(x) is the vector of N(0,σ 2 ) and any two sample data points x i and x j The covariance of is added in equation (3), JPEG2026031418000004.jpg10170Here, Cov is the covariance operation, and R(θ,x i , x j ) is the sample data point x i and x j is the correlation function that characterizes the spatial correlation of n ] T is the relevant parameter vector.

[0065] The correlation function may be a linear function, an exponential function, or a Gaussian function. The Gaussian function is generally calculated by Equation (4): JPEG2026031418000005.jpg28170 where N is the dimension of x and θ n is the unknown related parameter of the Kriging model, and d n represents the distance between sample data points xi and xj, as shown in equation (5).

[0066] Step S2043: Calculate the optimal solution of the relevant parameters using the Pelican optimization algorithm.

[0067] Specifically, after determining the correlation function, the prediction of the Kriging model can be further explained as follows: JPEG2026031418000006.jpg31170r T (x * ) can be expressed as follows: JPEG2026031418000007.jpg42170 According to the Kriging model, the unknown data point x * The mean square error or standard deviation s 2 predict, i.e., JPEG2026031418000008.jpg11170where, s 2 is the mean square error or standard deviation, which represents the predicted deviation between the Kriging model and the actual response value. n By solving the above equation, we can obtain the optimized Kriging model.

[0068] Related parameter θ n The mathematical model for the optimization is as follows: JPEG2026031418000009.jpg84170

[0069] Solve equation (11) using the Pelican optimization algorithm.

[0070] The Pelican Optimization Algorithm (POA) is a probabilistic metaheuristic algorithm that simulates the hunting behavior of pelicans. It has relatively good global search ability and optimal local discrimination ability. The process is divided into two stages: the search stage, in which the pelican approaches prey, and the exploitation stage, in which the pelican flies over the water surface.

[0071] (1) Initialization: The pelican population is initialized as follows: JPEG2026031418000010.jpg28170

[0072] The pelican population can be represented as follows: JPEG2026031418000011.jpg48170

[0073] The objective function value vector for the pelican population can be expressed as: JPEG2026031418000012.jpg26170

[0074] (2) Exploration phase, approaching the prey: The search phase is the stage in which pelicans move to the hunting area after determining the location of prey. The mathematical model of their prey approach strategy is as follows: JPEG2026031418000013.jpg45170

[0075] Objective function value F P If r improves on the position, the new position of the pelican is accepted, then we effectively update r as follows: JPEG2026031418000014.jpg32170

[0076] (3) Utilization phase, surface flight: The exploitation phase refers to the stage in which the pelican reaches the water surface, spreads its wings above the water, moves the fish upward, and then collects the prey in its guttural pouch. The mathematical model of the surface flight strategy is as follows: JPEG2026031418000015.jpg26170

[0077] This effectively updates the pelican's position as follows: JPEG2026031418000016.jpg38170

[0078] The Pelican algorithm calculates the relevant parameters θ n Obtain the optimal solution.

[0079] Step S2044: Optimize the initial proxy model using the optimal solution to obtain an optimized proxy model.

[0080] Specifically, the θ of the proxy model n Using a mathematical model of optimization, combined with the Pelican algorithm, nObtain the optimal solution of θ n The initial proxy model is optimized using the optimal solution of (1) to form an optimized Kriging proxy model.

[0081] Step S205: Under a preset operating condition, input the test parameter set for the corresponding operating condition into the frame structure finite element model and the optimized proxy model, respectively, to obtain the second stress and deformation simulation value for the corresponding operating condition and the stress and deformation prediction value for the corresponding operating condition, respectively. For details, please refer to step S105 in the embodiment shown in Figure 1, and the description will not be repeated here.

[0082] Step S206: if the second stress and deformation simulation value of the corresponding operating condition and the predicted stress and deformation value of the corresponding operating condition satisfy the preset accuracy condition, the predicted stress and deformation value of the corresponding operating condition is used to monitor the stress and deformation of the pumped storage power generation unit frame. For details, please refer to step S106 in the embodiment shown in Figure 1, and the description will not be repeated here.

[0083] The structural monitoring method for a pumped storage power generation unit frame in this embodiment involves, under preset operating conditions, using a Latin hypercube sampling algorithm to sample characteristic parameters of a simulation parameter set for corresponding operating conditions to obtain input data for the corresponding operating conditions, inputting the input data for the corresponding operating conditions and the first stress and deformation simulation values ​​for the corresponding operating conditions as sample data points into a preset initial proxy model to obtain relevant parameters of the initial proxy model, calculating an optimal solution for the relevant parameters using a Pelican optimization algorithm, optimizing the initial proxy model using the optimal solution, and obtaining an optimized proxy model, which provides a model basis for subsequent prediction of the stress and deformation conditions of the pumped storage power generation unit frame, thereby improving the prediction accuracy of the stress and deformation conditions.

[0084] This embodiment provides a structural monitoring method for a pumped storage power generation unit frame that can be applied to a pumped storage power generation unit. Figure 3 is a flowchart of the structural monitoring method for a pumped storage power generation unit frame according to an embodiment of the present invention. As shown in Figure 3, the process includes the following steps S301 to S308.

[0085] Step S301: Obtain structural parameters of the pumped storage power generation unit frame and operating parameters of multiple operating conditions. For details, please refer to step S201 in the embodiment shown in Figure 2, and the description will not be repeated here.

[0086] Step S302: construct a frame structure finite element model based on the structural parameters and the operating parameters of a plurality of operating conditions, extract characteristic parameters for the plurality of operating conditions based on the operating parameters of the plurality of operating conditions, and divide the characteristic parameters for the plurality of operating conditions into a test parameter set for the plurality of operating conditions and a simulation parameter set for the plurality of operating conditions. For details, please refer to step S202 in the embodiment shown in Figure 2, and the description will not be repeated here.

[0087] Step S303: Under a preset operating condition, perform a simulation analysis on a set of simulation parameters for the corresponding operating condition based on the frame structure finite element model to obtain a first stress and deformation simulation value for the corresponding operating condition of the pumped storage power generation unit frame. For details, please refer to step S203 in the embodiment shown in Figure 2, and the description will not be repeated here.

[0088] Step S304: Optimize the preset initial proxy model based on the simulation parameter set for the corresponding operating condition and the first stress and deformation simulation value for the corresponding operating condition to obtain an optimized proxy model. For details, see step S204 in the embodiment shown in FIG. 2, and the description will not be repeated here.

[0089] Step S305: Under a preset operating condition, input the test parameter set for the corresponding operating condition into the frame structure finite element model and the optimized proxy model, respectively, to obtain the second stress and deformation simulation value for the corresponding operating condition and the stress and deformation prediction value for the corresponding operating condition, respectively. For details, please refer to step S205 in the embodiment shown in Figure 2, and the description will not be repeated here.

[0090] Step S306: If the second stress and deformation simulation value for the corresponding operating condition and the stress and deformation prediction value for the corresponding operating condition satisfy the preset accuracy condition, the stress and deformation of the pumped storage power generation unit frame is monitored using the stress and deformation prediction value for the corresponding operating condition.

[0091] Specifically, the above step S306 includes the following steps S3061 to S3063.

[0092] Step S3061: Calculate a first relative error between the second simulated stress and deformation values ​​of the corresponding operating condition and the predicted stress and deformation values ​​of the corresponding operating condition.

[0093] Specifically, the absolute error between the second stress and deformation simulation value for the corresponding operating condition and the stress and deformation predicted value for the corresponding operating condition is calculated, and then the ratio between the absolute error and the second stress and deformation simulation value is calculated, and further multiplied by 100% to obtain the first relative error.

[0094] In step S3062, it is determined whether the first relative error satisfies a preset accuracy condition.

[0095] Specifically, the preset accuracy condition can be set according to actual circumstances and is not specifically limited here. For example, in this embodiment, the accuracy condition is set to be equal to or less than 10%, less than 10%, or equal to 10%. Whether the first relative error is equal to or less than 10% is determined.

[0096] The accuracy index of the model is the coefficient of determination R 2is selected, which is shown in the following equation (19). JPEG2026031418000017.jpg54170R 2 The closer to 1, the higher the predictive accuracy of the optimized proxy model.

[0097] Step S3063: if the first relative error satisfies the preset accuracy condition, the stress and deformation predicted values ​​of the corresponding operating conditions are used to monitor the stress and deformation of the pumped storage power generation unit frame.

[0098] Specifically, when the first relative error is 10% or less, the stress and deformation of the pumped storage power generation unit frame are monitored in real time using the predicted stress and deformation values ​​for the corresponding operating conditions.

[0099] In an alternative embodiment, the above step S3063 includes the following steps a1 to a3.

[0100] Step a1: if the first relative error satisfies the preset accuracy condition, collect the actual measured values ​​of stress and deformation of the pumped storage power generation unit frame under the corresponding operating condition.

[0101] Specifically, as shown in Figure 4, in order to more accurately verify the accuracy of the predicted stress and deformation values ​​for the corresponding operating conditions, if the first relative error is less than or equal to 10%, the actual measured stress and deformation values ​​of the pumped storage power generation unit frame for the corresponding operating conditions are collected when the pumped storage power generation unit for the corresponding operating conditions is operating.

[0102] Step a2: Calculate a second relative error between the actual measured stress and deformation values ​​for the corresponding operating conditions and the predicted stress and deformation values ​​for the corresponding operating conditions.

[0103] Specifically, the actual stress and deformation values ​​are obtained from real-time monitoring data collected by sensors attached to the in-situ frame of the pumped storage power generation unit. The absolute error between the actual stress and deformation values ​​of the corresponding operating conditions and the predicted stress and deformation values ​​of the corresponding operating conditions is calculated, and then the ratio between the absolute error and the actual stress and deformation values ​​is calculated, and further multiplied by 100% to obtain the second relative error.

[0104] Step a3: if the second relative error satisfies the preset accuracy condition, the stress and deformation of the pumped storage power generation unit frame are monitored using the predicted stress and deformation values ​​of the corresponding operating conditions.

[0105] Specifically, if the second relative error also satisfies the set accuracy condition, i.e., 10% or less, the stress and deformation of the pumped storage power generation unit frame are monitored in real time using the predicted stress and deformation values ​​for the corresponding operating conditions.

[0106] In the structural monitoring method for a pumped storage power generation unit frame of this embodiment, if the first relative error satisfies the predetermined accuracy condition, the accuracy verification of the second relative error between the actual stress and deformation values ​​for the corresponding operating conditions and the predicted stress and deformation values ​​for the corresponding operating conditions is continued, and if the second relative error satisfies the predetermined accuracy condition, the predicted stress and deformation values ​​for the corresponding operating conditions are used to monitor the stress and deformation of the pumped storage power generation unit frame, thereby improving the accuracy of the verification of the predicted stress and deformation values ​​for the corresponding operating conditions, and making the monitoring of the structure of the pumped storage power generation unit frame using the predicted stress and deformation values ​​for the corresponding operating conditions more accurate.

[0107] Step S307: if the first relative error does not satisfy the preset accuracy condition, add the second stress and deformation simulation values ​​as additional data to the sample data points to obtain additional sample data points, and update the optimized proxy model based on the additional sample data points.

[0108] Specifically, as shown in FIG. 4, if the first relative error does not satisfy the preset accuracy condition, the stress and deformation simulation values ​​are added as additional data to the sample data points to obtain additional sample data points, and the optimized proxy model is updated based on the additional sample data points, in order to further optimize the optimized model.

[0109] Step S308: if the second relative error does not satisfy the preset accuracy condition, update the frame structure finite element model using the preset correction coefficient, add the stress and deformation actual measurement values ​​as additional data to the sample data points to obtain additional sample data points, and update the optimized proxy model based on the additional sample data points.

[0110] Specifically, if the second relative error does not meet the predetermined accuracy condition, it means that the error between the predicted stress and deformation values ​​and the measured stress and deformation values ​​is large, reflecting the inaccuracy of the obtained predicted stress and deformation values. Furthermore, it also reflects the inaccuracy of the first simulated stress and deformation values ​​obtained by the frame structure finite element model. If the first simulated value is input into the predetermined initialization model for optimization, the obtained optimized proxy model will also be inaccurate. To improve the accuracy of the frame structure finite element model and the optimized proxy model, as shown in FIG. 5, the frame structure finite element model is updated using the predetermined correction coefficient, and the measured stress and deformation values ​​are added as additional data to the sample data points to obtain additional sample data points. The optimized proxy model is then updated based on the additional sample data points, thereby further improving the simulation accuracy of the frame structure finite element model and the prediction accuracy of the optimized proxy model.

[0111] In the method for monitoring the structure of a pumped storage power generation unit frame according to this embodiment, if the first relative error does not satisfy the predetermined accuracy condition, the stress and deformation simulation values ​​are added as additional data to the sample data points to obtain additional sample data points, and the optimized proxy model is updated based on the additional sample data points; or if the second relative error does not satisfy the predetermined accuracy condition, the frame structure finite element model is updated using a predetermined correction coefficient, the stress and deformation actual measurement values ​​are added as additional data to the sample data points to obtain additional sample data points, and the optimized proxy model is updated based on the additional sample data points, and the frame structure finite element model is updated using the predetermined correction coefficient, thereby further optimizing the frame structure finite element model and improving the simulation accuracy of the finite element model; and further optimizing the optimized proxy model using the additional sample data points and further improving the prediction accuracy of the optimized proxy model.

[0112] As one or more specific application examples of the embodiments of the present invention, a structure monitoring method for a pumped storage power generation unit frame will be further described with reference to FIGS. 4, 5 and 6, specifically as follows:

[0113] The specific process of the preferred embodiment shown in Figure 4: 1. Obtaining the structural parameters of the pumped storage power generation unit frame and the operating parameters under multiple operating conditions: Step S1: Identify the operating characteristics of the pumped storage power generation unit under each operating condition based on the upper-level scheduling command; Step S2: Collect historical monitoring data of the status of the pumped storage power generation unit; Step S3: Statistically analyze and identify the characteristic parameters of the working load of the pumped storage power generation unit frame; Step S4: Based on steps S1 to S3, determine the stress and deformation ranges of the pumped storage power generation unit frame, determine the dangerous operating conditions of the pumped storage power generation unit frame, and determine the characteristic parameters of the proxy model. The characteristic parameters include, but are not limited to, parameters such as rotation speed, flow rate, and working head. The characteristic parameters are divided into a simulation parameter set and a test parameter set.

[0114] 2. Static simulation analysis process based on finite element method: Step S1: Creating a finite element simulation model of the frame refinement: Based on the rules of geometric dimensions, material attributes, contact connection relationships, etc., the APDL parametric programming language is used to create a finite element simulation model of the frame structure including the frame and its accessories. The process of creating the frame structure finite element simulation model is as follows: Step S11: Based on the design data of the pumped storage power generation unit or three-dimensional scanning technology, a three-dimensional geometric solid model of the frame is created using CAD software; Step S12: Set the material attributes of the pumped storage power generation unit frame, including density, elastic modulus and Poisson's ratio; Step S13: Using the SOLID187 unit to discretize the three-dimensional geometric solid model of the pumped storage power generation unit frame to obtain a finite element mesh model, and using the APDL parametric programming language to create a refined frame structure finite element simulation model including the geometric dimensions of the frame, material attributes and its accessory members; Step S14: After analyzing the load conditions during actual operation of the pumped storage power generation unit frame, loads are applied. The pumped storage power generation unit frame is mainly subjected to the gravity of its own structure and auxiliary equipment and the axial water thrust. Gravity includes the weight of the frame itself, the weight of rotating parts such as runners, and the gravity of accessories such as thrust bearings and oil tanks. The axial water thrust includes the combined force of the water impact force acting on the runners and the buoyancy of the water. A rough estimate can be obtained based on the empirical formula for the axial water thrust of a Francis turbine, or an accurate value can be obtained by numerical simulation using CFD. The empirical formula is JPEG2026031418000018.jpg16170, where λ is a correction coefficient. According to experimental data in the related art, it is found that under normal operating conditions, λ increases with the increase in the output and flow rate of the pumped storage power generation unit. The value ranges from 0.7 to 0.8. It is recommended that the value be 0.75±0.01 at rated output, 0.77±0.01 when the rated output is exceeded, and a fixed value of 1±0.01 under load dump operating conditions. ω is the rotation speed of the pumped storage power generation unit in rpm, and Q is the flow rate of the unit in m 3 / s, H is the working head of the extraction unit frame in m, and D1 is the nominal diameter of the runner in m.

[0115] According to the frame structure finite element model, in the process of finite element analysis, the finite element model under different operating conditions is subjected to different loads, and the magnitude of the operating load needs to be determined according to the operating conditions of the operation.

[0116] Step S2: Based on the frame structure finite element model, a static simulation analysis of sample data points under key operating conditions is performed, the simulation output data is first stress deformation simulation data, the first stress deformation simulation data is used to optimize the predetermined initial proxy model, and the sample data points are obtained from initial sample input data determined by sampling from a simulation parameter set under multiple operating conditions using a Latin hypercube sampling algorithm.

[0117] Step S3: Based on the first stress-deformation simulation data output by the static analysis, a stress-deformation cloud map of the entire frame is obtained using the finite element method, and weak points in the frame can be identified from the stress-deformation cloud map.

[0118] 3. Real-time monitoring process based on optimized proxy model: Step S1: Within the range of the simulation parameter set, the Latin hypercube sampling algorithm is used to obtain initial sample input data, and the input data and the corresponding first stress / deformation simulation value are used as sample data points in combination with the first stress / deformation simulation value output by the finite element simulation result.

[0119] Step S2: Optimize the initial proxy model (i.e., Kriging model) preset by the sample data points to obtain an optimized proxy model. The specific process is as follows:

[0120] Step S21: inputting the input data of the corresponding operating conditions, the first stress and deformation simulation values ​​of the corresponding operating conditions as sample data points into a preset initial proxy model, and obtaining related parameters of the initial proxy model; The response values ​​and independent variables in the Kriging model satisfy the relationship in equation (2), JPEG2026031418000019.jpg14170where x=(x1,x2,…,x n ) T is the input data for any corresponding operating condition, n is the number of sample data points, y(x) is the corresponding predicted response value, and β i (i=1,2,…,n) are the estimated regression coefficients, and f i (x)(i=1,2,…,n) is the basis function vector of the polynomial, and z(x) is the vector of N(0,σ 2 ) and any two sample data points x i and x j The covariance of is added in equation (3), JPEG2026031418000020.jpg9170Here, Cov is the covariance operation, and R(θ,x i , x j ) is the sample data point x i and x j is the correlation function that characterizes the spatial correlation of n ] T is the relevant parameter vector, The correlation function may be a linear function, an exponential function, or a Gaussian function. A Gaussian function is generally calculated by equation (4): JPEG2026031418000021.jpg28170 where N is the dimension of x and θ n is the unknown related parameter of the Kriging model, and d n represents the distance between sample data points xi and xj, as shown in equation (5).

[0121] Step S22: Calculate the optimal solution of the relevant parameters using the Pelican optimization algorithm; Specifically, after determining the correlation function, the prediction of the Kriging model can be further explained as follows: JPEG2026031418000022.jpg31170r T (x * ) can be expressed as follows: JPEG2026031418000023.jpg42170 According to the Kriging model, the unknown data point x * The mean square error or standard deviation s 2 predict, i.e., JPEG2026031418000024.jpg11170where, s 2 is the mean square error or standard deviation, which represents the predicted deviation between the Kriging model and the actual response value, and is used in the Pelican optimization algorithm to calculate θ n By solving, we can obtain the optimized Kriging model, Related parameter θ n The mathematical model for the optimization is as follows: JPEG2026031418000025.jpg86170Solve equation (11) using the Pelican optimization algorithm.

[0122] Step S3: Optimize the initial proxy model using the optimal solution to obtain an optimized proxy model.

[0123] That is, the mathematical model of θn optimization of the proxy model is used, combined with the Pelican optimization algorithm, to obtain the optimal solution of θn, and form an optimized Kriging proxy model.

[0124] Step S4: As shown in Figure 4, under preset operating conditions, input the test parameter sets into the frame structure finite element model to perform simulation analysis, and input them into the optimized proxy model to perform stress and deformation prediction, respectively, to obtain second stress and deformation simulation values ​​and stress and deformation prediction values, and perform accuracy verification on the second stress and deformation simulation values ​​and stress and deformation prediction values ​​using Equation (19) to determine whether they meet the accuracy requirements. If they do not meet the accuracy requirements, execute S5. If they meet the accuracy requirements, monitor the stress and deformation of the pumped storage power generation unit frame using the stress and deformation prediction values ​​under the corresponding operating conditions. The accuracy index of the model is the coefficient of determination R 2 is selected, which is shown in the following equation (19). JPEG2026031418000026.jpg65170

[0125] Step S5: Add the stress and deformation simulation values ​​as additional data to the sample data points to obtain additional sample data points, and update the optimized proxy model based on the additional sample data points to further improve the prediction accuracy of the optimized proxy model.

[0126] In another preferred embodiment, as shown in Fig. 5, the remaining process is the same as the preferred embodiment shown in Fig. 4, and in this embodiment, the accuracy verification is set to two stages, the first stage is to verify the second stress and deformation simulation values ​​and the predicted stress and deformation values, and the second stage is to verify the actual stress and deformation values ​​and the predicted stress and deformation values. The actual stress and deformation values ​​are obtained from real-time monitoring data collected by sensors attached to the in-situ frame of the pumped storage power generation unit.

[0127] S4: Under the preset operating conditions, the test parameter sets are input into the frame structure finite element model to perform simulation analysis, and then input into the optimized proxy model to perform stress and deformation prediction, respectively, to obtain second stress and deformation simulation values ​​and stress and deformation prediction values, and perform relative error analysis between the second stress and deformation simulation values ​​and the stress and deformation prediction values ​​to determine whether the accuracy requirements are met.

[0128] S41: First-stage accuracy verification: This is the accuracy verification of the second stress and deformation simulation values ​​and the stress and deformation prediction values. If the accuracy requirements are not met, S5 is executed, and if the accuracy requirements are met, S42 is executed.

[0129] S42: Second-stage accuracy verification: This is the accuracy verification of the stress and deformation measured values ​​and the stress and deformation predicted values. If the accuracy requirements are not met, the frame structure finite element model is updated using a preset correction coefficient, and the finite element simulation analysis process is performed again.

[0130] S5: Add the stress and deformation simulation values ​​as additional data to the sample data points to obtain additional sample data points, and update the optimized proxy model based on the additional sample data points to further improve the prediction accuracy of the optimized proxy model.

[0131] In yet another preferred embodiment, as shown in Figure 6, the remaining process is the same as the preferred embodiment shown in Figure 4, and the accuracy verification is set to two stages, the first stage is to verify the second stress and deformation simulation values ​​and the predicted stress and deformation values, and the second stage is to verify the actual stress and deformation values ​​and the predicted stress and deformation values. The actual stress and deformation values ​​are obtained from real-time monitoring data collected by sensors attached to the in-situ frame of the pumped storage power generation unit.

[0132] The differences are explained below.

[0133] S4: Under the preset operating conditions, the test parameter sets are input into the frame structure finite element model to perform simulation analysis, and then input into the optimized proxy model to perform stress and deformation prediction, respectively, to obtain second stress and deformation simulation values ​​and stress and deformation prediction values, and perform relative error analysis between the second stress and deformation simulation values ​​and the stress and deformation prediction values ​​to determine whether the accuracy requirements are met.

[0134] S41: First-stage accuracy verification: This is the accuracy verification of the second stress and deformation simulation values ​​and the stress and deformation prediction values. If the accuracy requirements are not met, S5 is executed, and if the accuracy requirements are met, S42 is executed.

[0135] S42: Second-stage accuracy verification: This is the accuracy verification of the actual stress and deformation values ​​and the predicted stress and deformation values. If the accuracy requirements are not met, S6 is executed.

[0136] S5: Add the stress and deformation simulation values ​​as additional data to the sample data points to obtain additional sample data points, and update the optimized proxy model based on the additional sample data points to further improve the prediction accuracy of the optimized proxy model.

[0137] S6: Add the actual stress and deformation measurements as additional data to the sample data points to obtain additional sample data points, and update the optimized proxy model based on the additional sample data points to further improve the prediction accuracy of the optimized proxy model.

[0138] The structural monitoring method for the pumped storage power generation unit frame of this embodiment includes: performing a simulation analysis on a simulation parameter set for a corresponding operating condition using a frame structure finite element model; obtaining a first simulated stress and deformation value for the corresponding operating condition of the pumped storage power generation unit frame; obtaining an optimized proxy model based on the first simulated stress and deformation value and the extracted simulation parameter set for the corresponding operating condition; inputting a test parameter set for the corresponding operating condition into the frame structure finite element model and the optimized proxy model, respectively, under a preset operating condition; obtaining a second simulated stress and deformation value for the corresponding operating condition and a predicted stress and deformation value for the corresponding operating condition; and if the second simulated stress and deformation value for the corresponding operating condition and the predicted stress and deformation value for the corresponding operating condition meet the preset accuracy condition, using the predicted stress and deformation value for the corresponding operating condition to monitor the stress and deformation of the pumped storage power generation unit frame, thereby achieving the purpose of comprehensively and real-time monitoring of the stress and deformation conditions at each position of the pumped storage power generation unit frame under various operating conditions. Through the set accuracy verification, the prediction accuracy of the frame structure finite element model and the optimized proxy model is further improved, providing an accurate model basis for subsequent monitoring of the stress and deformation of the pumped storage power generation unit frame.

[0139] This embodiment further provides a structure monitoring device for a pumped storage power generation unit frame, which is used to realize the above-mentioned embodiments and preferred embodiments, and what has already been described will not be described again. In the following, the term "module" refers to a combination of software and / or hardware that can realize a predetermined function. Although it is preferable that the device described in the following embodiment is realized by software, it is also conceivable that it can be realized by hardware or a combination of software and hardware.

[0140] This embodiment provides a structure monitoring device for a pumped storage power generation unit frame, and as shown in FIG. 7, the device includes: an acquisition module 701 for acquiring structural parameters of the pumped storage power generation unit frame and operating parameters of a plurality of operating conditions; a construction module 702 for constructing a frame structure finite element model according to the structural parameters and the operating parameters of the plurality of operating conditions, extracting characteristic parameters for the plurality of operating conditions according to the operating parameters of the plurality of operating conditions, and dividing the characteristic parameters for the plurality of operating conditions into test parameter sets for the plurality of operating conditions and simulation parameter sets for the plurality of operating conditions; and a simulation module 703 for performing a simulation analysis on the simulation parameter sets of the corresponding operating conditions based on the frame structure finite element model under a predetermined operating condition, and obtaining first simulated stress and deformation values ​​of the corresponding operating condition of the pumped storage power generation unit frame. an optimization module 704 for optimizing a predetermined initial proxy model based on the characteristic parameters of the simulation parameter set for the corresponding operating condition and the first simulated stress and deformation values ​​for the corresponding operating condition to obtain an optimized proxy model; a test module 705 for inputting, under the predetermined operating condition, the test parameter set for the corresponding operating condition into the frame structure finite element model and the optimized proxy model, respectively, to obtain second simulated stress and deformation values ​​for the corresponding operating condition and predicted stress and deformation values ​​for the corresponding operating condition, respectively; and a monitoring module 706 for monitoring the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation values ​​for the corresponding operating condition when the second simulated stress and deformation values ​​for the corresponding operating condition and the predicted stress and deformation values ​​for the corresponding operating condition satisfy a predetermined accuracy condition.

[0141] The structural parameters of the pumped storage power generation unit frame include geometric dimensions, material attributes and structural connection relationships of the pumped storage power generation unit frame, and the operating parameters of the pumped storage power generation unit frame for multiple operating conditions include operating loads at the multiple operating conditions, and the construction module 702 is further used to construct a frame structural finite element model at the multiple operating conditions based on the geometric dimensions, material attributes, structural connection relationships and operating loads at the multiple operating conditions of the pumped storage power generation unit frame using a parametric programming language.

[0142] In some alternative embodiments, the simulation module 703 includes a first sampling unit for sampling a simulation parameter set for a corresponding operating condition using a Latin hypercube sampling algorithm at a preset operating condition to obtain input data for the corresponding operating condition, and a simulation unit for inputting the input data for the corresponding operating condition into a frame structure finite element model to perform static simulation analysis and obtain first stress and deformation simulation values ​​for the corresponding operating condition of the pumped storage power generation unit frame.

[0143] In some alternative embodiments, the optimization module 704 includes a second sampling unit for sampling characteristic parameters of the simulation parameter set for a corresponding operating condition using a Latin hypercube sampling algorithm at a preset operating condition to obtain input data for the corresponding operating condition; an input unit for inputting the input data for the corresponding operating condition and the first stress and deformation simulation values ​​for the corresponding operating condition as sample data points into a preset initial proxy model to obtain relevant parameters of the initial proxy model; a first calculation unit for calculating an optimal solution of the relevant parameters using the Pelican optimization algorithm; and an optimization unit for optimizing the initial proxy model using the optimal solution to obtain an optimized proxy model.

[0144] In some alternative embodiments, the monitoring module 706 includes a second calculation unit for calculating a first relative error between a second simulated stress and deformation value for the corresponding operating condition and a predicted stress and deformation value for the corresponding operating condition, a determination unit for determining whether the first relative error satisfies a predetermined accuracy condition, and a monitoring unit for monitoring the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation value for the corresponding operating condition if the first relative error satisfies the predetermined accuracy condition.

[0145] In some alternative embodiments, the monitoring unit includes a collection subunit for collecting actual measured stress and deformation values ​​of the pumped storage power generation unit frame for a corresponding operating condition when the first relative error satisfies a preset accuracy condition; a calculation subunit for calculating a second relative error between the actual measured stress and deformation values ​​for the corresponding operating condition and the predicted stress and deformation values ​​for the corresponding operating condition; and a monitoring subunit for monitoring the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation values ​​for the corresponding operating condition when the second relative error satisfies a preset accuracy condition.

[0146] The structure monitoring device for the pumped storage power generation unit frame further includes a first updating module for, when the first relative error does not satisfy the predetermined accuracy condition, adding simulated stress and deformation values ​​to the sample data points as additional data to obtain additional sample data points, and updating the optimized proxy model based on the additional sample data points; and, when the second relative error does not satisfy the predetermined accuracy condition, updating the frame structure finite element model using a predetermined correction coefficient, adding actual measured stress and deformation values ​​to the sample data points as additional data to obtain additional sample data points, and updating the optimized proxy model based on the additional sample data points.

[0147] Further functional explanations of the above modules and units are the same as those of the corresponding embodiments, so duplicate explanations will be omitted here.

[0148] The structural monitoring device for the pumped storage power generation unit frame in this embodiment is presented in the form of a functional unit, where unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory running one or more software or fixed programs, and / or other device capable of providing the above functionality.

[0149] The embodiment of the present invention further provides a computer device having a structure monitoring device for a pumped storage power generation unit frame as shown in FIG.

[0150] Referring to FIG. 8, FIG. 8 is a structural diagram of a computer device according to an alternative embodiment of the present invention. As shown in FIG. 8, the computer device includes one or more processors 10, memory 20, and interfaces, including high-speed and low-speed interfaces, for connecting each component. The components are communicatively connected to each other via different buses and may be mounted on a common motherboard or otherwise attached as needed. The processor can process instructions executed within the computer device, including instructions stored in or on memory for displaying GUI graphic information on an external input / output device (e.g., a display device coupled to the interface). In some alternative embodiments, multiple processors and / or multiple buses may be used along with multiple memories as needed. Similarly, multiple computer devices may be connected, each performing a portion of the required operations (e.g., functioning as a server array, a set of blade servers, or a multiprocessor system). FIG. 8 uses one processor 10 as an example.

[0151] 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 an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable logic gate array, a generic array logic, or any combination thereof.

[0152] Here, the memory 20 stores instructions executable by at least one processor 10, causing the at least one processor 10 to execute and realize the methods shown in the above embodiments.

[0153] Memory 20 may include a program storage area capable of storing an operating system and / or application programs required for at least one function, and a data storage area capable of storing data generated in response to use of the computing device. Memory 20 may also include high-speed random access memory and may further include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, memory 20 may optionally include memory located remotely from processor 10, and these remote memories may be connected to the computing device via a network. Examples of such networks include, but are not limited to, the Internet, a corporate intranet, a local area network, a mobile communications network, and combinations thereof.

[0154] Memory 20 may include volatile memory, such as random access memory, or may include non-volatile memory, such as flash memory, a hard disk, or a solid state drive, or memory 20 may include a combination of the above types of memory.

[0155] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30, and the output device 40 can be connected by a bus or other methods, and FIG. 8 shows the connection by a bus as an example.

[0156] The input device 30 can receive input numeric or character information and generate key signal inputs related to user settings and function control of the computing device, and can include, for example, a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 can include a display device, an auxiliary lighting device (e.g., LED), and a tactile feedback device (e.g., vibration motor), etc. The display device can include, but is not limited to, a liquid crystal display, a light-emitting diode display, and a plasma display. In some alternative embodiments, the display device can be a touch screen.

[0157] An embodiment of the present invention further provides a computer-readable storage medium, and the method according to the embodiment of the present invention may be implemented in hardware and firmware, or may be recordable in a storage medium, or may be implemented as computer code downloaded over a network, originally stored in a remote storage medium or a non-transitory machine-readable storage medium, but stored in a local storage medium, whereby the method described herein may be processed by software stored in a storage medium using a general-purpose computer, a special-purpose processor, or programmable or special-purpose hardware. Here, the storage medium may be a magnetic disk, optical disk, read-only memory, random-access memory, flash memory, hard disk, solid-state drive, etc., and may also include a combination of the above types of memory. It will be understood that a computer, processor, microprocessor controller, or programmable hardware may include a storage component capable of storing or receiving software or computer code, and when the software or computer code is accessed and executed by the computer, processor, or hardware, it implements the method described in the embodiment.

[0158] Although the embodiments of the present invention have been described with reference to the drawings, those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications fall within the scope defined by the appended claims.

Claims

1. A method for monitoring the structure of a pumped storage power generation unit frame, comprising: obtaining structural parameters of a pumped storage power generation unit frame and operating parameters for a plurality of operating conditions; constructing a frame structure finite element model based on the structural parameters and operating parameters of a plurality of operating conditions, extracting characteristic parameters for a plurality of operating conditions based on the operating parameters of the plurality of operating conditions, and dividing the characteristic parameters for the plurality of operating conditions into a test parameter set for a plurality of operating conditions and a simulation parameter set for a plurality of operating conditions, wherein the structural parameters of the pumped storage power generation unit frame include geometric dimensions, material attributes and structural connection relationships of the pumped storage power generation unit frame, and the operating parameters of the plurality of operating conditions of the pumped storage power generation unit frame include working loads under the plurality of operating conditions; Under a preset operating condition, performing a simulation analysis on a set of simulation parameters for the corresponding operating condition based on the frame structure finite element model to obtain a first stress and deformation simulation value for the corresponding operating condition of the pumped storage power generation unit frame; optimizing a preset initial proxy model according to the simulation parameter set of the corresponding operating condition and the first stress and deformation simulation value of the corresponding operating condition to obtain an optimized proxy model; In a preset operating condition, inputting a test parameter set of the corresponding operating condition into the frame structure finite element model and the optimized proxy model, respectively, to obtain a second stress and deformation simulation value of the corresponding operating condition and a stress and deformation prediction value of the corresponding operating condition, respectively; and monitoring the stress and deformation of the pumped storage power generation unit frame using the stress and deformation predicted value of the corresponding operating condition when the second stress and deformation simulation value of the corresponding operating condition and the predicted stress and deformation value of the corresponding operating condition satisfy a predetermined accuracy condition; constructing a frame structure finite element model based on the structural parameters and operating parameters of a plurality of operating conditions, A structural monitoring method for a pumped storage power generation unit frame, comprising the step of constructing a frame structure finite element model under a plurality of operating conditions based on the geometric dimensions, material attributes, structural connection relationships and operating loads under a plurality of operating conditions of the pumped storage power generation unit frame using a parametric programming language.

2. The step of performing a simulation analysis on a set of simulation parameters for a predetermined operating condition based on the frame structure finite element model to obtain a first stress and deformation simulation value for the corresponding operating condition of the pumped storage power generation unit frame, under the predetermined operating condition, includes: At a preset operating condition, sampling a simulation parameter set of the corresponding operating condition using a Latin hypercube sampling algorithm to obtain input data of the corresponding operating condition; 2. The method according to claim 1, further comprising: inputting the input data of the corresponding operating conditions into a frame structure finite element model to perform a static simulation analysis, and obtaining first stress and deformation simulation values ​​of the corresponding operating conditions of the pumped storage power generation unit frame.

3. the step of optimizing a preset initial proxy model based on the simulation parameter set of the corresponding operating condition and the first stress and deformation simulation value of the corresponding operating condition to obtain an optimized proxy model includes: In a preset operating condition, sampling characteristic parameters of the simulation parameter set of the corresponding operating condition using a Latin hypercube sampling algorithm to obtain input data of the corresponding operating condition; inputting the input data of the corresponding operating conditions and the first stress and deformation simulation values ​​of the corresponding operating conditions as sample data points into a preset initial proxy model to obtain relevant parameters of the initial proxy model; calculating an optimal solution for the relevant parameters using the Pelican optimization algorithm; and optimizing the initial proxy model using an optimal solution to obtain an optimized proxy model.

4. When the second stress and deformation simulation value for the corresponding operating condition and the predicted stress and deformation value for the corresponding operating condition satisfy a predetermined accuracy condition, the step of monitoring the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation value for the corresponding operating condition includes: calculating a first relative error between a second simulated stress and deformation value for the corresponding operating condition and a predicted stress and deformation value for the corresponding operating condition; determining whether the first relative error satisfies a preset accuracy condition; 4. The method according to claim 3, further comprising the step of: if the first relative error satisfies a preset accuracy condition, monitoring the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation values ​​of the corresponding operating condition.

5. When the first relative error satisfies a predetermined accuracy condition, the step of monitoring the stress and deformation of the pumped storage power generation unit frame using the stress and deformation predicted value of the corresponding operating condition includes: If the first relative error satisfies the predetermined accuracy condition, collecting actual measured values ​​of stress and deformation of the pumped storage power generation unit frame under the corresponding operating condition; Calculating a second relative error between the actual stress and deformation values ​​for the corresponding operating conditions and the predicted stress and deformation values ​​for the corresponding operating conditions; The method according to claim 4, further comprising the step of: if the second relative error satisfies a preset accuracy condition, monitoring the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation values ​​of the corresponding operating condition.

6. if the first relative error does not satisfy a preset accuracy condition, adding the second stress and deformation simulation values ​​as additional data to the sample data points to obtain additional sample data points, and updating the optimized proxy model based on the additional sample data points; Or, 6. The method of claim 5, further comprising: if the second relative error does not satisfy the predetermined accuracy condition, updating the frame structure finite element model using a predetermined correction coefficient; and adding the stress and deformation actual measurement values ​​as additional data to sample data points to obtain additional sample data points, and updating the optimized proxy model based on the additional sample data points.

7. A structure monitoring device for a pumped storage power generation unit frame, an acquisition module for acquiring structural parameters of the pumped storage power generation unit frame and operating parameters of a plurality of operating conditions; a construction module for constructing a frame structure finite element model based on the structural parameters and operating parameters of a plurality of operating conditions, extracting characteristic parameters for a plurality of operating conditions based on the operating parameters of the plurality of operating conditions, and dividing the characteristic parameters for the plurality of operating conditions into a test parameter set for a plurality of operating conditions and a simulation parameter set for a plurality of operating conditions, wherein the structural parameters of the pumped storage power generation unit frame include geometric dimensions, material attributes, and structural connection relationships of the pumped storage power generation unit frame, and the operating parameters of the plurality of operating conditions of the pumped storage power generation unit frame include working loads under the plurality of operating conditions; a simulation module for performing a simulation analysis on a set of simulation parameters for a corresponding operating condition based on the frame structure finite element model under a predetermined operating condition, to obtain a first stress and deformation simulation value for the corresponding operating condition of the pumped storage power generation unit frame; an optimization module for optimizing a preset initial proxy model according to the characteristic parameters of the simulation parameter set for the corresponding operating condition and the first stress and deformation simulation values ​​for the corresponding operating condition to obtain an optimized proxy model; a test module for inputting a test parameter set of a corresponding operating condition into the frame structure finite element model and the optimized proxy model, respectively, under a preset operating condition, to obtain a second stress and deformation simulation value of the corresponding operating condition and a stress and deformation prediction value of the corresponding operating condition, respectively; a monitoring module for monitoring the stress and deformation of the pumped storage power generation unit frame using the predicted stress and deformation values ​​for the corresponding operating conditions when the second simulated stress and deformation values ​​for the corresponding operating conditions and the predicted stress and deformation values ​​for the corresponding operating conditions satisfy a predetermined accuracy condition; constructing a frame structure finite element model based on the structural parameters and operating parameters of a plurality of operating conditions, A structural monitoring device for a pumped storage power generation unit frame, comprising a step of constructing a frame structure finite element model under a plurality of operating conditions based on the geometric dimensions, material attributes, structural connection relationships and operating loads under a plurality of operating conditions of the pumped storage power generation unit frame using a parametric programming language.

8. A computer device comprising: A computer device comprising a memory and a processor, the memory and the processor being communicatively connected, computer instructions stored in the memory, and the processor executing the computer instructions to perform the structural monitoring method for a pumped storage power generation unit frame according to any one of claims 1 to 6.

9. A computer-readable storage medium having stored thereon computer instructions for causing a computer to execute the method for monitoring the structure of a pumped storage power generation unit frame according to any one of claims 1 to 6.

Images