Vortex-induced vibration regulation and model training method, device, equipment and storage medium

Abstract

The present disclosure provides a vortex-induced vibration regulation and model training method, apparatus, equipment, and storage medium, which relate to the field of artificial intelligence technology, and in particular to the field of deep learning. The specific implementation scheme is: obtaining the vortex-induced vibration data of the vortex-induced vibration equipment in the candidate time period; inputting the vortex-induced vibration data into the vortex-induced vibration deep learning model to obtain the predicted vortex-induced vibration data of the vortex-induced vibration equipment in the next time period; obtaining the target vortex-induced vibration data that the vortex-induced vibration equipment is expected to achieve in the next time period; determining the control instructions based on the predicted vortex-induced vibration data and the target vortex-induced vibration data; and adjusting the vortex-induced vibration equipment using the control instructions. The present disclosure can improve the real-time performance of the regulation of the vortex-induced vibration equipment.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to fields such as deep learning. Background Art

[0002] Vortex-induced vibration (VIV) is a type of flow-induced vibration, the self-excited vibration of a structure caused by fluid flowing through it. For example, in engineering fields such as aviation, shipbuilding, and automotive, typical VIV phenomena include flutter caused by aerodynamic loads on aircraft wings, fatigue vibration of offshore oil pipelines caused by ocean currents, and the oscillation of span bridges under wind loads. Summary of the Invention

[0003] The present invention provides a vortex-induced vibration regulation and model training method, device, equipment and storage medium.

[0004] According to a first aspect of the present disclosure, a vortex-induced vibration regulation method is provided, comprising:

[0005] Acquire vortex-induced vibration data of the vortex-induced vibration equipment during a candidate period;

[0006] Inputting the vortex-induced vibration data into a vortex-induced vibration deep learning model to obtain predicted vortex-induced vibration data of the vortex-induced vibration device in the next time period;

[0007] Acquire target vortex-induced vibration data expected to be achieved by the vortex-induced vibration device in the next time period;

[0008] determining a control instruction based on the predicted vortex-induced vibration data and the target vortex-induced vibration data;

[0009] The vortex-induced vibration equipment is adjusted using the control instructions.

[0010] According to a second aspect of the present disclosure, a vortex-induced vibration deep learning model training method is provided, comprising:

[0011] Acquire a plurality of sample vortex-induced vibration data, wherein the sample vortex-induced vibration data is collected using a vortex-induced vibration device;

[0012] The vortex-induced vibration deep learning model is trained by using the multiple sample vortex-induced vibration data through the flow field control equation and the structural vibration control equation, wherein the structural vibration control equation is related to the flow field control equation.

[0013] According to a third aspect of the present disclosure, there is provided a vortex-induced vibration regulating device, comprising:

[0014] A first acquisition module is used to acquire vortex-induced vibration data of the vortex-induced vibration equipment in a candidate time period;

[0015] A data acquisition module is used to input the vortex-induced vibration data into a vortex-induced vibration deep learning model to obtain predicted vortex-induced vibration data of the vortex-induced vibration device in the next time period;

[0016] A second acquisition module is used to acquire target vortex-induced vibration data that the vortex-induced vibration device is expected to achieve in the next time period;

[0017] a determination module, configured to determine a control instruction based on the predicted vortex-induced vibration data and the target vortex-induced vibration data;

[0018] An adjustment module is used to adjust the vortex-induced vibration equipment using the control instruction.

[0019] According to a fourth aspect of the present disclosure, a vortex-induced vibration deep learning model training device is provided, comprising:

[0020] An acquisition module is used to acquire a plurality of sample vortex-induced vibration data, wherein the sample vortex-induced vibration data is collected by using a vortex-induced vibration device;

[0021] A training module is used to use the multiple sample vortex-induced vibration data to train a vortex-induced vibration deep learning model through a flow field control equation and a structural vibration control equation, wherein the structural vibration control equation is related to the flow field control equation.

[0022] According to a fifth aspect of the present disclosure, there is provided an electronic device, including:

[0023] at least one processor; and

[0024] a memory communicatively connected to the at least one processor; wherein,

[0025] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of the first aspect or the second aspect.

[0026] According to a sixth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method of any one of the first aspect or the second aspect.

[0027] According to a seventh aspect of the present disclosure, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the method according to any one of the first aspect or the second aspect.

[0028] The contents described in this section are not intended to identify the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings are provided to facilitate a better understanding of the present invention and do not constitute a limitation of the present disclosure.

[0030] Figure 1 is a flow chart of the vortex-induced vibration regulation method provided by the present disclosure;

[0031] Figure 2 is a schematic diagram of adjusting the vortex-induced vibration device in an embodiment of the present disclosure;

[0032] Figure 3 This is a flow chart of the vortex-induced vibration deep learning model training method provided by an embodiment of the present disclosure;

[0033] Figure 4 is a schematic diagram of constructing a vortex-induced vibration experimental device in an embodiment of the present disclosure;

[0034] Figure 5 Schematic diagram of a vortex-induced vibration deep learning model training method provided by an embodiment of the present disclosure;

[0035] Figure 6 Schematic diagram of the structure of the vortex-induced vibration regulating device provided by an embodiment of the present disclosure;

[0036] Figure 7 Schematic diagram of the structure of the vortex-induced vibration deep learning model training device provided by an embodiment of the present disclosure;

[0037] Figure 8 It is a block diagram of an electronic device used to implement the vortex-induced vibration regulation method or vortex-induced vibration deep learning model training method of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0038] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0039] The present disclosure provides a vortex-induced vibration regulation method, comprising:

[0040] Acquire vortex-induced vibration data of the vortex-induced vibration equipment during a candidate period;

[0041] Input the vortex-induced vibration data into the vortex-induced vibration deep learning model to obtain the predicted vortex-induced vibration data of the vortex-induced vibration equipment in the next period;

[0042] Obtain target vortex-induced vibration data that the vortex-induced vibration equipment is expected to achieve in the next period;

[0043] Determining control instructions based on the predicted vortex-induced vibration data and the target vortex-induced vibration data;

[0044] Use control instructions to adjust the vortex-induced vibration equipment.

[0045] In the embodiment of the present disclosure, the predicted vortex-induced vibration data of the vortex-induced vibration equipment in the next time period can be quickly obtained through the vortex-induced vibration deep learning model based on the vortex-induced vibration data of the vortex-induced vibration equipment in the candidate time period, and the predicted vortex-induced vibration data and the target vortex-induced vibration data expected to be achieved by the vortex-induced vibration equipment are used to determine the control instructions to adjust the vortex-induced vibration equipment, thereby improving the real-time performance of the adjustment of the vortex-induced vibration equipment.

[0046] Figure 1 This is a flow chart of the vortex-induced vibration regulation method provided by the present disclosure, referring to Figure 1 The vortex-induced vibration regulation method provided by the present disclosure may include:

[0047] S101, obtaining vortex-induced vibration data of a vortex-induced vibration device in a candidate time period.

[0048] The candidate time period may include the current moment, may include historical moments, or may include the current moment and historical moments.

[0049] The vortex-induced vibration device may include a device capable of generating vortex-induced vibrations. For example, the vortex-induced vibration device includes a vortex-induced vibration experimental device, that is, a vortex-induced vibration experimental device is constructed to generate vortex-induced vibrations.

[0050] Vortex-induced vibration data, i.e., data related to vortex-induced vibration, may include data about an object vibrating in a vortex-induced vibration device, such as the object's vibration amplitude and flow field information. In one implementation, the vortex-induced vibration device includes a vortex-induced vibration test device, and the vibrating object in the vortex-induced vibration test device is a test piece. The vortex-induced vibration data may include the test piece's vibration amplitude and flow field information, and the flow field information may include vibration velocity and pressure.

[0051] S102, inputting the vortex-induced vibration data into the vortex-induced vibration deep learning model to obtain the predicted vortex-induced vibration data of the vortex-induced vibration equipment in the next time period.

[0052] The next time period may include the next time instant after the current time instant, or multiple time instants after the current time instant.

[0053] A vortex-induced vibration deep learning model can be pre-trained. After obtaining the vortex-induced vibration data of the vortex-induced vibration device in the candidate time period, the vortex-induced vibration data is input into the vortex-induced vibration deep learning model to obtain the predicted vortex-induced vibration data of the vortex-induced vibration device in the next time period. Among them, the vortex-induced vibration deep learning model can be pre-trained using sample vortex-induced vibration data. For example, multiple sample vortex-induced vibration data are input into the vortex-induced vibration deep learning model respectively to obtain an output value for the sample vortex-induced vibration data, which can also be called a vortex-induced vibration prediction value, until the number of training times reaches a preset number or the output value obtained satisfies the constraint equation. The output value satisfying the constraint equation can be a value in the convergence domain of the loss value calculated based on the constraint equation or equal to 0.

[0054] In one possible implementation, the vortex-induced vibration deep learning model includes a flow field information neural network and a structural vibration neural network;

[0055] The step of inputting the vortex-induced vibration data into a vortex-induced vibration deep learning model to obtain predicted vortex-induced vibration data of the vortex-induced vibration device in the next time period includes:

[0056] The vortex-induced vibration data is input into a vortex-induced vibration deep learning model, the flow field information is obtained through the flow field information neural network, and the vibration amplitude is obtained through the structural vibration neural network.

[0057] The vortex-induced vibration deep learning model can include two networks: flow field information neural network and structural vibration neural network.

[0058] The acquired vortex-induced vibration data is input into the vortex-induced vibration deep learning model, and the flow field information such as vibration velocity (including horizontal velocity and vertical velocity) and pressure are obtained through the flow field control neural network. The vibration amplitude corresponding to the vortex-induced vibration data is obtained through the structural vibration neural network. The predicted vortex-induced vibration data of the vortex-induced vibration equipment in the next time period includes the flow field information obtained through the flow field control neural network and the vibration amplitude obtained through the structural vibration neural network.

[0059] In the embodiment of the present disclosure, the vortex-induced vibration deep learning model conveniently obtains the predicted vortex-induced vibration data of the vortex-induced vibration equipment in the next time period through two networks.

[0060] S103, obtaining target vortex-induced vibration data that the vortex-induced vibration equipment is expected to achieve in the next time period.

[0061] The target vortex-induced vibration data is the vortex-induced vibration data that the vortex-induced vibration equipment is expected to achieve in the next time period, and the target vortex-induced vibration data is acquired.

[0062] S104: Determine a control instruction based on the predicted vortex-induced vibration data and the target vortex-induced vibration data.

[0063] S105: Regulate the vortex-induced vibration equipment using control instructions.

[0064] The purpose of the adjustment is to make the actual vortex-induced vibration data of the vortex-induced vibration equipment in the next period reach the target vortex-induced vibration data as much as possible.

[0065] The control instruction is a control instruction for controlling the vortex-induced vibration device to achieve the target vortex-induced vibration data as much as possible in the next time period.

[0066] The control instructions may include instructions for directly controlling the vortex-induced vibration equipment, or may include instructions for controlling other equipment connected to the vortex-induced vibration equipment. By controlling the other equipment, the vortex-induced vibration equipment is driven to change, thereby enabling the vortex-induced vibration equipment to achieve the target vortex-induced vibration data as much as possible in the next time period.

[0067] A difference between the predicted vortex-induced vibration data and the target vortex-induced vibration data may be determined, and a control instruction may be determined based on the difference to eliminate the difference.

[0068] For example, the control instruction can also be understood as a control algorithm, which may include a control algorithm corresponding to the vibration amplitude, frequency, and constraint speed.

[0069] In an embodiment of the present disclosure, the predicted vortex-induced vibration data of the vortex-induced vibration equipment in the next time period can be quickly obtained through a vortex-induced vibration deep learning model based on the vortex-induced vibration data of the vortex-induced vibration equipment in the candidate time period. When there is a difference between the predicted vortex-induced vibration data and the target vortex-induced vibration data that the vortex-induced vibration equipment is expected to achieve, the control instructions are determined using the predicted vortex-induced vibration data and the target vortex-induced vibration data that the vortex-induced vibration equipment is expected to achieve, so as to adjust the vortex-induced vibration equipment. When there is a difference between the predicted vortex-induced vibration data and the target vortex-induced vibration data, the vortex-induced vibration equipment can be quickly adjusted, thereby improving the real-time performance of the adjustment of the vortex-induced vibration equipment.

[0070] In the disclosed embodiment, the vortex-induced vibration equipment and the vortex-induced vibration deep learning model are combined to realize the virtual-real combination, integrated verification and online adjustment of the vortex-induced vibration deep learning model and the vortex-induced vibration equipment such as test equipment.

[0071] In an optional embodiment, the vortex-induced vibration deep learning model can be trained using historical vortex-induced vibration data of the vortex-induced vibration device. The historical vortex-induced vibration data can include vortex-induced vibration data corresponding to a time before the candidate time period. This historical vortex-induced vibration data is used as sample vortex-induced vibration data, and the vortex-induced vibration deep learning model is trained using multiple sample vortex-induced vibration data.

[0072] For example, vortex-induced vibration equipment includes vortex-induced vibration experimental equipment.

[0073] In the embodiment of the present disclosure, the historical vortex-induced vibration data of the vortex-induced vibration equipment (the vortex-induced vibration data corresponding to the time before the above-mentioned candidate time period) can be used to train a vortex-induced vibration deep learning model. In this way, after obtaining the vortex-induced vibration data of the vortex-induced vibration equipment in the candidate time period, the vortex-induced vibration data is input into the vortex-induced vibration deep learning model to obtain the predicted vortex-induced vibration data of the vortex-induced vibration equipment in the next time period, and then the predicted vortex-induced vibration data and the target vortex-induced vibration data expected to be achieved by the vortex-induced vibration equipment are used to determine the control instructions to adjust the vortex-induced vibration equipment.

[0074] The trained vortex-induced vibration deep learning model is applied to the system-level control closed loop to realize the integration of the 3D (three-dimensional) complex fluid-solid coupling model in the 1D (one-dimensional) system-level simulation, wherein the 1D system-level simulation can include the vortex-induced vibration deep learning model and the vortex-induced vibration experimental equipment. Through the interaction between the vortex-induced vibration deep learning model and the vortex-induced vibration experimental equipment, the adjustment and verification of the vortex-induced vibration experimental equipment can be realized, thereby effectively realizing the closed loop of models of different fineness in one simulation process, which not only ensures the solution speed of the model, but also retains the high-order accuracy of some components. In addition, the vortex-induced vibration deep learning model trained using the historical vortex-induced vibration data of the vortex-induced vibration equipment can better fit the vortex-induced vibration equipment. In this way, the vortex-induced vibration data of the vortex-induced vibration equipment at the next moment can be better predicted by using the vortex-induced vibration deep learning model, that is, the predicted vortex-induced vibration data of the vortex-induced vibration equipment in the next period is more accurate, and then the predicted vortex-induced vibration data can be used to better adjust the vortex-induced vibration equipment.

[0075] In an optional embodiment, S104 may include:

[0076] The predicted vortex-induced vibration data and the target vortex-induced vibration data are compared; when the predicted vortex-induced vibration data is smaller than the target vortex-induced vibration data, a control instruction for increasing the vibration amplitude of the vortex-induced vibration equipment is determined; when the predicted vortex-induced vibration data is larger than the target vortex-induced vibration data, a control instruction for reducing the vibration amplitude of the vortex-induced vibration equipment is determined.

[0077] The situation where the predicted vortex-induced vibration data is smaller than the target vortex-induced vibration data can also be understood as the vortex-induced vibration data of the vortex-induced vibration equipment in the next time period predicted by the vortex-induced vibration deep learning model is smaller than the target vortex-induced vibration data. In order to make the actual vortex-induced vibration data of the vortex-induced vibration equipment in the next time period reach the target vortex-induced vibration data, the control instructions for increasing the vibration amplitude of the vortex-induced vibration equipment are determined. In this way, the adjustment of the vortex-induced vibration equipment can be achieved through the control instructions, so that the actual vortex-induced vibration data of the vortex-induced vibration equipment in the next time period can reach the target vortex-induced vibration data as much as possible.

[0078] Similarly, the situation where the predicted vortex-induced vibration data is greater than the target vortex-induced vibration data can also be understood as the vortex-induced vibration data of the vortex-induced vibration equipment in the next time period predicted by the vortex-induced vibration deep learning model is greater than the target vortex-induced vibration data. In order to make the actual vortex-induced vibration data of the vortex-induced vibration equipment in the next time period reach the target vortex-induced vibration data, the control instructions for reducing the vibration amplitude of the vortex-induced vibration equipment are determined. In this way, the adjustment of the vortex-induced vibration equipment can be achieved through the control instructions, so that the actual vortex-induced vibration data of the vortex-induced vibration equipment in the next time period can reach the target vortex-induced vibration data as much as possible.

[0079] In the embodiment of the present disclosure, by comparing the predicted vortex-induced vibration data with the target vortex-induced vibration data, a control instruction for increasing or decreasing the vibration amplitude of the vortex-induced vibration equipment is determined, so that the actual vortex-induced vibration data of the vortex-induced vibration equipment in the next time period is infinitely close to the target vortex-induced vibration data, thereby achieving accurate adjustment of the vortex-induced vibration equipment.

[0080] In an optional embodiment, the target vortex-induced vibration data includes vibration amplitude; the control instruction includes a control instruction for a motor, and the motor is a motor connected to the vortex-induced vibration device.

[0081] S105 may include:

[0082] Send control instructions to the motor.

[0083] The control instructions are used by the motor to execute the control instructions to adjust the length of the test piece in the vortex-induced vibration equipment. The length affects the vibration amplitude of the test piece in the vortex-induced vibration equipment.

[0084] The test piece may include a vibrating structure and a bracket, wherein the bracket is a structure of variable length. The bracket is connected to the vibrating structure, and the bracket's length affects the vibration amplitude of the vibrating structure. For example, the bracket's length may change the stiffness and / or natural frequency of the vibrating structure, thereby changing the vibration amplitude of the vibrating structure.

[0085] To put it simply, control instructions are sent to the motor, and the motor executes the control instructions to drive changes in the vortex-induced vibration equipment, such as driving the length of the test piece in the vortex-induced vibration equipment to become longer or shorter, thereby driving changes in the vibration amplitude of the vortex-induced vibration equipment.

[0086] Control instructions can also be understood as control strategies and control algorithms.

[0087] In an environment where the vortex-induced vibration deep learning model is integrated with vortex-induced vibration equipment, such as vortex-induced vibration test equipment, a control algorithm is added to achieve online control of the target, that is, to achieve online control of the vortex-induced vibration experimental equipment.

[0088] In the embodiment of the present disclosure, the adjustment of the vortex-induced vibration device is achieved through the combination of the vortex-induced vibration deep learning model, the motor and the vortex-induced vibration device. In addition, the vortex-induced vibration device can be more conveniently controlled through the motor, and the vortex-induced vibration device can be conveniently adjusted online.

[0089] In an optional embodiment, S104 may include:

[0090] Based on the predicted vortex-induced vibration data and the target vortex-induced vibration data, the current control quantity is determined; and through digital-to-analog conversion, the current control quantity is converted into an analog current control instruction.

[0091] S105 may include:

[0092] Send analog current control instructions to the motor.

[0093] The current control quantity can be understood as a digital signal. In order to adjust the physical vortex-induced vibration equipment, the digital signal can be converted into an analog signal through digital-to-analog conversion.

[0094] In one implementation, the current control amount may include data for directly controlling the current, or may also include data for adjusting the motor drive duty cycle to control the current.

[0095] The analog current control instruction is sent to the motor, and the motor executes the analog current control instruction to adjust the length of the test piece in the vortex-induced vibration equipment. The length affects the vibration amplitude of the test piece in the vortex-induced vibration equipment. In this way, the vibration amplitude of the test piece in the vortex-induced vibration experimental equipment can be adjusted.

[0096] The integration verification of deep learning models and vortex-induced vibration equipment such as vortex-induced vibration test equipment is carried out. Based on the vortex-induced vibration (VIV) deep learning model, a control strategy is established, and the length of the test piece in the vortex-induced vibration test equipment is changed by adjusting the current of the motor, thereby changing the stiffness and natural frequency of the vibration structure in the vortex-induced vibration test equipment, and then changing the vibration amplitude of the vibration structure.

[0097] In the embodiment of the present disclosure, the motor can be conveniently controlled by analog current control instructions, thereby driving the changes in the vortex-induced vibration device so that the vortex-induced vibration data of the vortex-induced vibration device in the next time period reaches the target vortex-induced vibration data as much as possible; in addition, the current control amount is determined based on the predicted vortex-induced vibration data and the target vortex-induced vibration data; and then the current control amount is converted into an analog current control instruction through digital-to-analog conversion. Through the combination of virtual and real, the output of the simulated deep learning model can be combined with the physical vortex-induced vibration device to adjust the vortex-induced vibration device.

[0098] Figure 2 It is a schematic diagram of adjusting the vortex-induced vibration device in the embodiment of the present disclosure. The vortex-induced vibration device can be a vortex-induced vibration experimental device, and the data of the vortex-induced vibration experimental device is collected to obtain the collected data, such as the vortex-induced vibration data of the above-mentioned vortex-induced vibration device in the candidate time period. In order to facilitate data processing, analog-to-digital conversion (AD) can be performed first, and the data after analog-to-digital conversion is input into the vortex-induced vibration deep learning model to obtain the predicted vortex-induced vibration data of the vortex-induced vibration device in the next time period, and obtain the target vortex-induced vibration data that the vortex-induced vibration device is expected to reach in the next time period. In this way, based on the predicted vortex-induced vibration data and the target vortex-induced vibration data, the control instruction can be determined to achieve current regulation, such as determining the current control amount based on the predicted vortex-induced vibration data and the target vortex-induced vibration data; through digital-to-analog conversion, the current control amount is converted into an analog current control instruction, and the analog current control instruction is sent to the motor, and the motor executes the analog current control instruction to drive the change of the vortex-induced vibration experimental device. Specifically, the motor executes analog current control instructions to adjust the length of the test piece in the vortex-induced vibration equipment. The length affects the vibration amplitude of the test piece in the vortex-induced vibration equipment. For example, the length of the bracket will change the stiffness and / or natural frequency of the vibration structure, thereby changing the vibration amplitude of the vibration structure.

[0099] The disclosed embodiments establish an interactive relationship between the control algorithm, deep learning model, and vortex-induced vibration test equipment, and adjust the vibration amplitude of the vortex-induced vibration test equipment based on the control algorithm, thereby achieving virtual-real integrated verification and online adjustment of the deep learning model and test equipment. The deep learning model is integrated with the real test equipment through internal mechanisms such as AD (analog-to-digital conversion) / DA (digital-to-analog conversion) and model calls. In this integrated environment, the control algorithm is added to achieve online control of the target.

[0100] After completing the training of the deep learning model, it can be compiled to support the call of the external system-level 1D (one-dimensional) model, such as compiling to generate exe (executable files), dll (dynamic link library), etc. At the same time, when defining the system control algorithm, such as PID adjustment, and combining the compiled file of the deep learning model with the control algorithm, this process requires the use of data acquisition and conversion equipment. That is, after completing the training of the vortex-induced vibration deep learning model, the vortex-induced vibration deep learning model can be called by compiling to generate exe, dll and other files, and the vortex-induced vibration data can be input into the vortex-induced vibration deep learning model to obtain the predicted vortex-induced vibration data of the vortex-induced vibration equipment in the next period; and based on the predicted vortex-induced vibration data and the target vortex-induced vibration data, the control instructions are determined; the vortex-induced vibration equipment is adjusted using the control instructions, and the vortex-induced vibration deep learning model is combined with the real vortex-induced vibration equipment. Through the combination of virtual and real, the vortex-induced vibration equipment can be adjusted, and then the vortex-induced vibration problem can be analyzed and solved in combination with the vortex-induced vibration equipment.

[0101] Compared with the traditional analysis of vortex-induced vibration and other problems, which usually uses professional tools of computational fluid dynamics (CFD), although it can ensure accuracy, the calculation speed is very slow. The physical information-based neural network (PINNs) method in the embodiment of the present disclosure highly approximates the solution process of complex fluid problems through neural networks under the premise of following physical equations. For the trained vortex-induced vibration deep learning model, it has fast reasoning capabilities. For example, the reasoning speed of the neural network for the next cycle after stabilization of the flow around a cylinder is about milliseconds, which far exceeds the solution speed of traditional CFD tools. Based on the characteristics of "order reduction" and fast reasoning of this model, the embodiment of the present disclosure combines the vortex-induced vibration deep learning model and integrates it with experimental verification, utilizes a neural network with physical information, and interacts with the test equipment to support online adjustment of the real equipment. Compared with pure black box data fitting based on data, the embodiment of the present disclosure uses a vortex-induced vibration deep learning model that relies on less data and retains the constraints of physical information. The constructed neural network (i.e., the vortex-induced vibration deep learning model) is interpretable. The disclosed embodiment uses vortex-induced vibration test equipment and the corresponding VIV deep learning model for integrated verification, which can be extended to various engineering problems in the industry, such as online prediction of structural damage, online monitoring of complex system performance and health management. It can provide a relatively efficient and feasible technical approach for the goals of "model reduction" and "digital twin" in the currently promoted "digital simulation".

[0102] The disclosed embodiments enable rapid inference through the vortex-induced vibration deep learning model, thereby increasing the speed of vortex-induced vibration analysis. Furthermore, in the disclosed embodiments, the accuracy of the predicted vortex-induced vibrations can be improved by using the trained vortex-induced vibration deep learning model to determine the vortex-induced vibration data of the vortex-induced vibration experimental equipment in the next time period. The vortex-induced vibration deep learning model is trained using the vortex-induced vibration data of the vortex-induced vibration experimental equipment, and can also ensure the accuracy of the vortex-induced vibration analysis, thereby increasing the calculation speed while ensuring accuracy.

[0103] In addition, in the related art, it is considered that the integration interface matching degree between 1D system-level tools and 3D component-level tools is not high, and the integration of 1D system-level models and 3D component-level models seriously reduces the calculation speed of the original 1D model. Therefore, it is impossible to integrate 3D models with real equipment to realize mechanism-based online monitoring in the related art. In the embodiment of the present disclosure, the vortex-induced vibration experimental equipment is adjusted based on the predicted vortex-induced vibration data obtained above, and the vortex-induced vibration experimental equipment can be accurately adjusted so that the vortex-induced vibration data of the vortex-induced vibration experimental equipment in the next time period can reach the target vortex-induced vibration data as much as possible, thereby improving the speed of analysis and processing while ensuring the accuracy of the analysis of vortex-induced vibration problems in the next time period. Compared with the integration means of CFD simulation and system-level 1D simulation in the related art, the method of analyzing vortex-induced vibration problems using a vortex-induced vibration deep learning model in the embodiment of the present disclosure can effectively realize the closed loop of models of different fineness in one simulation process, which not only ensures the solution speed of the model, but also retains the high-order accuracy of some components.

[0104] The disclosed embodiments utilize the rapid reasoning capabilities of deep learning models and, with the help of virtual-real interfaces such as digital-to-analog conversion equipment, can rapidly implement the integrated verification of vortex-induced vibration deep learning models and vortex-induced vibration test equipment, and, in combination with control algorithms, support online adjustment of test pieces.

[0105] The online monitoring based on monitoring data in related technologies is more of a "black box" model based on data, which does not fully consider the physical mechanism of the components. In the embodiment of the present disclosure, a vortex-induced vibration deep learning model is used to predict the vortex-induced vibration data of the vortex-induced vibration in the next time period, and the obtained predicted vortex-induced vibration data is used to adjust the vortex-induced vibration experimental equipment. The vortex-induced vibration deep learning model is based on the historical vortex-induced vibration data of the vortex-induced vibration experimental equipment, and is obtained through training of flow field control equations and structural vibration control equations, so that the vortex-induced vibration deep learning model fully considers the physical mechanism of the vortex-induced vibration experimental equipment, thereby improving the accuracy of the vortex-induced vibration deep learning model in processing the vortex-induced vibration experimental equipment.

[0106] The present disclosure also provides a method for training a vortex-induced vibration deep learning model, which may include:

[0107] Acquire a plurality of sample vortex-induced vibration data, where the sample vortex-induced vibration data is collected from a vortex-induced vibration device;

[0108] Using multiple sample vortex-induced vibration data, the vortex-induced vibration deep learning model is trained through the flow field control equation and the structural vibration control equation, where the structural vibration control equation is related to the flow field control equation.

[0109] In the embodiment of the present disclosure, a vortex-induced vibration deep learning model is obtained by training multiple sample vortex-induced vibration data, and the deep learning model can be used to predict the vortex-induced vibration data of the vortex-induced vibration equipment at subsequent times. Based on the predicted vortex-induced vibration data and the target vortex-induced vibration data, control instructions can be determined, and the vortex-induced vibration equipment can be adjusted using the control instructions.

[0110] The flow field control equations and the structural vibration control equations can also be understood as constraint equations for vortex-induced vibrations. The trained vortex-induced vibration deep learning model meets the physical requirements of vortex-induced vibrations. The vortex-induced vibration deep learning model training method in the embodiment of the present disclosure can be understood as a physical information-based neural network (PINNs). Under the premise of following the physical equations, the embodiment of the present disclosure highly approximates the solution process of complex fluid problems through the deep learning model.

[0111] In addition, the structural vibration control equation is related to the flow field control equation. In the process of training the vortex-induced vibration deep learning model by combining the flow field control equation and the structural vibration control equation related to the process control equation, the structural vibration control equation and the flow field control equation are combined with each other to jointly constrain the vortex-induced vibration learning model, thereby improving the accuracy of the trained vortex-induced vibration deep learning model.

[0112] The embodiments of the present disclosure can realize the processes of experimental data collection-deep learning model training-testing and model integration verification.

[0113] For experimental data collection, for example, a vortex-induced vibration test device is constructed, including a small test wind tunnel and test piece, a uniform flow device, and data collection equipment (including various sensors). The uniform flow provides different flow velocities, which, combined with the inherent characteristics of the test piece, produce different reduced velocities. Various sensors and other equipment deployed on the test piece collect and uniformly output information such as the test piece's vibration displacement and force for training deep learning models.

[0114] After collecting the data, that is, obtaining multiple sample vortex-induced vibration data, the vortex-induced vibration deep learning model is trained using the multiple sample vortex-induced vibration data through the flow field control equation and the structural vibration control equation, where the structural vibration control equation is related to the flow field control equation.

[0115] After obtaining the vortex-induced vibration deep learning model, the vortex-induced vibration data of the vortex-induced vibration equipment in the candidate time period can be obtained; the vortex-induced vibration data is input into the vortex-induced vibration deep learning model to obtain the predicted vortex-induced vibration data of the vortex-induced vibration equipment in the next time period; the target vortex-induced vibration data expected to be achieved by the vortex-induced vibration equipment in the next time period is obtained; based on the predicted vortex-induced vibration data and the target vortex-induced vibration data, the control instructions are determined; and the vortex-induced vibration equipment is adjusted using the control instructions. Specifically, for the integrated verification of the deep learning model and the test equipment, a control strategy is established based on the VIV deep learning model, and the effective length of the test piece in the test equipment is changed by adjusting the motor current, thereby changing the stiffness and natural frequency of the structure, and thus changing the vibration amplitude of the structure.

[0116] Figure 3 This is a flowchart of the vortex-induced vibration deep learning model training method provided by the embodiment of the present disclosure, referring to Figure 3 The vortex-induced vibration deep learning model training method provided by the embodiment of the present disclosure may include:

[0117] S301, obtaining a plurality of sample vortex-induced vibration data.

[0118] The sample vortex-induced vibration data is collected for the vortex-induced vibration equipment.

[0119] The sample vortex-induced vibration data collected from the vortex-induced vibration device, that is, the historical vortex-induced vibration data of the vortex-induced vibration device is collected as the sample vortex-induced vibration data. The historical vortex-induced vibration data may include the vortex-induced vibration data corresponding to the time before the candidate time period.

[0120] S302 , using multiple sample vortex-induced vibration data, through the flow field control equation and the structural vibration control equation, train the vortex-induced vibration deep learning model.

[0121] Among them, the structural vibration control equation is related to the flow field control equation.

[0122] The flow field control equation and the structural vibration control equation can be used as constraints, so that the trained vortex-induced vibration learning model is subject to the constraints of the flow field control equation and the structural vibration control equation. It can also be understood as combining the flow field control equation and the structural vibration control equation in the process of calculating the loss. For example, the flow field control equation and the structural vibration control equation can be used as loss functions, or the flow field control equation and the structural vibration control equation can be added on the basis of the preset loss function, that is, the preset loss function, the flow field control equation and the structural vibration control equation are used as the loss function.

[0123] In an optional embodiment, the vortex-induced vibration device includes a test wind tunnel, a test piece, a uniform flow device, and a collection device; the test wind tunnel is used to generate a uniform airflow; the uniform flow device is used to use the airflow to provide different flow velocities, and the flow velocities are used to act on the test piece to generate a reduced velocity; the collection device is used to collect vibration amplitude and flow field information of the vortex-induced vibration device;

[0124] Acquire multiple sample vortex-induced vibration data, including:

[0125] Obtain the vibration amplitude and flow field information of the vortex-induced vibration equipment at different times under different reduced velocities.

[0126] The flow field information may include vibration velocity and pressure.

[0127] A vortex-induced vibration test device can be built, including a test wind tunnel, a test piece, a uniform flow device, and a collection device.

[0128] The acquisition equipment may include laser equipment, sensors, etc. The vibration amplitude δ of the vortex-induced vibration experimental device can be acquired by laser acquisition, the vibration velocity of the vortex-induced vibration experimental device can be acquired by a flow velocity sensor, and the pressure P can be acquired by a pressure sensor, wherein the vibration velocity may include a horizontal velocity u and a vertical velocity v.

[0129] like Figure 4 As shown, the vortex-induced vibration experimental equipment includes a cylinder (01), a flexible scale bracket (02), a rigid bracket (03), a mounting rail (04), and a wind tunnel (05), which can also be understood as an experimental wind tunnel, a strain gauge (06), and a laser acquisition (07). The cylinder (01), the flexible scale bracket (02), and the rigid bracket (03) can be understood as the test piece of the vortex-induced vibration experimental equipment in the above embodiment, the cylinder (01) is the vibration structure, and the flexible scale bracket (02) and the rigid bracket (03) are the brackets. For example, the vortex-induced vibration experimental equipment can be aligned by a spirit level, and the vortex-induced vibration experimental equipment can be placed horizontally. The top view of the vortex-induced vibration experimental equipment is as shown in FIG. Figure 4 As shown, the vortex-induced vibration experimental equipment can also be viewed from a side view.

[0130] Among them, the length of the flexible scale bracket (02) can be adjusted. In the embodiment of the present disclosure, the length of the flexible scale bracket (02) can be adjusted to affect the vortex-induced vibration data of the vortex-induced vibration experimental equipment. For example, the length of the flexible scale bracket (02) will change the stiffness and / or natural frequency of the vibration structure, thereby changing the vibration amplitude of the vibration structure.

[0131] In the embodiment of the present disclosure, after the vortex-induced vibration data collected by the vortex-induced vibration experimental equipment is used to train the vortex-induced vibration deep learning model, the vortex-induced vibration deep learning model can be used to verify the vortex-induced vibration experimental equipment. Specifically, the above-mentioned Figure 1 In the embodiment shown, after obtaining the vortex-induced vibration data of the vortex-induced vibration device in the candidate time period, for example, the flow field information and vibration amplitude of the flexible scale bracket (02) at different lengths, different moments or different angles at different times (Time) are obtained, the vortex-induced vibration data are input into the vortex-induced vibration deep learning model, that is, input into the physical information-based neural network (PINNs), and the predicted vortex-induced vibration data of the vortex-induced vibration device in the next time period are obtained. Then, the predicted vortex-induced vibration data and the target vortex-induced vibration data expected to be achieved by the vortex-induced vibration device are used to determine the control instructions to adjust the vortex-induced vibration device. For example, the length of the flexible scale bracket (02) can be adjusted by an accelerometer, and the current control amount can be determined based on the predicted vortex-induced vibration data and the target vortex-induced vibration data; and the current control amount is converted into an analog current control instruction through digital-to-analog conversion.

[0132] Among them, because the length of the bracket will change the stiffness and / or natural frequency of the vibration structure, and thus will change the vibration amplitude of the vibration structure, therefore, in the process of determining the control instruction, the structural equivalent stiffness (EI) can be generalized to achieve amplitude adjustment guidance through experimental oscillation amplitude comparison (model vs experiment) to adjust the flexible scale bracket (02). Specifically, the difference between the predicted vortex-induced vibration data and the target vortex-induced vibration data can be determined, and based on the difference, the stiffness and / or natural frequency of the vibration structure that can eliminate the difference is determined, and then the stiffness and / or natural frequency of the vibration structure that can eliminate the difference is corresponded to the experimental oscillation amplitude comparison to be adjusted, and the experimental oscillation amplitude comparison to be adjusted is corresponded to the length of the flexible scale bracket (02) to be adjusted, and then the control instruction can be determined based on the length of the flexible scale bracket (02) to be adjusted. For example, a control instruction for adjusting the length of the test piece in the vortex-induced vibration device is executed by a motor connected to the vortex-induced vibration experiment.

[0133] Vortex-induced vibration (VIV) testing equipment can provide different flow velocities through a uniform flow, which, combined with the inherent characteristics of the test piece, produces different reduced velocities. Various sensors and other data acquisition devices deployed on the test piece collect and output information such as the test piece's vibration amplitude and applied force (such as pressure). This information can be used to train a VIV deep learning model.

[0134] The flow source equipment inside the experimental wind tunnel can generate a uniform airflow (such as 1m / s-3m / s) in the test section, thereby simulating different reduced velocities corresponding to the test piece. r The calculation formula is:

[0135]

[0136] Where u is the uniform flow velocity of the test piece in the test section, D is the characteristic length of the test piece, that is, the diameter, and f n is the natural frequency of the specimen, which is related to the stiffness and mass of the specimen.

[0137] In the embodiment of the present disclosure, sample vortex-induced vibration data can be easily and accurately obtained through the vortex-induced vibration equipment to train the vortex-induced vibration deep learning model.

[0138] In an optional embodiment, S302 may include:

[0139] For each sample vortex-induced vibration data, the sample vortex-induced vibration data is input into the vortex-induced vibration deep learning model to obtain the vortex-induced vibration prediction value corresponding to the sample vortex-induced vibration data; based on the vortex-induced vibration prediction value, the loss value is calculated through the flow field control equation and the structural vibration control equation; based on the loss value, the model parameters of the vortex-induced vibration deep learning model are adjusted until the preset training conditions are reached, and the trained vortex-induced vibration deep learning model is obtained.

[0140] The preset training conditions may include a loss value not being greater than a preset loss value, or may include a loss value converging to a value, or the number of training times reaching a preset number, and the preset number of times may be determined according to actual needs.

[0141] The sample vortex-induced vibration data is input into the vortex-induced vibration deep learning model to obtain the vortex-induced vibration prediction value corresponding to the sample vortex-induced vibration data. A loss value can be calculated based on the field control equation and the structural vibration control equation respectively, and then the loss values ​​calculated based on the field control equation and the structural vibration control equation are summed. The sum of the loss values ​​obtained is the loss value calculated by the flow field control equation and the structural vibration control equation. Then, the model parameters of the vortex-induced vibration deep learning model are adjusted based on the loss value until the preset training conditions are reached to obtain the trained vortex-induced vibration deep learning model.

[0142] Among them, based on the loss value, the model parameters of the vortex-induced vibration deep learning model can be adjusted using the adaptive moment estimation method (ADAM) gradient descent optimization algorithm.

[0143] In one possible implementation, after obtaining the loss value, the loss value can be compared with a preset loss value. If the loss value is not greater than the preset loss value, the preset training condition is reached, and the trained vortex-induced vibration deep learning model is obtained.

[0144] The flow field control equations and the structural vibration control equations can be understood as constraint equations for constraining the vortex-induced vibration deep learning model. By performing physical information constraints on the vortex-induced vibration depth from the perspective of flow field control and structural vibration control, the vortex-induced vibration deep learning model is trained using multiple sample vortex-induced vibration data and the flow field control equations and structural vibration control equations until the preset training conditions are reached, and the trained vortex-induced vibration deep learning model is obtained, which can improve the accuracy of the trained vortex-induced vibration deep learning model.

[0145] In an optional embodiment, the sample vortex-induced vibration data includes the vibration amplitude and flow field information of the test piece in the vortex-induced vibration experimental equipment, and the flow field information may include vibration velocity and pressure.

[0146] In the disclosed embodiment, the vortex-induced vibration deep learning model may include two networks: a flow field information neural network and a structural vibration neural network. For each sample vortex-induced vibration data, the flow field information in the sample vortex-induced vibration data may be input into the flow field information neural network, and the vibration amplitude may be input into the structural vibration neural network. The vortex-induced vibration prediction value corresponding to the sample vortex-induced vibration data includes the predicted flow field information obtained by the flow field information neural network and the vibration amplitude obtained by the structural vibration neural network. Based on the vortex-induced vibration prediction value, the loss value is calculated using the flow field control equation and the structural vibration control equation. Based on the loss value, the model parameters of the vortex-induced vibration deep learning model are adjusted until the preset training conditions are reached, thereby obtaining a trained vortex-induced vibration deep learning model.

[0147] like Figure 5As shown, for each sample vortex-induced vibration data, the flow field information in the sample vortex-induced vibration data, such as vibration velocity (including horizontal velocity and vertical velocity) and pressure, is input into the flow field control neural network, and the vibration amplitude is input into the structural vibration neural network to obtain the vortex-induced vibration prediction value corresponding to the sample vortex-induced vibration data. The loss value is calculated based on the vortex-induced vibration prediction value through the first constraint equation, the second constraint equation, the third constraint equation, and the fourth constraint equation; the model parameters of the vortex-induced vibration deep learning model are adjusted based on the loss value until the preset training condition is reached to obtain the trained vortex-induced vibration deep learning model. For example, after obtaining the loss value, the loss value can be compared with the preset loss value ε. If the loss value is greater than the preset loss value ε, the parameters of the vortex-induced vibration deep learning model are adjusted. Only the flow field control neural network can be adjusted, or only the structural vibration neural network can be adjusted, or both the flow field control neural network and the structural vibration neural network can be adjusted; if the loss value is not greater than the preset loss value, the preset training condition is reached, and the trained vortex-induced vibration deep learning model is obtained.

[0148] In an optional embodiment, the flow field control equation includes a first constraint equation, a second constraint equation, and a third constraint equation, and the structural vibration control equation includes a fourth constraint equation, and the fourth constraint equation is related to the first constraint equation, the second constraint equation, and the third constraint equation;

[0149] The first constraint equation is: t +u*u x +v*u y +P x -1 / R e *(u xx +u yy );

[0150] The second constraint equation is: t +u*v x +v*v y +P y -1 / R e *(v xx +v yy );

[0151] The third constraint equation is: x +v y ;

[0152] The fourth constraint equation is:

[0153]

[0154] Among them, u is the horizontal velocity, v is the vertical velocity, u x is the partial derivative of u in the x direction, u y is the partial derivative of u in the y direction, vx is the partial derivative of v in the x direction, v y is the partial derivative of v in the y direction, u t is the time derivative of the horizontal velocity, v t is the time derivative of vertical velocity, u xx is the second-order derivative of u in the x direction, u yy is the second-order partial derivative of u in the y direction, v xx is the second-order derivative of v in the x direction, v yy is the second-order derivative of v in the y direction, P x is the partial derivative of pressure P in the x direction, P y is the partial derivative of pressure in the y direction, R e is the Reynolds number, a dimensionless number that characterizes the inertia and viscosity of the fluid, m is the mass of the test piece, k is the equivalent damping of the test piece, c is the stiffness of the test piece, δ is the amplitude of the vibration of the test piece, is the vibration speed of the test piece, is the acceleration of the test piece vibration, D is the length of the test piece, and θ is the rotation angle of the cylinder in the test piece.

[0155] The vortex-induced vibration test equipment can be used in vortex-induced vibration experiments to analyze vortex-induced vibration problems. The reason why the test piece in the vortex-induced vibration test equipment vibrates is because it is subjected to a force perpendicular to the incoming flow direction, which is expressed as lift F. L , lift is related to the static pressure on the surface of the test piece and the flow field velocity. According to experience, the lift formula is expressed as:

[0156]

[0157] Where P is the static pressure at a point on the cylindrical surface, n x , n y is the y-direction projection of the normal vector at each point on the cylindrical surface, v x , v y is the partial derivative of the vertical velocity v at each point on the cylindrical surface with respect to x and y, u y is the partial derivative of the horizontal velocity u at each point on the cylindrical surface with respect to y, is the Reynolds number derivative, and ds is the arc length of the cylinder circumference.

[0158]

[0159] In the Reynolds number equation, u is the uniform incoming flow velocity of the test piece in the test section, D is the characteristic length of the test piece, and μ is the fluid viscosity.

[0160] By discretizing the above lift equation into angles, we can obtain the following lift equation:

[0161]

[0162] Structural vibration control equations:

[0163]

[0164] Where m is the mass of the specimen, k is the equivalent damping of the specimen, c is the stiffness of the specimen, and δ is the amplitude of the specimen vibration. is the vibration speed of the test piece, is the acceleration of the test piece vibration.

[0165] Flow field governing equations:

[0166]

[0167] By integrating the lift, structural vibration control equations and flow field control equations, we can obtain the unified vortex-induced vibration control equations, including:

[0168] The first constraint equation is: t +u*u x +v*u y +P x -1 / R e *(u xx +u yy );

[0169] The second constraint equation is: t +u*v x +v*v y +P y -1 / R e *(v xx +v yy );

[0170] The third constraint equation is: x +v y ;

[0171] The fourth constraint equation is:

[0172]

[0173] The Vortex-Induced Vibration (VIV) deep learning model can include two neural networks: a flow field control neural network and a structural vibration neural network. The flow field control neural network is used to predict flow field information, while the structural vibration neural network is used to predict structural vibration information, such as vibration amplitude. Furthermore, based on the automatic differentiation mechanism, the flow field partial derivative (PDE) equations, as well as equations for structural vibration and structural forces, are jointly constructed to form the comprehensive control equations of the VIV deep learning model. The deep learning loss function and network parameter optimization process are combined to establish the deep learning model, thus training the VIV learning model.

[0174] Through the first constraint equation, the second constraint equation, the third constraint equation and the fourth constraint equation, the vortex-induced vibration deep learning model is simulated more accurately from the perspective of flow field and structural vibration, thereby improving the accuracy of the vortex-induced vibration deep learning model in processing vortex-induced vibration.

[0175] In addition, the vortex-induced vibration deep learning model can be constrained by dynamic boundaries.

[0176] The dynamic boundary of the test piece motion in vortex-induced vibration can be expressed as:

[0177] (Motion Boundary):

[0178] Based on the essence of vortex-induced vibration (VIV) fluid-structure coupling, the fluid-structure coupling control equations are refined, integrating equations for lift, drag, structural vibration, flow field calculation, and the boundaries of structural dynamic motion into a single model. A flow field control neural network and a structural vibration neural network are established, and the control equations are comprehensively described using automatic differentiation. During the training process of the VIV deep learning model, the unified control equations and the motion boundaries represented by the equations are combined with a "semi-supervised training" approach to construct a VIV deep learning model. This model can simulate the flow field information and physical information of the test piece during the training time, and can also predict information at later times.

[0179] After the deep learning model is trained, it can be compiled to support the call of external system-level 1D models, such as compiling to generate exe, dll, and other files. At the same time, when defining the system's control algorithm, such as PID adjustment, the compiled file of the deep learning model is combined with the control algorithm. This process requires the use of data acquisition and conversion equipment. That is, after the vortex-induced vibration deep learning model is trained, it can be called by compiling to generate exe, dll, and other files. The vortex-induced vibration data is input into the vortex-induced vibration deep learning model to obtain the predicted vortex-induced vibration data of the vortex-induced vibration equipment in the next period; and based on the predicted vortex-induced vibration data and the target vortex-induced vibration data, the control instructions are determined; the vortex-induced vibration equipment is adjusted using the control instructions, and the vortex-induced vibration deep learning model is combined with the real vortex-induced vibration equipment. Through the combination of virtual and real, the vortex-induced vibration equipment can be adjusted, and then the vortex-induced vibration problem can be analyzed and solved in combination with the vortex-induced vibration equipment.

[0180] The disclosed embodiment proposes a comprehensive control equation of the fluid control equation, the structural lift and drag equations, and the structural force equations in vortex-induced vibration, and at the same time expresses the dynamic boundary equation of the structural vibration. In addition, a flow field control neural network and a structural vibration neural network are constructed, and the outputs of the two networks are comprehensively utilized to realize the construction of the comprehensive control equation through automatic differentiation, and the adam (Adaptive Moment Estimation) gradient descent optimization algorithm is used to realize the establishment of the VIV deep learning model. The training data of the VIV deep learning model comes from the test equipment, and after completing multiple reduced speed trainings, the system-level integration and online debugging of the VIV deep learning model and the test equipment are realized by determining the control algorithm. This breaks through the difficulties of traditional 3D models in realizing system-level integration verification and online debugging of virtual and real combinations.

[0181] Corresponding to the vortex-induced vibration regulation method provided in the above embodiment, the embodiment of the present disclosure also provides a vortex-induced vibration regulation device, such as Figure 6 As shown, the vortex-induced vibration regulating device may include:

[0182] The first acquisition module 601 is used to acquire vortex-induced vibration data of the vortex-induced vibration equipment in the candidate time period;

[0183] The data acquisition module 602 is used to input the vortex-induced vibration data into the vortex-induced vibration deep learning model to obtain the predicted vortex-induced vibration data of the vortex-induced vibration device in the next time period;

[0184] The second acquisition module 603 is used to obtain target vortex-induced vibration data that the vortex-induced vibration device is expected to achieve in the next period;

[0185] A determination module 604 is configured to determine a control instruction based on the predicted vortex-induced vibration data and the target vortex-induced vibration data;

[0186] The adjustment module 605 is used to adjust the vortex-induced vibration equipment using control instructions.

[0187] Optionally, the target vortex-induced vibration data includes a vibration amplitude; the control instruction includes a control instruction for a motor, and the motor is a motor connected to the vortex-induced vibration device;

[0188] The adjustment module 605 is specifically used to send control instructions to the motor, and the control instructions are used for the motor to execute the control instructions to adjust the length of the test piece in the vortex-induced vibration equipment. The length affects the vibration amplitude of the test piece in the vortex-induced vibration equipment.

[0189] Optionally, the determination module 604 is specifically configured to determine a current control variable based on the predicted vortex-induced vibration data and the target vortex-induced vibration data; and convert the current control variable into an analog current control instruction through digital-to-analog conversion;

[0190] The regulating module 605 is specifically configured to send analog current control instructions to the motor.

[0191] Optionally, the determination module 604 is specifically used to compare the predicted vortex-induced vibration data with the target vortex-induced vibration data; when the predicted vortex-induced vibration data is smaller than the target vortex-induced vibration data, determine the control instruction for increasing the vibration amplitude of the vortex-induced vibration equipment; when the predicted vortex-induced vibration data is larger than the target vortex-induced vibration data, determine the control instruction for reducing the vibration amplitude of the vortex-induced vibration equipment.

[0192] Optionally, the vortex-induced vibration deep learning model is trained using historical vortex-induced vibration data of vortex-induced vibration equipment.

[0193] Optionally, the data acquisition module 602 is specifically used to input the vortex-induced vibration data into a vortex-induced vibration deep learning model, obtain flow field information through the flow field information neural network, and obtain vibration amplitude through the structural vibration neural network.

[0194] Corresponding to the vortex-induced vibration deep learning model training method provided in the above embodiment, the embodiment of the present disclosure also provides a vortex-induced vibration deep learning model training device, such as Figure 7 As shown, this may include:

[0195] An acquisition module 701 is used to acquire a plurality of sample vortex-induced vibration data, where the sample vortex-induced vibration data is collected using a vortex-induced vibration device;

[0196] The training module 702 is used to train a vortex-induced vibration deep learning model using multiple sample vortex-induced vibration data through flow field control equations and structural vibration control equations, wherein the structural vibration control equations are related to the flow field control equations.

[0197] Optionally, the training module 702 is specifically used to input the sample vortex-induced vibration data into the vortex-induced vibration deep learning model for each sample vortex-induced vibration data to obtain a vortex-induced vibration prediction value corresponding to the sample vortex-induced vibration data; based on the vortex-induced vibration prediction value, calculate the loss value through the flow field control equation and the structural vibration control equation; based on the loss value, adjust the model parameters of the vortex-induced vibration deep learning model until the preset training conditions are reached, and obtain the trained vortex-induced vibration deep learning model.

[0198] Optionally, the flow field control equation includes a first constraint equation, a second constraint equation, and a third constraint equation, and the structural vibration control equation includes a fourth constraint equation, and the fourth constraint equation is related to the first constraint equation, the second constraint equation, and the third constraint equation;

[0199] The first constraint equation is: t +u*u x +v*u y +P x -1 / R e *(u xx +u yy );

[0200] The second constraint equation is: t +u*v x +v*v y +P y -1 / R e *(v xx +v yy );

[0201] The third constraint equation is: x +v y ;

[0202] The fourth constraint equation is:

[0203]

[0204] Among them, u is the horizontal velocity, v is the vertical velocity, u x is the partial derivative of u in the x direction, u y is the partial derivative of u in the y direction, v x is the partial derivative of v in the x direction, v y is the partial derivative of v in the y direction, u t is the time derivative of horizontal velocity, v t is the time derivative of vertical velocity, u xxis the second-order derivative of u in the x direction, u yy is the second-order partial derivative of u in the y direction, v xx is the second-order derivative of v in the x direction, v yy is the second-order derivative of v in the y direction, P x is the partial derivative of pressure P in the x direction, P y is the partial derivative of pressure in the y direction, R e is the Reynolds number, a dimensionless number that characterizes the inertia and viscosity of the fluid, m is the mass of the test piece, k is the equivalent damping of the test piece, c is the stiffness of the test piece, δ is the amplitude of the vibration of the test piece, is the vibration speed of the test piece, is the acceleration of the test piece vibration, D is the length of the test piece, and θ is the rotation angle of the cylinder in the test piece.

[0205] Optionally, the vortex-induced vibration device includes a test wind tunnel, a test piece, a uniform flow device, and a collection device; the test wind tunnel is used to generate a uniform airflow; the uniform flow device is used to use the airflow to provide different flow velocities, and the flow velocities are used to act on the test piece to generate a reduced velocity; the collection device is used to collect vibration amplitude and flow field information of the vortex-induced vibration device;

[0206] The acquisition module 701 is specifically used to obtain the vibration amplitude and flow field information of the vortex-induced vibration device at different times under different reduced velocities.

[0207] In the technical solutions disclosed herein, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0208] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0209] Figure 8 A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are provided as examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0210] like Figure 8As shown, the device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. Various programs and data required for the operation of the device 800 can also be stored in the RAM 803. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0211] Various components in device 800 are connected to I / O interface 805, including an input unit 806, such as a keyboard, mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a magnetic disk, optical disk, etc.; and a communication unit 809, such as a network card, modem, wireless communication transceiver, etc. The communication unit 809 allows device 800 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0212] The computing unit 801 can be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 801 performs the various methods and processes described above, such as the vortex-induced vibration regulation method or the vortex-induced vibration deep learning model training method. For example, in some embodiments, the vortex-induced vibration regulation method or the vortex-induced vibration deep learning model training method can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed on the device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the vortex-induced vibration regulation method or the vortex-induced vibration deep learning model training method described above can be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to execute the vortex-induced vibration regulation method or the vortex-induced vibration deep learning model training method in any other appropriate manner (for example, by means of firmware).

[0213] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0214] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0215] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0216] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0217] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0218] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0219] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not limited herein.

[0220] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A vortex-induced vibration regulation method, comprising: Acquire vortex-induced vibration data of the vortex-induced vibration equipment during a candidate period; Inputting the vortex-induced vibration data into a vortex-induced vibration deep learning model to obtain predicted vortex-induced vibration data of the vortex-induced vibration device in the next time period; Acquire target vortex-induced vibration data expected to be achieved by the vortex-induced vibration device in the next time period; determining a control instruction based on the predicted vortex-induced vibration data and the target vortex-induced vibration data; The vortex-induced vibration equipment is adjusted using the control instructions.

2. The method according to claim 1, wherein The target vortex-induced vibration data includes a vibration amplitude; the control instruction includes a control instruction for a motor, and the motor is a motor connected to the vortex-induced vibration device; The adjusting the vortex-induced vibration device by using the control instruction includes: The control instruction is sent to the motor, and the control instruction is used for the motor to execute the control instruction to adjust the length of the test piece in the vortex-induced vibration equipment, and the length affects the vibration amplitude of the test piece in the vortex-induced vibration equipment.

3. The method according to claim 2, wherein: The determining of the control instruction based on the predicted vortex-induced vibration data and the target vortex-induced vibration data includes: determining a current control amount based on the predicted vortex-induced vibration data and the target vortex-induced vibration data; Converting the current control quantity into an analog current control instruction through digital-to-analog conversion; The sending of the control instruction to the motor includes: The analog current control instruction is sent to the motor.

4. The method according to claim 1, wherein determining the control instruction based on the predicted vortex-induced vibration data and the target vortex-induced vibration data comprises: comparing the predicted vortex-induced vibration data with the target vortex-induced vibration data; determining a control instruction for increasing the vibration amplitude of the vortex-induced vibration device when the predicted vortex-induced vibration data is less than the target vortex-induced vibration data; In a case where the predicted vortex-induced vibration data is greater than the target vortex-induced vibration data, a control instruction for reducing the vibration amplitude of the vortex-induced vibration device is determined.

5. The method according to any one of claims 1 to 4, wherein: The vortex-induced vibration deep learning model is trained using historical vortex-induced vibration data of the vortex-induced vibration equipment.

6. The method according to any one of claims 1 to 4, wherein: The vortex-induced vibration deep learning model includes a flow field information neural network and a structural vibration neural network; The step of inputting the vortex-induced vibration data into a vortex-induced vibration deep learning model to obtain predicted vortex-induced vibration data of the vortex-induced vibration device in the next time period includes: The vortex-induced vibration data is input into a vortex-induced vibration deep learning model, the flow field information is obtained through the flow field information neural network, and the vibration amplitude is obtained through the structural vibration neural network.

7. A vortex-induced vibration deep learning model training method, comprising: Acquire a plurality of sample vortex-induced vibration data, wherein the sample vortex-induced vibration data is collected using a vortex-induced vibration device; Utilizing the multiple sample vortex-induced vibration data, a vortex-induced vibration deep learning model is trained through the flow field control equation and the structural vibration control equation, wherein the structural vibration control equation is related to the flow field control equation. The vortex-induced vibration deep learning model is used to input the vortex-induced vibration data into the vortex-induced vibration deep learning model after obtaining the vortex-induced vibration data of the vortex-induced vibration equipment in the candidate time period, so as to obtain the predicted vortex-induced vibration data of the vortex-induced vibration equipment in the next time period, so as to determine the control instructions based on the predicted vortex-induced vibration data and the target vortex-induced vibration data, and utilize the control instructions to adjust the vortex-induced vibration equipment, wherein the target vortex-induced vibration data is the target vortex-induced vibration data that the vortex-induced vibration equipment is expected to achieve in the next time period.

8. The method according to claim 7, wherein: The method of using the plurality of sample vortex-induced vibration data to train a vortex-induced vibration deep learning model through flow field control equations and structural vibration control equations includes: For each sample vortex-induced vibration data, inputting the sample vortex-induced vibration data into the vortex-induced vibration deep learning model to obtain a vortex-induced vibration prediction value corresponding to the sample vortex-induced vibration data; Based on the vortex-induced vibration prediction value, the loss value is calculated through the flow field control equation and the structural vibration control equation; The model parameters of the vortex-induced vibration deep learning model are adjusted based on the loss value until a preset training condition is reached to obtain a trained vortex-induced vibration deep learning model.

9. The method according to claim 8, wherein The flow field control equation includes a first constraint equation, a second constraint equation, and a third constraint equation; the structural vibration control equation includes a fourth constraint equation, and the fourth constraint equation is related to the first constraint equation, the second constraint equation, and the third constraint equation; The first constraint equation is: t +u*u x +v*u y +P x -1 / R e *(u xx +u yy ); The second constraint equation is: t +u*v x +v*v y +P y -1 / R e *(v xx +v yy ); The third constraint equation is: x +v y ; The fourth constraint equation is: Among them, u is the horizontal velocity, v is the vertical velocity, u x is the partial derivative of u in the x direction, u y is the partial derivative of u in the y direction, v x is the partial derivative of v in the x direction, v y is the partial derivative of v in the y direction, u t is the time derivative of horizontal velocity, v t is the time derivative of vertical velocity, u xx is the second-order derivative of u in the x direction, u yy is the second-order partial derivative of u in the y direction, v xx is the second-order derivative of v in the x direction, v yy is the second-order derivative of v in the y direction, P x is the partial derivative of pressure P in the x direction, P y is the partial derivative of pressure in the y direction, R e is the Reynolds number, a dimensionless number that characterizes the inertia and viscosity of the fluid, m is the mass of the test piece, k is the equivalent damping of the test piece, c is the stiffness of the test piece, δ is the amplitude of the vibration of the test piece, is the vibration speed of the test piece, is the acceleration of the test piece vibration, D is the length of the test piece, and θ is the rotation angle of the cylinder in the test piece.

10. The method according to any one of claims 7 to 9, wherein: The vortex-induced vibration equipment includes a test wind tunnel, a test piece, a uniform flow device, and a collection device; the test wind tunnel is used to generate a uniform airflow; the uniform flow device is used to use the airflow to provide different flow velocities, and the flow velocities are used to act on the test piece to generate a reduced velocity; The acquisition device is used to acquire the vibration amplitude and flow field information of the vortex-induced vibration device; The obtaining of a plurality of sample vortex-induced vibration data comprises: The vibration amplitude and flow field information of the vortex-induced vibration device at different times under different reduced speeds are obtained.

11. A vortex-induced vibration regulating device, comprising: A first acquisition module is used to acquire vortex-induced vibration data of the vortex-induced vibration equipment in a candidate time period; A data acquisition module is used to input the vortex-induced vibration data into a vortex-induced vibration deep learning model to obtain predicted vortex-induced vibration data of the vortex-induced vibration device in the next time period; A second acquisition module is used to acquire target vortex-induced vibration data that the vortex-induced vibration device is expected to achieve in the next time period; a determination module, configured to determine a control instruction based on the predicted vortex-induced vibration data and the target vortex-induced vibration data; An adjustment module is used to adjust the vortex-induced vibration equipment using the control instruction.

12. The device according to claim 11, wherein The target vortex-induced vibration data includes a vibration amplitude; the control instruction includes a control instruction for a motor, and the motor is a motor connected to the vortex-induced vibration device; The adjustment module is specifically used to send the control instruction to the motor, and the control instruction is used for the motor to execute the control instruction to adjust the length of the test piece in the vortex-induced vibration equipment, and the length affects the vibration amplitude of the test piece in the vortex-induced vibration equipment.

13. The device according to claim 12, wherein The determining module is specifically configured to determine a current control variable based on the predicted vortex-induced vibration data and the target vortex-induced vibration data; and convert the current control variable into an analog current control instruction through digital-to-analog conversion; The regulating module is specifically configured to send the analog current control instruction to the motor.

14. The device according to claim 11, wherein the determination module is specifically used to compare the predicted vortex-induced vibration data with the target vortex-induced vibration data; when the predicted vortex-induced vibration data is smaller than the target vortex-induced vibration data, determine a control instruction for increasing the vibration amplitude of the vortex-induced vibration device; when the predicted vortex-induced vibration data is larger than the target vortex-induced vibration data, determine a control instruction for reducing the vibration amplitude of the vortex-induced vibration device.

15. The device according to any one of claims 11 to 14, wherein The vortex-induced vibration deep learning model is trained using historical vortex-induced vibration data of the vortex-induced vibration equipment.

16. The device according to any one of claims 11 to 14, wherein The vortex-induced vibration deep learning model includes a flow field information neural network and a structural vibration neural network; The data acquisition module is specifically used to input the vortex-induced vibration data into the vortex-induced vibration deep learning model, obtain flow field information through the flow field information neural network, and obtain vibration amplitude through the structural vibration neural network.

17. A vortex-induced vibration deep learning model training device, comprising: An acquisition module is used to acquire a plurality of sample vortex-induced vibration data, wherein the sample vortex-induced vibration data is collected by using a vortex-induced vibration device; A training module is used to train a vortex-induced vibration deep learning model using the multiple sample vortex-induced vibration data through flow field control equations and structural vibration control equations, wherein the structural vibration control equations are related to the flow field control equations. The vortex-induced vibration deep learning model is used to input the vortex-induced vibration data of the vortex-induced vibration equipment in the candidate time period into the vortex-induced vibration deep learning model after obtaining the vortex-induced vibration data of the vortex-induced vibration equipment in the next time period, so as to determine the control instructions based on the predicted vortex-induced vibration data and the target vortex-induced vibration data, and use the control instructions to adjust the vortex-induced vibration equipment, and the target vortex-induced vibration data is the target vortex-induced vibration data that the vortex-induced vibration equipment is expected to achieve in the next time period.

18. The device according to claim 17, wherein The training module is specifically used to input each sample vortex-induced vibration data into the vortex-induced vibration deep learning model to obtain the vortex-induced vibration prediction value corresponding to the sample vortex-induced vibration data; based on the vortex-induced vibration prediction value, calculate the loss value through the flow field control equation and the structural vibration control equation; based on the loss value, adjust the model parameters of the vortex-induced vibration deep learning model until the preset training conditions are reached, and obtain the trained vortex-induced vibration deep learning model.

19. The device according to claim 18, wherein The flow field control equation includes a first constraint equation, a second constraint equation, and a third constraint equation; the structural vibration control equation includes a fourth constraint equation, and the fourth constraint equation is related to the first constraint equation, the second constraint equation, and the third constraint equation; The first constraint equation is: t +u*u x +v*u y +P x -1 / R e *(u xx +u yy ); The second constraint equation is: t +u*v x +v*v y +P y -1 / R e *(v xx +v yy ); The third constraint equation is: x +v y ; The fourth constraint equation is: Among them, u is the horizontal velocity, v is the vertical velocity, u x is the partial derivative of u in the x direction, u y is the partial derivative of u in the y direction, v x is the partial derivative of v in the x direction, v y is the partial derivative of v in the y direction, u t is the time derivative of horizontal velocity, v t is the time derivative of vertical velocity, u xx is the second-order derivative of u in the x direction, u yy is the second-order partial derivative of u in the y direction, v xx is the second-order derivative of v in the x direction, v yy is the second-order derivative of v in the y direction, P x is the partial derivative of pressure P in the x direction, P y is the partial derivative of pressure in the y direction, R e is the Reynolds number, a dimensionless number that characterizes the inertia and viscosity of the fluid, m is the mass of the test piece, k is the equivalent damping of the test piece, c is the stiffness of the test piece, δ is the amplitude of the vibration of the test piece, is the vibration speed of the test piece, is the acceleration of the test piece vibration, D is the length of the test piece, and θ is the rotation angle of the cylinder in the test piece.

20. The device according to any one of claims 17 to 19, wherein The vortex-induced vibration equipment includes a test wind tunnel, a test piece, a uniform flow device, and a collection device; the test wind tunnel is used to generate a uniform airflow; the uniform flow device is used to use the airflow to provide different flow velocities, and the flow velocities are used to act on the test piece to generate a reduced velocity; the collection device is used to collect the vibration amplitude and flow field information of the vortex-induced vibration equipment; The acquisition module is specifically used to obtain the vibration amplitude and flow field information of the vortex-induced vibration equipment at different times under different reduced speeds.

21. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 10.

22. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-10.

23. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 10.

Images