A wind farm digital twin system for building a wind environment test field

Abstract

The present invention provides a wind farm digital twin system for building a wind environment test field, which has a wind environment building laboratory, an Internet of Things platform and a wind farm digital twin. Based on the premise that the wind environment building laboratory determines the configuration state of the wind turbine array, the phenomenon of the hysteresis of the wind turbine state output is considered, and the wind turbine operation state and the wind farm state are mapped to the digital space through the wind farm digital twin. The non-steady-state numerical simulation is performed based on the simulated digital twin, and the wind turbine inlet boundary conditions are continuously simulated by computing power until the optimal solution is obtained in the cycle. The optimal solution is transmitted to the Internet of Things platform, and the Internet of Things platform controls the output wind speed of each wind turbine in the array in real time according to the optimal solution, so as to optimize the wind turbine parameters of the target space flow field at the next moment through cyclic iteration, and realize the accurate construction of the target non-steady-state wind field. The present invention realizes the interactive feedback between the real wind turbine and the virtual wind turbine, avoids the deviation of the manual regulation of the wind turbine, and effectively achieves the real-time intelligent control of the wind turbine.

Description

Technical Field

[0001] The invention belongs to the technical field of environmental simulation, and relates to a wind field digital twin system for creating a wind environment test field. Background Art

[0002] Complex environment simulation is an important part of the research and development and performance testing of autonomous intelligent unmanned systems represented by unmanned vehicles and drones. The creation of an environmental test site can accurately perceive the target unmanned system, collect multi-dimensional data information, and conduct multi-faceted tests on it, thereby improving its reliability and safety when it is officially put into use. It is an effective way to break through the current bottlenecks such as interoperability, autonomy, and training. The wind environment is one of the types of environments that need to be created.

[0003] The creation of a wind environment test field is to create a wind field for the target space. According to the needs of unmanned system development and performance testing, the wind field is non-steady-state. Some existing standardized or non-standard customized wind turbine equipment can generate different wind speeds and adjust them steplessly, but due to the mechanical structure limitations of the wind turbine (that is, the state output hysteresis phenomenon when switching output wind speed or starting and stopping), it is impossible to achieve a large range of wind speed adjustment within a short time step. In addition, a single wind turbine cannot achieve the creation of a non-steady-state wind field in a certain space. Each wind turbine needs to be responsible for a wind direction. The state presentation form of the reference wind direction diagram is as follows: Figure 1 As shown, it is necessary to configure the wind turbine array. Therefore, the key issue in creating a wind environment test field is how to determine the operating parameters of each wind turbine in the array in real time, that is, the output wind speed, under the premise of determining the configuration state of the wind turbine array, taking into account the phenomenon of hysteresis in the output of the wind turbine state, so as to achieve accurate creation of the target non-steady-state wind field. Since the non-steady-state wind environment test field for the development and performance testing of autonomous intelligent unmanned systems is a relatively cutting-edge application scenario, according to research, there is currently no technical solution to this problem.

[0004] Moreover, when the wind turbine output parameters and wind farm status at the previous moment are known, due to the hysteresis of the wind turbine status output, the wind turbine output parameters for creating the target wind farm status at the next moment can only be obtained by trial and error in the real physical space. This will greatly increase the cost of formulating wind farm creation strategies. If the time required to create the target non-steady-state wind farm is long or the time step of the wind farm status switching is small, it may not even be possible based on this method. Summary of the invention

[0005] Since the non-steady-state wind environment test field for autonomous intelligent unmanned system development and performance testing is a relatively cutting-edge application scenario, according to research, there is currently no technical solution to solve this problem. To solve this problem, a wind field digital twin system for the construction of a wind environment test field is provided. The present invention adopts the following technical solutions:

[0006] The present invention provides a wind farm digital twin system for building a wind environment test field, characterized in that it includes: a wind environment building laboratory; an Internet of Things platform; and a wind farm digital twin; wherein the wind environment building laboratory sets up a wind turbine array including a corresponding number of wind turbines according to wind direction and wind speed requirements, and each wind turbine is embedded with a sensor for real-time acquisition of wind turbine parameters when the corresponding wind turbine is running, and sends the wind turbine parameters to the Internet of Things platform, the Internet of Things platform performs data preprocessing on the wind turbine parameters collected by the wind environment building laboratory, and transmits the preprocessed wind turbine parameters to the wind farm digital twin, the wind farm digital twin models the wind environment building laboratory to generate a simulation model, and applies the wind turbine parameters read in real time from the Internet of Things platform to the wind farm digital twin. The fan inlet boundary condition is used in the simulation model as the fan inlet boundary condition at the current moment, and a predetermined time step is set to perform non-steady-state numerical simulation analysis to obtain the simulation analysis results. The timing parameters of the target spatial flow field are used as the optimization targets, and the fan parameters corresponding to the simulation analysis results are screened. The screened fan parameters are sent to the Internet of Things platform and saved at the same time. The Internet of Things platform will receive the screened fan parameters from the wind farm digital twin as the laboratory flow field data at the next moment, so as to generate control parameters to control the fan array of the wind environment creation laboratory. The wind environment creation laboratory will start the fan array at the next moment according to the control parameters of the Internet of Things platform, and the above cycle will be repeated to create an accurate target spatial flow field in the wind environment creation laboratory.

[0007] The wind farm digital twin system for creating a wind environment test field provided by the present invention may also have such technical features, wherein data preprocessing includes denoising and analyzing wind turbine parameters.

[0008] A wind field digital twin system for creating a wind environment test field provided by the present invention may also have such technical features, wherein the screening is specifically: comparing the simulation analysis results with the timing parameters of the target space flow field, once the simulation analysis results meet the timing parameters of the target space flow field, the fan parameters corresponding to the simulation analysis results are used as the laboratory flow field data at the next moment, once the simulation analysis results do not meet the timing parameters of the target space flow field, the simulation analysis results are discarded, and the wind field digital twin modifies the fan inlet boundary conditions to continue the simulation until the cycle obtains the simulation analysis results that match the timing parameters of the target space flow field.

[0009] A wind farm digital twin system for creating a wind environment test field provided by the present invention may also have such a technical feature, wherein the wind farm digital twin uses ANSYS parametric design language APDL to read wind turbine parameters and modify wind turbine inlet boundary conditions.

[0010] A wind farm digital twin system for creating a wind environment test field provided by the present invention may also have such technical features, wherein the modeling process of the simulation model is as follows: the wind farm digital twin first uses ANSYSICEM to model a laboratory with a wind turbine array to obtain a geometric model, then imports it into ANSYSFLUENT to construct a CFD numerical model, and uses DynamicROM technology to make the laboratory into a reduced-order model for system simulation to obtain a simulation model, thereby realizing real-time simulation.

[0011] Function and Effect of the Invention

[0012] According to the present invention, a wind farm digital twin system for creating a wind environment test field has three modules: a wind environment creation laboratory, an Internet of Things platform and a wind farm digital twin. On the premise that the configuration state of the wind turbine array is determined based on the wind environment creation laboratory, the phenomenon of hysteresis of the wind turbine state output is considered, and the wind turbine operating state and the wind farm state are mapped to the digital space through the wind farm digital twin. Non-steady-state numerical simulation is performed based on the simulated digital twin, and the wind turbine inlet boundary conditions are continuously simulated by relying on computing power until the cycle obtains the optimal solution. The optimal solution is transmitted to the Internet of Things platform, and the Internet of Things platform determines and controls the output wind speed of each wind turbine in the array of the wind environment creation room in real time according to the parameters of the optimal solution, so as to realize the precise creation of the target non-steady-state wind field, and finally optimizes the wind turbine parameters that can accurately create the target spatial flow field at the next moment through cyclic iteration. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is a schematic diagram of a wind turbine array;

[0014] Figure 2 is a structural schematic diagram of a wind farm digital twin system for creating a wind environment test field in an embodiment of the present invention;

[0015] Figure 3 It is a schematic diagram of the process of creating a wind environment test field based on the wind field digital twin system in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The technical problem to be solved by the present invention is as mentioned above, that is, on the premise of determining the configuration state of the wind turbine array, considering the phenomenon of hysteresis of the wind turbine state output, how to determine the output wind speed of each wind turbine in the array in real time to achieve accurate creation of the target non-steady-state wind field.

[0017] Specifically, according to the requirements of unmanned system development and testing, a required wind field database will first be provided, including the wind speed and wind direction that need to be created in the target space at each moment; according to the wind direction requirements, a wind turbine array containing the corresponding number of wind turbines will be configured; and a certain method will be used to achieve the creation of the required wind field presented by the wind field database in the target space through the wind turbine array. However, due to the hysteresis problem of wind turbine state output (such as the inability to reduce the output from 10m / s to 2m / s in a short time), if different wind speeds and wind directions are to be created in a time series, the output parameters of several wind turbines in the array need to be adjusted at the same time, and the determination of the array output parameters is related to the change requirements of the wind field before and after the moment. Therefore, it is necessary to rely on a certain method to achieve the specific output parameters of the wind turbine array at each moment.

[0018] The technical terms involved in the present invention are explained as follows:

[0019] Wind environment test field: Create a non-steady-state wind field in the target space for unmanned system development and performance testing.

[0020] Wind farm digital twin: Based on the simulation engine, the required non-steady-state wind farm is mapped to the digital space, and the operating parameters of the wind turbine array at each moment are calculated.

[0021] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the following is a specific description of the wind field digital twin system of the present invention for creating a wind environment test field in combination with embodiments and drawings.

[0022] <Example>

[0023] Figure 2 Schematic diagram of the structure of the wind farm digital twin system for creating a wind environment test field in an embodiment of the present invention

[0024] like Figure 2 As shown, the wind farm digital twin system 100 for creating a wind environment test field in this embodiment includes a wind environment creation laboratory 10, an Internet of Things platform 20 and a wind farm digital twin 30, which are communicatively connected with each other.

[0025] The wind environment creation laboratory 10 is equipped with a wind turbine array including a corresponding number of wind turbines according to the wind direction and wind speed requirements, wherein each wind turbine is embedded with a sensor. Once the wind turbine is running, the wind turbine parameters corresponding to the wind turbine operation can be collected in real time and sent to the Internet of Things platform 20.

[0026] The Internet of Things platform 20 is capable of receiving the wind turbine parameters transmitted by the wind environment creation laboratory 10 and the wind farm digital twin 20, performing preprocessing operations such as denoising and analysis on the wind turbine parameters collected in real time by the wind environment creation laboratory 10, and transmitting the preprocessed fan parameters to the wind farm digital twin 20; for the fan parameters received from the Internet of Things platform 20, control parameters are generated according to the fan parameters, and the control parameters are used to control the fan of the wind environment creation room 10 in real time.

[0027] The wind farm digital twin 30 is used to map the wind turbine operating status and wind farm status to the digital space, perform non-steady-state numerical simulation based on the simulated digital twin, and rely on computing power to continuously simulate the wind turbine inlet boundary conditions until the optimization target is reached, thereby outputting the wind turbine parameters that can accurately create the target spatial flow field at the next moment to the Internet of Things platform 20.

[0028] Figure 3 It is a schematic diagram of the process of creating a wind environment test field based on the wind field digital twin system in an embodiment of the present invention.

[0029] like Figure 3 As shown, the specific process of creating a wind environment test field based on the wind field digital twin system 100 is as follows:

[0030] First, the wind environment creation laboratory 10 sets up a wind turbine array according to the wind direction and wind speed requirements, collects the wind turbine parameters corresponding to the wind turbine operation in real time, and sends the collected wind turbine parameters to the Internet of Things platform 20.

[0031] The Internet of Things platform 20 then performs data preprocessing on the wind turbine parameters collected by the wind environment creation laboratory 10 , and transmits the preprocessed wind turbine parameters to the wind farm digital twin 30 .

[0032] The wind farm digital twin 30 performs non-steady-state numerical simulation analysis based on the wind turbine parameters received from the Internet of Things platform 20, and uses the timing parameters of the target spatial flow field as the optimization target, screens the wind turbine parameters corresponding to the simulation analysis results, and sends the screened wind turbine parameters to the Internet of Things platform 20 while saving them. Specifically:

[0033] First, the wind farm digital twin 30 of this embodiment uses ANSYS ICEM to model all graded laboratories in the wind environment creation laboratory 10 where wind turbine arrays are arranged to obtain a geometric model, and then imports it into ANSYS FLUENT to construct a CFD numerical model. The DynamicROM technology is used to make the laboratory into a reduced-order model for system simulation, which can greatly reduce the simulation time to achieve real-time simulation.

[0034] Secondly, the wind farm digital twin 30 applies the wind turbine parameters read from the IoT platform 20 to the simulation model containing the reduced-order model as the wind turbine inlet boundary conditions through the ANSYS parametric design language (APDL), sets a certain time step to perform unsteady numerical simulation analysis to obtain the laboratory flow field data at the next moment, and compares the derived results of the simulation analysis with the target spatial flow field:

[0035] When the simulation analysis result is inconsistent with the target spatial flow field, the result is discarded, and the wind farm digital twin 30 modifies the wind turbine inlet boundary conditions through the ANSYS parametric design language (APDL) and continues the simulation, as above, until the optimal solution is obtained in the cycle;

[0036] When the simulation analysis results are consistent with the target spatial flow field, the wind farm digital twin 30 directly outputs the wind turbine parameters to the Internet of Things platform 20 and saves the current flow field results as the initial flow field at the next moment.

[0037] Finally, the Internet of Things platform 20 uses the screened wind turbine parameters received from the wind farm digital twin 30 as the laboratory flow field data at the next moment, thereby generating control parameters to control the wind turbine array of the wind environment creation laboratory 10. The wind environment creation laboratory 10 starts the wind turbine array at the next moment according to the control parameters of the Internet of Things platform.

[0038] Through the above-mentioned iterative cycle, the fan parameters that can accurately create the target spatial flow field at the next moment are optimized.

[0039] Assume that the current demand is an extremely simplified target wind farm: a total of 6 seconds is required, with a time step of 2 seconds / wind farm state. The initial state is north wind 10m / s, which switches to southeast wind 3m / s after 2 seconds, and then switches to northeast wind 8m / s after another 2 seconds (i.e. the 4th second).

[0040] To create the initial wind farm state, the state parameters of the wind turbine responsible for the north wind can be set to 10m / s. However, if you want to switch to the southeast wind of 3m / s after 2 seconds, you can use the wind farm digital twin system 100 based on simulation in this embodiment to obtain the state parameter adjustment strategy of each wind turbine in the wind turbine array after 2 seconds. Then, at the 4th second, the same process is used to obtain the state parameter adjustment strategy of each wind turbine in the wind turbine array at that moment, thereby completing the creation of the target wind farm.

[0041] Example Function and Effect

[0042] According to the wind farm digital twin system for creating a wind environment test field provided by this embodiment, the system has three modules: a wind environment creation laboratory, an Internet of Things platform, and a wind farm digital twin. On the premise that the configuration status of the wind turbine array is determined based on the wind environment creation laboratory, the phenomenon of hysteresis of the wind turbine status output is considered, and the wind turbine operating status and the wind farm status are mapped to the digital space through the wind farm digital twin. Non-steady-state numerical simulation is performed based on the simulated digital twin, and the wind turbine inlet boundary conditions are continuously simulated by relying on computing power until the cycle obtains the optimal solution. The optimal solution is transmitted to the Internet of Things platform, and the Internet of Things platform determines and controls the output wind speed of each wind turbine in the array of the wind environment creation room in real time according to the parameters of the optimal solution, so as to realize the precise creation of the target non-steady-state wind field, and finally optimizes the wind turbine parameters that can accurately create the target spatial flow field at the next moment through cyclic iteration.

[0043] In summary, this embodiment creates an extreme wind farm solution based on the digital twin method, which comprehensively utilizes modeling and CFD simulation analysis software, sensors, digital twin platforms, Internet of Things platforms and other intelligent devices, realizes interactive feedback between real wind turbines and virtual wind turbines, avoids deviations in manual control of wind turbines, and effectively achieves real-time intelligent control of wind turbines.

[0044] The above embodiments are only used to illustrate specific implementation modes of the present invention, and the present invention is not limited to the description scope of the above embodiments.

Claims

1. A wind farm digital twin system for building a wind environment test field. It is characterized in that include: Wind environment creation laboratory; IoT platform; as well as Wind farm digital twin; The wind environment creation laboratory sets up a wind turbine array including a corresponding number of wind turbines according to the wind direction and wind speed requirements. Each wind turbine is embedded with a sensor for real-time acquisition of wind turbine parameters when the corresponding wind turbine is running, and sends the wind turbine parameters to the Internet of Things platform. The Internet of Things platform performs data preprocessing on the wind turbine parameters collected by the wind environment creation laboratory, and transmits the preprocessed wind turbine parameters to the wind farm digital twin. The wind farm digital twin models the wind environment creation laboratory to generate a simulation model, applies the wind turbine parameters read in real time from the Internet of Things platform to the simulation model as the wind turbine inlet boundary conditions at the current moment, sets a predetermined time step to perform non-steady-state numerical simulation analysis to obtain simulation analysis results, and uses the timing parameters of the target spatial flow field as the optimization target to screen the wind turbine parameters corresponding to the simulation analysis results, and sends the screened wind turbine parameters to the Internet of Things platform while saving them. The IoT platform uses the selected wind turbine parameters received from the wind farm digital twin as the laboratory flow field data at the next moment, thereby generating control parameters to control the wind turbine array of the wind environment creation laboratory. The wind environment creation laboratory starts the wind turbine array at the next moment according to the control parameters of the Internet of Things platform. The above cycle allows the wind environment creation laboratory to create an accurate target space flow field.

2. A wind farm digital twin system for creating a wind environment test field according to claim 1, characterized in that: in, The data preprocessing includes denoising and analyzing the fan parameters.

3. The wind farm digital twin system for creating a wind environment test field according to claim 1, characterized in that: in, The screening is specifically as follows: Compare the simulation analysis results with the time series parameters of the target spatial flow field, Once the simulation analysis results meet the time series parameters of the target spatial flow field, the fan parameters corresponding to the simulation analysis results are used as the laboratory flow field data at the next moment. Once the simulation analysis result does not meet the timing parameters of the target spatial flow field, the simulation analysis result is discarded, and the wind farm digital twin modifies the fan inlet boundary conditions to continue the simulation until the simulation analysis result that matches the timing parameters of the target spatial flow field is obtained in a cycle.

4. A wind farm digital twin system for creating a wind environment test field according to claim 3, characterized in that: in, The wind farm digital twin uses ANSYS parametric design language APDL to read wind turbine parameters and modify wind turbine inlet boundary conditions.

5. The wind farm digital twin system for creating a wind environment test field according to claim 1, characterized in that: in, The modeling process of the simulation model is as follows: The wind farm digital twin first uses ANSYS ICEM to model a laboratory with a wind turbine array to obtain a geometric model, then imports it into ANSYS FLUENT to build a CFD numerical model, and uses Dynamic ROM technology to make the laboratory into a reduced-order model for system simulation to obtain a simulation model, thereby realizing real-time simulation.

Images