Hardware-in-loop simulation method, equipment and product based on data regression model

Abstract

The invention provides a hardware-in-the-loop simulation method and device based on a data regression model, a medium and a product. The method comprises the steps that simulation data of a hardware-in-the-loop simulation system of a wind turbine generator is collected to serve as a training set of the data regression model; inputting the input data in the simulation data at different moments into the data regression model, taking the wind speed and wind direction data in the simulation data as the output data of the data regression model, and completing the training of the data regression model; inputting the data, collected in real time, of the wind turbine generator into the data regression model, and outputting reconstructed wind speed and reconstructed wind direction data; taking the reconstructed wind speed and the reconstructed wind direction data as the input of a hardware-in-the-loop simulation system of the wind turbine generator, and transmitting the output wind speed and wind direction data and the simulation operation state of the wind turbine generator to a control system of the wind turbine generator; the hardware-in-the-loop simulation system realizes a simulation process. The problem that the scene cannot be reproduced when wind speed measurement is adopted for simulation is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of wind power, and in particular to a hardware-in-the-loop simulation method, device, medium and product based on a data regression model. BACKGROUND

[0002] As an important part of the national sustainable development strategy, wind power is a clean energy technology that converts air kinetic energy into electrical energy. In recent years, the total installed capacity of wind power has steadily increased, with higher power levels and larger rotor diameters. This has placed higher demands on the stability and safety of wind turbine control systems.

[0003] Hardware-in-the-loop simulation is a key means of testing the reliability of wind turbine control systems. In the hardware-in-the-loop simulation method, the control system uses a real controller, and the controlled object is replaced by a high-speed real-time simulation system. This can avoid damage caused by real wind turbines participating in testing, and can simulate the control actions of the control system under extreme wind conditions to improve the design of the control system.

[0004] In addition to playing an irreplaceable role in the early design of control systems, hardware-in-the-loop simulation technology also plays a significant role after the unit is put into operation. Since wind speed and direction in nature are randomly changing and cannot fully meet the needs of wind turbine stable power generation, speed and power fluctuations are often found in actual operation. Simulating actual wind speed and direction through hardware-in-the-loop simulation technology often reproduces field problems, facilitating timely problem identification and control strategy adjustment.

[0005] In order to reproduce the actual operation problems of wind turbines, the actual wind speed and direction collected on site need to be used as the input of the simulation system. However, since the anemometer and wind vane are generally installed on the tail side of the cabin, they are affected by the rotation of the rotor, resulting in errors between the measured wind speed and the actual wind speed. Using measured wind speed for simulation cannot reproduce field problems, and the same control algorithm and wind speed and direction produce completely different unit operation results. SUMMARY

[0006] Based on the above problems, the present application provides a hardware-in-the-loop simulation method, device, medium and product based on a data regression model, which solves the technical problem that in the prior art, due to the fact that an anemometer and a wind vane are generally installed at the tail side of a cabin, they are affected by the rotation of a blade wheel, resulting in an error between the measured wind speed and the actual wind speed, simulation cannot reproduce the on-site problem, and the same control algorithm and wind speed and direction produce completely different unit operation effects. The simulation method reconstructs the wind speed and direction data collected on site by using a machine learning or neural network method, uses the reconstructed wind speed and direction data as the input of a fan simulation model, and uses a real control system to participate in the control of the simulation model. By comparing with the on-site unit operation data, the same control effect can be achieved, and the improvement and adjustment of the control strategy are completed.

[0007] The present application provides a hardware-in-the-loop simulation method based on a data regression model, comprising: Collecting simulation data of a hardware-in-the-loop simulation system of a wind turbine generator as a training set of a data regression model; Inputting the input data in the simulation data at different times into the data regression model, inputting the wind speed and direction data in the simulation data as the output data of the data regression model, and completing the training of the data regression model; Inputting the real-time collected data of the wind turbine generator into the data regression model, and outputting reconstructed wind speed and direction data; Using the reconstructed wind speed and direction data as the input of the hardware-in-the-loop simulation system of the wind turbine generator, and transmitting the output wind speed and direction data and the simulated operation state of the unit to the control system of the wind turbine generator; Downlinking the pitch instruction and torque instruction of the control system to the hardware-in-the-loop simulation system of the wind turbine generator, and the hardware-in-the-loop simulation system realizing the simulation process.

[0008] In addition, the input data in the simulation data at least includes: rotational speed, torque, pitch angle and power.

[0009] In addition, the inputting the input data in the simulation data at different times into the data regression model, inputting the wind speed and direction data in the simulation data as the output data of the data regression model, and completing the training of the data regression model comprises: According to Completing the training of the data regression model G, wherein, representing the generator rotational speed at time t, representing the generator torque at time t, representing the pitch angle at time t, representing the power generation at time t, representing the wind speed at time t, representing the wind direction at time t.

[0010] In addition, the data regression model is a machine learning model or a neural network model.

[0011] In addition, the data regression model adopts a long short-term memory network.

[0012] In addition, the input of the reconstructed wind speed and wind direction data to the hardware-in-the-loop simulation system of the wind turbine generator includes: The UDP communication protocol is adopted, and timestamp information is carried by data to synchronize the time of data with internal data of the control system.

[0013] In addition, if the pitch response delay exceeds a preset threshold and the torque control error exceeds a rated torque preset value, a retraining process of the data regression model is triggered.

[0014] The application further provides an electronic device, comprising: at least one processor; and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable 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 execute the hardware-in-the-loop simulation method based on the data regression model according to any one of the preceding embodiments.

[0015] The application further provides a storage medium storing computer instructions, and when a computer executes the computer instructions, the computer instructions are used to execute the hardware-in-the-loop simulation method based on the data regression model according to any one of the preceding embodiments.

[0016] The application further provides a computer program product comprising computer programs / instructions, and the computer programs / instructions are executed by a processor to implement the hardware-in-the-loop simulation method based on the data regression model according to any one of the preceding embodiments.

[0017] The application solves the technical problem that in the prior art, because an anemometer and a wind vane are generally installed at the tail side of a cabin and are affected by rotation of a blade, there is an error between a measured wind speed and an actual wind speed, simulation cannot reproduce a field problem, and the same control algorithm and wind speed and wind direction result in completely different wind turbine operation effects. The simulation method adopts a machine learning or neural network method to reconstruct wind speed and wind direction data collected in the field, the reconstructed wind speed and wind direction data are used as input of a wind turbine simulation model, a real control system is used to participate in simulation model control, and comparison with field wind turbine operation data can achieve the same control effect, thereby completing improvement and adjustment of a control strategy. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1A flowchart of a hardware-in-the-loop simulation method based on a data regression model is provided for an embodiment of the present application. Figure 2 A hardware-in-the-loop simulation process diagram of a wind turbine based on a data regression model is provided for an embodiment of the present application. Figure 3 A data regression model training flowchart is provided for an embodiment of the present application. Figure 4 A hardware structure diagram of an electronic device is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0019] The present application is described in further detail below in connection with specific embodiments and the accompanying drawings. It is intended that the specific embodiments merely be illustrative and not restrictive of the present application, the scope of which is to be determined by the claims.

[0020] Reference Figure 1 The present application proposes a hardware-in-the-loop simulation method based on a data regression model, comprising: Step S001, collecting simulation data of a hardware-in-the-loop simulation system of a wind turbine as a training set of a data regression model; Step S002, inputting input data in simulation data at different time into the data regression model, taking wind speed and wind direction data in the simulation data as output data of the data regression model, and completing training of the data regression model; Step S003, inputting real-time collected data of the wind turbine into the data regression model, and outputting reconstructed wind speed and reconstructed wind direction data; Step S004, taking the reconstructed wind speed and reconstructed wind direction data as input of the hardware-in-the-loop simulation system of the wind turbine, and transmitting output wind speed, wind direction data and simulated running state of the unit to a control system of the wind turbine; Step S005, issuing a variable pitch instruction and a torque instruction of the control system to the hardware-in-the-loop simulation system of the wind turbine, and the hardware-in-the-loop simulation system implements a simulation process.

[0021] The simulation running data of the hardware-in-the-loop simulation system of the wind turbine under various wind conditions and working conditions is collected as a training set for model training. As shown in Table I, representing a generator speed at time t, representing a generator torque at time t, representing a variable pitch angle at time t, representing a power generation at time t, representing a wind speed at time t, representing a wind direction at time t.

[0022] Table I: Input-output relationship of data regression model

[0023] A data regression model, which can be a machine learning model or a neural network model, is trained, and the key data of the unit operation at different times, such as rotation speed, torque, pitch angle, and power, are taken as input data of the training set, and the known wind speed and wind direction data in the simulation data are taken as output of the training set, and the training of the data regression model G is completed according to the following formula: The trained model G is used to reconstruct the wind speed and wind direction data on site, and the rotation speed, torque, pitch angle, and power data actually collected on site are taken as input of the model G, and the reconstructed wind speed and wind direction data are calculated The reconstructed wind speed and wind direction data are taken as input of the wind turbine simulation system, the wind speed, wind direction, and simulated operation state of the unit are transmitted to the control system through a real-time simulation interface, the pitch command and torque command of the control system are issued to the simulation system, the hardware-in-the-loop simulation process is completed, and the actual operation effect on site is reproduced, as shown in Figure 2 Figure 3

[0024] The present application provides a hardware-in-the-loop simulation method based on a data regression model, which solves the technical problem in the prior art that the wind speed meter and the wind vane are generally installed on the tail side of the cabin, and are affected by the rotation of the impeller, resulting in errors between the measured wind speed and the actual wind speed, and the simulation using the measured wind speed cannot reproduce the on-site problem, and the same control algorithm and wind speed and wind direction produce completely different unit operation effects. The simulation method reconstructs the wind speed and wind direction data collected on site by using a machine learning or neural network method, and the reconstructed wind speed and wind direction data are taken as input of the simulation model, and the real control system is used to participate in the simulation model control, and the same control effect can be achieved by comparing the on-site unit operation data, and the control strategy is improved and adjusted.

[0025] In one embodiment, the input data in the simulation data at least includes rotation speed, torque, pitch angle, and power.

[0026] When selecting input data, important data affecting the operation of the wind turbine are selected to better simulate.

[0027] In one embodiment, the input data in the simulation data at different times is input into the data regression model, the wind speed and wind direction data in the simulation data are taken as output data of the data regression model, and the training of the data regression model includes: according to the following formula:​​​​​​​​ Complete the training of the data regression model G, where represents the generator speed at time t, represents the generator torque at time t, represents the pitch angle at time t, represents the generated power at time t, represents the wind speed at time t, Represents the wind direction at time t.

[0028] Train a data regression model, which can be a machine learning model or a neural network model, to The key operating data of the unit, such as speed, torque, blade angle, power, etc., are used as the input data of the training set, and the known wind speed and wind direction data in the simulation data are used as the input data of the training set. As the output of the training set, according to Complete the training of the data regression model G.

[0029] By training the data regression model G, the data input to the hardware-in-the-loop simulation system of the wind turbine can be reconstructed, making the simulation closer to the real effect.

[0030] In one embodiment, the data regression model is a machine learning model or a neural network model. Both the machine learning model and the neural network model can be used as a training model.

[0031] In one embodiment, the data regression model uses a long short-term memory network.

[0032] The Long Short-Term Memory (LSTM) model can solve the problem of long-term dependencies: Through its unique gating mechanism, LSTM can effectively capture and process long-term dependencies in long sequence data, avoiding gradient vanishing and gradient exploding problems, and thus performs well in areas such as time series prediction and natural language processing.

[0033] In one embodiment, using the reconstructed wind speed and reconstructed wind direction data as input to a hardware-in-the-loop simulation system for a wind turbine generator system includes: Using UDP communication protocol, the data carries timestamp information to keep the data synchronized with the internal data of the control system.

[0034] In order to keep the data synchronized with the internal data of the control system, the reconstructed wind speed and reconstructed wind direction data also carry timestamp information, thereby ensuring the simulation effect.

[0035] In one embodiment, if the pitch response delay exceeds a preset threshold and the torque control error exceeds a preset value of the rated torque, a retraining process of the data regression model is triggered.

[0036] Optionally, if the pitch response delay is over 0.5s and the torque control error is over ±5% of the rated torque, a retraining process of the data regression model is triggered.

[0037] Referring to Figure 4 The present application further provides a hardware structure diagram of an electronic device, comprising: at least one processor 301; and a memory 302 in communication connection with the at least one processor 301; wherein The memory 302 stores instructions executable 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 hardware-in-the-loop simulation method based on the data regression model as described above.

[0038] Figure 4 The processor 301 is taken as an example in the foregoing embodiments.

[0039] The electronic device is preferably a controller of a vehicle. The electronic device can further comprise an input device 303 and a display device 304.

[0040] The processor 301, the memory 302, the input device 303 and the display device 304 can be connected through a bus or other means, and the connection through the bus is taken as an example in the figure.

[0041] The memory 302 is a non-volatile computer readable storage medium, which can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as the program instructions / modules corresponding to the hardware-in-the-loop simulation method based on the data regression model in the embodiments of the present application, for example, Figure 1 The processor 301 performs various functional applications and data processing by running the non-volatile software programs, instructions and modules stored in the memory 302, i.e. implements the hardware-in-the-loop simulation method based on the data regression model in the above embodiments.

[0042] The memory 302 can include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required by a function, and the data storage area can store data created according to the use of the data regression model based hardware-in-the-loop simulation method, etc. In addition, the memory 302 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state memory device. In some embodiments, the memory 302 can optionally include a memory disposed remotely with respect to the processor 301, and these remote memories can be connected to the device performing the data regression model based hardware-in-the-loop simulation method through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0043] The input device 303 can receive an input user click, and generate a signal input related to a user setting and a function control of the data regression model based hardware-in-the-loop simulation method. The display device 304 can include a display screen and the like display equipment.

[0044] When one or more modules stored in the memory 302 are run by the one or more processors 301, the data regression model based hardware-in-the-loop simulation method in any of the above method embodiments is executed.

[0045] An embodiment of the present application provides a storage medium, which stores computer instructions, when a computer executes the computer instructions, all steps of the data regression model based hardware-in-the-loop simulation method as described above are executed.

[0046] In the context of the present disclosure, the storage medium can be a tangible medium, which can contain or store programs for use by or in connection with an instruction execution system, device or apparatus. The storage medium can be a machine-readable signal medium or a machine-readable storage medium. Optionally, the storage medium can be a non-transitory computer readable storage medium, for example, the non-transitory computer readable storage medium can be a ROM, a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0047] The above description is only the principle and preferred embodiment of the present application. It should be noted that, for those skilled in the art, on the basis of the principle of the present application, several other variants can also be made, which should also be considered as the protection scope of the present application.

Claims

1. A hardware-in-the-loop simulation method based on a data regression model, characterized in that: include: Collect simulation data of the hardware-in-the-loop simulation system of the wind turbine as the training set of the data regression model; Input data from simulation data at different times is input into the data regression model, and wind speed and wind direction data in the simulation data is used as output data of the data regression model to complete the training of the data regression model; Input the real-time collected wind turbine data into the data regression model and output the reconstructed wind speed and reconstructed wind direction data; The reconstructed wind speed and direction data are used as inputs to the hardware-in-the-loop simulation system of the wind turbine, and the output wind speed, wind direction data and the simulated operating status of the turbine are transmitted to the control system of the wind turbine; The pitch control command and torque command of the control system are sent to the hardware-in-the-loop simulation system of the wind turbine generator set, and the hardware-in-the-loop simulation system realizes the simulation process.

2. The hardware-in-the-loop simulation method based on a data regression model according to claim 1, characterized in that: The input data in the simulation data include at least: rotational speed, torque, blade angle and power.

3. The hardware-in-the-loop simulation method based on a data regression model according to claim 1, characterized in that: Inputting the input data in the simulation data at different times into the data regression model, using the wind speed and wind direction data in the simulation data as the output data of the data regression model, and completing the training of the data regression model includes: according to Complete the training of the data regression model G, where represents the generator speed at time t, represents the generator torque at time t, represents the pitch angle at time t, represents the generated power at time t, represents the wind speed at time t, Represents the wind direction at time t.

4. The hardware-in-the-loop simulation method based on a data regression model according to claim 1, characterized in that: The data regression model is a machine learning model or a neural network model.

5. The hardware-in-the-loop simulation method based on a data regression model according to claim 1, characterized in that: The data regression model adopts a long short-term memory network.

6. The hardware-in-the-loop simulation method based on a data regression model according to claim 1, characterized in that: The method of using the reconstructed wind speed and reconstructed wind direction data as input to the hardware-in-the-loop simulation system of the wind turbine generator system includes: Using UDP communication protocol, the data carries timestamp information to keep the data synchronized with the internal data of the control system.

7. The hardware-in-the-loop simulation method based on a data regression model according to claim 1, characterized in that: If the pitch response delay exceeds the preset threshold and the torque control error exceeds the preset value of the rated torque, the retraining process of the data regression model is triggered.

8. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to at least one of the processors; wherein, The memory stores instructions that can be executed by at least one of the processors, and the instructions are executed by at least one of the processors to enable at least one of the processors to execute the hardware-in-the-loop simulation method based on the data regression model as described in any one of claims 1 to 7.

9. A storage medium, characterized in that: The storage medium stores computer instructions, and when a computer executes the computer instructions, it is used to execute the hardware-in-the-loop simulation method based on the data regression model according to any one of claims 1 to 7.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the hardware-in-the-loop simulation method based on the data regression model as described in any one of claims 1 to 7 is implemented.

Images