An RTLAB-based full-digital wind farm simulation method, system and device

Abstract

This invention discloses a fully digital wind farm simulation method, system, and device based on RTLAB. The method includes: initializing various basic parameters of the simulated wind farm by manually setting the model's input quantities or synchronizing data with the actual wind farm; performing preprocessing for yaw in various wind directions using deep learning and intelligent algorithms; establishing corresponding turbine-level models, corresponding wind farm-level models, and corresponding data interfaces for the simulated wind farm; initializing the platform communication software according to the required data volume and performing overall model pre-compilation processing; setting the software and algorithms in the server according to the actual wind farm data for real-time simulation. This invention integrates aerodynamic and electromagnetic characteristics to achieve a comprehensive wind farm model with aerodynamic, turbine, and electronic control linkage. Through an advanced algorithm server and cloud communication system, this system achieves synchronous simulation of a mirrored wind farm, and provides reference and assistance for the control of actual wind farms through wind speed prediction and wake algorithms.

Description

Technical Field

[0001] The technical field of this invention is wind power generation modeling and grid connection simulation analysis technology, and in particular, it relates to a fully digital wind farm simulation method, system and device based on RTLAB. Background Technology

[0002] my country will construct several offshore wind power bases with a capacity of tens of millions of kilowatts along its coast, and onshore wind power will also continue to develop, forming a situation of large-scale grid integration of new energy sources across multiple regions. This change in power structure will bring profound changes to the grid's form, operational control flexibility, and controllability, and also present greater challenges to the operation and stability control of the power system. Therefore, new large-scale wind farms connected to the AC / DC grid require rigorous steady-state and transient simulation demonstrations to ensure the safe and stable operation of the power system. On the other hand, maximizing grid-connected power and reducing generation costs are the goals of wind farm operators, requiring continuous refinement of wind farm power generation optimization control strategies. Digital simulation platforms for wind farms are essential tools for conducting research in both of these areas.

[0003] Wind turbine models span different domains, including aerodynamics, mechanics, and electricity. Large wind farms and wind farm clusters consist of hundreds of wind turbines. The randomness and distribution characteristics of natural wind energy, coupled with the wake effect of wind turbines, make the calculation of wind speed distribution across the entire farm computationally intensive and time-consuming, and it is difficult to grasp the changes in the operating status of wind turbines in real time. At the same time, wind turbines consist of aerodynamic parts of the rotor, transmission and power generation parts, and rectification and inversion parts. They have different operating control modes under natural wind speeds, and the control under fault ride-through is also more complex. Therefore, the computational resources required for the simulation of the entire farm are large, which most power system analysis software cannot handle. At present, the simulation tools for studying wind turbines from an electromechanical perspective mainly include electromagnetic simulation software such as MATLAB / Simulink, PASCAD, PASSAP, BPI, and DIgSILENT-powerfactory (without the entire farm), and front-end rotor simulation software such as FAST and Bladed. These can realize the simulation calculation of a single wind turbine or a small number of wind turbines, but they cannot perform simulation calculations of the entire medium and large-scale wind farm, nor can they consider the calculation of the aerodynamic parts. The simulation tools used to study wind power generation from an aerodynamic perspective are mainly FAST and others. They can perform offline wake analysis and calculation of wind farms and power generation assessment, but cannot be used for calculations after grid connection. Existing single-software simulations often separate gas and electricity. While integrated gas-mechanical-electric simulation platforms exist, such as Bladed-RTDS (primarily standalone), Bladed-Matlab (suitable for standalone operation but offline and not real-time), and ADPSS / ETSDAC (suitable for power system component simulation), they are often only applicable to single-unit studies. In multi-unit to field-level simulations, MATLAB-RTDS and MATLAB-RTLAB are currently the mainstream solutions. However, the wind speed inflow in these simulations is based on manual settings and lacks wake effects. Furthermore, current wind turbine models do not consider wind energy utilization under yaw conditions, resulting in a lack of understanding of wake effects within the wind farm during multi-unit simulations. Additionally, these simulations often neglect automatic generation control (AGC) systems and algorithms, failing to consider wind speed, wind direction, turbine coordinates, and grid commands. Therefore, a fully digital real-time integrated simulation platform for wind farms that encompasses wake, wind speed prediction, wind turbines, wind farm collection lines, intelligent station control, and grid connection does not currently exist. Current research on aerodynamic wakes is primarily based on small simulation platforms for single to three turbines, with simplified electromechanical simulations. Many wind farm simulations currently rely on aggregated models, which, while providing some accuracy in capacity and individual turbine characteristics, neglect the mutual influences of multiple turbines operating simultaneously. Current digital wind farm simulations lack versatility, focusing primarily on aerodynamic or electromechanical simulations, while simply using general mathematical models to replace less focused aspects. Simulation techniques typically utilize real-time turbine data, but these models lack real-time capabilities, cannot perform synchronous simulations, and suffer from low computational efficiency. Summary of the Invention

[0004] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, embodiments of the present invention provide a fully digital wind farm simulation method based on RTLAB, including:

[0008] The basic parameters of the simulated wind farm are initialized by manually setting the input quantities of the model or synchronizing data with the actual wind farm.

[0009] Based on the aforementioned basic parameters and the actual wind direction and wind speed rose diagram of the wind farm, yaw preprocessing for each wind direction is performed using deep learning and intelligent algorithms.

[0010] For the wind turbines deployed in the simulated wind farm, a corresponding turbine-level model and corresponding data interface are established; then, for the data of cables, substations, and control stations inside the simulated wind farm, a corresponding wind farm-level model and corresponding data interface are established.

[0011] The platform communication software is initialized according to the required data volume, and the overall model is pre-compiled.

[0012] Based on actual wind farm data, the software and algorithms within the server are configured, and real-time simulation begins.

[0013] As a preferred solution for a fully digital wind farm simulation method based on RTLAB, the following is provided:

[0014] The initialization settings include: obtaining various parameters through manual input or communication with the actual wind farm SCADA via a cloud server and inputting them into the wind farm-level data configuration file; classifying the various data and fixing the parameter data, which includes the wind turbine coordinates, wind turbine capacity, wind turbine rotor radius, wind turbine generator parameters, wind farm cable impedance, and constant parameters of the wind farm circuit settings; the parameter data also includes variable parameter data such as the wind direction and wind speed of the wind farm inflow.

[0015] The communication software, wind farm wake algorithm software, and wind farm automatic power generation control algorithm are initialized based on the aforementioned invariant parameters.

[0016] As a preferred solution for a fully digital wind farm simulation method based on RTLAB, the following is provided:

[0017] The wind direction yaw preprocessing includes: performing staged preprocessing on the wind farm's annual rose diagram and manually configurable parts in the basic parameters, where by default, a stage is defined as 5 degrees of wind direction and 0.5 m / s of wind speed. The optimal active yaw control parameters for the wind farm under each different stage are determined by a convolutional neural network and the DEA algorithm, and a lookup table file is generated. The lookup table file is loaded into the preliminary optimization lookup table module, and the corresponding data interface is set in the communication software to meet the communication requirements.

[0018] As a preferred solution for a fully digital wind farm simulation method based on RTLAB, the following is provided:

[0019] The establishment of the corresponding unit-level model and corresponding data interface includes: establishing the unit's aerodynamic model based on the wind turbine's mathematical model.

[0020] P t =0.5ρπR 2 ν 3 C p

[0021] in: ρ is air density, in kg / m³, taken as 1.25 kg / m³; R is the radius of the wind turbine blade, in m; v is the wind speed before entering the swept surface of the wind turbine, in m / s; C p The wind energy utilization coefficient of the wind turbine;

[0022]

[0023] in,

[0024] Where, ω t β is the rotor speed, and β is the blade pitch angle;

[0025] Then, the pitch model of the unit is established through the mathematical model of the pitch system:

[0026]

[0027] Where, β ref T serves as the reference input for the pitch angle. β The time constant of the time-varying pitch system;

[0028] The transfer function of the variable pitch system is:

[0029]

[0030] Where τ is the delay time of the variable pitch servo system;

[0031] The following formula was used to correct the parameters against the wind turbine's factory data:

[0032]

[0033]

[0034] Where λ is the tip speed ratio, β is the pitch angle, and c1-c8 are parameters obtained from the aerodynamic characteristics of the wind turbine.

[0035] A MATLAB / Simulink wind turbine motor model is established based on the wind turbine motor parameters within the invariant parameters. A MATLAB / Simulink converter switching model and corresponding average value model are established based on the DC bus and IGBT parameters. A grid-side transformer model is established based on the wind turbine transformer parameters.

[0036] Based on the DC bus parameters within the invariant parameters, establish constant speed control on the machine side, maximum power point tracking control on the machine side, reactive power control on the grid side, and constant power control of the variable pitch system; establish a data communication interface based on the data quantity and magnitude.

[0037] As a preferred solution for a fully digital wind farm simulation method based on RTLAB, the following is provided:

[0038] The establishment of the corresponding wind farm level model and corresponding data interface includes: establishing the data communication interface of the loop module; establishing the wind farm main transformer model and power grid model based on the booster station parameters in the invariant parameters; establishing the wind farm control module of the wind farm model; establishing the communication module of the wind farm model; and connecting different modules / models in the wind farm model with data / primary wiring through the Opt module of Rt-LAB to make the model conform to the RT-OPAL compilation standard.

[0039] The initialization settings for the platform communication software include: setting the communication software and data interface according to the data volume, quantity, and accuracy requirements, including data format and corresponding data names; and creating corresponding tables for data collection and input.

[0040] As a preferred solution for a fully digital wind farm simulation method based on RTLAB, the following is provided:

[0041] The software and algorithms within the configuration server include:

[0042] According to research requirements, the wake algorithm software for wind farms is initialized using corresponding wake models and wake deflection models. The Bastankhah wake model is represented as follows:

[0043]

[0044]

[0045]

[0046] Where Δu represents the wind speed loss caused by the wake at that location, U ∞ C represents the inflow velocity unaffected by the wake. T Here, β is the thrust coefficient of the wind turbine, β is the ratio of the diameter of the wake when it returns to normal atmospheric pressure to the diameter of the wind turbine rotor, d0 is the diameter of the wind turbine rotor, and k is the thrust coefficient of the wind turbine. w Where r is the wake expansion coefficient and r is the radius of the wind turbine impeller;

[0047] The Shapiro wake deflection model is expressed as follows:

[0048]

[0049] d w =d0+k w d0ln(1+e x / r-2 )

[0050]

[0051] Where v c Let θ be the lateral deflection velocity of the wake at a distance x behind the wind turbine rotor, d be the yaw angle, and d be the lateral deflection velocity. w Let y be the diameter of the wake at point x, and y be the diameter of the wake at point x. d Let x be the total deflection distance of the wake at point x;

[0052] The inflow velocity of each wind turbine in the wind farm under the initial condition is calculated iteratively by using the wind speed and wind direction in the variable parameter data, the coordinates and radius of the wind turbine in the constant parameters, and the preliminary optimization lookup table. The inflow wind speed of each wind turbine under the initial condition is obtained and imported into the table file for pre-reading.

[0053] Based on the preliminary optimization lookup table and the wind farm wake algorithm, further instruction allocation learning is carried out through convolutional neural networks for different grid commands to improve the automatic power generation control of wind farms under active yaw control.

[0054] Secondly, embodiments of the present invention provide a fully digital wind farm simulation system based on RTLAB, characterized in that it includes:

[0055] The parameter input module is used to initialize various basic parameters of the simulated wind farm by manually setting the input quantities of the model or by synchronizing data with the actual wind farm.

[0056] The wind direction and yaw preprocessing module is used to perform yaw preprocessing for each wind direction based on the basic parameters and the wind direction and wind speed rose diagram of the actual wind farm through deep learning and intelligent algorithms.

[0057] The modeling module is used to create corresponding unit-level models and data interfaces for wind turbines deployed in the simulated wind farm; then, it creates corresponding wind farm-level models and data interfaces for data on cables, substations, and central control stations within the simulated wind farm.

[0058] The model preprocessing module is used to initialize the platform communication software according to the required data volume and to perform overall model pre-compilation processing.

[0059] The real-time simulation module is used to set the software and algorithms in the server based on actual wind farm data and start real-time simulation.

[0060] Thirdly, embodiments of the present invention provide a computing device, including:

[0061] Memory and processor;

[0062] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the one or more programs are executed by the one or more processors, the one or more processors implement the fully digital wind farm simulation method based on RTLAB as described in any embodiment of the present invention.

[0063] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the aforementioned RTLAB-based fully digital wind farm simulation method.

[0064] The beneficial effects of this invention are as follows: This invention uses the RT-LAB nanosecond-level real-time simulator to perform real-time simulation of wind farm models, eliminating human error caused by relying on multiple combined turbine units for wind farm simulation and improving the accuracy of the simulation model; it achieves data synchronization with the SCADA system of the wind farm under study through a cloud server, realizing a synchronous mirror model of the wind farm, achieving online wind farm research and data analysis, and improving research efficiency; it establishes a real-time simulation platform for wind farms with the RT-LAB real-time simulator as the core, and establishes a hardware-in-the-loop system for real-time simulation through I / O interfaces and corresponding data boards. This system improves research efficiency and accuracy; it establishes a wind farm wind speed module based on the wake algorithm, calculating the inflow wind speed of each wind turbine in real time under the influence of wake and yaw through the status of each wind turbine via field control communication, reducing errors caused by manual configuration of wind speed input in wind farm simulation models; it modularizes each part of the real-time wind farm simulation platform to facilitate modification of requirements and expand the application range of the simulation platform; and it uses MATLAB / Simulink to modularly model wind turbines, establishing simulation model templates for different wind turbines and performing modular processing to improve the efficiency of manual configuration. Efficiency; Based on MODBUS / TCP communication, data communication software was compiled to achieve synchronous operation of the wind farm simulation platform and various algorithms; Automatic Generation Control (AGC) was specifically modified based on yaw control to adapt to the distribution of grid commands to each wind turbine under yaw conditions; An advanced algorithm server was established, and a data interface synchronized with the communication software was created to provide verification of various algorithms and conduct corresponding research; Through a SCADA wind farm monitoring system based on the wind farm server deployed on the simulation platform, wind farm station monitoring and data collection were achieved, providing a more realistic simulation mode for long-term real-time simulation operation; Through the communication linkage of MATLAB-RTLAB-SERVER-SCADA, the actual situation of wind farm unit-wind farm-SCADA-grid commands / constraints was simulated, greatly improving the realism of the simulation and facilitating further research; Through the analysis of actual wind farm SCADA data on the cloud server, and using the wind farm real-time simulation platform with RT-LAB as the core, wind speed prediction feedback for wind farm control in the future period helps to improve wind farm operating efficiency and reduce costs. Attached Figure Description

[0065] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0066] Figure 1This is an overall flowchart of the fully digital wind farm simulation method based on RTLAB described in the first embodiment of the present invention;

[0067] Figure 2 This is a schematic diagram of the wind turbine generator electrical model on the machine side in the fully digital wind farm simulation method based on RTLAB described in the first embodiment of the present invention;

[0068] Figure 3 This is a schematic diagram of the grid side of the electrical model of a single wind turbine unit in the fully digital wind farm simulation method based on RTLAB described in the first embodiment of the present invention;

[0069] Figure 4 This is a structural diagram of two types of wind turbine models in the fully digital wind farm simulation method based on RTLAB described in the first embodiment of the present invention;

[0070] Figure 5 This is a control diagram of a wind turbine generator set, taking a full-power unit as an example, in the fully digital wind farm simulation method based on RTLAB described in the first embodiment of the present invention.

[0071] Figure 6 This is a schematic diagram of the wind farm loop module before encapsulation in the RTLAB-based fully digital wind farm simulation method described in the first embodiment of the present invention.

[0072] Figure 7 This is a schematic diagram of the wind farm loop module after encapsulation in the RTLAB-based fully digital wind farm simulation method described in the first embodiment of the present invention.

[0073] Figure 8 This is a schematic diagram of the wind farm model in the fully digital wind farm simulation method based on RTLAB described in the first embodiment of the present invention;

[0074] Figure 9 This is a diagram of the user data interface and data allocation system in the fully digital wind farm simulation method based on RTLAB described in the first embodiment of the present invention;

[0075] Figure 10 This is a wind farm architecture diagram in the RTLAB-based fully digital wind farm simulation method described in the first embodiment of the present invention;

[0076] Figure 11 This is the overall architecture diagram of the real-time wind farm simulation platform in the RTLAB-based fully digital wind farm simulation method described in the first embodiment of the present invention;

[0077] Figure 12 This is a schematic diagram of MODBUS / TCP communication software in a simulation example of the fully digital wind farm simulation method based on RTLAB described in the second embodiment of the present invention.

[0078] Figure 13 This is a hardware architecture diagram of the RT-LAB-SCADA-SERVER-MATLAB joint real-time simulation platform in the simulation example of the fully digital wind farm simulation method based on RTLAB described in the second embodiment of the present invention.

[0079] Figure 14 This is a communication architecture diagram of the RT-LAB-SCADA-SERVER-MATLAB joint real-time simulation platform in the simulation example of the fully digital wind farm simulation method based on RTLAB described in the second embodiment of the present invention.

[0080] Figure 15 This is a local data architecture diagram of the RT-LAB-SCADA-SERVER-MATLAB joint real-time simulation platform in the simulation example of the fully digital wind farm simulation method based on RTLAB described in the second embodiment of the present invention.

[0081] Figure 16 This is a schematic diagram of wake / AGC algorithm communication and data transmission in a simulation example of the fully digital wind farm simulation method based on RTLAB described in the second embodiment of the present invention.

[0082] Figure 17 This is a flowchart of the wake algorithm in a simulation example of the fully digital wind farm simulation method based on RTLAB described in the second embodiment of the present invention.

[0083] Figure 18 This is a simulation example of the RTLAB-based fully digital wind farm simulation method described in the second embodiment of the present invention, which is a hardware architecture diagram of the RTLAB digital wind farm communication. Detailed Implementation

[0084] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0085] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0086] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0087] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0088] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0089] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0090] Example 1

[0091] Reference Figure 1-11 This is the first embodiment of the present invention, which provides a fully digital wind farm simulation method based on RTLAB, including:

[0092] S1: Initialize the basic parameters of the simulated wind farm by manually setting the input quantities of the model or synchronizing data with the actual wind farm;

[0093] Furthermore, various parameters can be obtained through manual input or by communicating with the actual wind farm's SCADA system via a cloud server and input into the wind farm-level data configuration file (EXCEL / TXT).

[0094] Specifically, the data is categorized into fixed parameters such as wind turbine coordinates, wind turbine capacity, wind turbine rotor radius, wind turbine generator parameters, wind farm cable impedance, and wind farm circuit settings, which are constant parameters during real-time simulation; and variable parameters such as wind direction and wind speed of the wind farm inflow, which may change during real-time simulation.

[0095] The MODBUS communication software, wind farm wake algorithm software, and wind farm automatic generation control (AGC) algorithm are initialized based on fixed parameter data.

[0096] It should be noted that reading data from the simulated wind farm can prepare for subsequent simulation and modeling.

[0097] S2: Based on the aforementioned basic parameters and the actual wind direction and wind speed rose diagram of the wind farm, preprocessing of yaw in each wind direction is performed using deep learning and intelligent algorithms;

[0098] Specifically, the DEA algorithm and convolutional neural network are used to preprocess fixed parameter data and variable parameter data to obtain lookup table files.

[0099] Furthermore, by using the annual rose diagram of the wind farm in both fixed and variable parameter data, and manually configurable staged preprocessing (defaulting to a stage of 5 degrees wind direction and 0.5 m / s wind speed), the optimal active yaw control parameters of the wind farm under each different stage are determined through deep learning (convolutional neural network) and intelligent algorithms (DEA algorithm) and a lookup table file is generated.

[0100] The lookup file is loaded into the preliminary optimization lookup module, and the corresponding data interface is set in the MODBUS communication software to meet the communication requirements.

[0101] It should be noted that the Automatic Generation Control (AGC) was specifically modified based on yaw control to adapt to the distribution of grid commands to each wind turbine under yaw conditions. Since active yaw control requires a large number of iterations, the optimal control strategy for different stages is obtained through staged preprocessing to reduce the number of iterations during real-time simulation and increase efficiency.

[0102] S3: For the wind turbines deployed in the simulated wind farm, establish the corresponding turbine-level model and corresponding data interface; then, for the data of cables, substations, and central control stations inside the simulated wind farm, establish the corresponding wind farm-level model and corresponding data interface.

[0103] Specifically, a single-unit model of the wind turbine is established using fixed parameter data.

[0104] Furthermore, an aerodynamic model of the unit is established based on the mathematical model of the wind turbine:

[0105] P t =0.5ρπR 2 ν 3 C p

[0106] in: ρ is the air density (kg / m3), taken as 1.25 kg / m3; R is the radius of the wind turbine blade (m); v is the wind speed before the air enters the swept surface of the wind turbine (i.e., the undisturbed wind speed) (m / s); Cp is the wind energy utilization coefficient of the wind turbine.

[0107]

[0108] In the formula

[0109]

[0110]

[0111] Where, ω t β is the rotor speed, and β is the blade pitch angle;

[0112] Next, the pitch model of the unit will be established using the mathematical model of the pitch system:

[0113]

[0114] Where, β ref T serves as the reference input for the pitch angle. β is the time constant of the time-varying pitch system.

[0115] Variable pitch systems exhibit hysteresis characteristics, and are generally simulated using a first-order inertial element with a delay. The transfer function of the variable pitch system can be obtained as follows:

[0116]

[0117] Where τ is the delay time of the variable pitch servo system.

[0118] It should be noted that the factor with the greatest impact on the aerodynamic characteristics of wind turbines in the above formula is the wind energy utilization coefficient C. p The following formula is typically used to correct the parameters of the wind turbine generator set based on its factory specifications, resulting in a formula that reflects its aerodynamic characteristics:

[0119]

[0120]

[0121] Where λ is the tip speed ratio, β is the pitch angle, and c1-c8 are parameters obtained based on the aerodynamic characteristics of the wind turbine. Aerodynamic data can be acquired through factory testing of the wind turbine or by using specialized aerodynamic software.

[0122] Furthermore, a MATLAB / Simulink wind turbine motor model is established based on the wind turbine motor parameters within the fixed parameter data, and a grid-side transformer substation model is established based on the wind turbine transformer substation parameters.

[0123] Specifically, a MATLAB / Simulink converter switching model and a corresponding average value model are established based on the DC bus and IGBT parameters within the fixed parameter data.

[0124] Based on parameters such as the DC bus in the fixed parameter data, establish constant speed control on the machine side, maximum power point tracking control on the machine side, reactive power control on the grid side, and constant power control of the variable pitch system.

[0125] Establish data communication interfaces based on the amount and magnitude of data.

[0126] Furthermore, based on data from the simulated wind farm's internal cables, substations, and control stations, a corresponding wind farm-level model was established using MATLAB / Simulink.

[0127] Specifically, a wind farm model is established using fixed parameter data.

[0128] Based on the loop data within the fixed parameter data, establish each loop module within the wind farm, including the wind turbine model within the loop and the π-type equivalent model of the cable within the loop.

[0129] Establish a data communication interface for the loop module.

[0130] Based on the booster station parameters in the fixed parameter data, establish the main transformer model of the wind farm and the power grid model.

[0131] A wind farm control module is established to create a wind farm model, which is used for real-time command allocation and user interaction interface.

[0132] A communication module for establishing a wind farm model is established, which exchanges data with MODBUS communication software through the MODBUS communication protocol and corresponding data interface.

[0133] The Opt module of Rt-LAB is used to connect different modules / models within the wind farm model for data / primary wiring, making the model conform to the RT-OPAL compilation standard.

[0134] It should be noted that different studies have different requirements for the accuracy, scale, and quantity of data, so it is necessary to set up the model data communication interface according to the requirements.

[0135] S4: Initialize the platform communication software according to the required data volume and perform overall model pre-compilation processing;

[0136] Specifically, the MODBUS communication software and data interface should be configured according to the data volume, quantity, and accuracy requirements, including the data format (INT, float, double, etc.) and the corresponding data name.

[0137] Create corresponding Excel spreadsheets based on the data for data collection and input.

[0138] The model is pre-compiled using the RT-OPAL software on the RT-LAB host computer.

[0139] It should be noted that the MATLAB / Simulink model itself does not meet the requirements of real-time simulation and has many aspects that can be optimized. The RT-OPAL software can compile the Simulink model specifically so that the model can be simulated in real time on the RT-LAB simulator (lower-level machine).

[0140] S5: Configure the software and algorithms on the server based on actual wind farm data, and begin real-time simulation.

[0141] Specifically, the software and algorithms within the SCADA server and advanced algorithm server are configured based on wind farm data, and the wake algorithm software for the wind farm is initialized.

[0142] The wake algorithm software for wind farms is initialized using the corresponding wake model and wake deflection model.

[0143] Bastankhah wake model:

[0144]

[0145]

[0146]

[0147] Where Δu is the wind speed loss caused by the wake at that location, U∞ is the inflow wind speed unaffected by the wake, CT is the wind turbine thrust coefficient, β here is the ratio of the wake diameter to the wind turbine rotor diameter when it returns to normal atmospheric pressure, d0 is the wind turbine rotor diameter, kW is the wake expansion coefficient, and r is the wind turbine rotor radius.

[0148] Shapiro wake deflection model:

[0149]

[0150] dw =d0+k w d0ln(1+e z / r-2 )

[0151]

[0152] Where vc is the lateral deflection velocity of the wake at a distance x behind the wind turbine rotor, θ is the yaw angle, dw is the diameter of the wake at x, and yd is the total deflection distance of the wake at x.

[0153] By using wind speed and direction data from variable parameters, wind turbine coordinates and radius data from fixed parameters, and preliminary optimization lookup tables, the inflow velocity of each wind turbine in the initial situation is iteratively calculated to obtain the inflow wind velocity of each wind turbine in the initial situation, and the result is imported into an EXCEL spreadsheet file for pre-reading.

[0154] Based on the initial optimization lookup table, the wind farm wake algorithm uses deep learning (convolutional neural network) to further learn command allocation for different grid commands (active and reactive power, voltage, frequency), thereby improving the automatic generation control (AGC) of wind farms under active yaw control.

[0155] The pre-compiled model is uploaded to the RT-LAB lower-level machine via RT-OPAL to start real-time simulation. Through the linkage of SCADA server, RT-LAB, MATLAB, MODBUS communication software, and various algorithm software, a real-time simulation platform for wind farms is formed.

[0156] It should be noted that when uploading the pre-compiled model to RT-LAB, the model needs to be segmented according to the model's characteristics (wind farm models are generally based on loop data), and the computational load of multiple computing cores (CPUs) should be reasonably allocated.

[0157] Furthermore, the RT-LAB lower-level machine is started, the compiled model is loaded into the OP5600 HIL BOX, and the model loading program is started.

[0158] The RT-LAB host computer allows users to perform real-time operations such as running, pausing, and reading data on the model through a user interface.

[0159] Start the software and algorithms within the SCADA server, SCADA client, and advanced algorithm server, and check the overall communication effectiveness through MODBUS communication software.

[0160] Input wind speed and direction information and collect the required research data.

[0161] Data can be input and collected through the wind farm control module, the human-computer interaction interface of various algorithms, the MODBUS communication software interface, and the corresponding EXCEL spreadsheet.

[0162] It should be noted that an advanced algorithm server was established, along with a data interface synchronized with the communication software, to provide verification of various algorithms and facilitate corresponding research. A SCADA wind farm monitoring system based on a wind farm server deployed on the simulation platform enables wind farm monitoring and data collection, providing a more realistic simulation mode for long-term real-time simulation operation. Communication linkage between MATLAB, RTLAB, SERVER, and SCADA allows for simulation of the actual situation of wind farm turbine-wind farm-SCADA-grid commands / constraints, greatly improving the realism of the simulation and facilitating further research. Analysis of actual wind farm SCADA data on the cloud server, utilizing a wind farm real-time simulation platform centered on RT-LAB, and providing wind speed prediction feedback for wind farm control over a future period, helps improve wind farm operating efficiency and reduce costs.

[0163] Example 2

[0164] Reference Figure 12-18 As an embodiment of the present invention, a fully digital wind farm simulation method based on RTLAB is provided. In order to verify the beneficial effects of the present invention, a simulation experiment is conducted for scientific demonstration.

[0165] The simulation system of this invention requires the following modules / models / devices:

[0166] MATLAB / SIMULINK:

[0167] This system includes the following modules / models:

[0168] Wind turbine single-unit model: It is used for single-unit grid-connected simulation analysis of wind turbines and outputs three-phase AC power.

[0169] Cable module: It is used to simulate the cables used to connect various devices in a wind farm and to transmit the three-phase AC power output from the loop module over long distances.

[0170] The single-unit model of the wind turbine includes the following modules:

[0171] Single-unit electrical model: It is used to simulate the generator, wind power converter, and step-up transformer of a wind turbine, and outputs three-phase AC power;

[0172] Front-end aerodynamic simulation module: It is used to simulate the front-end aerodynamic part of the wind turbine and to complete the maximum power point tracking of the wind turbine and calculate the power setpoint.

[0173] Machine-side converter control module: It is used to receive the power command from the front-end aerodynamic simulation module, control the machine-side converter of the wind turbine, realize the maximum power tracking of the wind turbine, and control the reactive power of the machine side.

[0174] Grid-side converter control module: It controls the grid-side converter of the wind turbine generator set, stabilizes the DC bus voltage of the grid-side converter, and controls the reactive power of grid connection.

[0175] Among them, the wind farm model includes the following modules:

[0176] Power system transmission line module: It is used to connect each modularized and encapsulated single wind turbine generator set model to achieve long-distance transmission of wind power.

[0177] Power grid module: It is used to simulate the large power grid model into which the wind farm is integrated.

[0178] Console subsystem module: It is used to provide display and interaction based on the RT-LAB user interface, and at the same time provide data interaction services between the SCADA server and the RT-LAB emulator.

[0179] Circuit module: It is used to simulate the single circuit of the wind farm, including the single wind turbine generator set model and the corresponding machine terminal transformer model, the power system transmission line module and the basic data interaction module.

[0180] Among them, the console subsystem model includes the following modules:

[0181] DSP data transmission module: It is used to transmit various data of the fan and the wind farm between the host computer and the platform, couple each platform to achieve linkage simulation, and transmit the required data to the SCADA wind farm monitoring system through the MODBUS protocol to provide demonstrations of various parameters.

[0182] Wind farm control module: It is used to perform real-time control / instruction allocation on the running or non-running wind farm model, and includes the wind farm AGC automatic generation control module based on Simulink.

[0183] Wind farm wind speed module: It is used to provide variable wind speed input so that the wind speed input of each fan meets the requirements; at the same time, it includes a wake calculation module, which can calculate the required wind speed input of each wind turbine generator set under the influence of wake according to the wind turbine generator set parameters.

[0184] AGC automatic generation control module: It is used to automatically allocate corresponding control instructions to the wind turbine generator sets in the field according to the grid instructions, and can be switched to the control instructions of the automatic generation control algorithm from the advanced algorithm server in the communication.

[0185] RT-LAB real-time emulator:

[0186] This system includes the following modules / devices:

[0187] Host computer (PC): It is used to run MATLAB / Simulink and RT-LAB client programs, and to compile and transfer models. It is also used to display the user interface and monitor the simulation process.

[0188] Lower-level machine (OP5600 HIL BOX): As the main body of the RT-LAB real-time simulator, it is used to receive the model compiled by the upper-level machine and allocate computing resources of the lower-level machine's computing core (CPU). At the same time, it performs data interaction / hardware-in-the-loop functions through various boards.

[0189] The lower-level machine contains the following main operating components:

[0190] Opal-RT FPGA board: It is used to manage analog / digital I / O with a sampling time of 100MHz.

[0191] Computing core (CPU): The computational core used for simulation.

[0192] server:

[0193] This system includes the following software / devices:

[0194] SCADA Server: It functions the same as the actual wind farm, interacting with the RT-LAB real-time simulator to monitor various data of the simulated wind farm.

[0195] SCADA User Terminal (PC): Serves as the user interface for SCADA, collecting, storing, and displaying various data.

[0196] MODBUS communication software and additional interfaces: It serves as the core of data interaction, integrating various platform data. Through MODBUS / TCP communication, it integrates data sent from various parts of the platform and feeds back data to the corresponding parts. Its additional interfaces are used for upgrade requirements.

[0197] Automatic Generation Control (AGC) for wind farms: It is used to automatically analyze data when receiving grid commands or when the wind farm's power generation does not meet grid constraints, and distribute the commands to each wind turbine in the wind farm.

[0198] Wind farm-level data configuration file: It is used to provide initial wind farm data, including coordinates, turbine model, turbine parameters, inflow wind speed, inflow azimuth angle and other initial data required for the wind farm simulation platform to run.

[0199] Wind farm wake algorithm software: It is used to correct the inflow wind speed of each wind turbine in the simulation wind farm based on the operating status of each wind turbine and the wind farm-level data configuration file.

[0200] Preliminary optimization lookup table: Using intelligent optimization algorithms, the optimal AWC (Active Wake Control) scheme for the wind farm under various wind directions and speeds is searched in stages of 5 degrees and 0.5 m / s, forming a preliminary lookup table, which reduces the iteration during simulation and improves simulation efficiency.

[0201] The deep learning-based intelligent optimization algorithm allocates yaw and pitch angle commands to the wind turbines in the field based on the preliminary optimization lookup data and the actual situation, so as to achieve optimal power generation under the AWC control of the wind farm.

[0202] The actual steps include: First, the basic parameters of the simulated wind farm (unit parameters, unit coordinates, i.e., wind turbine parameters, motor parameters, transformer parameters, power grid parameters, cable parameters and lengths of each part, primary wiring layout of the wind farm, wind speed, wind direction, wind turbine yaw angle, wind turbine pitch angle, etc.) are initialized by manually setting (i.e., manually setting the input quantities of the model, such as wind speed, wind direction, wind turbine coordinates, wind turbine yaw angle, wind turbine pitch angle, etc.) or by synchronizing data with the actual wind farm.

[0203] Based on the basic parameters and the wind farm rose diagram (i.e., the wind direction rose diagram and wind speed rose diagram of the wind farm throughout the year, which are the data that should exist in the actual wind farm), deep learning and intelligent algorithms are used to perform preprocessing of yaw in each wind direction.

[0204] For wind turbines deployed in simulated wind farms, corresponding unit-level models and data interfaces are established in MATLAB / Simulink.

[0205] For data such as internal cables, substations, and control stations in simulated wind farms, corresponding wind farm-level models and data interfaces were established in MATLAB / Simulink.

[0206] Initialize the platform communication software according to the required data volume.

[0207] The model is pre-compiled using the RT-OPAL software on the RT-LAB host computer.

[0208] Configure the software and algorithms within the SCADA server and advanced algorithm server based on wind farm data;

[0209] It should be noted that the SCADA server needs to pre-set the number, capacity, and circuits of wind turbines to match the different data received. The various algorithms in the advanced algorithm server also need to be pre-set according to the wind turbine coordinates, rotor diameter, wind direction, wind speed, etc. of the wind farm.

[0210] The pre-compiled model is uploaded to the RT-LAB lower computer via RT-OPAL to start real-time simulation. The real-time simulation platform for wind farm is formed by the linkage of SCADA server, RT-LAB, MATLAB, MODBUS communication software and various algorithm software.

[0211] Once the system is fully completed, real-time simulation can begin. Finally, wind speed and direction information will be input, and the necessary research data will be collected.

[0212] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A fully digital wind farm simulation method based on RTLAB, characterized in that, include: The basic parameters of the simulated wind farm are initialized by manually setting the input quantities of the model or synchronizing data with the actual wind farm. Based on the aforementioned basic parameters and the actual wind direction and wind speed rose diagram of the wind farm, yaw preprocessing for each wind direction is performed using deep learning and intelligent algorithms. For the wind turbines deployed in the simulated wind farm, a corresponding turbine-level model and corresponding data interface are established; then, for the data of cables, substations, and control stations inside the simulated wind farm, a corresponding wind farm-level model and corresponding data interface are established. The platform communication software is initialized according to the required data volume, and the overall model is pre-compiled. Based on actual wind farm data, the software and algorithms on the server are set up, and real-time simulation begins; The initialization settings include: obtaining various parameters through manual input or communication with the actual wind farm SCADA via a cloud server and inputting them into the wind farm-level data configuration file; classifying the various data and fixing the parameter data, which includes the wind turbine coordinates, wind turbine capacity, wind turbine rotor radius, wind turbine generator parameters, wind farm cable impedance, and constant parameters of the wind farm circuit settings; the parameter data also includes variable parameter data such as the wind direction and wind speed of the wind farm inflow. The communication software, wind farm wake algorithm software, and wind farm automatic power generation control algorithm are initialized according to the invariant parameters. The wind direction yaw preprocessing includes: performing staged preprocessing on the wind farm's annual rose diagram and manually configurable parts in the basic parameters, wherein by default, a stage is defined as 5 degrees of wind direction and 0.5 m / s of wind speed; using convolutional neural networks and DEA algorithms, the optimal active yaw control parameters of the wind farm under each different stage are determined and a lookup table file is generated; the lookup table file is loaded into the preliminary optimization lookup table module, and the corresponding data interface is set in the communication software to meet communication requirements; The establishment of the corresponding unit-level model and corresponding data interface includes: establishing the unit's aerodynamic model based on the wind turbine's mathematical model. Where: ρ is the air density, in kg / m³, taken as 1.25 kg / m³; R is the radius of the wind turbine blade, in m; v is the wind speed before entering the swept surface of the wind turbine, in m / s; C p The wind energy utilization coefficient of the wind turbine; Where, ω t β is the rotor speed, and β is the blade pitch angle; Then, the pitch model of the unit is established through the mathematical model of the pitch system: in, β ref This is the reference input for the pitch angle. T β Let be the time constant of the time-varying pitch system; the transfer function of the pitch system is: Where τ is the delay time of the variable pitch servo system; The following formula was used to correct the parameters against the wind turbine's factory data: Where λ is the tip speed ratio, β is the pitch angle, and c1-c8 are parameters obtained from the aerodynamic characteristics of the wind turbine. A MATLAB / Simulink wind turbine motor model is established based on the wind turbine motor parameters within the invariant parameters. A MATLAB / Simulink converter switching model and corresponding average value model are established based on the DC bus and IGBT parameters. A grid-side transformer model is established based on the wind turbine transformer parameters. Based on the DC bus parameters within the invariant parameters, establish constant speed control on the machine side, maximum power point tracking control on the machine side, reactive power control on the grid side, and constant power control of the variable pitch system; establish a data communication interface based on the data quantity and magnitude.

2. The fully digital wind farm simulation method based on RTLAB as described in claim 1, characterized in that, The establishment of the corresponding wind farm level model and corresponding data interface includes: establishing the data communication interface of the loop module; establishing the wind farm main transformer model and power grid model based on the booster station parameters in the invariant parameters; establishing the wind farm control module of the wind farm model; establishing the communication module of the wind farm model; and connecting different modules / models in the wind farm model with data / primary wiring through the Opt module of Rt-LAB to make the model conform to the RT-OPAL compilation standard. The initialization settings for the platform communication software include: setting the communication software and data interface according to the data volume, quantity, and accuracy requirements, including data format and corresponding data names; and creating corresponding tables for data collection and input.

3. The fully digital wind farm simulation method based on RTLAB as described in claim 2, characterized in that, The software and algorithms within the configuration server include: According to research requirements, the wake algorithm software for wind farms is initialized using corresponding wake models and wake deflection models. The Bastankhah wake model is represented as follows: Where △u is the wind speed loss caused by the wake at this point, U∞ is the inflow wind speed unaffected by the wake, CT is the wind turbine thrust coefficient, β here is the ratio of the wake diameter to the wind turbine rotor diameter when the wake returns to normal atmospheric pressure, d0 is the wind turbine rotor diameter, kW is the wake expansion coefficient, and r is the wind turbine rotor radius. The Shapiro wake deflection model is expressed as follows: Where v c Let θ be the lateral deflection velocity of the wake at a distance x behind the wind turbine rotor, d be the yaw angle, and d be the lateral deflection velocity. w Let y be the diameter of the wake at point x, and y be the diameter of the wake at point x. d Let x be the total deflection distance of the wake at point x; The inflow velocity of each wind turbine in the wind farm under the initial condition is calculated iteratively by using the wind speed and wind direction in the variable parameter data, the coordinates and radius of the wind turbine in the constant parameters, and the preliminary optimization lookup table. The inflow wind speed of each wind turbine under the initial condition is obtained and imported into the table file for pre-reading. Based on the preliminary optimization lookup table and the wind farm wake algorithm, further instruction allocation learning is carried out through convolutional neural networks for different grid commands to improve the automatic power generation control of wind farms under active yaw control.

4. The fully digital wind farm simulation method based on RTLAB as described in claim 3, characterized in that, The process of starting real-time simulation includes: when uploading the pre-compiled model to RT_LAB, segmenting the model according to its characteristics, allocating the computational load of multiple computing cores appropriately, starting the RT_LAB lower-level machine, loading the compiled model into the OP5600 HIL BOX, and starting the model loading program; performing real-time running, pausing, and data reading operations on the model through the user interface on the RT_LAB upper-level machine; starting the software and algorithms in the SCADA server, SCADA client, and advanced algorithm server, and checking the effectiveness of overall communication through MODBUS communication software.

5. A fully digital wind farm simulation system based on RTLAB, characterized in that, The system for implementing the RTLAB-based fully digital wind farm simulation method of claim 1 includes: The parameter input module is used to initialize various basic parameters of the simulated wind farm by manually setting the input quantities of the model or by synchronizing data with the actual wind farm. The wind direction and yaw preprocessing module is used to perform yaw preprocessing for each wind direction based on the basic parameters and the wind direction and wind speed rose diagram of the actual wind farm through deep learning and intelligent algorithms. The modeling module is used to create corresponding unit-level models and data interfaces for wind turbines deployed in the simulated wind farm; then, it creates corresponding wind farm-level models and data interfaces for data on cables, substations, and central control stations within the simulated wind farm. The model preprocessing module is used to initialize the platform communication software according to the required data volume and to perform overall model pre-compilation processing. The real-time simulation module is used to set the software and algorithms in the server based on actual wind farm data and start real-time simulation.

6. A computing device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the fully digital wind farm simulation method based on RTLAB as described in any one of claims 1 to 4.

7. A computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the RTLAB-based fully digital wind farm simulation method according to any one of claims 1 to 4.

Images