A real-time hardware-in-the-loop simulation method and system for wind turbine variable pitch control verification

Abstract

The application relates to the technical field of wind power generation control verification and real-time simulation, and discloses a real-time hardware-in-the-loop simulation method and system for wind turbine variable-pitch control verification. The method runs a wind turbine aeroelastic simulation system in a simulation server, and a physical simulation engine, a servo dynamics interface module and a modified dynamic link control library constitute a simulation side closed-loop interface; the dynamic link control library is built-in with an industrial communication client, and exchanges sensor data and control data with the servo dynamics interface module through a control interface array, and then establishes two-way data interaction with a communication server in an industrial programmable logic controller control cabinet through Ethernet. Real programmable logic controller hardware runs wind turbine control algorithms, completes variable-pitch proportional integral control, fault protection and state monitoring, and generates control commands which are returned to the simulation system through an Ethernet communication link, so that the wind turbine variable-pitch control strategy is verified under the participation of real control hardware.

Description

Technical Field

[0001] This invention relates to the fields of wind power generation control system testing, real-time simulation, and hardware-in-the-loop verification technology, and particularly to a real-time hardware-in-the-loop simulation method and system for verifying pitch control of large wind turbines. Background Technology

[0002] As wind turbine capacity continues to increase, the complexity of the turbine structure and the requirements for the control system also rise. Large wind turbines are generally difficult to use directly as research and experimental subjects for control strategy testing, while scaled-down models cannot fully represent the dynamic behavior of large turbines. Therefore, using high-resolution mathematical models of wind turbines to replace physical turbines and embedding real control hardware into the control closed loop has become a technical approach for researching control strategies for large wind turbines.

[0003] In wind turbine control, below rated operating conditions, electromagnetic torque control is primarily used to extract wind energy as much as possible; above rated operating conditions, collective pitch control is mainly used to maintain the speed and power within the rated range. Collective pitch control is typically based on the error between the generator speed and the rated speed, and after proportional-integral control and limiting processing, outputs a uniform pitch angle to the three pitch actuators. In a real programmable logic controller (PLC) control system, pitch control also needs to operate in conjunction with fault protection, condition monitoring, and electromagnetic control-related tasks to reflect the actual execution environment in the engineering control cabinet.

[0004] Pure digital simulation can be used to study wind turbine dynamics and control algorithms, but relying solely on it is insufficient to examine the performance of control algorithms in real control hardware. Hardware-in-the-loop systems (HILs) integrate mathematical models and physical control hardware within the same control closed loop, replacing large physical wind turbines with virtual ones, and enabling the real control hardware to perform low-level control and supervisory control tasks. Such systems must simultaneously meet the numerical solution requirements of the simulation model, the real-time task scheduling requirements of the control hardware, and the synchronization requirements between the simulation and control sides.

[0005] The difficulty of existing real-time hardware-in-the-loop verification lies in the fact that sensor data and control commands within the aeroelastic simulation program for wind turbines are typically exchanged through internal interfaces such as servo dynamics interface modules and dynamic link control libraries. However, the real programmable logic controller (PLC) control system needs to receive data and output control commands via industrial communication protocols. Relying solely on offline simulation or general signal interfaces makes it difficult to achieve stable bidirectional communication between the simulation side and the controller side without altering the actual control algorithm's operating environment. For pitch control verification based on PLCs, a real-time hardware-in-the-loop architecture that integrates the simulation program's source code-level control interface, industrial network communication, and the real control cabinet within the same closed loop is still needed. Summary of the Invention

[0006] The purpose of this invention is to provide a real-time hardware-in-the-loop simulation method and system for verifying pitch control of wind turbines, so that large wind turbines can be replaced by a wind turbine aeroelasticity simulation system running on a simulation server, and the real controller hardware in the industrial programmable logic controller control cabinet can perform pitch control, fault protection and status monitoring as part of the closed loop.

[0007] To achieve the above objectives, this invention runs a wind turbine aeroelasticity simulation system on a simulation server. The simulation system includes a physical simulation engine, a servo dynamics interface module, and a modified dynamic link control library. The physical simulation engine generates wind field, aerodynamic loads, and mechanical structure responses through the inflow wind module, aerodynamic module, and elastic dynamics module, and outputs sensor data to the servo dynamics interface module through internal data exchange. The servo dynamics interface module serves as the electrical and control interface layer, exchanging data with the modified dynamic link control library through a control interface array. The modified dynamic link control library contains control logic and an industrial communication client.

[0008] The data interaction method of this invention is as follows: a real controller hardware is deployed in the industrial programmable logic controller control cabinet, and the real controller hardware runs an industrial communication server and a real wind turbine control algorithm; the industrial communication server provides a data exposure interface and a publish-subscribe mechanism, and exchanges internal data with the real wind turbine control algorithm; the simulation server is connected to the industrial control cabinet through an Ethernet network, and the industrial communication client on the simulation side and the industrial communication server on the controller side conduct bidirectional communication, so that the sensor data on the simulation side is transmitted to the controller side, and the control commands on the controller side are sent back to the simulation system.

[0009] This invention also controls the closed-loop timing based on the operating conditions of real-time hardware-in-the-loop verification: the simulation system advances with a constant integral step size; the modified dynamic link control library reads sensor data and writes control data through the control interface array in each control interaction cycle; the industrial communication client and industrial communication server complete data publishing, subscription, or reading and writing according to a preset communication cycle; and the real wind turbine control algorithm completes pitch control calculations within the controller and outputs control commands. Through this timing relationship, the simulation server, network communication, and controller cabinet can be coordinated while meeting real-time requirements.

[0010] Compared to pure digital simulation, this invention involves real controller hardware, industrial communication protocols, and real wind turbine control algorithms in a closed-loop process, making it more suitable for studying large wind turbine pitch control systems. Compared to systems that only run the controller within the simulation software, this invention modifies the dynamic link control library and embeds an industrial communication client, connecting the simulation system's servo dynamics interface module and control interface array to the controller-side industrial communication server. This enables bidirectional exchange of sensor data from the simulation side and control commands from the controller side via an Ethernet network. The real control algorithm performs pitch proportional-integral control, fault protection, and status monitoring within the controller, facilitating the verification of the control algorithm's real-time performance and closed-loop control effectiveness in an industrial controller environment. Attached Figure Description

[0011] Figure 1 This is a flowchart of a real-time hardware-in-the-loop simulation method for verifying pitch control of wind turbines.

[0012] Figure 2 This is a schematic diagram of the overall architecture of a real-time hardware-in-the-loop simulation system for verifying pitch control of a wind turbine.

[0013] Figure 3 This is a schematic diagram of the synchronization timing for a real-time hardware-in-the-loop simulation used for verifying the pitch control of a wind turbine. Detailed Implementation

[0014] The present invention will be further described below with reference to the accompanying drawings. This embodiment describes the implementation of the present invention based on a hardware-in-the-loop architecture consisting of a wind turbine aeroelasticity simulation system, a modified dynamic link control library, an industrial communication link, and an industrial programmable logic controller control cabinet. Specific software libraries, communication cycles, and wind turbine parameters are used to illustrate the implementation conditions and are not intended to limit the scope of protection of the present invention.

[0015] like Figure 1 As shown, in step S101, an aeroelastic model of the wind turbine is established. This model is used to replace the physical wind turbine, and the physical simulation engine performs numerical solutions for the wind turbine's aerodynamics, elasticity, transmission, and control interfaces. The physical simulation engine calls the inflow wind module to generate wind input, calls the aerodynamic module to perform aerodynamic simulation, calls the elastic dynamics module to perform mechanical transmission and structural dynamics simulation, and sends the sensor data generated by the above modules to the servo dynamics interface module through internal data exchange.

[0016] Step S102: Configure the simulation server and simulation system. The simulation server provides the operating environment for the wind turbine aeroelasticity simulation system; the simulation system includes a physical simulation engine, a servo dynamics interface module, and a modified dynamic link control library. The servo dynamics interface module, as the electrical and control interface layer, receives sensor data output by the physical simulation engine and exchanges control interface data with the modified dynamic link control library through the control interface array.

[0017] Step S103: Configure the modified dynamic link control library. The modified dynamic link control library includes control logic and an industrial communication client. The control logic is used to maintain the calling relationship of the simulation control interface, and the industrial communication client is used to map simulation-side variables to communication variables; the two interact through the control interface array, enabling sensor data from within the simulation system to be transmitted to the network side, and enabling control commands received on the network side to be written back to the servo dynamics interface module.

[0018] Step S104: Configure the industrial programmable logic controller (PLC) control cabinet. The industrial control cabinet houses the actual controller hardware, which runs the industrial communication server and the actual wind turbine control algorithm. The industrial communication server is configured with a data exposure interface and a publish-subscribe mechanism to receive sensor data from the simulation side and provide control commands from the controller side. The actual wind turbine control algorithm exchanges data with the industrial communication server within the controller and serves as the control core of the actual electromagnetic control system.

[0019] Step S105: Establish an Ethernet industrial communication link. The simulation server and the industrial control cabinet are connected via an Ethernet network. The industrial communication client in the modified dynamic link control library establishes bidirectional communication with the industrial communication server in the actual controller hardware. During the communication process, sensor data from the simulation side is written to or published by the industrial communication client to the industrial communication server on the controller side. Control commands from the controller side are provided by the industrial communication server and read or subscribed to by the industrial communication client.

[0020] Step S106: Execute the real wind turbine control algorithm. The real wind turbine control algorithm receives sensor data such as generator speed, pitch angle, load status, and operating status from the simulation system; under rated operating conditions, the proportional-integral control (PIC) generates a unified pitch angle command based on the error between the generator speed and the rated speed, after PIC and limiting processing; fault protection triggers protection logic based on the operating status; and status monitoring monitors variables on both the simulation side and the controller side.

[0021] Step S107: The control commands are transmitted back and the simulation continues. The control commands generated by the real wind turbine control algorithm are written into the industrial communication server within the controller. The industrial communication client in the modified dynamic link control library receives the control commands via an Ethernet communication link and writes them into the servo dynamics interface module through the control interface array. In the next simulation step, the simulation system reads the control commands, updates the pitch control input, and continues to calculate the wind turbine's aeroelastic response.

[0022] Step S108: Output verification results. Through the above architecture, the simulation system in the simulation server replaces the actual wind turbine, the actual controller hardware in the industrial control cabinet executes the wind turbine control algorithm, and the Ethernet industrial communication link completes the bidirectional exchange of sensor data and control commands between the simulation side and the controller side. The verification results include pitch control commands, fault protection status, status monitoring data, and the wind turbine response output by the simulation system, used to evaluate the real-time performance and control effect of the actual controller's control algorithm in the hardware-in-the-loop closed loop.

[0023] like Figure 2 As shown, the overall system architecture of this invention includes a simulation server, an Ethernet network, and an industrial programmable logic controller (PLC) control cabinet. The simulation server runs a wind turbine aeroelasticity simulation system, which includes a physical simulation engine, a servo dynamics interface module, and a modified dynamic link control library. The physical simulation engine internally includes simulation modules for inflow wind, aerodynamics, and elasticity, and provides sensor data to the servo dynamics interface module through internal data exchange. The servo dynamics interface module exchanges data with the modified PLC through a control interface array. The modified PLC includes control logic and an industrial communication client. The Ethernet network serves as the industrial communication link, providing bidirectional communication between the industrial communication client on the simulation side and the industrial communication server on the controller side. The industrial control cabinet houses the actual controller hardware, which includes an industrial communication server and a real wind turbine control algorithm. The industrial communication server has a data exposure interface and a publish-subscribe mechanism. The real wind turbine control algorithm includes pitch proportional-integral control, fault protection, and status monitoring, and outputs control commands to the real electromagnetic control system.

[0024] like Figure 3 As shown, the synchronization timing of this invention includes a simulation side, an industrial communication side, and a controller control side. The simulation side progresses with a constant simulation step size and forms variables to be exchanged in the servo dynamics interface module; the industrial communication side transmits sensor data and control commands between the industrial communication client and the industrial communication server at a preset communication cycle; the controller control side executes the actual wind turbine control algorithm after receiving the sensor data and outputs control commands before the next control interaction cycle. The hardware maintains real-time operation in a closed-loop manner as long as the sum of the simulation calculation time, industrial communication time, and controller calculation time is less than the control interaction cycle.

Claims

1. A real-time hardware-in-the-loop simulation method for verifying pitch control of wind turbines, characterized in that, Includes the following steps: Run wind turbine aeroelasticity simulation software in a simulation server to build a simulation-side closed-loop system that includes a physical simulation engine, a servo dynamics interface module, and control middleware. The physical simulation engine is used to generate simulated wind field, aerodynamic load and wind turbine structural dynamic response data, and outputs the data as virtual sensor signals to the servo dynamics interface module. An industrial communication client is integrated into the control middleware. The internal variables of the servo dynamics interface module are converted into industrial communication protocol data through the data mapping interface and sent to the real industrial programmable logic controller (PLC) via industrial Ethernet. The wind turbine control algorithm is run in the real industrial PLC, receiving the virtual sensor signals and executing pitch control logic to generate pitch control commands. The real industrial PLC transmits the pitch control command back to the control middleware via the industrial Ethernet and writes it into the servo dynamics interface module, driving the physical simulation engine to advance to the next simulation step, thereby forming a closed-loop verification loop containing real control hardware.

2. The method according to claim 1, characterized in that, The control middleware is a dynamic link library or static link library obtained by extending the functionality of the native control interface library. The extension includes: retaining the calling timing of the native control algorithm interface, creating an independent communication thread in the background to run the industrial communication client, and realizing parallel interaction between simulation data and network data.

3. The method according to claim 1, characterized in that, The physical simulation engine includes an inflow wind module, an aerodynamic module, and an elastic dynamics module. The method further includes: calculating the load torque of the pitch actuator in real time through the elastic dynamics module and outputting it as a load feedback signal to the real industrial PLC to simulate the aeroelastic coupling characteristics of a real wind turbine.

4. The method according to claim 1, characterized in that, The industrial communication client interacts with the actual industrial PLC using a publish / subscribe mode or a master-slave polling mode, and the communication protocols include OPC UA, EtherCAT, or Modbus TCP / IP.

5. The method according to claim 1, characterized in that, The simulation system operates with a constant integral step size, and ensures that the sum of the simulation calculation time, industrial communication time, and controller calculation time is less than the preset control interaction cycle.

6. The method according to claim 5, characterized in that, The time margin of the preset control interaction cycle satisfies Δt=T0-m·trt-tcom-tplc, where T0 is the preset control interaction cycle, m is the number of simulation calculation steps in one control interaction cycle, trt is the actual calculation time of a single simulation time step, tcom is the industrial communication time, and tplc is the controller calculation time.

7. The method according to claim 1, characterized in that, The pitch control task uses the generator speed measurement value and the rated speed set value to form a control error, and generates a unified pitch angle command to be transmitted to the three pitch actuators through proportional-integral control with gain scheduling and anti-saturation structure.

8. The method according to claim 1, characterized in that, The real wind turbine control algorithm also includes fault protection and status monitoring. The fault protection triggers protection logic based on the operating status received by the controller. The status monitoring monitors the sensor data output by the simulation system and the control commands from the controller.

9. A real-time hardware-in-the-loop simulation system for verifying pitch control of wind turbines, characterized in that, include: A simulation server is configured to provide the operating environment for a wind turbine aeroelasticity simulation system; the simulation system is deployed on the simulation server and includes a physical simulation engine, a servo dynamics interface module, and a modified dynamic link control library. The physical simulation engine is configured to generate wind turbine sensor data based on wind field generation, aerodynamic simulation, and structural dynamics simulation. The servo dynamics interface module is configured as an electrical and control interface layer and exchanges data with the modified dynamic link control library through a control interface array. The modified dynamic link control library is configured to run control logic and an industrial communication client. The Ethernet network is configured to provide bidirectional communication between the industrial communication client and the industrial communication server on the controller side. The industrial programmable logic controller control cabinet is configured to house the real controller hardware and run the real wind turbine control algorithm, so that the real wind turbine control algorithm generates control commands based on the simulation-side sensor data and sends them back to the simulation system.

10. The system according to claim 9, characterized in that, The real controller hardware includes an industrial communication server, which is equipped with a data exposure interface and a publish / subscribe mechanism; the real wind turbine control algorithm includes a pitch proportional-integral control module, a fault protection module, and a status monitoring module. The real wind turbine control algorithm exchanges data with the industrial communication server within the controller and outputs control commands to the real electromagnetic control system.

Images