Wind farm dynamic sector management optimization method and system based on digital twin

Abstract

The present invention relates to a method and system for optimizing dynamic sector management of wind farms based on digital twins. The method comprises: establishing a weather research and forecasting model that takes into account the refined surface conditions of the wind farm; establishing a high-end dynamic aerodynamic simulation and virtual blade coupling model for the wind farm wake, simulating and calculating the wind farm parameters at the location where the wind turbine is placed, and verifying the calculated wake details through scanning laser wind measurement to obtain an accurate wake model for wind farm control; an impeller online health monitoring and diagnosis system based on machine hearing, which obtains wind turbine health status information and participates in wind turbine control optimization when the unit is in a healthy state; according to the influence of the wake, a single wind farm adopts adaptive optimization control and a wind farm group adopts dynamic wake control based on hearing. The present invention, with the basic features of knowing the wind, knowing the field, knowing the machine, and knowing the group control, realizes dynamic sector management of wind farms, effectively improving the intelligent monitoring of wind turbine impellers and the profitability of power generation in offshore wind farms.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and in particular to a wind farm dynamic sector management optimization method and system based on digital twins. Background Art

[0002] Offshore wind energy resources are abundant, with high wind speeds, low turbulence, and high power generation capacity. Furthermore, offshore wind power development does not occupy land resources, all of which provide favorable conditions for the development of offshore wind power. However, due to its large installed capacity, long offshore distance, harsh operating environment, and difficult inspection and maintenance, the aerodynamic coupling caused by the wind turbine wake interference effect reduces the total wind energy captured by the wind farm and increases turbine load fluctuations and vibration, which can easily cause component damage and shorten turbine life. Offshore wind farms require higher levels of intelligent monitoring and equipment reliability, and traditional onshore wind farm technical solutions no longer meet these operational requirements. Furthermore, wind turbine impellers are prone to cumulative damage over long service lives due to material aging, corrosion, fatigue, and long-term operation in harsh outdoor environments. This can lead to deformation of the blade trailing edge, cracking, or even fracture, resulting in the collapse of the entire turbine, causing significant economic losses and even casualties. Therefore, online monitoring and health assessment of wind turbine impellers are essential. Existing technologies, such as drone inspections, are difficult to implement due to the long offshore distance and harsh operating environment of offshore wind farms. Summary of the Invention

[0003] The purpose of the present invention is to provide a wind farm dynamic sector management optimization method and system based on digital twins, so as to achieve the purpose of improving quality and efficiency, reducing the cost of electricity, unmanned operation, less manpower on duty, and intelligent operation and maintenance.

[0004] The present invention provides a wind farm dynamic sector management optimization method based on digital twins, comprising:

[0005] Step 1: Establish a mesoscale weather research and forecast model that takes into account the surface conditions of the wind farm, and obtain downscaled and refined numerical weather and wind speed and direction forecast data for the wind farm;

[0006] Step 2: Using the obtained downscaled and refined numerical weather and wind speed and direction forecast data for the wind farm as initial and boundary conditions, a high-end dynamic aerodynamic simulation and virtual blade coupling model for the wind farm wake is established based on the wind farm terrain and unit layout. The wind farm parameters at the wind turbine placement location are simulated and calculated to obtain numerical calculation results of the wind farm wind turbine and terrain wakes. The calculated wake details are verified through scanning laser wind measurement, resulting in a precise wake model for wind farm control and wake visualization.

[0007] Step 3: The impeller online health monitoring and diagnosis system based on machine hearing obtains the health status information of the fan impeller, and proceeds to step 4 if the unit is in a healthy state;

[0008] Step 4: Analyze the wake interference status of wind turbines in the wind farm in real time based on the wake model. When the unit is not affected by the wake, a single-machine adaptive optimization control strategy is used to achieve maximum power. When the wake interference effect is detected in combination with the machine hearing device, a field group control strategy is implemented based on the flow field details provided by the wake model. During the implementation of the field group control strategy, the sound wave data obtained by the machine hearing device is used to correct the wake model to achieve maximum power for the entire field.

[0009] Furthermore, the field group control strategy in step 4 includes:

[0010] A numerical model simulation analysis of the wind farm is performed based on the load status of the wind turbines. If the simulation analysis results meet expectations, control instructions are issued to the relevant wind turbines.

[0011] Furthermore, the step 4 includes:

[0012] For onshore wind farms: Using one year of historical data showing healthy turbines and equipment temperatures within normal operating ranges, analyze the field's power output, noise, and vibration data at the same wind speed and different wind directions. Develop a numerical model and determine the wind direction corresponding to the sector. Based on the terrain and turbine layout, determine whether the sector formation is due to the wake of upstream wind turbines or terrain influences, including:

[0013] When the noise and vibration data model shows that the wind turbine is affected by the wake, it is determined that the wind turbine begins to be affected by the wake;

[0014] In the case where the upstream wind turbines generate wake influence to form a sector, the wake model is corrected according to the noise sound wave and vibration data model, and the wake influence and sector is determined to be fully covered or partially covered by the wind turbines according to the wake model, including:

[0015] When the wind speed corresponds to the wind turbine operating state in zone 2: if the sector influence range fully covers the wind turbine generator set, speed control and yaw control are adopted; if the sector influence range partially covers the wind turbine generator set, yaw control is adopted;

[0016] When the wind speed corresponds to the wind turbine operating state in the third zone: when the sector influence range is the full coverage of the wind turbine generator set, pitch control and yaw control are adopted; when the sector influence range is the partial coverage of the wind turbine generator set, yaw control is adopted;

[0017] In the case of a sector-shaped area generated by terrain, the wake influence and whether the sector is fully or partially covered by the wind turbine generator set are determined based on the characteristics of noise, sound waves and vibration data, including:

[0018] The affected wind turbines were shut down to prevent excessive load during operation, which would increase the failure rate and reduce the life of the units.

[0019] For offshore wind farms: Using one year of historical data showing healthy turbines and equipment temperatures within normal operating ranges, analyze the overall power output, noise, and vibration data at the same wind speed and different wind directions. Develop a numerical model to determine the wind direction corresponding to the wake pit, including:

[0020] When the noise and vibration data model shows that the wind turbine is affected by the wake, it is determined that the wind turbine begins to be affected by the wake;

[0021] In the case of wake impacts from upstream wind turbines, the wake model is calibrated based on the noise and vibration numerical model to determine whether the wake impact fully or partially covers the wind turbines, including:

[0022] When the wind speed corresponds to the operating state of the wind turbine in zone 2: when the sector range fully covers the wind turbine generator set, speed control and yaw control are adopted; when the sector range partially covers the wind turbine generator set, yaw control is adopted.

[0023] When the wind speed corresponds to the operating state of the wind turbine in the third zone: the sector influence range is the full coverage of the wind turbine generator set, and pitch control and yaw control are adopted; when the sector influence range is the partial coverage of the wind turbine generator set, yaw control is adopted.

[0024] The present invention also provides a wind farm dynamic sector management optimization system based on digital twins, comprising:

[0025] The wind knowledge module is used to establish a mesoscale weather research and forecast model that takes into account the surface conditions of wind farms, and obtain downscaled and refined numerical weather and wind speed and direction forecast data for wind farms;

[0026] The field knowledge module is used to use the acquired downscaled and refined numerical weather and wind speed and direction forecast data for the wind farm as initial and boundary conditions. Based on the wind farm terrain and unit layout, it establishes a high-end dynamic aerodynamic simulation and virtual blade coupling model for the wind farm wake. It simulates and calculates the wind farm parameters at the wind turbine placement location, obtains the numerical calculation results of the wind farm wind turbine and terrain wake, and verifies the calculated wake details through scanning laser wind measurement. It obtains a precise wake model for wind farm control and realizes wake visualization.

[0027] The Zhiji module is used in the impeller online health monitoring and diagnosis system based on machine hearing to obtain the health status information of the fan impeller and participate in the fan control optimization when the unit is in a healthy state;

[0028] The control module is used to detect and analyze the wake interference status of wind turbines in the wind farm in real time based on the wake model. When the unit is not affected by the wake, a single-machine adaptive optimization control strategy is adopted to achieve maximum power. When the wake interference effect is detected in combination with the machine hearing device, a field group control strategy is implemented according to the flow field details provided by the wake model. During the implementation of the field group control strategy, the acoustic wave data obtained by the machine hearing device is used to correct the wake model to achieve maximum power for the entire field.

[0029] Furthermore, the control module performs a wind farm numerical model simulation analysis according to the load status of the wind turbines, and issues a control instruction to the relevant wind turbines if the simulation analysis results meet expectations.

[0030] Through the above solution, through the digital twin-based wind farm dynamic sector management optimization method and system, with the basic characteristics of knowing the wind, knowing the field, knowing the machine, and group control, dynamic sector management of wind farms is realized, effectively improving the intelligent monitoring of wind turbine blades and the profitability of power generation in offshore wind farms.

[0031] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and implement it according to the contents of the specification, the following is a detailed description of the preferred embodiments of the present invention with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a flow chart of the wind farm dynamic sector management optimization method based on digital twin of the present invention;

[0033] Figure 2 This is a structural block diagram of the wind farm dynamic sector management optimization system based on digital twins of the present invention;

[0034] Figure 3 This is a flow chart of wind and field knowledge in one embodiment of the present invention. DETAILED DESCRIPTION

[0035] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0036] Ginseng Figure 1 As shown, this embodiment provides a wind farm dynamic sector management optimization method based on digital twins, including:

[0037] Step S1: Build a mesoscale weather research and forecasting model (a downscaled WRF wind farm model) that takes into account wind farm surface conditions, and obtain downscaled, refined numerical weather and wind speed and direction forecast data for the wind farm. Embedding the numerical weather research and forecasting (WRF) downscaling model within wind turbine and wake models enables the design of a new generation of mesoscale numerical weather and power forecasting systems, significantly improving forecast accuracy and providing initial and boundary conditions for subsequent field analysis.

[0038] In step S2, the obtained downscaled and refined numerical weather and wind speed and direction forecast data for the wind farm are used as initial and boundary conditions. Based on the wind farm terrain and unit layout, a high-end dynamic aerodynamic simulation and virtual blade coupling model of the wind farm wake is established, and the wind field parameters at the wind turbine placement location are simulated and calculated to obtain the numerical calculation results of the wind farm wind turbine and terrain wake. The calculated wake details are verified by scanning laser wind measurement to obtain a precise wake model for wind farm control and realize wake visualization.

[0039] In step S3, the machine-auditory impeller online health monitoring and diagnostic system acquires information about the wind turbine's impeller health status. If the unit is healthy, the system proceeds to step S4. This step significantly improves the targetedness and efficiency of on-site maintenance work; it accurately predicts faults and significantly reduces unit downtime.

[0040] Step S4: The wake interference status of the wind turbines in the wind farm is detected and analyzed in real time according to the wake model. When the unit is not affected by the wake, a single-machine adaptive optimization control strategy (including pitch control, speed control, and yaw control) is adopted to achieve maximum power. When the wake interference effect is detected in combination with the machine hearing device, a field group control strategy is implemented according to the flow field details provided by the wake model. During the implementation of the field group control strategy, the sound wave data obtained by the machine hearing device is used to correct the wake model to achieve maximum power for the entire field.

[0041] When a wind farm is free of wake influence, a single-unit adaptive optimization control system is used to maximize power. When a turbine within the wind farm is affected by wake, acoustic wave data captured from the blades is used to determine whether it is affected by wake and to calibrate the wake model in step S2, obtaining more accurate wake information for cluster control. Cluster control uses the turbine pitch angle, rotor speed, and nacelle yaw angle as inputs for real-time online control.

[0042] Without considering wake turbulence, this embodiment employs adaptive optimization control, which significantly improves turbine power generation efficiency, reduces turbine load, reduces vibration, and enhances turbine safety. Wind farm cluster control employs auditory-based dynamic wake control, using acoustic wave data as a basis for determining whether a wind farm is affected by wake turbulence and used to calibrate the wake model. This significantly improves model accuracy and makes the control process more accurate. This significantly improves control accuracy and power generation efficiency across the entire cluster, significantly reduces load on downwind turbines affected by wake turbulence, and significantly enhances safety.

[0043] This digital twin-based wind farm dynamic sector management optimization method, with its basic features of wind knowledge, field knowledge, machine knowledge, and group control, realizes dynamic sector management of wind farms and effectively improves the intelligent monitoring of wind turbine blades and the profitability of power generation in offshore wind farms.

[0044] In this embodiment, the field group control strategy in step 4 includes:

[0045] A numerical model simulation analysis of the wind farm is performed based on the load status of the wind turbines. If the simulation analysis results meet expectations, control instructions are issued to the relevant wind turbines.

[0046] In this embodiment, step S4 includes:

[0047] 1. When the unit is not affected by wake turbulence, the single-machine adaptive optimization control strategy is used to achieve maximum power.

[0048] 2. For onshore wind farms: Use one year of historical data showing healthy turbines and equipment temperatures within normal operating ranges to analyze the full-field power output, noise, and vibration data at the same wind speed and in different wind directions. Build a numerical model and determine the wind direction corresponding to the sector. Based on the terrain and turbine layout analysis, determine whether the sector formation is due to the wake of upstream wind turbines or the influence of terrain.

[0049] 2.1. When the noise and vibration data model shows that the wind turbine is affected by the wake, it is determined that the wind turbine begins to be affected by the wake.

[0050] 2.2. When the upstream wind turbines generate wake influence and form sectors, the wake model is calibrated based on the noise sound wave and vibration data model, and the wake influence and sector are determined to be fully covered or partially covered by the wind turbines based on the wake model.

[0051] 2.2.1. When the wind speed corresponds to the wind turbine operating state in zone 2: if the sector influence range fully covers the wind turbine generator set, variable speed control and yaw control are adopted; if the sector influence range partially covers the wind turbine generator set, yaw control is adopted.

[0052] 2.2.2. When the wind speed corresponds to the wind turbine operating state in zone three: if the sector influence range fully covers the wind turbine generator set, pitch control and yaw control are adopted; if the sector influence range partially covers the wind turbine generator set, yaw control is adopted.

[0053] 2.3. In the case of a sector-shaped area generated by terrain, the wake influence and whether the sector is fully or partially covered by the wind turbine generator set are determined based on the characteristics of noise sound waves and vibration data.

[0054] 2.3.1. Affected wind turbines shall be shut down to prevent excessive load during operation, which may increase the failure rate and reduce the life of the unit.

[0055] 3. For offshore wind farms: Use one year of historical data showing healthy turbines and equipment temperatures within normal operating ranges to analyze the full-field power output, noise, and vibration data at the same wind speed and in different wind directions, establish a numerical model, and determine the wind direction corresponding to the wake pit.

[0056] 3.1. When the noise and vibration data model shows that the wind turbine is affected by the wake, it is determined that the wind turbine begins to be affected by the wake.

[0057] 3.2. In the case of wake impact from upstream wind turbines, the wake model is calibrated based on the noise and vibration numerical model to determine whether the wake impact fully or partially covers the wind turbines.

[0058] 3.2.1. When the wind speed corresponds to the wind turbine operating state in zone 2: if the sector range fully covers the wind turbine generator set, speed control and yaw control are adopted; if the sector range partially covers the wind turbine generator set, yaw control is adopted.

[0059] 3.2.2. When the wind speed corresponds to the wind turbine operating state in zone 3: if the sector influence range fully covers the wind turbine generator set, pitch control and yaw control are adopted; if the sector influence range partially covers the wind turbine generator set, yaw control is adopted.

[0060] Ginseng Figure 2 、 Figure 3 As shown, the present invention also provides a wind farm dynamic sector management optimization system based on digital twin, including:

[0061] Wind Knowledge Module 10 (Wind Knowledge System) is used to establish a mesoscale weather research and forecast model (downscaled WRF wind field model) that takes into account the surface conditions of the wind farm, and obtain downscaled and refined numerical weather and wind speed and direction forecast data for the wind farm;

[0062] The field knowledge module 20 (field knowledge system) is used to use the obtained downscaled and refined numerical weather and wind speed and direction forecast data of the wind farm as the initial and boundary conditions, establish a high-end dynamic aerodynamic simulation and virtual blade coupling model of the wind farm wake based on the wind farm terrain and unit layout, simulate and calculate the wind field parameters of the wind turbine placement location, obtain the numerical calculation results of the wind farm wind turbine and terrain wake, and verify the calculated wake details through scanning laser wind measurement, obtain a precise wake model for wind farm control, and realize wake visualization.

[0063] The machine-aware module 30 (the machine-aware system) is used for the impeller online health monitoring and diagnosis system based on machine hearing. It obtains information on the health status of the wind turbine impeller and proceeds to step S4 if the unit is healthy. This step significantly helps improve the targetedness and efficiency of on-site operation and maintenance work; it has high fault prediction accuracy and can significantly reduce the number of unit failure hours.

[0064] Control module 40 (dynamic sector management system) is used to detect and analyze the wake interference status of wind turbines in the wind farm in real time based on the wake model. When the unit is not affected by the wake, a single-machine adaptive optimization control strategy (including pitch, speed, and yaw) is adopted to achieve maximum power. When the wake interference effect is detected in combination with the machine hearing device, a field group control strategy is implemented according to the flow field details provided by the wake model. During the implementation of the field group control strategy, the sound wave data obtained by the machine hearing device is used to correct the wake model to achieve maximum power for the entire field.

[0065] This digital twin-based wind farm dynamic sector management optimization system, with its basic features of knowing the wind, knowing the field, knowing the machine, and group control, realizes dynamic sector management of wind farms and effectively improves the intelligent monitoring of wind turbine blades and the profitability of power generation in offshore wind farms.

[0066] In this embodiment, the control module performs a wind farm numerical model simulation analysis based on the wind turbine load state, and issues a control instruction to the relevant wind turbine if the simulation analysis result meets expectations.

[0067] This digital twin-based wind farm dynamic sector management optimization system, with its basic features of knowing the wind, knowing the field, knowing the machine, and group control, realizes dynamic sector management of wind farms and effectively improves the intelligent monitoring of wind turbine blades and the profitability of power generation in offshore wind farms.

[0068] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A wind farm dynamic sector management optimization method based on digital twin, characterized in that: include: Step 1: Establish a mesoscale weather research and forecast model that takes into account the surface conditions of the wind farm, and obtain downscaled and refined numerical weather and wind speed and direction forecast data for the wind farm; Step 2: Using the obtained downscaled and refined numerical weather and wind speed and direction forecast data for the wind farm as initial and boundary conditions, a high-end dynamic aerodynamic simulation and virtual blade coupling model for the wind farm wake is established based on the wind farm terrain and unit layout. The wind farm parameters at the wind turbine placement location are simulated and calculated to obtain numerical calculation results of the wind farm wind turbine and terrain wakes. The calculated wake details are verified through scanning laser wind measurement, resulting in a precise wake model for wind farm control and wake visualization. Step 3: The impeller online health monitoring and diagnosis system based on machine hearing obtains the health status information of the fan impeller, and proceeds to step 4 if the unit is in a healthy state; Step 4: Analyze the wake interference status of wind turbines in the wind farm in real time based on the wake model. When the unit is not affected by the wake, a single-machine adaptive optimization control strategy is used to achieve maximum power. When the wake interference effect is detected in combination with the machine hearing device, a field group control strategy is implemented based on the flow field details provided by the wake model. During the implementation of the field group control strategy, the sound wave data obtained by the machine hearing device is used to correct the wake model to achieve maximum power for the entire field.

2. The wind farm dynamic sector management optimization method based on digital twin according to claim 1 is characterized in that: The field group control strategy described in step 4 includes: A numerical model simulation analysis of the wind farm is performed based on the load status of the wind turbines. If the simulation analysis results meet expectations, control instructions are issued to the relevant wind turbines.

3. The wind farm dynamic sector management optimization method based on digital twin according to claim 1 is characterized in that: The step 4 comprises: For onshore wind farms: Using one year of historical data showing healthy turbines and equipment temperatures within normal operating ranges, analyze the field's power output, noise, and vibration data at the same wind speed and different wind directions. Develop a numerical model and determine the wind direction corresponding to the sector. Based on the terrain and turbine layout, determine whether the sector formation is due to the wake of upstream wind turbines or terrain influences, including: When the noise and vibration data model shows that the wind turbine is affected by the wake, it is determined that the wind turbine begins to be affected by the wake; In the case where the upstream wind turbines generate wake influence to form a sector, the wake model is corrected according to the noise sound wave and vibration data model, and the wake influence and sector is determined to be fully covered or partially covered by the wind turbines according to the wake model, including: When the wind speed corresponds to the wind turbine operating state in zone 2: if the sector influence range fully covers the wind turbine generator set, speed control and yaw control are adopted; if the sector influence range partially covers the wind turbine generator set, yaw control is adopted; When the wind speed corresponds to the wind turbine operating state in the third zone: when the sector influence range is the full coverage of the wind turbine generator set, pitch control and yaw control are adopted; when the sector influence range is the partial coverage of the wind turbine generator set, yaw control is adopted; In the case of a sector-shaped area generated by terrain, the wake influence and whether the sector is fully or partially covered by the wind turbine generator set are determined based on the characteristics of noise, sound waves and vibration data, including: The affected wind turbines were shut down to prevent excessive load during operation, which would increase the failure rate and reduce the life of the units. For offshore wind farms: Using one year of historical data showing healthy turbines and equipment temperatures within normal operating ranges, analyze the overall power output, noise, and vibration data at the same wind speed and different wind directions. Develop a numerical model to determine the wind direction corresponding to the wake pit, including: When the noise and vibration data model shows that the wind turbine is affected by the wake, it is determined that the wind turbine begins to be affected by the wake; In the case of wake impacts from upstream wind turbines, the wake model is calibrated based on the noise and vibration numerical model to determine whether the wake impact fully or partially covers the wind turbines, including: When the wind speed corresponds to the wind turbine operating state in zone 2: the sector range is full coverage of the wind turbine generator set, and speed control and yaw control are adopted; when the sector range is partial coverage of the wind turbine generator set, yaw control is adopted; When the wind speed corresponds to the operating state of the wind turbine in the third zone: the sector influence range is the full coverage of the wind turbine generator set, and pitch control and yaw control are adopted; when the sector influence range is the partial coverage of the wind turbine generator set, yaw control is adopted.

4. A wind farm dynamic sector management optimization system based on digital twins, characterized by: include: The wind knowledge module is used to establish a mesoscale weather research and forecast model that takes into account the surface conditions of wind farms, and obtain downscaled and refined numerical weather and wind speed and direction forecast data for wind farms; The field knowledge module is used to use the acquired downscaled and refined numerical weather and wind speed and direction forecast data for the wind farm as initial and boundary conditions. Based on the wind farm terrain and unit layout, it establishes a high-end dynamic aerodynamic simulation and virtual blade coupling model for the wind farm wake. It simulates and calculates the wind farm parameters at the wind turbine placement location, obtains the numerical calculation results of the wind farm wind turbine and terrain wake, and verifies the calculated wake details through scanning laser wind measurement. It obtains a precise wake model for wind farm control and realizes wake visualization. The Zhiji module is used in the impeller online health monitoring and diagnosis system based on machine hearing to obtain the health status information of the fan impeller and participate in the fan control optimization when the unit is in a healthy state; The control module is used to detect and analyze the wake interference status of wind turbines in the wind farm in real time based on the wake model. When the unit is not affected by the wake, a single-machine adaptive optimization control strategy is adopted to achieve maximum power. When the wake interference effect is detected in combination with the machine hearing device, a field group control strategy is implemented according to the flow field details provided by the wake model. During the implementation of the field group control strategy, the acoustic wave data obtained by the machine hearing device is used to correct the wake model to achieve maximum power for the entire field.

5. The wind farm dynamic sector management optimization system based on digital twin according to claim 4 is characterized in that: The control module performs a wind farm numerical model simulation analysis based on the wind turbine load state, and issues a control instruction to the relevant wind turbine if the simulation analysis result meets expectations.

Images