Wind Farm Control Perturbation Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large-scale wind farms face challenges in optimizing power generation output due to the wake effect, where rear turbines produce smaller outputs compared to front turbines, and existing control methods struggle to efficiently adjust settings for multiple turbines to achieve optimal performance.
Innovation Solution
A method involving perturbation of control settings for wind turbines, followed by evaluation and updating based on gradient analysis, using filtered data from turbines meeting specific wind direction and velocity conditions to optimize overall farm output, allowing for faster convergence to optimal settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If individual optimization control is applied to each wind turbine, then the control setting for each turbine can be optimized, but it becomes difficult to apply the technique to large-scale wind farms with three or more turbines to achieve optimal overall adjustment
Solution Approach 1:
The patent combines individual turbine optimization with farm-level coordination by integrating wake effect models into a centralized control system. The control device merges data from multiple turbines and applies coordinated control strategies that optimize both individual turbine performance and overall farm output, resolving the contradiction between individual optimization precision and system complexity.
Solution Approach 2:
The control device performs multiple functions simultaneously: it monitors individual turbine performance, predicts wake effects across the farm, optimizes control settings for each turbine, and coordinates overall farm operation. This multi-functionality allows the system to achieve individual optimization while managing large-scale coordination without proportionally increasing complexity.
2Adaptability or versatility
If control settings are adjusted frequently to respond to wind variation, then the system can adapt to changing conditions, but the time required to reach optimal settings increases
Solution Approach 1:
The system performs preliminary actions by predicting future wake effects and pre-adjusting control settings before wind conditions fully change. The wake effect prediction model anticipates how wind variations will affect downstream turbines, allowing the control device to proactively optimize settings rather than reactively adjusting them, thus reducing convergence time while maintaining adaptability.
Solution Approach 2:
The control device implements continuous feedback loops that monitor actual turbine performance against predicted performance, automatically adjusting control settings to maintain optimality. This real-time feedback mechanism enables the system to quickly adapt to wind variations by learning from actual outcomes and correcting deviations, reducing the time to reach optimal settings through iterative refinement.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method of operating a wind farm including a plurality of wind turbines includes: a step of imparting a perturbation to a control setting of each of the wind turbines to be optimized; a step of, after imparting the perturbation, obtaining an evaluation value including at least a total output of the wind turbines to be optimized; and a step of updating the control setting for each of the wind turbines to be optimized on the basis of a gradient of the evaluation value. The step of obtaining an evaluation value includes, for each of the wind turbines to be optimized, employing, as effective data, the evaluation value obtained from the wind turbine satisfying a first condition which is required to be met by the wind turbines to be optimized at least in relation to wind direction.