Papers for

wind farm operators

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Bird species identification accuracy and latency vary by image resolution and model

What Input Resolution Is Required for Bird Species Identification, and What Is Its Latency Cost on an Edge Device? A Study of 14 Input Resolutions and Six Architectures with On-Device Measurements

Abstract: Bird-strike mitigation at wind farms requires identifying distant birds that span only tens of pixels, so the classifier's input resolution N is a design variable, not a fixed specification. We study it with a factorial design over 14 side lengths N (16 to 224), six architectures, two training and evaluation regimes and 30 random seeds -- 2,520 checkpoints and 5,040 evaluations -- plus latency measured on an NVIDIA Jetson Orin Nano. Four results. (1) The selected N depends on the target: 0.90 is met on validation by ResNet50 at N=112 in an estimated 1.85 ms (0.8980 on test) and 0.95 by DINOv2-L at N=144 in 12.70 ms; changing the model buys more accuracy than raising N (+5.93 versus +2.33 points at N=112). (2) The benefit of lowering N depends on the assumed preprocessing path: N=224 -> 80 saves 13.5% when each individual is decoded from its own file but 46.2% when the detector decodes the 4K frame once; the Pareto set grows from 20 to 23 configurations. (3) Accuracy must be measured on the deployed engine: half precision costs ViT-S/16 alone 4 to 7 points at N>=96 while the CNNs stay within 0.1 points, and with selection held at validation the choice differs at 26 of 176 targets. A broken FP16 engine can run faster than a correct one, undetectable from latency; admitting 14 ViT-S/16 FP32 configurations moves the recommendation over the 0.931-0.938 band and under the 10 ms budget. (4) ViT-L-scale models fit this device, but activations exceed the FP16 range; splitting the graph at transformer-block boundaries confines FP32 to the affected segments, making the deployed DINOv2-L chain 1.85x faster than the single-engine build. We also quantify how the regime-difference sign stabilises with seed count; a sensitivity split removing some forms of group sharing preserves all 14 non-trivial signs at the selection boundary.

Sun 13 SeptComputer Vision and Pattern RecognitionMachine Learning
The gist
Identifying bird species from images is important for preventing bird crashes at wind farms. The authors studied how the size of the input image affects the accuracy and speed of different AI models running on a small computer device. They found that bigger images do not always improve accuracy as much as changing the model can. Also, smaller images can save processing time depending on how the images are handled. The study highlights that measuring accuracy on the actual device is important because some settings that look good during testing can perform worse in real life.
Open 2609.14247v1

Offline reinforcement learning improves wind farm power under changing winds

Offline Reinforcement Learning for Wind Farm Control: A Wind Tunnel Study under Dynamic Wind Directions

Abstract: This paper addresses the wind farm power maximization problem in the presence of wind direction changes. Specifically, a model-free Modified Twin Delayed Deep Deterministic Policy Gradient with Behavior Cloning (MTD3-BC) algorithm is proposed to tackle this task through yaw control under varying wind direction conditions. MTD3-BC is an offline reinforcement learning (RL) algorithm that aims to infer good behavior from only a precollected offline dataset. Additionally, to ensure smooth and moderate yaw adjustments, a new action consistency term is introduced into the policy optimization objective. Unlike online RL methods, MTD3-BC does not require extensive interactions with a wind farm simulator during training, significantly reducing computational costs and training time. A wind tunnel experiment is conducted to validate the effectiveness of the algorithm under varying wind directions. The results demonstrate that MTD3-BC successfully mitigates wake effects, delivering farm-level power gains of approximately 10\% over the baseline greedy strategy and performance on par with a data-calibrated model-based wake-steering benchmark, while requiring no wake model and only a small fraction of the training cost of online RL. To our knowledge, this is the first time an offline RL wind farm control policy has been validated and demonstrated experimentally.

Fri 11 SeptMachine Learning
The gist
Wind farms can produce more electricity if turbines face the wind correctly, but wind directions often change. The authors developed a new way to control turbine angles using offline machine learning, which learns from existing data instead of needing lots of trial runs. They tested this method in a wind tunnel experiment and found that it increased power output by about 10% compared to simple controls. This approach also worked as well as state-of-the-art methods that need detailed wind models, while being faster and cheaper to train.
Open 2609.12905v1