Solar power plants have become widely accepted in most parts of the world, however, one of the concerns of some people is that birds will be killed by not roaming the structures. In a study published in 2016 by Argonne National Laboratory in Illinois (United States), it has been estimated that between 37,800 and 138,600 birds are killed each year in large-scale solar power plants. While this is a huge lineup, even the top end doesn’t compare to the death toll from tall buildings and moving vehicles. In any case, the Argonne researchers have launched a project to quantify the effects on birds, the Argonne national laboratory of the US Department of Energy is working on a way to monitor avian interaction with photovoltaic infrastructures.

Argonne enlisted the help of a computer vision system from BoulderAI, which includes a deep neural network camera that uses artificial intelligence to monitor and make decisions about bird activity. The system performs three tasks: detecting moving objects near solar panels; identify which of these objects are birds; and classify events (such as perching, flying through, or colliding with solar panels). Researchers use deep learning to build models, which allows computers to “teach” how to spot birds and their behaviors by training them on similar examples.

In a previous Argonne project, researchers trained computers to distinguish drones flying in the sky above their heads. The avian-solar interaction project will build on this capability, bringing new complexities, noted Adam Szymanski, a software engineer from Argonne, who developed the drone detection model. The cameras of the solar installations will be aimed at the panels rather than pointed upwards, so there will be more complex backgrounds. For example, the system will need to differentiate between birds and other moving objects in the field of view, such as clouds, insects or people.

First, the researchers will install cameras at one or two solar power sites, recording and analyzing the video. Hours of video will need to be processed and graded by hand to form the computer model.

The establishment

The Rock AI DNNCam is a supercomputer camera designed for harsh environments. It has an on-board computer, making it possible to digitize scenes, reduce data, and then store or transmit the reduced contextual data.

Boulder AI DNN Camera

The on-board camera allows operation in a serverless environment. has an on-board computer, so the camera can work in a serverless environment. The camera is a 4k, 12-bit, HDR, 60FPS, Sony StarVis IMX 334 8.3MP imaging sensor with a Theia TL410p motorized varifocal 4K lens. The camera is WIFI / Bluetooth / Ethernet compatible.

Teaching bird strikes

Bird strikes are actually quite rare in solar power plants, so to teach the camera how to determine what a strike is in relation to a flying or perched bird, a bird strike could be taught to the system at the l using an object, such as a toy bird. Once the model is trained, it will run internally in the cameras on a live video feed, ranking interactions on the fly.

Results

At the end of the project, Argonne will have developed a camera system capable of detecting, monitoring and reporting bird activities around solar installations. The system will also notify the personnel of the solar installation when collisions occur. The technology will then be ready for large-scale field trials in many solar installations, Hamada said.

And after

During this three-year project, the team plans to collect videos to generate training data and experiment with a camera. They expect to see presentable results in the winter of 2022 after a small-scale model evaluation in the fall of 2022. The resulting data could be used to detect trends and begin to answer key questions: certain types of birds more prone to impacts? Are collisions increasing at certain times of the day or year? Does the geographic location of solar panels play a role in the types of interactions? Do solar power installations provide viable habitat for birds? The tech framework can also be used to monitor other wildlife by recycling AI with appropriate data.

Written by Anne Fischer, Editorial Director, Novus Light Technologies today


Previous

Galaxy S21 Ultra reportedly features new camera system and Qualcomm's larger Max ultrasonic fingerprint sensor

Next

NASA's Chandra X-ray observatory high-resolution camera officially restored to full operation

Leave a Reply

Your email address will not be published.

Check Also