Scholarlink

<p>Aluminum recycling is entering a&#xa0;new phase within the circular economy. Despite significant advances in sensor-based technologies such as X‑ray transmission (XRT), X‑ray fluorescence (XRF), and laser-induced breakdown spectroscopy (LIBS), challenges remain due to the complexity of aluminum alloys and the throughput demands of the industry. Binder+Co addresses this issue with CLARITY AI, a&#xa0;new optical sorting approach that applies artificial intelligence (AI) and machine learning (ML) to interpret the material’s optical fingerprint. Using high-resolution imaging and deep learning algorithms, CLARITY AI distinguishes subtle differences in surface structure, reflectivity, and geometry to classify cast parts, sheets, and extrusion profiles. Unlike conventional methods, AI sorting does not require laborious chemical analyses, which results in a&#xa0;significant economic advantage. This enables high-throughput sorting of mixed aluminum scrap, such as Twitch, while reducing reliance on slow and costly downstream LIBS processes. To generate robust models, the system requires extensive training datasets collected from real-world sorting trials. The paper first outlines the initial conditions required for effective aluminum scrap sorting. To ensure the CLARITY AI machine is trained under realistic industrial conditions, several tests have been performed using different material classes distinguished by shape, color and form. The paper will explore the resulting output fractions and evaluate them in terms of quantity and purity. In conclusion, the paper demonstrates that an AI-based sorting process is able to generate product fractions of comparable quality to other sensor-based technologies such as XRT, XRF, and LIBS, but without the need for chemical analysis. A&#xa0;hybrid process design allows AI to perform coarse sorting, with optional LIBS refinement for high-precision alloy separation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Optische Sortierung von Post-Consumer-Aluminiumschrott in Legierungsgruppen mittels KI als Alternative zu XRF- und XRT-Technologien

  • 期刊论文

  • 发表日期：

    2026年1月22日

查看全文
  • Georg Weingrill,
  • Georg Schinnerl

摘要

Aluminum recycling is entering a new phase within the circular economy. Despite significant advances in sensor-based technologies such as X‑ray transmission (XRT), X‑ray fluorescence (XRF), and laser-induced breakdown spectroscopy (LIBS), challenges remain due to the complexity of aluminum alloys and the throughput demands of the industry. Binder+Co addresses this issue with CLARITY AI, a new optical sorting approach that applies artificial intelligence (AI) and machine learning (ML) to interpret the material’s optical fingerprint. Using high-resolution imaging and deep learning algorithms, CLARITY AI distinguishes subtle differences in surface structure, reflectivity, and geometry to classify cast parts, sheets, and extrusion profiles. Unlike conventional methods, AI sorting does not require laborious chemical analyses, which results in a significant economic advantage. This enables high-throughput sorting of mixed aluminum scrap, such as Twitch, while reducing reliance on slow and costly downstream LIBS processes. To generate robust models, the system requires extensive training datasets collected from real-world sorting trials. The paper first outlines the initial conditions required for effective aluminum scrap sorting. To ensure the CLARITY AI machine is trained under realistic industrial conditions, several tests have been performed using different material classes distinguished by shape, color and form. The paper will explore the resulting output fractions and evaluate them in terms of quantity and purity. In conclusion, the paper demonstrates that an AI-based sorting process is able to generate product fractions of comparable quality to other sensor-based technologies such as XRT, XRF, and LIBS, but without the need for chemical analysis. A hybrid process design allows AI to perform coarse sorting, with optional LIBS refinement for high-precision alloy separation.