Grün HPB Series AI-Driven Sorting System

Plastic bottle sorting faces a core challenge: the complexity of mixed waste streams. PET, HDPE, and other polymers are commingled with varying colors, inconsistent labels, and fluctuating contamination levels, which creates obvious efficiency and accuracy bottlenecks for manual sorting or single-sensor sorting systems. For recycling plants running PET whole-bottle recovery lines and MRFs processing municipal solid waste, the ability to efficiently sort by polymer type and target surface features in a single scan directly determines output purity, downstream recycling efficiency, and the commercial value of recovered materials.

The Grün HPB Series is an intelligent machine-vision-based sorting system tailored for this exact scenario. It integrates cutting-edge InGaAs-SWIR imaging with high-resolution RGB (HR-RGB) VIS imaging, paired with proprietary AI deep learning algorithms and high-performance air ejectors, to accurately identify and sort target polymers and surface features from mixed waste streams within milliseconds in a single pass.

Empowered by AI-driven surface feature recognition algorithms, the system can conduct deep learning training based on real production data, continuously optimizing the machine vision detection accuracy. This means the equipment’s performance is not fixed at the time of factory shipment; instead, it continuously adjusts and improves with the accumulation of production data, delivering stable and reliable sorting performance to help recycling plants and MRFs significantly boost recovery efficiency and commercial value.

Technical Features

● High-Precision Simultaneous Polymer & Color Detection System

✦ Cutting-edge InGaAs-SWIR Imaging: Efficiently identifies polymer types (PET/HDPE/PP, etc.) by analyzing the unique short-wave infrared spectral signatures of various polymers.

✦ High-Stability Halogen Illumination: Provides stable lighting across the full spectrum; its air-cooled design helps maintain consistent light output during operation and extends bulb lifespan.

✦ High-Resolution RGB (HR-RGB) VIS Imaging: Classifies bottles by color variations (clear/blue/green/contaminated, etc.) with precise chromatic discrimination.

✦ High-Intensity LED Illumination: Ensures a constant imaging environment with clear, stable images unaffected by external light changes, guaranteeing consistent detection results across different times and batches.

● Proprietary AI Deep Learning Algorithms 

✦ Visual Feature Recognition: Automatically classifies bottles, caps, labels, and contaminants (metal, silicone), complementing sensor-based material identification.

✦ AI Model Optimization: Continuously enhances machine vision sorting performance through deep learning, adapting to new materials and complex forms such as printed-label bottles.

✦ Reliable Handling of Challenging Scenarios: Ensures consistent sorting accuracy under diverse feed conditions, including reflective, black, and crushed bottles.

● High-Performance Belt Sorting

✦ Multi-Specification Model Portfolio:Available in three width specifications (1200 mm, 2000 mm, and 2800 mm) with a throughput range of 2–6 tons/hour, meeting varying capacity requirements.

✦ Precision Pneumatic Ejectors: Offered in three nozzle configurations (88, 160, and 224 nozzles) with a response time of ≤1 ms and an ejection accuracy of ±2 mm, ensuring efficient and reliable material separation at high speeds.

Applications

Typical Application Scenarios

Sorts whole bottles by polymer type (PET/HDPE/PP) and color (clear/blue/green).

● Municipal Solid Waste (MSW) Processing

Separates recyclable plastic bottles from mixed municipal solid waste in MRFs.

● Mixed Plastic Sorting

Categorizes commingled plastic packaging by material type to enhance the purity of downstream pelletizing.

Specifications