Hyperspectral Conveyor Sorting for Complex Excavated Waste Streams
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies are inadequate for high-throughput automated sorting of construction and demolition waste on-site, particularly in distinguishing between inert, non-inert, and hazardous materials, due to their reliance on fluorescence analysis and limited suitability for real-time characterization of complex waste compositions.
Innovation Solution
Employing a conveyor system equipped with a hyperspectral imager and real-time comparison capabilities to identify and sort excavated materials using hyperspectral signatures, coupled with a database and machine learning for precise classification and diversion to appropriate containers based on detected constituents.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If fluorescence analysis is used to identify materials on conveyor, then material classification is achieved, but the system is not suitable for high-throughput automated sorting of complex waste compositions
Solution Approach 1:
The patent transitions from fluorescence analysis to hyperspectral imaging, changing the detection parameter from fluorescent emission to reflected light spectral characteristics across multiple bands. This enables real-time identification of diverse materials including inert waste, non-inert waste, and contaminants without requiring material excitation, thereby increasing both throughput and adaptability to complex waste compositions
Solution Approach 2:
The patent replaces the fluorescence detection mechanism with a hyperspectral imaging system that captures reflected light across 100-200 nanometer wavelength ranges. This substitution eliminates the need for fluorescence excitation sources and enables faster, non-contact identification of materials, improving sorting speed and versatility for automated high-throughput applications
2Measurement precision
If real-time hyperspectral imaging is implemented for waste sorting, then accurate material classification is achieved, but system complexity increases
Solution Approach 1:
The hyperspectral imager is designed to perform multiple functions: identifying inert waste, detecting non-inert waste, locating contaminants, and classifying materials across different wavelength ranges (100 microns to 200 nanometers). This multi-functionality consolidates what would otherwise require multiple separate detection systems into a single device, managing complexity while maintaining high measurement precision
Solution Approach 2:
The patent introduces a database of hyperspectral signatures as an intermediary between the imager and classification algorithm. Pre-stored spectral signatures of known materials serve as reference patterns, enabling the system to accurately classify materials by comparing real-time measurements against known patterns, thereby simplifying the real-time decision-making process while maintaining high accuracy
3Productivity
If automated sorting system is deployed for construction waste, then recycling efficiency is enhanced, but difficulty in objectively determining presence of specific waste types remains
Solution Approach 1:
The system incorporates real-time feedback through hyperspectral imaging, where the imager continuously monitors the conveyor belt and provides immediate spectral data to the classification algorithm. This feedback loop enables objective, real-time detection and classification of waste constituents, allowing the system to automatically adjust sorting decisions based on actual material composition rather than relying on manual or subjective assessment
Solution Approach 2:
The patent replaces subjective visual inspection or simple color sensors with hyperspectral imaging that objectively measures reflected light across multiple wavelength bands. This substitution provides quantitative spectral data that enables rigorous, objective determination of material composition, accurately identifying specific waste types and contaminants based on their unique spectral fingerprints
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables rapid, accurate, and automated sorting of construction and demolition waste into inert, non-inert, and hazardous categories, enhancing recycling efficiency and compliance with environmental regulations by ensuring real-time detection and diversion of contaminants.
Implementation Method 1
a training sequence consisting of analyzing a plurality of reference samples and recording in a training database a) the spectral reflection signature acquired by the spectral analysis
Data Source
Figure 1
AI summary
The present invention relates to a method for automatically processing excavated material in the form of aggregate on a conveyor provided with a hyperspectral imager (10), characterised in that it comprises means for comparing, in real time, the imaged region (8) of the flow of excavated material with a hyperspectral base (15) of signatures characteristic of undesirable constituents, said means controlling a means (4) for deflecting the flow towards a secondary container (3) if undesirable constituents are detected in the imaged region (8) and/or a means (4) for deflecting the flow towards another secondary container (3) if desirable constituents are detected in the imaged region (8).