Battery Sorting via Dual-Energy X-Ray Spectral Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for automated sorting of used batteries and power cells face challenges in precision due to variations in mass, corrosion, and internal structure changes, leading to inefficiencies and inaccuracies, especially with large databases and high-resolution spectra requirements.
Innovation Solution
A method utilizing a fan-shaped X-ray beam to generate high-resolution digital images at two energies, analyzing gray spectra with 8-bit or higher resolution, and indexing distinctive regions for accurate type identification, with a model database branch for rapid comparison and sorting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution gray spectra with more than 16 shades of gray are used to reliably identify power cells, then measurement precision is improved, but processing time increases making it difficult to sort within 100 msec
Solution Approach 1:
The patent segments the gray spectrum into multiple resolution levels (16 shades and 256 shades). The system first compares images using 16-shade resolution for quick initial filtering, then proceeds to 256-shade resolution only for candidates that pass the initial filter, enabling both speed and accuracy
Solution Approach 2:
The patent applies partial action by using the lower-resolution 16-shade spectrum for the majority of comparisons and reserving the computationally intensive 256-shade spectrum only for selective detailed analysis of promising candidates, rather than applying full high-resolution analysis to all power cells
2Reliability
If a database of tens of thousands model gray spectra is created to account for positional variations, then reliability of identification is improved, but device complexity and processing load increase
Solution Approach 1:
The patent segments the large database into multiple branches organized by power cell type (cylindrical, rectangular, button cells). Each branch contains only the relevant model spectra for that type, reducing the search space from tens of thousands of models to a manageable subset specific to each category
Solution Approach 2:
The patent performs preliminary classification by power cell type before detailed spectral comparison. The system first identifies the geometric type of the power cell and then searches only within the corresponding branch of the database, preparing the appropriate model set in advance for the specific type being analyzed
3Productivity
If two-pixel resolution video images are used to generate model database, then processing speed is improved, but measurement precision deteriorates causing overlaps between similar-looking batteries
Solution Approach 1:
The patent segments the image processing into two stages: model database generation uses 2-pixel resolution for speed, while actual identification uses higher-resolution 1-pixel (or finer) images for accurate comparison, separating the requirements for speed and precision into different operational phases
Solution Approach 2:
The patent applies different image resolutions to different purposes: lower resolution (2-pixel) is used where speed is critical (model generation and initial filtering), while higher resolution is used where precision is critical (final identification and differentiation of similar batteries)
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
This approach achieves high accuracy (99%+) and increased sorting speed (up to 50 msec) by using a reduced database index and detailed gray spectra, effectively distinguishing between various battery types despite internal structure asymmetries and external variations.
Implementation Method 1
a) generating a fan-shaped X-ray beam; scanning the batteries with the fan-shaped X-ray beam
Implementation Method 2
c) capturing X-rays that pass through the battery with an X-ray detector and converting the X-rays into first and second digital images
Data Source
AI summary
Method for automatic sorting of batteries, the method including generating a fan-shaped X-ray beam; scanning the batteries with the fan-shaped X-Ray beam; for each battery, capturing X-rays that pass though the battery with an X-ray detector and converting the X-rays into first and second digital images, wherein the first digital image represents X-rays at a first energy, and the second digital image represents X-rays at a second energy; automatically analyzing the first and second digital images to determine a type of the battery by identifying characteristic features of each battery type based on a gray spectrum of at least 8 bit resolution that is looked up in a model database branch; and sorting the batteries by type.