JP2024508684A - plastic separation - Google Patents

Abstract

Systems and methods for classifying and separating plastic materials utilizing a vision system and one or more sensor systems, the systems and methods implementing a machine learning system to identify or classify each material. Each material may then be sorted into different groups based on such identification or classification.

Description

This application claims priority to US Provisional Patent Application No. 63/146,892 and US Provisional Patent Application No. 63/173,301. This application is a continuation-in-part of U.S. patent application Ser. No. 17/495,291, which is a continuation-in-part of U.S. patent application Ser. No. 17/227,245, which is a continuation-in-part of U.S. patent application Ser. No. 16/939,011, which is a continuation-in-part of U.S. patent application Ser. No. 10,722,922), which is a continuation-in-part of U.S. Patent Application No. 15/963,755 (issued as U.S. Patent No. 10,710,119). , which claims priority to U.S. Provisional Patent Application No. 62/490,219, which is filed under U.S. Patent Application No. 15/213,129 (issued as U.S. Pat. ), which claims priority to U.S. Provisional Patent Application No. 62/193,332, all of which are incorporated herein by reference. This application is also a continuation-in-part of U.S. patent application Ser. No. 17/491,415, which is a continuation-in-part of U.S. patent application Ser. 16/358,374 (issued as U.S. Patent No. 10,625,304), which is a divisional application of U.S. Patent Application No. 15/963,755 (issued as U.S. Patent No. 10,710,119). ) is a partial continuation application.

GOVERNMENT LICENSE RIGHTS This disclosure was made with support from the United States Government under Grant No. DE-AR0000422 awarded by the United States Department of Energy. The United States Government may have certain rights with respect to this disclosure.

TECHNICAL FIELD The present disclosure relates generally to the separation of solid waste, and more particularly to the separation of plastic debris from municipal or industrial solid waste.

This section is intended to introduce various aspects of technology that may be related to example embodiments of the present disclosure. This discussion is believed to help provide a framework that facilitates a better understanding of certain aspects of the disclosure. Accordingly, this section should be read in this light and should be understood as not necessarily an admission of prior art.

Recycling is the process of collecting and processing materials that would otherwise be thrown away (e.g., from a waste stream) so that they can be turned into new products or at least be disposed of more appropriately. Reduces the amount of waste sent to landfills, conserves natural resources such as wood, water, and minerals, increases economic security by leveraging domestic material sources, and reduces the need to collect new raw materials Recycling benefits local communities and the environment because it prevents pollution and saves energy. After collection, recyclables may be sent to a material recovery facility (“MRF”) where they are separated, cleaned, and processed into materials that can be used in manufacturing. As a result, a high-throughput automated separation platform that economically separates highly mixed waste streams would be beneficial across a variety of industries. Therefore, it is possible to identify, analyze and separate mixed industrial or municipal solid waste streams with high throughput and economically produce high-quality raw materials (which may also have low levels of micropollutants) for subsequent processing. There is a need for a cost-effective separation platform that can generate Typically, MRF cannot distinguish between many materials, so the separated materials are either limited to low-quality and low-priced markets, or are too time-consuming, labor-intensive and inefficient to be recycled economically. or limit the amount of material that can be recovered.

Municipal solid waste ("MSW") is a broad term that refers to waste streams that include domestic, commercial, and industrial sources. There are thousands of different materials and products within each of these categories. EPA reported that 267.8 million tons of MSW was generated in 2017. 35.37 million tons, or 13.2% of its total MSW weight, was composed of plastics. Of the 35.37 million tons of plastic, 2.96 million tons (8.4%) was recycled, 5.59 million tons (15.8%) was burned with energy recovery, and 26.82 million tons (75.8%) was reclaimed. It is clear that more plastics need to be recycled.

Plastic recycling is the reprocessing of plastic waste to create new and useful products. Almost all plastics are non-biodegradable and accumulate in the environment, so they need to be recycled. Currently, almost all recycling is done by remelting used plastics and regenerating them into new items: so-called mechanical recycling. This can lead to degradation of the polymer at a chemical level and also requires the plastic waste to be sorted by both color and polymer type before reprocessing, which is complex and expensive. Failure to do so may result in material properties becoming unstable, making them unattractive to industry. In an alternative approach known as raw material recycling, plastic waste can be converted back to its original chemicals and then reprocessed back into new plastic. This promises greater recycling, but at the expense of higher energy and capital costs. Plastic waste can also be burned instead of fossil fuels as part of energy recovery.

Currently, only some plastics can be recycled. When plastic is recycled, it is usually separated into different types of plastic. Recycling rates also vary depending on the type of plastic. Several types are commonly used, each with different chemical and physical properties. This creates differences in the ease of sorting and reprocessing, which affects the value and market size of recovered materials. Plastic packaging and products made from a single material (eg, polyethylene terephthalate (“PET”), high density polyethylene (“HDPE”), polypropylene (“PP”)) are more easily recycled. Plastics that are sometimes or rarely recyclable include polyvinyl chloride (“PVC”), low-density polyethylene (“LDPE”), linear low-density polyethylene (“LLDPE”), and polystyrene (“PS”). ) is included. Furthermore, plastic can only be recycled a limited number of times.

Modern single-flow MRFs and plastic reclaimers have large inflows of material, requiring processing equipment that can move and sort materials at high speeds. At the same time, the highest value is obtained from the purest and least polluted streams. To achieve these somewhat contradictory goals, today's single-stream MRF and reclaimers employ automated equipment to classify plastic packaging by its near-infrared ("NIR") signature, either through transmission or reflection. . These sensors rely on reflecting light from an external light source and can only see the surface of the material. Furthermore, only polymer information is obtained from this sensor. For example, NIR spectroscopy can distinguish between #1 type plastic, which is transparent and light blue PET, and #2 HDPE, #1 colored PET, #3 PVC, #4 LDPE, #5 PP, #6 PS, #7 Reject other plastics such as multilayer polymers, composite polymers, acrylics, and nylon. Additionally, NIR spectroscopy cannot accurately identify composite materials such as black or highly colored plastics, plastic-coated paper or multilayer packaging (made with polymeric multilayer films), resulting in misleading readings. There is a possibility that Most black plastics are colored using carbon. Black plastic is widely used in the automotive industry, electronics, food packaging, plastic bags, etc. But black plastic has the unfortunate side effect of not only absorbing visible light, but also absorbing the near-infrared part of the spectrum, making it invisible to NIR spectroscopy. So "secretly" the black plastic goes undetected into a "miscellaneous" bin at the end of the conveyor, where it is either burned for energy or thrown into a landfill.

In closed-loop, or primary, recycling, waste plastic is recycled into new items of similar quality and type (e.g., turning beverage bottles back into beverage bottles). However, continuous mechanical recycling of plastics without loss of quality is extremely difficult due to the cumulative degradation of the polymers and the risk of contaminant accumulation. Closed-loop recycling has been investigated for many polymers, but so far only plastic bottle recycling has been industrially successful.

In open-loop or secondary recycling (also called downcycling), the quality of the plastic decreases each time it is recycled, so the material cannot be recycled forever and ends up as waste. Recycling plastic bottles into fleece and other fibers is a common example and accounts for the majority of PET recycling. The reduction in polymer quality can be offset by mixing recycled plastics with virgin materials or compatibilized plastics when manufacturing new products.

Thermosetting polymers do not melt, but techniques have been developed to mechanically recycle them. This typically involves grinding the material and then mixing it with some type of binder to form a new composite material.

In raw material or tertiary recycling (also called chemical recycling), polymers are reduced to their chemical building blocks (monomers), which can then be polymerized back into new plastic. Thermal depolymerization and chemical depolymerization are two types of raw material recycling.

Energy recovery, also known as energy recycling or quaternary recycling, involves burning plastic waste instead of fossil fuels to produce energy.

A process has been developed that allows certain types of plastics to be used as a carbon source (instead of coke) in the recycling of steel scrap. Ground plastics may be used as construction aggregates or fillers in certain applications.

Plastic waste may be simply combusted as waste-derived fuel (“RDF”) in waste-to-energy processing, or it may first be chemically converted into a synthetic fuel. Both approaches either exclude PVC or use dichlorination techniques, since PVC produces large amounts of hydrogen chloride (HCl) when burned, which can corrode equipment and cause unwanted chlorination of fuel products. It is necessary to supplement this by introducing

Mixed plastic waste can be depolymerized to obtain synthetic fuel. Although it has a higher calorific value than the starting plastic and can burn more efficiently, it is still less efficient than fossil fuels. Various conversion techniques have been investigated, the most common of which is pyrolysis. The use of catalysts in pyrolysis results in clearer, higher value products. Compared to widespread incineration, plastics-to-fuel technologies have historically been less economically viable due to the high costs of collecting and sorting the plastics and the relatively low value of the fuels produced. I have been struggling with

US Published Patent Application No. 2022/0016675

Krizhevsky et al., “ImageNet Classification with Deep Convolutional Networks,” Proceedings of the 25th International Conference on Neural Information Processing Systems, December 3-6, 2012, Lake Tahoe, Nevada. LeCun et al., "Gradient-Based Learning Applied to Document Recognition," IEEE Proceedings, Institute of Electrical and Electronics Engineers (IEEE), November 1998.

As a result of the above, we have improved the process to separate all types of plastics, the ability to separate #3 to #7 type plastics, the ability to separate PVC, and the ability to separate plastic mixtures so that they can be recycled more efficiently. A new classification or ability to separate into fractions is desired.

Aspects of the present disclosure include capturing a first visual image of a first piece of material, resulting in a first image data packet for the first piece of material; capturing a second visual image of the piece, resulting in a second image data packet relating to the second piece of material, the first piece of material having a first chemical characteristic; the second piece of material has a second chemical feature that is different from the first chemical feature; using a machine learning system to process the first and second image data packets in response to a learned visual discrimination between pieces of material having different chemical characteristics; sorting the pieces of material into two different categories. The method may further include separating the first piece of material from the second piece of material according to the classification. The piece of material may also be a piece of plastic. The first chemical signature may be spectral data measured by a plurality of different sensor systems from at least one sample of a piece of plastic of the same type as the first piece of plastic, and the second chemical signature may be The spectral data may be measured by a plurality of different sensor systems from at least one sample of a piece of plastic of the same type as the second piece of plastic. The spectral data may relate to the non-visible spectrum. The plurality of different sensor systems may be selected from the group consisting of near-infrared ("NIR"), mid-wavelength infrared ("MWIR"), and x-ray fluorescence ("XRF") systems. Several different sensor systems include infrared (“IR”), Fourier transform IR (“FTIR”), forward looking infrared (“FLIR”), very near infrared (“VNIR”), near infrared (“NIR”), short Wavelength Infrared (“SWIR”), Long Wavelength Infrared (“LWIR”), Medium Wavelength Infrared (“MWIR” or “MIR”), X-Ray Transmission (“XRT”), Gamma Ray, Ultraviolet (“UV”), Fluorescent X radiation (“XRF”), laser-induced breakdown spectroscopy (“LIBS”), Raman spectroscopy, anti-Stokes Raman spectroscopy, gamma-ray spectroscopy, hyperspectral spectroscopy (e.g., beyond visible wavelengths), acoustic spectroscopy method, NMR spectroscopy, microwave spectroscopy, terahertz spectroscopy, differential scanning calorimetry (“DSC”), thermogravimetric analysis (“TGA”), capillary and rotational rheometry, optical microscopy and scanning electron microscopy (“SEM”) ), and chromatography. The first chemical signature may include measurements of organic and inorganic elements or molecules from at least one sample of a piece of plastic of the same type as the first piece of plastic, and the second chemical signature may include: It may include measurements of organic and inorganic elements or molecules from at least one sample of a piece of plastic of the same type as the second piece of plastic. The plastic pieces are Type #1 Polyethylene Terephthalate (“PET”), Type #2 High Density Polyethylene (“HDPE”), Type #3 Polyvinyl Chloride (“PVC”), Type #4 Low Density Polyethylene (“LDPE”) , Type #5 Polypropylene (“PP”), Type #6 Polystyrene (“PS”), Type #7 Other Polymers. The first piece of material may include polyvinyl chloride. The two different classifications may be different fractions.

Aspects of the present disclosure provide a first visual image of a first piece of material that provides a first image data packet for a first piece of material, and a second visual image that provides a second image data packet for a second piece of material. a second visual image of a piece of material, the first piece of material having a first chemical characteristic, and the second piece of material having a first chemical characteristic; A first and a second image are captured using a camera and a machine learning system pre-trained to visually identify pieces of material having different chemical characteristics. A data processing system configured to process data packets, the machine learning system detecting first and second pieces of material in response to learned visual discrimination between pieces of material having different chemical characteristics. A system is provided that includes a data processing system and a sorting device configured to separate a first piece of material from a second piece of material according to the fraction. The piece of material may also be a piece of plastic. The first chemical characteristic may be spectral data regarding non-visible spectra measured by a plurality of different sensor systems from at least one sample of a plastic piece of the same type as the first plastic piece; The chemical signature may include spectral data regarding non-visible spectra measured by a plurality of different sensor systems from at least one sample of a piece of plastic of the same type as the second piece of plastic. The plurality of different sensor systems may be from the group consisting of near-infrared ("NIR"), mid-wavelength infrared ("MWIR"), and x-ray fluorescence ("XRF") systems. Several different sensor systems include infrared (“IR”), Fourier transform IR (“FTIR”), forward looking infrared (“FLIR”), very near infrared (“VNIR”), near infrared (“NIR”), short Wavelength Infrared (“SWIR”), Long Wavelength Infrared (“LWIR”), Medium Wavelength Infrared (“MWIR” or “MIR”), X-Ray Transmission (“XRT”), Gamma Ray, Ultraviolet (“UV”), Fluorescent X radiation (“XRF”), laser-induced breakdown spectroscopy (“LIBS”), Raman spectroscopy, anti-Stokes Raman spectroscopy, gamma-ray spectroscopy, hyperspectral spectroscopy (e.g. beyond visible wavelengths), acoustic spectroscopy, NMR spectroscopy, microwave spectroscopy, terahertz spectroscopy, differential scanning calorimetry (“DSC”), thermogravimetric analysis (“TGA”), capillary and rotational rheometry, optical microscopy and scanning electron microscopy (“SEM”), and chromatography. The first chemical signature may include measurements of organic and inorganic elements or molecules from at least one sample of a piece of plastic of the same type as the first piece of plastic, and the second chemical signature may include: may include measurements of organic and inorganic elements or molecules from at least one sample of a plastic piece of the same type as the second plastic piece, the plastic piece being of type #1 polyethylene terephthalate (“PET”), type #2 High Density Polyethylene (“HDPE”), Type #3 Polyvinyl Chloride (“PVC”), Type #4 Low Density Polyethylene (“LDPE”), Type #5 Polypropylene (“PP”), Type #6 Polystyrene ( "PS"), Type #7 Other Polymers.

Aspects of the present disclosure include the steps of: using a plurality of different sensor systems to determine the chemical characteristics of each of a mixture of different pieces of plastic; capturing a visual image of each piece of plastic; digitally associating chemical signatures with visual images, determining specific fractions for separating plastic pieces, and using the visual images to determine which plastic pieces in the mixture belong to specific fractions; and training a machine learning system to visually identify pieces of plastic that fall within a particular fraction, the training comprising: The method includes the following steps: The control group may consist of captured visual image data for each of the identified plastic pieces. A fraction may be composed of a particular combination of organic and inorganic elements or molecules. The plurality of different sensor systems may be selected from the group consisting of near-infrared ("NIR"), mid-wavelength infrared ("MWIR"), and x-ray fluorescence ("XRF") systems. Mixtures of different plastic pieces are: Type #1 Polyethylene Terephthalate (“PET”), Type #2 High Density Polyethylene (“HDPE”), Type #3 Polyvinyl Chloride (“PVC”), Type #4 Low Density Polyethylene (“ LDPE"), Type #5 Polypropylene ("PP"), Type #6 Polystyrene ("PS"), Type #7 Other Polymers. Several different sensor systems include infrared (“IR”), Fourier transform IR (“FTIR”), forward looking infrared (“FLIR”), very near infrared (“VNIR”), near infrared (“NIR”), short Wavelength Infrared (“SWIR”), Long Wavelength Infrared (“LWIR”), Medium Wavelength Infrared (“MWIR” or “MIR”), X-Ray Transmission (“XRT”), Gamma Ray, Ultraviolet (“UV”), Fluorescent X radiation (“XRF”), laser-induced breakdown spectroscopy (“LIBS”), Raman spectroscopy, anti-Stokes Raman spectroscopy, gamma-ray spectroscopy, hyperspectral spectroscopy (e.g., beyond visible wavelengths), acoustic spectroscopy method, NMR spectroscopy, microwave spectroscopy, terahertz spectroscopy, differential scanning calorimetry (“DSC”), thermogravimetric analysis (“TGA”), capillary and rotational rheometry, optical microscopy and scanning electron microscopy (“SEM”) ), and chromatography.

1 is a schematic diagram of a sorting system configured in accordance with certain embodiments of the present disclosure. FIG. FIG. 3 illustrates an example representation of a control set of material pieces used in the training phase of a machine learning system. FIG. 3 illustrates a flowchart constructed in accordance with certain embodiments of the present disclosure. 1 is a simplified schematic diagram constructed in accordance with certain embodiments of the present disclosure; FIG. FIG. 3 is a diagram showing an example of chemical characteristics. FIG. 3 is a diagram showing an example of chemical characteristics. FIG. 3 illustrates a flowchart constructed in accordance with certain embodiments of the present disclosure. FIG. 3 illustrates a flowchart constructed in accordance with certain embodiments of the present disclosure. 1 is a block diagram of a data processing system configured in accordance with certain embodiments of the present disclosure. FIG.

Various detailed embodiments of the disclosure are disclosed herein. However, it is to be understood that the disclosed embodiments are merely illustrative of the disclosure and may be embodied in various alternative forms. Numbers are not necessarily to scale and some features may be exaggerated or minimized to illustrate details of particular components. Accordingly, the specific structural and functional details disclosed herein are not to be construed as limitations, but rather as a representative basis for teaching those skilled in the art to use the various embodiments of the present disclosure. It should only be construed as

As used herein, "materials" includes metals (ferrous and non-ferrous), alloys, plastics (such as those disclosed herein, known in the industry, or newly created in the future). (including but not limited to), rubber, foam, glass (including but not limited to borosilicate or soda lime glass, and various colored glasses), ceramics, paper, cardboard, Teflon (registered trademark) ), PE, wire bundles, insulated wires, rare earth elements, leaves, wood, plants, plant parts, textiles, biowaste, packaging, e-waste, batteries and accumulators, end-of-life vehicles, mines, construction, demolition Waste, crop waste, forest residue, purpose-grown grasses, wood energy crops, microalgae, urban food waste, food waste, hazardous chemical and biomedical waste, construction waste, farm waste , articles of biological origin, articles of non-biological origin, objects with a certain carbon content, other objects that may be found within municipal solid waste, and any of the sensor technologies disclosed herein. Any further types or classes disclosed herein that can be distinguished from each other by one or more sensor systems, including, but not limited to, Any item or object may be included, including, but not limited to, any other object, article, or material.

"Material" may include any item or object composed of a chemical element, a compound or mixture of chemical elements, or a compound or mixture of chemical elements, and the complexity of the compound or mixture can be as simple as They can range from simple to complex. As used herein, "element" means a chemical element of the Periodic Table of Elements, including elements that may be discovered after the filing date of this application. Within this disclosure, the terms "scrap," "scrap piece," "material," and "piece of material" may be used interchangeably.

As is well known in the art, a "polymer" is a substance or material composed of very large molecules or macromolecules made up of many repeating subunits. The polymer can be a natural polymer or a synthetic polymer found in nature.

A “multilayer polymeric film” is composed of two or more different compositions and may have a thickness of up to about 7.5 ⁻⁸ ×10 ⁻⁴ m. The layers are at least partially continuous and preferably, but possibly coextensive.

As used herein, the terms "plastic," "piece of plastic," and "piece of plastic material" (all of which may be used interchangeably) refer to a polymeric composition of one or more polymers and/or or refers to any object comprising or consisting of a multilayer polymeric film.

As used herein, the term "chemical signature" means a chemical signature produced by one or more analytical instruments that indicates the presence of one or more specific elements or molecules (including polymers) in a sample. refers to a unique pattern (such as a fingerprint spectrum) that is Elements or molecules may be organic and/or inorganic. Such analytical instruments include any of the sensor systems disclosed herein. According to embodiments of the present disclosure, one or more sensor systems disclosed herein may be configured to generate a chemical signature of a piece of material (eg, a piece of plastic).

As used herein, "fraction" refers to organic and/or inorganic elements or molecules, types of polymers, plastics, including all of the various classes and types of plastics disclosed herein. type, polymer composition, chemical characteristics of the plastic, physical properties of the plastic piece (e.g. color, transparency, strength, melting point, density, shape, size, manufacturing type, uniformity, response to stimuli, etc.) Refers to a combination. Non-limiting examples of fractions are LDPE plus a relatively high proportion of aluminum; LDPE and PP plus a relatively low proportion of iron; PP plus zinc; PE, PET, combination of HDPE; red LDPE plastic pieces of any type; any combination of plastic pieces except PVC; black plastic pieces; combination of #3 to #7 type plastics containing specified combinations of organic and inorganic molecules a combination of one or more different types of multilayer polymer films; a combination of specific plastics without specific contaminants or additives; any type of plastic with a melting point above a specified threshold; several specific types thermoset plastics; certain plastics that do not contain chlorine; combinations of plastics with similar densities; combinations of plastics with similar polarity; plastic bottles without caps attached, and vice versa. are different types of plastic pieces.

"Catalytic pyrolysis" includes the decomposition of a polymeric material by heating it in the presence of a catalyst in the absence of oxygen.

The term "predetermined" refers to something established or determined in advance.

"Spectral imaging" is imaging that uses multiple bands across the electromagnetic spectrum. A typical camera captures light across three wavelength bands of the visible spectrum, red, green, and blue (RGB), but spectral imaging includes a variety of techniques that include but go beyond RGB. Spectral imaging may use the infrared, visible, ultraviolet, and/or x-ray spectra, or combinations of the above. Spectral data, or spectral image data, is a digital data representation of a spectral image. Spectral imaging may include simultaneous acquisition of visible and non-visible band spectral data, illumination from outside the visible range, or use of optical filters to capture specific spectral ranges. It is also possible to capture hundreds of wavelength bands for each pixel of the spectral image.

As used herein, the term "image data packet" refers to a packet of digital data relating to a captured spectral image of an individual piece of material.

As used herein, the terms "identify" and "classify", "identify" and "classify", and their derivatives may be used interchangeably. As used herein, "classifying" a piece of material means determining (i.e., identifying) the type or class of material to which the piece of material belongs. For example, according to certain embodiments of the present disclosure, a sensor system (described further herein) can be configured to collect and analyze any type of information for classifying materials, Classifications are utilized within the system to include color, texture, hue, shape, brightness, weight, density, composition, size, uniformity, manufacturing type, chemical characteristics, predetermined proportions, radioactive characteristics, light, sound, or other signals, emitted from a piece of material, and/or reflected electromagnetic radiation ("EM"), including, but not limited to, responses to stimuli such as various fields. and/or material pieces can be selectively sorted according to a set of chemical properties (e.g., user-definable). As used herein, "manufacturing type" refers to metal parts formed by forging processes, cast metal parts (including, but not limited to, consumable casting, permanent casting, and powder metallurgy) Refers to the type of manufacturing process by which a piece of material was produced, such as a forging, material removal process, etc.

The type or class (ie, classification) of the material is user-definable and is not limited to known material classifications. The granularity of types or classes ranges from very coarse to very fine. For example, types or classes include relatively coarse-grained types or classes of plastics, ceramics, glasses, metals, and other materials, such as finer-grained types or classes of zinc, copper, brass, chrome plate, aluminum. various metals and metal alloys, such as, or among certain types or classes of relatively fine-grained plastics. Thus, a type or class may be used, for example, to distinguish between materials of significantly different composition, such as different types of plastics (e.g., any of the types #1 to #7), or to distinguish between, e.g. It may be configured to distinguish between materials of nearly identical composition, such as different subclasses of plastics that may fall within a plastic type. It should be appreciated that the methods and systems discussed herein can be applied to accurately identify/classify materials whose composition is completely unknown prior to classification.

Embodiments of the present disclosure improve plastics sorting capabilities through the fusion of multiple sensor technologies and machine learning systems. The limitations of sensor-based separation techniques arise from the use of a single sensor, as each sensor can only detect a narrow range of signals. The most common separator sensor types are eddy current, visible camera, x-ray transmission, near-infrared, and x-ray fluorescence (“XRF”). These are summarized in the table below.

However, MSW plastic pieces can be composed of one or more organic polymers, one or more inorganic elements, and come in a variety of colors, shapes, and sizes. Examples of these plastics include potato chip bags, squeezable juice boxes, some beverage containers, and the electromagnetic-sensitive packaging of electronic devices. Embodiments of the present disclosure utilize sensor-based techniques that can accomplish the separation of these different types of plastics into unique classifications that can account for their organic polymeric composition and/or their inorganic elemental composition. Generate new fractions from material flows. For example, a conversion chemist with a keen interest in the relative composition of polymers and inorganic elements can select one or more novel fractions and create specific products from recycled plastics separated into such fractions. It becomes like this. As a result, a fractionation system configured according to embodiments of the present disclosure can produce fractions that exceed what is possible with existing state-of-the-art fractionation techniques.

For example, certain embodiments of the present disclosure classify and/or separate predetermined fractions from bales of type #3-#7 plastic to create new products (e.g., through recycling methods) and/or fuels. may be configured. Examples of end uses for such fractions include, but are not limited to, gases (eg, C1-C4, etc.), fuels (eg, gasoline, diesel, etc.), and vacuum gas oils. However, it has never been successful to separate #3 to #7 type plastics based on their organic and inorganic element compositions.

Embodiments of the present disclosure may be configured to classify pieces of plastic material according to various different predetermined fractions or combinations of properties or types disclosed below and elsewhere within this disclosure.

Plastics are classified into three types according to their characteristics, such as chemical structure, polarity, and use.

Plastics can be classified according to their chemical structure and temperature behavior into thermoplastics, thermosets, and elastomers.

Regarding polarity, when atoms of different properties are present, electrons move toward the most electronegative atom in a covalent bond, creating a dipole. Polymers containing extremely electronegative atoms, such as CI, O, N, and F, become polar compounds and affect the properties of the material. As polarity increases, mechanical resistance, hardness, rigidity, heat resistance, water absorption/hygroscopicity, and chemical resistance improve, as well as permeability to polar compounds such as water vapor and adhesion/adhesion to metals. At the same time, increased polarity reduces thermal expansion, electrical insulation ability, tendency to accumulate electrostatic charges, and permeability to polar molecules (O ₂ , N ₂ ). In this way, different families of polyolefins, polyesters, acetals, halogenated polymers etc. can be distinguished.

Depending on the application, a third classification applies to thermoplastic materials. There are four types of plastics in this third category.

Standard plastic or commodity: A plastic that is produced and used in large quantities because of its price and various superior properties. Examples include copolymers of polyethylene ("PE"), polypropylene ("PP"), polystyrene ("PS"), polyvinyl chloride ("PVC"), or acrylonitrile butadiene styrene ("ABS").

Engineering plastics: used when good structure, transparency, self-lubricating properties, and thermal properties are required. Examples include polyamide (“PA”), polyacetal (“POM”), polycarbonate (“PC”), polyethylene terephthalate (“PET”), polyphenylene ether (“PPE”), and polybutylene terephthalate (“PBT”). There is.

Specialty plastics: Plastics with very specific properties, such as polymethyl methacrylate (“PMMA”), which has high transparency and light resistance, and polytetrafluoroethylene (Teflon®), which has excellent resistance to temperature and chemical products. ing.

High-performance plastics: Mainly thermoplastics with high heat resistance. In other words, it has excellent mechanical resistance to high temperatures, especially up to 150°C. Polyimide (“PI”), polysulfone (“PSU”), polyethersulfone (“PES”), polyarylsulfone (“PAS”), polyphenylene sulfide (“PPS”), and liquid crystal polymer (“LCP”) are highly Performance plastic.

Many plastic products have symbols that indicate the type of polymer from which they are made. These resin identification codes, often abbreviated as RIC, are used internationally. There are seven codes in total, six for the most common general purpose plastic types and one to cover everything else. These types are also referred to herein as polymer types #1-#7. Polymer type #1 refers to polyethylene terephthalate (“PET”), #2 refers to high density polyethylene (“HDPE”), #3 refers to polyvinyl chloride (“PVC”), and #4 refers to low density polyethylene (“HDPE”). #5 refers to polypropylene (“PP”); #6 refers to polystyrene (“PS”); #7 refers to other polymers not included in polymer types #1 to #6 ( For example, acrylic, polycarbonate ("PC"), polyactive fiber, polylactic acid, nylon, glass fiber, ABS). The EU maintains a similar nine-code list, which also includes ABS and polyamide.

PET plastic is used to make many common household items such as beverage bottles, medicine bottles, rope, clothing, and carpet fibers. HDPE plastics are often used to make containers for milk, motor oil, shampoo and conditioner, soap bottles, detergents, and bleach. PVC is used in all types of pipes and tiles, most commonly in plumbing pipes. LDPE products include cling film, sandwich bags, squeezable bottles, plastic grocery bags, etc. PP is used to make lunch boxes, margarine containers, yogurt pots, syrup bottles, medicine bottles, and PET bottle caps. Polystyrene products include disposable coffee cups, plastic food boxes, plastic cutlery, and packaging foam. Polycarbonate is used in baby bottles, CDs, medical storage containers, etc. Thus, according to embodiments of the present disclosure, a vision system implementing a machine learning system can be trained to identify and separate these different types of plastic based on the type of product manufactured. .

Plastic pieces can be classified according to the type of additives they may contain. Additives are compounds that are blended into plastics to improve performance and include stabilizers, fillers, and dyes. Clear plastics have the most value as they may still be dyed, while black or highly colored plastics have much less value as their inclusion may discolor the product. . Therefore, plastics may need to be sorted by both polymer type and color to obtain materials suitable for recycling.

Plastics can also be classified and sorted based on density. Certain polymers have similar density ranges (eg, PP and PE, or PET, PS, and PVC). If the plastic piece contains a high percentage of filler, it can affect the density.

Plastic waste can be broadly divided into two categories: industrial scrap (also called industrial waste resin) and used waste.

Plastic pieces can be sorted/sorted according to how they can be recycled. During mechanical recycling, plastics may be reprocessed at temperatures between 150 and 320°C depending on the polymer type, which can lead to undesirable chemical reactions that cause polymer degradation. This can reduce the physical properties and overall quality of the plastic, produce volatile low molecular weight compounds, create undesirable tastes and odors, and cause thermal discoloration. Accordingly, embodiments of the present disclosure may be configured to sort and separate plastic pieces such that such undesirable chemical reactions are avoided. Additives present within the plastic can accelerate this deterioration. For example, oxo-biodegradable additives aimed at improving the biodegradability of plastics can increase the degree of thermal degradation. Similarly, flame retardants can also have undesirable effects. Accordingly, embodiments of the present disclosure may be configured to sort and separate plastic pieces such that plastic pieces containing particular ones of such additives are discarded.

The quality of the product can also largely depend on how well the plastic was sorted. Many polymers are immiscible with each other when melted and phase separate during reprocessing (like oil and water). Products made from such blends contain many boundaries between different types of polymers, and weak cohesive forces across these boundaries result in poor mechanical properties. Accordingly, embodiments of the present disclosure may be configured to sort and sort plastic pieces such that certain immiscible plastic pieces are not sorted together into the same group.

Systems and methods described herein in accordance with certain embodiments of the present disclosure receive a heterogeneous mixture of pieces of material (e.g., any combination of various plastics disclosed herein), At least one piece of material in the heterogeneous mixture includes a different elemental composition (e.g., chemical signature) than one or more other pieces of material, and/or at least one piece of material in the heterogeneous mixture The piece is distinguishable from other pieces of material (e.g., visually distinguishable properties or characteristics, different chemical characteristics, etc.), and the systems and methods distinguish this piece of material from such other pieces of material. It is configured to identify/classify/separate into different groups. Embodiments of the present disclosure may be utilized to separate any type or class of materials or fractions defined herein.

Embodiments of the present disclosure may be implemented by physically placing (e.g., turning or discharging) pieces of material into individual containers or bins according to user-defined groupings (e.g., material type classifications or fractions). , is described herein as separating pieces of material into such distinct groups. By way of example, in certain embodiments of the present disclosure, pieces of material have physical properties that are distinguishable from the physical properties of other pieces of material (e.g., visually distinguishable properties or characteristics, different chemical characteristics, etc.). can be sorted into separate bins to separate pieces of material with

FIG. 1 illustrates an example system 100 configured in accordance with various embodiments of the present disclosure. Conveyor system 103 may be implemented to convey one or more streams of individual pieces of material 101 through system 100, thereby tracking, sorting, and delivering each individual piece of material 101 to a predetermined desired location. Can be separated into groups. Such a conveyor system 103 may be implemented using one or more conveyor belts on which the pieces of material 101 typically move at a predetermined constant speed. However, certain embodiments of the present disclosure are applicable to systems in which material pieces free fall through the various components of system 100 (or any other type of vertical separator), or to other types of systems, including vibratory conveyor systems. It may also be implemented with a conveyor system. Hereinafter, where applicable, conveyor system 103 may also be referred to as conveyor belt 103. In one or more embodiments, some or all of the communicating, stimulating, detecting, classifying, and sorting operations may be performed automatically, ie, without human intervention. For example, in system 100, one or more stimulation sources, one or more radiation detectors, classification modules, sorting devices, and/or other system components automatically perform these and other operations. It can be configured as follows.

Additionally, while FIG. 1 depicts a single stream of material pieces 101 on conveyor system 103, embodiments of the present disclosure may include multiple streams of material pieces moving the various components of system 100 toward each other. Can be implemented to pass in parallel. For example, as further described in U.S. Pat. can be distributed into streams. Accordingly, certain embodiments of the present disclosure are capable of simultaneously tracking, sorting, and separating streams of multiple such parallel moving pieces of material. According to certain embodiments of the present disclosure, the incorporation or use of singulators is not required. Alternatively, the conveyor system (e.g., conveyor system 103) may simply convey the mass of material pieces deposited on conveyor system 103 in a random manner.

According to certain embodiments of the present disclosure, pieces of material 101 may be fed onto conveyor system 103 using some suitable feeder mechanism (e.g., another conveyor system or hopper 102); The conveyor system 103 may transport pieces of material 101 through various components within the system 100. After the material pieces 101 are received by the conveyor system 103, an optional tumbler/vibrator/singulator 106 may be utilized to separate the individual material pieces from the collection of material pieces. In certain embodiments of the present disclosure, conveyor system 103 is operated to move at a predetermined speed by conveyor system motor 104. This predetermined speed may be programmable and/or adjustable by the operator in any known manner. Alternatively, monitoring of the predetermined speed of conveyor system 103 may be performed using position detector 105. In certain embodiments of the present disclosure, control of conveyor system motor 104 and/or position detector 105 may be performed by automatic control system 108. Such automated control system 108 may operate under the control of computer system 107 and/or functionality for performing automated control may be implemented in software within computer system 107.

Conveyor system 103 may be a conventional endless belt conveyor using a suitable conventional drive motor 104 to move the belt conveyor at a predetermined speed. Position detector 105 may be a conventional encoder and may be operably coupled to conveyor system 103 and automatic control system 108 to provide information corresponding to conveyor belt movement (e.g., speed). . Accordingly, pieces of material 101 are moved on conveyor system 103 through the use of controls to conveyor system drive motor 104 and/or automatic control system 108 (or including position detector 105), as further described herein. Once each of the materials are identified, they can be tracked by position and time (with respect to the various components of the system 100), so that each piece of material 101 can track the various components of the system 100 as it passes near it. components can be started/stopped. As a result, automatic control system 108 can track the position of each piece of material 101 as it moves along conveyor system 103.

Referring again to FIG. 1, certain embodiments of the present disclosure utilize a visual or optical recognition system 110 and/or a piece tracking device 111 as a means to track each piece of material 101 as it moves on conveyor system 103. can do. Vision system 110 may utilize one or more still or live-action cameras 109 to record the position (ie, position and timing) of each piece of material 101 on moving conveyor system 103. Vision system 110 may also or alternatively be configured to perform certain types of identification (e.g., classification) of all or a portion of material piece 101, as further described herein. . For example, such a vision system 110 may be utilized to capture or obtain information about each piece of material 101. For example, vision system 110 may classify material piece 101 according to a set of one or more properties (e.g., physical and/or chemical and/or radioactive, etc.), as described herein. The system 100 may be configured to capture or collect any type of information from the pieces of material available within the system 100 (eg, using a machine learning system) to separate and/or selectively sort. In accordance with certain embodiments of the present disclosure, vision system 110 provides visual images (one-dimensional , two-dimensional, three-dimensional, or holographic images). Such visual images captured by the optical sensor are stored as image data in a memory device (eg, initialized as an image data packet). According to certain embodiments of the present disclosure, such image data may represent images captured within optical wavelengths of light (i.e., wavelengths of light observable by a typical human eye). . However, alternative embodiments of the present disclosure may utilize sensor systems configured to capture images of materials comprised of wavelengths of light outside of the visual wavelengths of the human eye.

According to certain embodiments of the present disclosure, system 100 includes one or more sensor systems 120 that can be utilized alone or in combination with vision system 110 to classify/identify pieces of material 101. It can be implemented using Sensor system 120 may include sensor systems that utilize emitted or reflected electromagnetic radiation (e.g., infrared ("IR"), Fourier transform IR ("FTIR"), forward looking infrared ("FLIR"), very near infrared (" Near Infrared (“NIR”), Short Wave Infrared (“SWIR”), Long Wave Infrared (“LWIR”), Medium Wave Infrared (“MWIR” or “MIR”), X-ray Transmission (“XRT”) ), gamma rays, ultraviolet (“UV”), X-ray fluorescence (“XRF”), laser-induced breakdown spectroscopy (“LIBS”), Raman spectroscopy, anti-Stokes Raman spectroscopy, gamma-ray spectroscopy, hyperspectral spectroscopy (e.g., any range above visible wavelengths), acoustic spectroscopy, NMR spectroscopy, microwave spectroscopy, terahertz spectroscopy, and one-, two-, or three-dimensional imaging using any of the above. or by any other type of sensor technology, including but not limited to chemical or radioactive substances, to determine the chemical characteristics of the plastic pieces and/or classify the plastic pieces for separation. may be constructed using any type of sensor technology for. An implementation of an exemplary XRF system (eg, for use as sensor system 120 herein) is further described in US Pat. No. 10,207,296. XRF can be used within embodiments of the present disclosure to identify inorganic materials within a piece of plastic (eg, to include within a chemical signature).

The following sensor systems may also be used within certain embodiments of the present disclosure to determine the chemical characteristics of plastic pieces and/or to classify plastic pieces for separation.

Utilizes various previously disclosed forms of infrared spectroscopy to provide information about the base polymer of any plastic material and other components present in the material (such as mineral fillers, copolymers, polymer blends, etc.) This gives each piece of plastic a unique chemical signature.

Differential Scanning Calorimetry (“DSC”) is a thermal analysis technique that captures the unique thermal transitions of each material that are generated during heating of the material being analyzed.

Thermogravimetric analysis ("TGA") is another thermal analysis technique that provides quantitative information about the composition of plastic materials in terms of polymer proportions, other organic components, mineral fillers, carbon black, etc.

Capillary and rotational rheometry can determine the rheological properties of polymeric materials by measuring their creep resistance and deformation resistance.

Optical microscopy and scanning electron microscopy (“SEM”) can be used to measure the number and thickness of layers in multilayer materials (e.g., multilayer polymer films), the dispersion size of pigment or filler particles in the polymer matrix, the It can provide information about the structure of the analyzed material with respect to defects, interfacial morphology between components, etc.

Chromatography (LC-PDA, LC-MS, LC-LS, GC-MS, GC-FID, HS-GC, etc.) removes residual monomers, residual solvents from inks and adhesives, deteriorating substances, and UV stabilizers. , antioxidants, plasticizers, anti-slip agents, and other trace components of plastic materials can be quantified.

Although FIG. 1 is shown in combination with a vision system 110 and one or more sensor systems 120, embodiments of the present disclosure may be implemented using any of the sensor technologies disclosed herein or currently available. Note that it may be implemented with any combination of sensor systems utilizing other sensor technologies or other sensor technologies developed in the future. Although FIG. 1 is shown as including one or more sensor systems 120, implementation of such sensor systems is optional within certain embodiments of the present disclosure. In certain embodiments of the present disclosure, a combination of both vision system 110 and one or more sensor systems 120 may be used to classify pieces of material 101. In certain embodiments of the present disclosure, any combination of one or more different sensor technologies disclosed herein may be used to classify pieces of material 101 without utilizing vision system 110. . Additionally, embodiments of the present disclosure provide that the output of the sensor/vision system is processed within a machine learning system (as further disclosed herein) to classify/identify materials from a heterogeneous mixture of materials. ), and may include any combination of one or more sensor systems and/or vision systems that can then be separated from each other.

According to alternative embodiments of the present disclosure, the vision system 110 and/or sensor system determines which pieces of material 101 are of the type to be separated by the system 100 (e.g., contain certain contaminants, additives, or undesirable physical substances). a plastic piece containing a characteristic (e.g., an attached container cap formed of a different type of plastic than the container), and may be configured to transmit a signal to identify and reject such piece of material. . In such a configuration, the identified pieces of material 101 may be redirected/ejected utilizing one of the mechanisms described below for physically redirecting the separated pieces of material into individual bins. I can do it.

In certain embodiments of the present disclosure, the piece tracking device 111 and associated control system 112 determine the location (i.e., position and timing) of each piece of material 101 on the moving conveyor system 103 as well as the proximity of the piece tracking device 111. may be utilized and configured to measure the size and/or shape of each piece of material 101 passing through. Exemplary operation of such a piece tracking device 111 and control system 112 is further described in US Pat. No. 10,207,296. Alternatively, as previously disclosed, vision system 110 may be utilized to track the position (i.e., position and timing) of each piece of material 101 transported by conveyor system 103. Accordingly, certain embodiments of the present disclosure may be implemented without a piece tracker (eg, piece tracker 111) to track pieces of material.

In certain embodiments of the present disclosure that implement one or more sensor systems 120, the sensor system 120 is configured such that the vision system 110 determines the chemical composition, relative chemical composition, and/or manufacturing type of each piece of material 101. , may be configured to assist in identifying the piece of material 101 as it passes near the sensor system 120. Sensor system 120 may include an energy emitting source 121 that may be powered by a power source 122, for example, to stimulate a response from each of the pieces of material 101.

According to certain embodiments of the present disclosure implementing an XRF system as sensor system 120, source 121 is an in-line X-ray fluorescence (“IL- XRF) tubes. Such an IL-XRF tube may include a separate X-ray source dedicated to one or more streams (eg, singulation) of the transported piece of material. In such a case, the one or more detectors 124 may be implemented as XRF detectors that detect X-ray fluorescence from the pieces of material 101 within each of the individualized streams. Examples of such XRF detectors are further described in US Pat. No. 10,207,296.

In certain embodiments of the present disclosure, sensor system 120 may emit an appropriate sensing signal toward each piece of material 101 as each piece of material 101 passes near radiation source 121. The one or more detectors 124 may be arranged and configured to sense/detect one or more properties from the piece of material 101 in a manner appropriate to the type of sensor technology utilized. One or more detectors 124 and associated detector electronics 125 capture these received sensed characteristics and perform signal processing to generate digitized information representative of the sensed characteristics (e.g. , spectral data), and the digitized information may then be analyzed and used to assist vision system 110 in classifying each of the pieces of material 101 in accordance with certain embodiments of the present disclosure. . This classification may be performed within the computer system 107, and then the material pieces 101 are sorted (e.g., redirected/ can be utilized by the automatic control system 108 to operate one of the N (N>1) fractionators 126...129 for discharging). Four sorting devices 126...129 and four sorting bins 136...139 associated with these sorting devices are shown in FIG. 1 by way of non-limiting example only.

Existing sorters for plastics are designed to separate materials in a binary manner, with air nozzles at the end of the conveyor discharging the identified class of plastic into one of two bins. For example, if four classes of plastics need to be separated, the entire stream would need to be conveyed four times to such a binary separator, including four times when removing one object in the stream. It takes twice as long. According to embodiments of the present disclosure, the system 100 allows multiple classifications of plastics to be sorted in a single pass.

The sorting device may include any well-known method for redirecting selected material pieces 101 to a desired location, including but not limited to redirecting material pieces 101 from a conveyor belt system to a plurality of sorting bins. It may include a mechanism. For example, a sorting device may utilize air jets, each air jet being assigned to one or more classifications. When one of the air jets (e.g., 127) receives a signal from the automatic control system 108, that air jet directs the pieces of material 101 from the conveyor system 103 to its corresponding sorting bin (e.g., 137). Release the air flow to be diverted/exhausted.

Robotically remove pieces of material from the conveyor belt, push pieces of material from the conveyor belt (e.g., using a paintbrush type plunger), open any openings (e.g., trap doors) into which pieces of material may fall into the conveyor system 103 or using other mechanisms to redirect/eject material pieces, such as using air jets to redirect pieces of material into different bins as they fall off the end of a conveyor belt. You can also. The term pusher device, as used herein, refers to any suitable type of mechanical pushing mechanism (such as an ACME screwdrive), pneumatic pushing mechanism, or air jet pushing mechanism, pneumatic, mechanical, or other means. can be used to refer to any form of device that can be actuated to dynamically move objects on or from a conveyor system/device. Some embodiments may include multiple pusher devices located at different locations and/or having different turning path orientations along the path of the conveyor system. In various different implementations, these sorting systems described herein can determine which pusher devices (if any) to activate depending on the classification of the material pieces performed by the machine learning system. can. Further, the determination of which pusher device to actuate may be based on the detected presence and/or characteristics of other objects that may be present in the turning path of the pusher device at the same time as the target item. Additionally, even in installations where singulation along the conveyor system is not perfect, the disclosed sorting system will recognize when multiple objects are not properly separated and may separate objects in close proximity to A pusher device to be activated can be dynamically selected from a plurality of pusher devices based on the pusher device that provides a suitable turning path. In some embodiments, the object identified as the target object may represent material to be diverted from the conveyor system. In other embodiments, objects identified as target objects represent material that should be allowed to remain on the conveyor system so that non-target objects can be redirected instead.

In addition to the N sorting bins 136...139 into which pieces of material 101 are diverted/discharged, the system 100 also stores pieces of material that have not been sorted/discharged from the conveyor system 103 into any of the aforementioned sorting bins 136...139. A container or bin 140 for receiving 101 may also be included. For example, a piece of material 101 is transferred from the conveyor system 103 to the N sorting bins 136... 139 may not be diverted/ejected. Therefore, bin 140 may serve as a default container into which unsorted material pieces are dumped. Alternatively, bin 140 may be used to receive one or more classifications of material pieces that are not intentionally assigned to any of the N sorting bins 136...139. These pieces of material may be further sorted according to other characteristics and/or by another sorting system.

Depending on the different classifications of desired material pieces, multiple classifications can be mapped to a single sorting device and associated sorting bins. In other words, there does not need to be a one-to-one correlation between classification and sorting bins. For example, a user may desire to separate certain categories of materials into the same sorting bin (eg, different types of plastics included in the fractions, etc.). To accomplish this sorting, when pieces of material 101 are classified as falling into a predetermined classification group (eg, fraction), the same sorting device can be operated to separate them into the same sorting bin. Such combinatorial fractionation may be applied to produce any desired combination of fractionated material pieces. The classification mapping may be programmed by a user (eg, using a sorting algorithm operated by computer system 107 (see, eg, FIG. 7)) to generate such desired combinations. Additionally, the classification of material pieces is user-definable and is not limited to any particular known classification of material pieces (eg, fractions as disclosed herein).

Conveyor system 103 may include a carousel (not shown) so that unsorted pieces of material are returned to the beginning of system 100 and passed through system 100 again. Further, the system 100 can specifically track each piece of material 101 as it moves on the conveyor system 103 so that the system 100 can sort it after a predetermined number of cycles by the system 100 (or after a piece of material 101 has been collected in the bin 140). Some sorting device (eg, sorting device 129) may be implemented to guide/eject the material pieces 101 so as not to fail.

In certain embodiments of the present disclosure, the conveyor system 103 has a first belt conveying pieces of material past a vision system 110 and a second belt carrying a sensor system 120 implemented for a second sorting. It may also be divided into a plurality of belts arranged in series, for example two belts, which convey particular separated pieces of material therethrough. Furthermore, such second conveyor belt may be at a lower height than the first conveyor belt such that pieces of material fall from the first belt onto the second belt.

In certain embodiments of the present disclosure implementing sensor system 120, light source 121 may be located above the detection area (i.e., above conveyor system 103). However, certain embodiments of the present disclosure may place the light source 121 and/or detector 124 at other locations that still produce acceptable sensed/detected physical characteristics.

The systems and methods described herein can be applied to sort and/or separate individual pieces of material having any of a variety of sizes and shapes. Although the systems and methods described herein are primarily described in connection with the separation of individual pieces of material, the systems and methods described herein are not limited thereto. Such systems and methods can be used to stimulate and/or detect release from multiple materials simultaneously. For example, multiple singulated streams may be conveyed in parallel, as opposed to streams of singulated material being conveyed in series along one or more conveyor belts. . Each stream may be on the same belt or on different belts arranged in parallel. Further, the pieces of material may be randomly distributed on (eg, across and along) one or more conveyor belts. Accordingly, the systems and methods described herein can be used to simultaneously stimulate and/or detect radiation from multiple pieces of material. In other words, multiple pieces of material may be treated as a single part, rather than each piece of material being considered individually. Accordingly, multiple pieces of material can be sorted and sorted together (eg, diverted/ejected from a conveyor system).

Although the systems and methods described herein are primarily described in connection with the separation of material pieces, such systems and methods are not limited to that application. These may be used for other applications, such as identifying elements within a piece of material (eg, contaminants, etc.) and determining the composition of a piece of material.

As mentioned above, certain embodiments of the present disclosure may implement one or more vision systems (eg, vision system 110) to identify, track, and/or classify pieces of material. According to embodiments of the present disclosure, such a vision system may operate alone to identify and/or classify and separate pieces of material, or may operate to identify and/or classify and separate pieces of material. may operate in combination with one or more sensor systems (eg, sensor system 120). If a sorting system (e.g., system 100) is configured to operate solely with such vision system 110, sensor system 120 may be omitted from system 100 (or may simply be turned off).

Regardless of the sensed characteristics/type of information captured of the pieces of material, the information (e.g., image data packets) is then sent to a computer system (e.g., a computer) to identify and/or classify each piece of material. system 107) and may be processed by the machine learning system. Such machine learning systems include neural networks (e.g., artificial neural networks, deep neural networks, convolutional neural networks, recurrent neural networks, autoencoders, reinforcement learning, etc.), fuzzy logic, artificial intelligence ("AI"), and deep learning. algorithms, deep structured learning hierarchical learning algorithms, support vector machines (“SVM”) (e.g., linear SVM, nonlinear SVM, SVM regression, etc.), decision tree learning (e.g., classification and regression trees (“CART”)), ensemble techniques (e.g. ensemble learning, random forests, bagging and pasting, patches and subspaces, boosting, stacking, etc.), dimensionality reduction (e.g. projection, manifold learning, principal component analysis, etc.) and/or deep machine learning algorithms, ref. deeplearning.net website (including all software, publications, and hyperlinks to available software referenced within this website), which is incorporated herein by Any well-known machine learning system can be implemented, including those that implement Limited examples include Python, OpenCV, Inception, Theano, Torch, PyTorch, Pylearn2, Numpy, Blocks, TensorFlow, MXNet, Caffe, Lasagne, Keras, Chainer, Matla b Deep Learning, CNTK, MatConvNet (for computer vision applications) MATLAB® toolbox for implementing convolutional neural networks), DeepLearnToolbox (Matlab toolbox for deep learning (manufactured by Rasmus Berg Palm)), BigDL, Cuda-Convnet (convolutional (more generally feedforward) neural networks) Deep Belief Networks, RNNLM, RNNLIB-RNNLIB, matrbm, deeplearning4j, Elearn.lsh, deepmat, MShadow, Matplotlib, SciPy, CXXNET, Nengo-Nengo, Elearn, cudamat, Gnumpy, 3-way factorization RBM and mcRBM, mPoT (Python code to train models of natural images using CUDAMat and Gnumpy), ConvNet, Elektronn, OpenNN, NeuralDesigner, Theano Generalized Hevian Learning, Apache Singa, Lightnet, and One example is SimpleDNN.

According to certain embodiments of the present disclosure, machine learning may be performed in two stages. For example, although training is initially performed, this training can be performed offline in that the system 100 is not utilized to perform the actual sorting/separation of material pieces (e.g., FIGS. 3-4 ). The system 100 allows a homogeneous set (also referred to herein as a control sample) of pieces of material (i.e., having the same type or class of material or falling within the same predetermined fraction) to be collected (e.g., by conveyor system 103). ) passing through the system 100 and training the machine learning system such that all such pieces of material may not be separated, but may be collected in a common bin (e.g., bin 140, etc.) can be used for. Alternatively, training may be performed at another location remote from system 100, including using some other mechanism to collect sensed information (properties) of a control set of pieces of material. During this training phase, algorithms within the machine learning system extract features from the captured information (eg, using image processing techniques well known in the art). Non-limiting examples of training algorithms include, but are not limited to, linear regression, gradient descent, feedforward, polynomial regression, learning curves, regularized learning models, and logistic regression. An algorithm within the machine learning system learns the relationships between materials and their features/properties (e.g., those captured by the vision system and/or sensor system) that are received by the system 100 and then classified according to the desired classification. It is this training phase that creates a knowledge base for later classification of the heterogeneous mixture of material pieces that can be separated. Such a knowledge base may include one or more libraries, each library containing parameters (e.g., neural network parameters) utilized by the machine learning system in classifying pieces of material. included. For example, one particular library may include parameters configured by a training stage to recognize and classify one or more materials of a particular type or class, or of a predetermined fraction. According to certain embodiments of the present disclosure, such a library may be input into a machine learning system, and a user of system 100 may adjust certain of the parameters to adjust the operation of system 100. (e.g., adjusting threshold effectiveness for how well a machine learning system recognizes a particular piece of material from a heterogeneous mixture of materials).

As shown in FIG. 2, during the training phase, a plurality of pieces of material 201 of one or more particular types, classifications, or fractions of material are control samples, and the algorithm within the machine learning system or may be delivered (eg, by conveyor system 203) through a vision system and/or one or more sensor systems to detect, extract, and learn features representative of a class of materials. For example, each of the pieces of material 201 has gone through such a training step that an algorithm within a machine learning system has undergone such a training step to detect, recognize, and classify such pieces of plastic. individual plastic pieces of a type, class, or predetermined fraction. When training a vision system (eg, vision system 110), it is trained to visually identify pieces of material. This creates a library of parameters specific to one or more particular types, classes, or fractions of plastic materials. The same processing can then be performed on different types, classes, or fractions of plastic pieces, creating a library of parameters specific to that type, class, or fraction, and so on. For each type, class, or fraction of plastic classified by the machine learning system, any number of example plastic pieces of that type, class, or fraction of plastic may be passed through the system. Given the captured sensing information as input data, the algorithm within the machine learning system uses N classifiers, each testing one of N different material types, classes, or fractions. The machine learning system is configured to detect any type, class, or fraction of material, including any type, class, or fraction of material found within MSW or any other material disclosed herein. Note that a person can be “educated” (trained).

Once the algorithm is established and the machine learning system has learned (trained) enough (e.g., within a user-defined statistical confidence level) to differentiate between material classifications (e.g., visually discernible differences), it then The library is implemented in a material classification/segregation system (e.g., system 100) used to identify and/or classify material pieces from a heterogeneous mixture of material pieces (e.g., as contained within MSW). , and, if separation is performed, such sensitive pieces of material may be separated.

Techniques for constructing, optimizing, and utilizing machine learning systems are known to those skilled in the art, as described in the relevant literature. Examples of such literature include the following publications: Krizhevsky et al., "ImageNet Classification with Deep Convolutional Networks," Proceedings of the 25th International Conference on Neural Information Processing Systems, December 3-6, 2012, Nevada and LeCun et al., "Gradient-Based Learning Applied to Document Recognition," Proceedings of the IEEE, Institute of Electrical and Electronics Engineers (IEEE), November 1998, both of which are incorporated by reference in their entirety. incorporated into the book.

In one example technique, data captured by a visual or sensor system regarding a particular piece of material is transmitted to a data processing system (configured) implementing a machine learning system (e.g., data processing system 3400 of FIG. 9). ) can be treated as an array of data values. For example, the data may be spectral data captured by a digital camera or other type of sensor system for a particular piece of material and processed as an array of data values (eg, an image data packet). Each data value can be represented as a single number or as a series of numbers representing the value. These values may be multiplied by the neuron's weight parameters (eg, using a neural network) and a bias may be added. This can affect the nonlinearity of neurons. The resulting number output by a neuron can be treated like a value by multiplying this output by the weight value of the subsequent neuron, optionally adding a bias, and again affecting the nonlinearity of the neuron. I can do it. Each iteration of such processing is known as a "layer" of the neural network. The final output of the final layer may be interpreted as the probability of the presence or absence of material in the captured data associated with the piece of material. Examples of such processing can be found in the aforementioned references “ImageNet Classification with Deep Convolutional Networks” and “Gradient-Based Learning Applied to Document Recognition”. ion” are explained in detail in both.

According to certain embodiments of the present disclosure, in which the neural network is implemented as a final layer (a "classification layer"), the final set of neuron outputs represents the probability that a piece of material is associated with the captured data. be trained as such. In operation, it is determined that the piece of material is actually associated with the captured data if the probability that the piece of material is associated with the captured data exceeds a user-specified threshold. These techniques can be extended to determine whether a subregion of a particular captured data belongs to one type of material or another. It is also possible to determine whether the material belongs to a type of material. This process is known as segmentation, and can be applied to neural networks, such as those known as "fully convolutional" neural networks, or networks that contain convolutional parts (i.e., are partially convolutional), even if they are not fully convolutional. Techniques exist in the literature that use . This allows the location and size of the material to be determined.

It should be understood that this disclosure is not limited only to machine learning techniques. Other common techniques for material classification/identification can also be used. For example, sensor systems utilize spectroscopic techniques using multispectral or hyperspectral cameras to detect the presence or absence of a type, class, or fraction of a material by examining the spectral emissions of the material (i.e., spectral imaging). A signal indicating absence may be provided. A spectral image of a piece of material may also be used in a template matching algorithm, in which a database of spectral images is compared to the acquired spectral image and the presence or absence of a particular type of material is detected from the database. The histogram of the captured spectral image can also be compared to a database of histograms. Similarly, bag-of-words models can be used with feature extraction techniques such as scale-invariant feature transforms (“SIFT”) to compare extracted features between captured spectral images and spectral images in a database. You can also do it.

Accordingly, as disclosed herein, certain embodiments of the present disclosure utilize one or more Provides identification/classification of multiple different types, classes or fractions of materials. According to certain embodiments, machine learning techniques are utilized to train (i.e., configure) a neural network to identify one or more different types, classes, or fractions of a variety of materials. . Spectral images and other types of sensing information are captured from materials (e.g., those moving on a conveyor system), and based on the identification/classification of such materials, the systems described herein identify which materials. Pieces can be left on the conveyor system and it can be determined which pieces of material should be diverted/removed from the conveyor system (e.g., either placed in a collection bin or diverted to another conveyor system) ).

In accordance with certain embodiments of the present disclosure, a machine learning system for an existing facility (e.g., system 100) is configured to generate new types of materials by replacing a current set of neural network parameters with a new set of neural network parameters. , classes, or fractions can be dynamically reconfigured to identify/classify characteristics.

One point to mention here is that according to certain embodiments of the present disclosure, the detected/captured features/properties (e.g., spectral images) of a piece of material are not necessarily merely particularly distinguishable or distinguishable. They do not have to be physical properties; they may be abstract formulas that can only be expressed mathematically, or they may not be expressible mathematically at all; nevertheless, machine learning systems use spectral data. can be configured to look for patterns that can classify the control samples during the training phase. Additionally, the machine learning system may take subsections of the captured information (eg, spectral images, etc.) of the piece of material and attempt to find correlations between predefined classifications.

According to certain embodiments of the present disclosure, instead of utilizing a training phase in which control samples of material pieces are passed by a vision system and/or sensor system, training of the machine learning system utilizes labeling/annotation techniques. The data/information of the material pieces is captured by the vision/sensor system and the user is provided with a label or annotation that identifies each piece of material to categorize the pieces of material within a heterogeneous mixture of material pieces. Enter labels or annotations that are used to create libraries used by machine learning systems.

Referring to FIGS. 3-6, embodiments of the present disclosure utilize multiple sensors in a manner that uniquely identifies various types, classes, or fractions of plastics, such that they can be separated by organic and inorganic chemical composition. Combining or fusing technologies, such as any combination of visual (“VIS”), XRF, NIR, and MWIR. However, because these pieces of plastic within the MSW come in different sizes and shapes, the signals generated from these various sensors can have large differences between sensors. Therefore, the combination of machine learning and the fusion of various sensor technologies will improve the classification accuracy of these signals even in the presence of such large variance. Because implementing multiple different sensors in a system can increase the cost of the system and also reduce the sorting speed, certain embodiments of the present disclosure are designed to implement fewer sensor systems (and thus lower A system (e.g., system 100) can be implemented to increase economic viability (eg, system 100) with low capital costs and operating costs, yet still adequately segregate materials.

FIG. 4 is a simplified schematic of a system (e.g., system 100) in which pieces of material (e.g., pieces of plastic) 401 are conveyed by a conveyor system 403 past a sensor system that captures spectral data from each piece of material 401. Show the diagram. In this non-limiting example, the sensor systems are a camera 410 (eg, vision system 110) that captures visible image data of each piece of material 401, an XRF system 411, a NIR system 412, and a MWIR system 413. However, it is noted that any of the other sensor systems disclosed herein can be utilized in any combination.

Referring to FIGS. 3 and 4, chemical characteristics of the material are determined at processing block 301 using one or more sensor systems. The sensed/detected/captured signals from the sensor systems are combined (eg, in a multidimensional data array) for each material to create a chemical signature. An XRF sensor system can determine the presence of inorganic elements or molecules within a piece of plastic, while a combination of one or more other sensor systems such as NIR or MWIR can determine the presence of organic elements or molecules within a piece of plastic. I want to be reminded that I can make decisions. At processing block 302, a visual image of each piece of material is captured. At processing block 303, the captured visual image of each piece of material (ie, its associated image data) is associated with its determined chemical characteristics (ie, spectral image data). 5 and 6 show non-limiting exemplary representations of chemical characteristics and associated image data for two different types of plastic materials (a bag of potato chips and an electronics package). As can be easily seen, different types or classes of plastic pieces have different (unique) chemical characteristics, and these generate plastic waste segregation and/or classification (which may be user-defined). Utilized within embodiments of the present disclosure to. According to embodiments of the present disclosure, a control group of a particular type or class of plastic pieces is used to train a machine learning system to associate a particular chemical characteristic with a particular type or class of plastic pieces. This can be done through the system shown below.

For example, for the example shown in FIG. Images may be processed to train machine learning systems.

Processing block 304 may include separating the plastic pieces into one or more fractions. There are various ways to create these fractions. One method is to create a first layer based on major elements, then a second, or even third layer based on minor elements. For example, fractions can be determined first by polymer type and then branched into inorganic elements such as aluminum or zinc. Other exemplary fractions can then be created for blends of polymers and branched into their inorganic elemental compositions. There are also computational techniques such as principal component analysis, K-means clustering, unsupervised learning and semi-semi-supervised learning to perform this type of clustering and determine fractions. Fractions are further defined herein.

At processing block 305, after the fractions have been determined, the plastic pieces associated with the fractions can be separated (eg, manually) to create a control group for each fraction. Because each fraction was measured with a sensor system, each control group contains chemical information about that piece. A vision system (eg, vision system 110) may be used to train a machine learning system to identify those fractions. Using this method, chemical data within the plastic is converted into visual features that a machine learning system can learn and classify. Performing classification based on visual images using system 100 also separates plastics by chemical composition. This method works when two objects look different and have different chemical compositions. If two objects are the same or very similar but have different chemical compositions, two or more sensor systems (eg, VIS and XRF, etc.) may be used to perform the classification.

Because the determined fractions may constitute any desired variety of specific organic and/or inorganic elements or molecules, the process 300 separates a heterogeneous mixture of different plastic pieces into one or more different types. or may be utilized to train a machine learning system implemented within a sorting system to be configured to generate at least one fraction containing plastic particles of a class. For example, if a machine learning system is trained to identify any piece of plastic that contains a specified combination of organic and/or inorganic elements or molecules, then once the separation is complete, the separated fractions will contain: Multiple plastic chip bags related to different brands of chips, as all plastic pieces are not identical (i.e. each plastic chip bag is composed of organic and/or inorganic elements or molecules defined by a given fraction) ) may be included.

FIG. 7 depicts a flowchart diagram illustrating an exemplary embodiment of a process 3500 for classifying/separating pieces of material utilizing a vision system and/or one or more sensor systems, in accordance with certain embodiments of the present disclosure. . Process 3500 may be performed to classify a heterogeneous mixture of plastic pieces into any combination of predetermined types, classes, and/or fractions. Process 3500 may be configured to operate within any embodiment of the disclosure described herein, including system 100 of FIG. The operations of process 3500 may include hardware and/or hardware, including within a computer system (e.g., computer system 3400 of FIG. 9) that controls the system (e.g., computer system 107, vision system 110, and/or sensor system 120 of FIG. 1). or may be performed by software. At process block 3501, pieces of material may be deposited onto a conveyor system. At processing block 3502, the position of each piece of material on the conveyor system is detected to track each piece of material as it moves through system 100. This may be performed by vision system 110 (e.g., by distinguishing pieces of material from underlying conveyor system material while communicating with a conveyor system position detector (eg, position detector 105)). Alternatively, material tracking device 111 may be used to track pieces of material. Alternatively, a system can be used that can generate a light source (including but not limited to visible light, UV, and IR) and includes a detector that can be used to locate the piece of material. At processing block 3503, sensed information/characteristics of the piece of material are captured/obtained as the piece of material moves near one or more of the vision system and/or sensor system. At processing block 3504, a vision system (e.g., implemented within computer system 107) as described above may perform pre-processing of the captured information, utilizing the captured information to identify the piece of material. Respective information can be detected (extracted) (e.g. from the background (e.g. conveyor belt); in other words, pre-processing can be used to distinguish between pieces of material and the background). Well-known image processing techniques such as dilation, thresholding, and contouring can be utilized to identify pieces of material as distinct from the background. At processing block 3505, segmentation may be performed. For example, the captured information may include information regarding one or more pieces of material. Additionally, certain pieces of material may be located at the seams of the conveyor belt when the image is captured. Therefore, in such cases it may be desirable to separate images of individual pieces of material from the background of the image. In the exemplary technique of processing block 3505, the first step is to apply high contrast of the image, such that the background pixels are reduced to substantially all black pixels, at least for the piece of material. Some pixels are brightened to substantially all white pixels. The white image pixels of the piece of material are expanded to cover the size of the entire piece of material. After completing this step, the position of the material piece becomes a high-contrast image with all white pixels on a black background. A contour algorithm can then be utilized to detect the boundaries of the piece of material. Boundary information is saved and the border position is transferred to the original image. Segmentation is then performed on regions of the original image that are larger than the previously defined boundaries. In this way, pieces of material are identified and separated from the background.

At optional processing block 3506, the pieces of material are placed on a conveyor near a piece tracking device and/or sensor system to track each piece of material and/or to determine the size and/or shape of the piece of material. may be carried along the system, which may be useful if an XRF system or other spectroscopic sensor is also implemented within the fractionation system. At processing block 3507, post-processing may be performed. Post-processing may include resizing the captured information/data for use in a machine learning system. This post-processing involves changing certain characteristics (e.g., enhancing image contrast, changing the image background, applying filters, etc.) in a way that enhances the machine learning system's ability to classify the pieces of material. ) may also be included. At processing block 3509, the size of the data may be changed. Under certain circumstances, it may be desirable to resize data to match the data input requirements of a particular machine learning system, such as a neural network. For example, neural networks may require image sizes that are much smaller than the size of images captured with typical digital cameras (eg, 225x255 pixels or 299x299 pixels). Additionally, the smaller the size of the input data, the less processing time is required to perform the classification. Therefore, smaller data sizes may ultimately improve the throughput of system 100 and increase its value.

At processing blocks 3510 and 3511, each piece of material is identified/classified based on the sensed/detected characteristics. For example, processing block 3510 uses one or more machine learning algorithms to compare the extracted features to previously generated features stored in a knowledge base (e.g., generated during a training phase). and, based on such comparison, assigns a best matching classification to each of the pieces of material. Machine learning system algorithms may process the captured information/data hierarchically using automatically trained filters. The filter responses are successfully combined in the next level of the algorithm until the probabilities are obtained in the final step. At processing block 3511, these probabilities may be used for each of the N classifications to determine which of the N sorting containers each piece of material should be sorted into. For example, each of the N classifications can be assigned to one sorting container, and the piece of material under consideration is sorted into that container corresponding to the classification that returns a probability greater than a predefined threshold. In embodiments of the present disclosure, such predetermined threshold may be preset by the user. If neither probability is greater than a predetermined threshold, the particular piece of material may be classified into an outlier bin (eg, sort bin 140).

Next, at processing block 3512, the sorting device corresponding to the classification of the material pieces is activated. Between the time the image of the material piece was captured and the time the sorter was activated, the material piece moved from near the vision and/or sensor system to a location downstream of the conveyor system (e.g. speed). In embodiments of the present disclosure, activation of the sorter is such that the sorter is activated when a piece of material passes a sorter that is mapped to the classification of that piece of material, and the piece of material is directed from the conveyor system to the associated sorting container. Timed for conversion/emission. In embodiments of the present disclosure, activation of the sorting device may be timed by respective position detectors that detect when pieces of material pass in front of the sorting device and transmit signals that enable activation of the sorting device. . At process block 3513, the sorting container corresponding to the activated sorting device receives the diverted/discharged material pieces.

FIG. 8 depicts a flowchart diagram illustrating an exemplary embodiment of a process 800 for separating pieces of material, according to certain embodiments of the present disclosure. Process 800 may be configured to operate within any embodiment of the disclosure described herein, including system 100 of FIG. Process 800 may be configured to operate in conjunction with process 3500. For example, in accordance with certain embodiments of the present disclosure, processing blocks 803 and 804 may be configured such that the efforts of vision system 110 implemented in conjunction with a machine learning system to classify and/or separate pieces of material are performed using machine learning. It can be incorporated into processing 3500 (eg, operating in series or parallel with processing blocks 3503-3510) for combination with sensor systems (eg, sensor system 120) that are not implemented in conjunction with the system.

The operations of process 800 may be performed by hardware and/or software, including within a computer system (eg, computer system 3400 of FIG. 9) controlling a system (eg, computer system 107 of FIG. 1). At process block 801, pieces of material may be deposited onto a conveyor system. Next, at optional processing block 802, the pieces of material are placed on a piece of material tracking device and/or an optical imaging system to track each piece of material and/or determine the size and/or shape of the piece of material. It may be transported along a conveyor system within the vicinity. At processing block 803, as the piece of material moves near the sensor system, the piece of material is probed with EM energy (waves) or other type of stimulus appropriate for the particular type of sensor technology utilized in the sensor system. , can be stimulated. At processing block 804, physical characteristics of the piece of material are sensed/detected and captured by a sensor system. At processing block 805, for at least a portion of the pieces of material, the type of material is identified/classified based (at least in part) on the captured characteristics, combined with classification by a machine learning system in conjunction with vision system 110. I can do it.

Next, if sorting of the material pieces is to be performed, at processing block 806, the sorting device corresponding to the classification of the material pieces is activated. Between sensing a piece of material and activating the sorter, the piece of material moved from near the sensor system to a position downstream of the conveyor system at the conveying speed of the conveyor system. In certain embodiments of the present disclosure, activation of the sorting device is such that when a piece of material passes through a sorting device mapped to a classification of the material piece, the sorting device is activated and the piece of material is transferred from the conveyor system to its associated sorting container. Timed to change direction/eject. In certain embodiments of the present disclosure, activation of the sorting device is timed by respective position detectors that detect when a piece of material passes in front of the sorting device and send a signal that enables activation of the sorting device. can be done. At process block 807, the sorting container corresponding to the activated sorting device receives the diverted/discharged material pieces.

According to certain embodiments of the present disclosure, multiple at least portions of system 100 may be linked together in series to perform multiple iterations or layers of fractionation. For example, when two or more systems 100 cooperate in this manner, the conveyor system can be implemented with a single conveyor belt or multiple conveyor belts, and the conveyor system can deliver pieces of material to a first set of pieces of material of a heterogeneous mixture of materials. into a first set of one or more containers (e.g., sorting bins 136...139) by a sorting device (e.g., first automatic control system 108 and associated one or more sorting devices 126...129). conveying the pieces of material past a first vision system (and, according to certain embodiments, a sensor system) configured to a second vision system (and, according to certain embodiments, another sensor system) configured to sort the two sets of material pieces into a second set of one or more sorting bins; and transport it. Further discussion of such multi-step fractionation is provided in US Patent Application Publication No. 2022/0016675, which is incorporated herein by reference.

Such a series 100 may include any number of such systems linked together in such a manner. According to certain embodiments of the present disclosure, each continuous system may be configured to separate a different classification or type of material than the previous system.

According to various embodiments of the present disclosure, different types, classes, or fractions of materials are each classified by different types of sensors for use in a machine learning system, and pieces of material in a scrap or waste stream are classified by different types of sensors for use in a machine learning system. can be combined to classify.

According to various embodiments of the present disclosure, data (e.g., spectral data) from two or more sensors is combined using a single or multiple machine learning systems to perform classification of material pieces. be able to.

According to various embodiments of the present disclosure, multiple sensor systems can be attached to a single conveyor system, and each sensor system can utilize a different machine learning system. According to various embodiments of the present disclosure, multiple sensor systems can be attached to different conveyor systems, and each sensor system can utilize a different machine learning system.

Certain embodiments of the present disclosure may be configured to produce a mass of material having less than a predetermined weight or volume percent content of a particular element or material after fractionation.

According to various embodiments of the present disclosure, any combination of different types of sensor systems may be utilized to identify/classify and possibly separate materials disclosed herein. For example, each image or spectroscopic sensor disclosed herein is used to generate data from information/properties sensed from a piece of material, as processed by a machine learning system specific to that sensor system. can be done. Alternatively, any sensor system can be used without processing by a machine learning system, with processing by a machine learning system, or a combination of both.

According to various embodiments of the present disclosure, different types, classes, and/or fractions of materials are each classified by different types of sensor systems for use in machine learning systems, and pieces of material in a waste stream are classified by different types of sensor systems for use in machine learning systems. can be combined to classify.

Referring now to FIG. 9, a block diagram illustrating a data processing ("computer") system 3400 in which aspects of embodiments of the present disclosure may be implemented is shown. (The terms "computer," "system," "computer system," and "data processing system" may be used interchangeably herein). Aspects of computer system 107, automatic control system 108, sensor system 120, and/or vision system 110 may be configured similarly to computer system 3400. Computer system 3400 can use a local bus 3405 (eg, peripheral component interconnect ("PCI") local bus architecture). Any suitable bus architecture may be utilized, such as Accelerated Graphics Port ("AGP") or Industry Standard Architecture ("ISA"), among others. One or more processors 3415, volatile memory 3420, and non-volatile memory 3435 may be connected to local bus 3405 (eg, via a PCI bridge (not shown)). An integrated memory controller and cache memory may be coupled to one or more processors 3415. One or more processors 3415 may include one or more central processor units and/or one or more graphics processor units 3401 and/or one or more tensor processing units. Additional connections to local bus 3405 can be made through direct interconnection of components or add-in boards. In the illustrated example, a communications (e.g., network (LAN)) adapter 3425, an I/O (e.g., small computer system interface ("SCSI") host bus) adapter 3430, and an expansion bus interface (not shown) are configured The elements can be connected to local bus 3405 by direct connections. An audio adapter (not shown), a graphics adapter (not shown), and a display adapter 3416 (coupled to display 3440) are connected to local bus 3405 (e.g., by an add-in board inserted into an expansion slot). obtain.

User interface adapter 3412 can provide connections for a keyboard 3413 and mouse 3414, a modem (not shown), and additional memory (not shown). I/O adapter 3430 can provide connections to a hard disk drive 3431, tape drive 3432, and CD-ROM drive (not shown).

An operating system runs on one or more processors 3415 and may be used to coordinate and control various components within computer system 3400. In FIG. 9, the operating system may be a commercially available operating system. An object-oriented programming system (eg, Java, Python, etc.) can be executed in conjunction with the operating system to provide calls to the operating system from programs (eg, Java, Python, etc.) running on system 3400. Instructions for the operating system, object-oriented operating system, and programs may be located on non-volatile memory 3435 storage, such as a hard disk drive 3431, and loaded into volatile memory 3420 for execution by processor 3415. Good too.

Those skilled in the art will understand that the hardware in FIG. 9 may vary depending on the implementation. In addition to or in place of the hardware shown in FIG. 9, other internal hardware or peripherals, such as flash ROM (or equivalent non-volatile memory) or optical disk drives, may also be used. Additionally, any of the processes of this disclosure may be applied to a multiprocessor computer system or performed by multiple such systems 3400. For example, training of the machine learning system may be performed by the first computer system 3400, while operations of the system 100 for classification may be performed by the second computer system 3400.

As another example, computer system 3400 may be a standalone device configured to be bootable independent of some type of network communication interface, whether or not computer system 3400 includes any type of network communication interface. It may be a system. As a further example, computer system 3400 may be an embedded controller comprised of ROM and/or flash ROM that provides non-volatile memory for storing operating system files or user-generated data.

The example shown in FIG. 9 and the examples above are not intended to imply any architectural limitations. Additionally, the computer program form of aspects of the present disclosure may reside on any computer-readable storage medium used by a computer system (i.e., floppy disk, compact disk, hard disk, tape, ROM, RAM, etc.).

As described herein, embodiments of the present disclosure may be implemented to perform various functions described for identifying, tracking, classifying, and/or separating pieces of material. Such functionality may be implemented in aspects of the aforementioned computer system 107, vision system 110, sensor system 120, and/or automated control system, such as within one or more data processing systems (e.g., data processing system 3400 of FIG. 9). It can be implemented in hardware and/or software, such as 108. However, the functionality described herein is not limited to implementation on any particular hardware/software platform.

As will be understood by those skilled in the art, aspects of the present disclosure may be embodied as systems, processes, methods, and/or computer program products. Accordingly, various aspects of the present disclosure may be implemented in an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, microcode, etc.), or as generally referred to herein as a "circuit." Embodiments may combine software and hardware aspects called "circuitries," "modules," or "systems." Additionally, aspects of the present disclosure may take the form of a computer program product embodied in one or more computer readable storage media having computer readable program code embedded therein. (However, any combination of one or more computer-readable media may be utilized. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium.)

The computer-readable storage medium may be, for example, an electronic, magnetic, optical, electromagnetic, infrared, biological, atomic, or semiconductor system, apparatus, controller, or device, or any suitable combination thereof. Without limitation, the computer-readable storage medium itself is not a transitory signal. More specific examples (non-exhaustive list) of computer-readable storage media include electrical connections having one or more wires, portable computer diskettes, hard disks, random access memory ("RAM") (e.g., FIG. RAM 3420), read-only memory ("ROM") (e.g., ROM 3435 in Figure 9), erasable programmable read-only memory ("EPROM" or flash memory), fiber optic, portable compact disk read-only memory ("CD-ROM"), ''), optical storage, magnetic storage (eg, hard drive 3431 in FIG. 9), or any suitable combination of the foregoing. In the context of this document, a computer-readable storage medium is any tangible medium that contains or can store a program for use by or in connection with an instruction execution system, apparatus, controller, or device. good. Program code embodied on a computer readable signal medium can be implemented using any suitable medium, including but not limited to wireless, wired, fiber optic cable, RF, etc., or any suitable combination of the foregoing. can be sent.

A computer-readable signal medium may include a propagating data signal having computer-readable program code embodied therein, eg, as part of a baseband or carrier wave. Such propagated signals may take any of a variety of forms including, but not limited to, electromagnetic, optical, or any suitable combination thereof. A computer-readable signal medium is any computer-readable medium that is not a computer-readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, controller, or device. It's good.

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, processes, and program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code that includes one or more executable program instructions for implementing the specified logical functions. Note also that, depending on the implementation, the functions depicted in the blocks may be performed out of the order shown in the figures. For example, two blocks shown in succession may actually be executed substantially simultaneously, or the blocks may be executed in the reverse order, depending on the functionality involved.

In the discussion herein, flowchart techniques may be described as a series of sequential operations. The order of the operations and the parties performing the operations may be changed at will without departing from the scope of the teachings. Operations may be added, deleted, or modified in several ways. Similarly, operations can be reordered or looped. Further, although processes, methods, algorithms, etc. may be described in a sequential order, such processes, methods, algorithms, or any combination thereof may be operable to be performed in a different order. There may be. Additionally, some actions within a process, method, or algorithm may be performed simultaneously (e.g., actions performed in parallel) at least at some point, in whole, in part, or in any combination thereof. May be executed.

Software-implemented modules executed by various types of processors (e.g., GPU 3401, CPU 3415) may include, for example, one or more physical blocks of computer instructions that may be organized as objects, procedures, or functions, or Logic blocks may be included. However, the executable files of the identified modules do not have to be physically located together, and may contain disparate instructions stored in different locations, which are logically combined. and incorporate the module to achieve the module's specified purpose. In fact, a module of executable code may be a single instruction, or many instructions, and may be distributed across different code segments, between different programs, and across multiple memory devices. . Similarly, operational data (e.g., materials classification libraries described herein) may be identified and illustrated herein in modules, embodied in any suitable format, and of any suitable type. May be organized in data structures. Operational data may be collected as a single data set or distributed across different locations including different storage devices. The data may provide electronic signals over a system or network.

These program instructions are such that instructions executed via a processor (e.g., GPU 3401, CPU 3415) of a computer or other programmable data processing device perform the functions/operations specified in the blocks of the flowcharts and/or block diagrams. provided to one or more processors and/or controllers of a general purpose computer, special purpose computer, or other programmable data processing device (e.g., a controller) to create a circuit or means for implementing the machine. can be produced.

Each block in the block diagrams and/or flowchart diagrams, and combinations of blocks in the block diagrams and/or flowchart diagrams, may represent a dedicated hardware-based system (e.g., one or more graphical Note also that it can be implemented by a processor unit (which may include a processing unit (such as GPU 3401)) or by a combination of dedicated hardware and computer instructions. For example, a module can be implemented as a custom VLSI circuit or gate array, off-the-shelf semiconductors such as logic chips, or hardware circuits that include transistors, controllers, or other discrete components. Modules can also be implemented with programmable hardware devices such as field programmable gate arrays, programmable array logic, programmable logic devices, and the like.

Computer program code, or instructions, for performing operations of aspects of the present disclosure may be implemented in an object-oriented programming language such as Java, Smalltalk, Python, C++, the "C" programming language or similar programming language, or as disclosed herein. Machine learning software can be written in any combination of one or more programming languages, including traditional procedural programming languages, such as any machine learning software. The program code may be executed entirely on a user's computer system as a stand-alone software package, partially executed on the user's computer system, and partially executed on the user's computer system (e.g., a computer system used separately). It may be executed partially on a remote computer system (such as a computer system used to train a sensor system), or entirely on a remote computer system or server. In the latter scenario, the remote computer system may be connected to the user's computer system through any type of network, including a local area network ("LAN") or wide area network ("WAN"), or may be connected to an external It may also be connected to a computer system (eg, via the Internet using an Internet service provider).

These program instructions may be used by a computer system, other program products, etc., such that the instructions stored on a computer-readable medium produce a product that includes instructions that implement the functions/operations specified in the blocks of the flowcharts and/or block diagrams. The information may also be stored on a computer-readable storage medium that enables a capable data processing apparatus, controller, or other device to function in a particular manner.

Program instructions may also include a computer, computer, or other programmable device such that instructions executed on a computer or other programmable device provide operations for implementing the functions/operations specified in the blocks of the flowcharts and/or block diagrams. loaded into another programmable data processing device, controller, or other device to cause the computer, other programmable device, or other device to perform a sequence of operational steps to produce a computer-implemented process; You can also do that.

A host may include one or more databases to store and provide access to data for various implementations. Those skilled in the art will appreciate that for security reasons, any database, system, or component of this disclosure may include any combination of databases or components in a single location or multiple locations; It will also be appreciated that the database or system may include any of a variety of suitable security features, such as firewalls, access codes, encryption, decryption, and the like. The database can be any type of database, such as relational, hierarchical, object-oriented, etc. Common database products that can be used to implement a database include IBM's DB2, database products available from Oracle Corporation, Microsoft Access from Microsoft Corporation, or other database products. Databases can be organized in any suitable manner, such as data tables or lookup tables.

Specific data association (e.g., for each piece of material processed by the sorting system described herein) can be accomplished through any data association technique known and practiced in the art. For example, association can be performed manually or automatically. Automatic association techniques may include, for example, database searches, database merging, GREP, AGREP, SQL, and the like. The association step can be accomplished, for example, by a database merge function using key fields of the manufacturer's and retailer's respective data tables. Key fields partition the database according to the high-level class of objects defined by the key field. For example, a particular class can be designated as a key field in both a first data table and a second data table, and the two data tables can be merged based on the key field's class data. In these embodiments, the data corresponding to each key field of the merged data table is preferably the same. However, data tables that are not identical but have similar data in key fields can also be merged using, for example, AGREP.

As used herein, we refer to "configuring" a device or referring to a device "configured" to perform some function. It should be appreciated that this may include selecting predefined logical blocks and logically associating them to provide particular logical functions, including monitoring or control functions. It may also include programming the computer software-based logic of the retrofit controller, wiring individual hardware components, or a combination of any or all of the foregoing.

The description herein describes programming, software modules, user selections, network transactions, database queries, database structures, hardware modules, hardware circuits, hardware chips, to provide a thorough understanding of embodiments of the present disclosure. Many specific details are provided, such as examples of , controllers, etc. However, one of ordinary skill in the art will recognize that the present disclosure may be practiced without one or more of the specific details or using other methods, components, materials, and the like. In other instances, well-known structures, materials, or operations may not be shown or described in detail to avoid obscuring aspects of the disclosure.

Those skilled in the art will appreciate that the various settings and parameters of the components of system 100 (including neural network parameters) can be used to determine the types of materials to be sorted and sorted, the desired sorting and sorting results, and the type of equipment being used. , it should be understood that it may be customized, optimized, and reconfigured over time based on empirical results of previous classifications, data that becomes available, and other factors.

Throughout this specification, references to "one embodiment," "embodiment," or similar terminology indicate that a particular feature, structure, or characteristic described in connection with the embodiment is present in at least one implementation of the present disclosure. It means included in the form. Therefore, all appearances of the terms "in one embodiment," "in an embodiment," "embodiment," "particular embodiment," "various embodiments," and similar terms throughout this specification refer to the same embodiment. Although it may refer to the form, it is not necessarily the same thing. Additionally, the described features, structures, aspects, and/or characteristics of this disclosure may be combined in any suitable manner in one or more embodiments. Accordingly, in some cases one or more features may be removed from the claimed combination, even if the features were originally claimed to work in a particular combination. , the claimed combinations may be directed to subcombinations or variations of subcombinations.

Benefits, advantages, and solutions to problems are described herein with respect to specific embodiments. However, the benefit, advantage, solution to the problem, and the elements from which the benefit, advantage, solution may arise or become more prominent are important, necessary, or essential to some or all of the claims. or may not be interpreted as a required feature or element. Moreover, no component described herein is required to practice the present disclosure, unless explicitly described as essential or critical.

Although there are many details contained in this specification, these should not be construed as limitations on the scope of the disclosure or the claims, but rather on features specific to particular implementations of the disclosure. Should be construed as illustrative. The headings herein may not be intended to limit the disclosure, embodiments of the disclosure, or other matter disclosed under the heading.

As used herein, the term "or" is intended to be inclusive, and "A or B" includes A or B, as well as both A and B. As used herein, the term "and/or" when used in the context of a list of entities refers to the entities present alone or in combination. Thus, for example, the phrase "A, B, C, and/or D" includes A, B, C, and D individually, but includes all combinations and subcombinations of A, B, C, and D. Also included.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the disclosure. As used herein, the singular forms "a," "an," and "the" may be intended to include the plural forms as well, unless the context clearly dictates otherwise.

Corresponding structures, materials, acts, and all means or steps plus equivalents of functional elements in the following claims are specifically claimed for performing the functions in combination with other claim elements. It may be intended to include any structure, material, or operation.

As used herein, "controller", "processor", "memory", "neural network", "interface", "separator", "device", "pushing mechanism", "pusher apparatus", " Terms such as "image sensor," "bin," "receptacle," "system," and "circuit" each refer to non-common device elements that are recognized and understood by those skilled in the art, and are used herein to ．． S. C. 112(f) as an accidental or casual term.

As used herein with respect to a specified property or situation, "substantially" refers to a degree of deviation that is sufficiently small so as not to measurably impair the specified property or situation. The exact degree of deviation allowed may depend on the particular circumstances.

As used herein, for convenience, multiple items, structural elements, components, exemplary fractions, and/or materials may be presented in a common list. However, these lists should be interpreted as if each member of the list were individually identified as a separate and unique member. Accordingly, individual members of such a list should not be construed as being equivalent in nature to other members of the same list solely on the basis of their expression within a common group, without any indication to the contrary.

Unless otherwise defined, all technical and scientific terms (such as acronyms used for polymers or chemical elements in the periodic table) used herein will be understood by those skilled in the art to which the subject matter disclosed herein belongs. has the same meaning as commonly understood. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety, unless a specific document is cited. In case of conflict, the present specification, including definitions, will control. Additionally, the materials, methods, and examples (eg, fractions listed, plastics) are for illustrative purposes only and are not intended to be limiting.

Unless otherwise noted herein, many details regarding specific materials, processing practices, and circuits are conventional and can be found in computing, electronics, and software technology textbooks and other sources. .

Unless otherwise indicated, all numbers expressing quantities of components, reaction conditions, etc. used in the specification and claims are to be understood as modified in all cases by the term "about" Should. Accordingly, unless indicated to the contrary, the numerical parameters set forth in this specification and the appended claims may vary depending on the desired properties sought to be obtained by the subject matter disclosed herein. This is the approximate value obtained.

100 System 101 Material piece 102 Hopper 103 Conveyor system, conveyor belt 104 Conveyor system motor 105 Position detector 106 Tumbler/vibrator/singulator 107 Computer system 108 Automatic control system 109 Live action camera 110 Optical recognition system, vision system 111 Material piece tracking device , material tracking device 112 control system 120 sensor system 121 radiation source 122 power supply 124 detector 125 detector electronics 126...129 sorting device 136...139 sorting bin 140 bin, sorting container 201 material piece 203 conveyor system 401 plastic piece 403 conveyor system 410 VIS
411 XRF
412 NIR
413 MWIR
3400 Data processing system, computer system 3401 Graphics processor unit, GPU
3405 Local bus 3412 User interface adapter 3413 Keyboard 3414 Mouse 3415 Processor (CPU)
3416 Display adapter 3420 Volatile memory, RAM
3425 Network (LAN) adapter 3430 I/O adapter 3431 Hard disk drive 3432 Tape drive 3435 Non-volatile memory, ROM
3440 display

Claims (23)

capturing a first visual image of a first piece of material, resulting in a first image data packet relating to the first piece of material;
capturing a second visual image of a second piece of material, resulting in a second image data packet relating to the second piece of material, the first piece of material being exposed to a first chemical the second piece of material has a second chemical characteristic different from the first chemical characteristic;
processing the first and second image data packets using a machine learning system previously trained to visually identify pieces of material having different chemical characteristics;
using the machine learning system to classify the first and second pieces of material into two different classifications depending on the learned visual discrimination between the pieces of material having different chemical characteristics;
method including. 2. The method of claim 1, further comprising separating the first piece of material from the second piece of material according to the classification. 3. The method of claim 2, wherein the piece of material is a piece of plastic. The first chemical signature includes spectral data measured by a plurality of different sensor systems from at least one sample of a piece of plastic of the same type as the first piece of plastic; 4. The method of claim 3, comprising spectral data measured by said plurality of different sensor systems from at least one sample of a plastic piece of the same type as the two plastic pieces. 5. The method of claim 4, wherein the spectral data relates to the non-visible spectrum. 5. The method of claim 4, wherein the plurality of different sensor systems are selected from the group consisting of near-infrared ("NIR"), mid-wavelength infrared ("MWIR"), and x-ray fluorescence ("XRF") systems. . The plurality of different sensor systems include infrared (“IR”), Fourier transform IR (“FTIR”), forward looking infrared (“FLIR”), very near infrared (“VNIR”), near infrared (“NIR”), Short Wave Infrared (“SWIR”), Long Wave Infrared (“LWIR”), Mid Wave Infrared (“MWIR” or “MIR”), X-ray Transmission (“XRT”), Gamma Rays, Ultraviolet (“UV”), Fluorescence X-rays (“XRF”), laser-induced breakdown spectroscopy (“LIBS”), Raman spectroscopy, anti-Stokes Raman spectroscopy, gamma-ray spectroscopy, hyperspectral spectroscopy (e.g. beyond visible wavelengths), acoustics Spectroscopy, NMR spectroscopy, microwave spectroscopy, terahertz spectroscopy, differential scanning calorimetry (“DSC”), thermogravimetric analysis (“TGA”), capillary and rotational rheometry, optical microscopy and scanning electron microscopy (“SEM”) 5. The method of claim 4, wherein the method is selected from the group consisting of: The first chemical characteristic includes measurements of organic and inorganic elements or molecules from at least one sample of a piece of plastic of the same type as the first piece of plastic; 4. The method of claim 3, comprising measurements of organic and inorganic elements or molecules from at least one sample of the same type of plastic piece as the plastic piece. The plastic pieces may include Type #1 Polyethylene Terephthalate (“PET”), Type #2 High Density Polyethylene (“HDPE”), Type #3 Polyvinyl Chloride (“PVC”), and Type #4 Low Density Polyethylene (“LDPE”). ), Type #5 Polypropylene (“PP”), Type #6 Polystyrene (“PS”), Type #7 Other Polymers. 4. The method of claim 3, wherein the first piece of material comprises polyvinyl chloride. 2. The method of claim 1, wherein the two different classifications are different fractions. a first visual image of said first piece of material resulting in a first image data packet relating to a first piece of material; and a first visual image of said second piece of material resulting in a second image data packet relating to a second piece of material. a camera configured to capture a visual image of a second piece of material, wherein the first piece of material has a first chemical characteristic; and the second piece of material has a first chemical characteristic. a camera having a second chemical characteristic different from
a data processing system configured to process the first and second image data packets using a machine learning system pre-trained to visually identify pieces of material having different chemical characteristics; wherein the machine learning system classifies the first and second pieces of material into two different fractions depending on the learned visual discrimination between the pieces of material having different chemical characteristics; system and
a sorting device configured to separate the first piece of material from the second piece of material according to the fraction;
A system equipped with 13. The system of claim 12, wherein the piece of material is a piece of plastic. the first chemical signature includes spectral data regarding non-visible spectra measured by a plurality of different sensor systems from at least one sample of a plastic piece of the same type as the first plastic piece; 14. The system of claim 13, wherein features include spectral data regarding non-visible spectra measured by the plurality of different sensor systems from at least one sample of a piece of plastic of the same type as a second piece of plastic. 15. The system of claim 14, wherein the plurality of different sensor systems are selected from the group consisting of near-infrared ("NIR"), mid-wavelength infrared ("MWIR"), and x-ray fluorescence ("XRF") systems. . The plurality of different sensor systems include infrared (“IR”), Fourier transform IR (“FTIR”), forward looking infrared (“FLIR”), very near infrared (“VNIR”), near infrared (“NIR”), Short Wave Infrared (“SWIR”), Long Wave Infrared (“LWIR”), Mid Wave Infrared (“MWIR” or “MIR”), X-ray Transmission (“XRT”), Gamma Rays, Ultraviolet (“UV”), Fluorescence X-rays (“XRF”), laser-induced breakdown spectroscopy (“LIBS”), Raman spectroscopy, anti-Stokes Raman spectroscopy, gamma-ray spectroscopy, hyperspectral spectroscopy (e.g. beyond visible wavelengths), acoustics Spectroscopy, NMR spectroscopy, microwave spectroscopy, terahertz spectroscopy, differential scanning calorimetry (“DSC”), thermogravimetric analysis (“TGA”), capillary and rotational rheometry, optical microscopy and scanning electron microscopy (“SEM”) 15. The system of claim 14, wherein the system is selected from the group consisting of: The first chemical characteristic includes measurements of organic and inorganic elements or molecules from at least one sample of a piece of plastic of the same type as the first piece of plastic; measurements of organic and inorganic elements or molecules from at least one sample of a plastic piece of the same type as a plastic piece, said plastic piece being Type #1 polyethylene terephthalate (“PET”), Type #2 High Density Polyethylene (“HDPE”), Type #3 Polyvinyl Chloride (“PVC”), Type #4 Low Density Polyethylene (“LDPE”), Type #5 Polypropylene (“PP”), Type #6 Polystyrene (“PS”), 14. The system of claim 13, selected from the group consisting of: Type #7 Other polymers. determining the chemical characteristics of each of the mixture of different plastic pieces using a plurality of different sensor systems;
capturing a visual image of each said piece of plastic;
digitally associating the chemical characteristics of each piece of plastic with the visual image;
determining a specific fraction for separating the plastic pieces;
using the visual image to identify which of the plastic pieces in the mixture have chemical characteristics that correspond to the particular fraction;
training a machine learning system to visually identify plastic pieces that fall into said particular fraction, said training being performed using a control group created from said identified plastic pieces; step,
method including. 19. The method of claim 18, wherein the control group is comprised of captured visual image data for each of the identified plastic pieces. 19. The method of claim 18, wherein the fraction is composed of a specific combination of organic and inorganic elements or molecules. 19. The method of claim 18, wherein the plurality of different sensor systems are selected from the group consisting of near-infrared ("NIR"), mid-wavelength infrared ("MWIR"), and X-ray fluorescence ("XRF") systems. . The mixture of different plastic pieces includes Type #1 Polyethylene Terephthalate (“PET”), Type #2 High Density Polyethylene (“HDPE”), Type #3 Polyvinyl Chloride (“PVC”), Type #4 Low Density Polyethylene ( 22. The method of claim 21, wherein the polymer is selected from the group consisting of: "LDPE"), type #5 polypropylene ("PP"), type #6 polystyrene ("PS"), type #7 other polymers. The plurality of different sensor systems include infrared (“IR”), Fourier transform IR (“FTIR”), forward looking infrared (“FLIR”), very near infrared (“VNIR”), near infrared (“NIR”), Short Wave Infrared (“SWIR”), Long Wave Infrared (“LWIR”), Mid Wave Infrared (“MWIR” or “MIR”), X-ray Transmission (“XRT”), Gamma Rays, Ultraviolet (“UV”), Fluorescence X-rays (“XRF”), laser-induced breakdown spectroscopy (“LIBS”), Raman spectroscopy, anti-Stokes Raman spectroscopy, gamma-ray spectroscopy, hyperspectral spectroscopy (e.g. beyond visible wavelengths), acoustics Spectroscopy, NMR spectroscopy, microwave spectroscopy, terahertz spectroscopy, differential scanning calorimetry (“DSC”), thermogravimetric analysis (“TGA”), capillary and rotational rheometry, optical microscopy and scanning electron microscopy (“SEM”) 19. The method of claim 18, wherein the method is selected from the group consisting of:

Images