This invention relates to sorting systems. In particular, this invention relates to an optical materials sorting system, method, and apparatus for recyclable materials.
Recycling is one of the most environmentally sensible solutions to waste disposal and resource conservation. Many different types of materials can be recycled for reuse, including metals, plastics and paper.
In paper recycling, white copy paper is one of the most valuable of all recyclable paper grades, while newsprint and file folder stock, which have a high concentration of groundwood, are less valuable grades and at times are considered to be contaminants when found in white copy paper. Thus, during the recycling of high grade white paper, it is desirable that the feed be devoid of high groundwood-content material, which would degrade the finished product. Typically paper sorting is undertaken by workers, who visually identify inferior grades and contaminated white copy paper on a conveyor carrying mixed stock and manually separate inferior grades from the higher value white copy paper.
The separated product inevitably contains an undesirably high content of the inferior grades because visual discrimination is often not very effective, particularly when the conveyor is moving at a high speed. Manual sorting is also undesirable for security reasons, where for example the paper to be recycled contains confidential documents destined for shredding.
Sorting can be done automatically by colour detection. However, it is common to encounter groundwood contaminants in the white paper copy paper grade when sorting is based on color, because sensitivity limitations and obscuring of the stock by graphics can result in newsprint and white-colored file folders being graded as white paper.
Furthermore, paper destined for recycling is typically received by a recycler as an incoming volume of mixed paper waste that may be categorised in terms of content as white paper, colour paper, mixed paper and non-paper content. White paper comprises office waste consisting of substantially white paper. Colour paper is office waste without white content. Mixed paper, which comprises paper waste that is not white paper or colour paper, may comprise paper waste with groundwood content, carbon paper, colour folders, adhesives (labels), newspaper, coated magazines, wrap paper, brown kraft paper and waxed paper. It will be understood that “waste” may include pre-consumer, post-consumer, pre-industrial, post-industrial, and recyclable materials, which may be shredded, non-shredded, and commingled, and is not restricted to post-consumer waste paper.
The white, colour, and mixed categories of mixed paper waste may be shredded and combined in various ratios to produce a number of different grades. Five of those possible sorted shredded paper grades are as follows: Computer Print Out (CPO), Super Sorted Office Waste (SSOW also known as OP1), Office Waste (OW also known as SOP, OP2), High Grade Mix (HGM) and Low Grade Mix (LGM or known as filestock or OP3). Each of these paper grades is defined based on specific minimum standards of grade composition in various industry trade publications to meet paper mills' specifications. The mills pay different prices per ton of shredded paper depending upon the grade. Grades may be further defined by ISRI Scrap Specification Circular, distributed by the Institute of Scrap Recycling Industries, Inc., www.isri.org. The grades referenced herein are used only as examples, and are not intended to be limiting in any way.
While mills will accept sorted shredded paper that exceeds the composition specified, the recycler is not necessarily rewarded for providing an enhanced grade of sorted shredded paper. Sorted shredded paper that fails to meet a specified grade is categorised in the first grade whose specifications it exceeds.
The current practice in the paper recycling industry is to simply maximise production of the highest grade of sorted shredded paper, and then maximise production of the next highest grade, and so on. The result of following such a strategy is the inefficient production of sorted shredded paper grades, such that revenue for sorting a given inventory of paper waste is not maximised. This practice leads to excess production of low grade sorted shredded paper that actually exceeds grade requirements. For example, in a manual sorting operation for sorted shredded paper grades of CPO (85% by weight white paper, 15% by weight colour paper), SSOW (15% white paper, 85% colour paper), and OW (10% white paper, 85% colour paper, 5% mixed paper), it has been found that the OW grade shredded paper generated typically comprises far more white paper than is actually required to meet specifications.
In drawings which illustrate by way of example only a preferred embodiment of the invention,
a and 8b are schematic representations of methods for sorting the input paper waste by content.
a,
9
b, 9c, 9d, and 9e are schematic representations of the production of a selection of shredded paper grades from the sorted input paper waste.
Groundwood contains the polymer lignin, which absorbs ultra violet (UV) light. One of the steps in the manufacture of paper is delignification, which involves the removal of lignin from wood pulp. However, the extent of delignification varies depending on the type of paper being manufactured. Paper products such as newsprint and file folders contain substantially higher amounts of lignin compared to regular white copy paper. The higher the lignin content, such as in newsprint, the stronger the absorption in the UV region of the spectrum. Newsprint and other types of paper containing high amounts of groundwood can thus be identified based on the absorption of UV light, which can be determined inferentially by the diffuse reflectance of the paper in the UV range of the spectrum.
When a beam of light impinges on the surface of a sheet of paper, it is either absorbed or diffusely reflected depending on the quality of the paper. The degree of absorption and reflectance varies with the wavelength of the light and the ‘whiteness’ of the paper. Spectral diffuse reflectance measurements show that regular white copy paper has a substantially higher diffuse reflectance compared to groundwood in the UV region. This difference is due to the lignin content, which has a high affinity for UV light, resulting in a very strong absorbance in the UV region.
A perfect diffusing surface reflects all light incident upon it.
In measuring % R the colour, intensity and orientation of the incident light beam should be exactly the same for both the samples and the reference. This eliminates the need to measure the intensity of the incident light.
The curves in
[% R]TS=[% R]high groundwood+([% R]low white−[% R]high groundwood)/2 (2)
where
[% R]TS is the Threshold Relative Reflectance,
[% R]high groundwood is the upper limit of groundwood relative reflectance range,
and
[% R]low white is the lower limit of white relative reflectance range.
Using formula (2), [% R]TS has been calculated to be 44% in the 200-300 nm UV light range. The acceptance/rejection algorithm can be expressed as follows;
IF [% R]<[% R]
TS
SHEET=“GROUNDWOOD”
ELSE
SHEET=“WHITE”
The detection device 10, illustrated in
In the embodiment shown a plano convex lens 14 collects and collimates the diffusely reflected light. The collimated light is filtered though a UV filter, for example a U-340 Hoya (trademark) filter 16, which isolates the 300-400 nm band. It will thus be appreciated that the light source 12 may be a full spectrum or standard “white” light source which illuminates the sheet 2 in many spectral ranges outside of the UV range, for example visible light regions of the spectrum, because the filter 16 blocks all wavelengths except for those in the selected UV region. Even where the light source 12 is a UV light source confined substantially to the selected UV light region, the filter 16 would be necessary to block light in the visible region of the spectrum produced through fluorescence.
The filter 16 is positioned in front of an aperture 22 through the wall of an opaque detector housing 11 containing a suitable optical detector 24, for example a gallium nitride detector, positioned directly behind the aperture 22. The housing 11 is positioned in a convenient location with the aperture 22 facing the sheet of paper 2. The housing 11 prevents ambient light from striking the detector 24, so that the detector 24 generates a photoelectric current which is directly proportional to the intensity of light diffusely reflected from the paper sheet 2.
In the paper separating apparatus of the invention the detection device 10 is disposed adjacent to the conveyor 4 such that incident light emitted from the light source 12 and diffusely reflected off of the paper sheet 2 traverses the aperture 22 and strikes the detector 24. In the embodiment shown the sheets 2 are supported on top of the conveyor 4 by gravity, and the device 10 is thus positioned above the conveyor 4. In other embodiments, where for example the sheets 2 are retained on the conveyor 4 by air suction or other mechanical means, the conveyor 4 may be oriented in any fashion and the detection device 10 would be positioned accordingly, so that incident light from the light source 12 reflects off of the sheet 2 and enters the aperture 22 of the detector housing 11.
The level of the electrical signal generated by the detector 24 may be measured by any suitable instrument and manually compared against a reference level. However, the preferred embodiment of the invention is automatic and includes a data processing device, such as a personal computer or microcomputer 30, which processes the reflectance values generated by the detector 24 and operates the ejection system 40 in the paper separating apparatus. The photoelectric current generated by the detector 24 is carried by a suitable connection to an analog input port of an analog to digital converter 32, which digitizes the signal and outputs the digital equivalent signal to the computer 30, for example through a parallel connection. The computer 30 is programmed to receive data representative of the intensity of each signal, calculate the relative reflectance of the sheet 2 and apply the algorithm set out above to determine whether the sheet 2 is classified as ‘white’ or ‘groundwood’.
Preferably the computer 30 is also programmed to display the absolute value of digitized signal, the computed value of % R and the type of paper (‘white’ or ‘groundwood’) on a monitor 30a. The computer 30 is further programmed to output either a high or low logic signal depending on which of the two classes of sheets (white or groundwood) has been detected, the high logic signal being an ejection signal which activates the ejection system 40. Preferably, because groundwood normally occurs in smaller quantities in the mixed stock, groundwood is selected as the grade to be ejected. Thus, for example, the paper separating apparatus may be used to separate white from non-white input paper waste, or white and colour paper waste from mixed paper waste, if mixed paper waste may be identified by a higher average lignin content than non-mixed paper waste.
Thus, when the output from the computer 30 is low, no ejection signal is generated; when the computer 30 determines that a sheet of paper 2 falls into the ‘groundwood’ class, it outputs an ejection signal. In the embodiment shown the ejection signal is transmitted back to the AD converter 32 over the parallel cable connection with the computer 30, and the AD converter 32 in turn sends a digital output signal to the relay switch 42 to activate the ejection system 40. Alternatively, the ejection signal may be transmitted directly to a relay card associated with the relay switch 42 to activate the ejection system 40.
The ejection system 40 releases a momentary blast of air to divert a sheet 2 of groundwood paper from the primary collection bin 6 to a rejection bin 8. The bins 6, 8 are thus positioned adjacent to the end of a conveyor 4 such that a sheet 2 reaching the end of the conveyor 4 will fall into the primary collection bin 6 unless diverted to the rejection bin 8, as shown in
The ejection system 40 comprises relay switch 42, one terminal of which is connected to a two-way normally closed solenoid valve 44 and the other terminal of which is connected to one terminal of a power supply, for example a conventional 120 V mains power supply 43. The other terminal of the power supply is connected to the solenoid valve 44 to form a circuit through the relay switch 42, solenoid valve 44 and power supply. A compressor 46 is in communication with an air nozzle 48 connected to the outlet port of the solenoid valve 44 and directed at the terminal end of the conveyor 42, to divert a rejected sheet of paper 2 into the rejection bin 8.
In operation, the conveyor 4 is activated and sheets 2 of paper are fed onto the conveyor 4 upstream of the detector housing 20, either manually or by any suitable mechanical feeding means (not shown). The conveyor 4 conveys the sheets 2, one at a time, past the detector housing 20. As each sheet 2 passes the housing 20 light from the light source 12, collimated by the lens 14, strikes the sheet 2 and reflects through the filter 16 into the aperture 22, striking the detector 24. The detector 24 generates an analog electrical signal proportional to the intensity of light striking the detector, which signal is digitized by AD converter 32 and output to the computer 30.
The computer 30, having been programmed with the appropriate algorithms and a reference level, calculates the percent reflectance of the sheet 2 and compares the % R value to the preprogrammed acceptance-rejection threshold. If the calculated % R is above the acceptance-rejection threshold, the computer 30 outputs a logic low signal (or no signal) and the sheet 2 is conveyed to the terminal end of the conveyor 4 where it falls into the primary collection bin 6.
If the calculated % R is above the acceptance-rejection threshold, the computer 30 outputs a logic high signal to activate the ejection system 40. When the relay switch 42 receives the high input signal from the computer 30, the relay switch 42 closes and completes the circuit, causing a current to pass through the solenoid valve 44, which in turn opens to release a blast of air from the compressor 46 through the air nozzle 48. The distance between the air nozzle 48 and detection device 10, along with the speed of the conveyor 4, determines the time lapse between detection and ejection. The relay switch 42 may be activated after a suitable delay interval, to account for the time taken between detection of the % R from the sheet 2 and conveyance of the sheet to the terminal end of the conveyor 4, however with the conveyor 4 set to a high enough speed and the detection device 10 suitably positioned near the terminal end of the conveyor 4, a delay may be unnecessary.
The air blast causes the sheet of paper 2 to be diverted into the rejection bin 8, which may for example be positioned beneath the terminal end of the conveyor 4, as shown in
In the preferred embodiment the duration of the air blast is equal to the length of time the sheet of paper is “viewed” by the detection device 10. The computer 30 can be programmed to time the interval between the start and end of a reflectance signal from the AD converter 32, which respectively correspond to the sheet 2 entering and leaving the view field of the detector 24, and to maintain the high output signal for this interval in order to ensure that the sheet 2 is properly diverted into the rejection bin 8.
Referring to
From the singulation apparatus 60, the paper sheets 2 are conveyed, for example on a conveyor belt 66, to a buffer 70. It will be understood that all conveyor belts and similar means described herein are preferably enclosed to reduce the likelihood that paper sheets 2, particularly in shredded paper form or in dust form, will escape the system anywhere but through the intended exits and passages. The buffer 70 preferably comprises a sensor, such as a weight or volume sensor, to detect when the buffer 70 is nearing a predetermined maximum capacity of paper sheets 2. The sensor is connected to a signal generator 72, which may comprise lights or other signaling devices to be observed by an operator, or an electric signal generator for use in communicating with the controller operating the actuator 64. When the sensor determines that the buffer 70 is nearing its maximum capacity, it may cause the signal generator 72 to signal to an operator that the buffer is full so that remedial action may be taken (for example, to cease feeding the shredder or singulation apparatus 60 with input paper waste). Alternatively, or in addition to this action, the signal generator may send a signal to the controller for the actuator 64 to activate and close the gate 62, thus preventing further sheets 2 from exiting the singulation apparatus 60 via the exit 61. Any buildup of paper sheets 2 in the singulation apparatus 60 may be removed or ejected through the ejection passage 65. The ejection passage 65 may be connected to a reject bin or baler or the rejection bin 8, for example via a conveyor belt (not shown). This portion of the system thus prevents the buffer 70 from becoming overloaded from the singulation apparatus feed.
From the buffer 70, the paper sheets 20 are deposited on a separation table 90. The separation table 90 comprises a perforated surface, in which the perforations are preferably sized to allow undersized pieces of paper to fall through the table 90 and into a collector 100, which in turn directs the collected reject paper to another ejection passage 102. The ejection passage 102 may be connected to the same reject bin or baler or rejection bin 8 as the earlier ejection passage 65. The separation table 90 may be a vibrating conveyor, which conveys the paper sheets 2 from a first end 92 to a second end 94, while further separating the paper sheets 2 and removing smaller pieces.
From the separation table 90, the paper sheets 2 are preferably transferred to a conveyor 110 and then passed through a detection device 10 such as that illustrated in
While the ejection system 40 employs blasts of air to divert unacceptable sheets 2 to the rejection bin 8, the deflector 120 may be actuated to divert all sheets 2 to the rejection bin 8. This deflector 120 may thus be used to divert the incoming sheets 2 when the detection device 10 determines that there is a high count of unacceptable content arriving on the conveyor 110. A high count of unacceptable content (defects) may be defined as being a certain count of rejected sheets 2 within a specified unit length of time. Since rejected sheets 2 are normally ejected using a blast of air, a persistent high count of rejected sheets 2 may be taxing on the air compressor feeding the air nozzle 48.
The deflector 120 is preferably positioned between the conveyor 110 and the primary collection bin 6 and is pivotably mounted so that it may rotate between the default position and a deflecting position, as shown in phantom as deflector 120′ in
The predetermined signal may be discontinued, or the further cancel signal transmitted, a fixed period of time after the commencement of the predetermined signal, for example 1 or 2 seconds. Alternatively, the predetermined signal may be discontinued or the cancel signal transmitted when the detection device 10 determines that the count of unacceptable content has dropped sufficiently, in accordance with predetermined thresholds.
This embodiment of the optical sorting system thus provides three means for mechanically diverting the input stream of paper sheets 2 from a path leading to the primary collection bin 8: first at the singulation-buffer feed, next at the separation table, and finally at the detection device. The gate 62 at the singulation apparatus 60 prevents overload of the buffer 70 by diverting the sheets 2 towards an ejection passage 65. The separation table 90 allows for smaller pieces of paper 2, which may not be large enough to be effectively detected by the detection device 10, to be removed from the input to the detection device 10 and instead redirected to an ejection passage 102. Finally, the deflector 120 may be actuated to divert all paper 2 away from the primary collection bin 6 and into the rejection bin 8. The system thus described allows for efficient removal of reject or unsuitable paper 2 and efficient collection of acceptable quality paper sheets 2. It will be appreciated that this system may not only be used in conjunction with the system for detecting materials with a certain composition of lignin, as described above, but also with systems for optically sorting white paper from colour paper, such as that described in Canadian Patent Application No. 2,406,300, which is incorporated herein by reference. Thus it will be appreciated that the device and apparatus of the present invention can be adopted to sort and separate other fibrous objects or objects containing lignin fiber besides paper, utilizing the principles of the invention to differentiate between lignin levels in two or more categories. The invention may also be used to sort and separate paper into more than two categories, by defining a plurality of classification thresholds, increasing the logic output options (for example by outputting to the ejection device a multiple-bit word rather than a high or low signal), and providing a sufficient number ejection systems (i.e. at least one less than the number of classifications) to separate materials of the different classes.
In a further embodiment, a system and method for sorting paper waste to provide an optimum combination of grades of sorted shredded paper waste is provided.
Referring to
Portions of sorted input waste 830, 840, 850 are then recombined in specified ratios 880a, 880b, 880c, 880d, and 880e at step 870 to produce graded sorted shredded paper. With reference to
The ideal amount of each grade to be produced is determined by considering the composition of the incoming paper waste, the sorting efficiency of the sorting team, the composition specifications of the sorted shredded paper grades, and the current market price for each grade. An estimate of sorting efficiency may be determined by comparing the ideal grade amounts with the actual grade amounts produced.
The ideal amounts of each grade to be produced may be determined by a linear optimisation method such as linear programming. First, initial pre-known starting conditions are defined: the relative weight of each of the sorted input paper types 930, 940, 950, and the market price of each specified grade of sorted shredded paper composed of one or more of the sorted input paper types 930, 940, 950. The relative weights of the input paper types may account for non-paper waste as well, such as binders, metal clamps, fasteners, and actual garbage content that is not recyclable as a paper product. The calculations may provide for an estimated garbage content, for example 5% by weight. The value to be maximized, Z, is the sum of the revenue generated based on the market price per weight of each grade of sorted shredded paper multiplied by the weight of each grade of sorted shredded paper to be produced:
Z=(Price/tonne of CPO)×(No. tonnes CPO)+(Price/tonne of SSOW)×(No. tonnes SSOW)+(Price/tonne of OW)×(No. tonnes OW)+(Price/tonne of HGM)*(No. tonnes HGM)+(Price/tonne of LGM)×(No. tonnes LGM)
Next, constraints are defined: for example, the weight of any input paper for each grade of sorted shredded paper must be greater than or equal to zero, and the total weight of sorted shredded paper produced must be less than or equal to the weight of sorted input paper. Preferably, the total weight of input waste paper is equal to the weight of output sorted shredded paper, taking into account the estimated garbage content. Also preferably, optional constraints may be defined, such as target output weights of specified grades of sorted shredded paper (for example, if no SSOW grade is to be produced, then the weight of that sorted shredded paper grade is equal to zero). Further constraints may also be defined; for example, if it is known that LGM grade sorted shredded paper output is typically 2% of the total paper waste input, this further constraint setting LGM output to 2% of the total input may be imposed. Certain constraints may be defined as binding, meaning the constraint must be met exactly. Other constraints may be non-binding. A constraint that 0 metric tons of SSOW is to be produced, for example, would preferably be defined as a binding constraint, while a constraint that the total weight of input waste paper is equal to the weight of output sorted shredded paper would preferably be defined as non-binding.
As an example, if it is expected that the target grade composition will be exactly met (100% efficiency), sample specifications for output may be provided as follows:
In the table above, white paper is considered to be higher quality input paper waste than colour paper, and colour paper is higher quality input paper waste than mixed paper. Accordingly, a sorted shredded paper grade is met if the proportions of the higher quality input paper waste are met or exceeded, and the proportion of lower quality input paper waste is not exceeded. Thus, a composition of 85% white paper, 13% colour paper, and 2% mixed paper does not meet the minimum 98% standard for CPO; while it does meet the minimum standard for white paper for SSOW, it exceeds the maximum 0% standard of mixed paper; therefore, such a composition would be categorized as OW grade, since it meets or exceeds the minimum white paper composition of 10%, and does not exceed the maximum 5% limit for mixed paper.
Given the above constraints and starting conditions, the optimal output of sorted shredded paper grades is determined such that Z is maximised. This step may be carried out using linear programming techniques, and may preferably be carried out using a numerical method package such as Microsoft Excel Solver™ running on a computer. Use of a numerical method package is preferable, as changes may be made to various constraints or starting conditions and the optimal output and Z recalculated. It will be appreciated that a paper sorting system may be provided with a processor and input and output means for carrying out the method described above. This processor may communicate with sorting machinery, such as the optical characterization systems described below, in order to produce the desired optimal output.
Once the optimal output of each sorted shredded paper grade is determined, the input paper waste 930, 940, 950 may be combined in the amounts defined by the optimal output to produce sorted shredded paper for further processing. Since the optimal grade allotment is known from the outset, workers can avoid under or over sorting the inventory.
However, turning again to
This sampling method comprising one or more optical characterization steps may be performed inline with the further sorting process carried out on the entire input paper waste. The input paper waste may be passed through a series of detectors in an optical characterization system, as described above. Once the sampling is complete, the optimal composition of the sorted shredded paper is determined as described above, and the same optical characterization systems are configured to then sort the input paper waste in accordance with that optimal composition. A first set of detectors, for example, may be used to generate output sorted shredded paper comprising the highest quality of output to be produced; the highest quality output is then output to a baler for delivery to a mill for further processing, while the remaining input paper waste that is separated at the first set of detectors is diverted to a second set of detectors. The second set of detectors is used to generate output sorted shredded paper comprising the next highest quality of output shredded paper to be produced, which in turn is directed to another baler. A further optical characterization system may be provided to further sort the input paper waste that remains in accordance with any remaining grades of sorted shredded paper to be produced and baled. Once it is determined by a processor controlling the optical characterization systems that sufficient quality of a given grade of sorted shredded paper has been produced, the processor may be configured to allow all remaining input paper waste to pass through the optical characterization system associated with that grade, or to reconfigure all of the optical characterization systems in accordance with the remaining output to be generated. A schematic of an exemplary system is shown in
In a further embodiment, the sampling may comprise an estimation of paper content based on the known characteristics of the source of the input paper waste. Input paper waste is collected from one or more client sources. The source of input paper waste may vary according to client source; for example, some clients may produce paper waste that generally comprises a certain percent composition range of mixed paper or colour paper compared to white paper, and this range may be known based on prior sampling techniques or prior shredded paper grade composition. Therefore, by estimating the weight of the input paper waste from each client, it is possible to determine the optimal output from the input paper waste, and to use the processor to control the optical characterization systems described above.
A preferred embodiment of the invention having been thus described by way of example only, it will be apparent to those skilled in the art that certain modifications and adaptations may be made without departing from the scope of the invention. For example, it will be apparent that the deflector 120 may be employed in optical sorting systems used to sort objects other than paper.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/CA2007/000826 | 5/11/2007 | WO | 00 | 11/12/2008 |
Number | Date | Country | |
---|---|---|---|
60747161 | May 2006 | US | |
60747169 | May 2006 | US |