Note: Descriptions are shown in the official language in which they were submitted.
SYSTEMS AND METHODS FOR PRODUCING RECYCLED PLASTIC
PELLETS FROM LARGE ROLLS OF PLASTIC SHEETS
TECHNICAL FIELD
[0001] Example embodiments relate to methods and systems for
recycling plastic
waste by diverting recyclable plastic waste away from landfills and towards a
complete
recycling.
BACKGROUND
[0002] Plastics are polymers manufactured from non-renewable crude
oil.
However, once consumed, the plastic material is not easy to dispose, as it
does not
biodegrade naturally. Consequently, plastic waste is a large strain on
existing disposal
methods, which include recycling, landfill and incineration.
[0003] The plastics industry is very wasteful. In Canada for example,
86% of all
durable and non-durable plastic waste is sent to landfills with only 9%
diverted for
recycling. Landfill space is becoming scarce and expensive, a problem
exacerbated by
the fact that plastic waste is more voluminous than other waste types. Many
safely
situated landfills and disposal areas near urban areas, in particular, are
already filled to
capacity or approaching full capacity. Therefore, in order to properly dispose
of many
items, new disposal sites must be found. In most cases, these new sites are
further away,
less desirable than the existing sites, or are in foreign countries that are
starting to ban
plastic imports and/or place restrictions for imported plastics and/or
imposing increased
taxes on imported plastics.
[0004] Plastic is an important new tool in agriculture. It is
improving
productivity, shortening the growing season and facilitating crop cultivation
in non-
traditional growing areas. It is also providing new storage systems for
forages, and grain
crops. It was estimated that 2000 tonnes (4.4 million lbs) of plastic films
were used for
the storage of Canadian forages and grains back in 1991. This was
approximately 8% of
1
Date Recue/Date Received 2021-11-23
the 24,000 tonnes (53 million lbs) of plastic film used in the agricultural
and construction
sectors. The use of plastics in agriculture has seen a rapid growth since 1991
because
agricultural plastics have low capital cost; flexible storage options; and
extend the life of
existing storage because of use as plastic film liners.
[0005] For example, plastic films or plastic mulches are widely known and
used
in different agricultural tasks, the main one of these being mulching.
Mulching is a
technique that is used to protect crops and soil from the action of
atmospheric agents,
which in addition to other effects, reduce the quality of fruit, dry out soil,
cool the ground
and carry away or wash away fertilizers, increasing costs. This cultivation
technique has
considerably increased the economic performance of plantations and is
applicable to a
wide variety of crops.
[0006] Agricultural plastic film is mainly low density polyethylene
(LDPE)
which means that it is recyclable. However, used plastic film is bulky and
cannot be
transported very effectively. To reduce transport cost for this bulky
material, the long
sheets of plastics are compacted using small square baler that uses tine forks
for cross
feeding into the baling chamber on the farm. While this decreases the bulk and
makes
transportation easier, the resulting compacted volume still is complicated to
process.
[0007] Another drawback for recycling agricultural plastic films is
that it arrives
at the collection points with a great amount of dirt and soil that these
plastic incorporate
during their use. Agricultural plastic films must be cleaned before being
converted into
pellets for film or formed into moulded products like plastic lumber and fence
posts.
Incoming agricultural plastic films must be inspected for contamination and
are accepted
or rejected depending on the level of contamination. Contamination includes
dirt, sand,
stones, grease, vegetation, water, other types of plastic, glue, tape and
ultraviolet (UV)
light degradation. If the film has lost its flexibility and is crinkly, it has
serious ultraviolet
light damage. UV damage can severely limit the recyclability of agricultural
plastic film.
2
Date Recue/Date Received 2021-11-23
[0008] Because of this, there is an increasing level of pressure to
keep as much
material as possible out of landfills and increase the efficiency of recycling
of the plastic
waste, and increase the ability to recycle complex plastic waste.
[0009] Accordingly, there is a need for systems and methods increase
efficiency
of the plastic recycling process by simultaneously diverting recyclable
plastics away
from landfills and increasing the availability of recycled plastics for the
end user.
SUMMARY
[0010] In an example embodiment, there is provided systems and methods
for
converting agricultural plastic waste feedstocks into recycled plastic
pellets. The
products from the systems and methods include high-density polyethylene
(HDPE), low-
density polyethylene (LDPE), polypropylene (PP) and polystyrene (PS).
[0011] In another example embodiment, there is provided systems and
methods
to provide one or more of cutting size, temperature, and rates of processing
of the
agricultural plastic feedstocks.
[0012] In another example embodiment, the systems and methods to comprise
using vision technology to sort and dry the plastic agricultural feedstocks.
In one aspect,
vision technology is used to inspect the plastic materials to detect
suitability for recycling
as well as ensure complete dry washing. In some aspects, the vision technology
by way
of a camera or spectroscopic Charge Coupled Device (CCD). In some aspects, the
camera is a light image camera, a multispectral camera, an infrared (IR)
camera, or an
ultraviolet (UV) camera.
[0013] In some example embodiments, there is provided systems and
methods
using computer vision machine learning to automatically: 1) sort plastic prior
to cutting
that it is i) suitable for cutting, ii) needs cleaning prior to cutting, or
iii) not usable
(discard); and 2) Determine whether plastic being dry washed in a rolling drum
to
remove impurities has been completed or is below an impurity threshold.
3
Date Recue/Date Received 2021-11-23
[0014] In some example embodiment, the dry washing comprises rotating
the
camera with the rolling drum so that the images can be captured more stably so
that the
computer machine learning vision technology can determine that the dry filter
has been
completed or is below an impurity threshold.
BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figures 1A and 1B are a schematic of a method for recycling
plastic waste
according to an example embodiment; and
[0016] Figure 2 is a process flow diagram of the method for recycling
plastic
waster according to an example embodiment.
DETAILED DESCRIPTION
[0017] Reference will be made below in detail to exemplary
embodiments,
examples of which are illustrated in the accompanying drawings. The same
reference
numerals may be used throughout the drawings.
[0018] Example embodiments include systems and methods for recycling
plastic
waste or complex agriculture film plastic waste to create quality pellets to
meet industry
standard of optical, mechanical and chemical properties.
[0019] Figures 1A and 1B illustrate an example method 100 for
recycling plastic
waste or complex agriculture film plastic waste to create quality pellets to
meet industry
standard of optical, mechanical and chemical properties. Figure 2 is a process
flow
diagram of the method for recycling plastic waster according to an example
embodiment.
In some examples, at least part of the method 100 is performed by at least one
processor
(or computer). The plastic waste recycling system 100 begins when agricultural
plastic
film waste 10 is transported to the recycling facility. At a sorting stage,
the sorting is by
a sorter 20 and an optical screener 22 which comprises optical sorting
technologies
including, by not limited to, vision technology. Vision technology can include
camera,
spectroscopic Charge Coupled Device (CCD), or x-ray. In some examples, the
camera is
4
Date Recue/Date Received 2021-11-23
a light image camera, a multispectral camera, an infrared (IR) camera, or an
ultraviolet
(UV) camera. In some examples, the camera can be configured to detect a
particular or
narrow band of spectra, or can detect multiple types or a wide band of
spectra.
[0020] The plastic feedstock 10 is received in a roll in some
examples. For
example, the plastic feedstock 10 can be rolled into a bale using a baler, and
then
received by the process in the form of rolled plastic.
[0021] The optical screener 22 is configured to obtain information to
determine
whether plastic feedstock 10 are 1) sufficiently clean of contaminants and
suitable for
shredding, 2) not sufficiently clean and needs an initial cleaning to remove
any
contaminants prior to shredding, or 3) not usable (discard) when there is
significant
contamination. The sorter 20 diverts the suitable plastic feedstock 10 that
meets or is
below this contamination threshold for shredding, or diverts the plastic
feedstock 10 that
does not meet or is above this contamination threshold for either initial
cleaning or for
discard. Significant contamination includes that which has permanently altered
the
plastic feedstock (e.g. UV damage) and/or cannot be removed by cleaning. The
optical
screener 22 can use the above-noted vision technology. The optical screener 22
can
detect which pixel locations of the camera correspond to the plastic feedstock
10. The
optical screener 22 can estimate an amount of impurity for those pixel
locations that
correspond to the plastic feedstock 10. In some examples, the optical screener
22 can
estimate and/or identify of the type of material of the impurities. In some
examples, the
optical screener 22 can detect the type of plastic material of the plastic
feedstock 10. In
other examples, the type of plastic material (e.g. LDPE or HDPE) of the
plastic feedstock
10 is pre-set and has known optical properties that are detectable by the
optical screener
22. In further examples, the impurities include dirt, sand, stones, grease,
vegetation,
water, glue, and tape.
[0022] The optical screener 22 can include a neural network. In some
embodiments, the optical screener 22 may categorize the plastic feedstock 10
by
algebraic functions, stochastic functions, decision tree protocols, or machine
learning
5
Date Recue/Date Received 2021-11-23
techniques which may include neural networks (including deep neural networks,
DNN),
random forest, collaborative filtering, support vector networks.
[0023] A neural network consists of neurons. A neuron is a
computational unit
that uses Xs and an intercept of 1 as inputs. An output from the computational
unit may
be:
hwb (x) = f (WT x)= f(s + b)
s=1, 2, ... n, n is a natural number greater than 1, Ws is a weight of xs, b
is an offset (i.e.
bias) of the neuron and f is an activation function (activation functions) of
the neuron and
used to introduce a nonlinear feature to the neural network, to convert an
input of the
neuron to an output. The output of the activation function may be used as an
input to a
neuron of a following convolutional layer in the neural network. The
activation function
may be a sigmoid function. The neural network is formed by joining a plurality
of the
foregoing single neurons. In other words, an output from one neuron may be an
input to
another neuron. An input of each neuron may be associated with a local
receiving area of
a previous layer, to extract a feature of the local receiving area. The local
receiving area
may be an area consisting of several neurons.
[0024] A deep neural network (DNN) is also referred to as a multi-
layer neural
network and may be understood as a neural network that includes a first layer
(generally
referred to as an input layer), a plurality of hidden layers, and a final
layer (generally
referred to as an output layer). A layer is considered to be a fully connected
layer when
there is a full connection between two adjacent layers of the neural network.
To be
specific, all neurons at an ith layer is connected to any neuron at an 0+.
sjth
layer. In the
DNN, more hidden layers enable the DNN to depict a complex situation in the
real
world. Training of the deep neural network is a weight matrix learning
process. A final
purpose of the training is to obtain a trained weight matrix (a weight matrix
consisting of
learned weights W of a plurality of layers) of all layers of the deep neural
network.
6
Date Recue/Date Received 2021-11-23
[0025] The neural network can be trained using labelled data pairings
between
one or more inputs and one or more outputs. For example, the labelled data
pairings can
include inputs of images of plastic feedstock 10 and outputs of: 1)
sufficiently clean and
suitable for shredding, 2) not sufficiently clean and needs an initial
cleaning prior to
shredding, or 3) not usable (discard). Various pixels of the plastic feedstock
10 in the
image can be labelled with any of these outputs. In addition, various pixels
of the plastic
feedstock 10 in the image can be labelled with the particular type of
material(s) of the
impurity(ies).
[0026] Sufficiently contaminant-free plastic feedstock 10 is
processed at a
shredding stage where a shredder 30 is adapted to shred the plastic feedstock
10 to
produce suitably-sized plastic sheets 12 (also called regrind 12). Regrind 12
is subjected
to an optical screener 32 comprising vision technology to obtain information
about the
size of the regrind 12. If the shredded pieces are too large, this can
decrease washing
efficiency and may leave contaminants behind and if the shredded pieces are
too small,
this can create wastage as small pieces get stuck in mechanical equipment and
introduces
more input wastage.
[0027] In one embodiment, the regrind 12 has a shred size of on or
about 10 x 10
cm which provides one optimal point because this can improve the speed and
efficiency
of washing process. Any regrind 12 that is determined to be too large is
further
shredded. When it is established that the regrind 12 is suitably-sized, the
regrind 12
leaves the shredder 30 and the shredding stage.
[0028] Regrind 12 travels by conveyor belt 14 to a dry wash stage to
remove
debris such as for example, sand, stones, glass, paper, etc. The dry wash
stage comprises
a hot air tumbler 40 which tumbles the regrind 12 and an optical screener 42.
The
optical screener 42 is configured to obtain visual information about whether
the regrind
12 is sufficiently free from debris and meets or is below an impurity
threshold. In
example embodiments, the optical screener 42 includes vision technology. In
one
example embodiment, the optical screener 42 is a rotatably-mounted camera 44
which
therefore rotates with the rotating drum so as to obtain clearer images and/or
with greater
7
Date Recue/Date Received 2021-11-23
stability (less motion artifacts). In some example embodiments, the rotation
of the
camera 44 can be synchronous or slightly out of synchronicity with the
rotating drum.
When the visual information is collected, and it is established that the
regrind 12 is
sufficiently free from impurities and is at or is below the desired impurity
threshold, the
regrind 12 leaves the hot air tumbler 40.
[0029] In another example embodiment, the camera 44 of the optical
screener 42
is statically mounted to a frame or other stationary part of the hot air
tumbler 40.
[0030] The regrind 12 then proceeds a wet wash stage for wet washing
by a line
washer 50 to remove any remaining debris and is then dried using a spinner 52
to reduce
the moisture.
[0031] In one embodiment, after the wet wash stage, the regrind 12
may proceed
to a fine shredding stage comprising a fine shredder 60 and an optical
screener 62.
[0032] The regrind 12 then proceeds to a heating and extruding stage
comprising
a heater 70 and an extruder 72. The main heating extruder 70 incorporates
numerous
heating stages with smooth temperature gradients. In one preferred embodiment,
there
are about 4 to about 6 heating steps so that the regrind 12 is gradually
melted and the
regrind 12 at the beginning stages is firm enough to be pushed through the
extruder.
[0033] If temperature is too high at the beginning of the heating
stages, the
regrind plastic melts too fast and becomes too soft such that the extruder 72
cannot
continue to push the volume through efficiently.
[0034] The initial temperature selected is based on type of material
input. For
LDPE, the temperature gradient starts at 130 C and ends at 150 C. For LDPE, a
firmer
consistency is achieved if temperature gradient is set to 130 C or about 130 C
at the first
stage, and gradually increases through two more stages to reach 150 C or about
150 C at
the last stage of extrusion. For HDPE, the temperature gradient starts at 190
C or about
8
Date Recue/Date Received 2021-11-23
190 C and gradually increases through two more stages to reach 210 C or about
210 C.
An extruded product 16 is produced at the end of the heating and extruding
stage.
[0035] The extruded product 16 undergoes a filtering stage to pass
the extruded
product 16 through two 100 micron filtering mesh 80 of inox 304 (SAE 304
stainless
steel per SAE International). As plastic enters the mesh, contaminants can
fill up the
mesh screens and prevent a good flow. Internal pressure is built up as more
contaminant
is trapped. Pressure at filtering stage is carefully observed to maintain flow
efficiency.
Pressure of no more than 12 MegaPascal (MPa) is maintained by replacing mesh
filters
to lower the pressure and prevent buildup of contaminant and machine damaging.
In
some examples, this preserves the flow and increase production efficiency as
well as
material properties for suitable plastic pellets 18.
[0036] The various embodiments presented above are merely examples
and are
in no way meant to limit the scope of this disclosure. Variations of the
innovations
described herein will be apparent to persons of ordinary skill in the art,
such variations
being within the intended scope of the present disclosure. In particular,
features from one
or more of the above-described embodiments may be selected to create
alternative
embodiments comprises of a sub-combination of features which may not be
explicitly
described above. In addition, features from one or more of the above-described
embodiments may be selected and combined to create alternative embodiments
comprised of a combination of features which may not be explicitly described
above.
Features suitable for such combinations and sub-combinations would be readily
apparent
to persons skilled in the art upon review of the present disclosure as a
whole. The subject
matter described herein intends to cover and embrace all suitable changes in
technology.
9
Date Recue/Date Received 2021-11-23
