CN110980036A

CN110980036A - Intelligent garbage classification device and classification method thereof

Info

Publication number: CN110980036A
Application number: CN202010007854.8A
Authority: CN
Inventors: 王治华; 徐向红; 黄云风; 刘宝顺; 于小岭
Original assignee: Shazhou Professional Institute of Technology
Current assignee: Shazhou Professional Institute of Technology
Priority date: 2020-01-06
Filing date: 2020-01-06
Publication date: 2020-04-10

Abstract

The invention discloses an intelligent garbage classification device, wherein the interior of a garbage box body is of a double-layer structure, a conveyor belt, a shock absorber, a motor and a roller are arranged in a first-layer cavity, the roller is respectively connected with the shock absorber and the motor and is positioned above the conveyor belt, a garbage storage classification plate, a guide rail, a moving block, a mechanical arm, an infrared probe, a spectrum analyzer and an electric motor are arranged in a second-layer cavity, the mechanical arm is arranged below the guide rail in a sliding mode through the moving block, the electric motor is arranged on the upper portion of the mechanical arm and is connected with the spectrum analyzer, and the infrared probe is arranged at the bottom of the spectrum analyzer. By adopting the mode, the intelligent garbage classification device and the classification method thereof provided by the invention adopt the infrared spectrum technology to identify and classify the plastic garbage and the paper garbage, are convenient to recycle, avoid secondary pollution, reduce the subsequent treatment procedures and are convenient to operate and implement in life and industry.

Description

Intelligent garbage classification device and classification method thereof

Technical Field

The invention relates to the field of garbage classification, in particular to an intelligent garbage classification device and a classification method thereof.

Background

Currently, the total production of plastics and paper products worldwide exceeds millions of tons. This creates a global environmental problem, namely the environmental pollution caused by waste plastics and cartons. The waste plastics and the paper boxes are difficult to naturally degrade and have no affinity to the natural environment, so that the recycling of the waste plastics and the paper boxes is actively promoted by all administrative mansions.

Today, the problem of contamination of waste plastics and cartons is solved in three ways: recycling, landfill treatment and development of degradable plastic cartons. From the aspect of environmental protection, the recycling of the waste plastics and the paper boxes can eliminate environmental pollution, and meanwhile, precious energy and resources can be obtained, thereby bringing obvious environmental and social benefits. Meanwhile, the reasonable technology can also generate certain economic benefit, and the recycling of the waste plastics and the paper boxes is in accordance with the current Chinese situation because the plastic paper boxes are in the period of inundation nowadays.

The recovery processing method of the waste mixed plastic and the paper box mainly comprises four methods: melting regeneration, combustion heat utilization, pyrolysis conversion and modification of recovered chemicals. However, in the process of recycling and processing waste plastics, a plurality of outstanding problems are encountered, and the paper boxes are mainly burnt by fire directly:

(1) secondary pollution and corrosion to equipment can occur in the combustion heat-taking process, and a large amount of harmful gas and black smoke are generated along with the combustion heat-taking process. The prior art for taking heat by combustion in China is not mature, and a large amount of funds are needed to support incineration equipment;

(2) the catalyst has the failure problem in the catalytic cracking process, because the waste plastics have poor heat-conducting property, and part of the mixed non-pyrolyzable substances coke and deactivate the surface of the catalyst in the cracking production of fuel oil, and in addition, hydrogen chloride generated in the PVC pyrolysis process can also cause catalyst poisoning;

(3) recycling chemical products and sorting waste plastics before simple regeneration.

The identification and sorting of the waste plastics are bottleneck links which restrict the recycling of the waste plastics at the present stage. If different types of plastics are mixed together for utilization without discrimination, the product performance is easily reduced, the treatment process efficiency is low, the technical complexity is increased, even waste products occur, and secondary pollution is generated. Therefore, the identification and separation of waste plastics and paper is a primary task for recycling.

Disclosure of Invention

The invention mainly solves the technical problem of providing an intelligent garbage classification device, which adopts an infrared spectrum technology to identify and classify plastic garbage and paper garbage, is convenient to recycle, avoids secondary pollution, reduces subsequent treatment procedures, saves a large amount of time, and is convenient to operate and implement in life and industry.

In order to solve the technical problems, the invention adopts a technical scheme that: the intelligent garbage classification device is provided, the upper end of the garbage box body is provided with a garbage inlet, the interior of the garbage box body is of a double-layer structure and comprises a first layer cavity and a second layer cavity which are communicated with each other, a conveyor belt, a shock and dispersion device, an electric motor and a roller are arranged in the first layer cavity, the roller is respectively connected with the shock and dispersion device and the electric motor and is positioned above the conveyor belt, a garbage placing and classifying plate, a guide rail, a moving block, a mechanical arm, an infrared probe, a spectrum analyzer and an electric motor are arranged in the second layer cavity, the guide rail is arranged at the top of the second layer cavity, the mechanical arm is arranged below the guide rail in a sliding manner through the moving block, the garbage placing and classifying plate is horizontally arranged at one inner wall of the garbage box body and is positioned below the mechanical arm, and the electric motor is arranged at the upper part of the mechanical arm and is connected with the spectrum, the infrared probe is arranged at the bottom of the spectrum analyzer.

In a preferred embodiment of the present invention, the bottom of the garbage can is respectively provided with a plastic discharge port, a paper discharge port and other garbage discharge ports.

In a preferred embodiment of the present invention, the plastic material outlet is located on the left side of the other garbage outlets, and the paper material outlet is located on the right side of the other garbage outlets.

In a preferred embodiment of the present invention, a plastic recycling bin, a paper recycling bin and other garbage recycling bins are correspondingly disposed below the plastic material outlet, the paper material outlet and other garbage outlets.

In a preferred embodiment of the present invention, the spectrum analyzer is a fourier near infrared spectrometer.

In order to solve the technical problem, the invention adopts another technical scheme that: the classification method of the intelligent garbage classification device comprises the following specific steps:

a. garbage enters the first-layer cavity inside the garbage can body from the garbage inlet,

b. the vibration scattering device in the first layer cavity drives the roller through the motor, and the garbage is fluctuated up and down through the rolling of the roller, so that the garbage is scattered in a vibration mode;

c. the garbage scattered by vibration falls to the upper end of a garbage storage and classification plate in the cavity of the second layer by using a conveyor belt;

d. the spectrum analyzer and the infrared probe freely move in the second cavity through the guide rail, and the infrared probe scans all the garbage at the upper end of the garbage storage classifying plate;

e. distinguishing and processing the scanned result by a spectrum analyzer, and identifying and classifying the plastic garbage and the paper garbage;

f. when the spectral reaction is identified to be at 730cm, the garbage is judged to be plastic garbage, when the spectral reaction is scanned to be between 1500-1600cm, the garbage is judged to be paper garbage,

g. controlling a mechanical arm to pick up according to the judgment result, respectively clamping the plastic garbage and the paper garbage, then placing the clamped plastic garbage and paper garbage into corresponding plastic discharge ports and paper discharge ports, and respectively dropping the clamped plastic garbage and paper garbage into corresponding plastic recovery barrels and paper recovery barrels;

h. after all plastic waste and paper class rubbish are categorised, other rubbish is directly pressed from both sides through the arm again and is put into other rubbish discharge gates in the middle of, falls into other garbage recycling bin, accomplishes the classification of all rubbish.

In a preferred embodiment of the present invention, the method for determining process includes the steps of:

step 1: collecting an original spectrum of a plastic and paper standard sample by using a spectrum analyzer;

step 2: performing infrared spectrum K-M conversion on the acquired original spectrum of the plastic and paper standard sample;

and step 3: performing principal component analysis on the spectrum data subjected to the K-M conversion processing to extract a characteristic spectrum;

and 4, step 4: using Fisher to judge and establish a judgment model for the extracted characteristic spectrum;

and 5: and identifying and classifying the plastics and the paper by applying a discrimination model.

In a preferred embodiment of the present invention, the original spectrum of the plastic is 730cm, and the original spectrum of the paper is 1500-1600 cm.

In a preferred embodiment of the present invention, in the step 4, a BP neural network model is used instead of Fisher discrimination to establish a discriminant model.

In a preferred embodiment of the present invention, in the step 4, a PNN neural network model is used to replace Fisher discrimination to establish a discrimination model.

The invention has the beneficial effects that: the intelligent garbage classification device and the classification method thereof adopt the infrared spectrum technology to identify and classify the plastic garbage and the paper garbage, are convenient to recycle and treat, avoid secondary pollution, reduce the subsequent treatment procedures, save a large amount of time and are convenient to operate and implement in life and industry.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without inventive efforts, wherein:

FIG. 1 is a schematic structural diagram of an intelligent garbage classification device according to a preferred embodiment of the present invention;

FIG. 2 is an IR spectrum of a plastic;

FIG. 3 is an infrared spectrum of paper;

the labels in the figures are: 1. garbage bin, 2, rubbish entry, 3, conveyer belt, 4, first floor cavity, 5, second floor cavity, 6, shake and loose ware, 7, motor, 8, cylinder, 9, rubbish put thing classification board, 10, guide rail, 11, movable block, 12, arm, 13, infrared probe, 14, spectral analysis appearance, 15, electric motor, 16, plastics discharge gate, 17, paper discharge gate, 18, other rubbish discharge gate, 19, plastics recycling bin, 20, paper recycling bin, 21, other rubbish recycling bin.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

As shown in fig. 1, an embodiment of the present invention includes:

the utility model provides an intelligence waste classification device, includes rubbish box 1, the upper end of rubbish box 1 be provided with rubbish entry 2, the inside of rubbish box 1 divide into bilayer structure, include through the first layer cavity 4 and the second floor cavity 5 that are linked together.

The first-layer cavity 4 is internally provided with a conveyor belt 3, a shock absorber 6, a motor 7 and a roller 8, and the roller 8 is respectively connected with the shock absorber 6 and the motor 7 and is positioned above the conveyor belt 3.

The garbage storage classifying plate is characterized in that a garbage storage classifying plate 9, a guide rail 10, a moving block 11, a mechanical arm 12, an infrared probe 13, a spectrum analyzer 14 and an electric motor 15 are arranged in the second-layer cavity 5, the guide rail 10 is arranged at the top of the second-layer cavity 5, the mechanical arm 12 is arranged below the guide rail 10 in a sliding mode through the moving block 11, the garbage storage classifying plate 9 is horizontally arranged on one inner wall of the garbage box body 1 and located below the mechanical arm 12, the electric motor 15 is arranged on the upper portion of the mechanical arm 12 and connected with the spectrum analyzer 14, and the infrared probe 13 is arranged at the bottom of the spectrum analyzer 14.

The bottom of the garbage can body 1 is respectively provided with a plastic discharge port 16, a paper discharge port 17 and other garbage discharge ports 18. The plastic material outlet 16 is positioned at the left side of other garbage outlet 18, and the paper material outlet 17 is positioned at the right side of other garbage outlet 18.

Wherein, the plastic discharge port 16, the paper discharge port 17 and other garbage discharge ports 18 are respectively and correspondingly provided with a plastic recycling bin 19, a paper recycling bin 20 and other garbage recycling bins 21.

In this embodiment, the spectrum analyzer 14 is a fourier near infrared spectrometer; the arm 12 has three segments that can be freely bent and extended.

The invention also provides a classification method of the intelligent garbage classification device, which comprises the following specific steps:

When the spectral response is scanned at 730cm, it is determined as plastic, and the mechanical arm 12 then picks it up and places it into the plastic recycling bin 19 on the left side. When the distance reaches 1500-1600cm, the paper is judged to be paper, and the mechanical arm 12 is extended to clamp the paper and place the paper into the paper recycling bin 20 on the right side. In another case, when the plastic and paper are overlapped, the infrared scanning can be performed perspective, the mechanical arm 12 is extended out, namely a needle which is extended out by 0.5mm when the mechanical arm is about to touch the garbage is inserted into the garbage by 0.2mm, and then the mechanical arm is bent to hook the garbage, so that two different types of garbage are prevented from being clamped into the same garbage recycling bin together. After all plastics and paper class are categorised, other rubbish will be directly put into other garbage collection bucket 21 in the middle of, and is convenient succinct.

The method for judging and processing comprises the following steps:

As shown in FIGS. 2 and 3, the original spectrum of the plastic is 730cm, and the original spectrum of the paper is 1500-1600 cm.

Further, in the step 4, a Back Propagation (BP) neural network model is used for replacing Fisher to judge and establish a judgment model, or, in the step 4, a PNN neural network model is used for replacing Fisher to judge and establish a judgment model. The analysis process of the BP neural network model or the PNN neural network model is simple, and the practicability is strong.

The method comprises the steps of collecting an original spectrum of a plastic and paper standard sample by a Fourier near-infrared spectrometer, extracting characteristic wavelengths by range value conversion and principal component analysis of the original spectrum, establishing a discrimination model according to the reflectivity corresponding to the characteristic wavelengths, and performing discrimination analysis on an unknown sample by using the established discrimination model. The characteristic wavelength which can represent most of information of the original spectrum is extracted, the calculation amount of subsequent processing is reduced, and the calculation time is saved. Meanwhile, the near infrared spectrum analysis technology has low requirements on samples, and the method which is roughly classified is practical for the society.

Plastics in identification have specific characteristics: PVC plastics contain a large amount of ester-group-containing plasticizers, so that cartons and plastics can be conveniently and simply classified effectively.

The near infrared spectrum of the plastic and paper samples after K-M conversion is subjected to principal component analysis, two spectrogram information is extracted, and the difference between the two spectrograms is large, so that the two spectrograms can be distinguished at a glance.

Example 1:

the surface of the waste plastic is cleaned and polished by No. 360 abrasive paper, then cut into small blocks with the size of 2mm by 2mm, and finally the number of the small blocks is measured by a diffuse reflection mode at the room temperature of 20-25 ℃.

The invention adopts FT S6000 type Fourier infrared spectrometer produced by American Bio-rad company to collect the near infrared spectrum of a sample, the spectrum wavelength range is 730cm, and the K-M conversion is carried out on the original spectrum of the collected standard sample.

Carrying out principal component analysis on a near infrared spectrum of waste plastics and carton samples serving as a standard sample set after K-M conversion by using MATLAB to obtain principal component scores and load coefficients, wherein the accumulated contribution rate of the first 4 principal components reaches 96.62%, the accumulated contribution rate can represent most characteristic information of an original spectrum, the contribution rate of the first principal component is 75.68%, so that 18 characteristic wavelengths are selected by taking the load coefficient of the first principal component as a main component and comprehensively considering the load coefficients of the first 4 principal components, the 18 characteristic wavelength data of the extracted plastics of the 86 standard sample sets are imported into IBM SPSS statics 19 software, and a Fisher discriminant model is established. The established Fisher discriminant model correctly classifies 96.5% of the initial grouping cases and 89.5% of the cross-validation grouping cases. And then, the established Fisher discrimination model is used for carrying out identification analysis on 74 unknown samples, and only 3 discrimination errors exist in the discrimination and classification of the plastic and paper samples.

Example 2:

referring to the method in

embodiment

1, 18 pieces of characteristic wavelength data of plastics of 86 standard sample sets extracted are used as input of a BP neural network model, the BP neural network model is trained, the number of trained BP neural network model input layer nerves is 18, the number of hidden layer nerves is 20, the number of output layer nerves is 6, a target error is 0.01, a transfer function of the hidden layer adopts a double tangent S-type transfer function tan sig, an output layer adopts a purelin function, a learning function tranlmm is selected, and the training times are 100. And then carrying out prediction classification on the unknown sample set by using the trained BP neural network model. The discrimination accuracy of the BP neural network model on the standard sample set reaches 98.84%, the prediction accuracy on the plastic and paper box sample set reaches 73.33%, and the requirement on waste plastic and paper classification can be basically met.

Example 3:

referring to the method in example 1, the PNN neural network model is trained by using 18 characteristic wavelength data of plastics of the extracted 86 standard sample sets as input of the PNN neural network model, and the dispersion constant of the trained PNN neural network model is 0.1, and the maximum number of neurons in the middle layer is 86. And then carrying out prediction classification on the unknown sample set by using the trained PNN network model. The PNN network training model achieves 100% of classification accuracy of the training set and 97.67% of classification accuracy of the prediction set, and meets the requirements for waste plastic and paper classification.

In conclusion, the intelligent garbage classification device and the classification method thereof adopt the infrared spectrum technology to identify and classify the plastic garbage and the paper garbage, are convenient to recycle and treat, avoid secondary pollution, reduce the subsequent treatment procedures, save a large amount of time and are convenient to operate and implement in life and industry.

The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by the present specification, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims

1. An intelligent garbage classification device is characterized by comprising a garbage box body, wherein a garbage inlet is formed in the upper end of the garbage box body, the interior of the garbage box body is of a double-layer structure and comprises a first layer cavity and a second layer cavity which are communicated with each other, a conveying belt, a shock dissipating device, an electric motor and a roller are arranged in the first layer cavity, the roller is respectively connected with the shock dissipating device and the electric motor and is positioned above the conveying belt, a garbage placing classification plate, a guide rail, a moving block, a mechanical arm, an infrared probe, a spectrum analyzer and an electric motor are arranged in the second layer cavity, the guide rail is arranged at the top of the second layer cavity, the mechanical arm is arranged below the guide rail in a sliding manner, the garbage placing classification plate is horizontally arranged on one inner wall of the garbage box body and is positioned below the mechanical arm, the electric motor is arranged on the upper portion of the mechanical arm and is connected with the spectrum analyzer, the infrared probe is arranged at the bottom of the spectrum analyzer.

2. The intelligent garbage classification device according to claim 1, wherein the bottom of the garbage can body is respectively provided with a plastic discharge port, a paper discharge port and other garbage discharge ports.

3. The intelligent garbage classification device according to claim 2, wherein the plastic material outlet is positioned at the left side of other garbage outlets, and the paper material outlet is positioned at the right side of other garbage outlets.

4. The intelligent garbage classification device according to claim 3, wherein a plastic recycling bin, a paper recycling bin and other garbage recycling bins are correspondingly placed below the plastic discharging port, the paper discharging port and other garbage discharging ports respectively.

5. The intelligent garbage classification device according to claim 1, wherein the spectrum analyzer is a Fourier near infrared spectrometer.

6. The classification method of the intelligent garbage classification device according to one of the claims 1 to 5, characterized by comprising the following specific steps:

7. The classification method of an intelligent garbage classification device according to claim 6, wherein the method for discriminating processing comprises the following steps:

8. The method as claimed in claim 7, wherein the original spectrum of the plastic is 730cm, and the original spectrum of the paper is 1500-1600 cm.

9. The method according to claim 1, wherein in step 4, a BP neural network model is used to replace Fisher's discrimination to establish a discrimination model.

10. The classification method of an intelligent garbage classification device according to claim 1, wherein in the step 4, a PNN neural network model is used to replace Fisher discrimination to establish a discrimination model.