Waste plastic sorting device, sorting method and program product

Abstract

A device for determining the material of waste plastics, comprising: a first determination unit that determines whether the spectrum of the emitted light of the light irradiated on the conveying path of the conveyor by the illumination device and detected by the mid-infrared camera is the spectrum of a waste plastic sheet or the spectrum of the conveying path; a second determination unit that extracts two kinds of characteristic data from the spectrum of the waste plastic sheet determined by the first determination unit; and a third determination unit that discriminates the material of the waste plastic sheet based on the characteristic data extracted by the second determination unit.

Description

Technical Field

[0001] The present disclosure relates to a device for determining the material of waste plastics, a method for determining the material, and a program for determining the material. Background Art

[0002] In the reprocessing of waste plastics, in order to achieve material recycling, it is sought that the non-target objects mixed in the sorted products are less and the purity is higher. In addition, when the raw material contains an expensive raw material, it is sought to be able to sort the expensive raw material in a non-missing manner. In addition, it is also sought to effectively determine and sort the black plastics that have traditionally had to be thermally recycled because they could not be distinguished, in order to recycle their materials.

[0003] Patent Document 1 describes irradiating an object to be sorted with infrared light and receiving the reflected light from the object to be sorted, and determining the resin type of the object to be sorted by a pattern matching method using the spectrum of the reflected light.

[0004] <Prior Art Documents>

[0005] <Patent Documents>

[0006] Patent Document 1: Japanese Unexamined Patent Application Publication No. 2018-100903 Summary of the Invention

[0007] <Problems to be Solved by the Invention>

[0008] However, in the conventional methods described in Patent Document 1 and the like, there is room for improvement in the determination accuracy of the material.

[0009] An object of the present disclosure is to provide a device for determining the material of waste plastics, a method for determining the material, and a program for determining the material, which can improve the determination accuracy of the material of waste plastics.

[0010] <Means for Solving the Problems>

[0011] A device for determining the material of waste plastics according to one aspect of an embodiment of the present invention includes: an irradiation unit that irradiates waste plastic pieces conveyed on a conveyance path with light; a reflection spectrum detection unit that receives the reflected light of the light irradiated by the irradiation unit to detect the spectrum of the reflected light; a first determination unit that determines whether the spectrum detected by the reflection spectrum detection unit is the spectrum of the waste plastic pieces or the spectrum of the conveyance path; a second determination unit that extracts a feature amount from the spectrum determined by the first determination unit to be the spectrum of the waste plastic pieces; and a third determination unit that discriminates the material of the waste plastic pieces based on the feature amount extracted by the second determination unit.

[0012] <Effects of the Invention>

[0013] According to the present disclosure, a material determination device, a material determination method, and a material determination program can be provided, which can improve the determination accuracy of the material of waste plastics. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 FIG. 1 is a perspective view showing a schematic configuration of a waste plastic material determination device according to an embodiment.

[0015] Figure 2 FIG. Figure 1 2 is a side view of the waste plastic material determination device shown in FIG. 1.

[0016] Figure 3 FIG. Figure 1 3 is a plan view of the waste plastic material determination device shown in FIG. 1.

[0017] Figure 4 FIG. 4 is a functional block diagram of the discrimination device.

[0018] Figure 5 FIG. 5 is a flowchart of the waste plastic material discrimination process according to an embodiment.

[0019] Figure 6 FIG. 6 is a diagram showing a method for extracting a calibration spectrum.

[0020] Figure 7 FIG. 7 is a diagram showing an example of a process of cutting out a wavelength region with characteristics from a reflected wave spectrum.

[0021] Figure 8 FIG. 8 is a diagram showing an example of extracting characteristic data.

[0022] Figure 9 FIG. 9 is a diagram showing an example of material determination using a decision tree.

[0023] Figure 10 FIG. 10 is a plan view showing a first material sorting method performed by the material determination device of the present embodiment.

[0024] Figure 11 FIG. 11 is a plan view showing a second material sorting method performed by the material determination device of the present embodiment.

[0025] Figure 12 FIG. 12 is a plan view showing a third material sorting method performed by the material determination device of the present embodiment.

[0026] Figure 13 FIG. 13 is a plan view showing a fourth material sorting method performed by the material determination device of the present embodiment.

[0027] Figure 14 FIG. 14 is a plan view showing a fifth material sorting method performed by the material determination device of the present embodiment.

[0028] Figure 15 This is a diagram showing an example of the operation screen of the material determination device. Detailed implementation manners

[0029] Hereinafter, the implementation manners will be described with reference to the accompanying drawings. For the convenience of understanding the description, the same reference numerals are given to the same components as much as possible in each drawing, and repeated descriptions are omitted.

[0030] It should be noted that in the following description, the x-direction, y-direction, and z-direction are mutually perpendicular directions. The x-direction and y-direction are horizontal directions, and the z-direction is the vertical direction. The x-direction is the conveying direction of the conveying path 3 of the conveyor 2. The y-direction is the width direction of the conveying path 3 of the conveyor 2. In addition, hereinafter, for the convenience of description, the positive z-direction side may sometimes be represented as the upper side, and the negative z-direction side may be represented as the lower side.

[0031] Refer to Figures 1 to 3 , and the schematic configuration of the waste plastic material determination device 1 according to the implementation manner will be described. Figure 1 This is a perspective view showing the schematic configuration of the waste plastic material determination device 1 according to the implementation manner. Figure 2 This is Figure 1 a side view of the waste plastic material determination device 1 shown in Figure 3 This is Figure 1 a plan view of the waste plastic material determination device 1 shown in. Here, the case where the waste plastic to be determined is black waste plastic and the two materials S1, S2 (represented by quadrilateral and triangular marks in Figures 1 to 3 ) are mixed will be taken as an example for description. Hereinafter, the black waste plastic pieces of the two materials S1 and S2 may be aggregated and represented by the symbol S.

[0032] The waste plastic material determination device 1 for black waste plastic mainly includes a vibratory feeder 8 as an example of a supply unit and a conveyor 2 as an example of a conveying unit. The vibratory feeder 8 sequentially supplies the black waste plastic pieces S1 and S2, and the conveyor 2 conveys the black waste plastic pieces S1 and S2 supplied by the vibratory feeder 8. For example, the crushed black waste plastic pieces S1 and S2 are supplied to the vibratory feeder 8 via a feed hopper or the like. The vibratory feeder 8 vibrates the placement surface on which the black waste plastic pieces S1 and S2 are placed, so as to supply them to the conveyor 2 while preventing the black waste plastic pieces S1 and S2 from overlapping each other. The conveyor 2 has a conveying path 3 on its upper surface, and conveys the black waste plastic pieces S1 and S2 on the conveying path 3 in a direction away from the vibratory feeder 8.

[0033] In addition, the material determination device 1 mainly includes a lighting device 10 as an example of the irradiation unit, a mid-infrared camera 4 as an example of the reflection spectrum detection unit, and a discrimination device 5. The lighting device 10 irradiates infrared rays onto the black waste plastic sheets S1 and S2. The mid-infrared camera 4 detects the reflection spectra from the black waste plastic sheets S1 and S2. The discrimination device 5 identifies the materials of the black waste plastic sheets S1 and S2 based on the reflection spectra detected by the mid-infrared camera 4. The lighting device 10 has a lamp 10A (see Figure 6 ) such as a halogen tungsten lamp as an infrared light source, and irradiates infrared rays from the lamp 10A onto the black waste plastic sheets S1 and S2. In addition, the lighting device 10 is arranged to let the reflected light from the black waste plastic sheets S1 and S2 enter the mid-infrared camera 4, and is arranged on both sides (or one side above) in the flow direction of the conveyor 2 relative to the mid-infrared camera 4.

[0034] For example, as Figure 1 shown, one mid-infrared camera 4 can measure across the entire area in the width direction of the conveyor 2, and can be divided into multiple (for example, 318) areas in the width direction to receive the near-infrared reflected light from the black waste plastic sheets S1 and S2, so as to measure the spectra of the reflected light for each area. The mid-infrared camera 4 is composed of, for example, a camera with a spectroscope in the wavelength region of mid-infrared rays of 3 μm or more. The mid-infrared camera 4 measures at a scanning frequency of, for example, 230 Hz, and sends 318 spectral data to the discrimination device 5 for each scan. The discrimination device 5 outputs the material determination results of the 318 individual areas to the spraying control unit 6 described later based on the 318 spectral data received from the mid-infrared camera 4.

[0035] In addition, the material determination device 1 is provided with a nozzle 7 on the downstream side in the conveying direction of the conveyor 2, which sprays air in a lateral or oblique direction perpendicular to the conveying direction. A plurality of (for example, 318) nozzles 7 are arranged side by side in the width direction of the conveyor 2, and the operation of each nozzle is controlled by the spraying control unit 6. The spraying control unit 6 sorts the black waste plastic sheets S1 and S2 and makes them fall into, for example, a plurality of areas (such as recycling hoppers, etc.) divided by a partition plate 9 according to the material determination results received from the discrimination device 5, so as to collect the waste plastic of the required material. In other words, in this embodiment, the spraying control unit 6, the nozzle 7, and the partition plate 9 function as a collection device 12 for collecting the waste plastic sheets of the required material from the waste plastic sheets flowing through the conveying path 3 of the conveyor 2 based on the material determination results obtained by the discrimination device 5.

[0036] The operation of the material determination device 1 will be described. When, for example, the crushed black waste plastic pieces S1 and S2 are supplied to the vibrating feeder 8 via a hopper for input materials, etc., the vibrating feeder 8 applies vibration to the supplied black waste plastic pieces S1 and S2 and conveys them downstream without overlapping and supplies them to the conveyor 2.

[0037] The black waste plastic pieces S1 and S2 on the conveying path 3 on the upper surface of the conveyor 2 are conveyed along the conveying direction on the positive x-direction side while being irradiated with infrared light from the lighting device 10 at a position where the mid-infrared camera 4 can take pictures. The mid-infrared camera 4 receives the reflected light of the infrared rays emitted from the lighting device 10 and reflected by the black waste plastic pieces S1 and S2, and outputs the light reception result (data of the light reception spectrum) to the discrimination device 5.

[0038] The discrimination device 5 identifies the materials of the black waste plastic pieces S1 and S2 based on the light reception result input from the mid-infrared camera 4. It should be noted that the details of the material determination method performed by the discrimination device 5 will be described later with reference to Figures 4 to 9 The discrimination device 5 outputs the material identification result to the injection control unit 6.

[0039] The injection control unit 6 selects the nozzle 7 corresponding to the material from among the multiple arranged nozzles 7, measures the timing, and sends a control signal. The nozzle 7 that receives the control signal opens the nozzle orifice and injects air. By injecting air from the nozzle 7 at an appropriate timing according to the discrimination result of the discrimination device 5, it is possible to separate and recover the materials of the picking objects and the non-picking objects.

[0040] In Figure 2 、 Figure 3 In the example of, the black waste plastic piece S1 on the conveyor 2 is subjected to the air from the air nozzle 7 that has received the control signal, and is blown into the collection device 12 provided for each material and recovered. In addition, since the black waste plastic piece S2 on the conveyor 2 is not subjected to the air from the nozzle 7, it is recovered into a collection device 12 different from that of the black waste plastic piece S1. In this way, by the injection and stop of the nozzle 7, it is possible to sort and recover the black waste plastic pieces of multiple materials for each material.

[0041] With reference to Figures 4 to 9 The method for discriminating the materials of waste plastics performed by the discrimination device 5 will be described. Figure 4 This is the functional block diagram of the discrimination device 5.

[0042] As Figure 4 shown, the discrimination device 5 has a preprocessing unit 51, a first determination unit 52, a second determination unit 53, and a third determination unit 54.

[0043] The preprocessing unit 51 performs preprocessing such as correction and processing on the reflection spectra of the black waste plastic sheets S1 and S2 detected by the mid-infrared camera 4. For example, the preprocessing unit 51 corrects the detected reflection spectra using the spectra measured under conditions where the reflected light is brighter and the spectra measured under conditions where the reflected light is darker. The "darker conditions" refer to conditions that are relatively darker compared to the above-mentioned "brighter conditions". In addition, the preprocessing unit 51 performs processing to cut out a predetermined frequency range from the corrected spectra.

[0044] The first determination unit 52 determines whether the spectrum detected by the mid-infrared camera 4 is the spectrum of the waste plastic sheets S1 and S2 or the spectrum of the conveying path 3 of the conveyor 2. The first determination unit 52 performs the determination using the learned One Class SVM (Support Vector Machine).

[0045] One Class SVM is a type of SVM that is a classification algorithm in machine learning. In SVM, the support vectors of each class (the positions closest to other classes in the training data) are used as a reference, and the recognition boundary is set so as to maximize the Euclidean distance. In addition, in the case where the features are non-linear features, a kernel is used to map the data to a feature space. By appropriately selecting the kernel, it is possible to draw a recognition boundary even when the data arrangement is complex.

[0046] In One Class SVM, a method called the kernel trick is used for the training data of one type to map the data to the feature space of a high-dimensional space. At this time, since the training data is mapped in a manner that is arranged far from the origin, data that is not similar to the original training data gathers near the origin. This property is used to distinguish normal data (conveyor 2) from abnormal data (objects (waste plastic sheets S1 and S2)).

[0047] By using One Class SVM with excellent pattern recognition ability in the first determination unit 52, it is possible to accurately identify whether the reflection spectrum is the spectrum reflected by the waste plastic sheets S1 and S2 or the spectrum reflected by the conveying path 3 of the conveyor 2. It should be noted that a supervised learning classification method of machine learning other than One Class SVM may also be applied in the first determination unit 52.

[0048] The second determination unit 53 extracts feature data Score1 and Score2 (feature quantities) from the spectra determined to be the spectra of waste plastic sheets by the first determination unit 52. The second determination unit 53 performs the determination using the learned PLS (Partial Least Squares).

[0049] PLS is a regression algorithm for supervised learning in machine learning, which performs regression analysis only between a few principal components among the principal components calculated from the explanatory variables and the target variable. In PLS, the principal components are calculated in such a way that the covariance with the target variable is large. In this embodiment, based on the explanatory variables of the reflection spectra judged to be reflected by the waste plastic sheets S1 and S2, two characteristic data, Score1 and Score2, are calculated using PLS.

[0050] By using PLS in the second determination unit 53, it is possible to reduce the multi-dimensional explanatory variables of the reflection spectrum to a small number of features, so that appropriate characteristic data Score1 and Score2 that are easier to distinguish can be extracted. It should be noted that a multivariate analysis method of machine learning other than PLS can also be applied in the second determination unit 53.

[0051] The third determination unit 54 discriminates the materials of the waste plastic sheets S1 and S2 corresponding to the spectrum based on the two characteristic quantities Score1 and Score2 of the reflection spectrum extracted by the second determination unit 53. The third determination unit 54 uses a learned decision tree for determination. A decision tree is a classification algorithm for supervised learning. A decision tree uses a tree structure to represent the rules for classifying the target variable and is generally used for classification problems.

[0052] By using the decision tree in the third determination unit 54, it is possible to accurately discriminate the materials of the waste plastic sheets S1 and S2 based on the two characteristic quantities Score1 and Score2 of the reflection spectrum. It should be noted that a supervised learning classification method of machine learning other than the decision tree can also be applied in the third determination unit 54.

[0053] The discrimination device 5 can be physically configured as a computer system including a CPU (Central Processing Unit), a RAM (Random Access Memory) as a main storage device, a ROM (Read Only Memory), a communication module, an auxiliary storage device, and the like. Figure 4 Each function of the shown discrimination device 5 is realized by reading a predetermined computer software (material determination program) into the CPU, RAM, etc., causing each hardware to operate under the control of the CPU, and reading and writing data in the RAM. That is, by executing the material determination program according to this embodiment on a computer, the discrimination device 5 functions as Figure 4 the preprocessing unit 51, the first determination unit 52, the second determination unit 53, and the third determination unit 54.

[0054] The material discrimination program of this embodiment is stored in a storage device provided in a computer, for example. It should be noted that the material discrimination program can be configured such that part or all of it is transmitted via a transmission medium such as a communication line, and received and recorded (including installed) by a communication module or the like provided in the computer. In addition, the material discrimination program can be configured such that part or all of it is recorded (including installed) in the computer from a state stored in a removable storage medium such as a CD-ROM, DVD-ROM, or flash memory.

[0055] The discrimination device 5 can be a circuit composed of an analog circuit, a digital circuit, or an analog / digital hybrid circuit. In addition, a control circuit for controlling each function of the discrimination device 5 can be provided. The implementation of each circuit can be performed by an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0056] Similarly, the injection control unit 6 can also be physically configured as a computer system including a CPU, RAM, ROM, communication module, auxiliary storage device, etc., and its functions can be realized by reading a predetermined computer software into the CPU, RAM, etc.

[0057] Figure 5 It is a flowchart of the material discrimination process for waste plastics according to the embodiment. Figure 5 Each process in the shown flowchart is executed by the discrimination device 5.

[0058] In step S01, the spectrum S org (n, w) obtained by the mid-infrared camera 4 is acquired by the preprocessing unit 51. Here, n is the number of sensors (the number of spectral detection regions divided by the mid-infrared camera 4 in the width direction of the conveyor 2). When the number of sensors is 318, integers 0 to 317 corresponding to each detection region are used. w is the wavelength of the spectrum. In this embodiment, a total of 131 wavelengths are set at an increment of 20 (nm) between 2700 (nm) and 5300 (nm), and integers 0 to 130 corresponding to each wavelength are used. In other words, S org (n, w) represents the value of the spectral intensity of wavelength w in the nth spectral detection region along the width direction of the conveyor 2.

[0059] In step S02, the spectrum S org (n, w) acquired in step S01 is corrected by the preprocessing unit 51, and the corrected spectrum S cor(n, w). Through this calibration, it is possible to absorb the differences in the characteristics of the spectral intensity caused by changes in the concentrations of water vapor and carbon dioxide in the measurement space, the temperatures of the black waste plastic sheets S1 and S2 to be measured, the aging deterioration of the lighting device 10 and the mid-infrared camera 4, the position on the conveyor 2, etc. The calibrated spectrum S cor (n, w) can be calculated, for example, by the following formula (1).

[0060] [Equation 1]

[0061]

[0062] Here, W ref (n, w) is the first calibration spectrum measured under the condition that the reflected light is bright. D ref (n, w) is the second calibration spectrum measured under the condition that the reflected light is darker than the above-mentioned brighter condition. These calibration spectra W ref (n, w), D ref (n, w) can be extracted, for example, when calibrating the mid-infrared camera 4 before performing the material discrimination process.

[0063] Figure 6 shows the calibration spectra W ref (n, w), D ref (n, w). As Figure 6 shown, by setting the calibration plate 11 for obtaining the calibration spectrum in the shooting area of the mid-infrared camera 4 on the conveying path 3 of the conveyor 2 and detecting the spectrum of the reflected light obtained by the mid-infrared camera 4, the calibration spectra W ref (n, w), D ref (n, w) can be obtained.

[0064] In the case of the first calibration spectrum W ref (n, w) measured under the condition that the reflected light is bright, place the calibration plate 11 (aluminum, stainless steel, etc.) for reflecting all wavelengths in the mid-infrared region, and in the state where the lighting device 10 is lit, for all sensors (n = 0, 1, 2,..., 317), obtain the data of all wavelengths (w = 0 (2700), 1 (2720), 2 (2740),..., 130 (5300)).

[0065] In the case of the second calibration spectrum D refIn the case of (n, w), a calibration plate 11 (aluminum, stainless steel, etc.) for reflecting all wavelengths in the mid-infrared region is placed, and in the state where the lighting device 10 is turned off (or in the state where the shutter of the camera is closed), data for all wavelengths (w = 0 (2700), 1 (2720), 2 (2740), …, 130 (5300)) are acquired for all sensors (n = 0, 1, 2, …, 317).

[0066] For example, as Figure 6 indicated by the dashed arrow in, the calibration plate 11 is preferably arranged to be movable between the position of the shooting area of the mid-infrared camera 4 on the conveying path 3 of the conveyor 2 where the calibration spectra W ref (n, w), D ref (n, w) are acquired and the standby position outside the shooting area of the mid-infrared camera 4 or the irradiation range of the lighting device 10. In other words, the calibration plate 11 is preferably capable of being fixed at a predetermined position within the field of view of the mid-infrared camera 4 and a predetermined position outside the field of view, and being movable between the two predetermined positions. The calibration plate 11 is preferably processed so that the surface roughness of the main surface receiving the light from the lighting device 10 is large and rough. Thereby, the occurrence of the halo of the reflected light can be suppressed.

[0067] In addition, the conveyor 2 can be stopped when acquiring the calibration spectra. In this case, if due to some failure in the operation of the calibration plate 11, the calibration plate 11 is not correctly arranged at the position of the shooting area of the mid-infrared camera 4, the temperature of the irradiated infrared part on the conveying path 3 of the conveyor 2 will rise due to the infrared rays of the lighting device 10, and there is a possibility of burning or catching fire. Therefore, it is preferable to provide an interlock device so as not to irradiate infrared rays from the lighting device 10 when the calibration plate 11 is not fixed within the field of view of the mid-infrared camera 4.

[0068] It should be noted that, as Figure 6 shown, the lighting device 10 has a lamp 10A (sheathed heater, carbon lamp, Kanthal lamp, etc.) as a light source of infrared rays, and a reflector 10B for collecting the heat of the lamp 10A. The lamp 10A is formed to extend in the width direction (y direction) of the conveyor 2 and is arranged to emit infrared rays in all directions around the axis along the y-axis. The reflector 10B is arranged on the opposite side of the conveying path 3 of the conveyor 2 with respect to the lamp 10A and is formed to be curved circumferentially around the axis of the lamp 10A, whereby the infrared rays emitted from the lamp 10A to the opposite side of the conveyor 2 can be collected and reflected and sent to the conveyor 2 side. The reflector 10B is made of, for example, aluminum, stainless steel, or a component with aluminizing, etc.

[0069] ReturnFigure 5 , in step S03, the wavelength region where features exist is cut out from the corrected spectrum S cor (n, w) by the preprocessing unit 51. Figure 7 FIG. is an example showing the process of cutting out the wavelength region where features exist from the reflected wave spectrum. Figure 7 The horizontal axis of represents the wavelength (nm) of the spectrum, and the vertical axis represents the intensity of the spectrum at each wavelength. Figure 7 FIG. shows an example of the spectra of the respective materials ABS, HIPS, PP, and PE. And, in Figure 7 the example of, the spectra in the wavelength regions of 3250 to 3750 (nm) and 4400 to 4600 (nm) are cut out. In Figure 7 the example of, the range of the cut-out wavelength region is shown by the shaded pattern.

[0070] Return Figure 5 , in step S04, by the first determination unit 52, using the spectrum that was corrected in step S02 and where the wavelength region where features exist was cut out in step S03, it is determined whether each spectrum is the conveyor belt (transport path 3) of the conveyor 2 or an object (waste plastic) on the transport path 3. In the present embodiment, the first determination unit 52 performs the determination using the learned OneClass SVM.

[0071] In step S05, by the second determination unit 53, using the learned PLS, two types of feature data Score1 and Score2 are extracted from the spectra determined to be objects (waste plastic) in step S4. Figure 8 FIG. is a diagram showing an example of the extraction of feature data. Figure 8 The horizontal axis of represents the first feature data (Score1), and the vertical axis represents the second feature data (Score2). Figure 8 FIG. shows Figure 7 the extraction examples of the four materials ABS, HIPS, PP, and PE exemplified in. As Figure 8 shown, it can be seen that: in the two-dimensional space obtained from the two feature data Score1 and Score2, the regions drawn for each material can be distinguished. It should be noted that the number of feature data can be other than two.

[0072] Return to Figure 5 , in step S06, by the third determination unit 54, based on the two types of feature data Score1 and Score2 extracted in step S5, the material is discriminated using the learned decision tree. Figure 9 FIG. is a diagram showing an example of material determination using a decision tree. In the present embodiment, in order to finally identify the four materials (PE, PP, ABS, HIPS), as Figure 9As shown, the decision tree has two levels of conditional branches. In the first level, the function f1(Score1, Score2) of the conditional branch is used to divide the dataset of the feature data Score1 and Score2 into two groups G1 and G2. In the second level, the function f2(Score1, Score2) of the conditional branch is used to further divide one of the groups G1 into two groups G11 and G12. In the second level, the function f3(Score1, Score2) of the conditional branch is used to further divide the other group G2 into two groups G21 and G22. Therefore, the dataset of the feature data Score1 and Score2 is classified into four groups G11, G12, G21, and G22, and the materials of each group are determined to be PE, PP, ABS, and HIPS, respectively.

[0073] As described above, the discrimination device 5 of the waste plastic material determination device 1 according to the present embodiment includes a first determination unit 52, a second determination unit 53, and a third determination unit 54. The first determination unit 52 determines whether the spectrum of the emitted light of the light irradiated on the conveying path 3 of the conveyor 2 by the illumination device 10 detected by the mid-infrared camera 4 is the spectrum of the waste plastic sheet S or the spectrum of the conveying path 3. The second determination unit 53 extracts two kinds of feature data Score1 and Score2 from the spectrum determined by the first determination unit 52 to be the spectrum of the waste plastic sheet S. The third determination unit 54 discriminates the materials S1 and S2 of the waste plastic sheet S based on the feature data Score1 and Score2 extracted by the second determination unit 53.

[0074] With this configuration, it is possible to obtain the output information of the material of the waste plastic through three-stage determination processing and data screening, including the screening of the spectrum of the black waste plastic sheet S through the object discrimination by the first determination unit 52, the dimensionality reduction from the spectral information to the feature data through the feature quantity extraction by the second determination unit 53, and the classification processing by the third determination unit 54, from the input information of the reflection spectrum. Therefore, the waste plastic material determination device 1 according to the present embodiment can determine the material of the waste plastic on the basis of considering various conditions, and can perform the determination in detail in a manner spanning multiple stages, thereby improving the determination accuracy of the material of the waste plastic.

[0075] In addition, the discrimination device 5 of the waste plastic material determination device 1 according to the present embodiment includes a preprocessing unit 51. The preprocessing unit 51 uses the first calibration spectrum W ref (n, w) measured under the condition where the reflected light is bright and the second calibration spectrum D ref (n, w) measured under the condition relatively darker than the bright condition to correct the reflection spectrum S org (n, w) detected by the mid-infrared camera 4.

[0076] By using, for example, the formula (1) in this way, the calibration spectrum W ref (n, w) and D ref (n, w) are used to calibrate the reflection spectrum S org (n, w), so that the differences in the characteristics of the spectral intensity caused by the temperature of the black waste plastic sheets S1 and S2 of the measurement object, the aging deterioration of the mid-infrared camera 4, the position on the conveyor 2, etc. can be suppressed. Therefore, by using the calibrated spectrum S cor (n, w) for the learning and determination of the first determination unit 52, the second determination unit 53, and the third determination unit 54, the determination accuracy of the material of the waste plastic can be further improved.

[0077] In addition, the pretreatment unit 51 further performs processing to cut out a predetermined frequency range from the calibrated spectrum S cor (n, w), and outputs the processed spectrum to the first determination unit 52.

[0078] With this configuration, since a part with a strong correlation with the material of the waste plastic can be extracted from the spectrum and used for the learning and determination of the first determination unit 52, the second determination unit 53, and the third determination unit 54, the mixing of noise that hinders learning or determination can be reduced, and the determination accuracy of the material of the waste plastic can be further improved.

[0079] It should be noted that although in this embodiment, the pretreatment unit 51 performs two processes, that is, the calibration process of the reflection spectrum S org (n, w) and the process of cutting out a predetermined frequency range, it may also be configured to perform only one of the two processes.

[0080] In addition, although in this embodiment, the case where the waste plastic to be determined for its material is black waste plastic S is taken as an example for description, it may also be other colored waste plastics such as red or blue. In addition, waste plastics of different colors may be mixed and used.

[0081] Refer to Figures 10 to 14 , and the method for sorting waste plastics of the desired material by the material determination device 1 of this embodiment will be described. Figure 10 It is a plan view showing the first material sorting method performed by the material determination device 1 of this embodiment. Figure 10 It shows the one corresponding to and simplified from the plan view of the material determination device 1 shown in Figure 3 . In the figures after Figure 10 , as an example of the sorting method, an example in which a plastic mixture mixed with 5 materials (1), (2), (3), (4), and (5) is taken as the sorting object will be described.

[0082] In Figure 10 the example of, it is illustrated the configuration of a conveying path of a single system formed on the conveyor 2 and the collecting device 12 without being divided in the width direction (y-direction) of the conveyor 2. In the collecting device 12, by the ejection and stop of the nozzle 7, the waste plastics are distinguished with the partition plate 9 shown in Figure 2 etc. as the boundary, and the sorting objects are roughly classified into two categories. Therefore, in order to sort the plastic mixture mixed with 5 kinds of materials as the sorting object by each single material respectively, it is necessary to repeatedly perform the process of classifying into one of the collecting devices 12-1 (for example, the device for recovering the waste plastics obtained by ejecting air from the nozzle 7) in the collecting device 12 according to each category. In other words, as shown in Figure 10 , first, only the material (1) is classified from the plastic mixture and recovered by the collecting device 12-1. At this time, the remaining plastic mixture collected in the other collecting device 12-2 is mixed with the other 4 kinds of materials (2) to (5). Then, the remaining plastic mixture mixed with 4 kinds is put into the material determination device 1 again, and any one of (2) to (5) is classified. By repeating this step 4 times, the 5 kinds of materials (1) to (5) can be distinguished respectively.

[0083] Figure 11 is a plan view showing the second material sorting method performed by the material determination device 1A of the present embodiment. In Figure 11 the following figures, the conveying path 3 of the conveyor 2 is divided into two systems, a first system and a second system, in the width direction. More specifically, the inlet (vibrating feeder 8), the conveyor 2, and the collecting device 12 are each divided into two parts in the width direction. It should be noted that the inlet 8 and the conveyor 2 do not have two constituent elements, but a partition device or the like is provided on a single constituent element to prevent mixing between systems. For example, the conveying path 3 of the conveyor 2 can be divided into two systems by providing a partition wall along the conveying direction at a position approximately in the center of the width direction.

[0084] In the following description, the first system is represented by the subscript A, and the second system is represented by the subscript B. In addition, the elements equivalent to Figure 10 the collecting device 12-1 are denoted as "collecting device A1" and "collecting device B1", and the elements equivalent to Figure 10 the collecting device 12-2 are denoted as "collecting device A2" and "collecting device B2".

[0085] In Figure 11 the example of, the waste plastic sheets of the same material are collected by the first system and the second system. For example, as shown in Figure 11As shown, a plastic mixture mixed with five materials (1) to (5) is supplied to inlets A and B of the first system and the second system respectively. Material determination is performed in each system, and then waste plastic sheets of the same material (1) are collected by collection devices A1 and B1 respectively. In addition, in collection devices A2 and B2, waste plastics mixed with the remaining materials (2) to (5) are collected.

[0086] Figure 12 is a plan view showing a third material sorting method performed by the material determination device 1B of the present embodiment. In Figure 12 example, in the first system, waste plastic sheets of the first material are collected, and the remaining waste plastic sheets are supplied to the second system. In the second system, waste plastic sheets of the second material are collected from the remaining waste plastic sheets. In Figure 12 example, various mixed materials can be classified into three materials: the first material, the second material, and other materials.

[0087] In Figure 12 example, a plastic mixture mixed with five materials (1) to (5) is supplied to inlet A of the first system. Material determination is performed by conveyor A of the first system, and waste plastic sheets of material (1) are collected in collection device A1. In addition, in collection device A2, waste plastics mixed with the remaining materials (2) to (5) are collected.

[0088] Next, the waste plastics mixed with the remaining materials (2) to (5) collected in collection device A2 are transported to inlet B of the second system through transport device 13 and supplied to inlet B. Material determination is performed by conveyor B of the second system, and waste plastic sheets of material (2) are collected in collection device B1. In collection device B2, waste plastics mixed with the remaining materials (3) to (5) are collected.

[0089] Figure 13 is a plan view showing a fourth material sorting method performed by the material determination device 1C of the present embodiment. In Figure 13 example, in the first system, waste plastic sheets of the first material and a small amount of other materials are collected, and the collected waste plastic sheets are supplied to the second system. In the second system, waste plastic sheets of the first material are collected from the waste plastic sheets of the first material and a small amount of other materials. In Figure 13 example, plastic sheets of a predetermined single material can be sorted with high purity.

[0090] In Figure 13In the example, a plastic mixture mixed with five materials (1) to (5) is supplied to the inlet A of the first system. The materials are judged by the conveyor A of the first system, and the waste plastic flakes of materials (1) and trace amounts of (2) to (5) are collected in the collection device A1. In addition, in the collection device A2, waste plastics mixed with the remaining materials (2) to (5) and trace amounts of (1) are collected.

[0091] Next, the waste plastics mixed with material (1) and trace amounts of (2) to (5) collected by the collection device A1 are transported to the inlet B of the second system through the transport device 13 and supplied to the inlet B. The materials are judged by the conveyor B of the second system, and material (1) is sorted again in the collection device B1 and the waste plastic flakes of material (1) are collected. The purity of the material (1) collected by the collection device B1 is higher than that of the material collected by the collection device A1. In the collection device B2, waste plastics mixed with the remaining materials (1) to (5) are collected.

[0092] Figure 14 It is a plan view showing the fifth material sorting method performed by the material judging device 1D of the present embodiment. In Figure 14 In the example, in the first system, waste plastic flakes of the first material and trace amounts of other materials are excluded, and the remaining waste plastic flakes after exclusion are supplied to the second system. In the second system, waste plastic flakes of the first material and trace amounts of other materials are further excluded from the other waste plastics to collect waste plastic flakes that do not contain the first material. In Figure 14 In the example, plastic flakes of a predetermined material can be more reliably selected from the mixed materials.

[0093] In Figure 14 In the example, a plastic mixture mixed with five materials (1) to (5) is supplied to the inlet A of the first system. The materials are judged by the conveyor A of the first system, and the waste plastic flakes of material (1) and trace amounts of (2) to (5) are collected by the collection device A1. In addition, in the collection device A2, waste plastics mixed with the remaining materials (2) to (5) and a small amount of (1) are collected.

[0094] Next, the waste plastics mixed with materials (2) to (5) and a small amount of (1) collected by the collection device A2 are transported to the inlet B of the second system through the transport device 13 and supplied to the inlet B. The materials are judged by the conveyor B of the second system, and material (1) is sorted again by the collection device B1 and the waste plastic flakes mixed with material (1) and trace amounts of materials (2) to (5) are collected. In the collection device B2, waste plastics mixed with the remaining materials (2) to (5) and trace amounts of (1) are collected.

[0095] Figure 15 This is a diagram showing an example of the operation screen of the material determination device 1. Figure 15 The operation screen shown is displayed on a display device provided, for example, in the main body of the material determination device 1. As Figure 15 shown, on the operation screen, the material names of the plastics to be sorted are listed, and spraying can be performed according to the above-mentioned first system ( Figure 15 "primary" in Figure 15 ") and the second system ( Figure 15 "secondary" in

[0096] respectively to individually select the materials to be sorted. The display device for displaying the operation screen is, for example, a touch screen, and it can be set to switch to the "ON" display by operating, such as pressing the "OFF" display of the "spray selection" column, so that in the case of this material (

[0097] ABS in

[0098] ), the nozzle 7 sprays air and is differentiated by the collection device. In addition, on the operation screen, "input raw material area ratio" can also be set, and according to the determination result of the material determination process, the ratio of each material mixed in the material is displayed.

[0099] 1, 1A, 1B, 1C, 1D: Material determination devices for waste plastics;

[0100] 2: Conveyor;

[0101] 3: Conveyor path;

[0102] 4: Mid-infrared camera (reflection spectrum detection unit);

[0103] 5: Discrimination device;

[0104] 51: Preprocessing unit;

[0105] 52: First determination unit;

[0106] 53: Second determination unit;

[0107] 54: Third determination unit;

[0108] 12, 12-1, 12-2, A1, A2, B1, B2: Collection devices;

[0109] S1, S2: Black waste plastic sheets.

Claims

1. A sorting device for waste plastics, comprising: An irradiation unit that irradiates light onto waste plastic sheets conveyed on a conveying path; A reflection spectrum detection unit that receives the reflected light of the light irradiated by the irradiation unit to detect the spectrum of the reflected light; A preprocessing unit that corrects the spectrum detected by the reflection spectrum detection unit using a first calibration spectrum measured under a condition where the reflected light is brighter and a second calibration spectrum measured under a condition darker than the brighter condition; A first determination unit that determines whether the spectrum detected by the reflection spectrum detection unit corrected by the preprocessing unit is the spectrum of the waste plastic sheet or the spectrum of the conveying path; A second determination unit that extracts a feature quantity from the spectrum determined by the first determination unit to be the spectrum of the waste plastic sheet; A third determination unit that discriminates the material of the waste plastic sheet based on the feature quantity extracted by the second determination unit; and A sorting unit that controls the timing of jetting air onto the waste plastic sheets conveyed on the conveying path according to the material determination result obtained by the third determination unit to sort the waste plastic sheets and make the waste plastic sheets fall into multiple areas, thereby sorting and collecting the waste plastic sheets by each material; The preprocessing unit corrects the spectrum using the following formula: [Equation 1] Among them, S org (n, w) is the spectrum detected by the reflection spectrum detection unit, W ref (n, w) is the first calibration spectrum, D ref (n, w) is the second calibration spectrum, S cor (n, w) is the spectrum after calibration.

2. The sorting device for waste plastics according to claim 1, wherein: The preprocessing unit performs a process of cutting out a predetermined frequency range from the corrected spectrum, The first determination unit uses the spectrum processed by the preprocessing unit for the determination.

3. The sorting device for waste plastics according to claim 1 or 2, wherein: The first determination unit uses a learned One Class SVM for the determination.

4. The sorting device for waste plastics according to claim 1 or 2, wherein: The second determination unit uses a learned PLS for the determination.

5. The sorting device for waste plastics according to claim 1 or 2, wherein: The third determination unit uses a learned decision tree for the determination.

6. The sorting device for waste plastics according to claim 1 or 2, wherein the sorting unit comprises: A collection device that collects waste plastic sheets of one material from the waste plastic sheets flowing through the conveying path based on the determination result of the third determination unit.

7. The sorting device for waste plastics according to claim 6, wherein: The conveying path is divided into two systems, a first system and a second system, in the width direction.

8. The sorting device for waste plastics according to claim 7, wherein: Waste plastic sheets of the same material are collected in the first system and the second system.

9. The sorting device for waste plastics according to claim 7, wherein: In the first system, waste plastic sheets of a first material are collected, and the remaining waste plastic sheets are supplied to the second system. In the second system, waste plastic sheets of a second material are collected from the remaining waste plastic sheets.

10. The sorting device for waste plastics according to claim 7, wherein: In the first system, waste plastic sheets of a first material and trace amounts of other materials are collected, and the collected waste plastic sheets are supplied to the second system. In the second system, waste plastic sheets of the first material are collected from the waste plastic sheets of the first material and trace amounts of other materials.

11. The waste plastic sorting device according to claim 7, wherein In the first system, waste plastic sheets of a first material and trace amounts of other materials are excluded, and the remaining waste plastic sheets after the exclusion are supplied to the second system. In the second system, waste plastic sheets of the first material and trace amounts of other materials are further excluded from the other waste plastic sheets to collect waste plastic sheets that do not contain the first material.

12. A method for sorting waste plastic, comprising: An irradiation step of irradiating light on waste plastic sheets conveyed on a conveying path; A reflected spectrum detection step of receiving the reflected light of the light irradiated in the irradiation step to detect the spectrum of the reflected light; A preprocessing step of correcting the spectrum detected in the reflected spectrum detection step using a first calibration spectrum measured under a condition where the reflected light is brighter and a second calibration spectrum measured under a condition darker than the brighter condition; A first determination step of determining whether the spectrum detected in the reflected spectrum detection step corrected in the preprocessing step is the spectrum of the waste plastic sheet or the spectrum of the conveying path; A second determination step of extracting a feature amount from the spectrum determined to be the spectrum of the waste plastic sheet in the first determination step; A third determination step of discriminating the material of the waste plastic sheet based on the feature amount extracted in the second determination step; and A sorting step of controlling the timing of jetting air onto the waste plastic sheets conveyed on the conveying path according to the material determination result of the third determination step to sort the waste plastic sheets and make the waste plastic sheets fall into a plurality of areas, thereby sorting and collecting the waste plastic sheets by each material. In the preprocessing step, the spectrum is corrected using the following formula: [Equation 2] Among them, S org (n, w) is the spectrum detected in the reflection spectrum detection step, W ref (n, w) is the spectrum for the first correction, D ref (n, w) is the spectrum for the second correction, S cor (n, w) is the spectrum after correction.

13. A program product, including a waste plastic sorting program, the sorting program enabling a computer to implement the following functions: An irradiation function of irradiating light on waste plastic sheets conveyed on a conveying path; A reflected spectrum detection function of receiving the reflected light of the light irradiated using the irradiation function to detect the spectrum of the reflected light; A preprocessing function of correcting the spectrum detected by the reflected spectrum detection function using a first calibration spectrum measured under a condition where the reflected light is brighter and a second calibration spectrum measured under a condition darker than the brighter condition; A first determination function of determining whether the spectrum detected by the reflected spectrum detection function corrected by the preprocessing function is the spectrum of the waste plastic sheet or the spectrum of the conveying path; A second determination function of extracting a feature amount from the spectrum determined to be the spectrum of the waste plastic sheet using the first determination function; A third determination function that discriminates the material of the waste plastic sheet based on the feature amount extracted by the second determination function; and A sorting function that controls the timing of blowing air onto the waste plastic sheet conveyed on the conveying path according to the material determination result obtained by the third determination function, sorts the waste plastic sheet, and causes the waste plastic sheet to fall into a plurality of areas, thereby sorting and collecting the waste plastic sheet by each material. The preprocessing function corrects the spectrum using the following formula: [Equation 3] where S org (n, w) is the spectrum detected by the reflection spectrum detection function, W ref (n, w) is the spectrum for the first correction, D ref (n, w) is the spectrum for the second correction, S cor (n, w) is the spectrum after correction.

Images