Sorting method and sorting device for waste, manufacturing method for cement clinker, and manufacturing facility for cement clinker

Abstract

To provide a sorting method and a sorting device for waste capable of sufficiently stably continuing an operation of a combustion apparatus when using the waste as a heat source.SOLUTION: A sorting method is for waste for fuel including plastic. The waste sorting method includes: a calibration curve preparation step of preparing a calibration curve for obtaining prediction information Y on a water content of the waste on the basis of a feature amount x obtained from electromagnetic wave absorption information on each of a plurality of samples including the plastic and having different water contents and an actual measurement value y' of the water content of each of the plurality of samples; a prediction step of deriving the prediction information Y on the water content of the waste from the feature amount x obtained from the electromagnetic wave absorption information on the waste and the calibration curve; and a sorting step of sorting the waste into a plurality of waste groups including first sorted waste and second sorted waste where average value of the water content of the second sorted waste is lower than that of the first sorted waste on the basis of the prediction information Y.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a method and apparatus for sorting waste, and a method and production facility for producing cement clinker.

Background Art

[0002] In order to reduce the use of fossil fuels and build a recycling-oriented society, it has been studied to utilize waste including waste plastic as a heat source such as fuel. In Patent Document 1, in a power generation system using an incinerator using waste plastic as fuel, the total calorific value of the fuel is calculated using image analysis, and the speed of the conveyor is controlled to make the calorific value of the fuel input into the combustion inlet constant, and a technique for operating the incinerator stably has been proposed.

[0003] In Patent Document 2, a technique for calculating the combustion calorific value of a plastic mixture based on the near-infrared absorption spectrum of the plastic mixture composed of a plurality of types of plastics and an evaluation function predetermined based on the absorption spectra of a plurality of reference samples with known combustion calories has been proposed. In Patent Document 3, moisture measuring means for measuring the moisture content contained in the waste is provided, the calorific value of the waste is calculated based on the product of the bulk specific gravity and the moisture content of the waste, and an attempt has been made to increase or decrease the feeding speed of the waste according to the calorific value. In Patent Document 4, a moisture meter for measuring the moisture content of the waste is provided in the line for supplying the waste to the incinerator, and a technique for predicting the combustion calorific value of the waste based on the moisture content has been proposed.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

SUMMARY OF THE INVENTION

PROBLEMS TO BE SOLVED BY THE INVENTION

[0005] Waste may be exposed to wind and rain during storage or transportation. Since waste includes hygroscopic substances such as paper, wood chips, and cloth, it may contain relatively high moisture in some cases. When such waste is used as a heat source, there is a concern that not only the calorific value fluctuates but also the combustion becomes unstable due to moisture. It is expected that the influence will be greater as the proportion of waste in the heat source increases. As a method for continuously measuring the moisture content of such waste, it is conceivable to use a commercially available non-contact moisture meter presented in Patent Document 4. However, in the case of waste containing plastic, it is expected to be difficult to ensure sufficient prediction accuracy when predicting calories from the measured moisture content value.

[0006] The present invention provides a waste separation method and a separation device capable of continuously operating a combustion device sufficiently stably when waste is used as a heat source. The present invention also provides a method for producing cement clinker and a production facility for cement clinker capable of continuously producing cement clinker sufficiently stably even when waste is used as a heat source.

MEANS FOR SOLVING THE PROBLEMS

[0007] One aspect of the present invention is a method for separating waste for fuel containing plastic, A method for separating waste for fuel containing plastic, Based on the feature amount x obtained from the electromagnetic wave absorption information of each of a plurality of samples containing plastic and having different moisture contents, and the actually measured value y' of the moisture content of each of the plurality of samples, a calibration curve creating step for creating a calibration curve for obtaining prediction information Y of the moisture content of the waste, A prediction step of deriving prediction information Y of the moisture content of the waste from the feature amount x obtained from the electromagnetic wave absorption information of the waste and the calibration curve, Based on the prediction information Y, a sorting step of sorting the waste into a plurality of waste groups including first sorted waste and second sorted waste having an average moisture content lower than that of the first sorted waste is provided.

[0008] In the above sorting method, a calibration curve is created based on the electromagnetic wave absorption information, and the prediction information Y of the moisture content is obtained from the calibration curve and the feature quantity x obtained from the electromagnetic wave absorption information of the waste to be sorted. Therefore, the moisture content of the waste to be sorted can be predicted with high accuracy. And it has a sorting step of sorting the waste based on the prediction information Y of the moisture content predicted with high accuracy. For this reason, the waste will be sorted into a plurality of waste groups according to the moisture content. Since the waste is sorted according to the moisture content that affects the operation of the combustion device, for example, if the first sorted waste and the second sorted waste are used in separate combustion devices, the operation of each combustion device can be continued sufficiently stably. Also, for example, if the waste group with a low moisture content is used as a heat source in the combustion device as it is, and the waste with a high moisture content is used as a heat source in the combustion device after being dried, the operation of the combustion device can be continued efficiently and sufficiently stably.

[0009] One aspect of the present invention is a method for sorting waste for fuel containing plastic, a calibration curve creation step of creating a plurality of calibration curves for obtaining the prediction information Y of the moisture of the waste based on the feature quantity x obtained from the electromagnetic wave absorption information of each of a plurality of samples containing plastic and having different moisture contents and compositions other than moisture, and the measured value y' of the moisture content of each of the plurality of samples; a prediction step of deriving the prediction information Y of the moisture content of the waste from the feature quantity x obtained from the electromagnetic wave absorption information of the waste and at least one of the plurality of calibration curves; Based on the prediction information Y, a sorting step of sorting the waste into a plurality of waste groups including first sorted waste and second sorted waste having an average moisture content lower than that of the first sorted waste is provided.

[0010] In the above separation method, a plurality of calibration curves are created based on the electromagnetic wave absorption information, and prediction information Y of the moisture content is obtained from at least one of the plurality of calibration curves and the feature quantity x obtained from the electromagnetic wave absorption information of the waste to be separated. Therefore, the moisture content of the waste to be separated can be predicted with high accuracy. And there is a separation step of separating based on the prediction information Y of the moisture content predicted with high accuracy. For this reason, the waste will be separated into a plurality of waste groups according to the moisture content. Since the waste is separated according to the moisture content that affects the operation of the combustion device, for example, if the first separated waste and the second separated waste are used in separate combustion devices, the operation of each combustion device can be continued sufficiently stably. Further, for example, if the second separated waste with a low moisture content is used as a heat source in the combustion device as it is, and the first separated waste with a high moisture content is used as a heat source in the combustion device after being dried, the operation of the combustion device can be continued efficiently and sufficiently stably.

[0011] One aspect of the present invention is a method for manufacturing cement clinker, which has a heating step of heating cement raw materials with a combustion device in a manufacturing facility for cement clinker, and uses at least one of the plurality of waste groups separated by any of the above separation methods as fuel for the combustion device. A method for manufacturing cement clinker is provided.

[0012] In the above method for manufacturing cement clinker, at least one of the plurality of waste groups separated by any of the above separation methods is used as a heat source for the combustion device. Since the plurality of waste groups are separated based on the prediction information Y of the moisture content predicted with high accuracy, if such a waste group is used as fuel, the operation fluctuation of the combustion device can be suppressed. Therefore, in the above method for manufacturing cement clinker, the manufacturing of cement clinker can be continued sufficiently stably.

[0013] One aspect of the present invention is a separation device for waste for fuel containing plastic, A calibration curve creation unit that creates a calibration curve for obtaining prediction information Y on the moisture content of the waste using a feature amount x obtained from the electromagnetic wave absorption information of each of a plurality of samples containing plastic and having different moisture contents, and the measured value y' of the moisture content of each of the plurality of samples; An information acquisition unit that acquires the electromagnetic wave absorption information of the waste; An information processing unit that obtains prediction information Y on the moisture content of the waste from the feature amount x obtained from the electromagnetic wave absorption information of the waste and the calibration curve created by the calibration curve creation unit; A sorting unit that sorts the waste into a plurality of waste groups including a first sorted waste and a second sorted waste having an average moisture content lower than that of the first sorted waste based on the prediction information Y. A waste sorting device is provided.

[0014] In the above sorting device, the information processing unit creates a calibration curve based on the electromagnetic wave absorption information, and obtains prediction information Y on the moisture content of the waste from the calibration curve and the feature amount x. Therefore, the moisture content of the waste to be sorted can be predicted with high accuracy. In the sorting unit, the waste is sorted into a plurality of waste groups based on the prediction information Y on the moisture content predicted with high accuracy. Since the waste is sorted according to the moisture content that affects the operation of the combustion device, for example, if the first sorted waste and the second sorted waste are used in separate combustion devices, the operation of each combustion device can be continued sufficiently stably. Further, for example, if the second sorted waste with a low moisture content is used as a heat source in the combustion device as it is, and the first sorted waste with a high moisture content is used as a heat source in the combustion device after being dried, the operation of the combustion device can be continued efficiently and sufficiently stably.

[0015] One aspect of the present invention is a waste sorting device for fuel containing plastic, A calibration curve creation unit that creates a plurality of calibration curves for obtaining prediction information on the moisture content of the waste using a feature amount x obtained from the electromagnetic wave absorption information of each of a plurality of samples containing plastic and having different moisture contents, and the measured value y' of the moisture content of each of the plurality of samples; An information acquisition unit that acquires the electromagnetic wave absorption information of the waste; An information processing unit that obtains prediction information Y on the moisture content of the waste from a feature amount x obtained from the electromagnetic wave absorption information of the waste and at least one of the plurality of calibration curves created by the calibration curve creation unit; A sorting unit that sorts the waste into a plurality of waste groups including first sorted waste and second sorted waste having an average moisture content lower than that of the first sorted waste based on the prediction information Y. Provided is a waste sorting device comprising:

[0016] In the above sorting device, the information processing unit creates a plurality of calibration curves based on electromagnetic wave absorption information, and obtains prediction information Y on the moisture content from at least one of the plurality of calibration curves and the feature amount x. Therefore, the moisture content of the waste to be sorted can be predicted with high accuracy. In the sorting unit, the waste is sorted into a plurality of waste groups based on the prediction information Y on the moisture content of the waste predicted with high accuracy. Since the waste is sorted according to the moisture content that affects the operation of the combustion device, for example, if the first sorted waste and the second sorted waste are used in separate combustion devices, the operation of each combustion device can be continued sufficiently stably. Further, for example, if the second sorted waste with a low moisture content is used as a heat source in the combustion device as it is, and the first sorted waste with a high moisture content is used as a heat source in the combustion device after being dried, the operation of the combustion device can be continued efficiently and sufficiently stably.

[0017] One aspect of the present invention provides a cement clinker manufacturing facility including any of the above sorting devices and a combustion device, and using at least one of a plurality of waste groups as fuel for the combustion device.

[0018] In the above cement clinker manufacturing facility, at least one of the plurality of waste groups sorted by any of the above sorting devices is used as fuel for the combustion device. Since the plurality of waste groups are sorted based on the prediction information Y on the moisture content predicted with high accuracy, the temperature variation of the combustion device can be suppressed by using such waste groups as a heat source. Therefore, in the above cement clinker manufacturing facility, the production of cement clinker can be continued sufficiently stably.

Advantages of the Invention

[0019] According to the present invention, when using waste as a heat source, it is possible to provide a waste sorting method and a sorting device that can continuously operate the combustion device with sufficient stability. Further, the present invention can provide a method for manufacturing cement clinker and a manufacturing facility for cement clinker that can continuously manufacture cement clinker with sufficient stability even when using waste as a heat source.

Brief Description of the Drawings

[0020]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Mode for Carrying Out the Invention

[0021] Hereinafter, an embodiment of the present invention will be described with reference to the drawings as appropriate. However, the following embodiments are examples for explaining the present invention and are not intended to limit the present invention to the following contents. In the description, the same reference numerals are used for the same elements or elements having the same function, and redundant descriptions may be omitted as appropriate. Also, the positional relationships such as up, down, left, and right are based on the positional relationships shown in the drawings unless otherwise specified. Furthermore, the dimensional ratios of each element are not limited to the ratios shown in the drawings. Note that a numerical range indicated by "~" between the upper limit value and the lower limit value is a numerical range including the upper limit value and the lower limit value.

[0022] A waste separation method according to an embodiment is a method for separating waste for fuel containing plastic, including a calibration curve creation step of creating a calibration curve for predicting the moisture content of waste based on a feature amount x obtained from the electromagnetic wave absorption information of each of a plurality of samples containing plastic and having different moisture contents, and the measured value y' of the moisture content of each of the plurality of samples; a prediction step of deriving prediction information Y of the moisture content from the feature amount x obtained from the electromagnetic wave absorption information of the waste to be separated and the calibration curve; and a separation step of separating the waste into a plurality of waste groups including a first separated waste and a second separated waste having an average moisture content lower than that of the first separated waste based on the prediction information Y.

[0023] The plurality of samples used for creating the calibration curve may be samples obtained by sampling a part of the waste to be separated, or may be prepared by separately collecting and mixing each component contained in the waste. The moisture contents of the plurality of samples are different from each other. Also, the moisture content varies as in the case of waste, and it is preferably in the same numerical range. Thereby, the moisture content can be predicted with high accuracy. From the same viewpoint, the sample preferably has the same composition as the waste. However, since the prediction accuracy of the moisture content may be affected by the composition other than moisture, when the composition other than moisture of the waste to be separated varies, it is preferable to use a plurality of samples having different compositions other than moisture. The measured value y' of the moisture content of each sample may be a value measured by a known method. Examples of the measurement method include JIS Z 7302-2:1999 "Waste Solidification Fuel - Part 3: Moisture Test Method", etc. On the other hand, the moisture content of the waste to be separated may be unknown.

[0024] The waste only needs to contain things that have become unnecessary and are discarded. Examples of the contained components include garbage, bulky waste, ash, sludge, manure, waste oil, and other contaminants. More specifically, examples include plastic, paper, wood chips, stone, glass, metal, cloth, and leather. The plurality of samples used for creating the calibration curve may also contain the same components. Examples of the plastic include polyolefin plastics and polyester plastics. The moisture content of the waste and the samples may be, for example, 0.1 to 50% by weight, or may be 0.5 to 40% by weight.

[0025] The electromagnetic wave absorption information may be information correlated with the moisture content. For example, since water has absorption peaks near wavelengths of 980 nm, 1180 nm, and 1450 nm, it may be near-infrared absorption information. The electromagnetic wave absorption information may be a near-infrared absorption spectrum including a wavelength range of 900 to 1700 nm. Among the above-mentioned absorption peaks of water, the absorption peak at 1450 nm is the largest. Therefore, it is more preferable that the electromagnetic wave absorption information is a near-infrared absorption spectrum including a wavelength range of 1440 to 1460 nm. Also, the electromagnetic wave absorption information may be a microwave absorption spectrum. The feature quantity x may be the absorbance of an absorption peak near a wavelength of 980 nm, 1180 nm, or 1450 nm, or may be the absorbance of a microwave absorption peak.

[0026] The feature quantity x may be the intensity of the absorption peak. The intensity may be the peak height or the peak area. It may also be the integrated intensity in a predetermined wavelength range. The feature quantity x may be calculated by performing any of addition, subtraction, multiplication, or division using a plurality of electromagnetic wave absorption information including at least one electromagnetic wave absorption information capable of detecting water. For example, it may be any of the sum, difference, product, or quotient of the electromagnetic wave absorption information a and the electromagnetic wave absorption information b obtained in different wavelength regions, or may be the average value of the electromagnetic wave absorption information a and b. "Any of addition, subtraction, multiplication, or division" includes cases where two or more operations such as addition and multiplication are performed.

[0027] The feature amount x may be a value obtained from a near-infrared absorption spectrum in a predetermined wavelength range including a water absorption peak. For example, it may be the maximum value of the absorbance of the near-infrared absorption spectrum in the predetermined wavelength range. In this specification, the maximum value of the absorbance means that the electromagnetic wave is most absorbed at that wavelength. In this specification, "near-infrared rays" are electromagnetic waves in the wavelength range of 780 to 2500 nm.

[0028] Since the waste may be stored outdoors or may be sprinkled with water during handling for fire prevention, the moisture content varies greatly. When the moisture content varies, the temperature in the combustion device may vary or a fire may occur. In the above sorting method, by sorting waste with different moisture contents into a plurality based on the prediction information Y of the moisture content to obtain a plurality of waste groups (sorted waste) with different moisture contents, the operation fluctuations of the combustion device can be suppressed.

[0029] The prediction information Y of the moisture content of the waste may be the predicted value y of the moisture content itself or a numerical value related to the predicted value y. For example, it may be a calculated value obtained by multiplying the predicted value of the moisture content by a coefficient.

[0030] FIG. 1 shows an example of a separation device for implementing the separation method of the present embodiment. The separation device 100 in FIG. 1 is a separation device for waste containing plastic, and includes a calibration curve creation unit 60 for creating a calibration curve for predicting the moisture content using the feature amount x obtained from the electromagnetic wave absorption information of each of a plurality of samples containing plastic and having different moisture contents, and the measured value y' of the moisture content of each of the plurality of samples, a crushing unit 10 for crushing the waste containing plastic, a conveying unit 12 for conveying the crushed waste, an information acquisition unit 15 having a light source 14 for irradiating the waste with electromagnetic waves and a detector 16 for detecting the electromagnetic wave absorption information of the waste while being conveyed by the conveying unit 12, an information processing unit 20 for deriving the feature amount x of the waste from the electromagnetic wave absorption information of the waste and outputting prediction information Y of the moisture content of the waste from the feature amount x of the waste and the calibration curve, a control unit 30 for outputting a control signal based on the prediction information Y output from the information processing unit 20, and a separation unit 40 for separating the waste into a plurality of separated wastes including at least a first separated waste and a second separated waste having an average moisture content lower than that of the first separated waste by the control signal from the control unit 30. Note that it is not essential to perform the above separation method using this separation device, and it may be performed using another separation device. Further, the above separation device may be configured to separate waste by a separation method different from the above separation method.

[0031] The waste separation method may include a crushing step of crushing the waste to be separated. The crushing step may be performed in the crushing unit 10. Further, the prediction step can also be subdivided into an information acquisition step of acquiring the electromagnetic wave absorption information of the waste, and a moisture content prediction step of deriving the feature amount x from the electromagnetic wave absorption information of the waste and predicting the moisture content of the waste from the feature amount x and the calibration curve created in the calibration curve creation step. An example of the separation method may include, as shown in FIG. 2, a calibration curve creation step S1 of creating a calibration curve for predicting the moisture content of the waste based on the feature amount x obtained from the near-infrared absorption information of a plurality of samples with known moisture contents and the measured value y' of the moisture content of the samples, a crushing step S2 of crushing the waste, an information acquisition step S3, a moisture content prediction step S4, and a separation step S5.

[0032] In the calibration curve creation step S1, a calibration curve is created based on the characteristic quantity x of a plurality of samples containing plastic and having different moisture contents, and the actually measured value y' of the moisture content of each sample. The characteristic quantity x can be obtained from electromagnetic wave absorption information correlated with the moisture content.

[0033] The characteristic quantity x may be obtained by preprocessing the electromagnetic wave absorption information. The preprocessing may include at least one selected from the group consisting of trimming, smoothing, second-order differentiation, and normalization by the maximum value. Thereby, the prediction accuracy of the moisture content of the waste can be made sufficiently high. For example, trimming or smoothing may be performed to reduce the influence of noise from the light source of the electromagnetic wave. Also, by performing second-order differentiation or normalization by the maximum value, the influence of the spectrum shifting up and down due to the intensity of the reflected light can be reduced.

[0034] The characteristic quantity x may be the characteristic quantity x1 derived from the electromagnetic wave absorption information specific to water and the electromagnetic wave absorption information specific to components other than water (first components) contained in the waste. The electromagnetic wave absorption information specific to water may be the absorbance of the near-infrared absorption peak around a wavelength of 980 nm, 1180 nm, or 1450 nm, or may be the absorbance of the microwave absorption peak by water. The components other than water may be polyolefin-based plastics or polyester-based plastics that are often contained in the waste. Examples of polyolefin-based plastics include polyethylene, polypropylene, and the like. Examples of polyester-based plastics include polyethylene terephthalate, polybutylene terephthalate, polyethylene naphthalate, polytrimethylene terephthalate, polycyclohexylene dimethylene terephthalate, and the like.

[0035] For example, polyethylene has an absorption spectrum with characteristic peaks near wavelengths of 1200 nm, 1400 nm, 1725 nm, 1775 nm, and 1850 nm. Thus, in the near-infrared region, peaks of components other than water contained in the waste are often detected. Therefore, by adopting the feature quantity x1 derived using not only the electromagnetic wave absorption information specific to water but also the electromagnetic wave absorption information specific to components other than water contained in the waste, the prediction accuracy of the moisture content of the waste 11 can be further improved. For example, if the feature quantity x1 is the ratio of the absorbance (I w ) of the absorption peak derived from water to the absorbance (I1) of the absorption peak derived from components other than water (the first component) (for example, I1 / I w , or I w / I1), or something derived from a calculation formula including this, the influence of the absorption peak derived from the first component can be reduced, and the prediction accuracy of the moisture content can be further improved.

[0036] When creating a plurality of calibration curves for each main component of the sample, a part of the plurality of calibration curves may be created based on the feature quantity x1 and the measured value y’ of the moisture content, and the other part of the plurality of calibration curves may be created based on the feature quantity x and the measured value y’ of the moisture content. That is, the plurality of calibration curves may use different feature quantities from each other.

[0037] The calibration curve creation step S1 may be performed using the calibration curve creation unit 60 in FIG. 1. In the calibration curve creation unit 60, a calibration curve is created based on the feature quantity x or the feature quantity x1 (hereinafter, may also be simply referred to as “feature quantity x”) obtained from the electromagnetic wave absorption information of a plurality of samples and the measured value y’ of the moisture content of each sample. The calibration curve creation unit 60 may be equipped with the same devices as the crushing unit 10, the conveying unit 12, and the information acquisition unit 15 shown in FIG. 1. When using a sampled waste as the sample, a calibration curve may be created using the electromagnetic wave absorption information acquired by the information acquisition unit 15 and the measured value y’ of the moisture content of the sampled sample.

[0038] To create a calibration curve for predicting the moisture content of waste using the feature quantity x obtained from the electromagnetic wave absorption information of a sample and the measured value y' of the moisture content of the sample, various linear or non-linear regression analysis methods can be used. For example, as linear regression methods, there are least squares approximation for an arbitrary polynomial model corresponding to the feature quantity x, ridge regression, lasso regression, and elastic net, etc. As non-linear regression methods, there are k-nearest neighbor method and decision tree method, etc. Among these, from the viewpoint of reducing the computational load when calculating the calibration curve and the ease of interpretation by graph display, it is preferable to use linear regression by the least squares method. In this case, the calibration curve may be obtained by simple regression analysis with the feature quantity x as the explanatory variable and the measured value y' as the objective variable. By using a sample similar in composition to the waste 11 to be sorted, the prediction accuracy of the moisture content of the waste can be increased.

[0039] From the viewpoint of sufficiently increasing the prediction accuracy of the moisture content of the waste 11, it is better to have more combinations of the feature quantity x and the measured value y' of the sample. In the x-y' diagram (graph) with the feature quantity x on the horizontal axis and the measured value y' on the vertical axis, the number of plots of the coordinates (x, y') is preferably 10 or more, more preferably 30 or more, and even more preferably 50 or more. The coefficient of determination (R 2 ) of the calibration curve obtained by regression analysis is preferably 0.4 or more, more preferably 0.5 or more, and even more preferably 0.6 or more. The larger the coefficient of determination, the higher the prediction accuracy of the moisture content of the waste can be. From the same viewpoint, the RMSE (Root Mean Square Error) is preferably 2.8 or less, more preferably 2.5 or less, and even more preferably 2.0 or less. From the same viewpoint, the MAE (Mean Absolute Error) is preferably 4.2 or less, more preferably 3.8 or less, and even more preferably 3.5 or less.

[0040] In the calibration curve creation step S1, the calibration curve can be created in this way. The created calibration curve may be stored in the storage 76 of the information processing unit 20.

[0041] In the crushing process S2, as shown in FIG. 1, the waste is crushed, for example, by the crushing unit 10. Examples of the crusher provided in the crushing unit 10 include a mill, a shredder, a crusher, and the like. Through the crushing process S2, waste 11 containing plastic with a particle size of, for example, 35 mm or less is obtained. By irradiating waste 11 of such a size with electromagnetic waves, the moisture content can be predicted with a sufficiently high accuracy. This is because when the size is large, the moisture and contained components of the waste may vary greatly between the inside and the surface.

[0042] From the perspective of obtaining more accurate prediction results, the particle size of the waste 11 (sample) may be 30 mm or less, or may be 25 mm or less. The particle size of the waste 11 (sample) can be determined as the diameter of the circumscribed circle that circumscribes the fuel depicted in the two-dimensional image. Upstream of the crushing unit 10, a coarse crusher, a pneumatic separator, a magnetic separator, etc. may be installed to remove metals, etc. from the waste raw material. Performing the crushing process is not essential, and the received waste may be directly conveyed by the conveying unit 12 and the information acquisition process S3 may be performed by the information acquisition unit 15.

[0043] As shown in FIG. 3, the information acquisition unit 15 that performs the information acquisition process S3 includes a light source 14 that irradiates the waste 11 crushed by the crushing unit 10 with electromagnetic waves and a detector 16 that detects near-infrared rays. The information acquisition unit 15 may acquire the absorption spectrum (electromagnetic wave absorption information) of the waste containing plastic. Examples of the light source 14 include those capable of irradiating electromagnetic waves including near-infrared rays and / or microwaves. For example, a halogen lamp, an infrared LED lamp, an incandescent bulb, sunlight, a microwave generator, etc. may be mentioned.

[0044] The detector 16 is not particularly limited as long as it can detect electromagnetic wave absorption information used for deriving the feature quantity x. For example, an infrared sensor, an infrared camera, a multi-band camera, a hyperspectral camera, a microwave meter, etc. can be mentioned. Among these, since the near-infrared spectrum can be measured with high wavelength resolution, a hyperspectral camera can be preferably used. As the hyperspectral camera, for example, those manufactured by RESONON or HySpex can be used. By including the near-infrared absorption spectrum in the electromagnetic wave absorption information, a feature quantity x highly correlated with the measured value y' of the moisture content of the waste 11 can be obtained.

[0045] The wavelength range of the near-infrared absorption spectrum acquired by the information acquisition unit 15 may be 800 to 2500 nm, or may be 900 to 1700 nm. Thereby, a feature quantity x highly correlated with the measured value y' of the moisture content of the waste 11 can be obtained, and the prediction accuracy of the moisture content of the waste can be improved. If necessary, a band-pass filter, a low-pass filter, a high-pass filter, etc. that transmit only electromagnetic waves of a specific wavelength may be attached to the detector 16 to adjust the wavelength of the electromagnetic waves incident on the detector 16.

[0046] In the moisture content prediction step S4, a feature quantity x used to obtain the prediction information Y of the moisture content of the waste 11 is obtained from the electromagnetic wave absorption information of the waste obtained in the information acquisition step S3. The feature quantity x of the waste 11 can be derived in the same manner as the feature quantity x of the sample in the calibration curve creation step S1. When deriving the feature quantity x of the waste 11, the electromagnetic wave absorption information of the waste 11 may be pre-processed in the same manner as in the calibration curve creation step S1. By performing the same pre-processing as in the calibration curve creation step S1, the prediction accuracy of the moisture content of the waste 11 can be made sufficiently high. The derivation of the feature quantity x of the waste 11 may be performed by the information processing unit 20 shown in FIG. 3. In the moisture content prediction step S4, the feature quantity x is substituted into the calibration curve obtained in the calibration curve creation step S1 to obtain the prediction information Y of the moisture content of the waste 11.

[0047] In a modified example, as shown in FIG. 4, an identification step S35 for identifying the composition of the waste may be provided between the information acquisition step S3 and the moisture content prediction step S4. The calibration curve may vary according to the ratio of components other than water (first component) contained in the waste 11. For this reason, in the calibration curve creation step S1, a plurality of calibration curves may be created for each main component contained in the sample. The main component refers to the component having the highest ratio among the components contained in the sample. For example, a first calibration curve when a polyolefin-based plastic is the main component, a second calibration curve when a polyester-based plastic is the main component, and a third calibration curve when other components are the main component may be created.

[0048] In an example of the identification step S35, a calibration curve is selected along the flowchart shown in FIG. 5. In the identification step S35, the main component contained in the waste 11 is identified based on the electromagnetic wave absorption information obtained in the information acquisition step S3. For the identification of the main component, the same type of electromagnetic wave absorption information as that used for deriving the feature amount x in the information acquisition step S3 may be used. For example, the feature amount x may be derived and the main component may be identified using the wavelength range including the peak of water and the peak of the main component as the absorption spectrum of near-infrared rays. In the flowchart of FIG. 5, first, it is determined whether or not the main component of the waste 11 is a polyolefin-based plastic (S351). If it is determined that the main component is a polyolefin-based plastic, the first calibration curve is selected (S352). If it is determined that the main component of the waste is not a polyolefin-based plastic, it is determined whether or not the main component is a polyester-based plastic (S353). If it is determined that the main component is a polyester-based plastic, the second calibration curve is selected (S354). If the main component is a component other than a polyolefin-based plastic and a polyester-based plastic, the third calibration curve is selected (S355). In this way, in the identification step S35, one calibration curve is selected according to the identification result of the composition of the waste 11.

[0049] The number of calibration curves before selection is not particularly limited, and the components contained in the waste 11 (sample) may be further subdivided, and four or more calibration curves may be used. Also, the identification of the main components may be performed by means other than using the same electromagnetic wave absorption information as the feature quantity x. Also, it is not always essential to identify the substance. For example, as an identification result, one calibration curve may be selected according to the position of the absorption peak of the near-infrared absorption spectrum. The plurality of calibration curves may use different feature quantities x. In this case, prediction information Y on the moisture content of the waste 11 may be derived using different types of feature quantities x (for example, feature quantity x1) for each calibration curve.

[0050] When having the identification step S35, in the moisture content prediction step S4, prediction information Y on the moisture content of the waste 11 is derived from the feature quantity x and the calibration curve selected in the identification step S35. The derivation of the feature quantity x, the derivation of the identification result, and the derivation of the prediction information Y on the moisture content of the waste 11 may be performed by the information processing unit 20 shown in FIGS. 1 and 3. The information processing unit 20 can be configured as a normal computer system. An example of the hardware configuration of the information processing unit 20 has at least one processor 72, a memory 74, a storage 76, and an input / output port 78, as shown in FIG. 3. In the storage 76, computer software (for example, analysis software) for realizing each function may be recorded.

[0051] The information processing unit 20 may be configured such that, by loading such computer software onto hardware such as the processor 72 and the memory 74, the input / output port 78 and the input / output device 82 operate under the control of the processor 72. The storage 76 may be a computer-readable recording medium such as a hard disk, a non-volatile semiconductor memory, a magnetic disk, or an optical disk. In the storage 76, the actually measured value y' of the moisture content of the sample and the calibration curve obtained in the calibration curve creation step S1 may be stored.

[0052] Memory 74 temporarily stores programs, data, operation results of the processor 72, etc. loaded from the storage 76. The processor 72 cooperates with the memory 74 to perform preprocessing of electromagnetic wave absorption information such as the electromagnetic wave absorption spectrum acquired by the information acquisition unit 15, derivation of the feature quantity x, identification of the waste 11, selection of a calibration curve according to the identification result, calculation of the prediction information Y of the moisture content of the waste 11 from the feature quantity x of the waste and the selected calibration curve, etc. A program may be executed. When a plurality of calibration curves are used for calculating the prediction information Y of the moisture content, the average value of the predicted values y of the moisture content derived from the plurality of calibration curves may be used as the prediction information Y of the moisture of the waste 11. The input / output port 78 performs input / output of electrical signals with the control unit 30, the input / output device 82, etc. according to commands from the processor 72.

[0053] The calibration curve creation unit 60 may be configured as a normal computer system similar to the information processing unit 20. However, one computer system may have the functions of both the calibration curve creation unit 60 and the information processing unit 20. The calibration curve creation unit 60 derives a correlation formula for each main component contained in the sample from the feature quantity x and the measured value y', and stores it in the storage. When the main component of the waste 11 is identified by the electromagnetic wave absorption information from the information acquisition unit 15, at least one correlation formula is selected from the plurality of correlation formulas stored in the storage, and the prediction information Y of the moisture content of the waste 11 is derived from the correlation formula and the feature quantity x obtained from the waste 11.

[0054] As shown in FIGS. 1 and 3, the information processing unit 20 outputs the prediction information Y of the moisture content of the waste 11 to the control unit 30. The control unit 30 outputs a control signal for operating the dampers 41, 42 in FIG. 1 based on the input value from the information processing unit 20. The control unit 30 may have, for example, a processor, a memory, a storage, and an input / output port, etc. similar to the information processing unit 20. The control unit 30 only needs to have a configuration capable of deriving a control signal and outputting the control signal. It should be noted that it is not essential to configure the information processing unit 20 and the control unit 30 as separate hardware, and they may be configured as one hardware.

[0055] The separation process S5 can be performed in the separation unit 40 of FIG. 1. The separation unit 40 includes dampers 41 and 42 connected to the conveying unit 12, a tank 51 for storing the first separated waste, and a tank 52 for storing the second separated waste. In this example, the waste 11 is separated into the first separated waste and the second separated waste. Note that the number of separated waste (the number of waste groups) to be separated is not particularly limited, and the waste 11 may be separated into three or more waste groups according to the predicted moisture content information Y.

[0056] The average value of the predicted value y of the moisture content in the first separated waste stored in the tank 51 is, for example, a weight % or more. The average value of the predicted value y of the moisture content in the second separated waste stored in the tank 52 is, for example, less than a weight %. However, waste with a predicted value y of the moisture content less than a weight % may be mixed into the tank 51. Waste with a predicted value y of the moisture content of a mass % or more may be mixed into the tank 52. It is sufficient that the arithmetic mean value of the predicted value y of the moisture content of the first separated waste is a weight % or more, and the arithmetic mean value of the predicted value y of the moisture content of the second separated waste is less than a weight %. The average value of the predicted value y is the arithmetic mean value of a plurality of predicted values y. a weight % may be, for example, 5 weight %, 10 weight %, or 15 weight %.

[0057] As shown in FIG. 1, the tanks 51 and 52 are respectively connected to outlet paths 51a and 52a. The first separated waste and the second separated waste once stored in the tanks 51 and 52 may flow through the outlet paths 51a and 52a and be conveyed to the first combustion device and the second combustion device downstream. The conveyance may be performed by a belt conveyor or gas (air) pressure feeding, or may be performed by gravity fall. When the separation device 100 and the combustion device are in separate locations, each separated waste derived from the tank may be transported to the tank at the location where the combustion device is installed using a truck. Note that not all of the separated waste needs to be used as waste such as an incinerator. One type of the plurality of separated waste, or a part of each separated waste, may be used for other purposes.

[0058] In an example of the separation process S5, the waste is separated into two waste groups by the separator 40 according to the flowchart shown in FIG. 6. Before starting the separation, the dumpers 41 and 42 are closed (S11). When the predicted water content value y is a weight % or more (for example, 15 weight % or more) (S12), the dumper 41 is opened (S13) and introduced into the tank 51 as the first separated waste (S14). When the predicted water content value y is less than a weight % (for example, less than 15 weight %), the dumper 42 is opened (S15) and introduced into the tank 52 as the second separated waste (S16). In this way, the waste is separated into two separated wastes according to the predicted water content value y.

[0059] After the plurality of separated wastes are once stored in the tanks 51 and 52, they are burned as fuel in the first combustion device and the second combustion device. The first combustion device and the second combustion device may be an incinerator, a boiler, a preheating furnace in a cement clinker manufacturing facility, or a cement kiln. It is not essential to introduce the first separated waste and the second separated waste into separate combustion devices. For example, they may be introduced from separate locations of the same incinerator or the introduction amounts may be adjusted individually. In this way, while continuously operating the combustion device such as an incinerator stably, the waste can be effectively used as fuel to reduce the amount of fossil fuel used. In addition, in the combustion device (the first combustion device and the second combustion device), the waste may be burned with a burner provided in the combustion device. Even when burned with a burner, the waste is still used in the combustion device.

[0060] A method for manufacturing cement clinker according to an embodiment includes a heating step of preheating a cement raw material in a preheating furnace and firing it in a cement kiln. In the heating step, at least one of the separated fuels (the first separated waste, the second separated waste) including the waste separated into a plurality by the above-described separation method is used to heat at least one of the combustion devices of the cement kiln and the preheating furnace.

[0061] The above-described method for manufacturing cement clinker may be carried out using the cement clinker manufacturing facility 300 shown in FIG. 7. However, it is not essential to carry out the above manufacturing method using this cement clinker manufacturing facility 300, and it may be carried out using another cement clinker manufacturing facility. Further, the cement clinker manufacturing facility 300 may be configured to manufacture cement clinker by a manufacturing method different from the above-described method for manufacturing cement clinker.

[0062] The cement clinker manufacturing facility 300 shown in FIG. 7 includes a waste sorting device 100 and a cement clinker manufacturing device 200. The cement clinker manufacturing device 200 includes an inlet 201 into which a cement raw material is introduced, four cyclones C1, C2, C3, C4 (preheaters) for preheating the cement raw material introduced from the inlet 201, a calciner 230 for calcining the cement raw material, a cement kiln 240 for firing the preheated and calcined cement raw material to produce cement clinker, and a clinker cooler 250 for cooling the cement clinker produced in the cement kiln 240 and discharging the cooled cement clinker.

[0063] The inlet 201 is provided at the connection portion between the cyclone C1 and the cyclone C2. The cement raw material introduced from the inlet 201 may contain, for example, at least one selected from the group consisting of incineration ash, coal ash, limestone, an iron source, slag, and waste. The waste introduced from the inlet 201 is a cement raw material and does not correspond to fuel use. The cement raw material introduced from the inlet 201 is heated while flowing through the cyclone C1, the cyclone C2, the cyclone C3, the riser duct 234, the calciner 230, and the cyclone C4, and is introduced into the kiln end 242 of the cement kiln 240.

[0064] The kiln end 242 of the cement kiln 240 and the precalciner 230 are connected by a rising duct 234. An extraction pipe 236 for extracting the kiln exhaust gas in the rising duct 234 is connected to the rising duct 234. Downstream of the extraction pipe 236, a chlorine bypass section 237 having a cooler, a bag filter, etc. is installed, and dust contained in the extracted gas (kiln exhaust gas) extracted by the extraction pipe 236 is recovered. The gas after the dust is removed in the chlorine bypass section 237 is introduced into a circulation gas line 252 connecting the clinker cooler 250 and the precalciner 230 and is introduced into the precalciner 230. By having the chlorine bypass section 237, volatile components such as chlorine-based compounds and alkalis can be reduced from within the cement clinker production apparatus 200. In a modified example, the extraction pipe 236 may be connected to the kiln end 242 or may be connected to the boundary portion between the rising duct 234 and the kiln end 242.

[0065] The temperature inside the cement kiln 240 is, for example, 1400°C to 1500°C. The burner 244 provided at one end of the cement kiln 240 is supplied with the second separated waste and fossil fuel from the tank 52. As the fossil fuel, coal such as pulverized coal, coal coke, oil coke, heavy oil, recycled oil, etc. can be used. In the present embodiment, since the second separated waste used as the fuel of the cement kiln 240 has a lower moisture content than the waste before separation, the ratio of the second separated waste to the fossil fuel can be made sufficiently high. Therefore, the amount of fossil fuel used can be reduced. Also, the second separated waste has less variation in moisture content compared to the waste before separation. For this reason, while using a sufficient amount of the second separated waste as fuel, the stable operation of the cement clinker production apparatus 200 can be continued. From the viewpoint of further stabilizing the operation of the cement clinker production apparatus 200, the average value of the moisture content (measured value) of the second separated waste is preferably less than 15% by weight. Thereby, the variation in the quality of the obtained cement clinker can be sufficiently reduced.

[0066] The operation control of the cement kiln 240 may be performed by a measuring unit M that measures the torque when rotating the cement kiln 240 or the power consumption of the rotating motor, an adjusting unit 222 that adjusts the supply amount of the fossil fuel, and a control unit 224 that outputs a control signal for adjusting the adjusting unit 222 based on the measured value of the measuring unit M. At this time, the first separated waste may be introduced into the burner 244 in the maximum amount within the range where the operation control of the cement kiln 240 is possible. In this case, the supply amount of the fossil fuel may be adjusted so that the measured value of the measuring unit M falls within the target range. Thereby, while effectively utilizing the second separated waste, the stable operation of the cement kiln 240 can be continued. However, in a modified example, the introduction amount of the second separated waste into the burner 244 may be adjusted by a flow rate adjusting unit 52V according to the predicted information Y on the moisture content of the second separated waste and / or the operation status of the cement kiln 240. Also, the second separated waste and the fossil fuel may be introduced into different burners.

[0067] The temperature in the calciner 230 is, for example, 850°C to 900°C. The first separated waste containing waste and fossil fuel are supplied to the calciner 230 from the tank 51. Although the first separated waste has a higher moisture content than the second separated waste, the calciner 230 can continue to operate stably because it has a higher allowable value of moisture content than the cement kiln 240. As the fossil fuel, pulverized coal, oil coke, heavy oil, recycled oil, etc. can be used. Further, a non-fossil fuel may be supplied to the calciner 230. As the non-fossil fuel, waste oil, waste tires, waste clay, meat and bone meal, sludge, etc. can be used.

[0068] The operation control of the pre - calciner 230 may be performed by a measurement unit T that measures the temperature of the pre - calciner 230, an adjustment unit 231 that adjusts the supply amount of fossil fuel, and a control unit 232 that outputs a control signal for adjusting the adjustment unit 231 based on the measurement value of the measurement unit T. At this time, the first separated waste may be introduced into the pre - calciner 230 in the maximum amount within the range where the operation control of the pre - calciner 230 is possible. Then, the supply amount of fossil fuel is adjusted so that the measurement value of the measurement unit T maintains the target range. Thereby, while effectively utilizing the first separated waste as fuel for the pre - calciner 230, the stable operation of the pre - calciner 230 can be continued. However, in a modified example, the introduction amount of the first separated waste into the pre - calciner 230 may be adjusted by a flow rate adjustment unit 51V according to the predicted information Y on the moisture content of the first separated waste and / or the operation status of the pre - calciner 230. The first separated waste and fossil fuel may be introduced into a burner attached to the pre - calciner 230 respectively and burned in the pre - calciner 230.

[0069] By controlling the flow rate adjustment units 51V and 52V, the first separated waste and the second separated waste may be introduced into the cement clinker manufacturing apparatus 200 at a certain flow rate ratio. Thereby, the operation of the cement clinker manufacturing apparatus 200 can be sufficiently stabilized.

[0070] The flow rate adjustment units 51V and 52V may each be composed of a control unit (not shown) that controls a valve and its opening degree. For example, a load cell, an impact line flow meter, etc. may be used to measure the supply amount, and the flow rate of each separated fuel may be adjusted by feedback control. The flow rate adjustment units 51V and 52V are not particularly limited as long as they can adjust the flow rates of the first separated waste and the second separated waste respectively.

[0071] The introduction destination of the second separated waste may be switched from the burner 244 to the pre - calciner 230 (or a burner attached to the pre - calciner 230) according to the operation status of the cement clinker manufacturing apparatus 200 and the inventory balance of the first separated waste and the second separated waste. The switching of the introduction destination of the second separated waste can be performed, for example, by operating a three - way valve 52T based on a control signal from the control unit 30.

[0072] According to the method for manufacturing cement clinker and the manufacturing equipment for cement clinker of the present embodiment, waste containing various components is separated based on the predicted information Y of the water content obtained based on the characteristic quantity x, and the separated waste obtained by the separation is used as fuel for the combustion device. Therefore, the production of cement clinker can be stably continued. Further, according to the operating state of the cement clinker manufacturing apparatus, the introduction destination of each separated waste can be changed or the introduction amount can be adjusted, enabling flexible operation adjustment. Thereby, the variation in the quality of the cement clinker can be sufficiently reduced. Furthermore, the consumption of fossil fuels can be reduced, and the effective utilization of waste can be achieved.

[0073] As described above, several embodiments of the present disclosure have been described, but the present disclosure is not limited to the above embodiments. For example, in the example of the manufacturing equipment for cement clinker in FIG. 7, a separation device 100 is provided, but the present disclosure is not limited thereto. For example, a separation device that separates waste into three or more waste groups may be provided. The description content of the waste separation device is also applicable to the waste separation method, and the description content of the waste separation method is also applicable to the waste separation device. The description content of the cement clinker manufacturing equipment is also applicable to the cement clinker manufacturing method, and the description content of the cement clinker manufacturing method is also applicable to the cement clinker manufacturing equipment.

[0074] The present disclosure includes the following contents [1] to

[20] . [1] A method for separating waste for fuel containing plastic, a calibration curve creation step of creating a calibration curve for obtaining predicted information Y of the moisture content of the waste based on a characteristic quantity x obtained from electromagnetic wave absorption information of a plurality of samples each containing plastic and having different moisture contents, and the measured value y' of the moisture content of each of the plurality of samples; a prediction step of deriving predicted information Y of the moisture content of the waste from the characteristic quantity x obtained from the electromagnetic wave absorption information of the waste and the calibration curve; A waste separation method having a separation step of separating the waste into a plurality of waste groups including first separated waste and second separated waste having an average moisture content lower than that of the first separated waste, based on the prediction information Y. [2] A method for separating waste for fuel containing plastic, A calibration curve creation step of creating a plurality of calibration curves for obtaining prediction information Y on the moisture content of the waste, based on a feature amount x obtained from electromagnetic wave absorption information of each of a plurality of samples containing plastic and having different moisture contents and compositions other than moisture, and an actually measured value y' of the moisture content of each of the plurality of samples; A prediction step of deriving prediction information Y on the moisture content of the waste from the feature amount x obtained from the electromagnetic wave absorption information of the waste and at least one of the plurality of calibration curves; A waste separation method having a separation step of separating the waste into a plurality of waste groups including first separated waste and second separated waste having an average moisture content lower than that of the first separated waste, based on the prediction information Y. [3] Having an identification step of identifying the waste according to its composition, In the separation step, the waste is separated into the plurality of waste groups based on the prediction information Y by one calibration curve selected from the plurality of calibration curves according to the identification result of the identification step. The waste separation method according to [2]. [4] The feature amount x is a feature amount x1 derived from electromagnetic wave absorption information specific to water and electromagnetic wave absorption information specific to a first component different from the water contained in the waste, The plurality of calibration curves created in the calibration curve creation step include a first calibration curve created using the feature amount x1 and the actually measured value y' of the moisture content of each of the plurality of samples, When the waste is identified as containing the first component as a main component in the identification step, in the separation step, the waste is separated into the plurality of waste groups based on the prediction information Y on the moisture content of the waste derived using the feature amount x1 obtained from the electromagnetic wave absorption information of the waste and the first calibration curve. The waste separation method according to [3]. [5] The waste sorting method according to [4], wherein the feature quantity x1 is derived from the ratio of the near-infrared absorption intensity of the first component to water, or a calculation formula including the same. [6] The waste sorting method according to [4] or [5], wherein the first component is a polyolefin-based plastic or a polyester-based plastic. [7] The waste sorting method according to any one of [1] to [6], wherein the feature quantity x is the intensity of an absorption peak included in the wavelength range of 1440 to 1460 nm, or is derived from a calculation formula including the same. [8] The waste sorting method according to any one of [1] to [7], wherein the waste and the sample include the plastic and at least one selected from the group consisting of paper, wood chips, stone, glass, metal, cloth, and leather. [9] The waste sorting method according to any one of [1] to [8], which has an information acquisition step of irradiating the waste to be sorted, which is being conveyed, with electromagnetic waves including near-infrared rays and acquiring a near-infrared absorption spectrum as the electromagnetic wave absorption information.

[10] A method for producing cement clinker, which has a heating step of heating cement raw materials with a combustion device in a cement clinker production facility, A method for producing cement clinker, wherein at least one of the plurality of waste groups sorted by the sorting method according to any one of [1] to [9] above is used as fuel for the combustion device.

[11] The combustion device includes a precalciner and a cement kiln, The method for producing cement clinker according to

[10] , wherein the fuel used in the cement kiln includes the first sorted waste, and the fuel used in the precalciner includes the second sorted waste.

[12] A waste sorting device for fuel containing plastic, A calibration curve creation unit that creates a calibration curve for obtaining predicted moisture information Y of the waste using a feature quantity x obtained from the electromagnetic wave absorption information of each of a plurality of samples containing plastic and having different moisture contents, and the actually measured value y' of the moisture content of each of the plurality of samples; An information acquisition unit that acquires the electromagnetic wave absorption information of the waste; An information processing unit that obtains prediction information Y on the moisture content of the waste from the characteristic quantity x obtained from the electromagnetic wave absorption information of the waste and the calibration curve created by the calibration curve creation unit; A waste sorting device comprising: a sorting unit that sorts the waste into a plurality of waste groups including a first sorted waste and a second sorted waste having an average moisture content lower than that of the first sorted waste, based on the prediction information Y.

[13] A waste sorting device for fuel containing plastic, A calibration curve creation unit that creates a plurality of calibration curves for obtaining prediction information on the moisture of the waste, using the characteristic quantity x obtained from the electromagnetic wave absorption information of each of a plurality of samples containing plastic and having different moisture contents, and the measured value y' of the moisture content of each of the plurality of samples; An information acquisition unit that acquires the electromagnetic wave absorption information of the waste; An information processing unit that obtains prediction information Y on the moisture content of the waste from the characteristic quantity x obtained from the electromagnetic wave absorption information of the waste and at least one of the plurality of calibration curves created by the calibration curve creation unit; A waste sorting device comprising: a sorting unit that sorts the waste into a plurality of waste groups including a first sorted waste and a second sorted waste having an average moisture content lower than that of the first sorted waste, based on the prediction information Y.

[14] The information acquisition unit identifies the waste according to its composition, The sorting unit sorts the waste into the plurality of waste groups based on the prediction information Y by one calibration curve selected from the plurality of calibration curves selected according to the identification result in the information acquisition unit. The waste sorting device according to

[13] .

[15] The characteristic quantity x is a characteristic quantity x1 derived from the electromagnetic wave absorption information unique to water and the electromagnetic wave absorption information unique to a first component different from the water contained in the waste, The plurality of calibration curves created by the calibration curve creation unit include a first calibration curve created using the characteristic quantity x1 and the measured value y' of the moisture content of each of the plurality of samples; When the information processing unit identifies the waste as containing the first component as the main component, the sorting unit sorts the waste into the plurality of waste groups based on the predicted information Y of the moisture content of the waste derived using the feature amount x1 obtained from the electromagnetic wave absorption information of the waste and the first calibration curve. The waste sorting device according to

[14] .

[16] The information acquisition unit acquires a spectrum including near-infrared absorption information as the electromagnetic wave absorption information of the waste by using a light source that irradiates electromagnetic waves including near-infrared rays and a detector that detects the near-infrared rays. The waste sorting device according to any one of

[12] to

[15] .

[17] A crushing unit that crushes the waste, and a conveying unit that conveys the waste obtained by crushing by the crushing unit, The information acquisition unit acquires the electromagnetic wave absorption information of the waste conveyed by the conveying unit. The waste sorting device according to any one of

[12] to

[16] .

[18] A cement clinker manufacturing facility including any one of the sorting devices according to

[12] to

[17] and a combustion device, and using at least one of the plurality of waste groups as fuel for the combustion device.

[19] The combustion device includes a cement kiln and a precalciner, Using the second sorted waste sorted by the sorting device as fuel for the cement kiln and using the first sorted waste as fuel for the precalciner. The cement clinker manufacturing facility according to

[18] .

[20] The second sorted waste is supplied from at least one of the plurality of tanks to the cement kiln, and the first sorted waste is supplied from at least one of the plurality of tanks different from the tank that supplies the second sorted waste to the precalciner. The cement clinker manufacturing facility according to

[18] or

[19] .

Example

[0075] The content of the present invention will be described in more detail with reference to the examples, but the present invention is not limited to the following examples.

[0076] (Example 1) The following devices were used as the equipment constituting the conveyance unit, light source, information acquisition unit, and information processing unit in the separation device. · Conveyance unit: Movable stage (belt conveyor) · Light source that irradiates electromagnetic waves including near-infrared rays: Halogen lamp · Information acquisition unit: Hyperspectral camera (manufactured by RESONON, product name: PikaNIR-320) · Information processing unit: Personal computer (installed with control / analysis software "SPECTRONON")

[0077] Three hundred and twenty-eight types of waste samples with known moisture content were prepared. The measured value y' of the moisture content of each waste sample was measured by a method conforming to JIS Z 7302-2:1999 "Solidified fuel from waste - Part 3: Moisture test method" using a constant-temperature dryer. For each of the 328 types of waste samples, the near-infrared absorption spectrum was measured with a hyperspectral camera. In the measurement, each waste sample (about 100 g) was placed in an aluminum tray and installed on the movable stage. A halogen lamp and a hyperspectral camera were installed above the movable stage. While irradiating the reference sample with electromagnetic waves including near-infrared rays from the halogen lamp, the tray was moved by the movable stage, and the near-infrared absorption spectrum (measurement wavelength range: 880 to 1710 nm) of the waste sample in the tray was obtained by line scanning in a direction perpendicular to the moving direction.

[0078] Using analysis software (SPECTRONON manufactured by RESONON), preprocessing of the near-infrared absorption spectra of the obtained waste samples (328 types) was performed. In the preprocessing, first, the data of the near-infrared absorption spectra were limited to the wavelength range of 960 to 1710 nm to reduce noise. Then, the absorption intensity at each wavelength in the near-infrared absorption spectrum with reduced noise was divided by the maximum value of the absorption intensity in the spectrum (normalization by the maximum value). This reduced the influence of the baseline variation.

[0079] From the near-infrared absorption spectra of each of the waste samples (328 types) after pretreatment, the absorbance at a wavelength of 1450 nm was determined. This was used as the feature quantity x in Example 1. Out of the 328 types of waste samples, 225 types randomly selected were used as samples for calibration curve creation. The remaining waste samples (103 types) were used for verification of the agreement between the predicted value y and the measured value y' of the moisture content.

[0080] Figure 8(A) is a graph showing the relationship between the feature quantity x of the waste samples (225 types) for calibration curve creation on the horizontal axis and the measured value y' of the moisture content on the vertical axis. The regression equation of the measured value y' of the moisture content and the feature quantity x obtained by the least squares method was as shown in the following formula (1A). y’ = -57.891x + 56.463 (1A)

[0081] From the results in Figure 8(A), it was confirmed that there was a good correlation between the feature quantity x and the measured value y' of the moisture content. Based on this result, the following formula (1B) was obtained as the calibration curve used for predicting the moisture content of the waste. y = -57.891x + 56.463 (1B)

[0082] Using the waste samples (103 types) for verification as waste, the predicted value y of the moisture content was calculated using each feature quantity x and the above formula (1B). The predicted value y was obtained by substituting each feature quantity x of the waste samples (103 types) for verification into the above formula (1B).

[0083] Figure 8(B) is a graph showing the relationship between the obtained predicted value y of the moisture content and the measured value y' of the moisture content of the waste samples (103 types) for verification. As shown in Figure 8(B), the predicted value y of the moisture content and the measured value y' of the moisture content were distributed near the diagonal line in the graph, and it was confirmed that they had a good correlation. From this result, it was confirmed that the moisture content of the waste could be predicted with high accuracy using the feature quantity x (absorbance at a wavelength of 1450 nm) and the calibration curve of the above formula (1B). From the data plotted in Figure 8(B), RMSE (Root Mean Square Error) and MAE (Mean Absolute Error) were obtained. The results were as shown in Table 1.

[0084] (Example 2) The 328 types of waste samples used in Example 1 were visually sorted by hand into three types: plastic, paper, wood chips, and others. After that, the near-infrared absorption spectrum of the plastic was measured. According to the measurement results, out of the 328 types of waste samples, 112 samples contained polyolefin-based plastic as the main component and had a content of 40% by weight or more. Out of the 112 samples identified as containing polyolefin-based plastic as the main component from the 328 types, 76 samples were used to create a calibration curve. The remaining samples (36 types) were used to verify the agreement between the predicted value y of the moisture content and the measured value y'. The feature quantity x was the same as in Example 1 (absorbance at a wavelength of 1450 nm).

[0085] Figure 9(A) is a graph showing the relationship between the feature quantity x of the waste samples (76 types) for creating a calibration curve on the horizontal axis and the measured value y' of the moisture content on the vertical axis. The regression equation of the measured value y' of the moisture content and the feature quantity x obtained by the least squares method was as shown in the following formula (2A). y’=-74.455x+69.167 (2A)

[0086] From the results of Figure 9(A), it was confirmed that there was a good correlation between the feature quantity x and the measured value y' of the moisture content. Based on this result, the following formula (2B) was obtained as the calibration curve used to predict the moisture content of waste with polyolefin-based plastic as the main component. y=-74.455x+69.167 (2B)

[0087] The waste samples for verification (36 types) were used as waste, and the predicted value y of the moisture content was calculated using the respective feature quantity x and the above formula (2B). The predicted value y was obtained by substituting the respective feature quantity x of the waste samples for verification (36 types) into the above formula (2B).

[0088] Figure 9(B) is a graph showing the relationship between the predicted value y of the obtained moisture content and the measured value y' of the moisture content of the waste samples (36 types) for verification. As shown in Figure 9(B), the predicted value y of the moisture content and the measured value y' of the moisture content are distributed near the diagonal line in the graph, and it was confirmed that they have a good correlation. From this result, it was confirmed that the moisture content of the waste can be predicted with high accuracy by the characteristic quantity x (absorbance at a wavelength of 1450 nm) and the calibration curve of the above formula (2B). The RMSE and MAE were obtained from the data plotted in Figure 9(B). The results were as shown in Table 1.

[0089] (Example 3) From each of the near-infrared absorption spectra after the pretreatment measured in Example 1, in addition to the absorbance at a wavelength of 1450 nm, the absorbance at a wavelength of 1210 nm was obtained. Polyolefin plastics have an absorption peak at a wavelength of 1210 nm. The ratio of the absorbance at a wavelength of 1450 nm to the absorbance at a wavelength of 1210 nm was defined as the characteristic quantity x1 of Example 3. The calibration curve was created and the calibration curve was verified in the same manner as in Example 2 except that this characteristic quantity x1 was used.

[0090] Figure 10(A) is a graph showing the relationship between the characteristic quantity x1 of the waste samples (76 types) for creating the calibration curve on the horizontal axis and the measured value y' of the moisture content on the vertical axis. The regression equation of the measured value y' of the moisture content and the characteristic quantity x1 obtained by the least squares method was as shown in the following formula (3A). y’ = 39.904x1 - 32.94 (3A)

[0091] From the results of Figure 10(A), it was confirmed that there is a good correlation between the characteristic quantity x1 and the measured value y' of the moisture content. Based on this result, the following formula (3B) was obtained as the calibration curve used for predicting the moisture content of the waste mainly composed of polyolefin plastics. y = 39.904x1 - 32.94 (3B)

[0092] Similar to Example 2, using 36 types of waste samples for verification as waste, the predicted value y of the moisture content was calculated using each feature quantity x1 and the above formula (3B). Figure 10(B) is a graph showing the relationship between the obtained predicted value y of the moisture content and the measured value y' of the moisture content of the waste samples for verification (36 types). As shown in Figure 10(B), the predicted value y of the moisture content and the measured value y' of the moisture content are distributed near the diagonal line in the graph, and it was confirmed that they have a better correlation than Figure 9(B). The RMSE and MAE were obtained from the data plotted in Figure 10(B). The results were as shown in Table 1.

[0093] (Example 4) According to the measurement results of the near-infrared absorption spectrum of the plastics after manual sorting in Example 2, among the 328 types of waste samples used in Example 1, there were 50 samples that contained polyester-based plastics as the main component and had a content of 40% by weight or more. Among the 50 samples identified as containing polyester-based plastics as the main component from the 328 types of waste samples, 34 types were used for calibration curve creation. The remaining samples (16 types) were used for verification of the agreement between the predicted value y and the measured value y' of the moisture content. The feature quantity x was the same as in Example 1.

[0094] Figure 11(A) is a graph showing the relationship between the feature quantity x of the waste samples for calibration curve creation (34 types) on the horizontal axis and the measured value y' of the moisture content on the vertical axis. The regression equation of the measured value y' of the moisture content and the feature quantity x obtained by the least squares method was as shown in the following formula (4A). y’=-37.427x+34.585 (4A)

[0095] From the results of Figure 11(A), it was confirmed that there was a good correlation between the feature quantity x and the measured value y' of the moisture content. Based on this result, the following formula (4B) was obtained as the calibration curve used for predicting the moisture content of waste with polyester-based plastics as the main component. y=-37.427x+34.585 (4B)

[0096] Using the waste samples for verification (16 types) as waste, the predicted value y of the moisture content was calculated using each feature quantity x and the above formula (4B). The predicted value y was obtained by substituting each feature quantity x of the waste samples for verification (16 types) into the above formula (4B).

[0097] Figure 11(B) is a graph showing the relationship between the obtained predicted value y of the moisture content and the measured value y' of the moisture content of the waste samples for verification (16 types). As shown in Figure 11(B), the predicted value y of the moisture content and the measured value y' of the moisture content are distributed near the diagonal line in the graph, and it was confirmed that they have a good correlation. From this result, it was confirmed that the moisture content of the waste can be predicted with high accuracy by the calibration curve of the feature quantity x (absorbance at a wavelength of 1450 nm) and the above formula (4B). The RMSE and MAE were obtained from the data plotted in Figure 11(B). The results were as shown in Table 1.

[0098] (Example 5) From each of the near-infrared absorption spectra after the pretreatment measured in Example 1, in addition to the absorbance at a wavelength of 1450 nm derived from water, the absorbance at a wavelength of 1400 nm was determined. The polyester-based plastic has an absorption peak at a wavelength of 1400 nm. The ratio of the absorbance at a wavelength of 1450 nm to the absorbance at a wavelength of 1400 nm was defined as the feature quantity x1 of Example 5. The calibration curve was created and the calibration curve was verified in the same manner as in Example 4, except that this feature quantity x1 was used.

[0099] Figure 12(A) is a graph showing the relationship between the feature quantity x1 of the waste samples for calibration curve creation (34 types) on the horizontal axis and the measured value y' of the moisture content on the vertical axis. The regression equation of the measured value y' of the moisture content and the feature quantity x1 obtained by the least squares method was as shown in the following formula (5A). y’=78.341x1-75.656 (5A)

[0100] From the results of Figure 12(A), it was confirmed that there is a good correlation between the feature quantity x1 and the measured value y' of the moisture content. Based on this result, the following formula (5B) was obtained as the calibration curve used for predicting the moisture content of the waste mainly composed of polyester-based plastic. y = 78.341x1 - 75.656 (5B)

[0101] Similar to Example 4, using 16 types of waste samples for verification as waste, the predicted value y of the moisture content was calculated using each characteristic quantity x1 and the above formula (5B). Figure 12(B) is a graph showing the relationship between the obtained predicted value y of the moisture content and the measured value y' of the moisture content of 16 types of waste samples for verification. As shown in Figure 12(B), the predicted value y of the moisture content and the measured value y' of the moisture content are distributed near the diagonal line in the graph, and it was confirmed that they have a better correlation than Figure 11(B). The RMSE and MAE were obtained from the data plotted in Figure 12(B). The results were as shown in Table 1.

[0102] (Example 6) According to the manual sorting performed in Example 2, among the 328 types of waste samples used in Example 1, there were 107 types of samples that contained paper and wood chips as the main components and had a content of 40% by weight or more. Among the 107 types of samples identified as containing paper and wood chips as the main components from the 328 types of waste samples, 72 types were used for calibration curve creation. The remaining samples (35 types) were used for verification of the agreement between the predicted value y and the measured value y' of the moisture content. The characteristic quantity x was the same as in Example 1.

[0103] Figure 13(A) is a graph showing the relationship between the characteristic quantity x of 72 types of waste samples for calibration curve creation on the horizontal axis and the measured value y' of the moisture content on the vertical axis. The regression equation of the measured value y' of the moisture content and the characteristic quantity x obtained by the least squares method was as shown in the following formula (6A). y’ = -89.711x + 84.607 (6A)

[0104] From the results of Figure 13(A), it was confirmed that there is a good correlation between the characteristic quantity x and the measured value y' of the moisture content. Based on this result, the following formula (6B) was obtained as the calibration curve used for predicting the moisture content of waste mainly composed of polyester-based plastics. y = -89.711x + 84.607 (6B)

[0105] Using the 35 waste samples for verification as waste, the predicted value y of the moisture content was calculated using each characteristic quantity x and the above formula (6B). The predicted value y was obtained by substituting each characteristic quantity x of the 35 waste samples for verification into the above formula (6B).

[0106] Figure 13(B) is a graph showing the relationship between the obtained predicted value y of the moisture content and the measured value y' of the moisture content of the 35 waste samples for verification. As shown in Figure 13(B), the predicted value y of the moisture content and the measured value y' of the moisture content are distributed near the diagonal line in the graph, and it was confirmed that the predicted value y of the moisture content and the measured value y' of the moisture content have a good correlation. From this result, it was confirmed that the moisture content of the waste can be predicted with high accuracy by the characteristic quantity x (absorbance of water at a wavelength of 1450 nm) and the calibration curve of the above formula (6B). The RMSE and MAE were obtained from the data plotted in Figure 13(B). The results were as shown in Table 1.

[0107] (Example 7) In addition to the absorbance at a wavelength of 1450 nm derived from water, the absorbance at a wavelength of 1700 nm was obtained from each of the near-infrared absorption spectra after the pretreatment measured in Example 1. Paper and wood chips have an absorption peak at a wavelength of 1700 nm. The ratio of the absorbance at a wavelength of 1700 nm to the absorbance at a wavelength of 1450 nm was defined as the characteristic quantity x1 of Example 7. The calibration curve was created and the calibration curve was verified in the same manner as in Example 6 except that this characteristic quantity x1 was used.

[0108] Figure 14(A) is a graph showing the relationship between the characteristic quantity x1 of the 72 waste samples for calibration curve creation on the horizontal axis and the measured value y' of the moisture content on the vertical axis. The regression equation of the measured value y' of the moisture content and the characteristic quantity x1 obtained by the least squares method was as shown in the following formula (7A). y’=72.348x1-64.61 (7A)

[0109] From the results of Fig. 14(A), it was confirmed that there was a good correlation between the feature quantity x1 and the measured value y' of the moisture content. Based on this result, the following formula (7B) was obtained as the calibration curve used for predicting the moisture content of waste mainly composed of paper and wood chips. y = 72.348x1 - 64.61 (7B)

[0110] Similar to Example 6, 35 waste samples for verification were used as waste, and the predicted value y of the moisture content was calculated using the respective feature quantity x1 and the above formula (7B). Fig. 14(B) is a graph showing the relationship between the obtained predicted value y of the moisture content and the measured value y' of the moisture content of 35 waste samples for verification. As shown in Fig. 14(B), the predicted value y of the moisture content and the measured value y' of the moisture content were distributed near the diagonal line in the graph. RMSE and MAE were obtained from the data plotted in Fig. 14(B). The results were as shown in Table 1.

[0111]

Table 1

[0112] As shown in Table 1, it was confirmed that in each example, the moisture content of the waste could be predicted with sufficiently high accuracy. Also, it was confirmed that Examples 2 to 7 where discrimination was performed could predict the moisture content with even higher accuracy than Example 1. Further, it was confirmed that when plastic was the main component, by using the ratio of water to the main component as the feature quantity, the moisture content could be predicted with even higher accuracy.

Explanation of Reference Signs

[0113] 10…Crushing section, 11…Waste, 12…Conveyor section, 14…Light source, 15…Information acquisition section, 16…Detector, 20…Information processing section, 30…Control section, 40…Separation section, 41, 42…Dampers, 51, 52…Tanks, 52T…Three-way valve, 51V, 52V…Flow regulators, 51a, 52a…Outlet paths, 60…Calibration curve creation section, 72…Processor, 74…Memory, 76…Storage, 78…Input / output port, 82…Input / output device, 100…Separator, 200…Cement clinker manufacturing apparatus, 201…Inlet, 222…Regulating section, 224…Control section, 230…Prerotary kiln, 231…Regulating section, 232…Control section, 234…Rising duct, 236…Exhaust pipe, 237…Chlorine bypass section, 240…Cement kiln, 242…Kiln end, 244…Burner, 250…Clinker cooler, 252…Circulation gas line, 300…Cement clinker manufacturing facility.

Claims

1. A method for separating waste for fuel containing plastic, comprising: a calibration curve creation step of creating a calibration curve for obtaining predicted moisture information Y of the waste based on a feature quantity x obtained from electromagnetic wave absorption information of each of a plurality of samples containing plastic and having different moisture contents, and an actually measured value y' of the moisture content of each of the plurality of samples; a prediction step of deriving predicted moisture information Y of the moisture content of the waste from the feature quantity x obtained from the electromagnetic wave absorption information of the waste and the calibration curve; and a separation step of separating the waste into a plurality of waste groups including a first separated waste and a second separated waste having an average moisture content lower than that of the first separated waste based on the predicted information Y.

2. A method for separating waste for fuel containing plastic, comprising: a calibration curve creation step of creating a plurality of calibration curves for obtaining predicted moisture information Y of the waste based on a feature quantity x obtained from electromagnetic wave absorption information of each of a plurality of samples containing plastic and having different moisture contents and compositions other than moisture, and an actually measured value y' of the moisture content of each of the plurality of samples; a prediction step of deriving predicted moisture information Y of the moisture content of the waste from the feature quantity x obtained from the electromagnetic wave absorption information of the waste and at least one of the plurality of calibration curves; and a separation step of separating the waste into a plurality of waste groups including a first separated waste and a second separated waste having an average moisture content lower than that of the first separated waste based on the predicted information Y.

3. having an identification step of identifying the waste according to its composition, wherein in the separation step, the waste is separated into the plurality of waste groups based on the predicted information Y by one calibration curve selected from the plurality of calibration curves according to the identification result of the identification step.

4. The feature quantity x is a feature quantity x1 derived from electromagnetic wave absorption information specific to water and electromagnetic wave absorption information specific to a first component different from the water contained in the waste, and the plurality of calibration curves created in the calibration curve creation step include a first calibration curve created using the feature quantity x1 and the actually measured value y' of the moisture content of each of the plurality of samples. When the waste is identified as containing the first component as the main component in the identification step, in the separation step, the waste is separated into the plurality of waste groups based on the predicted information Y of the moisture content of the waste derived using the characteristic quantity x1 obtained from the electromagnetic wave absorption information of the waste and the first calibration curve. The waste separation method according to claim 3.

5. The characteristic quantity x1 is derived from the ratio of the near-infrared absorption intensities of the first component and water, or a calculation formula including the same. The waste separation method according to claim 4.

6. The first component is a polyolefin-based plastic or a polyester-based plastic. The waste separation method according to claim 4 or 5.

7. The characteristic quantity x is the intensity of an absorption peak included in the wavelength range of 1440 to 1460 nm, or is derived from a calculation formula including the same. The waste separation method according to any one of claims 1 to 5.

8. The waste and the sample include the plastic and at least one selected from the group consisting of paper, wood chips, stone, glass, metal, cloth, and leather. The waste separation method according to any one of claims 1 to 5.

9. An information acquisition step of irradiating the waste to be separated, which is being conveyed, with electromagnetic waves including near-infrared rays and acquiring a near-infrared absorption spectrum as the electromagnetic wave absorption information. The waste separation method according to any one of claims 1 to 5.

10. A method for manufacturing cement clinker, comprising a heating step of heating cement raw materials in a combustion device in a cement clinker manufacturing facility. A method for manufacturing cement clinker, wherein at least one of the plurality of waste groups separated by the separation method according to any one of claims 1 to 5 is used as fuel for the combustion device.

11. The combustion device includes a precalciner and a cement kiln. The fuel used in the cement kiln includes the first separated waste, and the fuel used in the precalciner includes the second separated waste. The method for manufacturing cement clinker according to claim 10.

12. A waste separation device for fuel containing plastic. A calibration curve creation unit that creates a calibration curve for obtaining predicted information Y of the moisture content of the waste using the characteristic quantity x obtained from the electromagnetic wave absorption information of each of a plurality of samples containing plastic and having different moisture contents, and the actually measured value y' of the moisture content of each of the plurality of samples. An information acquisition unit that acquires electromagnetic wave absorption information of the waste; An information processing unit that obtains prediction information Y on the moisture content of the waste from a feature quantity x obtained from the electromagnetic wave absorption information of the waste and the calibration curve created by the calibration curve creation unit; A sorting unit that sorts the waste into a plurality of waste groups including a first sorted waste and a second sorted waste having an average moisture content lower than that of the first sorted waste based on the prediction information Y. A waste sorting device comprising:

13. A waste sorting device for fuel containing plastic, A calibration curve creation unit that creates a plurality of calibration curves for obtaining prediction information on the moisture of the waste using a feature quantity x obtained from the electromagnetic wave absorption information of a plurality of samples each containing plastic and having different moisture contents, and the measured value y' of the moisture content of each of the plurality of samples; An information acquisition unit that acquires electromagnetic wave absorption information of the waste; An information processing unit that obtains prediction information Y on the moisture content of the waste from a feature quantity x obtained from the electromagnetic wave absorption information of the waste and at least one of the plurality of calibration curves created by the calibration curve creation unit; A sorting unit that sorts the waste into a plurality of waste groups including a first sorted waste and a second sorted waste having an average moisture content lower than that of the first sorted waste based on the prediction information Y. A waste sorting device comprising:

14. The information acquisition unit identifies the waste according to its composition, The sorting unit sorts the waste into the plurality of waste groups based on the prediction information Y by one calibration curve selected from the plurality of calibration curves selected according to the identification result in the information acquisition unit. The waste sorting device according to claim 13.

15. The feature quantity x is a feature quantity x1 derived from electromagnetic wave absorption information unique to water and electromagnetic wave absorption information unique to a first component different from the water contained in the waste, The plurality of calibration curves created by the calibration curve creation unit include a first calibration curve created using the feature quantity x1 and the measured value y' of the moisture content of each of the plurality of samples, When the waste is identified as containing the first component as a main component in the information processing unit, the sorting unit uses the feature quantity x1 obtained from the electromagnetic wave absorption information of the waste and the first calibration curve. The waste sorting device according to claim 14, wherein the waste is sorted into the plurality of waste groups based on the prediction information Y on the moisture of the waste derived therefrom.

16. The waste sorting device according to any one of claims 12 to 15, wherein the information acquisition unit acquires a spectrum including near-infrared absorption information as the electromagnetic wave absorption information of the waste by using a light source that irradiates electromagnetic waves including near-infrared rays and a detector that detects the near-infrared rays.

17. It includes a crushing unit that crushes the waste and a conveying unit that conveys the waste obtained by crushing by the crushing unit. The waste sorting device according to any one of claims 12 to 15, wherein the information acquisition unit acquires the electromagnetic wave absorption information of the waste conveyed by the conveying unit.

18. A cement clinker manufacturing facility comprising the sorting device according to any one of claims 12 to 15 and a combustion device, and using at least one of the plurality of waste groups as fuel for the combustion device.

19. The combustion device includes a cement kiln and a calciner. The cement clinker manufacturing facility according to claim 18, wherein the second sorted waste sorted by the sorting device is used as fuel for the cement kiln, and the first sorted waste is used as fuel for the calciner.

20. The cement clinker manufacturing facility according to claim 18, wherein the second sorted waste is supplied from at least one of the plurality of tanks to the cement kiln, and the first sorted waste is supplied from at least one of the plurality of tanks different from the tank that supplies the second sorted waste to the calciner.

Images