Sorting system and sorting method

Abstract

A sorting system according to one aspect of the present disclosure comprises: a sorting device for sorting a mixture containing a plurality of types of materials by each type of material; a first computation device provided with an acquisition unit that acquires mixture information, operation state information, and sorting result information, a first computation unit that performs a first computation process on the mixture information, the operation state information, and the sorting result information, and a transmission unit that transmits information on which the first computation process has been performed; and a second computation device provided with a second computation unit that computes sorting conditions of the sorting device on the basis of the information transmitted from the first computation device, and a transmission unit that transmits information indicating the sorting conditions of the sorting device computed by the second computation unit to the first computation device, wherein the first computation device is provided with a reception unit that receives information indicating the sorting conditions transmitted by the second computation device, and a control unit that controls the sorting device on the basis of the sorting conditions.

Description

Sorting system and sorting method

[0001] The present disclosure relates to a sorting system and a sorting method.

[0002] Conventionally, there has been known a technique for separating a mixture containing multiple types of objects into individual types of objects. For example, one such separation technique is an electrostatic separation device that electrically charges multiple types of objects contained in the mixture and separates specific objects as objects to be recycled by electrostatic separation. For example, the electrostatic separation device described in Patent Document 1 listed below is known as a technique for improving the accuracy of electrostatic separation.

[0003] The electrostatic separation device in Patent Document 1 agitates a group of plastic pieces containing a mixture of multiple types of plastic pieces, charges each plastic piece, drops the charged plastic pieces into an electrostatic field formed by a pair of electrodes, and collects the dropped plastic pieces in a collection box. This electrostatic separation device has an imaging unit between the pair of electrodes and the collection box that captures images of the falling plastic pieces, binarizes the images of the plastic pieces captured by the imaging unit, calculates the weight percentage of the plastic pieces in the image, and creates a drop distribution that distributes the weight percentage of the plastic pieces according to the drop position. Furthermore, this electrostatic separation device has a partition plate in the collection box to separate the plastic pieces by type, and the position of the partition plate is optimized.

[0004] International Publication No. 2022-195712

[0005] In the electrostatic separation device described above, it is necessary to collect sufficient data to optimally control the position of the partition plates, etc. However, immediately after a customer installs an electrostatic separation device or when there is a significant change in the materials fed into the electrostatic separation device, it takes time to collect data to optimally control the position of the partition plates, etc., and there is a problem that separation efficiency decreases during the period until sufficient data is collected.

[0006] The present disclosure has been made in consideration of these circumstances, and aims to provide a sorting system and a sorting method that can increase the sorting efficiency of a sorting device by utilizing sorting conditions obtained from outside.

[0007] The present disclosure has been made to solve the above-mentioned problems, and one aspect of the present disclosure is a sorting system comprising: a sorting device that sorts a mixture containing multiple types of materials by material type; an acquisition unit that acquires mixture information about the mixture, operating status information that indicates the operating status of the sorting device, and sorting result information that indicates the sorting result of the sorting device; a first calculation unit that performs a first calculation process on the mixture information, the operating status information, and the sorting result information acquired by the acquisition unit; and a transmission unit that transmits information resulting from the first calculation process performed by the first calculation unit; and a second calculation unit that calculates sorting conditions for the sorting device based on the information transmitted from the first calculation device, and a transmission unit that transmits information indicating the sorting conditions for the sorting device calculated by the second calculation unit to the first calculation device, wherein the first calculation device comprises a receiving unit that receives the information indicating the sorting conditions transmitted by the second calculation device, and a control unit that controls the sorting device based on the sorting conditions.

[0008] Another aspect of the present disclosure is a sorting method including the steps of: a sorting device that sorts a mixture containing multiple types of materials by material type; generating mixture information regarding the mixture in response to a supply of the mixture to be sorted; a first calculation device performing a first calculation process on the mixture information and transmitting information resulting from the first calculation process; a second calculation device calculating sorting conditions for the sorting device based on the information transmitted from the first calculation device and transmitting information indicating the sorting conditions for the sorting device to the first calculation device; and a step of the first calculation device controlling the sorting device to sort the mixture to be sorted based on the sorting conditions of the sorting device.

[0009] According to the present disclosure, the sorting efficiency of the sorting device can be increased by utilizing sorting conditions obtained from outside.

[0010] FIG. 1 is a block diagram showing an example of an electrostatic separation system in a first embodiment. FIG. 2 is a block diagram showing an example of a learning unit 308 in the first embodiment. FIG. 3 is a diagram showing an example of an electrostatic separation device 100 in the first embodiment. FIG. 4 is a block diagram showing an example of the functional configuration of a first calculation unit 204 in the first embodiment. FIG. 5 is a sequence diagram showing an example of the operation of the electrostatic separation system 1 in an embodiment. FIG. 6 is a flowchart showing an example of the processing of a first calculation unit 200 and a second calculation unit 300 in an embodiment. FIG. 7 is a diagram showing an example of a gravity-type separation device 500 in a second embodiment. FIG. 8 is a diagram showing an example of an optical separation device 600 in a third embodiment.

[0011] Hereinafter, a sorting system and a sorting method to which the present invention is applied will be described with reference to the drawings.

[0012] 1 is a block diagram showing an example of an electrostatic separation system 1 according to a first embodiment. The electrostatic separation system 1 includes, for example, a plurality of electrostatic separation devices 100A, 100B, 100C, ..., a plurality of first arithmetic units 200A, 200B, 200C, ..., a second arithmetic unit 300, and a database unit 400. In the following description, a plurality of electrostatic separation devices will be collectively referred to as "electrostatic separation device 100," and a plurality of first arithmetic units will be collectively referred to as "first arithmetic unit 200."

[0013] The electrostatic separation device 100, the first arithmetic device 200, the second arithmetic device 300, and the database device 400 each have a communication interface (not shown) such as a network interface card (NIC) or a wireless communication module for connecting to a communication network NW such as the Internet. The communication network NW may include, for example, a general-purpose network such as the Internet, and a private network such as local 5G or Wi-Fi (registered trademark).

[0014] In the embodiment, an electrostatic separation device 100 is described that uses static electricity to separate recyclable materials from a mixture containing multiple types of resins, but the present invention is not limited to this. The electrostatic separation system 1 of the embodiment may be any separation device that separates a mixture containing multiple types of materials by material type. The electrostatic separation system 1 of the embodiment can also be applied to a separation device that separates specific objects from a mixture containing metal or glass, or a separation device that separates specific objects from a mixture containing food.

[0015] The electrostatic separation device 100 separates materials to be recycled from a mixture (hereinafter also referred to as raw material) containing multiple types of resins (plastics). The materials to be recycled include multiple types of plastic pieces with different charging characteristics. The electrostatic separation device 100 includes, for example, a detection unit 102, a separation unit 104, a receiving unit 106, and a transmission unit 108. The detection unit 102 detects sensing data. The sensing data is data indicating the state of the mixture. The separation unit 104 performs an operation to separate plastic pieces from the mixture. The receiving unit 106 receives control commands from the first calculation unit 200. The transmission unit 108 transmits the sensing data to the first calculation unit 200.

[0016] The first arithmetic unit 200 is, for example, an information processing device having a function called an edge server. The first arithmetic unit 200 is installed, for example, in correspondence with the electrostatic separation device 100 installed in one factory. The first arithmetic unit 200 and the electrostatic separation device 100 are connected by a communication line such as a LAN line installed in the factory.

[0017] The first calculation device 200 includes, for example, an acquisition unit 202, a first calculation unit 204, a receiving unit 206, a transmitting unit 208, and a control unit 210. The first calculation unit 204 and the control unit 210 are functional units realized by a processor, such as a CPU (Central Processing Unit), executing a program stored in a program memory. The acquisition unit 202 is connected to the electrostatic separation device 100, for example, via a LAN line. The acquisition unit 202 acquires mixture information, operating status information, and sorting result information. The mixture information is information about the mixture in the electrostatic separation device 100. The mixture information may include, for example, at least one of the type, composition, and particle size of the material. The operating status information is information indicating the operating status of the electrostatic separation device 100. The sorting result information is information indicating the sorting result of the electrostatic separation device 100.

[0018] The first calculation unit 204 performs a first calculation process on at least one of the mixture information, operation status information, and sorting result information acquired by the acquisition unit 202. The first calculation process is, for example, a process of extracting at least some information from the mixture information, operation status information, and sorting result information. The first calculation process is, for example, a process of reducing the amount of data to be transmitted to the second calculation device 300 by organizing the mixture information, operation status information, and sorting result information. The transmission unit 208 transmits the information that has undergone the first calculation process from the first calculation unit 204.

[0019] The receiving unit 206 receives the sorting conditions from the second arithmetic unit 300. The control unit 210 controls the electrostatic separation device 100 based on the sorting conditions received by the receiving unit 206. The sorting conditions are information that serve as target values ​​for operating the electrostatic separation device 100. Specifically, the sorting conditions include a target value for the voltage value to be applied to the electric field generating unit 130, a target value for the partition plate position, etc.

[0020] The second arithmetic unit 300 is, for example, an information processing device that functions as a server that performs processing in response to a request from the first arithmetic unit 200 and transmits the processing result to the first arithmetic unit 200. The second arithmetic unit 300 is, for example, one second arithmetic unit corresponding to a plurality of first arithmetic units 200.

[0021] The second arithmetic unit 300 includes, for example, a receiving unit 302, a second arithmetic unit 304, a trained model 306, a learning unit 308, and a transmitting unit 310. The second arithmetic unit 304 and the learning unit 308 are functional units realized by a processor such as a CPU executing a program stored in a program memory. The receiving unit 302 receives information transmitted from the first arithmetic unit 200. The second arithmetic unit 304 calculates the sorting conditions of the electrostatic separation device 100 based on the information transmitted from the first arithmetic unit 200. The trained model 306 is a machine learning model used in the second arithmetic processing. The learning unit 308 performs processing to train the trained model 306. The transmitting unit 310 transmits the sorting conditions of the electrostatic separation device 100 calculated by the second arithmetic unit 304 to the first arithmetic unit 200.

[0022] The second calculation unit 304 inputs information transmitted from the first calculation device 200 into the trained model 306, and calculates the sorting conditions based on the output of the trained model 306. The trained model 306 learns the mixture information, operating state information, and sorting result information transmitted by each of the multiple first calculation devices 200 as learning data. Note that the second calculation unit 304 may calculate the sorting conditions by calculation processing using an arithmetic expression or rule-based calculation processing without using the trained model 306.

[0023] The database device 400 is an information processing device including a data storage device, a processor for constructing a database (DB), and the like. The database device 400 stores, for example, a user database 410, a sorting device database 412, and a trained model 414. The user database 410 is a database that associates user information, which identifies the user, industry, product, etc. of the electrostatic separation device 100, with the trained model 414. The sorting device database 412 is a database that associates mixture information, operating status information (sensing data), and sorting result information for each electrostatic separation device 100. The trained model 414 is information for identifying the trained model, such as information indicating the training data and processing parameters for identifying a machine learning model. The trained model 414 may be constructed by type, composition, or particle size of resin pieces. The trained model 414 may be constructed for each user, industry, or product.

[0024] The electrostatic separation system 1 according to the embodiment includes a plurality of electrostatic separation devices 100A, 100B, and 100C, a plurality of first arithmetic units 200A, 200B, and 200C corresponding to the plurality of electrostatic separation devices 100A, 100B, and 100C, respectively, and a second arithmetic unit 300 corresponding to the plurality of first arithmetic units 200A, 200B, and 200C. The second arithmetic unit 304 calculates the separation conditions for the single electrostatic separation device 100 based on a plurality of separation conditions transmitted by the plurality of first arithmetic units 200A, 200B, and 200C. The electrostatic separation system 1 is not limited to this, and may include, for example, one each of the electrostatic separation device 100, the first arithmetic unit 200, and the second arithmetic unit 300.

[0025] FIG. 2 is a block diagram showing an example of the learning unit 308 according to the first embodiment. The learning unit 308 inputs the sorting conditions of one electrostatic separation device 100 as learning data based on the sensing data acquired from the plurality of first arithmetic units 200A, 200B, and 200C, the processing results of the first arithmetic processing, and the sorting conditions calculated by the second arithmetic unit 304, and updates the processing parameters of the trained model 306 based on the output from the trained model 306. The processing parameters are, for example, filters (also referred to as weights or biases) included in a neural network, but are not limited thereto. The machine learning method used in the trained model 306 may be commonly used deep learning or random forest, for example.

[0026] 3 is a diagram showing an example of an electrostatic separation device 100 according to the first embodiment. The electrostatic separation device 100 includes, for example, a feed-side detection unit 110A, a recovery-side detection unit 110B, a charging cylinder 120, a vibrating feeder 122, an electric field generation unit 130, a recovery box 140, and a control device 150.

[0027] A raw material (mixture) containing multiple types of plastic pieces to be sorted is supplied to the charging cylinder 120. For example, a predetermined amount of raw material is fed into the charging cylinder 120 per unit time. In the present embodiment, the raw material includes plastic pieces containing a mixture of multiple materials obtained by, for example, crushing the housings of home appliances such as air conditioners, refrigerators, and washing machines to be recycled. In the present embodiment, the plastic pieces are ABS (Acrylonitrile Butadiene Styrene) pieces and PS (Polystyrene) pieces, which have different charging characteristics. The plastic pieces are, for example, approximately 5 mm square.

[0028] The charging cylinder 120 agitates the raw material by rotating. Plastic pieces A and B contained in the raw material rub against each other and become electrically charged inside the charging cylinder 120 as they rotate and agitate. The electrically charged plastic pieces A and B each have a polarity and amount of charge according to their triboelectric series. Specifically, plastic piece A (e.g., ABS) is positively charged, and plastic piece B (e.g., PS) is negatively charged.

[0029] Plastic pieces A and B, charged by charging cylinder 120, are discharged into vibrating feeder 122. Positively charged plastic piece A and negatively charged plastic piece B are attracted to each other by electrostatic force. Vibrating feeder 122 vibrates plastic pieces A and B up and down while pushing them toward electric field generator 130, causing plastic pieces A and B to separate and fall from the tip of vibrating feeder 122. Some of plastic pieces A and B are collected from vibrating feeder 122 and fed into separating and feeding device 124.

[0030] The electric field generating unit 130 generates an electrostatic field. The electric field generating unit 130 includes, for example, a pair of electrodes 131 and 132 and a DC power supply 133. Each of the pair of electrodes 131 and 132 is formed, for example, in a flat plate shape. The pair of electrodes 131 and 132 are arranged opposite each other, sandwiching the path along which the plastic pieces A and B fall. A ground voltage is applied to the electrode 131. The DC power supply 133 applies a DC voltage between the pair of electrodes 131 and 132, thereby generating an electrostatic field between the pair of electrodes 131 and 132. The DC voltage between the electrodes 131 and 132 is adjusted by a control signal output from the control device 150.

[0031] Plastic pieces A and B that fall from vibrating feeder 122 are attracted to one of the pair of electrodes 131 and 132 by electrostatic force according to their charge state, such as polarity and charge amount, and fall. In other words, plastic pieces A and B fall to different positions depending on their charge state. In this embodiment, plastic piece A is positively charged and therefore is attracted to electrode 131 and falls. Plastic piece B is negatively charged and therefore is attracted to electrode 132 and falls.

[0032] The dropped plastic pieces A and B are collected in a collection box 140. The collection box 140 collects the plastic pieces A and B. The collection box 140 includes, for example, partition plates 141 and 142 and a partition drive device 143. The partition plates 141 and 142 are movable, for example, in the directions of the arrows in the figure. The movement of the partition plates 141 and 142 adjusts the size of the three areas formed by the partition plates 141 and 142. The partition drive device 143 includes a drive mechanism such as a motor. The partition drive device 143 supplies drive force to the partition plates 141 and 142 based on a control signal supplied from the control device 150.

[0033] The partition plates 141 and 142 may be moved manually. In the embodiment, the partition plates 141 and 142 are described as partitions that separate the areas that contain the material induced by the electrostatic force, but they are not limited to a plate shape and may have other shapes.

[0034] Plastic pieces A and B dropped from the vibrating feeder 122 are collected in one of three areas formed by the partition plates 141 and 142 depending on their charge state. Positively charged plastic pieces A are collected mainly in the area on the partition plate 141 side. Negatively charged plastic pieces B are collected mainly in the area on the partition plate 142 side. Plastic pieces that are not sufficiently charged are collected in the area between the partition plates 141 and 142. The plastic pieces collected in each area are transported by a conveyor (not shown) and stored in separate containers. The electrostatic separation device 100 may return the plastic pieces collected in the area between the partition plates 141 and 142 to the supply section 220 and perform separation again.

[0035] The input side detection unit 110A includes, for example, a plastic type identification sensor 111, a charge amount sensor 112, a plastic type identification sensor 113, and a mass sensor 114.

[0036] The plastic type identification sensor 111 detects sensing data for identifying the plastic type of the plastic pieces fed into the charging cylinder 120. The plastic type identification sensor 111 is, for example, an optical sensor that captures an image of the mixture. The optical sensor may be one that generates image information, such as a near-infrared camera or a visible light camera. The plastic type identification sensor 111 may also be a sensor for detecting particle size or foreign matter.

[0037] The charge sensor 112 is, for example, a pass-through Faraday cage sensor. The charge sensor 112 drops charged plastic pieces from above and detects an electrostatic induction voltage waveform (sensing data) generated as the plastic pieces fall. The electrostatic induction voltage waveform is analyzed by the first calculation unit 204 and converted into an amount of charge. The plastic type identification sensor 113 is, for example, an optical sensor that captures an image of the mixture. The plastic type identification sensor 113 detects sensing data for identifying the plastic type of the plastic pieces that have passed through the charge sensor 112. The mass sensor 114 detects sensing data for determining the mass of the plastic pieces that have passed through the charge sensor 112.

[0038] The collection-side detection unit 110B includes, for example, a plastic type identification sensor 115 and a mass sensor 116. The plastic type identification sensor 115 is, for example, an optical sensor that captures an image of the mixture. The plastic type identification sensor 115 detects sensing data for identifying the plastic type of the plastic pieces collected in each of the three areas formed by the partition plates 141 and 142 in the collection box 140. The mass sensor 116 detects sensing data for determining the mass of the plastic pieces collected in each of the three areas formed by the partition plates 141 and 142.

[0039] The control device 150 controls each unit in the electrostatic separation device 100 and performs various calculations. In response to acquiring sensing data detected by the input side detection unit 110A and the recovery side detection unit 110B, the control device 150 immediately transmits at least one of the acquired sensing data to the first calculation device 200. The control device 150 controls the operation of at least one of the charging cylinder 120, the vibrating feeder 122, the electric field generation unit 130, and the recovery box 140 in accordance with a control command received from the first calculation device 200.

[0040] 4 is a block diagram showing an example of the functional configuration of the first calculation unit 204 in the first embodiment. As described above, the electrostatic separation device 100 includes the input-side detection unit 110A, the recovery-side detection unit 110B, and the control device 150. The input-side detection unit 110A detects, for example, near-infrared images or visible images, charge amount information, and mass information as sensing data, and the control device 150 outputs the sensing data to the first calculation unit 204. The first calculation unit 204 includes, for example, an image processing unit 204a and data conversion units 204b and 204c to process the sensing data detected by the input-side detection unit 110A.

[0041] As described above, the electrostatic separation device 100 includes the plastic type identification sensor 111 as an optical sensor that captures an image of the mixture to be separated and generates an image signal as mixture information, and the image processing unit 204a of the first calculation unit 204 generates mixing ratio data indicating the mixing ratio of the mixture based on the image signal generated by the plastic type identification sensor 111. The image processing unit 204a may, for example, analyze the infrared image signal detected by the plastic type identification sensor 111 and convert it into mixing ratio data indicating the mixing ratio of plastic pieces in the mixture. In this way, the image processing unit 204a converts the infrared image signal into mixing ratio data with a smaller data volume than an infrared image.

[0042] As described above, the electrostatic sorting device 100 is equipped with a plastic type identification sensor 111 as an optical sensor that captures an image of the mixture to be sorted and generates an image signal as mixture information, and the image processing unit 204a of the first calculation unit 204 extracts a portion of the image signal generated by the plastic type identification sensor 111, and the transmission unit 108 transmits a portion of the image signal to the second calculation device 300.

[0043] The first calculation unit 204 may analyze the visible image signal and convert it into particle size data and foreign matter data of the mixture. As a result, the first calculation unit 204 outputs particle size data and foreign matter data having a smaller data volume than the visible image. The first calculation unit 204 may also convert a particle passing waveform signal detected by a particle sensor (not shown) for the mixture supplied to the charging tube 120 into a fluid passing amount, and output the particle size and foreign matter data.

[0044] The electrostatic separation device 100 includes a charge sensor 112 that generates, as mixture information, a charge signal that changes in accordance with the charge of the mixture being separated. The electrostatic separation device 100 adjusts the magnitude of the voltage applied to electrodes 131, 132 (electrode pair) that apply an electrostatic force to the charged mixture, or the position of partition plates 141, 142 that separate areas containing the mixture induced by the electrostatic force. The data conversion unit 204b of the first calculation unit 204 analyzes the charge signal generated by the charge sensor 112 to generate charge data indicating the charge of the mixture. The data conversion unit 204b inputs an electrostatic induction voltage waveform from the plastic type identification sensor 113 and converts it into charge data. As a result, the data conversion unit 204b outputs charge data with a smaller data capacity than the electrostatic induction voltage waveform.

[0045] The data conversion unit 204c receives the mass signal from the mass sensor 114 and converts it into mass data, thereby outputting charge amount data having a smaller data capacity than the mass signal.

[0046] The electrostatic separation device 100 may be equipped with a flow rate sensor that generates, as mixture information, a flow rate signal that changes according to the flow rate of the mixture to be separated. The flow rate sensor may be, for example, a sensor that detects the amount of raw material fed into the charging cylinder 120. It may be a sensor integrated with the plastic type identification sensor 111, or may be a flow rate sensor provided separately from the plastic type identification sensor 111. The data conversion unit 204d of the first calculation unit 204 may analyze the flow rate signal generated by the flow rate sensor to generate flow rate information indicating the flow rate of the mixture and transmit the flow rate information to the second calculation unit 300.

[0047] The control device 150 outputs operating status information indicating the operating status of the electrostatic separation device 100, such as partition plate position information, voltage information, MID circulation information, charging cylinder information, and feeder speed information, to the data processing unit 204e. The partition plate position information is information indicating the positions of the partition plates 141 and 142. The voltage information is information indicating the voltage between the electrodes 131 and 132. The MID circulation information is information indicating the amount of plastic pieces circulated from the area between the partition plates 141 and 142 to the charging cylinder 120. The charging cylinder information is information indicating the rotation speed of the charging cylinder 120. The feeder speed information is information indicating the vibration speed of the vibrating feeder 122. The data processing unit 204e, for example, performs predetermined data processing and transmits operating status data linking the partition plate position information, voltage information, MID circulation information, charging cylinder information, and feeder speed information for each time period to the second calculation device 300.

[0048] The first calculation unit 204 includes, for example, an image processing unit 204f and a data conversion unit 204g for processing sensing data detected by the recovery-side detection unit 110B. The image processing unit 204f analyzes the infrared image detected by the plastic type identification sensor 115 and converts it into composition ratio data indicating the composition ratio of the mixture. As a result, the image processing unit 204f outputs composition ratio data with a smaller data volume than the infrared image. The data conversion unit 204g inputs a mass signal from the mass sensor 116 and converts it into mass data. As a result, the data conversion unit 204c outputs charge amount data with a smaller data volume than the mass signal.

[0049] 5 is a sequence diagram showing an example of the operation of the electrostatic separation system 1 according to the embodiment. First, when the supply of raw materials to be separated is started (step S100), the electrostatic separation device 100 acquires sensing data detected by the plastic type identification sensor 111 and transmits the data to the first arithmetic unit 200 (step S102).

[0050] The first calculation unit 204 executes a first calculation process using the sensing data acquired by the acquisition unit 202 (step S104). The transmission unit 208 transmits the processing result of the first calculation process to the second calculation device 300 (step S106).

[0051] The control device 150 immediately transmits each sensing data to the first arithmetic device 200 in response to acquiring sensing data from each of the plastic type identification sensor 111, the charge amount sensor 112, the plastic type identification sensor 113, and the mass sensor 114. In response to receiving each of the plurality of sensing data, the first arithmetic device 204 immediately performs a first arithmetic process on each of the plurality of sensing data, and transmits a plurality of processing results corresponding to each of the plurality of sensing data to the second arithmetic device 300.

[0052] The receiving unit 302 of the second calculation device 300 receives the processing result of the first calculation process, and the second calculation unit 304 executes the second calculation process using the processing result of the first calculation process (step S108) to calculate the sorting conditions for the raw materials to be sorted. The transmitting unit 310 transmits information indicating the sorting conditions as the processing result of the second calculation process to the first calculation device 200 (step S110).

[0053] The second calculation unit 304 performs advanced automatic sorting by sensing the raw material mixture ratio, sorting results, and specific charge distribution of the plastic pieces based on the processing results of the first calculation process. The specific charge is a physical property value obtained by dividing the charge amount of plastic pieces charged by friction by their mass. Even for plastic pieces of the same type, the specific charge does not have a constant value but varies depending on the shape of each plastic piece and the way they rub against each other during frictional charging. There is a correlation between the specific charge distribution and the distribution of the falling positions of the plastic pieces after passing between the electrodes 131 and 132. Therefore, the second calculation unit 304 calculates the specific charge distribution and predicts the optimal positions of the partition plates 141 and 142 based on the specific charge distribution.

[0054] The electrostatic separation device 100 collects a portion of the mixed plastic immediately after it has been frictionally charged by the charging tube 120, and then uses the separation and charging device 124 to feed the collected plastic pieces one by one into the plastic type identification sensor 113. The sensing data obtained by the plastic type identification sensor 113 (optical sensor) and the mass sensor 114 is then transmitted to the second calculation device 300 via the first calculation device 200. The second calculation device 300 predicts the distribution of the drop positions of the plastic pieces based on the specific charge distribution, thereby determining the optimal positions (sorting conditions) of the partition plates 141 and 142, and remotely controls the partition drive device 143.

[0055] The receiving unit 206 of the first calculation device 200 receives information indicating the sorting conditions from the second calculation device 300, and the transmitting unit 208 transmits the information indicating the sorting conditions to the electrostatic separation device 100 as a control command (step S112).

[0056] The receiving unit 106 of the electrostatic separation device 100 receives the control command, and the control device 150 controls the separation unit 104, which includes the charging cylinder 120, the vibrating feeder 122, the electric field generation unit 130, and the recovery box 140, so that the electrostatic separation device 100 separates single plastics (step S114). The recovery-side detection unit 110B detects sensing data, and the control device 150 transmits the sensing data acquired from the recovery-side detection unit 110B to the first calculation unit 200 (step S116).

[0057] The first calculation unit 204 performs a first calculation process using the sensing data transmitted by the first calculation device 200 (step S118), and transmits the processing result of the first calculation process to the second calculation device 300 (step S120).

[0058] The second calculation device 300 updates the selection conditions by performing a second calculation process using the processing results of the first calculation process (step S122), and transmits the updated selection conditions to the first calculation device 200 (step S124).

[0059] The electrostatic separation device 100 also acquires sensing data from the plastic type identification sensor 115 and mass sensor 116 for the plastic pieces collected in the three regions formed by the partition plates 141 and 142 after sorting. The second calculation unit 304 calculates the sorting results for the plastic pieces for each of the three regions formed by the partition plates 141 and 142. The sorting results may be, for example, purity and collected amount. The second calculation unit 304 compares the predicted value of the distribution of the drop positions of the plastic pieces with the actual measured value of the distribution of the drop positions of the plastic pieces to recalculate the optimal positions (sorting conditions) of the partition plates 141 and 142. In this way, the second calculation unit 304 improves the accuracy of determining the positions of the partition plates 141 and 142 in response to fluctuations in the raw material.

[0060] The receiving unit 206 of the first calculation device 200 receives information indicating the updated sorting conditions from the second calculation device 300, and the transmitting unit 208 transmits the information indicating the sorting conditions as a control command to the electrostatic separation device 100 (step S126). The receiving unit 106 of the electrostatic separation device 100 receives the control command, and the control device 150 controls the separation unit 104, which includes the charging cylinder 120, the vibrating feeder 122, the electric field generation unit 130, and the collection box 140. In this way, the electrostatic separation device 100 separates the raw materials to be separated (step S128).

[0061] 6 is a flowchart showing an example of processing by the first arithmetic unit 200 and the second arithmetic unit 300 according to the embodiment. The first arithmetic unit 200 determines whether the electrostatic separation device 100 is operating (step S200). If the electrostatic separation device 100 is operating (step S200: YES), the first arithmetic unit 200 performs the processing described with reference to FIG. 5 (step S202).

[0062] When the electrostatic separation device 100 is stopped (step S200: NO), the first calculation unit 200 determines whether there is untransmitted data (step S204). The untransmitted data is image signals other than a part of the image signals when a part of the image signals generated by the plastic type identification sensor 111 or the plastic type identification sensor 113 is extracted. When the electrostatic separation device 100 is operating, the untransmitted data may be raw information used to calculate the calculation result information when the calculation result information obtained by the first calculation process performed by the first calculation unit 204 is being transmitted.

[0063] The first calculation unit 204 acquires untransmitted data from the electrostatic separation device 100 (step S206). The first calculation unit 204 may acquire raw information for the first calculation process from an internal storage device. The transmission unit 208 transmits the acquired untransmitted data to the second calculation device 300 (step S208). The reception unit 302 of the second calculation device 300 receives the untransmitted data, and the learning unit 308 trains the trained model 306 using the received untransmitted data (step S210).

[0064] As described above, the electrostatic separation system 1 according to the embodiment includes a plurality of electrostatic separation devices 100, a plurality of first calculation devices 200 that transmit information resulting from first calculation processing performed on the mixture information, operating state information, and separation result information acquired by the electrostatic separation devices 100, and a second calculation device 300 that calculates separation conditions for the electrostatic separation devices 100 based on the information transmitted from the first calculation devices 200, and the first calculation device 200 controls the electrostatic separation device 100 based on the separation conditions transmitted by the second calculation device 300. According to this electrostatic separation system 1, by utilizing separation conditions acquired from outside the electrostatic separation device 100, it is possible to utilize sensing data and separation condition data of other electrostatic separation devices 100 even in a situation where there is little sensing data or separation condition data, such as immediately after the electrostatic separation device 100 is installed, and therefore the separation efficiency of the electrostatic separation device 100 can be improved. Furthermore, according to the electrostatic separation system 1, the computational load of the electrostatic separation device 100 and the second calculation device 300 can be reduced, and communication delays can be reduced, compared to sending data directly from the electrostatic separation device 100 to the second calculation device 300.

[0065] Second Embodiment FIG. 7 is a diagram showing an example of a gravity-based separator 500 according to a second embodiment. While the separator according to the first embodiment described above is the electrostatic separator 100, the separator may be a gravity-based separator 500. The gravity-based separator 500 is a separator that performs separation by utilizing differences in specific gravity according to the types of resin pieces contained in the mixture to be separated. The gravity-based separator 500 separates plastic pieces contained in the mixture by, for example, a wet gravity separation method. The gravity-based separator 500 includes, for example, a float-sink separator 510 and a jig separator 520.

[0066] The float / sink sorting section 510 uses water as a medium to separate plastic pieces by floating PP, which is lighter than water, and sinking PS, ABS, and other plastics, which are heavier than water. Plastic pieces floating on the water in the float / sink sorting section 510 are collected. Plastic pieces that have settled in the float / sink sorting section 510 are transported to the jig sorting section 520.

[0067] The jig sorting unit 520 generates a pulsating water current by transmitting a piston motion to the water, forming layers of different specific gravity. The jig sorting unit 520 separates the light-specific gravity PS and ABS mixture from the heavy-specific gravity flame-retardant plastic, and collects the light-specific gravity plastic pieces and the heavy-specific gravity plastic pieces separately.

[0068] As described above, the first calculation unit 200 and the second calculation unit 300 are connected to the gravity-type sorting device 500. The first calculation unit 204 performs a first calculation process on at least a portion of the information: mixture information related to the mixture in the gravity-type sorting device 500, operating status information indicating the operating status of the gravity-type sorting device 500, and sorting result information indicating the sorting results of the gravity-type sorting device 500, and transmits the processing results of the first calculation process to the second calculation unit 300. The second calculation unit 300 calculates sorting conditions for the gravity-type sorting device 500 by the second calculation process and transmits the calculated sorting conditions to the first calculation unit 200. The first calculation unit 200 controls the gravity-type sorting device 500 based on the sorting conditions received from the second calculation unit 300.

[0069] Third Embodiment FIG. 8 is a diagram showing an example of an optical sorting device 600 according to a third embodiment. While the sorting device according to the first embodiment described above is the electrostatic sorting device 100, the sorting device may be an optical sorting device 600. The optical sorting device 600 is an optical sorting device that includes an optical sensor (X-ray source 620, detector 622) that captures images of resin pieces contained in a mixture to be sorted, and performs sorting by utilizing differences in light absorption characteristics according to the type of resin piece. The optical sorting device 600 includes, for example, a conveyor unit 610, an X-ray source 620, a detector 622, an air gun unit 630, recovery units 640 and 642, and a control device 650.

[0070] The optical sorting device 600 irradiates the mixture conveyed by the conveyor unit 610 with X-rays from the X-ray source 620 and detects the transmitted light with the detector 622. This generates an X-ray transmission image signal of the mixture to be sorted. The X-ray transmission image signal is, for example, an image in which a shadow is projected on the bromine-containing plastic. The X-ray transmission image signal is transmitted as sensing data by the control device 650 to the first arithmetic unit 200. The first arithmetic unit 204 transmits the processing results of the first arithmetic processing on the sensing data to the second arithmetic unit 300. The second arithmetic unit 300 transmits sorting conditions indicating the operation of the air gun unit 630 to the first arithmetic unit 200 based on the processing results of the first arithmetic processing. The first arithmetic unit 200 controls the air gun unit 630 based on the sorting conditions received from the second arithmetic unit 300. As a result, the optical sorting device 600 sorts, for example, plastics containing brominated flame retardants into the collection section 642 and plastics not containing brominated flame retardants into the collection section 640.

[0071] Although each embodiment and variant have been described, these are merely examples and are not intended to limit the scope of the present invention. For example, one of the embodiments or variants, or a part of each embodiment or a part of each variant, may be combined with one or more other embodiments or one or more other variants to realize one aspect of the present invention.

[0072] 1...electrostatic separation system, 100, 100A, 100B, 100C...electrostatic separation device, 102...detection unit, 104...sorting unit, 106...receiving unit, 108...transmitting unit, 110A...feed-in side detection unit, 110B...recovery-side detection unit, 111, 113, 115...plastic type identification sensor, 112...charge amount sensor, 114, 116...mass sensor, 120...charging cylinder, 122...vibration feeder, 124...separation and feeding device, 130...electric field generation unit, 131, 132...electrode, 133...DC power supply, 140...recovery box, 141, 142...partition plate, 143, 150...driving device, 200, 200A, 200B, 200C...first calculation unit, 202...acquisition unit, 204...first calculation unit, 204a, 20 4f...image processing unit, 204b, 204c, 204d, 204g...data conversion unit, 204e...data processing unit, 206...receiving unit, 208...transmitting unit, 210...control unit, 220...supply unit, 300...second calculation unit, 302...receiving unit, 304...second calculation unit, 306...trained model, 308...learning unit, 310...transmitting unit, 400...database device, 410...user database, 412...sorting device database, 414...trained model, 500...specific gravity sorting device, 510...floating / sinking sorting unit, 520...jig sorting unit, 600...optical sorting device, 610...conveyor unit, 620...X-ray source, 622...detector, 630...air gun unit, 640, 642...recovery unit, 650...control unit

Claims

1. A sorting system comprising: a sorting device that sorts a mixture containing multiple types of materials by material type; a first arithmetic unit comprising: an acquisition unit that acquires mixture information regarding the mixture, operating status information that indicates the operating status of the sorting device, and sorting result information that indicates the sorting results of the sorting device; a first arithmetic unit that performs a first arithmetic process on the mixture information, operating status information, and sorting result information acquired by the acquisition unit; and a transmission unit that transmits information resulting from the first arithmetic process performed by the first arithmetic unit; and a second arithmetic unit that calculates sorting conditions for the sorting device based on information transmitted from the first arithmetic unit, and a transmission unit that transmits information indicating the sorting conditions for the sorting device calculated by the second arithmetic unit to the first arithmetic unit, wherein the first arithmetic unit comprises: a receiving unit that receives the information indicating the sorting conditions transmitted by the second arithmetic unit, and a control unit that controls the sorting device based on the sorting conditions.

2. A sorting system as described in claim 1, comprising: a plurality of sorting devices; a plurality of first arithmetic units corresponding to each of the plurality of sorting devices; and one second arithmetic unit corresponding to the plurality of first arithmetic units, wherein the second arithmetic unit calculates sorting conditions for one of the sorting devices based on information received from the plurality of first arithmetic units.

3. The sorting system described in claim 2, wherein the second calculation unit inputs information transmitted from the first calculation device into a trained model and calculates the sorting conditions based on the output of the trained model, and the trained model is trained using the mixture information, the operating status information, and the sorting result information transmitted by each of the multiple first calculation devices as training data.

4. The sorting system according to claim 1, wherein the sorting device is equipped with an optical sensor that captures an image of the mixture to be sorted and generates an image signal as the mixture information, and the first calculation unit generates mixing ratio data indicating the mixing ratio of the mixture based on the image signal generated by the optical sensor.

5. The sorting system according to claim 1, wherein the sorting device is equipped with an optical sensor that captures an image of the mixture to be sorted and generates an image signal as the mixture information, the first calculation unit extracts a portion of the image signal generated by the optical sensor, and the transmission unit transmits a portion of the image signal to the second calculation unit.

6. The sorting system according to claim 5, wherein the transmitting unit transmits information other than part of the image signal to the second computing unit when the sorting device is stopped operating.

7. The sorting system of claim 1, wherein the transmitting unit transmits calculation result information in which the first calculation process is performed by the first calculation unit when the sorting device is operating, and transmits raw information used to calculate the calculation result information when the sorting device is not operating.

8. The sorting system according to claim 1, wherein the sorting device is equipped with a flow sensor that generates, as the mixture information, a flow rate signal that changes according to the flow rate of the mixture to be sorted, and the first calculation unit analyzes the flow rate signal generated by the flow rate sensor to generate flow rate information indicating the flow rate of the mixture.

9. The sorting system according to claim 1, wherein the sorting device is an electrostatic sorting device that includes an electric charge sensor that generates, as the mixture information, an electric charge signal that changes according to the electric charge of the mixture to be sorted, and adjusts the magnitude of the voltage applied to an electrode pair that applies an electrostatic force to the charged mixture, or the position of a partition that separates an area that contains the mixture induced by the electrostatic force, and wherein the first calculation unit analyzes the electric charge signal generated by the electric charge sensor to generate electric charge information that indicates the electric charge of the mixture.

10. The sorting system according to claim 1, wherein the sorting device is a gravity-based sorting device that sorts by utilizing differences in specific gravity according to the types of resin pieces contained in the mixture to be sorted, and the first calculation unit performs a first calculation process on at least part of the information: mixture information regarding the mixture in the gravity-based sorting device, operating status information indicating the operating status of the gravity-based sorting device, and sorting result information indicating the sorting results of the gravity-based sorting device.

11. The sorting system described in claim 1, wherein the sorting device is an optical sorting device equipped with an optical sensor that captures images of resin pieces contained in the mixture to be sorted and performs sorting by utilizing differences in absorption characteristics according to the type of resin pieces captured by the optical sensor, and the first calculation unit performs first calculation processing on at least part of information including mixture information regarding the mixture in the optical sorting device, operating status information indicating the operating status of the optical sorting device, and sorting result information indicating the sorting results of the optical sorting device.

12. A sorting method comprising the steps of: a sorting device that sorts a mixture containing multiple types of materials by type of material; generating mixture information regarding the mixture in response to the mixture being supplied to the sorting device; a first calculation device performing a first calculation process on the mixture information and transmitting information resulting from the first calculation process; a second calculation device calculating sorting conditions for the sorting device based on the information transmitted from the first calculation device and transmitting information indicating the sorting conditions for the sorting device to the first calculation device; and a first calculation device controlling the sorting device to sort the mixture based on the sorting conditions of the sorting device.

Images