A retired wind power blade fiber high-value intelligent sorting system and process

Abstract

The application discloses a kind of retired wind power blade fiber high-value intelligent sorting system and process.The intelligent sorting system includes feed unit, electromagnetic vibration type fiber screening unit, conveying unit, fiber dispersion unit, fiber arrangement unit, photoelectric identification unit, pneumatic sorting unit, impurity removal recovery bin and four-stage recovery bin.Through the sorting system of the application, a continuous processing process is constructed, in which fiber materials are gradually converted from a scattered mixed state to a dispersed state and then to an interval ordered arrangement state, so that the originally difficult to identify fiber mixture is converted into an ordered fiber flow that can be continuously detected and accurately sorted, providing a reliable premise for subsequent intelligent grading, thereby achieving efficient identification and sorting of retired wind power blade fibers with different physical and chemical characteristics, improving grading accuracy and recycling value, and promoting the resource utilization of retired wind power blade fibers.

Description

Technical Field

[0001] This invention belongs to the field of composite material recycling technology, specifically, it relates to an intelligent sorting system and process for high-value fiber recycling of decommissioned wind turbine blades based on multi-sensor fusion. Background Technology

[0002] With the large-scale development of the wind power industry, a large number of wind turbine blades are being decommissioned after reaching the end of their service life. The number of decommissioned wind turbine blades worldwide exceeds 100,000 tons annually, and their main material is glass fiber reinforced polymer (GFRP). These decommissioned wind turbine blades are numerous and bulky, and if not properly disposed of, they will occupy a large amount of land resources and pose a potential threat to the environment.

[0003] In the construction industry, modifying retired wind turbine blades for use in cement concrete or asphalt mixtures can reduce production costs and improve performance. The resource recovery process for retired wind turbine blades typically includes dismantling, cutting, crushing, grading, refining, and grinding. Existing methods use high-efficiency fine crushers to further transform sheet materials into fiber powder mixtures. Screening yields relatively uniform retired wind turbine blade fibers. However, current methods for reusing these fibers are rather crude, failing to achieve automated sorting based on precise fiber length and resin content. At best, the mixed fibers can only be classified according to a rough size range (e.g., coarse / medium / fine), without further subdivision and grading based on specific length, thickness, type, and resin content. This makes it difficult for the blade fiber mixture to match the fiber performance requirements of different value-added applications, hindering efficient grading and utilization and reducing the value of recycling.

[0004] Traditional fiber sorting methods suffer from several problems: 1) low efficiency due to manual sorting; 2) inaccurate grading due to a single detection parameter; and 3) detection distortion caused by fiber entanglement. Currently, existing patents lack fiber sorting methods that combine both physical and chemical characteristics.

[0005] In view of the above problems and challenges, there is an urgent need for an innovative method for processing decommissioned wind turbine blade fibers. This method can efficiently identify the physical characteristics (length and thickness) and chemical characteristics (resin content) of the fibers to achieve graded recycling, promote precise matching for different value-added application scenarios, reduce labor and energy costs, and promote the recycling and utilization of decommissioned wind turbine blades and recycled products. Summary of the Invention

[0006] To address the challenge of automating and accurately classifying the mixture of decommissioned wind turbine blade fibers obtained from mechanical crushing according to the performance requirements of different value-added application scenarios, thus affecting efficient classification and utilization, this invention provides a high-value intelligent sorting system and process for decommissioned wind turbine blade fibers. This system enables efficient identification and sorting of decommissioned wind turbine blade fibers with different physical and chemical characteristics, improving classification accuracy and recycling value, promoting the resource utilization of decommissioned wind turbine blade fibers, and achieving length detection deviation control within ±0.5mm and resin content detection error control within 3%.

[0007] To achieve the above-mentioned objectives, the present invention adopts the following technical solution: A smart sorting system for high-value fiber recycling in decommissioned wind turbine blades includes: The feeding unit is a feeding funnel with a feeding port, which is used to quantitatively transport the crushed decommissioned wind turbine blade fiber material to the electromagnetic vibration fiber screening unit. An electromagnetic vibrating fiber screening unit is used for vibratory conveying, screening, impurity removal, and preliminary classification of fiber materials from decommissioned wind turbine blades. It is positioned below the feed hopper. This electromagnetic vibrating fiber screening unit includes an electromagnetic excitation unit, an elastic support mechanism, a vibrating table, a screen assembly, and a collection hopper. It has a compact structure. The screen assembly is embedded in the vibrating table to form an integrated screening structure, which is inclined, with the upper end being the feed end and the lower end the discharge end. At least one set of elastic support mechanisms is provided at both the feed and discharge ends. The collection hopper is positioned below this integrated structure. The screen assembly uses a fine mesh structure. The electromagnetic excitation unit is fixed to the vibrating table. The screen assembly and the vibrating platform are preferably located below or behind the high end of the vibrating table surface. They are driven by a high-frequency, small-amplitude alternating electromagnetic force to generate stable reciprocating micro-amplitude vibrations. The vibration frequency and amplitude can be adjusted according to the characteristics of the fiber material. When the electromagnetic excitation unit is working, under the combined action of vibration and slope, dust, balsa wood chips and other environmental or structural impurities can pass through the screen holes and fall down to achieve dust and impurity removal. Larger fiber materials cannot pass through the screen and are conveyed to the discharge end along the slope of the screen, thereby achieving continuous feeding and pre-screening. Under the action of electromagnetic vibration, continuous material conveying and impurity separation are achieved. The conveying units are respectively located between the electromagnetic vibration fiber screening unit and the fiber dispersion unit, between the fiber arrangement unit and the photoelectric recognition unit, and within the pneumatic sorting unit. They include a main conveyor belt, an identification conveyor belt, and a sorting conveyor track. The main conveyor belt is located between the electromagnetic vibration fiber screening unit and the fiber dispersion unit. The identification conveyor belt is located between the discharge end of the fiber arrangement unit and the photoelectric recognition unit, used to receive the arranged fibers and stably convey them to the photoelectric recognition unit at a preset pace and preset posture. The sorting conveyor track is located within the pneumatic sorting unit and behind the photoelectric recognition unit, used to convey the fibers that have completed identification and grading to the corresponding pneumatic nozzle's blowing position at a preset speed to achieve subsequent grading and sorting. The fiber dispersion unit is a three-section strip material vibrating conveyor. The decommissioned wind turbine blade fiber material, after being impurity removed by the electromagnetic vibrating fiber screening unit, is conveyed to the fiber dispersion unit for dispersion by the main conveyor belt of the conveying unit. The feed end of the main conveyor belt is located below the discharge end of the electromagnetic vibrating fiber screening unit, and the other end of the main conveyor belt is the discharge end, which is connected to the feed port of the fiber dispersion unit. The main conveyor belt is used to receive the decommissioned wind turbine blade fiber material after being screened and impurity removed by the electromagnetic vibrating fiber screening unit and continuously convey it to the fiber dispersion unit. The fiber arrangement unit comprises 12 to 24 vibration-assisted sequential fiber feeding mechanisms, preferably 16, used to further guide, limit, and output the dispersed fibers according to a rhythm. Each vibration-assisted sequential fiber feeding mechanism includes a receiving vibrating base plate, a high-pressure limiting component, a tapered inlet, a rhythmic feeding component, and a slit outlet. The receiving vibrating base plate receives and conveys the dispersed fiber material from the previous stage. The high-pressure limiting component is positioned above the feed end of the receiving vibrating base plate to limit the fiber accumulation thickness. The tapered inlet is located on the side of the high-pressure limiting component away from the feed end, used to limit the fiber accumulation thickness. The fiber is arranged in a staggered pattern; the cycle-shifting material feeder is located outside the discharge end of the vibrating base plate and reciprocates or pushes intermittently at a preset frequency (0.5~5Hz, preferably 1~3Hz, specifically determined according to the equipment cycle) to apply a periodic feeding action to the fiber; the slit discharge port is connected to the discharge end of the vibrating base plate and is used to control the fiber to be released in batches according to the cycle, thereby changing the strip fiber from a continuous and scattered conveying state to an intermittent and orderly output state, and finally evenly arranged into multiple parallel and uniformly spaced output tracks; the number of output tracks is 12~24, preferably 16; A photoelectric identification unit, located at the end of the identification conveyor belt behind the discharge end of the fiber arrangement unit, is used for dual-channel detection of the decommissioned wind turbine blade fiber material processed by the fiber dispersion unit and the fiber arrangement unit. The photoelectric identification unit includes a visual detection module, a spectral detection module, a transmission track, and a data fusion processing unit. The visual detection module and the spectral detection module are arranged sequentially along the fiber transmission direction to continuously detect the same fiber, thereby identifying fibers of different grades. The visual detection module is equipped with an industrial camera and a visual detection light source to collect fiber length and morphology information. The spectral detection module is equipped with a near-infrared spectrometer and a spectral detection light source to collect information on resin residue on the fiber surface. The data fusion processing unit communicates with both detection modules and has a built-in fusion classification model, preferably a machine learning classification model, and more preferably a neural network classification model. It performs fusion analysis on the detection data from both modules and automatically classifies the fibers into four grades (A to D) based on a preset grading threshold. The pneumatic sorting unit is located behind the photoelectric recognition unit and is used to pneumatically sort the decommissioned wind turbine blade fibers after they have been identified by the photoelectric recognition unit, so as to realize multi-level fine recycling of decommissioned wind turbine blade fibers. The impurity removal and recovery bin is located below the electromagnetic vibrating fiber screening unit and is used to receive the impurities separated by the electromagnetic vibrating fiber screening unit. The four-stage recycling bin is located behind the pneumatic sorting unit and is used to receive different grades of decommissioned wind turbine blade fibers after being sorted by the pneumatic sorting unit.

[0008] Furthermore, the feeding hopper of the present invention is equipped with a weighing sensor or weighing module to realize quantitative batch feeding. When the preset feeding weight is reached, the discharge port or unloading mechanism provided below the feeding hopper is opened to realize quantitative batch feeding.

[0009] Furthermore, in the electromagnetic vibration fiber screening unit of the present invention, the screen assembly is preferably a stainless steel screen with an aperture of 1~3mm.

[0010] Furthermore, the surfaces of the main conveyor belt, identification conveyor belt, and sorting conveyor track in the conveying unit are all coated with a wear-resistant polymer coating to reduce wear and adhesion during fiber conveying and to ensure the stability of the fiber conveying process. The conveying speed of the main conveyor belt is 0.1~0.5 m / s, preferably 0.20~0.35 m / s. The conveying speed of the identification conveyor belt is matched with the discharge rhythm of the fiber arrangement unit to ensure that the fibers maintain an orderly interval when entering the photoelectric identification unit. The sorting conveyor track corresponds one-to-one with the position of the pneumatic nozzle array to provide a stable running trajectory and blowing trigger reference for fibers of each grade.

[0011] Furthermore, the fiber dispersion unit of the present invention is a three-section strip-shaped material vibration conveying device, which adopts a three-layer stepped stainless steel grooved toothed plate structure with decreasing height. The spacing between the toothed plates of each layer is set in a gradient decreasing manner, and the groove depth of the toothed plate matches the spacing of the corresponding layer of toothed plates. Each layer of stainless steel grooved toothed plates is inclined, and an amplitude controller is provided on the outer side of the structure. The vibration frequency of the fiber dispersion unit is adjusted in real time by the amplitude controller. The material slides along the inclined toothed plate to the next layer under the action of vibration and gravity. The vibration frequency is set to 10~50Hz. Among them, the toothed plate spacing of the first layer (upper layer) is 8~10mm, and the groove depth is 6~8mm, which is used to accommodate and initially disperse large fiber clusters; the toothed plate spacing of the second layer (middle layer) is 4~6mm, and the groove depth is 4~6mm, which is used to constrain and separate medium fiber clusters; the toothed plate spacing of the third layer (lower layer) is 1~2mm, and the groove depth is 3~4mm, which is used to regularize single fibers or a small number of fibers to form a stable fiber flow.

[0012] Furthermore, in the photoelectric recognition unit, the visual detection module and the spectral detection module are provided with a closed detection cavity to isolate the influence of ambient light fluctuations on the detection results; the visual detection light source is a visible LED surface light source or a line light source, used to provide stable illumination for the industrial camera to improve the extraction accuracy of fiber contour and edge information; the spectral detection light source is a halogen light source or a near-infrared dedicated broadband light source, used to provide stable spectral excitation conditions for the near-infrared spectrometer; the two detection modules are arranged independently at separate workstations and are calibrated for light intensity separately; the data fusion processing unit is also used to perform background processing on the detection results based on the empty track background signal and the standard sample signal. Correction and drift compensation are implemented to improve detection stability. The transmission track is a parallel track structure corresponding to the output track in the upstream fiber arrangement unit, with the number of tracks matching the number of upstream output tracks. Each track has a width of 1-2 mm and is used to transmit dispersed fibers and ensure accurate extraction of fiber information by each detection module. Furthermore, the transmission track is made of a wear-resistant, low-reflection, and optically stable material to reduce background interference and improve detection accuracy. At the visual inspection station, the track detection background is preferably a black matte background; at the spectral inspection station, the track bearing surface is preferably made of a low-reflection material with a stable background response in the near-infrared band. Further, the industrial camera is used to acquire fiber image information and, combined with an image segmentation algorithm, identify fiber length and morphological features. The near-infrared spectrometer has a detection wavelength range of 900-1700 nm and is used to detect the resin residue rate on the fiber surface. The data fusion processing unit is used to match and fuse the length and morphological data acquired by the industrial camera with the resin content data acquired by the near-infrared spectrometer based on fiber running time, position encoding, and transmission speed, and complete fiber grade identification based on the fusion classification model.

[0013] Furthermore, as an optional implementation in the specification, the data fusion processing unit, in addition to completing fiber grade identification, can further predict the physical and mechanical properties of the fiber based on the identification results and generate a performance digital profile for the corresponding batch of fiber; the performance digital profile may include indicators such as predicted tensile strength retention rate and predicted interface performance index, which are used to provide data support for the storage management, quality traceability and subsequent high-value utilization of recycled fibers.

[0014] Furthermore, the pneumatic sorting unit includes a three-stage pneumatic nozzle array and a sorting conveyor track; the sorting conveyor track is located behind the photoelectric recognition unit and is connected to the transmission track in the upstream photoelectric recognition unit; the sorting conveyor track includes a first-stage sorting conveyor belt, a second-stage sorting conveyor belt, and a third-stage sorting conveyor belt arranged sequentially along the material conveying direction, and a set of pneumatic nozzle arrays is set at the end of each sorting conveyor belt, thereby forming a three-stage pneumatic nozzle array; each set of pneumatic nozzle arrays includes multiple nozzles that correspond one-to-one with each transmission track, and the number of nozzles is the same as the number of transmission tracks. If there are 16 tracks, there are 16 nozzles per stage, for a total of 48 nozzles across the three stages. Preferably, the nozzle response delay is no more than 50ms, and the compressed air pressure is no less than 0.6MPa. Further, the first-stage pneumatic nozzle array is used to sort Grade A fibers. When Grade A fibers are conveyed to the end of the first-stage sorting conveyor belt, the first-stage pneumatic nozzle array is triggered, blowing them into the Grade A recycling bin. The second-stage pneumatic nozzle array is used to sort Grade B fibers. When Grade B fibers reach the end of the second-stage sorting conveyor belt, the second-stage pneumatic nozzle array is triggered, blowing them into the Grade B recycling bin. The third-stage pneumatic nozzle array is used to sort Grade C fibers. When Grade C fibers reach the end of the third-stage sorting conveyor belt, the third-stage pneumatic nozzle array is triggered, blowing them into the Grade C recycling bin. Materials not separated by the above nozzle arrays are classified as Grade D materials and fall into the Grade D recycling bin. The Grade D materials include ultra-short fibers with a length below a set lower limit, powdery materials, and rejects that do not meet the Grade A, B, or C criteria. In practical applications, the fiber grading threshold can be adjusted according to the recycling purpose, end-use application scenario, and fiber performance requirements.

[0015] Furthermore, the four-level recycling bins include a level A recycling bin, a level B recycling bin, a level C recycling bin, and a level D recycling bin; weighing modules are respectively installed below the four-level recycling bins and the impurity removal recycling bins for measuring the recycled materials at each level.

[0016] Furthermore, the present invention also provides a high-value intelligent sorting process for decommissioned wind turbine blade fibers, comprising the following steps: (1) Feeding: Weigh and monitor the fiber material of the decommissioned wind turbine blades in the feeding unit. When the weight of the material reaches the set threshold, open the discharge port to feed the material quantitatively. (2) Impurity removal pretreatment: The decommissioned wind turbine blade fiber material conveyed by the feeding unit is sent to the electromagnetic vibration fiber screening unit for impurity removal to remove dust, powder, wood chips and other impurities, so as to avoid impurities clogging the equipment or interfering with subsequent testing. (3) Precision dispersion and arrangement: The impurity-removed fiber material is sequentially fed into the fiber dispersion unit and the fiber arrangement unit. The fiber clumps are gradually disintegrated through the stepped vibrating toothed plate, and the orderly fiber flow is formed through the rhythmic feeding and multi-channel guiding structure, ultimately ensuring that a single fiber or a small number of fibers can be independently identified. (4) Dual-sensor fusion detection and intelligent grading: The arranged fibers are sent into the photoelectric recognition unit and transmitted in multiple tracks at a constant speed; the industrial camera simultaneously acquires fiber images and extracts length information, and the near-infrared spectrometer simultaneously acquires resin content information; the data fusion processing unit performs fusion analysis on the above data and automatically divides the fibers into four grades A to D according to the preset grading threshold or classification model. (5) Pneumatic precision diversion and recycling metering: Based on the grading results output by the photoelectric recognition unit, the corresponding level nozzle is controlled to trigger at the preset position to blow the corresponding level fiber to the corresponding recycling bin, and the metering is completed through the weighing module.

[0017] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention addresses the characteristics of wind turbine blade fibers after mechanical crushing that are prone to agglomeration, entanglement, coupling and accumulation, strong morphological dispersion and difficulty in direct individual detection. By setting up an electromagnetic vibration fiber screening unit, a fiber dispersion unit, a fiber arrangement unit and a conveying unit, a continuous processing process is constructed to transform fiber materials from a scattered mixed state to a dispersed state and then to an intermittent ordered arrangement state. This transforms the originally difficult-to-identify fiber mixture into an ordered fiber flow that can be continuously detected and accurately sorted, providing a reliable premise for subsequent intelligent grading. (2) Through the synergistic effect of the three-section strip material vibration conveying device and the vibration-assisted fiber timing feeding mechanism, the present invention realizes the step-by-step decoupling, combing and regular output of retired wind turbine blade fibers from fiber clumps and fiber clusters to single fibers or a small number of fibers, effectively reducing the overlapping, blocking and overlapping phenomena between fibers, reducing the problems of missed detection, false detection and detection distortion caused by agglomerated conveying, thereby improving the stability and accuracy of fiber detection and grading; (3) This invention employs a dual-channel fusion detection method combining a visual detection module and a spectral detection module. The visual detection module is used to acquire fiber length and morphology information, while the spectral detection module is used to acquire information on resin residue on the fiber surface, thereby achieving synergistic characterization of the physical and chemical characteristics of retired wind turbine blade fibers. Compared with existing methods that sort fibers based solely on a single size parameter or a single detection signal, this invention can more comprehensively reflect the fiber state and improve the accuracy and reliability of fiber grade identification. (4) In this invention, the visual detection module and the spectral detection module are arranged back and forth along the fiber transmission direction, and the multi-source detection data of the same fiber are matched and fused by combining position coding, running time and transmission speed, which ensures the consistency of the detection object; at the same time, through measures such as light shield or closed detection cavity, independent light source arrangement and background correction and drift compensation, the influence of ambient light fluctuation and background noise on the detection results is effectively reduced, further improving the detection accuracy and engineering implementation stability; (5) The present invention adopts a multi-level pneumatic nozzle array corresponding to the sorting and conveying track. After the fiber grade identification is completed, it can perform fixed-point, rapid and continuous spraying and diversion of different grade fibers according to the identification results, so that the detection results and sorting actions are effectively connected. It has the advantages of fast response speed, small contact interference and strong continuous operation capability, thereby improving the efficiency and stability of automated fine sorting of fibers in retired wind turbine blades. (6) The present invention can achieve finer-grained intelligent grading based on key parameters such as the length, morphology and resin residue of retired wind turbine blade fibers, breaking through the limitation of the prior art which only classifies according to a rough size range. This is conducive to improving the adaptability between different grades of recycled fibers and target application scenarios, and improving the high-value utilization level of retired wind turbine blade fibers. (7) The data fusion processing unit of the present invention can generate a digital profile of the performance of the corresponding batch of fibers on the basis of fiber grade identification, providing data support for quality evaluation, storage management, flow traceability and high value-added reuse of recycled fibers, thereby improving the manageability and reliability of recycled fiber application process. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the intelligent sorting system described in this invention.

[0019] Figure 2 This is a side view schematic diagram of the electromagnetic vibration screening unit structure in the intelligent sorting system described in this invention.

[0020] Figure 3 This is a top view schematic diagram of the screen assembly in the electromagnetic vibration screening unit described in this invention.

[0021] Figure 4This is a top view schematic diagram of the fiber dispersion unit in the intelligent sorting system described in this invention.

[0022] Figure 5 This is a schematic diagram of the fiber dispersion unit structure in the intelligent sorting system described in this invention.

[0023] Figure 6 This is a top view schematic diagram of the fiber arrangement unit in the intelligent sorting system described in this invention.

[0024] Figure 7 This is a schematic diagram of a single channel fiber arrangement unit in the intelligent sorting system described in this invention.

[0025] Figure 8 This is a top view schematic diagram of the working state of the photoelectric recognition unit in the intelligent sorting system described in this invention.

[0026] Figure 9 This is a schematic diagram of the operation of a single track in the photoelectric recognition unit of the intelligent sorting system described in this invention.

[0027] Figure 10 This is a side view schematic diagram of the working state of the photoelectric recognition unit in the intelligent sorting system described in this invention.

[0028] Figure 11 This is a side view of the pneumatic sorting unit in the intelligent sorting system described in this invention.

[0029] Figure 12 This is a schematic diagram of a single conveyor belt in the pneumatic sorting unit of the intelligent sorting system described in this invention.

[0030] Figure 13 This is a framework diagram of the intelligent identification and pneumatic sorting control system of the intelligent sorting system described in this invention.

[0031] The attached figures are labeled as follows: 1. Feed hopper; 2. Electromagnetic vibrating screening unit; 201. Electromagnetic excitation unit; 202. Elastic support mechanism; 203. Vibrating table; 204. Screen assembly; 205. Collection hopper; 3. Conveying unit; 301. Main conveyor belt; 302. Identification conveyor belt; 303. Sorting conveyor track; 4. Fiber dispersion unit; 401. First layer stepped stainless steel grooved toothed plate; 402. Second layer stepped stainless steel grooved toothed plate; 403. Third layer stepped stainless steel grooved toothed plate. 5. Fiber Alignment Unit, 501 High-Pressure Material Component, 502 Gradient Inlet, 503 Material-Bearing Vibrating Base Plate, 504 Cyclic Material Pushing Component, 505 Slit Outlet, 6. Photoelectric Recognition Unit, 601 Industrial Camera, 602 Near-Infrared Spectrometer, 603 Transmission Track, 604 Visual Inspection Light Source, 605 Spectral Inspection Light Source, 606 Enclosed Detection Chamber, 607 Data Fusion Processing Unit, 7. Pneumatic Sorting Unit, 701 Pneumatic Nozzle, 8 Impurity Removal and Recovery Chamber, 9. Four-Stage Recovery Chamber. Detailed Implementation

[0032] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0033] Unless otherwise specified in this technical solution, the component model, material name, connection structure, control method, and other features are considered to be common technical features disclosed in the prior art.

[0034] In the description of this invention, it should be understood that the terms "upper", "lower", "vertical", "horizontal", "top", "bottom", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0035] In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified. Furthermore, the terms "installed," "connected," and "linked" should be interpreted broadly; for example, they can refer to a fixed connection, a detachable connection, or an integrated connection; they can refer to a bolted connection or a welded connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0036] To efficiently achieve precise grading of decommissioned wind turbine blades and accurately match the fiber physicochemical parameter requirements of different value-added application scenarios, thereby maximizing resource utilization, this invention provides an intelligent sorting system for high-value fibers in decommissioned wind turbine blades. The structure is described in [reference needed]. Figures 1 to 12 As shown, it includes: 1- Feeding unit (feeding funnel), equipped with weighing sensor or weighing module to achieve quantitative batch feeding; 2-Electromagnetic vibration screening unit, located below the feeding unit, includes electromagnetic excitation unit 201, elastic support mechanism 202, vibration table 203, screen assembly 204, and collection funnel 205, used to remove environmental and structural impurities through high-frequency vibration. 3- Conveyor belt, including main conveyor belt 301, identification conveyor belt 302, and sorting conveyor track 303, adopts wear-resistant polymer coating on the belt surface and is used for continuous conveying of processed fibrous materials; 4- Fiber dispersion unit, including a first layer of stepped stainless steel grooved toothed plate 401, a second layer of stepped stainless steel grooved toothed plate 402 and a third layer of stepped stainless steel grooved toothed plate 403, achieves the gradual disintegration of fiber clusters through multi-level vibration dispersion. 5- Fiber arrangement unit, including high pressure limiting component 501, tapered inlet 502, material bearing vibration base plate 503, rhythmic material feeding component 504 and slit outlet 505, is used to further guide and limit the dispersed fibers and output them in rhythm, so that they enter multiple output tracks corresponding to the downstream transmission track 603 in an intermittent and orderly state. The 6-photoelectric recognition unit includes: 601-industrial camera, 602-near-infrared spectrometer, 603-transmission track, 604-visual inspection light source, 605-spectral detection light source, 606-enclosed detection cavity, and 607-data fusion processing unit; wherein, the industrial camera 601 and the visual inspection light source 604 constitute a visual inspection module, and the near-infrared spectrometer 602 and the spectral detection light source 605 constitute a spectral detection module. The two are arranged back and forth along the fiber transmission direction for continuous detection of the same fiber; the enclosed detection cavity 606 is used to isolate external light interference, and the data fusion processing unit 607 is used to fuse and analyze the detection data and output fiber grade identification results and performance digital files; 7-Airflow sorting unit, including pneumatic nozzle 701, for rapidly blowing and separating fibers of different grades according to the identification results; 8-Impurity removal and recovery bin, used for independent collection of screened impurities; 9-Four-stage recycling bins are used to receive fiber materials from grades A to D respectively, and a weighing module can be installed below each recycling bin to achieve graded measurement.

[0037] Figure 13 The control system framework for photoelectric recognition and pneumatic sorting in Embodiment 1 of the present invention is shown. An industrial camera 601 and a near-infrared spectrometer 602 respectively acquire image and spectral information of the fibers and transmit them to a data fusion processing unit 607. The data fusion processing unit 607 preprocesses, extracts, matches, and fuses the image and spectral data, outputting the grade identification result of the corresponding fiber. The control unit further calculates the trigger time of the corresponding nozzle by combining the fiber's track position, conveying speed, distance from the detection station to the nozzle array, and nozzle response delay, and controls the corresponding pneumatic nozzle 701 to perform a blowing action at a preset position to guide fibers of different grades into the corresponding recycling bins, thereby achieving integrated control of detection, grading, and sorting.

[0038] In this embodiment, the present invention provides a set of fiber grading rules for retired wind turbine blades. It should be noted that the following grading rules, their corresponding recycling destinations and application scenarios are exemplary embodiments in the specification, used to illustrate how the present invention grades fibers based on fiber length and resin content, and are not intended to be necessary limitations on the scope of the claims. In practical applications, the length threshold, resin content threshold, number of grades, and recycling destination can be adjusted according to different recycling purposes, end-use scenarios, and fiber performance requirements.

[0039] Table 1 Example of fiber grading It should be noted that, in this embodiment, the decommissioned wind turbine blade fibers include, but are not limited to, glass fibers and carbon fibers. The length and resin content classification rules shown in Table 1 above are only an example. In practical applications, the length range, resin content threshold, and number of grades can be flexibly set according to different application requirements. In some embodiments, when only fiber length needs to be classified, the spectral detection module can be omitted, only the visual detection module can be retained, and the fibers can be divided into multiple grades according to the length range. For example, they can be divided into four categories according to length: long fibers, medium-length fibers, short fibers, and powdered fibers. ① Long fibers (>5mm, mostly from the "lightly crushed" stage of fine crushing, with good fiber integrity retention; long fibers have less loss of mechanical properties (tensile strength, toughness), and are preferentially used in scenarios requiring material strength; they have high recycling added value and can be used in recycled composite materials. ② Medium-length fibers (1-5mm, from a combination of medium and fine crushing processes, with partial fiber breakage). Medium-length fibers have moderate mechanical properties and cannot meet the requirements of high-end composite materials. Primarily used in scenarios requiring moderate strength and sensitive to cost, this type has the largest recycling volume and widest application, and can be used as a filler in building materials. ③ Short fibers (0.1-1mm, from fine crushing and grinding processes, with severe fiber breakage): Short fibers have significantly reduced mechanical properties and are mainly used for "functional fillers," not relying on strength, but utilizing their physical properties (such as heat resistance, wear resistance, and adsorption). They can be used as fillers in coatings and adhesives. ④ Powdered fibers (<0.1mm, from deep grinding processes, with fibers completely crushed into powder): Powdered fibers have almost no mechanical properties and are mainly used for "functional additives" or low-end fillers. They have the lowest added value from recycling but can achieve "zero waste." The above classification methods also fall within the scope of protection of this invention.

[0040] In this embodiment, the electromagnetic vibrating screening unit 2 pre-treats the input decommissioned wind turbine blade fibers. Specifically, the fibers enter the electromagnetic vibrating screening unit 2 from the feeding unit 1, and under the drive of the electromagnetic excitation unit 201, the screen assembly 204 generates high-frequency micro-amplitude vibration. The screen assembly 204 is preferably made of stainless steel, and the aperture can be configured according to the particle size of the material, preferably 1-3 mm, and exemplarily 2 mm. Through screening, residual dust, attached dirt, sawdust, and other impurities in the fiber mixture can be removed, thereby preventing impurities from clogging subsequent equipment or affecting the optical detection accuracy. The fibers after impurity removal fall into the conveyor belt 3 and are transported to the downstream unit by the main conveyor belt 301; the main conveyor belt 301 is preferably made of wear-resistant polymer coated surface, and the conveying speed is preferably 0.1-0.5 m / s.

[0041] In this embodiment, the main conveyor belt 301 transports the impurity-removed fibers to the fiber dispersion unit 4 and the fiber arrangement unit 5. The fiber dispersion unit 4 (see...) Figure 4 , Figure 5 The structure employs a three-layer stepped stainless steel grooved toothed plate, comprising a first-layer stepped stainless steel grooved toothed plate 401, a second-layer stepped stainless steel grooved toothed plate 402, and a third-layer stepped stainless steel grooved toothed plate 403. The spacing between the toothed plates in each layer is designed to decrease in a gradient: Upper layer (401): coarse tooth spacing (10mm), under high-frequency vibration (30Hz) driven by a piezoelectric ceramic controller, initially loosens and disperses large fiber clumps; Middle layer (402): medium tooth spacing (5mm), under a higher vibration frequency (40Hz), effectively separates medium-sized fiber clusters; Lower layer (403): fine tooth spacing (1.5mm), at the highest vibration frequency (50Hz), outputs mainly single or very few (2-3) fiber flows. This process achieves the gradual loosening and dissociation of large fiber clumps into more dispersed fiber flows under vibration. The fiber arrangement unit 5 (see...) Figure 6 7) Includes a high-pressure limiting component 501, a tapered inlet 502, a material-bearing vibrating base plate 503, a rhythmic material-pushing component 504, and a slit outlet 505, used to further guide and limit the dispersed fibers and output them in rhythm, so that they enter the subsequent multiple parallel transmission tracks 603 in a more orderly state. For some specific implementation methods, please refer to Figures 6 to 7 The fiber arrangement unit 5 can be configured as a multi-channel structure, preferably including 12 to 24 channels, more preferably 16 channels, to improve recognition efficiency and sorting capability.

[0042] In this embodiment, the arranged fibers of the decommissioned wind turbine blades enter the photoelectric identification unit 6 along multiple parallel transmission tracks 603. The photoelectric identification unit 6 (see...) Figures 8-10The system includes an industrial camera 601, a near-infrared spectrometer 602, a transmission track 603, a visual inspection light source 604, a spectral detection light source 605, a closed detection cavity 606, and a data fusion processing unit 607. The industrial camera 601 and the visual inspection light source 604 together constitute a visual inspection module, used to acquire fiber image information and identify fiber length and morphological features. The near-infrared spectrometer 602 and the spectral detection light source 605 together constitute a spectral detection module, used to acquire fiber surface spectral information and detect resin residue rate. The visual inspection module and the spectral detection module are arranged back-to-back along the fiber transmission direction for continuous inspection of the same fiber. The closed detection cavity 606 is located outside the visual inspection module and the spectral detection module to isolate the influence of ambient light fluctuations and stray light on the detection results, thereby improving detection stability and repeatability. The data fusion processing unit 607 is connected to the industrial camera 601 and the near-infrared spectrometer 602 respectively. It is used to match and fuse the length and morphology data acquired by the industrial camera 601 with the resin content data acquired by the near-infrared spectrometer 602 based on the fiber running time, position encoding and transmission speed, and output the fiber grade identification result according to the preset grading rules or classification model.

[0043] In some specific embodiments, the transmission track 603 is a parallel track structure corresponding to the upstream output track, with the number of tracks matching the number of upstream output tracks, and the width of a single track can be 1~2 mm. The transmission track 603 is made of a wear-resistant, low-reflection, and optically stable material to reduce background interference and improve detection accuracy; wherein, in the visual inspection station, the track detection background is preferably a black matte background; in the spectral inspection station, the track bearing surface is preferably made of a low-reflection material with a stable background response in the near-infrared band.

[0044] In some specific embodiments, the visual inspection light source 604 is a visible light surface light source or a line light source, used to provide stable illumination for the industrial camera 601 to improve the extraction accuracy of fiber contour and edge information; the spectral inspection light source 605 is a halogen light source or a near-infrared dedicated broadband light source, used to provide stable spectral excitation conditions for the near-infrared spectrometer 602. The visual inspection light source 604 and the spectral inspection light source 605 are arranged independently and can be calibrated for light intensity separately. The data fusion processing unit 607 can also perform background correction and drift compensation on the detection results based on the empty track background signal and the standard sample signal to further improve the detection accuracy.

[0045] In some specific embodiments, the data fusion processing unit 607 may employ a machine learning classification model, preferably a neural network classification model. In addition to outputting fiber grade, the data fusion processing unit 607 can further establish a mapping relationship between fiber length, resin content, and mechanical properties, generating a performance digital profile for the corresponding batch of fibers. The performance digital profile may include indicators such as predicted tensile strength retention rate and predicted interface performance index, providing data support for the storage management, quality traceability, and subsequent high-value utilization of recycled fibers.

[0046] In this embodiment, the fibers identified by the photoelectric recognition unit 6 continue to enter the pneumatic sorting unit 7. The pneumatic sorting unit 7 (see...) Figure 11 , Figure 12 The system includes a multi-stage pneumatic nozzle array and a sorting conveyor track 303. Each stage of the pneumatic nozzle array is used to sort fibers of different grades. As an example, the first-stage pneumatic nozzle array can be used to sort grade A fibers, the second-stage pneumatic nozzle array can be used to sort grade B fibers, and the third-stage pneumatic nozzle array can be used to sort grade C fibers. Materials not separated by the nozzles are classified as grade D materials and enter the grade D recycling bin. The grade D materials typically include short fibers with high resin content, ultra-short fibers, powdered materials, and other materials that do not meet the grade A, B, or C criteria.

[0047] In this embodiment, the A, B, C, and D grade recycling bins together constitute a four-stage recycling bin 9, and the impurity removal recycling bin 8 is used to collect and screen impurities. In some embodiments, a weighing module can be installed below each recycling bin to realize the measurement of recycled materials according to grade or category.

[0048] The second aspect of this invention provides a high-value intelligent sorting process for decommissioned wind turbine blade fibers, which is implemented using the aforementioned intelligent fine grading system for decommissioned wind turbine blade fibers, and includes the following steps: (1) Feeding: Weigh and monitor the fiber material of the decommissioned wind turbine blades in the feeding unit. When the weight of the material reaches the set threshold, open the discharge port to feed the material quantitatively. (2) Impurity removal pretreatment: The decommissioned wind turbine blade fiber material conveyed by the feeding unit is sent to the electromagnetic vibration fiber screening unit for impurity removal to remove dust, powder, wood chips and other impurities, so as to avoid impurities clogging the equipment or interfering with subsequent testing. (3) Precision dispersion and arrangement: The impurity-removed fiber material is sequentially fed into the fiber dispersion unit and the fiber arrangement unit. The fiber clumps are gradually disintegrated through the stepped vibrating toothed plate, and the orderly fiber flow is formed through the rhythmic feeding and multi-channel guiding structure, ultimately ensuring that a single fiber or a small number of fibers can be independently identified. (4) Dual-sensor fusion detection and intelligent grading: The arranged fibers are sent into the photoelectric recognition unit and transmitted in multiple tracks at a constant speed; the industrial camera simultaneously acquires fiber images and extracts length information, and the near-infrared spectrometer simultaneously acquires resin content information; the data fusion processing unit performs fusion analysis on the above data and automatically divides the fibers into four grades A to D according to the preset grading threshold or classification model. (5) Pneumatic precision diversion and recycling metering: Based on the grading results output by the photoelectric recognition unit, the corresponding level nozzle is controlled to trigger at the preset position to blow the corresponding level fiber to the corresponding recycling bin, and the metering is completed through the weighing module.

[0049] Example 1 This embodiment was conducted at a wind turbine blade recycling base. The following method was used: Figure 1 The intelligent sorting system for high-value processing of decommissioned wind turbine blade fibers shown herein is used to classify and sort the mixture of mechanically crushed decommissioned wind turbine blade fibers. The processed material is mainly glass fiber.

[0050] In this embodiment, the parameters of the fiber dispersion unit 4 and the fiber arrangement unit 5 are set as follows: The fiber dispersion unit 4 is a three-stage strip-shaped material vibrating conveyor. The upper stepped stainless steel grooved toothed plate 401 has a tooth spacing of 10 mm, a vibration frequency of 30 Hz, and an amplitude of 5 mm; the middle stepped stainless steel grooved toothed plate 402 has a tooth spacing of 5 mm, a vibration frequency of 40 Hz, and an amplitude of 3 mm; the lower stepped stainless steel grooved toothed plate 403 has a tooth spacing of 1.5 mm, a vibration frequency of 50 Hz, and an amplitude of 1 mm. Through the three-layer gradient decreasing structure, the fiber clusters are gradually loosened and disintegrated.

[0051] The fiber arrangement unit 5 is a vibration-assisted fiber timing feeding mechanism. Its output end is correspondingly set with 16 parallel transmission tracks 603. The track spacing is 2mm and the single track width is 1.5mm, so as to achieve orderly arrangement and stable output of fibers.

[0052] In this embodiment, the photoelectric recognition unit 6 is located downstream of the fiber arrangement unit 5, and mainly includes: optical detection components: ① an industrial high-speed CMOS camera 601 (resolution 2448×2048 pixels, frame rate 250FPS), with a ring-shaped LED white light source (illuminance ≥1000 lux) for acquiring fiber image information; ② a near-infrared spectrometer 602 (wavelength range 900–1700nm, sampling interval 2nm) 2 for acquiring fiber surface spectral information; ③ a data processing unit 607: equipped with a BP neural network classification model, the training dataset contains 100,000 sets of calibrated fiber samples (length error ±0.1mm, resin content error ±0.5%), which performs fusion analysis on fiber length and resin content and outputs grading results.

[0053] In this embodiment, the airflow sorting unit 7 includes a three-stage pneumatic nozzle array, each stage of which is equipped with 16 independent nozzles. The compressed air pressure is 0.6 MPa, and the nozzle response time is no more than 50 ms. It is used to sort and recycle fibers of different grades.

[0054] Using the above system, according to Figure 1 The process shown involves the following steps: processing the finely crushed mixture of decommissioned blade fibers (glass fiber content >95%): Table 2 Workflow of this embodiment The output fibers from each recycling bin were sampled and tested (1 kg per group), and the results are shown in Table 3: Table 3. Results of fiber testing at each grade after sorting As shown in Table 3, the system of the present invention can effectively classify the fibers of retired wind turbine blades according to fiber length and resin content, so that different grades of fibers can be adapted to different value-added application scenarios, thereby improving the resource utilization efficiency and application value of recycled fibers.

Claims

1. A high-value intelligent sorting system for decommissioned wind turbine blade fibers, characterized in that, include: The feeding unit is a feeding funnel (1) with a feeding port; An electromagnetic vibrating fiber screening unit (2) is placed below the feed hopper (1). The electromagnetic vibrating fiber screening unit (2) includes an electromagnetic excitation unit (201), an elastic support mechanism (202), a vibrating table (203), a screen assembly (204), and a collection hopper (205). The screen assembly (204) is embedded in the vibrating table (203) to form an integrated screening structure. The integrated structure is inclined, with the high end being the feed end and the low end being the discharge end. At least one set of elastic support mechanisms (202) is provided at the feed end and the discharge end, respectively. The collection hopper (205) is placed below the integrated structure. The screen assembly (204) adopts a fine mesh structure. The electromagnetic excitation unit (201) is fixed below or behind the high end of the vibrating table (203). The conveying unit (3) is respectively set between the electromagnetic vibration fiber screening unit (2) and the fiber dispersion unit (4), between the fiber arrangement unit (5) and the photoelectric recognition unit (6), and within the pneumatic sorting unit (7), including a main conveyor belt (301), an identification conveyor belt (302), and a sorting conveyor track (303); the main conveyor belt (301) is set between the electromagnetic vibration fiber screening unit (2) and the fiber dispersion unit (4), the identification conveyor belt (302) is set between the discharge end of the fiber arrangement unit (5) and the photoelectric recognition unit (6); the sorting conveyor track (303) is set within the pneumatic sorting unit (7) and located behind the photoelectric recognition unit (6); The fiber dispersion unit (4) is a three-section strip material vibration conveying device. The feed end of the main conveyor belt (301) is located below the discharge end of the electromagnetic vibration fiber screening unit (2). The other end of the main conveyor belt (301) is the discharge end, which is connected to the feed port of the fiber dispersion unit (4). The fiber arrangement unit (5) consists of 12 to 24 vibration-assisted fiber sequential feeding mechanisms; each vibration-assisted fiber sequential feeding mechanism includes a material-bearing vibration base plate (503), a high-pressure limiting component (501), a gradually narrowing inlet (502), a rhythmic feeding component (504), and a slit outlet (505); wherein, the material-bearing vibration base plate (503) is used to receive and transport the fiber material after the previous stage dispersion, and the high-pressure limiting component (501) is set above the feeding end of the material-bearing vibration base plate (503); The tapered inlet (502) is located on the side in front of the high-pressure limiting material component (501) away from the feed end; the rhythmic feeding component (504) is located outside the discharge end of the material-bearing vibrating base plate (503) and reciprocates or pushes intermittently according to a preset frequency; the slit discharge port (505) is connected to the discharge end of the material-bearing vibrating base plate (503) and is used to control the fibers to be released in batches according to the rhythm, and finally evenly distributed into multiple parallel and spaced output tracks; the number of output tracks is 12 to 24. The photoelectric recognition unit (6) is located at the end of the recognition conveyor belt (302) behind the discharge end of the fiber arrangement unit (5). The photoelectric recognition unit (6) includes a vision detection module, a spectral detection module, a transmission track (603) and a data fusion processing unit (607). The vision detection module and the spectral detection module are arranged back and forth along the fiber transmission direction. The vision detection module is equipped with an industrial camera (601) and a vision detection light source (604). The spectral detection module is equipped with a near-infrared spectrometer (602) and a spectral detection light source (605). The data fusion processing unit (607) communicates with both detection modules and has a built-in fusion classification model. It performs fusion analysis on the detection data of the two modules and automatically classifies the fibers into four levels, A to D, according to the preset grading threshold. The pneumatic sorting unit (7) is located behind the photoelectric recognition unit (6); The impurity removal and recovery bin (8) is located below the electromagnetic vibration fiber screening unit (2); The fourth-level recovery bin (9) is located behind the pneumatic sorting unit (7).

2. The intelligent sorting system for high-value fiber recycling of decommissioned wind turbine blades according to claim 1, characterized in that, The feed hopper is equipped with a weighing sensor or weighing module.

3. The intelligent sorting system for high-value fiber recycling of decommissioned wind turbine blades according to claim 1, characterized in that, In the electromagnetic vibration fiber screening unit, the screen assembly uses a stainless steel screen with a aperture of 1~3mm.

4. The intelligent sorting system for high-value fiber recycling of decommissioned wind turbine blades according to claim 1, characterized in that, The surfaces of the main conveyor belt, identification conveyor belt, and sorting conveyor track in the conveying unit are all coated with a wear-resistant polymer coating. The conveying speed of the main conveyor belt is 0.1~0.5 m / s; The conveying speed of the identification conveyor belt is matched with the discharge cycle of the fiber arrangement unit; The sorting and conveying track corresponds one-to-one with the position of the pneumatic nozzle array.

5. The intelligent sorting system for high-value fiber recycling of decommissioned wind turbine blades according to claim 1, characterized in that, The fiber dispersion unit is a three-section strip material vibration conveying device, which adopts a three-layer height-decreasing stepped stainless steel groove toothed plate structure. The spacing between each layer of toothed plates is set in a gradient decreasing manner, and the groove depth of the toothed plate matches the spacing of the corresponding layer of toothed plates. Each layer of stainless steel groove toothed plate is inclined, and an amplitude controller is provided on the outside of its structure. The vibration frequency set by the amplitude controller is 10~50Hz; The first layer of stepped stainless steel grooved toothed plates has a tooth spacing of 8~10mm and a groove depth of 6~8mm; the second layer of stepped stainless steel grooved toothed plates has a tooth spacing of 4~6mm and a groove depth of 4~6mm; the third layer of stepped stainless steel grooved toothed plates has a tooth spacing of 1~2mm and a groove depth of 3~4mm.

6. The intelligent sorting system for high-value fiber recycling of decommissioned wind turbine blades according to claim 1, characterized in that, The preset frequency in the fiber arrangement unit is 0.5~5Hz, and the specific frequency is determined according to the equipment cycle time.

7. The intelligent sorting system for high-value fiber recycling of decommissioned wind turbine blades according to claim 1, characterized in that, In the photoelectric recognition unit, the visual detection module and the spectral detection module are provided with a closed detection cavity on their outer sides. The two detection modules are arranged independently at their respective workstations and are calibrated for light intensity separately. The visual inspection light source is a visible LED surface light source or a line light source; The spectral detection light source is a halogen light source or a dedicated broadband light source for near-infrared light, and the detection wavelength range of the near-infrared spectrometer is 900~1700nm; The data fusion processing unit has a built-in classification model that is a machine learning classification model, which is used to perform background correction and drift compensation on the detection results based on the empty track background signal and the standard sample signal; the transmission track is a parallel track structure corresponding to the output track in the upstream fiber arrangement unit, the number of tracks is the same as the number of upstream output tracks, and the width of a single track is 1~2mm. The transmission track is made of a wear-resistant, low-reflection, and optically stable material. At the visual inspection station, the track inspection background is a black matte background, and at the spectral inspection station, the track bearing surface is made of a low-reflection material with a stable background response in the near-infrared band.

8. The intelligent sorting system for high-value fiber recycling of decommissioned wind turbine blades according to claim 1, characterized in that, The pneumatic sorting unit includes a three-stage pneumatic nozzle array and a sorting conveyor track. The sorting conveyor track is located behind the photoelectric recognition unit and connects to the conveyor track in the upstream photoelectric recognition unit. The sorting conveyor track includes a first-stage sorting conveyor belt, a second-stage sorting conveyor belt, and a third-stage sorting conveyor belt arranged sequentially along the material conveying direction. Each stage of the sorting conveyor belt has a set of pneumatic nozzle arrays at its end, thus forming a three-stage pneumatic nozzle array. Each set of pneumatic nozzle arrays includes multiple nozzles that correspond one-to-one with each conveyor track, and the number of nozzles is the same as the number of conveyor tracks. The nozzle response delay is no more than 50ms, and the compressed air pressure is no less than 0.6MPa; The three-stage pneumatic nozzle array is respectively matched with the corresponding recycling bin in the four-stage recycling bin, and is used to spray and separate the fiber material step by step according to the grade identification result output by the photoelectric identification unit. The first-stage pneumatic nozzle array is used to sort grade A fibers, the second-stage pneumatic nozzle array is used to sort grade B fibers, the third-stage pneumatic nozzle array is used to sort grade C fibers, and the material that is not blown and separated by the above nozzle arrays is classified as grade D material and falls into the grade D recycling bin.

9. The intelligent sorting system for high-value fiber processing of decommissioned wind turbine blades according to claim 1, characterized in that, The four-level recycling bins include Class A, Class B, Class C, and Class D recycling bins; Weighing modules are installed below the four-level recycling bins and the impurity removal recycling bins, respectively.

10. A high-value intelligent sorting process for decommissioned wind turbine blade fibers based on the intelligent sorting system according to any one of claims 1 to 9, characterized in that, Includes the following steps: (1) Feeding: Weigh and monitor the fiber material of the decommissioned wind turbine blades in the feeding unit. When the weight of the material reaches the set threshold, open the discharge port to feed the material quantitatively. (2) Impurity removal pretreatment: The decommissioned wind turbine blade fiber material conveyed by the feeding unit is sent to the electromagnetic vibration fiber screening unit for impurity removal, removing dust, powder, wood chips and other impurities; (3) Precision dispersion and arrangement: The impurity-removed fiber material is sequentially fed into the fiber dispersion unit and the fiber arrangement unit. The fiber clumps are gradually disintegrated through the stepped vibrating toothed plate, and the orderly fiber flow is formed through the rhythmic feeding and multi-channel guiding structure, ultimately ensuring that a single fiber or a small number of fibers can be independently identified. (4) Dual-sensor fusion detection and intelligent grading: The arranged fibers are sent into the photoelectric recognition unit and transmitted in multiple tracks at a constant speed; the industrial camera simultaneously acquires fiber images and extracts length information, and the near-infrared spectrometer simultaneously acquires resin content information; The data fusion processing unit performs fusion analysis on the above data and automatically classifies the fibers into four levels, A to D, based on preset grading thresholds or classification models. (5) Pneumatic precision diversion and recycling metering: Based on the grading results output by the photoelectric recognition unit, the corresponding level nozzle is controlled to trigger at the preset position to blow the corresponding level fiber to the corresponding recycling bin, and the metering is completed through the weighing module.

Images