Thermosetting plastic crushing and recycling processing method and system

Abstract

The invention relates to a thermosetting plastic crushing and recycling processing method and system.The method comprises the steps that U1, in the thermosetting plastic crushing and recycling process, data information of the discharging speed of thermosetting plastic is collected, and data information of the rotating speed of a roller of a magnetic separator and data information of the rotating speed of a belt of an eddy current sorting machine are obtained in real time; and U2, on the basis of the data information of the rotating speed of the roller of the magnetic separator, the data information of the rotating speed of the belt of the eddy current separator and the data information of the blanking speed of the thermosetting plastic, predicting a screening result of the thermosetting plastic by adopting an improved two-way recurrent neural network fused attention multivariable regression prediction algorithm. And obtaining data information of the predicted screening result of the thermosetting plastic. According to the device, metal particles or nonferrous metal powder in the thermosetting plastic can be accurately screened, high-quality completion of a subsequent process is guaranteed, and the recycling efficiency and the recycling quality of the thermosetting plastic are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of thermosetting plastic crushing and recycling, and in particular to a method and system for thermosetting plastic crushing and recycling processing. Background Art

[0002] In today's era of abundant materials but increasingly scarce resources, the recycling and reuse of waste materials are not only a manifestation of environmental responsibility but also a key link in promoting sustainable development. The concept of turning waste into treasure is gradually taking root in people's hearts and becoming the direction of joint efforts from all sectors of society.

[0003] Among them, for the recycling of thermosetting plastics, during the recycling process, it is first necessary to screen metal particles or non-ferrous metal powders in the plastics. If the screening is not thorough enough, it will lead to a high impurity content in subsequent processes, which is not conducive to full recycling and reuse. Not only will the recycling quality decrease, but also the process needs to be reprocessed, wasting resources and having low efficiency. Therefore, how to accurately screen and recycle thermosetting plastics has become an urgent problem for us to solve. Summary of the Invention

[0004] In view of the above problems, the present invention provides a method and system for thermosetting plastic crushing and recycling processing, which can not only accurately screen metal particles or non-ferrous metal powders in thermosetting plastics, ensure the high-quality completion of subsequent processes, but also improve the recycling efficiency and quality of thermosetting plastics.

[0005] In order to achieve the above object and other related objects, the technical solutions provided by the present invention are as follows:

[0006] A method for thermosetting plastic crushing and recycling processing, the method comprising:

[0007] U1. During the thermosetting plastic crushing and recycling process, collect data information on the falling speed of thermosetting plastics, and obtain data information on the drum rotation speed of the magnetic separator and data information on the belt rotation speed of the eddy current separator in real time;

[0008] U2. Based on the data information on the drum rotation speed of the magnetic separator, the data information on the belt rotation speed of the eddy current separator, and the data information on the falling speed of thermosetting plastics, use an improved bidirectional recurrent neural network fusion attention multi-variable regression prediction algorithm to predict the screening result of thermosetting plastics, and obtain data information on the predicted screening result of thermosetting plastics;

[0009] U3. Based on the data information on the predicted screening result of thermosetting plastics, use an improved triangular topology aggregation optimization algorithm to optimize the screening result of thermosetting plastics, and obtain data information on the optimized screening result of thermosetting plastics;

[0010] U4. Based on the data information of the screening results of the optimized thermosetting plastics, construct an adaptive feedback adjustment function F to control and adjust the blanking speed, the rotation speed of the magnetic separator drum, and the belt speed of the eddy current separator, and output the data information of the adjusted blanking speed, the rotation speed of the magnetic separator drum, and the belt speed of the eddy current separator.

[0011] Further, the adaptive feedback adjustment function F is

[0012]

[0013] where f1 is the adaptive feedback adjustment control function of the blanking speed, and α 11 、α 12 and α 13 are the dynamic adjustment factors of the blanking speed, f2 is the adaptive feedback adjustment control function of the rotation speed of the magnetic separator drum, and α 21 、α 22 and α 23 are the dynamic adjustment factors of the rotation speed of the magnetic separator drum, f3 is the adaptive feedback adjustment control function of the belt speed of the eddy current separator, and α 31 、α 32 and α 33 are the dynamic adjustment factors of the belt speed of the eddy current separator, and x is the data information of the screening results of the optimized thermosetting plastics.

[0014] Further, the dynamic adjustment factors α 11 、α 12 and α 13 of the blanking speed are

[0015]

[0016] The dynamic adjustment factors α 21 、α 22 and α 23 of the rotation speed of the magnetic separator drum are

[0017]

[0018] The dynamic adjustment factors α 31 、α 32 and α 33 of the belt speed of the eddy current separator are

[0019]

[0020] where x is the data information of the screening results of the optimized thermosetting plastics.

[0021] Further, the dynamic adjustment factors α11 and the constraint conditions of α 12 and α 13 are as follows:

[0022]

[0023] The dynamic adjustment factor α of the drum rotation speed of the magnetic separator 21 and α 22 and α 23 The constraint function g is as follows:

[0024] The dynamic adjustment factor α of the belt rotation speed of the eddy current separator 31 and α 32 and α 33 The constraint function h is as follows:

[0025]

[0026] Among them, the value range of the constraint function g is (0, 1), and the value range of the constraint function h is (1, 2).

[0027] Furthermore, in step U2, the prediction of the screening results of thermosetting plastics by using the improved multivariate regression prediction algorithm that fuses attention with a bidirectional recurrent neural network includes:

[0028] U21. Based on the data information of the drum rotation speed of the magnetic separator, the data information of the belt rotation speed of the eddy current separator, and the data information of the falling speed of the thermosetting plastics, construct the preliminary processing sequence function Q of the thermosetting plastics,

[0029]

[0030] Among them, y1 is the data information of the drum rotation speed of the magnetic separator, y2 is the data information of the belt rotation speed of the eddy current separator, y3 is the data information of the falling speed of the thermosetting plastics, and β1, β2, and β3 are weight coefficients, which characterize the drum rotation speed of the magnetic separator, the belt rotation speed of the eddy current separator, and the falling speed, and obtain the data information of the preliminary processing sequence of the thermosetting plastics;

[0031] U22. Input the data information of the preliminary processing sequence of the thermosetting plastics into the bidirectional recurrent neural network model for training and learning, predict the preliminary screening results of the thermosetting plastics, and obtain the data information of the preliminary screening results of the thermosetting plastics;

[0032] U23. Based on the data information of the preliminary screening results of the thermosetting plastics, construct the multivariate regression prediction function W of the attention mechanism,

[0033]

[0034] Among them, z is the data information of the preliminary screening result of the thermosetting plastic, and γ1, γ2, and γ3 are attention mechanism factors, which are used to predict the screening result of the thermosetting plastic, and the data information of the screening result of the predicted thermosetting plastic is obtained.

[0035] Further, the attention mechanism factors γ1, γ2, and γ3 are

[0036]

[0037] Among them, z is the data information of the preliminary screening result of the thermosetting plastic.

[0038] Further, in step U3, the optimization of the screening result of the thermosetting plastic by using the improved triangular topology aggregation optimization algorithm includes:

[0039] U31. Based on the data information of the screening result of the predicted thermosetting plastic, initialize the population, determine the population size and the variable dimension D, and obtain the data information of the initialized population.

[0040] U32. Based on the data information of the initialized population, establish the triangular topology unit function S,

[0041]

[0042] Among them, r is the data information of the initialized population, and η1, η2, and η3 are random numbers, which are used to characterize the triangular topology unit of the population, and the data information of the triangular topology unit of the population is obtained.

[0043] U33. Based on the data information of the triangular topology unit of the population, establish the objective optimization function G,

[0044]

[0045] Among them, a is the data information of the triangular topology unit of the population, p1(a) is the global aggregation function of the population, p2(a) is the local aggregation function of the population, and λ1, λ2, and λ3 are the weight coefficients of the objective optimization, which are used to optimize the screening result of the thermosetting plastic, and the data information of the optimized screening result of the thermosetting plastic is obtained.

[0046] Further, the global aggregation function p1(a) of the population is

[0047]

[0048] The local aggregation function p2(a) of the population is

[0049]

[0050] Among them, a is the data information of the triangular topological unit of the population.

[0051] Furthermore, the method further includes:

[0052] U5. After being screened by eddy current and entering the liquid nitrogen tank for cold extraction, and then through primary crushing and secondary crushing, it enters the waste collection tank.

[0053] To achieve the above object and other related objects, the present invention also provides a system for implementing the thermosetting plastic crushing and recycling processing method described in any one of the above, and the system includes:

[0054] A data acquisition module, which is used to collect the data information of the falling speed of thermosetting plastics, and to obtain in real time the data information of the rotating speed of the magnetic separator drum and the data information of the belt speed of the eddy current separator;

[0055] A prediction module for the screening result of thermosetting plastics, which is connected to the data acquisition module, and is used to predict the screening result of thermosetting plastics by using an improved multivariate regression prediction algorithm that combines a bidirectional recurrent neural network with attention, so as to obtain the data information of the predicted screening result of thermosetting plastics;

[0056] An optimization module for the screening result of thermosetting plastics, which is connected to the prediction module for the screening result of thermosetting plastics, and is used to optimize the screening result of thermosetting plastics by using an improved triangular topological aggregation optimization algorithm, so as to obtain the data information of the optimized screening result of thermosetting plastics;

[0057] An adaptive feedback control and adjustment module, which is connected to the optimization module for the screening result of thermosetting plastics, and is used to construct an adaptive feedback adjustment function F to control and adjust the falling speed, the rotating speed of the magnetic separator drum and the belt speed of the eddy current separator, and output the data information of the adjusted falling speed, the rotating speed of the magnetic separator drum and the belt speed of the eddy current separator.

[0058] The present invention has the following positive effects:

[0059] 1. By using an improved multivariate regression prediction algorithm that combines a bidirectional recurrent neural network with attention to predict the screening result of thermosetting plastics, and combining with an improved triangular topological aggregation optimization algorithm to optimize the screening result of thermosetting plastics, the present invention can not only accurately screen metal particles or non-ferrous metal powders in thermosetting plastics to ensure the high-quality completion of subsequent processes, but also does not require manual participation during the screening process, reducing the labor intensity of workers, thereby improving the recycling efficiency of thermosetting plastics.

[0060] 2. The present invention controls and adjusts the feeding speed, the rotating speed of the drum of the magnetic separator, and the belt speed of the eddy current separator by constructing an adaptive feedback adjustment function F. It can not only perform feedback adaptive control and adjustment on the feeding speed, the magnetic separator, and the eddy current separator according to the predicted screening results, improve the screening accuracy, but also improve the recycling efficiency and quality of thermosetting plastics. Description of the Drawings

[0061] Figure 1 It is a schematic flow chart of the method of the present invention;

[0062] Figure 2 It is a schematic flow chart of the improved multi-variable regression prediction algorithm of the bidirectional recurrent neural network integrating attention of the present invention;

[0063] Figure 3 It is a schematic flow chart of the improved triangular topology aggregation optimization algorithm of the present invention;

[0064] Figure 4 It is a schematic diagram of the system framework of the present invention;

[0065] Figure 5 It is a schematic diagram of the overall structure of the processing of the present invention;

[0066] Figure 6 It is a schematic diagram of the overall processing technological process of the present invention;

[0067] Figure 7 It is a schematic diagram of the structure of the feeding device of the present invention;

[0068] Figure 8 It is a schematic diagram of the structure of the magnetic separator of the present invention;

[0069] Figure 9 It is a schematic diagram of the structure of the eddy current separator of the present invention;

[0070] Figure 10 It is a schematic diagram of the structure of the drying hood of the present invention.

[0071] Explanation of the reference numerals in the figure: 1 - feeding device, 11 - cutting blade, 12 - belt conveyor line, 2 - magnetic separator, 3 - eddy current separator, 4 - ring rail conveying trolley, 5 - liquid nitrogen cold extraction device, 6 - primary crushing device, 7 - negative pressure conveying device, 8 - secondary crushing device, 9 - waste collection tank, 10 - drying hood. Detailed Embodiments

[0072] The following describes exemplary embodiments of the present disclosure with reference to the accompanying drawings. Various details of the embodiments of the present disclosure are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, descriptions of well-known functions and structures are omitted in the following description for clarity and conciseness.

[0073] Embodiment 1: As Figure 1 shown, a method for crushing and recycling thermosetting plastics, the method comprising:

[0074] U1. During the crushing and recycling of thermosetting plastics, collect data information on the falling speed of thermosetting plastics, and obtain in real time data information on the rotational speed of the magnetic separator drum and data information on the belt speed of the eddy current separator;

[0075] U2. Based on the data information on the rotational speed of the magnetic separator drum, the data information on the belt speed of the eddy current separator, and the data information on the falling speed of thermosetting plastics, use an improved multi-variable regression prediction algorithm that combines a bidirectional recurrent neural network with attention to predict the screening result of thermosetting plastics, and obtain data information on the predicted screening result of thermosetting plastics;

[0076] U3. Based on the data information on the predicted screening result of thermosetting plastics, use an improved triangular topology aggregation optimization algorithm to optimize the screening result of thermosetting plastics, and obtain data information on the optimized screening result of thermosetting plastics;

[0077] U4. Based on the data information on the optimized screening result of thermosetting plastics, construct an adaptive feedback adjustment function F to control and adjust the falling speed, the rotational speed of the magnetic separator drum, and the belt speed of the eddy current separator, and output data information on the adjusted falling speed, the rotational speed of the magnetic separator drum, and the belt speed of the eddy current separator.

[0078] In this embodiment, the adaptive feedback adjustment function F is

[0079]

[0080] where f1 is the adaptive feedback adjustment control function for the falling speed, α 11 、α 12 and α 13 are dynamic adjustment factors for the falling speed, f2 is the adaptive feedback adjustment control function for the rotational speed of the magnetic separator drum, α 21 、α 22 and α 23α is the dynamic adjustment factor of the drum speed of the magnetic separator, f3 is the adaptive feedback regulation control function of the belt speed of the eddy current separator, α 31 、α 32 and α 33 are the dynamic adjustment factors of the belt speed of the eddy current separator, and x is the data information of the screening result of the optimized thermosetting plastic.

[0081] In this embodiment, the dynamic adjustment factors α 11 、α 12 and α 13 for the

[0082]

[0083] The dynamic adjustment factors α 21 、α 22 and α 23 for the

[0084]

[0085] The dynamic adjustment factors α 31 、α 32 and α 33 for the

[0086]

[0087] where x is the data information of the screening result of the optimized thermosetting plastic.

[0088] In this embodiment, the constraint conditions for the dynamic adjustment factors α 11 、α 12 and α 13 are

[0089]

[0090] The constraint function g for the dynamic adjustment factors α 21 、α 22 and α 23 is

[0091] The constraint function h for the dynamic adjustment factors α 31 、α 32 and α 33 is

[0092]

[0093] Among them, the value range of the constraint function g is (0, 1), and the value range of the constraint function h is (1, 2).

[0094] In this embodiment, as Figure 5 or Figure 6 or Figure 7 or Figure 8 or Figure 9 or Figure 10 shown, the discharge port of the blanking device 1 is connected to the feed port of the magnetic separator 2, the discharge port of the magnetic separator 2 is connected to the feed port of the eddy current separator 3, a ring rail conveying trolley 4 is arranged on one side of the eddy current separator 3, a liquid nitrogen cold extraction device 5 is arranged below the ring rail conveying trolley 4, a primary crushing device 6 is arranged on one side of the liquid nitrogen cold extraction device 5, the discharge port of the primary crushing device 6 is connected to the feed port of the negative pressure conveying device 7, the discharge port of the negative pressure conveying device 7 is connected to the feed port of the secondary crushing device 8, the discharge port of the secondary crushing device 8 is connected to the feed port of the waste collection tank 9. The bagged waste residue is lifted and conveyed above the blanking port, the waste residue bag is cut open by the blade 11, the waste residue enters the blanking port, is received by the funnel onto the belt conveyor 12, and then conveyed backward. After the paint stripping, the waste residue is all large particle substances, impurities are removed by magnetic separation and eddy current separation, and then cold extraction treatment is carried out with liquid nitrogen (to make the waste residue in a low-temperature embrittlement state), then primary crushing is carried out to crush the large particle waste residue into fine particle powder, which is conveyed through the vacuum conveying pipe, and then secondary crushing is carried out to grind the waste residue (200 - 300 mesh fine powder), and finally it is collected and recycled with the receiving bucket. A drying cover 10 is arranged outside the primary crushing device 6 and the secondary crushing device 8. The method further includes:

[0095] U5. After being screened by the eddy current and entering the liquid nitrogen tank for cold extraction, and then after primary crushing and secondary crushing, it enters the waste collection tank.

[0096] In this embodiment, as Figure 4 shown, the present invention also provides a system for implementing the thermosetting plastic crushing and recycling processing method described in any one of the above. The system includes: a data acquisition module, which is used to collect the data information of the blanking speed of the thermosetting plastic, and obtain the data information of the roller rotation speed of the magnetic separator and the belt rotation speed of the eddy current separator in real time;

[0097] a prediction module for the screening result of the thermosetting plastic, connected to the data acquisition module, which is used to predict the screening result of the thermosetting plastic by using an improved multivariate regression prediction algorithm that combines attention with a bidirectional recurrent neural network, and obtain the data information of the predicted screening result of the thermosetting plastic;

[0098] An optimization module for the screening results of thermosetting plastics, connected to the prediction module for the screening results of thermosetting plastics, is used to optimize the screening results of thermosetting plastics by adopting an improved triangular topology polymerization optimization algorithm, and obtain the data information of the optimized screening results of thermosetting plastics;

[0099] An adaptive feedback control and regulation module, connected to the optimization module for the screening results of thermosetting plastics, is used to construct an adaptive feedback regulation function F to control and regulate the falling speed, the rotating speed of the magnetic separator drum, and the belt speed of the eddy current separator, and output the data information of the regulated falling speed, the rotating speed of the magnetic separator drum, and the belt speed of the eddy current separator.

[0100] Example 2: On the basis of a method for crushing and recycling thermosetting plastics in Example 1, the present invention will be further described and illustrated below.

[0101] As Figure 1 shown, a method for crushing and recycling thermosetting plastics, the method includes:

[0102] U1. During the crushing and recycling process of thermosetting plastics, collect the data information of the falling speed of thermosetting plastics, and obtain in real time the data information of the rotating speed of the magnetic separator drum and the data information of the belt speed of the eddy current separator;

[0103] U2. Based on the data information of the rotating speed of the magnetic separator drum, the data information of the belt speed of the eddy current separator, and the data information of the falling speed of thermosetting plastics, adopt an improved multi-variable regression prediction algorithm that combines a bidirectional recurrent neural network with attention to predict the screening results of thermosetting plastics, and obtain the data information of the predicted screening results of thermosetting plastics;

[0104] U3. Based on the data information of the predicted screening results of thermosetting plastics, adopt an improved triangular topology polymerization optimization algorithm to optimize the screening results of thermosetting plastics, and obtain the data information of the optimized screening results of thermosetting plastics;

[0105] U4. Based on the data information of the optimized screening results of thermosetting plastics, construct an adaptive feedback regulation function F to control and regulate the falling speed, the rotating speed of the magnetic separator drum, and the belt speed of the eddy current separator, and output the data information of the regulated falling speed, the rotating speed of the magnetic separator drum, and the belt speed of the eddy current separator.

[0106] In this embodiment, as Figure 2 shown, in step U2, the adoption of an improved multi-variable regression prediction algorithm that combines a bidirectional recurrent neural network with attention to predict the screening results of thermosetting plastics includes:

[0107] U21. Based on the data information of the drum rotation speed of the magnetic separator, the data information of the belt rotation speed of the eddy current separator, and the data information of the falling speed of the thermosetting plastic, construct the preliminary processing sequence function Q of the thermosetting plastic,

[0108]

[0109] where y1 is the data information of the drum rotation speed of the magnetic separator, y2 is the data information of the belt rotation speed of the eddy current separator, y3 is the data information of the falling speed of the thermosetting plastic, and β1, β2, and β3 are weight coefficients, which characterize the drum rotation speed of the magnetic separator, the belt rotation speed of the eddy current separator, and the falling speed, to obtain the data information of the preliminary processing sequence of the thermosetting plastic;

[0110] U22. Input the data information of the preliminary processing sequence of the thermosetting plastic into the bidirectional recurrent neural network model for training and learning, predict the preliminary screening result of the thermosetting plastic, and obtain the data information of the preliminary screening result of the thermosetting plastic;

[0111] U23. Based on the data information of the preliminary screening result of the thermosetting plastic, construct the multivariate regression prediction function W of the attention mechanism,

[0112]

[0113] where z is the data information of the preliminary screening result of the thermosetting plastic, and γ1, γ2, and γ3 are attention mechanism factors, which predict the screening result of the thermosetting plastic to obtain the data information of the predicted screening result of the thermosetting plastic.

[0114] In this embodiment, the attention mechanism factors γ1, γ2, and γ3 are

[0115]

[0116] where z is the data information of the preliminary screening result of the thermosetting plastic.

[0117] In this embodiment, as Figure 3 shown, in step U3, the optimization of the screening result of the thermosetting plastic by using the improved triangular topology aggregation optimization algorithm includes:

[0118] U31. Based on the data information of the predicted screening result of the thermosetting plastic, initialize the population, determine the population size and the variable dimension D, and obtain the data information of the initialized population;

[0119] U32. Based on the data information of the initialized population, establish the triangular topology unit function S,

[0120]

[0121] Among them, r is the data information of the population after initialization, η1, η2, and η3 are random numbers, which characterize the triangular topological units of the population to obtain the data information of the triangular topological units of the population;

[0122] U33. Based on the data information of the triangular topological units of the population, establish the objective optimization function G,

[0123]

[0124] Among them, a is the data information of the triangular topological units of the population, p1(a) is the global aggregation function of the population, p2(a) is the local aggregation function of the population, and λ1, λ2, and λ3 are the weight coefficients of the objective optimization, which optimize the screening results of the thermosetting plastics to obtain the data information of the optimized screening results of the thermosetting plastics.

[0125] In this embodiment, the global aggregation function p1(a) of the population is

[0126]

[0127] The local aggregation function p2(a) of the population is

[0128]

[0129] Among them, a is the data information of the triangular topological units of the population.

[0130] In this embodiment, the present invention provides a computer-readable storage medium, on which a computer program is stored that is programmed or configured to execute any one of the thermosetting plastic crushing and recycling processing methods.

[0131] Any reference to memory, storage, database, or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), among others.

[0132] In summary, the present invention can not only accurately screen metal particles or non-ferrous metal powders in thermosetting plastics to ensure the high-quality completion of subsequent processes, but also improve the recycling efficiency and quality of thermosetting plastics.

[0133] The above specific embodiments do not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present disclosure shall be included within the protection scope of the present disclosure.

Claims

1. A method for crushing and recycling thermosetting plastics, characterized in that: The method comprises: U1. In the process of thermosetting plastic crushing and recycling, collect data information on the falling speed of thermosetting plastic, and obtain data information on the drum speed of the magnetic separator and the belt speed of the eddy current separator in real time; U2. Based on the data information of the drum speed of the magnetic separator, the belt speed of the eddy current separator and the falling speed of the thermosetting plastic, an improved bidirectional recurrent neural network fusion attention multivariate regression prediction algorithm is used to predict the screening results of the thermosetting plastic, and the data information of the predicted screening results of the thermosetting plastic is obtained; U3. Based on the predicted data information of the screening results of the thermosetting plastics, the screening results of the thermosetting plastics are optimized using an improved triangle topology aggregation optimization algorithm to obtain optimized data information of the screening results of the thermosetting plastics; U4. Based on the data information of the screening results of the optimized thermosetting plastics, an adaptive feedback adjustment function F is constructed to control and adjust the blanking speed, the rotation speed of the magnetic separator drum and the belt speed of the eddy current separator, and output the data information of the adjusted blanking speed, the rotation speed of the magnetic separator drum and the belt speed of the eddy current separator.

2. The method for crushing and recycling thermosetting plastics according to claim 1, characterized in that: The adaptive feedback adjustment function F is: Among them, f1 is the adaptive feedback control function of the blanking speed, α 11 , α 12 and α 13 is the dynamic adjustment factor of the material dropping speed, f2 is the adaptive feedback control function of the speed of the magnetic separator drum, α 21 , α 22 and α 23 is the dynamic adjustment factor of the drum speed of the magnetic separator, f3 is the adaptive feedback regulation control function of the belt speed of the eddy current separator, α 31 , α 32 and α 33 is the dynamic adjustment factor of the belt speed of the eddy current separator, and x is the data information of the screening results of the optimized thermosetting plastic.

3. The method for crushing and recycling thermosetting plastics according to claim 2, characterized in that: The dynamic adjustment factor α of the blanking speed 11 , α 12 and α 13 for, The dynamic adjustment factor α of the magnetic separator drum speed 21 , α 22 and α 23 for, The dynamic adjustment factor α of the belt speed of the eddy current separator 31 , α 32 and α 33 for, Wherein, x is the data information of the screening results of the optimized thermosetting plastics.

4. The method for crushing and recycling thermosetting plastics according to claim 2, characterized in that: The dynamic adjustment factor α of the blanking speed 11 , α 12 and α 13 The constraints are: The dynamic adjustment factor α of the magnetic separator drum speed 21 , α 22 and α 23 The constraint function g is, The dynamic adjustment factor α of the belt speed of the eddy current separator 31 , α 32 and α 33 The constraint function h is, Among them, the value range of the constraint function g is (0,1), and the value range of the constraint function h is (1,2).

5. The thermosetting plastic crushing and recycling processing method according to claim 1, characterized in that: In step U2, the use of the improved bidirectional recurrent neural network fused attention multivariate regression prediction algorithm to predict the screening results of thermosetting plastics includes: U21. Based on the data information of the drum speed of the magnetic separator, the belt speed of the eddy current separator and the falling speed of the thermosetting plastic, a preliminary processing sequence function Q of the thermosetting plastic is constructed. Among them, y1 is the data information of the magnetic separator drum speed, y2 is the data information of the eddy current separator belt speed, y3 is the data information of the thermosetting plastic blanking speed, β1, β2 and β3 are weight coefficients, and the magnetic separator drum speed, eddy current separator belt speed and blanking speed are characterized to obtain the data information of the preliminary processing sequence of the thermosetting plastic; U22. Inputting the data information of the preliminary processing sequence of the thermosetting plastic into the bidirectional recurrent neural network model for training and learning, predicting the preliminary screening results of the thermosetting plastic, and obtaining the data information of the preliminary screening results of the thermosetting plastic; U23. Based on the data information of the preliminary screening results of the thermosetting plastic, a multivariate regression prediction function W of the attention mechanism is constructed. Among them, z is the data information of the preliminary screening results of thermosetting plastics, γ1, γ2 and γ3 are attention mechanism factors, and the screening results of thermosetting plastics are predicted to obtain the data information of the predicted screening results of thermosetting plastics.

6. The method for crushing and recycling thermosetting plastics according to claim 5, characterized in that: The attention mechanism factors γ1, γ2 and γ3 are, Wherein, z is the data information of the preliminary screening results of thermosetting plastics.

7. The thermosetting plastic crushing and recycling processing method according to claim 1, characterized in that: In step U3, the optimization of the screening results of thermosetting plastics by using the improved triangular topology aggregation optimization algorithm includes: U31. Based on the data information of the predicted screening results of the thermosetting plastics, the population is initialized, the population size and the variable dimension D are determined, and the data information of the initialized population is obtained; U32. Based on the data information of the initialized population, a triangle topology unit function S is established. Among them, r is the data information of the initialized population, η1, η2 and η3 are random numbers, which characterize the triangular topological unit of the population to obtain the data information of the triangular topological unit of the population; U33. Based on the data information of the triangular topological unit of the population, establish the target optimization function G, Among them, a is the data information of the triangular topological unit of the population, p1(a) is the global aggregation function of the population, p2(a) is the local aggregation function of the population, λ1, λ2 and λ3 are the weight coefficients of the target optimization, and the screening results of thermosetting plastics are optimized to obtain the data information of the optimized screening results of thermosetting plastics.

8. The method for crushing and recycling thermosetting plastics according to claim 7, characterized in that: The global aggregation function p1(a) of the population is, The local aggregation function p2(a) of the population is, Among them, a is the data information of the triangular topological unit of the population.

9. The thermosetting plastic crushing and recycling processing method according to claim 1, characterized in that: The method further comprises: U5. After eddy current screening, it enters the liquid nitrogen tank for cold extraction, and then after primary and secondary crushing, it enters the waste collection tank.

10. A system for implementing the thermosetting plastic crushing and recycling processing method according to any one of claims 1 to 9, characterized in that: The system comprises: The data acquisition module is used to collect data information on the falling speed of thermosetting plastics, and obtain data information on the drum speed of the magnetic separator and the belt speed of the eddy current separator in real time; A prediction module for the screening results of thermosetting plastics, connected to the data acquisition module, is used to predict the screening results of thermosetting plastics by using an improved bidirectional recurrent neural network fused with attention multivariate regression prediction algorithm to obtain data information of the predicted screening results of thermosetting plastics; An optimization module for the screening results of thermosetting plastics, connected to the prediction module for the screening results of thermosetting plastics, is used to optimize the screening results of thermosetting plastics by using an improved triangle topology aggregation optimization algorithm to obtain data information of the screening results of the optimized thermosetting plastics; The adaptive feedback control and regulation module is connected to the optimization module of the screening results of the thermosetting plastic, and is used to construct an adaptive feedback regulation function F, control and regulate the blanking speed, the rotation speed of the magnetic separator drum and the belt speed of the eddy current separator, and output data information of the adjusted blanking speed, the rotation speed of the magnetic separator drum and the belt speed of the eddy current separator.

Images