Equipment recycling method and system

Abstract

The present invention provides an equipment recycling processing method, which obtains attribute information of smart devices and establishes a hybrid graph disassembly model, generates disassembly information according to the hybrid graph disassembly model and disassembly constraint relationship, constructs a recycling network model of the smart devices according to the attribute information of the smart devices and saves the recycling model to an equipment recycling cloud service platform, establishes a disassembly cloud service platform based on the disassembly information and the recycling network model, and the disassembly cloud service platform receives a disassembly request of a disassembly task and allocates disassembly resources to complete equipment recycling. The disassembly model can accurately disassemble part-related information and express the assembly structure and disassembly information of the product in detail. A mathematical matrix is ​​introduced based on the hybrid graph to express the constraint relationship of discarded smart devices to establish a hybrid graph disassembly model. The disassembly sequence can achieve the goal of minimizing disassembly workload and cost, avoids some non-removable constraints or uses destructive methods, and improves the work efficiency of smart device recycling.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an equipment recycling method and system. Background Art

[0002] Discarded smart devices are causing increasingly severe environmental problems, posing huge challenges to product designers, manufacturers and recyclers. For discarded electronic products, they will not disappear or decompose on their own, and the incineration facilities in landfills continue to occupy human living space and pollute the environment. It is well known that the best way to deal with discarded products is to recycle them. However, only a very small number of products can be fully recycled. The main goal of discarded product treatment is to maximize the proportion of parts and materials reused, recycled and recreated in products, reduce the part that is landfilled or incinerated, and achieve beneficial protection of the natural environment.

[0003] There is great uncertainty in the recycling of parts and raw materials used in discarded products, and third-party recycling companies often do not have key information related to recycling, such as product structure, parts materials, etc., which makes it impossible to fully recycle discarded products, making product designers, manufacturers and waste product recycling organizations lack professional knowledge and skills for complex product recycling and lack sufficient information to make decisions on recycling equipment. Summary of the invention

[0004] In view of this, the present invention provides an equipment recycling and processing method and system, which can improve the disassembly efficiency and recycling value of large quantities of waste smart devices. By establishing a recycling cloud service platform, not only can the sharing of product life cycle information be achieved, but also the unified allocation of disassembly and recycling resources can be achieved, thereby improving the equipment recycling efficiency. The following technical solutions are specifically adopted to achieve this.

[0005] In a first aspect, the present invention provides a method for recycling equipment, comprising the following steps:

[0006] Acquire the attribute information of the smart device and establish a hybrid graph disassembly model, wherein the hybrid graph disassembly model uses a digital matrix to represent the constraint relationship between parts;

[0007] Generate disassembly information according to the hybrid graph disassembly model and the disassembly constraint relationship, wherein the disassembly information includes disassembly time, disassembly tools and disassembly sequence;

[0008] A recycling network model of smart devices is constructed based on the attribute information of smart devices and the recycling model is saved to the device recycling cloud service platform, wherein the physical resources in the cloud platform include recycling-related facilities and knowledge resources, and cloud service requests are decomposed by the cloud platform. According to the specific resource types requested, the decision module dynamically organizes and allocates service resources to form a recycling cloud service;

[0009] A disassembly cloud service platform is established based on the disassembly information and recycling network model. The disassembly cloud service platform receives disassembly requests for disassembly tasks and allocates disassembly resources to complete equipment recycling.

[0010] As a further improvement of the above technical solution, disassembly information is generated according to the hybrid graph disassembly model and the disassembly constraint relationship, including:

[0011] The function of the disassembly time T of the smart device is expressed as Where i and k are the i-th and k-th parts in the smart device. represents the basic time of disassembly condition i to part k, t ik It represents the replacement coefficient of the disassembly tool from disassembling part i and part k. When the same disassembly tool is used to disassemble part i and part k, t ik =0, otherwise t ik =1,d ik It represents the coefficient of variation in the disassembly direction from part i to part k. When the disassembly directions of part i and part k are the same, d ik =0, otherwise d ik =1, They are the single replacement time of the disassembly tool and the single change time of the disassembly direction, and the auxiliary time for disassembling from part i to k;

[0012] According to the preset schedule of disassembly tool replacement, the single replacement time of the disassembly tool of the smart device is set as follows: when the disassembly tool from part i to part k remains unchanged When a single disassembly tool is changed from part i to part k When N disassembly tools are changed from part i to part k

[0013] The disassembly of the intelligent device is divided into six disassembly directions, which are set as +X, -X, +Y, -Y, +Z, and -Z respectively. The single change time of the disassembly direction of the parts in the intelligent device is set as: when the disassembly direction from part i to part k remains unchanged When the disassembly direction changes 90° from part i to part k When the disassembly direction changes 180° from part i to part k

[0014] As a further improvement of the above technical solution, a recycling profit function is established Where n is the total number of disassemblies, r i represents the recycling income of dismantled parts i, a d is the disassembly cost of old smart devices, a p is the purchase cost of the old smart devices, a t represents the transportation cost of waste smart devices, a hThe inventory cost of obsolete smart devices;

[0015] Disassembly costa d The cost of dismantling old smart devices, including labor costs and depreciation costs of dismantling tools, and the cost of dismantling old smart devices where a r The labor cost of the demolition personnel, a l Depreciation of disassembly tools;

[0016] A target optimization model is established based on the disassembly time and recycling profit. The function expression of the target optimization model is: Where F is the target optimization function, λ1 is the disassembly time weight coefficient, and λ2 is the recovery profit weight coefficient.

[0017] As a further improvement of the above technical solution, a recycling network model of smart devices is constructed according to the attribute information of smart devices and the recycling model is saved to the device recycling cloud service platform, including:

[0018] The difference between the time when the smart device is collected and the time when the smart device starts to be used is recorded as t1, which is used to determine the newness of the device model. The operation time of the smart device is recorded as t2. If the smart device has the ability to record operating parameters, the setting of the operating parameters determines the level of workload. The workload is recorded as L, and the workload includes three levels: light load, medium load and heavy load;

[0019] If the workload level is automatically generated by the intelligent device, the structure of the part or module self-test is pass and fail. Passing the self-test means that it is working normally, and failing the self-test means that there is an abnormality in the part or module.

[0020] As a further improvement of the above technical solution, a disassembly mixed graph of the robot arm is established according to the mixed graph disassembly model, including:

[0021] It is assumed that n is the total number of parts and connectors in the equipment, part m is the reference part, k is the disassembled part, list s is the sequence of disassembled parts that meet the disassembly conditions, list q is the initial population N×n matrix of the disassembly sequence, and N is the number of the initial population;

[0022] Determine the model of the smart device, obtain the relevant information for disassembly and determine the reference part m, establish a mixed graph disassembly model, generate the connection matrix G1 and the priority matrix G2 to obtain the disassembly conditions;

[0023] Initialize the number of good populations N, assign i = 0 and clear the list q, determine whether i is less than N, if so, according to the part detachable conditions, put any part k that meets the detachable conditions into the fission s; if not, output the list q and terminate the program;

[0024] Determine whether n is greater than 0. If so, delete part k, update the connection matrix G1 and the optimization matrix G2, n-1 and go to re-determination. If not, obtain the list s and store it in another table q.

[0025] As a further improvement of the above technical solution, the fitness function is used as a criterion for testing the quality of individuals in the population. The disassembly time and recycling profit are used as the fitness function to search for the minimum value of the disassembly time and the maximum value of the recycling profit. The expression of the fitness function f(x) is:

[0026] The weighted summation is used to integrate the disassembly time and recycling profit into a fitness function. The disassembly time and recycling profit are normalized to obtain the weighted summation fitness function T v =(TT min ) / (T max -T min ), P v =(P max -P) / (P max -P min ), f = μ1T v +μ2P v , where T max 、T min are the maximum and minimum disassembly time, P max and P min For the minimum recovery profit, μ1 and μ2 are T v , P v The weight coefficient of the relative weight is μ1+μ2=1.

[0027] As a further improvement of the above technical solution, the weighted sum is used to integrate the disassembly time and recycling profit into a fitness function, including:

[0028] Input the original data, randomly generate N feasible groups coded by real numbers, form the initial population, and calculate the fitness values ​​of all individuals in the population according to the weighted sum fitness function;

[0029] In the initial population, individuals with high fitness values ​​are selected as parents to form an individual set B. According to the crossover probability P B According to the crossover operator, the adjacent individuals are crossovered to generate a new individual B1, where the crossover operator uses a single-point crossover method to arbitrarily select a crossover point, divide the parent individual into two parts, and exchange the gene information behind the crossover point of the parent individual;

[0030] According to the mutation probability P mAccording to the mutation operator, the new individual B1 is mutated to obtain the new individual B2, and the optimal individual is retained. According to the population individual exchange, the individuals between the two populations are exchanged, and the number of calls is randomly generated. After the exchange, a new generation of population is generated. The mutation operator is to select two mutation points in the chromosome at random, delete the gene segment between the mutation points, and then generate a gene segment that satisfies the disassembly constraint relationship according to the initial disassembly sequence. The gene segment is inserted into the deleted area and a new individual is regenerated. Then the individual satisfies the disassembly constraint relationship.

[0031] Determine whether the end condition has reached the maximum number of iterations. If it is satisfied, output the result. Otherwise, return to execute the calculated fitness value. The real number encoding represents a gene for each disassembled part, and the number of disassembled parts is equal to its encoding length.

[0032] As a further improvement of the above technical solution, the process of disassembly sequence optimization also includes:

[0033] Set the population size and iteration times of the algorithm, encode the disassembled parts of the scrapped smart devices with real numbers, randomly generate N feasible particles with real number encoding to form the initial population, calculate the fitness values ​​of all ion individuals in the population, compare them, and use the particle corresponding to the minimum value as the initial value of tourPbest and tourGbest;

[0034] Determine whether the end condition has reached the maximum number of iterations. If it is satisfied, update the ion, otherwise output the result. The updating of the ion includes updating the individual optimal particle tourPbest and the population optimal particle tourGbest according to the fitness value of the ion in the loyal dog. During the iteration process, the particle corresponding to the minimum fitness value is taken as the historical optimal of tourPbest and tourGbest.

[0035] According to the crossover operator, a crossover operation is performed on the particles and the individual optimal particle tourPbest and the group optimal particle tourGbest to generate new particles, an interference check is performed on the new particles C generated by the crossover operation to generate new particles D, and fitness values ​​are calculated for all ions D generated by the interference check operation;

[0036] The fitness values ​​of the new particle D and the new particle are compared, the particle with the smallest fitness value is retained, and the retained particle is mutated to optimize the disassembly sequence.

[0037] As a further improvement of the above technical solution, the fitness value of the excellent particle retained after each generation of mutation is compared with the historical tourPbest. If the fitness value of the excellent particle is smaller, the particle is used as tourPbest, otherwise the historical optimal tourPbest is retained. The tourPbest of the excellent particle is compared with the optimal value tourGbest of the group. If its fitness value is smaller, it is used as tourGbest, otherwise the historical optimal tourGbest is retained.

[0038] In a second aspect, the present invention further provides an equipment recycling and processing system, comprising:

[0039] An acquisition module is used to acquire the attribute information of the intelligent device and establish a hybrid graph disassembly model, wherein the hybrid graph disassembly model uses a digital matrix to represent the constraint relationship between parts;

[0040] A first generating module is used to generate disassembly information according to the hybrid graph disassembly model and the disassembly constraint relationship, wherein the disassembly information includes disassembly time, disassembly tools and disassembly sequence;

[0041] The second generation module is used to build a recycling network model of the smart device according to the attribute information of the smart device and save the recycling model to the device recycling cloud service platform, wherein the physical resources in the cloud platform include recycling-related facilities and knowledge resources, and the cloud service request is decomposed by the cloud platform. According to the specific resource type requested, the decision module dynamically organizes and allocates service resources to form a recycling cloud service;

[0042] The processing module is used to establish a disassembly cloud service platform based on the disassembly information and the recycling network model. The disassembly cloud service platform receives the disassembly request of the disassembly task and allocates disassembly resources to complete the equipment recycling.

[0043] The present invention provides an equipment recycling processing method and system, which obtains the attribute information of the intelligent device and establishes a hybrid graph disassembly model, generates disassembly information according to the hybrid graph disassembly model and the disassembly constraint relationship, constructs a recycling network model of the intelligent device according to the attribute information of the intelligent device and saves the recycling model to an equipment recycling cloud service platform, establishes a disassembly cloud service platform based on the disassembly information and the recycling network model, and the disassembly cloud service platform receives the disassembly request of the disassembly task and allocates disassembly resources to complete the equipment recycling. The disassembly model can accurately disassemble the relevant information of the parts and express the assembly structure and disassembly information of the product in detail. The hybrid graph disassembly model is established by introducing a mathematical matrix to express the constraint relationship of the discarded intelligent device based on the hybrid graph. The disassembly sequence can achieve the goal of minimizing the disassembly workload and cost, avoids some non-disassembly constraints or adopts destructive methods to reduce the total disassembly workload and cost, and also improves the work efficiency of intelligent device recycling. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0045] Figure 1 A flow chart of the equipment recycling method provided by the present invention;

[0046] Figure 2 A process diagram for constructing the fitness function provided by the present invention;

[0047] Figure 3 A process diagram for disassembly sequence optimization provided by the present invention;

[0048] Figure 4 This is a structural block diagram of the equipment recycling and processing system provided by the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0051] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0052] See also Figure 1 The present invention provides a method for recycling equipment, comprising the following steps:

[0053] S10: Acquire attribute information of the smart device and establish a hybrid graph disassembly model, wherein the hybrid graph disassembly model uses a digital matrix to represent the constraint relationship between parts;

[0054] S11: generating disassembly information according to the hybrid graph disassembly model and the disassembly constraint relationship, wherein the disassembly information includes disassembly time, disassembly tools and disassembly sequence;

[0055] S12: constructing a recycling network model of the smart device according to the attribute information of the smart device and saving the recycling model to the device recycling cloud service platform, wherein the physical resources in the cloud platform include recycling-related facilities and knowledge resources, and the cloud service request is decomposed by the cloud platform, and the decision module dynamically organizes and allocates service resources according to the specific resource type requested to form a recycling cloud service;

[0056] S13: A disassembly cloud service platform is established based on the disassembly information and the recycling network model. The disassembly cloud service platform receives a disassembly request of a disassembly task and allocates disassembly resources to complete equipment recycling.

[0057] In this embodiment, the process of generating disassembly information according to the hybrid graph disassembly model and the disassembly constraint relationship includes: the function of the disassembly time T of the smart device is expressed as: Where i and k are the i-th and k-th parts in the smart device. represents the basic time of disassembly condition i to part k, t ik It represents the replacement coefficient of the disassembly tool from disassembling part i and part k. When the same disassembly tool is used to disassemble part i and part k, t ik =0, otherwise t ik =1,d ik It represents the coefficient of variation in the disassembly direction from part i to part k. When the disassembly directions of part i and part k are the same, d ik =0, otherwise d ik =1, They are the single replacement time of the disassembly tool of the part, the single change time of the disassembly direction, and the auxiliary time of disassembling from part i to k; according to the preset schedule of disassembly tool replacement, the single replacement time of the disassembly tool of the smart device is set as: when the disassembly tool from part i to part k remains unchanged When a single disassembly tool is changed from part i to part k When N disassembly tools are changed from part i to part k The disassembly of the intelligent device is divided into six disassembly directions, which are set as +X, -X, +Y, -Y, +Z, and -Z respectively. The single change time of the disassembly direction of the parts in the intelligent device is set as: when the disassembly direction from part i to part k remains unchanged When the disassembly direction changes 90° from part i to part k When the disassembly direction changes 180° from part i to part k

[0058] It should be noted that the recovery profit function is established Where n is the total number of disassemblies, r i represents the recycling income of dismantled parts i, a dis the disassembly cost of old smart devices, a p is the purchase cost of the old smart devices, a t represents the transportation cost of waste smart devices, a h The inventory cost of obsolete smart devices;

[0059] Disassembly costa d The cost of dismantling old smart devices, including labor costs and depreciation costs of dismantling tools, and the cost of dismantling old smart devices where a r The labor cost of the demolition personnel, a l Depreciation of disassembly tools;

[0060] A target optimization model is established based on the disassembly time and recycling profit. The function expression of the target optimization model is: Where F is the target optimization function, λ1 is the disassembly time weight coefficient, and λ2 is the recovery profit weight coefficient.

[0061] It should be understood that the difference between the time when the smart device is collected and the time when the smart device begins to be used is recorded as t1, which is used to determine the newness of the device model. The operation time of the smart device is recorded as t2. If the smart device has the ability to record operating parameters, the setting of the operating parameters determines the level of the workload. The workload is recorded as L, and the workload includes three levels: light load, medium load and heavy load; if the level of the workload is automatically generated by the smart device, the structure of the self-inspection of the parts or modules is pass and fail. Passing the self-inspection means that it works normally, and failing the self-inspection means that there is an abnormality in the parts or modules. Taking mobile phones as an example, the recycling of used mobile phones based on Internet technology mainly includes offline recycling and online mobile platform recycling. The factors that affect the operating efficiency of the online recycling platform for used mobile phones mainly include information security, recycling prices, word-of-mouth effect, recycling logistics and recycling channels. As the complexity of products increases, the number of parts will also increase accordingly. Generally, the number of disassembly sequences will show an exponential growth. For smartphones, which have a complex structural layout, a large number of small precision parts, and a variety of constraints between parts, we choose to use intelligent methods and priority constraint methods to obtain the disassembly sequence of smartphones, use human-computer interaction to obtain disassembly-related information of smartphones, and simplify the generation problem of disassembly sequences based on the knowledge framework and rich experience of experts. We use priority constraints to analyze and summarize the constraint relationships between all parts of smartphones, use intelligent algorithms to search sequences, obtain all feasible initial disassembly sequences, and then select the optimal or approximately optimal disassembly sequence that meets the conditions according to the given optimization goal, thereby improving the accuracy of disassembly during the recycling process of smart devices.

[0062] Optionally, a disassembly mixed graph of the robot arm is established according to the mixed graph disassembly model, including:

[0063] It is assumed that n is the total number of parts and connectors in the equipment, part m is the reference part, k is the disassembled part, list s is the sequence of disassembled parts that meet the disassembly conditions, list q is the initial population N×n matrix of the disassembly sequence, and N is the number of the initial population;

[0064] Determine the model of the smart device, obtain the relevant information for disassembly and determine the reference part m, establish a mixed graph disassembly model, generate the connection matrix G1 and the priority matrix G2 to obtain the disassembly conditions;

[0065] Initialize the number of good populations N, assign i = 0 and clear the list q, determine whether i is less than N, if so, according to the part detachable conditions, put any part k that meets the detachable conditions into the fission s; if not, output the list q and terminate the program;

[0066] Determine whether n is greater than 0. If so, delete part k, update the connection matrix G1 and the optimization matrix G2, n-1 and go to re-determination. If not, obtain the list s and store it in another table q.

[0067] In this embodiment, the fitness function is used as a criterion for testing the quality of individuals in the population. The disassembly time and the recycling profit are used as the fitness function to search for the minimum value of the disassembly time and the maximum value of the recycling profit. The expression of the fitness function f(x) is: The weighted summation is used to integrate the disassembly time and recycling profit into a fitness function. The disassembly time and recycling profit are normalized to obtain the weighted summation fitness function T v =(TT min ) / (T max -T min ), P v =(P max -P) / (P max -P min ), f = μ1T v +μ2P v , where T max 、T min are the maximum and minimum disassembly time, P max and P min For the minimum recovery profit, μ1 and μ2 are T v , P v The weight coefficient of the relative weight is μ1+μ2=1.

[0068] It should be noted that the disassembly of the current part is defined according to the constraints between the structural components of the smartphone. When the part is disassembled, the constraint relationship of the part needs to be deleted, the connection matrix and the priority matrix need to be updated, and then the disassembly of the next part needs to be judged. When the scrapped smartphone is completely disassembled, the initial population cannot fully extract the information of the disassembly mixed graph. The feasible disassembly sequences of scrapped smartphones are as high as tens of thousands, and these disassembly sequences have large differences in disassembly time, disassembly cost, etc. A large number of feasible disassembly sequences need to be optimized according to different evaluation indicators, namely, disassembly time and recycling profit, to obtain the optimal or approximately optimal sequence for device disassembly. Using dual populations for optimization can increase the randomness of the feasible population, expand the search space of the population, speed up the search rate, and improve the quality of the optimization solution.

[0069] See also Figure 2 , the weighted sum is used to integrate the disassembly time and recycling profit into a fitness function, including:

[0070] S20: Input the original data, arbitrarily generate N feasible groups encoded by real numbers, form an initial population, and calculate the fitness values ​​of all individuals in the population according to the weighted sum fitness function;

[0071] S21: Select individuals with high fitness values ​​from the initial population as parents to form an individual set B. B According to the crossover operator, the adjacent individuals are crossovered to generate a new individual B1. The crossover operator uses a single-point crossover method to arbitrarily select a crossover point and divide the parent individual into two parts.

[0072] Exchange the genetic information behind the crossover point of the parent individuals;

[0073] S22: According to the mutation probability P m According to the mutation operator, the new individual B1 is mutated to obtain the new individual B2, and the optimal individual is retained. According to the population individual exchange, the individuals between the two populations are exchanged, and the number of calls is randomly generated. After the exchange, a new generation of population is generated. The mutation operator is to select two mutation points in the chromosome at random, delete the gene segment between the mutation points, and then generate a gene segment that satisfies the disassembly constraint relationship according to the initial disassembly sequence. The gene segment is inserted into the deleted area and a new individual is regenerated. Then the individual satisfies the disassembly constraint relationship.

[0074] S23: Determine whether the end condition has reached the maximum number of iterations. If it is satisfied, output the result. Otherwise, return to execute the fitness value calculation, where the real number encoding represents a gene for each disassembled part, and the number of disassembled parts is equal to its encoding length.

[0075] In this embodiment, during the disassembly process, there are constraints between product parts. In order to ensure that the generated disassembly sequence meets the tea break conditions, interference check operations are introduced after the cross operation and the mutation operation, that is, the method of generating the initial sequence performs a cyclic judgment check. After the mutation operation, the generated particles may not meet the disassembly constraints. After the mutation operation, a second interference check operation is performed. The mechanism of the second interference check operation is different from that of the first interference check operation, which will not be repeated here.

[0076] See also Figure 3 , the process of disassembly sequence optimization also includes:

[0077] S30: setting the population size and iteration times of the algorithm, encoding the disassembled parts of the scrapped smart devices with real numbers, arbitrarily generating N feasible particles with real number encoding to form an initial population, calculating the fitness values ​​of all ion individuals in the population, comparing them, and taking the particle corresponding to the minimum value as the initial value of tourPbest and tourGbest;

[0078] S31: Determine whether the end condition reaches the maximum number of iterations. If it is satisfied, update the ion, otherwise output the result, wherein updating the ion includes updating the individual optimal particle tourPbest and the population optimal particle tourGbest according to the fitness value of the ion in the loyal dog. During the iteration process, the particle corresponding to the minimum fitness value is used as the historical optimal of tourPbest and tourGbest;

[0079] S32: performing a crossover operation on the particles and the individual optimal particle tourPbest and the group optimal particle tourGbest according to the crossover operator to generate new particles, performing interference check on the new particles C generated by the crossover operation to generate new particles D, and calculating fitness values ​​for all ions D generated by the interference check operation;

[0080] S33: Compare the fitness values ​​of the new particle D and the new particle, retain the particle with the smallest fitness value, and perform mutation operation on the retained particle to optimize the disassembly sequence.

[0081] In this embodiment, the fitness value of the excellent particle retained after each generation of mutation is compared with the historical tourPbest. If the fitness value of the excellent particle is smaller, the particle is used as tourPbest, otherwise the historical optimal tourPbest is retained. The tourPbest of the excellent particle is compared with the group optimal value tourGbest. If its fitness value is smaller, it is used as tourGbest, otherwise the historical optimal tourGbest is retained. The smartphone disassembly sequence is optimized according to the basic intelligent algorithm to improve the optimization performance.

[0082] See also Figure 4The present invention also provides an equipment recycling and processing system, comprising:

[0083] An acquisition module is used to acquire the attribute information of the intelligent device and establish a hybrid graph disassembly model, wherein the hybrid graph disassembly model uses a digital matrix to represent the constraint relationship between parts;

[0084] A first generating module is used to generate disassembly information according to the hybrid graph disassembly model and the disassembly constraint relationship, wherein the disassembly information includes disassembly time, disassembly tools and disassembly sequence;

[0085] The second generation module is used to build a recycling network model of the smart device according to the attribute information of the smart device and save the recycling model to the device recycling cloud service platform, wherein the physical resources in the cloud platform include recycling-related facilities and knowledge resources, and the cloud service request is decomposed by the cloud platform. According to the specific resource type requested, the decision module dynamically organizes and allocates service resources to form a recycling cloud service;

[0086] The processing module is used to establish a disassembly cloud service platform based on the disassembly information and the recycling network model. The disassembly cloud service platform receives the disassembly request of the disassembly task and allocates disassembly resources to complete the equipment recycling.

[0087] In this embodiment, the basis for the operation of the resource layer recycling cloud service platform mainly includes physical resources, knowledge resources and supporting resources. Physical resources are hardware facilities provided to the society by enterprises with recycling service capabilities, such as disassembly robots, integrated recycling equipment, etc. The semantic method is used to establish ontology models for describing physical resources and knowledge resources, such as test equipment, disassembly equipment and recycling equipment for describing physical resources, and disassembly knowledge ontology model, material knowledge ontology model and recycling equipment ontology model for describing knowledge resources. According to user requests, product recycling resource optimization allocation and knowledge resource decision-making services are realized. For product recycling services composed of physical resources, the decision-making module comprehensively evaluates and allocates the allocated recycling resources from the aspects of recycling cost, recycling income, environmental characteristics and social benefits, and combines various required physical resources for recycling cloud service requests. After receiving the request, the service layer of the recycling cloud service platform decomposes the requested cloud service, mainly into requests for physical resources and requests for knowledge resources, and calls the product recycling resource decision module and the knowledge resource decision module to make decisions on the requested services, combines and encapsulates the decision results, and provides cloud service responses in the form of visual charts. The service layer also has the functions of monitoring and pricing recycling cloud services, so as to improve the intelligent and efficient management of equipment recycling.

[0088] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for recycling equipment, characterized in that: The following steps are involved: Acquire the attribute information of the smart device and establish a hybrid graph disassembly model, wherein the hybrid graph disassembly model uses a digital matrix to represent the constraint relationship between parts; Generate disassembly information according to the hybrid graph disassembly model and the disassembly constraint relationship, wherein the disassembly information includes disassembly time, disassembly tools and disassembly sequence; A recycling network model of smart devices is constructed based on the attribute information of smart devices and the recycling model is saved to the device recycling cloud service platform, wherein the physical resources in the cloud platform include recycling-related facilities and knowledge resources, and cloud service requests are decomposed by the cloud platform. According to the specific resource types requested, the decision module dynamically organizes and allocates service resources to form a recycling cloud service; A disassembly cloud service platform is established based on disassembly information and recycling network model. The disassembly cloud service platform receives disassembly requests of disassembly tasks and allocates disassembly resources to complete equipment recycling. A recycling network model of smart devices is constructed based on the attribute information of smart devices and the recycling model is saved to the device recycling cloud service platform, including: The difference between the time when the smart device is collected and the time when the smart device starts to be used is recorded as t1, which is used to determine the newness of the device model. The operation time of the smart device is recorded as t2. If the smart device has the ability to record operating parameters, the setting of the operating parameters determines the level of workload. The workload is recorded as L, and the workload includes three levels: light load, medium load and heavy load; If the workload level is automatically generated by the intelligent device, the structure of the part or module self-test is pass and fail. Passing the self-test means that the work is normal, and failing the self-test means that there is an abnormality in the part or module; The disassembly mixed graph of the robot arm is established according to the mixed graph disassembly model, including: It is assumed that n is the total number of parts and connectors in the equipment, part m is the reference part, k is the disassembled part, list s is the sequence of disassembled parts that meet the disassembly conditions, list q is the initial population N×n matrix of the disassembly sequence, and N is the number of the initial population; Determine the model of the smart device, obtain the relevant information for disassembly and determine the reference part m, establish a mixed graph disassembly model, generate the connection matrix G1 and the priority matrix G2 to obtain the disassembly conditions; Initialize the number of good populations N, assign i = 0 and clear the list q, determine whether i is less than N, if so, according to the part detachable conditions, put any part k that meets the detachable conditions into the fission s; if not, output the list q and terminate the program; Determine whether n is greater than 0. If so, delete part k, update the connection matrix G1 and the optimization matrix G2, n-1 and go to re-determine. If not, get the list s and store it in another table q; The fitness function is used as a criterion for testing the quality of individuals in the population. The disassembly time and recycling profit are used as the fitness function to search for the minimum disassembly time and the maximum recycling profit. The expression of the fitness function f(x) is: The weighted summation is used to integrate the disassembly time and recycling profit into a fitness function. The disassembly time and recycling profit are normalized to obtain the weighted summation fitness function T v =(TT min ) / (T max -T min ), P v =(P max -P) / (P max -P min ), f = μ1T v +μ2P v , where T max 、T min are the maximum and minimum disassembly time, P max and P min For the minimum recovery profit, μ1 and μ2 are T v , P v The weight coefficient of the relative weight is μ1+μ2=1.

2. The equipment recycling method according to claim 1, characterized in that: Generate disassembly information based on the hybrid graph disassembly model and disassembly constraint relationship, including: The function of the disassembly time T of the smart device is expressed as Where i and k are the i-th and k-th parts in the smart device. represents the basic time of disassembly condition i to part k, t ik It represents the replacement coefficient of the disassembly tool from disassembling part i and part k. When the same disassembly tool is used to disassemble part i and part k, t ik =0, otherwise t ik =1,d ik It represents the coefficient of variation in the disassembly direction from part i to part k. When the disassembly directions of part i and part k are the same, d ik =0, otherwise d ik =1, They are the single replacement time of the disassembly tool and the single change time of the disassembly direction, and the auxiliary time for disassembling from part i to k; According to the preset schedule of disassembly tool replacement, the single replacement time of the disassembly tool of the smart device is set as follows: when the disassembly tool from part i to part k remains unchanged When a single disassembly tool is changed from part i to part k When N disassembly tools are changed from part i to part k The disassembly of the intelligent device is divided into six disassembly directions, which are set as +X, -X, +Y, -Y, +Z, and -Z respectively. The single change time of the disassembly direction of the parts in the intelligent device is set as: when the disassembly direction from part i to part k remains unchanged When the disassembly direction changes 90° from part i to part k When the disassembly direction changes 180° from part i to part k 3. The equipment recycling method according to claim 2, characterized in that: Also includes: Establishing the recovery profit function Where n is the total number of disassemblies, r i represents the recycling income of dismantled parts i, a d is the disassembly cost of old smart devices, a p is the purchase cost of the old smart devices, a t represents the transportation cost of waste smart devices, a h The inventory cost of obsolete smart devices; Disassembly costa d The cost of dismantling old smart devices, including labor costs and depreciation costs of dismantling tools, and the cost of dismantling old smart devices where a r The labor cost of the demolition personnel, a l Depreciation of disassembly tools; A target optimization model is established based on the disassembly time and recycling profit. The function expression of the target optimization model is: Where F is the target optimization function, λ1 is the disassembly time weight coefficient, and λ2 is the recovery profit weight coefficient.

4. The equipment recycling method according to claim 1, characterized in that: The weighted sum is used to integrate the disassembly time and recycling profit into a fitness function, including: Input the original data, randomly generate N feasible groups coded by real numbers, form the initial population, and calculate the fitness values ​​of all individuals in the population according to the weighted sum fitness function; In the initial population, individuals with high fitness values ​​are selected as parents to form an individual set B. According to the crossover probability P B According to the crossover operator, the adjacent individuals are crossovered to generate a new individual B1, where the crossover operator uses a single-point crossover method to arbitrarily select a crossover point, divide the parent individual into two parts, and exchange the gene information behind the crossover point of the parent individual; According to the mutation probability P m According to the mutation operator, the new individual B1 is mutated to obtain the new individual B2, and the optimal individual is retained. According to the population individual exchange, the individuals between the two populations are exchanged, and the number of calls is randomly generated. After the exchange, a new generation of population is generated. The mutation operator is to select two mutation points in the chromosome at random, delete the gene segment between the mutation points, and then generate a gene segment that satisfies the disassembly constraint relationship according to the initial disassembly sequence. The gene segment is inserted into the deleted area and a new individual is regenerated. Then the individual satisfies the disassembly constraint relationship. Determine whether the end condition has reached the maximum number of iterations. If it is satisfied, output the result. Otherwise, return to execute the calculated fitness value. The real number encoding represents a gene for each disassembled part, and the number of disassembled parts is equal to its encoding length.

5. The equipment recycling method according to claim 1, characterized in that: The process of disassembly sequence optimization also includes: Set the population size and iteration times of the algorithm, encode the disassembled parts of the scrapped smart devices with real numbers, randomly generate N feasible particles with real number encoding to form the initial population, calculate the fitness values ​​of all ion individuals in the population, compare them, and use the particle corresponding to the minimum value as the initial value of tourPbest and tourGbest; Determine whether the end condition has reached the maximum number of iterations. If it is satisfied, update the ion, otherwise output the result. The updating of the ion includes updating the individual optimal particle tourPbest and the population optimal particle tourGbest according to the fitness value of the ion in the loyal dog. During the iteration process, the particle corresponding to the minimum fitness value is taken as the historical optimal of tourPbest and tourGbest. According to the crossover operator, a crossover operation is performed on the particles and the individual optimal particle tourPbest and the group optimal particle tourGbest to generate new particles, an interference check is performed on the new particles C generated by the crossover operation to generate new particles D, and fitness values ​​are calculated for all ions D generated by the interference check operation; The fitness values ​​of the new particle D and the new particle are compared, the particle with the smallest fitness value is retained, and the retained particle is mutated to optimize the disassembly sequence.

6. The equipment recycling method according to claim 5, characterized in that: Compare the fitness value of the excellent particle retained after each generation of mutation with the historical tourPbest. If the fitness value of the excellent particle is smaller, then the particle is used as tourPbest, otherwise the historical optimal tourPbest is retained. Compare the tourPbest of the excellent particle with the optimal tourGbest of the group. If its fitness value is smaller, then it is used as tourGbest, otherwise the historical optimal tourGbest is retained.

7. An equipment recycling system according to any one of claims 1 to 6, characterized in that: include: An acquisition module is used to acquire the attribute information of the intelligent device and establish a hybrid graph disassembly model, wherein the hybrid graph disassembly model uses a digital matrix to represent the constraint relationship between parts; A first generating module is used to generate disassembly information according to the hybrid graph disassembly model and the disassembly constraint relationship, wherein the disassembly information includes disassembly time, disassembly tools and disassembly sequence; The second generation module is used to build a recycling network model of the smart device according to the attribute information of the smart device and save the recycling model to the device recycling cloud service platform, wherein the physical resources in the cloud platform include recycling-related facilities and knowledge resources, and the cloud service request is decomposed by the cloud platform. According to the specific resource type requested, the decision module dynamically organizes and allocates service resources to form a recycling cloud service; The processing module is used to establish a disassembly cloud service platform based on the disassembly information and the recycling network model. The disassembly cloud service platform receives the disassembly request of the disassembly task and allocates disassembly resources to complete the equipment recycling.

Images