Full-automatic assembly line efficiency improving system for solar thermal collector
By designing a fully automatic assembly line efficiency improvement system for solar collectors and using intelligent optimization algorithms to optimize process scheduling, the problems of low production efficiency and low equipment utilization of traditional assembly lines are solved, and the production efficiency is improved and the efficient use of resources is achieved.
Patent Information
- Application Number
- CN202510205400.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-24
- Publication Date
- 2025-05-30
AI Technical Summary
Due to the lack of flexible allocation mechanism of traditional solar collectors, the production efficiency is low and the equipment utilization rate hovers at a low level, resulting in extended production cycles and waste of resources.
A fully automatic assembly line efficiency improvement system is designed, including process analysis unit, scheduling optimization unit and process scheduling unit. The process analysis unit conducts detailed analysis of each process and constructs the process sequence; the scheduling optimization unit uses intelligent optimization algorithm to encode the process sequence into chromosomes, constructs the fitness function, and optimizes the process scheduling scheme through tournament selection, partial matching crossover and mutation operations.
The optimization of process scheduling has been achieved, which significantly improves production efficiency and equipment utilization, shortens the production cycle, and saves computing resources and time costs.
Smart Images
Figure CN120065946A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the production of solar collectors, and more particularly to a system for improving the efficiency of a fully automatic production line of solar collectors. Background Art
[0002] New energy production is an important technology. In the field of solar collector production, the efficient operation of a fully automatic production line is crucial for improving production capacity and economic benefits. However, there are many challenges in the actual production of its many complex processes, such as component assembly, heat-absorbing coating treatment, shell encapsulation, etc.
[0003] Among them, the disadvantages of the traditional fixed process sequence arrangement mode are obvious. Due to the lack of a flexible allocation mechanism, the production efficiency is often in trouble. The delay of a certain previous process will cause the subsequent processes to wait for a long time, resulting in the idle waste of equipment and manpower, greatly lengthening the overall production cycle. In the heat-absorbing coating treatment process, if the equipment preheating does not meet the standard, the subsequent assembly process can only stop, and during this period, other equipment is idle and cannot be reasonably allocated to other time-consuming processes, resulting in the equipment utilization rate remaining at a low level, seriously hindering the release of the production line's production capacity and the efficient use of resources. To solve this technical problem, we provide a system for improving the efficiency of a fully automatic production line of solar collectors. Summary of the Invention
[0004] The purpose of the present invention is to provide a system for improving the efficiency of a fully automatic production line of solar collectors to solve the problems raised in the above background art.
[0005] To achieve the above object, a system for improving the efficiency of a fully automatic production line of solar collectors is provided, including a process analysis unit, a scheduling optimization unit, and a process scheduling unit;
[0006] The process analysis unit is used to analyze each process in the fully automatic production line of solar collectors, determine the processing time range, required equipment type and quantity, parallel processing process relationship, and mandatory sequential constraint between processes of each process, and construct them into a process sequence;
[0007] The scheduling optimization unit uses an intelligent optimization algorithm to encode the process sequence into chromosomes, obtaining the encoded representation of the processes. Then, it constructs a fitness function based on the production efficiency target of the assembly line to obtain a quantitative index for evaluating the quality of the process scheduling plan. Next, it selects the individual with the highest fitness from the current population into the next-generation population through the tournament selection method, and performs a crossover operation on the selected individuals using the partially mapped crossover method according to the set crossover probability to obtain a new process scheduling combination. Then, it mutates the chromosome genes in the new process scheduling combination according to the set mutation probability. Finally, it terminates the algorithm iteration process based on the mutation operation result by setting the upper limit of the iteration times to obtain the optimal process scheduling plan;
[0008] The process scheduling unit is used to output the optimal process scheduling plan and apply it to the actual production control of the solar collector full-automatic assembly line.
[0009] As a further improvement of this technical solution, when constructing the process sequence, the process analysis unit stores the parameter information of each process in the form of a data structure table. Among them, the processing time range is stored in two fields of the minimum value and the maximum value, the required equipment type and quantity are stored in independent fields respectively, the process relationship that can be processed in parallel is stored in the form of an associative array, and the mandatory precedence constraint between processes is stored in the process sequence structure.
[0010] As a further improvement of this technical solution, the scheduling optimization unit includes an encoding initial module. When encoding the process sequence into chromosomes, the encoding initial module arranges the genes corresponding to the processes on the chromosomes in the order of the inherent numbers of the processes, specifically as follows:
[0011] Obtain the process sequence information constructed by the process analysis unit, where each process has a unique fixed number;
[0012] Create an empty chromosome data structure with a length equal to the length of the pre-set process sequence, and starting from the first process in the process sequence, place the genes corresponding to each process in the corresponding positions of the chromosome in the order of the numbers.
[0013] As a further improvement of this technical solution, the scheduling optimization unit includes a search mutation module. The production efficiency targets in the search mutation module include minimizing the total production cycle and maximizing the equipment utilization rate, specifically as follows:
[0014] The total production cycle is obtained by accumulating the start time, processing time, waiting time, and transfer time between processes. The equipment utilization rate is calculated by statistically calculating the ratio of the working time of each equipment during the production cycle to the total time;
[0015] The waiting time between the processes is determined based on the difference between the end time of the previous process and the start time of this process, and the transfer time is determined based on the standard time-consuming for the process to transfer between different workstations on the assembly line.
[0016] As a further improvement of this technical solution, the way that the search and mutation module constructs a fitness function according to the production efficiency target of the assembly line to obtain a quantitative index for evaluating the quality of the process scheduling scheme specifically includes:
[0017] Starting from the first process, calculate the start time, end time of each process in turn, as well as the waiting and transfer time between it and the next process, and finally obtain the total time for the entire assembly line to complete all processes.
[0018] For each piece of equipment, count its working time during the entire production process, including the actual processing time, start-up time, preheating time, and maintenance time in each process, and obtain the utilization rate of the equipment based on this.
[0019] Integrate the utilization rates of all equipment, use the weighted average method to obtain the overall equipment utilization rate, and construct a fitness function based on the overall equipment utilization rate and the total time.
[0020] As a further improvement of this technical solution, the way that the search and mutation module performs a crossover operation on the selected individuals using the partially matched crossover method according to the set crossover probability to obtain a new process scheduling combination is as follows:
[0021] Suppose the current population contains N individuals, and each individual represents a process scheduling scheme and corresponds to a fitness function. Each time k individuals are selected for fitness function comparison in each round, for a total of N / k rounds, and the individual with the largest fitness function in each round is selected to form a candidate set for the next generation population.
[0022] Set the crossover probability, and select the individuals participating in the crossover from the candidate set according to the crossover probability, group them in pairs. For each pair of individuals, set their chromosome length, and generate two different random numbers within a preset range as the crossover point positions.
[0023] For each pair of parental chromosomes in each group, determine the gene segments between their crossover points, then exchange the gene segments to obtain intermediate chromosomes, and perform conflict processing on the intermediate chromosomes to obtain two offspring chromosomes.
[0024] As a further improvement of this technical solution, the search and mutation module then performs a mutation operation on the genes in the chromosome according to the set mutation probability, specifically as follows:
[0025] Set the mutation probability, generate a random number for the offspring chromosome, and the random number is used to determine whether the chromosome participates in the mutation operation;
[0026] For the chromosome participating in mutation, randomly select a gene position in the chromosome for mutation to obtain the mutated offspring chromosome.
[0027] As a further improvement of this technical solution, the scheduling optimization unit includes a control output module, and the control output module terminates the algorithm iteration process by setting the upper limit of the number of iterations to obtain the optimal process scheduling plan, specifically as follows:
[0028] Set the upper limit of the number of iterations I by minimizing the total production cycle and maximizing the equipment utilization rate max , in the first iteration, use the encoding initial module to encode the process sequence into a chromosome, generate the initial population, and calculate the fitness function of each individual;
[0029] In subsequent iterations, based on the previous generation population, select individuals with higher fitness to enter the next generation population through the tournament selection method, then perform crossover operations on the selected individuals according to the set crossover probability using the partially matched crossover method, and then mutate the genes in the chromosome according to the set mutation probability. After obtaining the new generation population, calculate the fitness function of each new individual in the same way;
[0030] After each iteration completes the population update and fitness calculation, compare the fitness functions of all individuals in the current population, find the individual with the highest fitness function, record this individual and its fitness function, and determine whether the current number of iterations reaches the set upper limit of the number of iterations, and output the chromosome combination at the end of the iteration.
[0031] Compared with the prior art, the beneficial effects of the present invention are:
[0032] In a system for improving the efficiency of a fully automatic solar collector production line, the scheduling optimization unit brings key improvements to the fully automatic solar collector production line. Among them, in the process encoding link, the process sequence is encoded into a chromosome, which effectively improves the optimization efficiency. The construction of the fitness function organically combines the minimization of the total production cycle and the maximization of the equipment utilization rate, accurately quantifies the quality of the process scheduling plan, provides a clear optimization path for the algorithm, ensures rapid and accurate positioning of the efficient plan, greatly saves computing resources and time costs, improves production efficiency, and finally through the cooperation of the tournament selection method, the partially matched crossover method and the mutation operation, the optimal process combination is screened out layer by layer to improve the equipment utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Figure 1 It is the overall block diagram of the present invention.
[0034] The meanings of the various labels in the figure are:
[0035] 1. Process analysis unit; 2. Scheduling optimization unit; 21. Initial coding module; 22. Search and mutation module; 23. Control output module; 3. Process scheduling unit. Detailed implementation mode
[0036] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.
[0037] The present invention provides a system for improving the efficiency of a full-automatic production line of solar collectors. Please refer to Figure 1 as shown, including a process analysis unit 1, a scheduling optimization unit 2, and a process scheduling unit 3;
[0038] The process analysis unit 1 is used to analyze each process in the full-automatic production line of solar collectors, determine the processing time range, required equipment type and quantity, parallelizable process relationships, and mandatory precedence constraints between processes, and construct them into a process sequence.
[0039] The full-automatic production line of solar collectors has numerous and complex processes. Traditional data management methods are difficult to clearly and accurately present process details. Using this data structure table can centrally integrate various process parameters. When constructing a process sequence, the differences in the characteristics of different processes need to be distinguished by special fields. The minimum and maximum values of the processing time range are conducive to arranging production time periods and resources. The equipment requirement fields help with overall equipment planning. The parallelizable process relationships are stored in an associative array, which fits its one-to-many logical characteristics and can effectively sort out the parallel relationships between processes, providing a precise and well-organized data basis for subsequent scheduling optimization and overcoming the problems of scattered data and fuzzy relationships in the traditional mode.
[0040] Create a data structure table. Each row in the table corresponds to a process in the full-automatic production line of solar collectors. The table structure stores the process parameters in categories, which is clear at a glance and inputs reliable data for the scheduling optimization unit 2.
[0041] For the processing time range, set two fields, "minimum processing time" and "maximum processing time", in the table. For example, for a certain component assembly process, after analysis, its shortest processing time is 10 minutes and the longest is 15 minutes. Then fill in 10 and 15 in these two fields of the corresponding row, reducing the complexity and error rate of data management.
[0042] Required equipment type and quantity fields. For each process, specify the name of the required equipment type and record it in the "Required equipment type" field, such as "welding equipment"; at the same time, fill in the quantity of this type of equipment required for this process in the "Required equipment quantity" field, for example, 2 units.
[0043] The relationship of processes that can be processed in parallel is stored through an associative array. An associative array is constructed. The key of the array is the process number of the current process, and the value is an array that contains all the process identifiers that can be processed in parallel with it. For example, if process A can be processed in parallel with process C and process D, then in the associative array, with "A" as the key, the corresponding value array is ["C", "D"]. The scheduling algorithm arranges the process order according to the precise time range, allocates resources according to the equipment requirements, and processes the processes in parallel according to the parallel relationship, avoiding resource idleness and process conflicts, and improving the overall scheduling accuracy and efficiency.
[0044] The mandatory precedence constraints between processes are stored using a process sequence structure. Each process is regarded as a node in the process sequence. If process B must be carried out after process A, then a directed edge is created in the process sequence from node A to node B. By traversing all such precedence relationships between processes, a complete process sequence is constructed, thus clearly representing the sequential logic of the entire assembly line process. In this way, through this data structure table and related storage methods, the process analysis unit can effectively organize and manage the key parameter information of each process, providing an accurate and easy-to-process data basis for the subsequent scheduling optimization unit.
[0045] The scheduling optimization unit 2 uses an intelligent optimization algorithm to encode the process sequence into chromosomes to obtain the encoded representation of the process. Then, according to the production efficiency goal of the assembly line, a fitness function is constructed to obtain a quantitative index for evaluating the quality of the process scheduling scheme. Then, the individual with the highest fitness is selected from the current population into the next generation population through the tournament selection method, and the selected individuals are crossed using the partially mapped crossover method according to the set crossover probability to obtain a new process scheduling combination. Then, the chromosome genes in the new process scheduling combination are mutated according to the set mutation probability. Finally, according to the results of the mutation operation, the algorithm iteration process is terminated by setting the upper limit of the number of iterations to obtain the optimal process scheduling scheme.
[0046] First, the encoding initial module 21 obtains the constructed process sequence information from the process analysis unit 1. Each process has a unique inherent number. The processes in the solar collector production assembly line include "Part processing 1 (number 01)", "Part processing 2 (number 02)", "Assembly (number 03)", "Inspection (number 04)".
[0047] Then, the initial encoding module 21 creates an empty chromosome data structure with a length equal to the length of the pre-set process sequence. Starting from the first process in the process sequence, in the order of numbering, the genes corresponding to each process are sequentially placed at the corresponding positions on the chromosome. The gene corresponding to "Part Processing 1 (numbered 01)" is placed at the first position on the chromosome, the gene corresponding to "Part Processing 2 (numbered 02)" is placed at the second position on the chromosome, and so on, until all the genes corresponding to the processes are arranged in order in the chromosome.
[0048] In this process, it is necessary to ensure that the mapping relationship between genes and processes is accurate and error-free, and the length of the chromosome always remains consistent with the length of the process sequence. This encoding method presents the process sequence in a concise form that is convenient for subsequent algorithm operations, providing a basic data structure support for the selection, crossover, and mutation of intelligent optimization algorithms in the search and optimization process of process scheduling schemes, enabling the algorithm to effectively explore and evaluate different process permutations and combinations to seek the optimal process scheduling scheme, thereby improving the production efficiency and resource utilization rate of the fully automatic solar collector production line.
[0049] Since the efficient operation of the fully automatic solar collector production line not only depends on a short total production cycle but also requires full utilization of equipment resources and improvement of equipment utilization rate, only focusing on a single indicator cannot comprehensively evaluate the pros and cons of the process scheduling scheme. Therefore, it is necessary to construct a fitness function that comprehensively considers the total production cycle and equipment utilization rate, and the intelligent optimization algorithm requires a quantified fitness function to evaluate the quality of each individual for selection, crossover, and mutation operations.
[0050] The search and mutation module 22 starts from the first process and sequentially calculates the start time, end time, and waiting and transfer times between each process and the next process, and finally obtains the total time T(S) for the entire production line to complete all processes. The total production cycle is obtained by accumulating the start time, processing time, and waiting and transfer times between processes. The equipment utilization rate is calculated by statistically calculating the ratio of the working time of each equipment during the production cycle to the total time;
[0051] By analyzing in detail the time parameters of each process and their mutual relationships, the impact of different scheduling schemes on the overall production duration can be accurately determined, avoiding evaluation biases caused by ignoring certain time factors;
[0052] For each piece of equipment i, its working time UT i during the entire production process is statistically calculated, including the actual processing time, start-up time, preheating time, and maintenance time in each process. Let the total available time of equipment i be AT i , then the utilization rate of equipment i This makes the evaluation of device resources more comprehensive and meticulous, helping to identify scheduling schemes where the device appears busy but actually has low utilization, so as to optimize them. Among them, the waiting time between processes is calculated based on the difference between the end time of the previous process and the start time of this process, and the transmission time is determined based on the standard time-consuming for the process to transfer between different workstations on the assembly line.
[0053] The search and mutation module 22 synthesizes the utilization rates of all devices and obtains the overall device utilization rate by using the weighted average method. where m is the total number of devices, and ω i is the weight of device i. The purpose of the fitness function is to comprehensively evaluate the quality of the process scheduling scheme, so that the scheme with a short total production cycle and high device utilization rate has a higher fitness. Therefore, the mathematical expression of the fitness function α(S) is where β and χ are weight coefficients. Through this fitness function, for each process scheduling scheme S, a fitness function α(S) can be calculated. This value is the quantitative index for evaluating the quality of the process scheduling scheme. In the intelligent optimization algorithm, different process scheduling schemes are compared and selected according to the fitness function. The scheme with a higher fitness function is better, thus guiding the algorithm to continuously search and optimize, and finally finding the optimal process scheduling scheme to improve the production efficiency of the solar collector full-automatic assembly line.
[0054] Among them, it is necessary to first clarify the number of individuals included in the current population. Let the number of individuals included in the current population be N individuals. Each individual represents a process scheduling scheme and corresponds to a fitness function calculated through the fitness function. At the same time, according to the setting, the predetermined number k selected each time during the tournament selection is 10% - 20% of the population size N. Let k = 15.
[0055] The search and mutation module 22 uses a random number generator to randomly select k = 15 individuals without repetition from the N individuals in the current population to increase the probability of obtaining excellent gene combinations. For these 15 selected individuals, their respective fitness functions α(S i ) are calculated respectively according to the constructed fitness function, guiding the algorithm to approach a better solution, where i = 1, 2,..., 15, and S i represents the process scheduling scheme corresponding to the i-th selected individual.
[0056] Then, the sizes of these 15 fitness functions are compared respectively, and the individual with the highest fitness function is selected and denoted as S max1, which is used to preserve the characteristics of high-quality solutions, so as to gradually optimize the overall quality of the population, and take it as the excellent individuals selected in the first round of the tournament to enter the candidate set of the next-generation population. Repeat the above processes of extraction, calculation of the fitness function, comparison and selection, and perform N / k rounds of tournament selection. In each round, an individual S with the highest fitness is selected. maxj , and jointly form the candidate set of the next-generation population.
[0057] Set the crossover probability to p c , and from the candidate set of the next-generation population selected by the tournament selection method, determine which individuals participate in the crossover operation according to the probability p c , balance the exploration and exploitation capabilities of the algorithm, and avoid excessive or insufficient gene recombination, as follows:
[0058] Generate a random number r between 0 and 1 for each individual. n , if r n < p c , then this individual participates in the crossover operation. Suppose it is finally determined that m individuals participate in the crossover. Group the m individuals participating in the crossover operation in pairs. If m is odd, there will be a group with only one individual, and this individual does not participate in the crossover and directly enters the next-generation population. For each pair of two individuals, assume the length of their chromosomes is L, and use a random number generator to generate two different random numbers r 1 、r 2 within the range of 1 to L - 1 as the crossover point positions, which is convenient for gene segment exchange operations, increases the randomness and diversity of gene combinations, and creates more potential excellent process scheduling schemes.
[0059] For each pair of two parental chromosomes P 1 、P 2 , search the mutation module 22 to determine their gene segments between the crossover points r 1 、r 2 . Denote the gene segment of the parental P 1 as P 1f = [P 1 [r 1 , P 1 [r 1 + 1], …, P 1 [r 2 ; Denote the gene segment of the parental P 2 as P 2f = [P 2 [r 1 , P 2 [r 1 + 1],..., P 2 [r 2 .
[0060] Then, gene segment exchange is carried out to obtain the intermediate chromosome M 1 and M 2 , where the construction of M 1 is to first copy the genes in P 1 except for P 1f in order, and then insert the genes in P 2f into the corresponding positions according to the relative order of the genes in P 2f in P 2 ; similarly, M 2 is to first copy the genes in P 2 except for P 2f , and then insert the genes in P 1f in relative order, fuse the gene characteristics of different individuals, introduce new scheduling ideas, so as to expand the solution space and increase the possibility of finding a better solution.
[0061] Let P 1 = [1, 2, 3, 4, 5, 6, 7, 8], P 2 = [5, 4, 6, 7, 8, 1, 2, 3], the crossover point r 1 = 2, r 2 = 5, then P 1f = [2, 3, 4, 5], P 2f = [4, 6, 7, 8], where the construction of M 1 is specifically as follows:
[0062] First, copy the genes in P 1 except for P 1f to get [1], and then insert the genes in P 2f in order to get M 1 = [1, 4, 6, 7, 8, 6, 7, 8], where 4 corresponds to the second position in P 2 , 6 corresponds to the third position, and so on. Similarly, M 2 = [5, 2, 3, 4, 5, 1, 2, 3].
[0063] Then, conflict handling is performed on the intermediate chromosomes M 1 and M 2 to ensure that legal offspring chromosomes are obtained. Since exchanging gene segments may cause some genes to appear repeatedly or be missing in the chromosome, it is necessary to correct them through mapping relationships. For M 1 , check for duplicate genes, such as 6, 7, 8 in the above example. After finding the duplicate genes, replace them in M 2 according to the positions of the corresponding genes in P 1 . For example, the second 6 in M 1 was originally in the third position in P 2 , corresponding to P 1If it is 3 in , replace it with 3. Similarly, replace the second 7 with 4 and the second 8 with 5 to obtain the offspring chromosome C. 1 , then C 1 = [1, 4, 3, 4, 5, 3, 4, 5]. Through the above complete tournament selection method combined with the partially matched crossover method, the search and mutation module 22 can, on the basis of ensuring the inheritance of excellent genes, generate new gene combinations through crossover operations, thereby obtaining new process scheduling combinations, increasing the diversity of the population, guiding the intelligent optimization algorithm to search for a better process scheduling scheme, and effectively improving the production efficiency of the fully automatic solar collector assembly line.
[0064] Taking C 1 = [1, 4, 3, 4, 5, 3, 4, 5] as an example, if gene 1 corresponds to "Part processing 1" and gene 4 corresponds to "Assembly", then the new process scheduling combination is to first perform "Part processing 1", then "Assembly", and then "Part processing 3". This new process scheduling combination may change the waiting time between processes, equipment utilization, etc., thereby affecting the production efficiency of the entire assembly line. It is the key for the intelligent optimization algorithm to explore a better solution.
[0065] The search and mutation module 22 sets the mutation probability to p m , for the chromosome C 1 corresponding to the new process scheduling combination obtained through the crossover operation, generate a random number r between 0 and 1 for each chromosome m , if r m < p m , then this chromosome participates in the mutation operation. Suppose the random number r 1 corresponding to C m1 < 0.1, then C 1 participates in the mutation.
[0066] In C 1 = [1, 4, 3, 4, 5, 3, 4, 5], randomly generate a position number. Suppose it is the 3rd position. Mutate the gene 3 at this position, that is, replace it with another gene that has not appeared in this chromosome. Suppose the gene 3 is mutated to gene 7 to obtain the mutated chromosome C 1 ' = [1, 4, 7, 4, 5, 3, 4, 5]. This completes the transformation from the chromosome corresponding to the new process scheduling combination to the mutated chromosome. The mutated chromosome represents a new process scheduling possibility and will continue to participate in the subsequent algorithm iteration process, such as recalculating the fitness function, performing selection, crossover, etc., to continuously explore a better process scheduling scheme to improve the production efficiency of the fully automatic solar collector assembly line.
[0067] In the first iteration, the encoding initial module 21 encodes the process sequence into chromosomes to generate an initial population, and calculates the fitness function of each individual. In subsequent iterations, based on the previous generation population, individuals with higher fitness are selected into the next generation population through the tournament selection method. Then, the partially matched crossover method is used to perform crossover operations on the selected individuals according to the set crossover probability, and the genes in the chromosomes are mutated according to the set mutation probability to update the individuals in the population. After obtaining a new generation population, the fitness function of each new individual is also calculated.
[0068] After each iteration completes population update and fitness calculation, compare the fitness functions of all individuals in the current population, and find the individual with the highest fitness function, and record this individual and its fitness function.
[0069] Among them, after each iteration ends, the control output module 23 judges the current iteration number I θ whether it has reached the set upper limit of the iteration number I max , if I θ < I max , then continue with the next iteration. If I θ = I max , then terminate the iteration process. After the iteration terminates, from the previously recorded optimal individuals and their fitness functions for each iteration, screen out again the individual with the highest fitness function, and the process scheduling scheme corresponding to it is the final optimal process scheduling scheme sought.
[0070] The control output module 23 traverses the list of all recorded optimal individual fitness functions, finds the individual corresponding to the maximum value, and restores this individual to the actual process arrangement order according to the correspondence between genes and processes, and it can be applied to the actual production control of the solar collector full-automatic production line, so as to improve the production efficiency of the production line based on the established optimization goal.
[0071] The process scheduling unit 3 is used to output the optimal process scheduling scheme and apply it to the actual production control of the solar collector full-automatic production line.
[0072] In the present invention, the process analysis unit deeply analyzes various aspects of the process to determine the processing time range of each process, the details of the required equipment, the parallel and sequential relationships, constructs a process sequence in a data structure table, laying a foundation for the follow-up. The encoding initial module 21 encodes the processes into chromosomes according to the inherent numbers of the processes, starts the intelligent optimization process. The search mutation module 22 constructs a fitness function based on the total production cycle and equipment utilization rate according to the production efficiency target of the assembly line, retains superior individuals through tournament selection, generates new process combinations through partially matched crossover, jumps out of the local optimal solution through mutation operations, and conducts iterative improvement, and applies the optimal solution to the actual production control of the assembly line. Each unit cooperates with each other to improve the production efficiency of the solar collector assembly line.
[0073] The foregoing has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and the descriptions in the specification are only preferred examples of the present invention, and are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of the present invention claimed is defined by the appended claims and their equivalents.
Claims
1. A solar collector fully automatic assembly line efficiency improvement system, characterized in that: It includes a process analysis unit (1), a scheduling optimization unit (2) and a process scheduling unit (3); The process analysis unit (1) is used to analyze each process of the solar collector automatic assembly line, determine the processing time interval of the process, the type and quantity of equipment required, the relationship between the processes that can be processed in parallel, and the mandatory sequence constraints between the processes, and generate a process sequence; The scheduling optimization unit (2) uses an intelligent optimization algorithm to encode the process sequence and express the process sequence in a chromosome form; constructs a fitness function based on the production efficiency target of the assembly line to quantitatively evaluate the quality of the process scheduling plan; selects the individuals with the best fitness from the current population through a tournament selection method to enter the next generation; performs a crossover operation on the selected individuals using a partial matching crossover method according to a set crossover probability to generate a new process scheduling combination; performs a mutation operation on the chromosome gene of the offspring according to a set mutation probability; terminates the algorithm iteration process based on a set upper limit of the number of iterations, and outputs the optimal process scheduling plan; The process scheduling unit (3) is used to output an optimized process scheduling plan and apply it to the actual production control of the solar thermal collector full-automatic assembly line.
2. The solar collector fully automatic assembly line efficiency improvement system according to claim 1, characterized in that: When constructing a process sequence, the process analysis unit (1) uses a data structure table to store parameter information of each process, wherein the processing time range is stored through a minimum value field and a maximum value field, the required equipment type and quantity are recorded in independent fields, the process relationship that can be processed in parallel is represented by an associative array, and the mandatory sequence constraints between processes are stored in a sequence structure.
3. The solar collector fully automatic assembly line efficiency improvement system according to claim 1 is characterized by: The scheduling optimization unit (2) comprises a coding initialization module (21), and the coding initialization module (21) is used to encode the process sequence into chromosomes, and the operation process is as follows: Acquire process sequence information constructed by the process analysis unit (1), wherein each process has a unique fixed number; A chromosome data structure whose length is consistent with the length of the preset process sequence is constructed, and the gene corresponding to each process is mapped to the corresponding position of the chromosome in sequence according to the order of the process's inherent number.
4. The solar collector fully automatic assembly line efficiency improvement system according to claim 1, characterized in that: The scheduling optimization unit (2) includes a search variation module (22) for optimizing process scheduling based on production efficiency targets, wherein the production efficiency targets include minimizing the total production cycle and maximizing the equipment utilization rate, specifically: The total production cycle is calculated and determined by accumulating the start time, processing time, waiting time between processes and process transfer time of each process; The equipment utilization rate is calculated based on the ratio of the effective working time of the equipment to the total time during the production cycle; The waiting time between processes is determined according to the time difference between the completion time of the previous process and the allowed start time of the next process, and the transfer time is calculated based on the standard transfer time of the process between different workstations in the assembly line.
5. The solar collector fully automatic assembly line efficiency improvement system according to claim 4 is characterized by: The search mutation module (22) constructs a fitness function based on the production efficiency target of the assembly line to quantitatively evaluate the quality of the process scheduling scheme, specifically including: Starting from the first process, calculate the start time, end time, waiting time and transmission time between each process and the subsequent processes, so as to determine the total production time for the assembly line to complete all processes; Statistics are collected on the working time of each equipment during the production process, including actual processing time, startup time, warm-up time and maintenance time, so as to calculate the equipment utilization rate; Combining the utilization rates of all equipment, the weighted average method is used to calculate the overall equipment utilization, and a fitness function is constructed based on the overall equipment utilization and the total production time.
6. The solar collector fully automatic assembly line efficiency improvement system according to claim 5, characterized in that: The search mutation module (22) uses the partial matching crossover method to perform a crossover operation on the selected individuals according to the set crossover probability to obtain a new process scheduling combination, which is specifically as follows: Assume that the current population contains N individuals, each of which represents a process scheduling scheme and corresponds to a fitness function. In each round, k individuals are selected for fitness function comparison, for a total of N / k rounds. The individual with the largest fitness function in each round is selected to form the candidate set for the next generation population. Set the crossover probability, and select individuals participating in the crossover from the candidate set according to the crossover probability, group them into two groups, set the chromosome lengths of the two individuals in each group, and generate two different random numbers within the preset range as the crossover point positions; For each group of two parent chromosomes, their gene segments between the crossover points are determined, and then the gene segments are exchanged to obtain an intermediate chromosome, and the conflict of the intermediate chromosome is resolved to obtain two daughter chromosomes.
7. The solar collector fully automatic assembly line efficiency improvement system according to claim 6, characterized in that: The search mutation module (22) then performs mutation operations on the genes in the offspring chromosomes in the new process scheduling combination according to the set mutation probability, as follows: Set the mutation probability, and generate a random number for the offspring chromosome, which is used to determine whether the chromosome participates in the mutation operation; For the daughter chromosomes involved in the mutation, a gene position in the daughter chromosome is randomly selected for mutation to obtain a mutated new daughter chromosome.
8. The solar collector fully automatic assembly line efficiency improvement system according to claim 3 is characterized by: The scheduling optimization unit (2) comprises a control output module (23), wherein the control output module (23) terminates the iteration process of the optimization algorithm by setting an upper limit of the number of iterations, and generates an optimal process scheduling scheme. The specific process is as follows: The upper limit of the number of iterations is set to minimize the total production cycle and maximize the equipment utilization rate. In the first iteration, the process sequence is encoded into chromosomes using the encoding initial module (21), an initial population is generated, and the fitness function of each individual is calculated; In subsequent iterations, based on the previous generation population, the individuals with the highest fitness are selected through the tournament selection method to enter the next generation population, and then the partial matching crossover method is used to perform crossover operations on the selected individuals according to the set crossover probability, and then the genes in the chromosomes are mutated according to the set mutation probability. After obtaining the new generation population, the fitness function of each new individual in the new generation population is also calculated; After completing the population update and fitness calculation in each iteration, compare the fitness functions of all individuals in the current population, find the individual with the highest fitness function, record the individual and its fitness function, and determine whether the current number of iterations reaches the set upper limit of iterations. If so, output the chromosome combination when the iteration terminates.