Silkworm breeding base low-carbon planning method based on NSGA-II model

Abstract

The invention relates to a silkworm breeding base low-carbon planning method based on an NSGA-II model. The silkworm breeding base low-carbon planning method is used for improving the overall operation efficiency. The silkworm breeding base low-carbon planning method specifically comprises the following steps: a data preprocessing step; establishing a data target; determining constraint conditions; and a model algorithm iteration step. According to the silkworm breeding base layout optimization method, in the multi-objective optimization solving process, the biological survival rate is comprehensively considered in the logistics cost and carbon emission objectives, the problem that an existing silkworm breeding base layout optimization method depends on a specific layout mode can be effectively solved, the applicability and flexibility of the algorithm under different layout requirements and environment conditions are improved, and the silkworm breeding base layout optimization method is suitable for large-scale popularization and application. And reference significance is also provided for other biological layout researches.

Description

Technical Field

[0001] The invention relates to the technical field of biological production layout, and more specifically, to a low-carbon planning method for a silkworm breeding base based on a NSGA-II model. Background Art

[0002] In recent years, low-carbon economy and sustainable development have become research hotspots in the global agricultural and industrial fields. The rationality of the layout of silkworm breeding bases not only affects production efficiency, but is also directly related to carbon emissions and resource utilization. However, traditional base layout methods often ignore the influence of biological characteristics, which can easily lead to resource waste and biosafety issues. Therefore, there is an urgent need for a silkworm breeding base layout optimization method that can combine biological characteristics and low-carbon goals.

[0003] On the basis of analyzing the existing layout, the present invention pays special attention to multiple factors that are different from the general production workshop layout optimization research. First, the present invention studies the survival rate factor in biological breeding. In the process of mulberry leaf transportation, the survival rate of silkworms is a key factor because the survival rate directly affects the demand for mulberry leaves by the silkworm body. The higher the survival rate, the mulberry leaf consumption of the silkworm body will also increase, resulting in an increase in the demand for mulberry leaf transportation. Therefore, in the mulberry leaf transportation research of the present invention, the efficient operation of the logistics system is ensured by dividing the logistics volume into mulberry leaf logistics affected by the survival rate of silkworms and other logistics not affected by the survival rate of silkworms. In addition, as the survival rate of silkworms in the breeding farm increases, their number increases, and accordingly, the amount of mulberry leaves required will also increase, resulting in an increase in the cost of mulberry leaves. Secondly, carbon emissions are taken into account in the layout problem. In addition to the common equipment operation energy consumption and site energy consumption in general production workshops, the metabolic heat during the growth of silkworms is also studied.

[0004] Current intelligent optimization algorithms usually face conflicts and balances between multiple objectives when constructing optimization objective functions. Taking the optimization of biological layout of silkworm breeding bases as an example, it is necessary not only to reduce logistics costs and carbon emissions, but also to ensure biological safety and efficient operation. However, many existing intelligent optimization algorithms are often troubled by problems such as local convergence and premature convergence when dealing with multi-objective functions, making it difficult for the algorithm to find the true optimal solution globally. Therefore, it is of great significance to design an optimization that can effectively avoid local optimal traps and achieve global optimal solutions. Summary of the invention

[0005] The present invention overcomes the shortcomings of the prior art, provides a low-carbon planning method for a silkworm breeding base based on the NSGA-II model, which realizes low-carbon optimization of the layout of a silkworm breeding base and improves the overall operational efficiency.

[0006] The technical solution of the present invention is as follows:

[0007] A low-carbon planning method for a silkworm breeding base based on the NSGA-II model specifically comprises the following steps:

[0008] S1 Data preprocessing step: obtain the basic data of the silkworm breeding base, classify the basic data, determine the layout constraints of the classified data; collect parameters related to the growth environment of silkworms;

[0009] S2 Establish data target step: Combine historical data of parameters related to the silkworm's growth environment to evaluate the impact on the survival rate of silkworms; establish a multi-objective optimization model for the layout of silkworm breeding bases, with the goal of reducing operating costs and carbon emissions;

[0010] S3 step of determining constraints: layout constraints include spacing constraints, boundary constraints, special location constraints, and biological isolation constraints;

[0011] The spacing constraint is that the minimum spacing between any two facilities in the horizontal direction or in the vertical direction should be greater than a certain threshold;

[0012] The boundary constraint is that each work unit cannot exceed the size of the workshop when it is laid out in the workshop;

[0013] Special location constraints are certain conditions that must be met by special operation units in the base, including the silkworm tray cleaning and disinfection room as a special operation unit, which must be close to ventilation;

[0014] Biological isolation constraints To ensure that there is no cross infection between breeding areas of different silkworm ages, the regional layout is designed according to the principle of biosafety. The actual distance between areas of different silkworm ages must be greater than the preset minimum biological isolation distance;

[0015] S4 model algorithm iteration steps: instantiate the decision variables of the NSGA-II model, generate population individuals, and determine the layout plan; each element of the population individual represents the location coordinates or layout status of the facility in the base; the specific steps for confirming the layout plan are as follows:

[0016] S4.1 Population differentiation step: Divide the population into two sub-populations, sub-population 1 and sub-population 2;

[0017] Subpopulation 1 is a population close to the border: it is used to optimize the silkworm tray cleaning and disinfection area when it needs to be close to the base border. The facility is initially placed near the layout border to ensure that the facility maintains a minimum distance from the border.

[0018] Subpopulation 2 is a population close to the border: it is used to optimize the silkworm tray cleaning and disinfection area when it is close to the border but not close to it. The facility is randomly allocated in the area close to the border.

[0019] S4.2 Initial parent generation judgment processing steps: determine whether the two sub-populations are merged to generate the initial parent population; if so, proceed to the next step to combine the parent and sub-populations; if not, perform non-dominated sorting and generate sub-populations through selection, crossover, and mutation;

[0020] S4.3 New parent generation judgment processing steps: After the parent generation and the offspring population are combined in step S4.2, it is determined whether a new parent generation is generated; if so, selection, crossover and mutation are performed to generate the offspring population; if not, the objective function value is calculated, a fast non-dominated sort is performed, and then the congestion degree is calculated;

[0021] S4.4 Individual sorting step: According to the multi-objective values ​​(logistics cost and carbon emissions) of each individual, the population is non-dominated and the non-dominated sorting level of each individual is determined;

[0022] S4.5 New generation parent step: adopt the improved elite retention strategy to retain its excellent individuals and generate a new generation parent population;

[0023] S4.6 Determine the layout plan step: return to step 4.3 with the new generation parent population as input and continue iterating; if the number of iterations reaches the preset maximum threshold MaxIter or meets other convergence conditions, the iteration ends and the non-dominated solution set of the Pareto algorithm is output to determine the layout plan.

[0024] Furthermore, the basic data in the S1 data acquisition step include the location, size, inter-facility logistics volume, mulberry leaf quality, silkworm survival rate over the years, equipment energy consumption, and carbon emission factor information of each facility.

[0025] Furthermore, parameters related to the growth environment of silkworms include temperature and humidity, light, air circulation and epidemic prevention measures.

[0026] Furthermore, the specific formula for reducing operating costs in step S2 is as follows:

[0027]

[0028] MinZ o =MinZ1+MinZ2+MinZ3 Formula (4)

[0029] Among them, Z0 represents the total daily operating cost of the base, Z1 represents the total daily transportation cost of the base, Z2 represents the total daily mulberry leaf cost of the base, and Z3 represents the daily construction depreciation cost of the base;

[0030] C ij represents the unit material handling cost per unit distance between facility i and facility j, d ij represents the distance between facilities i and j, t represents the tth day in the growth cycle of silkworms, p represents a growth cycle of silkworms, Bt represents the number of trays of silkworms per day for t-year-old silkworms, B0 represents the initial number of trays of silkworms on the first day of age, ρ t represents the number of mulberry leaf trays required for each tray of silkworms at age t per day, ρ0 represents the initial number of mulberry leaf trays required for each tray of silkworms at day 1, and λ t represents the average survival rate of silkworms in a cycle, f′ j represents the amount of materials handled between facilities i and j except for mulberry leaves, and n is the total number of facilities;

[0031] q t is the mass of mulberry leaves, c s represents the unit cost of mulberry leaves;

[0032] c c It represents the construction cost per unit area of ​​the base, a represents the total length of the sericulture base, b represents the total width of the sericulture base, and ε is the depreciation coefficient.

[0033] Furthermore, the carbon emissions in step 2 include equipment operation energy consumption, site energy consumption and silkworm metabolic heat, and the optimization formula is as follows:

[0034]

[0035] Among them, ω represents the carbon emission coefficient, that is, the ratio of carbon dioxide emissions generated by unit energy consumption; P represents the total power input of the equipment inside the base; T d represents the daily working hours of the equipment; CDD represents the annual working days of the equipment; t a Indicates the daily operation time of the air conditioning equipment inside the base; E k Indicates the site energy consumption coefficient of the base, that is, the energy consumption coefficient of the air conditioner in the base during operation, and the seasonal weight coefficient; α k Indicates the air conditioning energy consumption ratio in a certain season; l i represents the length of facility i, w i Represents the width of facility i.

[0036] N is the number of silkworms in the base, q is the metabolic heat of each silkworm, β is an adjustment factor to represent the additional impact of silkworm metabolic heat on air conditioning energy consumption, and λ t It represents the average survival rate of silkworms in one cycle.

[0037] Furthermore, the survival rate of silkworms is expressed by mathematical expectation, and the survival days T(t), ambient temperature X(r), and mulberry leaf freshness Q(h) are used as fuzzy inputs. The specific formula is as follows:

[0038]

[0039] Among them, t ranges from 0 to 26 days of survival, r ranges from 25 to 31 degrees Celsius, and h ranges from 0 to 24 hours of storage. Through experiments, the input ambient temperature and mulberry leaf freshness are derived from the above fuzzy rules to generate the corresponding survival rate E(λ t ) Fuzzy output surface. By analyzing the surface, it is found that when the temperature reaches 28℃ and the freshness of mulberry leaves is 0 hours, the survival rate reaches the maximum value.

[0040] Furthermore, the spacing constraint satisfies the following formula:

[0041]

[0042] Among them, x is the horizontal direction, H ij is the minimum spacing in the horizontal direction, y is the vertical direction, V ij is the minimum vertical spacing, l i represents the length of facility i, w i Represents the width of facility i.

[0043] Furthermore, the boundary constraint satisfies the following formula:

[0044] and

[0045] and

[0046] S is the minimum distance between the facility and the boundary.

[0047] Furthermore, the special position constraint satisfies the following formula:

[0048] P s ={(x i ,y i )|x i =x s ,y i =y s}

[0049] Among them, (x s ,y s ) is the preset position of the cleaning and disinfection room.

[0050] Furthermore, the biological isolation constraint satisfies the following formula:

[0051] d ij ≥R ij

[0052] Among them, d ij is the actual distance between different silkworm age areas, R ij The preset minimum biological isolation distance.

[0053] Compared with the prior art, the present invention has the following advantages:

[0054] The present invention can effectively solve the problem of dependence on specific layout patterns in existing silkworm breeding base layout optimization methods by taking biological survival rate into account in the process of multi-objective optimization solution, improve the applicability and flexibility of the algorithm under different layout requirements and environmental conditions, and provide reference significance for other biological layout research. In addition, the improved NSGA-II model improves the elite retention mechanism by designing a dual-population co-evolution strategy and combining a multi-objective dispersion strategy with a migration distance screening strategy to ensure the diversity of the population and the uniformity of the solution distribution, significantly enhancing the adaptability and convergence speed of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a schematic diagram of the facility layout coordinates;

[0056] Figure 2 is the fuzzy output surface plot of survival rate;

[0057] Figure 3 It is the flow chart of IMNSGA-II;

[0058] Figure 4 It is the Pareto front graph of IMNSGA-II and other algorithms;

[0059] Figure 5 This is the optimized layout of the mulberry silkworm base. DETAILED DESCRIPTION

[0060] The embodiments of the present invention are described in detail below, wherein the same or similar reference numerals throughout represent the same or similar elements or elements with similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention and cannot be used as a limitation of the present invention.

[0061] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the field to which the present invention belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

[0062] The reference numerals of the steps mentioned in each embodiment are only for the convenience of description and have no substantial sequence connection. Different steps in each specific embodiment can be combined in different sequences to achieve the purpose of the invention. In addition, some structures, modules and model methods not described in detail in the present invention can be implemented by conventional technical means, so they will not be described in detail.

[0063] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0064] like Figures 1 to 5 As shown, Figure 5 For example, the base is 120 meters long, 90 meters wide, and covers an area of ​​10,800 square meters. The production base is a regular rectangle, mainly including mulberry leaf storage room 1, 1-3 age silkworm room 2, silkworm tray replacement area 3, young silkworm feeding area 4, old silkworm feeding area 5, 4-age silkworm room 6, old silkworm tray storage room 7, young silkworm tray disinfection area 8, 5-age silkworm room 9, young silkworm tray storage room 10, old silkworm tray disinfection area 11, cocooning area 12, silkworm cocoon storage room 13, silkworm cocoon disinfection area 14, quality inspection center 15, a total of 15 areas.

[0065] A specific low-carbon planning method for a silkworm breeding base based on the NSGA-II model aims to minimize operating costs and carbon emissions, where the total layout area is variable. By incorporating biological characteristics into layout planning, the problems of biosafety and operational efficiency that traditional layout optimization methods have not fully considered are effectively solved. Under the premise of meeting the constraints, the corresponding work units can be randomly arranged at any position in the workshop, and the layout area can be minimized. This method shows excellent optimization performance in solving low-carbon biological layout optimization problems with specific position constraints, and can generate feasible layout plans more quickly and efficiently. Specifically, it includes the following steps:

[0066] S1 Data preprocessing steps: obtain the basic data of the silkworm breeding base, classify the basic data (including the classification of silkworm houses, mulberry leaf storage rooms, silkworm tray disinfection areas and other facilities), determine the layout constraints of the classified data; collect parameters related to the growth environment of silkworms. The basic data includes information such as the location, size, inter-facility logistics volume, mulberry leaf quality, silkworm survival rate over the years, equipment energy consumption, and carbon emission factors of each facility. That is, to achieve the construction of a mathematical model with the operation unit as the layout object, in order to minimize the operating logistics cost, carbon emissions and operation unit, the enveloping rectangular area is the objective function, and the decision variables (the location coordinates of each facility in this scheme) and constraints are determined.

[0067] S2 Establish data target step: collect parameters related to the silkworm's growth environment, such as temperature and humidity, light, air circulation and epidemic prevention measures, and combine the historical data of parameters related to the silkworm's growth environment to evaluate the impact on the silkworm's survival rate;

[0068] A multi-objective optimization model for the layout of silkworm breeding bases is established, with the goal of reducing operating costs and carbon emissions. The specific formula for reducing operating costs and minimizing operating costs is as follows:

[0069]

[0070] MinZ o =MinZ1+MinZ2+MinZ3 Formula (4)

[0071] Among them, Z0 represents the total daily operating cost of the base, Z1 represents the total daily transportation cost of the base, Z2 represents the total daily mulberry leaf cost of the base, and Z3 represents the daily construction depreciation cost of the base;

[0072] C ij represents the unit material handling cost per unit distance between facility i and facility j, d ij represents the distance between facilities i and j, t represents the tth day in the growth cycle of silkworms, p represents a growth cycle of silkworms, B t represents the number of trays of silkworms per day for t-year-old silkworms, B0 represents the initial number of trays of silkworms on the first day of age, ρ t represents the number of mulberry leaf trays required for each tray of silkworms at age t per day, ρ0 represents the initial number of mulberry leaf trays required for each tray of silkworms at day 1, and λ t represents the average survival rate of silkworms in a cycle, f′ ij It represents the material handling volume except mulberry leaves between facilities i and j, and n is the total number of facilities.

[0073] q t is the mass of mulberry leaves, c s Represents the unit cost of mulberry leaves.

[0074] c c It represents the construction cost per unit area of ​​the base, a represents the total length of the sericulture base, b represents the total width of the sericulture base, and ε is the depreciation coefficient.

[0075] Formula (1) is used to describe the goal of minimizing the material handling cost. Material handling mainly includes mulberry leaf transportation and other material transportation except mulberry leaves. In the refined silkworm breeding mode, the amount of mulberry used is determined by the number of surviving silkworms. As the survival rate of silkworms increases, the transportation volume of mulberry leaves also increases accordingly. f′ ij It represents the cost caused by the handling requirements of other materials besides mulberry leaves, including silkworm feeding, silkworm tray cleaning and silkworm cocooning logistics.

[0076] Formula (2) is used to describe the goal of minimizing the cost of mulberry leaf consumption. Because the greater the demand for mulberry leaves by silkworms, the greater the cost of mulberry leaf production. That is, the cost of mulberry leaves is directly related to the demand for mulberry leaves by silkworms. The more surviving silkworms there are, the greater the demand for mulberry leaves, which in turn leads to an increase in the cost of mulberry leaves.

[0077] Formula (3) describes the goal of minimizing the construction depreciation cost. Since the construction of the base involves a large amount of infrastructure investment, the construction cost has an important impact on the overall benefits of the base. Therefore, in the planning stage, it is necessary not only to pay attention to the breeding cost, but also to comprehensively consider the long-term impact of the initial construction investment on the benefits of the base, and strive to achieve the best balance between construction scale and operational efficiency. Given that silkworm breeding is subject to seasonal restrictions, the annual effective production cycle CDD is about 240 days. According to relevant regulations, the depreciation period of industrial plants is generally 20 to 40 years. The depreciation period is selected as 40 years, so the depreciation coefficient ε can be determined accordingly:

[0078]

[0079] The goal of minimizing carbon emissions includes considerations of equipment operation energy consumption, site energy consumption, and silkworm metabolic heat. Specifically, site energy consumption mainly involves the energy consumption of the entire base, including lighting, ventilation, heating, cooling, cleaning and maintenance, and other aspects. In the process of silkworm breeding, silkworms need to grow in a relatively stable temperature. Since silkworms perform physiological activities such as breathing and digestion during growth, these activities will generate heat, causing the base temperature to rise. In order to maintain a relatively stable room temperature, air conditioning needs to work continuously, so air conditioning energy consumption is one of the main factors of base energy consumption.

[0080] The specific carbon emission optimization formula is as follows:

[0081]

[0082] Among them, ω represents the carbon emission coefficient, that is, the ratio of carbon dioxide emissions generated by unit energy consumption; P represents the total power input of the equipment inside the base; T d represents the daily working hours of the equipment; CDD represents the annual working days of the equipment; t a Indicates the daily operation time of the air conditioning equipment inside the base; E k Indicates the site energy consumption coefficient of the base, that is, the energy consumption coefficient of the air conditioner in the base during operation, and the seasonal weight coefficient; α k Indicates the air conditioning energy consumption ratio in a certain season; l i represents the length of facility i, w i Represents the width of facility i.

[0083] N is the number of silkworms in the base, q is the metabolic heat of each silkworm (the energy consumption per unit time is in W for easy calculation), β is an adjustment factor to represent the additional impact of silkworm metabolic heat on air conditioning energy consumption, and λ t It represents the average survival rate of silkworms in one cycle.

[0084] The survival rate of silkworms is affected by the nonlinear combination of multiple factors such as ambient temperature, ventilation conditions, mulberry leaf quality, disease prevention and control, breeding density and breeding technology. Due to practical difficulties, it is not possible to carry out large-scale experiments for systematic analysis, so a fuzzy control system is used for research. The survival days T(t) (0-26 days), ambient temperature X(r) (25-31℃), mulberry leaf freshness Q(h) (storage time 0-24 hours) are used as fuzzy inputs, and the survival rate λ(t) is used as the fuzzy output.

[0085] λ t =F(T,X,Q)=F(T(t),X(r),Q(h))t∈(0,26),j∈(0,24)

[0086] The survival rate of silkworms is a dynamic process with time-varying characteristics. Relying only on the survival rate data of a certain day may be affected by accidental factors and short-term fluctuations, and cannot fully reflect its long-term survival situation. According to small-scale practice, the survival rate generally fluctuates within a specific range, so it is necessary to use mathematical expectation for research and expression. The specific formula is as follows:

[0087]

[0088] The fuzzy subset of ambient temperature X is set to {L, M, H}, the fuzzy subset of mulberry leaf freshness Q is set to {L, M, H}, and the fuzzy subset of survival rate λ_t is set to {VL, L, M, H, VH}. It is known from small-scale practice that the suitable survival temperature of silkworms is about 28°C, so M in the fuzzy subset of ambient temperature X is set to 28, L to 25, and H to 31. In addition, the storage time is set to 0–24 hours according to the freshness of the stored mulberry leaves, M is set to 12, L to 24, and H to 0. Through experiments, it is obtained that the input ambient temperature and mulberry leaf freshness are derived from the above fuzzy rules to generate the corresponding survival rate E(λ t ) Fuzzy output surface. By analyzing the surface, it is found that different temperatures and mulberry leaf freshness correspond to different survival rates, and they vary within a reasonable range. Finally, it is found that when the temperature reaches 28℃ and the mulberry leaf freshness is 0 hours, the survival rate reaches the maximum value.

[0089] S3: Determine the constraint condition step: Layout constraints include spacing constraints, boundary constraints, special location constraints, and biological isolation constraints.

[0090] The spacing constraint is that the minimum spacing between any two facilities in the horizontal direction or in the vertical direction should be greater than a certain threshold. The spacing constraint satisfies the following formula:

[0091]

[0092] Among them, x is the horizontal direction, H ij is the minimum spacing in the horizontal direction, y is the vertical direction, V ij is the minimum vertical spacing, l i represents the length of facility i, w i Represents the width of facility i.

[0093] Preferably, the boundary constraint satisfies the following formula:

[0094] and

[0095] and

[0096] S is the minimum distance between the facility and the boundary.

[0097] Special location constraints are certain conditions that must be met by special operation units in the base. For example, the silkworm tray cleaning and disinfection room, as a special operation unit, must be close to ventilation. Special location constraints satisfy the following formula:

[0098] P s ={(x i ,y i )|x i =x s ,y i =y s}

[0099] Among them, (x s ,y s ) is the preset position of the cleaning and disinfection room.

[0100] To ensure that there is no cross infection between breeding areas of different silkworm ages, the regional layout is designed according to the principle of biosafety. The actual distance between areas of different silkworm ages must be greater than the preset minimum biological isolation distance. The biological isolation constraint satisfies the following formula:

[0101] d ij ≥R ij

[0102] Among them, d ij is the actual distance between different silkworm age areas, R ijis the preset minimum biological isolation distance. Since young silkworms have weaker resistance to external pathogens, and the metabolic products of older silkworms are easily used as a medium for the spread of pathogens, areas of different silkworm ages need to be clearly isolated. Figure 5 The mark 2 in the figure is set as the silkworm room for young silkworms, and the marks 6 and 9 are set as the silkworm rooms for old silkworms.

[0103] S4 model algorithm iteration steps: instantiate the decision variables of the NSGA-II model (the location coordinates of each facility are used as decision variables, and the upper and lower boundaries of the variables define the location range where the facility can be placed in the workshop to ensure that the facility does not exceed the workshop boundary and maintains a certain minimum spacing), generate population individuals, and determine the layout plan. Among them, each element of the population individual represents the location coordinates of the facility in the base. The specific steps to confirm the layout plan are as follows:

[0104] S4.1 Population differentiation step: Divide the population into two sub-populations, sub-population 1 and sub-population 2;

[0105] Subpopulation 1 is a population close to the border: it is used to optimize the silkworm tray cleaning and disinfection area when it needs to be close to the base border. The facility is initialized and placed near the layout border to ensure that the facility maintains a minimum distance from the border.

[0106] Subpopulation 2 is a population close to the border: it is used to optimize the silkworm tray cleaning and disinfection area when it needs to be close to the border but not tight. The facility is randomly allocated in the area close to the border to meet flexible layout requirements.

[0107] Specific as Figure 5 In the example, a special boundary constraint is imposed on facility 8. One part of the population ensures that facility 8 is close to the boundary, and the other part of the population ensures that facility 8 is close to the boundary.

[0108] S4.2 Initial parent generation judgment processing steps: determine whether the two sub-populations are to be merged and generate the initial parent population; if a merge is required, directly generate the initial parent population and proceed to the next step; if not, perform non-dominated sorting on the current population, and on this basis, generate the sub-population through selection, crossover and mutation operations.

[0109] Since the target values ​​of the initial individuals are relatively dispersed in terms of operating costs and carbon emissions, some individuals may perform better in terms of cost, while others may be superior in terms of carbon emissions. The superiority and inferiority relationship between different individuals is not yet clear. When each individual's dominance counter np and dominance set Sp are used, the parent population Pt is quickly non-dominated sorted according to the optimization goal, and all individuals are traversed to calculate the dominance relationship of each individual. Individuals that are not dominated by any other individual are added to the first-level non-dominated set Sp0, and individuals dominated by individuals in the previous layer are divided layer by layer into the second-level and lower-level non-dominated sets (Sp 1,,Sp2,...). After optimization, the location coordinates of the individual are gradually optimized, tending to reduce operating costs and carbon emissions. For example, the location of the facility may gradually move closer to the demand center to reduce transportation costs, while avoiding over-concentration that leads to reduced coverage efficiency.

[0110] S4.3 New parent generation judgment processing steps: After the parent generation and the offspring population are combined in step S4.2, it is determined whether a new parent generation is generated; if so, selection, crossover and mutation are performed to generate the offspring population; if not, the objective function value (minimizing operating costs, minimizing carbon emissions, and silkworm survival rate) is recalculated, and the crowding is calculated quickly and non-dominatedly; when calculating the crowding, the total crowding of each individual needs to be calculated. Sort the individuals according to the crowding and store the results in chs.

[0111] Among them, crossover is to perform binary crossover on the new population to generate offspring individuals. That is, two parent individuals are randomly selected, crossover calculations are performed, and two new offspring individuals are generated. The frequency of crossover operations is controlled by crossover probability to maintain the diversity of solutions. During the crossover process, it is ensured that the generated offspring variables are within the feasible range and meet the constraints of the problem.

[0112] The mutation operation of the offspring population is a polynomial mutation method. By introducing moderate disturbances to the offspring individuals generated by the crossover operation, the diversity and search ability of the population are further increased. The variables of each offspring individual are mutated independently with the mutation probability, and the variables are perturbed according to the predefined mutation distribution, which must ensure that the mutated individuals meet the constraints.

[0113] S4.4 Individual sorting step: According to the multi-objective values ​​of each individual (such as the operating cost and carbon emissions as mentioned above), the population is non-dominated and the non-dominated sorting level of each individual is determined.

[0114] S4.5 New generation parent step: Use the improved elite retention strategy to retain its excellent individuals and generate a new generation of parent population. The improved elite retention strategy specifically uses a multi-objective decentralized weight strategy and a migration distance screening strategy to improve the traditional elite retention strategy.

[0115] (1) Multi-objective decentralized weight strategy

[0116] This strategy combines the target weight and the distribution characteristics of individuals in the target space to give priority to retaining individuals in sparse areas of high-weight targets, thereby improving population diversity and balancing the distribution of solutions. Specifically, in the environmental selection stage, individuals in sparse areas are given priority based on the crowding distance of individuals on high-weight targets to ensure the uniform distribution of solutions in the target space. Before calculating the crowding distance, all objective function values ​​need to be normalized to eliminate the influence of different target dimensions on the distance calculation:

[0117]

[0118] In the formula, F i,k Represents individual x i The crowding distance of the kth objective function, ρ i,k represents the weight of crowding distance of different objective functions when calculating the total crowding distance, f i,k Represents individual x i The function value on the kth target, f k,max and f k,min They represent the maximum and minimum values ​​of the kth objective function respectively.

[0119] Secondly, the weight is adjusted dynamically according to the range of variation of the objective function value. The larger the range of variation of the objective function, the more significant its contribution to the optimization direction, so the assigned weight is also increased accordingly.

[0120]

[0121] In the formula, Represents the sum of the range of variation of all objective function values.

[0122] Finally, for the sparse areas in the high-weight targets, the priority of retaining the individuals is increased by adjusting the weight priority, which ultimately makes the diverse individuals in the sparse areas occupy a more important position in the population update.

[0123]

[0124] Where w′ k represents the adjusted weight, F k represents the average crowding distance of the kth target in the entire population.

[0125] (2) Migration distance screening strategy

[0126] The migration distance measures the displacement of an individual between two consecutive generations and reflects the exploration potential and adaptive changes of an individual by calculating the distance between the solutions of the individual in the tth generation and the t-1th generation.

[0127] M d =||x i,t -x i,t-1 ||

[0128] Among them, ‖.‖ represents the Euclidean distance between individuals. In order to prevent individuals from changing too much in two consecutive generations (migration distance is too large) and deviating from the direction of high-quality solutions, wasting search resources, or too small migration distance leads to excessive concentration of the population and loss of exploration ability, the upper and lower limit thresholds of migration distance are introduced:

[0129]

[0130] Where M d Represents individual x i The migration distance, x i,t Represents individual x i In the tth generation solution (layout position), M min Represents the minimum effective migration distance, M max Represents the maximum permissible migration distance.

[0131] S4.6 Determine the layout plan step: return to step 4.3 with the new generation parent population as input and continue iterating; if the number of iterations reaches the preset maximum threshold MaxIter, the iteration ends and the non-dominated solution set of the Pareto algorithm is output to determine the layout plan.

[0132] Specifically, when an individual in the model violates a constraint, a penalty function will be applied to the individual. The penalty function includes the penalty factor punish imposed on violations of boundary constraints, facility spacing constraints, special location constraints, and biological isolation constraints, as well as the penalty value penalty of a specific facility (facility 8), thereby increasing its target value and making its fitness worse.

[0133] The value in the fitness is the weighted sum of the objective function value and the penalty value. First, the two objective functions are standardized to scale their magnitudes to the same standard. The sum of the standardized objective function value array and the penalty value is used as the fitness value of the individual.

[0134] The center coordinates of the facilities are extracted from the optimal individuals, and the vertex coordinates of the facilities and their envelope areas are calculated. Then, the logistics cost, carbon emissions, and envelope area of ​​the facility layout are evaluated, and the boundary and distance constraints are checked, and the layout is displayed on a graphical interface.

[0135] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0136] The descriptions of several embodiments of the application are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the application, and these improvements and modifications should also be regarded as within the scope of protection of the present invention. Therefore, the scope of protection of the patent of this application shall be based on the attached claims.

Claims

1. A low-carbon planning method for silkworm breeding base based on NSGA-II model, characterized by: The specific steps include: S1 Data preprocessing steps: obtain the basic data of the silkworm breeding base, classify the basic data, determine the layout constraints of the classified data; collect parameters related to the growth environment of silkworms; S2 Establish data target step: Combine historical data of parameters related to the silkworm's growth environment to evaluate the impact on the survival rate of silkworms; establish a multi-objective optimization model for the layout of silkworm breeding bases, with the goal of reducing operating costs and carbon emissions; S3 step of determining constraints: layout constraints include spacing constraints, boundary constraints, special location constraints, and biological isolation constraints; The spacing constraint is that the minimum spacing between any two facilities in the horizontal direction or in the vertical direction should be greater than a certain threshold; the boundary constraint is that each work unit cannot exceed the size of the workshop when it is laid out in the workshop; Special location constraints are certain conditions that must be met by special operation units in the base, including the silkworm tray cleaning and disinfection room as a special operation unit, which must be close to ventilation; Biological isolation constraints To ensure that there is no cross infection between breeding areas of different silkworm ages, the regional layout is designed according to the principle of biosafety. The actual distance between areas of different silkworm ages must be greater than the preset minimum biological isolation distance; S4 model algorithm iteration steps: instantiate the decision variables of the NSGA-II model, generate population individuals, and determine the layout plan; each element of the population individual represents the location coordinates or layout status of the facility in the base; the specific steps for confirming the layout plan are as follows: S4.1 Population differentiation step: Divide the population into two sub-populations, sub-population 1 and sub-population 2; Subpopulation 1 is a population close to the border: it is used to optimize the silkworm tray cleaning and disinfection area when it needs to be close to the base border. The facility is initially placed near the layout border to ensure that the facility maintains a minimum distance from the border. Subpopulation 2 is a population close to the border: it is used to optimize the silkworm tray cleaning and disinfection area when it is close to the border but not close to it. The facility is randomly allocated in the area close to the border. S4.2 Initial parent generation judgment processing steps: determine whether the two sub-populations are merged to generate the initial parent population; if so, proceed to the next step to combine the parent and sub-populations; if not, perform non-dominated sorting and produce sub-populations through selection, crossover, and mutation. S4.3 New parent generation judgment processing steps: After the parent generation and the offspring population are combined in step S4.2, it is determined whether a new parent generation is generated; if so, selection, crossover and mutation are performed to generate the offspring population; if not, the objective function value is calculated, a fast non-dominated sort is performed, and then the congestion degree is calculated; S4.4 Individual sorting step: According to the multi-objective values ​​of each individual, the population is non-dominated sorted to determine the non-dominated sorting level of each individual; S4.5 New generation parent step: adopt the improved elite retention strategy to retain its excellent individuals and generate a new generation parent population; S4.6 Determine the layout plan step: return to step 4.3 with the new generation parent population as input and continue iterating; if the number of iterations reaches the preset maximum threshold MaxIter, the iteration ends and the non-dominated solution set of the Pareto algorithm is output to determine the layout plan.

2. According to claim 1, a low-carbon planning method for silkworm breeding base based on improved NSGA-II algorithm is characterized in that: The basic data in the S1 data acquisition step include the location, size, inter-facility logistics volume, mulberry leaf quality, silkworm survival rate over the years, equipment energy consumption, and carbon emission factor information of each facility.

3. The low-carbon planning method for silkworm breeding base based on improved NSGA-II algorithm according to claim 1 is characterized in that: Parameters related to the silkworm's growth environment include temperature and humidity, light, air circulation and epidemic prevention measures.

4. The low-carbon planning method for silkworm breeding base based on the improved NSGA-II algorithm according to claim 1 is characterized in that: The specific formula for reducing operating costs in step S2 is as follows: MinZ o = MinZ1 + MinZ2 + MinZ3 Formula (4) Among them, Z0 represents the total daily operating cost of the base, Z1 represents the total daily transportation cost of the base, Z2 represents the total daily mulberry leaf cost of the base, and Z3 represents the daily construction depreciation cost of the base; c ij represents the unit material handling cost per unit distance between facility i and facility j, d ij represents the distance between facilities i and j, t represents the tth day in the growth cycle of silkworms, p represents a growth cycle of silkworms, B t represents the number of trays of silkworms per day for t-year-old silkworms, B0 represents the initial number of trays of silkworms on the first day of age, ρ t represents the number of mulberry leaf trays required for each tray of silkworms at age t per day, ρ0 represents the initial number of mulberry leaf trays required for each tray of silkworms at day 1, and λ t represents the average survival rate of silkworms in a cycle, f′ j represents the amount of materials handled between facilities i and j, except for mulberry leaves; n is the total number of facilities; q t is the mass of mulberry leaves, c s represents the unit cost of mulberry leaves; c c It represents the construction cost per unit area of ​​the base, a represents the total length of the sericulture base, b represents the total width of the sericulture base, and ε is the depreciation coefficient.

5. The low-carbon planning method for silkworm breeding base based on improved NSGA-II algorithm according to claim 1 is characterized in that: The carbon emissions in step 2 include equipment operation energy consumption, site energy consumption and silkworm metabolic heat. The optimization formula is as follows: Among them, ω represents the carbon emission coefficient, that is, the ratio of carbon dioxide emissions generated by unit energy consumption; P represents the total power input of the equipment inside the base; T d represents the daily working hours of the equipment; CDD represents the annual working days of the equipment; t a Indicates the daily operation time of the air conditioning equipment inside the base; E k Indicates the site energy consumption coefficient of the base, that is, the energy consumption coefficient of the air conditioner in the base during operation, and the seasonal weight coefficient; α k Indicates the air conditioning energy consumption ratio in a certain season; l i represents the length of facility i, w i represents the width of facility i; N is the number of silkworms in the base, q is the metabolic heat of each silkworm, β is an adjustment factor to represent the additional impact of silkworm metabolic heat on air conditioning energy consumption, and λ t It represents the average survival rate of silkworms in one cycle.

6. The low-carbon planning method for silkworm breeding base based on improved NSGA-II algorithm according to claim 1 is characterized in that: The survival rate of silkworms is expressed by mathematical expectation, with the survival days T(t), ambient temperature X(r), and mulberry leaf freshness Q(h) as fuzzy inputs. The specific formula is as follows: Among them, t ranges from 0 to 26 days of survival, r ranges from 25 to 31 degrees Celsius, and h ranges from 0 to 24 hours of storage. Through experiments, the input ambient temperature and mulberry leaf freshness are derived from the above fuzzy rules to generate the corresponding survival rate E(λ t ) Fuzzy output surface. By analyzing the surface, it is found that when the temperature reaches 28℃ and the freshness of mulberry leaves is 0 hours, the survival rate reaches the maximum value.

7. The low-carbon planning method for silkworm breeding base based on improved NSGA-II algorithm according to claim 1 is characterized in that: The spacing constraint satisfies the following formula: Among them, x is the horizontal direction, H ij is the minimum spacing in the horizontal direction, y is the vertical direction, V ij is the minimum vertical spacing, l i represents the length of facility i, w i Represents the width of facility i.

8. The low-carbon planning method for silkworm breeding base based on improved NSGA-II algorithm according to claim 7 is characterized in that: The boundary constraints satisfy the following formula: and and S is the minimum distance between the facility and the boundary.

9. The low-carbon planning method for silkworm breeding base based on improved NSGA-II algorithm according to claim 8 is characterized in that: The special position constraint satisfies the following formula: P s ={(x i ,and i )|x i =x s ,and i =and s } Among them, (x s ,y s ) is the preset position of the cleaning and disinfection room.

10. The low-carbon planning method for silkworm breeding base based on improved NSGA-II algorithm according to claim 9, characterized in that: The biological isolation constraint satisfies the following formula: d ij ≥R ij Among them, d ij is the actual distance between different silkworm age areas, R ij The preset minimum biological isolation distance.

Images