A multi-temperature layer food low-carbon packing optimization method based on a hybrid intelligent algorithm

Abstract

This invention discloses a low-carbon food packing optimization method based on a hybrid intelligent algorithm, belonging to the field of cold chain logistics loading optimization technology. The method includes the following steps: S1, obtaining the size, weight, temperature requirements, and vulnerability attributes of the food to be packed, forming a candidate set of zones; S2, generating a zone allocation scheme based on the cargo set and the candidate set of zones; S3, constructing a set of available space, performing posture matching for each food item, and obtaining candidate packing schemes; S4, calculating space utilization, refrigerant usage, and weight distribution balance based on the candidate packing schemes, forming an evaluation result; S5, iteratively optimizing the packing sequence, posture encoding, and zone allocation scheme based on the evaluation result, and outputting the optimized packing result. This invention achieves direct evaluation and optimization of refrigerant usage by jointly calculating the food volume and unit refrigerant consumption coefficient in the frozen and refrigerated zones.

Description

Technical Field

[0001] This invention relates to the field of cold chain logistics loading optimization technology, and in particular to a low-carbon packaging optimization method for multi-temperature food based on a hybrid intelligent algorithm. Background Technology

[0002] With the increasing demand for co-transporting fresh, frozen, and refrigerated foods, multi-temperature cold chain packing is gradually shifting from simple, experience-based loading to optimized loading oriented towards constraints and objectives. Existing technologies typically address the varying temperature requirements of different foods during storage and transportation by pre-defining fixed temperature zones and packing according to manual rules. Some solutions also consider packing volume utilization or transportation stability, but in multi-temperature co-transport scenarios, it remains difficult to simultaneously achieve optimal space utilization, refrigerant usage, and balanced weight distribution.

[0003] The existing technology has at least two drawbacks: First, fixed partitioning and static packing strategies make it difficult to dynamically adjust the partition length according to the temperature compatibility and volume combination of food, resulting in insufficient utilization of available container space, and redundant refrigerant configuration in low-temperature and refrigerated partitions, leading to high refrigerant consumption; Second, existing packing schemes tend to focus on volume filling and pay insufficient attention to the weight distribution on the bottom of the container, which can easily lead to problems such as localized unbalanced loading, uneven stress on the front and rear axles, or decreased stability during transportation.

[0004] Therefore, there is an urgent need to provide a low-carbon packaging optimization method for multi-temperature food based on hybrid intelligent algorithms to solve the above problems. Summary of the Invention

[0005] The technical problem to be solved by this invention is to overcome the above-mentioned prior art: First, fixed partitioning and static packing strategies are difficult to dynamically adjust the partition length according to the temperature compatibility and volume combination of food, resulting in insufficient utilization of available container space, and redundant refrigerant configuration in low-temperature and refrigerated partitions, leading to high refrigerant consumption; Second, existing packing schemes tend to focus on volume filling and pay insufficient attention to the weight distribution on the bottom of the container, which easily leads to problems such as localized unbalanced loading, uneven force on the front and rear axles, or decreased stability during transportation. This invention provides a low-carbon packing optimization method for multi-temperature food based on a hybrid intelligent algorithm.

[0006] To solve the above-mentioned technical problems, one technical solution adopted by the present invention is: to provide a method for optimizing low-carbon packaging of multi-temperature-layer food based on a hybrid intelligent algorithm, comprising the following steps: S1. Obtain the size, weight, temperature requirements, and fragility attributes of the food to be packed, form a cargo set according to temperature requirements and compatibility relationships, and form a candidate set of partitions according to the internal dimensions of the container and movable partitions. S2. Based on the cargo set and the partition candidate set, group and sort the food, generate a packing sequence, and generate a partition allocation scheme for food that is compatible with multiple temperature zones. S3. Construct an available space set based on the packing sequence and partitioning allocation scheme, and perform posture matching for each food item to obtain candidate packing schemes; S4. Calculate the space utilization rate, refrigerant usage and weight distribution balance according to the candidate packing schemes, and form the evaluation results by combining temperature layer matching, boundary constraints, overlap constraints and load constraints. S5. Based on the evaluation results, iteratively optimize the packing sequence, attitude coding, and partition allocation scheme, and output the optimized packing result.

[0007] The present invention is further configured such that: the candidate set of partitions in step S1 includes a frozen partition, a refrigerated partition, and a room temperature partition; the specific steps for forming the candidate set of partitions are as follows: the information of the food to be packed includes the length, width, height, weight, temperature requirements, and whether it is a fragile food; according to the temperature requirements and compatibility relationship, the food is divided into frozen food, frozen / refrigerated compatible food, refrigerated food, refrigerated / room temperature compatible food, and room temperature food; according to the internal dimensions of the container and the movable partitions, a candidate set of partitions for the frozen partition, the refrigerated partition, and the room temperature partition is formed.

[0008] The present invention is further configured such that the method for generating the packing sequence in step S2 is as follows: S201. Based on the cargo set and the partition candidate set, the food is divided into frozen food group, frozen / refrigerated compatible food group, refrigerated food group, refrigerated / room temperature compatible food group and room temperature food group according to the temperature category to which the food belongs. Volume information, weight information and vulnerability attributes are extracted in each food group to form a sorting base set corresponding to each food group. S202. Based on the sorting base set, arrange the food in each food group in a group according to temperature priority and volume descending order. The temperature priority is used to allow food groups with stricter temperature requirements to enter the arrangement process before food groups with more lenient temperature requirements. The volume descending order is used to allow food with a volume greater than a preset threshold to enter the arrangement process before food with a volume not exceeding the preset threshold, thus forming the group sorting result corresponding to each food group. S203. Based on the sorting results within the group and combined with the temperature layer connection relationship between adjacent food groups, the sorting results within the group are spliced ​​between groups to obtain an initial packing sequence. The temperature layer connection relationship is used to maintain the continuity of temperature layer changes between the end position of the previous food group and the start position of the next food group, and to reserve a sequence segment for the insertion and adjustment of compatible foods. S204. Based on the initial packing sequence and the sequence segment, insert and test the compatible foods in the frozen / refrigerated compatible food group and the refrigerated / room temperature compatible food group to form a transitional arrangement between the compatible foods and the adjacent temperature layer foods in terms of volume and weight distribution, thereby obtaining the packing sequence. The transitional arrangement is used to reduce temperature layer jumps and local off-center loading during subsequent partitioning.

[0009] The present invention is further configured such that the step of generating the partition allocation scheme in step S2 is as follows: S205. Based on the packing sequence, extract compatible food and its corresponding adjacent temperature zone food from the packing sequence to form a set to be allocated, and determine the current capacity margin, current load margin and current refrigerant load status of the frozen zone, refrigerated zone and room temperature zone respectively according to the partition candidate set to form an allocation base set corresponding to the set to be allocated. S206. Based on the allocation base set, determine the space occupancy change value, refrigerant usage change value, and weight distribution change value for each compatible food in the set to be allocated after entering different allowed zones. Then, process the space occupancy change value, the refrigerant usage change value, and the weight distribution change value according to a preset combination rule to obtain a comprehensive change value. Finally, generate the zone adaptation order based on the comprehensive change value. S207. According to the partition adaptation order, each compatible food is sequentially assigned to the corresponding allowed partition. After each compatible food is assigned, the current capacity margin, the current load margin, and the current refrigerant load status are updated to form a stage allocation result. The next compatible food in the stage allocation result continues to be assigned using the updated current capacity margin, current load margin, and current refrigerant load status, so that the partition allocation process remains continuously adjusted. S208. Based on the stage allocation results, the compatible foods in each allowed zone are reviewed. If the space utilization rate, refrigerant usage, or weight distribution balance of any allowed zone containing a compatible food does not meet the corresponding preset requirements, the compatible food is moved to the next allowed zone in the zone adaptation order and the stage allocation results are updated again until each compatible food is only allocated to one allowed temperature zone, and the zone allocation scheme is output.

[0010] The present invention is further configured such that: in step S3, an available space set is constructed based on the packing sequence and the partitioning allocation scheme, and posture matching and trial packing are performed on each food item in the available space set; the posture matching may include checking the various orthogonal placement postures of the food item one by one, and selecting a position from the space positions that meet the preset placement conditions for trial packing to obtain a candidate packing scheme.

[0011] The present invention is further configured such that: the space utilization rate in step S4 is evaluated by the ratio of the total volume of the food to be packed to the effective volume of the container; In step S4, the amount of refrigerant used is evaluated by the cumulative value of the food volume allocated to the frozen zone and the refrigerated zone and the corresponding unit refrigerant consumption coefficient. The amount of refrigerant used in the room temperature zone is not included. The weight distribution balance in step S4 is evaluated by dividing the bottom surface of the container into multiple statistical regions and calculating the discrete index composed of the weight deviation of each statistical region.

[0012] The present invention is further configured such that step S4 includes applying preset support constraints and fragility constraints to the candidate packing schemes, and penalizing or eliminating candidate packing schemes that have insufficient bottom support or whose fragile food exceeds the preset pressure limit.

[0013] The present invention is further configured such that: step S5 adopts a preset non-dominated sorting genetic optimization strategy to jointly optimize the space utilization rate, the refrigerant usage, and the weight distribution balance.

[0014] The beneficial effects of this invention are as follows: 1. This invention improves space utilization during multi-temperature layer transportation by forming a dynamic set of candidate partitions through movable partitions and combining them with a partition allocation scheme for compatible food. 2. This invention achieves direct evaluation and optimization of refrigerant usage by jointly calculating the food volume and unit refrigerant consumption coefficient in the frozen and refrigerated zones; 3. This invention uses weight deviation evaluation of the bottom statistical area to control the load distribution in each area of ​​the container and improve the balance of weight distribution; 4. This invention combines spatial partitioning and non-dominated sorting genetic optimization to obtain optimized packing results that balance space, energy consumption, and balance, while satisfying temperature matching, boundary constraints, and load constraints. Attached Figure Description

[0015] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a diagram illustrating the cargo classification and sorting structure of the present invention; Figure 3 This is a schematic diagram of the spatial segmentation method of the present invention; Figure 4 This is a flowchart of the non-dominated sorting genetic optimization strategy of the present invention. Detailed Implementation

[0016] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby providing a clearer and more explicit definition of the scope of protection of the present invention.

[0017] Please see Figure 1 - Figure 4 A method for optimizing low-carbon food packaging across multiple temperature zones based on a hybrid intelligent algorithm includes the following steps: S1. Obtain the size, weight, temperature requirements, and fragility attributes of the food to be packed, form a cargo set according to temperature requirements and compatibility relationships, and form a candidate set of partitions according to the internal dimensions of the container and movable partitions. S2. Based on the cargo set and the candidate set of partitions, group and sort the food, generate a packing sequence, and generate a partition allocation scheme for food that is compatible with multiple temperature zones. S3. Construct a set of available spaces based on the packing sequence and partitioning scheme, and perform posture matching for each food item to obtain candidate packing schemes; S4. Calculate the space utilization rate, refrigerant usage and weight distribution balance according to the candidate packing schemes, and form the evaluation results by combining temperature layer matching, boundary constraints, overlap constraints and load constraints. S5. Based on the evaluation results, iteratively optimize the packing sequence, attitude coding, and partition allocation scheme, and output the optimized packing results.

[0018] This method achieves collaborative packing of multi-temperature food products through partition candidate construction, packing sequence generation, candidate packing formation, three-objective evaluation, and iterative optimization. It improves space utilization, reduces refrigerant usage, and enhances weight distribution balance, thereby improving loading efficiency, carbon emissions, and transportation stability.

[0019] In this embodiment, step S4 is a step of quantitatively evaluating the candidate packing schemes. The quantitative evaluation includes at least three objectives: space utilization, refrigerant usage, and weight distribution balance. The space utilization objective is to maximize the proportion of the total volume of packed food in the effective volume of the container. The refrigerant usage objective is to minimize the cumulative refrigerant consumption corresponding to the frozen and refrigerated zones. The weight distribution balance objective is to minimize the variance of the total weight of goods relative to the average weight in the four statistical areas on the bottom of the container. To make the evaluation results calculable and comparable, in step S4, objective functions for space utilization, refrigerant usage, and weight distribution balance are constructed respectively. The evaluation results are then formed based on the calculation results of the objective functions, combined with temperature layer matching, boundary constraints, overlap constraints, and load constraints.

[0020] In step S1, the candidate set of partitions includes frozen partitions, refrigerated partitions, and ambient temperature partitions. The specific steps for forming the candidate set of partitions are as follows: the information of the food to be packed includes the length, width, height, weight, temperature requirements, and whether it is a fragile food; based on temperature requirements and compatibility, the food is divided into frozen food, frozen / refrigerated compatible food, refrigerated food, refrigerated / ambient temperature compatible food, and ambient temperature food; based on the internal dimensions of the container and movable partitions, the candidate set of partitions for frozen partitions, refrigerated partitions, and ambient temperature partitions is formed; movable partitions are used to change the length of each partition to balance space utilization and refrigerant usage.

[0021] The method for generating the binning sequence in step S2 is as follows: S201. Based on the cargo set and the candidate set of partitions, the food is divided into frozen food group, frozen / refrigerated compatible food group, refrigerated food group, refrigerated / room temperature compatible food group and room temperature food group according to the temperature category to which the food belongs. Volume information, weight information and vulnerability attributes are extracted in each food group to form a sorting base set corresponding to each food group. S202. Based on the sorting base set, arrange the food in each food group in the order of temperature layer priority and volume descending order. Temperature layer priority is used to make food groups with stricter temperature requirements enter the arrangement process before food groups with more lenient temperature requirements. Volume descending order is used to make food with a volume greater than a preset threshold enter the arrangement process before food with a volume less than a preset threshold, thus forming the sorting result within each food group. S203. Based on the sorting results within the group and the temperature layer connection relationship between adjacent food groups, the sorting results within the group are spliced ​​between groups to obtain the initial packing sequence. The temperature layer connection relationship is used to maintain the continuity of temperature layer changes between the end position of the previous food group and the start position of the next food group, and to reserve sequence segments for the insertion and adjustment of compatible foods. S204. Based on the initial packing sequence and sequence segment, insert and test the compatible foods in the frozen / refrigerated compatible food group and the refrigerated / room temperature compatible food group to form a transitional arrangement between the compatible foods and the adjacent temperature layer foods in terms of volume distribution and weight distribution, thereby obtaining the packing sequence. The transitional arrangement is used to reduce temperature layer jumps and local off-center loading during subsequent zoning allocation.

[0022] The steps for generating the partition allocation scheme in step S2 are as follows: S205. Based on the packing sequence, extract compatible food and its corresponding adjacent temperature layer food from the packing sequence to form a set to be allocated. Then, based on the partition candidate set, determine the current capacity margin, current load margin, and current refrigerant load status of the frozen partition, refrigerated partition, and ambient temperature partition respectively to form an allocation basis set corresponding to the set to be allocated. S206. Based on the allocation base set, determine the space occupancy change value, refrigerant usage change value, and weight distribution change value for each compatible food in the allocation set after entering different allowable zones. Then, process the space occupancy change value, refrigerant usage change value, and weight distribution change value according to the preset combination rules to obtain the comprehensive change value. Finally, generate the zone adaptation order based on the comprehensive change value. S207. According to the partition adaptation order, each compatible food is sequentially assigned to the corresponding allowed partition. After each compatible food is assigned, the current capacity margin, current load margin, and current refrigerant load status are updated to form a phased allocation result. The next compatible food in the phased allocation result continues to use the updated current capacity margin, current load margin, and current refrigerant load status for allocation, so that the partition allocation process remains continuously adjusted. S208. Based on the phase allocation results, the compatible food in each allowed zone is reviewed. If the space utilization rate, refrigerant usage, or weight distribution balance of any allowed zone containing a compatible food does not meet the corresponding preset requirements, the compatible food is moved to the next allowed zone in the zone adaptation order and the phase allocation results are updated again until each compatible food is only allocated to one allowed temperature zone, and the zone allocation scheme is output.

[0023] In step S3, an available space set is constructed based on the packing sequence and partition allocation scheme, and each food item is matched and trial-packed in the available space set. The matching of postures may include checking the various orthogonal placement postures of the food item one by one, and selecting a position from the space positions that meet the preset placement conditions for trial packing to obtain a candidate packing scheme.

[0024] The available space set in step S3 is constructed using a space partitioning method. After each food placement, the occupied space is deleted and a new remaining space is generated. The candidate placement location is selected according to the principle of minimizing the remaining space and satisfying the boundary constraints, so as to improve the subsequent space utilization target value and provide a basis for reducing the refrigerant usage target value and reducing the weight distribution balance target value.

[0025] During the trial assembly process, the length, width, and height of each food item are spatially matched and determined. Let the first... The length of each food item is Width is The height is The weight is Then the first Volume of food It can be represented as: = × ×

[0026] in: Indicates the first Individual food items; Indicates the first The length of each food item; Indicates the first The width of each food item; Indicates the first The height of each food item; Indicates the first The weight of each food item; Indicates the first The volume of each food item.

[0027] In this embodiment, it is set that Indicates the partition number. Indicates the first Whether the food item was allocated to the first The partition, when the first The food items were allocated to the first Set the value to 1 if there is a partition, otherwise set it to 0; let... Indicates the first The length of each partition Indicates the first The width of each partition, Indicates the first The height of each zone. Zone parameters are used to characterize the placement conditions of food in different temperature zones.

[0028] During the trial packing process, it is also determined whether the food exceeds the boundary of its assigned zone and whether any two food items overlap in space; trial packing results that exceed the boundary or overlap are not retained as candidate packing schemes. Preferably, after each food trial packing is completed, the original occupied space is split and new usable space is generated to form an updated set of usable space.

[0029] In step S4, the space utilization rate is evaluated by the ratio of the total volume of the food to be packed to the effective volume of the container. In step S4, a space utilization objective function is used to evaluate the space utilization of candidate packing schemes. The space utilization objective function uses the effective volume of the container as a benchmark, representing the proportion of the total volume of packed food within the effective container volume, and its optimization direction is to maximize space utilization. The space utilization objective function is:

[0030] The parameters are defined as follows: Indicates space utilization rate; This indicates the total quantity of food items to be packed. Indicates the first Individual food items; Indicates the first The volume of each food item; This indicates the effective volume of the container.

[0031] The aforementioned space utilization objective function is used to characterize the degree to which candidate packing schemes utilize the internal space of the container, and its optimization direction is to maximize space utilization.

[0032] In step S4, the amount of refrigerant used is evaluated by the cumulative value of the food volume allocated to the frozen and refrigerated zones and the corresponding unit refrigerant consumption coefficient. The amount of refrigerant used in the room temperature zone is not included. In step S4, to evaluate the refrigerant consumption of candidate packing schemes under different temperature zones, a refrigerant usage objective function is adopted. Refrigerant usage is directly related to the low-carbon objective. The refrigerant usage objective function is based on the food volume allocated to the freezing and refrigeration zones, multiplied by the unit refrigerant consumption coefficient of the corresponding temperature zone, and then accumulated. Its optimization direction is to minimize refrigerant usage. The refrigerant usage objective function is:

[0033] in, Indicates the amount of refrigerant used; This indicates the total quantity of food items to be packed. Indicates the first Individual food items; The refrigerant consumption coefficient per unit for the refrigeration zone; Indicates the first The volume of each food item; Indicates the first The allocation variable for whether a food item is assigned to the frozen section, when the first food item... A value of 1 is used when a food item is assigned to the frozen section, otherwise a value of 0 is used. Indicates the first The allocation variable for whether a food item is assigned to a refrigerated section, when the first food item... A food item is assigned a value of 1 when it is placed in the refrigerated section, and 0 otherwise.

[0034] The above refrigerant usage objective function is used to characterize the refrigerant consumption level of the candidate packing scheme under the premise of meeting the temperature requirements. Its optimization direction is to minimize the refrigerant usage. The refrigerant usage is not included in the normal temperature zone because it does not require refrigerant configuration.

[0035] In step S4, the weight distribution balance is evaluated by dividing the bottom surface of the container into multiple statistical regions and calculating the discrete index formed by the weight deviation of each statistical region. The smaller the discrete index, the more balanced the weight distribution.

[0036] In step S4, to evaluate the load distribution of candidate packing schemes across different areas of the container bottom, a weight distribution balance objective function is adopted. Based on the modeling criteria in document 2, the container bottom is divided into four statistical regions. The total weight of goods in each region is calculated, and the variance of the total weight in each region relative to the average weight is used as the weight distribution balance objective value. The optimization direction is to minimize this variance. The weight distribution balance objective function is:

[0037] The parameters are defined as follows: This represents the target value for balanced weight distribution. Indicates the first Statistical regions; Indicates the first The total weight of goods within each statistical region; This represents the average weight across the four statistical regions, and

[0038] This indicates the total quantity of food items to be packed. ; Indicates the first The weight of each food item; Indicates the first Is the food item located in the [number]th [item]? Regional variables for the nth statistical region, when the nth statistical region The food item is located in the... The value is 1 if there is a statistical range, otherwise it is 0.

[0039] The aforementioned weight distribution balance objective function is used to characterize the degree of load distribution balance of the candidate packing scheme in the four statistical regions on the bottom of the container. Its optimization direction is to minimize the variance of the total weight in the four statistical regions.

[0040] Step S4 further includes applying preset support constraints and fragility constraints to candidate packing schemes, and penalizing or eliminating candidate packing schemes that have insufficient bottom support or whose fragile food exceeds the preset pressure limit.

[0041] In this embodiment, support constraints and vulnerability constraints are used to further screen candidate packing schemes to ensure that while improving space utilization, reducing refrigerant usage, and reducing the target value of weight distribution balance, the candidate packing schemes still have acceptable loading stability.

[0042] In step S5, a preset non-dominated sorting genetic optimization strategy is used to jointly optimize space utilization, refrigerant usage, and weight distribution balance.

[0043] In step S5, space utilization, refrigerant usage, and weight distribution balance are used as joint optimization objectives to iteratively optimize the packing sequence, attitude encoding, and partition allocation scheme. Step S5 employs a non-dominated sorting genetic optimization strategy to search for multiple objectives in parallel. Instead of pre-combining space utilization, refrigerant usage, and weight distribution balance into a single weighted objective, it retains candidate packing schemes under different objective combinations through non-dominated sorting and crowding comparison to obtain Pareto optimization results with higher space utilization, lower refrigerant usage, and smaller weight distribution balance objective values.

[0044] Example This embodiment uses a 20ft standard container as the transport carrier. The container's interior is divided along its length into frozen, refrigerated, and ambient temperature zones by movable partitions. The food to be packed includes frozen food, frozen / refrigerated compatible food, refrigerated food, refrigerated / ambient temperature compatible food, and ambient temperature food.

[0045] First, in step S1, the length, width, height, weight, temperature requirements, and fragility attributes of the food to be packed are obtained, and a cargo set is formed according to temperature requirements and compatibility relationships. Simultaneously, a candidate partition set is formed based on the container's internal dimensions and movable partitions. The candidate partition set includes at least a frozen partition, a refrigerated partition, and a normal temperature partition.

[0046] Secondly, in step S2, the food is grouped and sorted according to the cargo set and the candidate partition set to generate a packing sequence, and a partition allocation scheme is generated for foods that are compatible with multiple temperature zones. Specifically, the food is first grouped into frozen food group, frozen / refrigerated compatible food group, refrigerated food group, refrigerated / room temperature compatible food group, and room temperature food group. Then, within each group, the food is arranged in descending order of volume, and the allowed partitions are determined for compatible foods through the partition allocation scheme.

[0047] Then, in step S3, a set of available space is constructed based on the packing sequence and partition allocation scheme. For each food item, posture matching is performed and trial packing is conducted to obtain candidate packing schemes. During the trial packing process, it is determined whether the food item exceeds the partition boundary, overlaps with other food items, or meets the partition capacity conditions. After each trial packing is completed, the originally occupied space is deleted and new remaining space is generated to update the set of available space.

[0048] Next, in step S4, the space utilization rate, refrigerant usage, and weight distribution balance of the candidate packing schemes are calculated respectively, and the evaluation results are formed by combining temperature layer matching, boundary constraints, overlap constraints, and load constraints. Specifically, the space utilization rate is calculated using a space utilization objective function, with maximizing space utilization as the optimization direction; the refrigerant usage is calculated using a refrigerant usage objective function, with minimizing refrigerant usage as the optimization direction; and the weight distribution balance is calculated using a weight distribution balance objective function, with minimizing the variance of the total weight of goods relative to the average weight in the four statistical areas on the bottom of the container as the optimization direction. Preferably, the bottom of the container is divided into four statistical areas, and the total weight of goods in each statistical area is counted separately to reduce the risk of uneven loading.

[0049] Finally, in step S5, the packing sequence, attitude coding and partition allocation scheme are iteratively optimized based on the evaluation results. The space utilization, refrigerant usage and weight distribution balance are jointly optimized using a preset non-dominated sorting genetic optimization strategy, and the optimized packing results are output.

[0050] Through the above implementation methods, in the scenario of transporting food in multiple temperature zones in the same container, the space utilization rate of the container can be improved on the one hand, and the refrigerant usage in the frozen and refrigerated zones can be reduced on the other hand. At the same time, the variance of the total weight of goods in the four statistical areas on the bottom of the container relative to the average weight can be reduced, thereby obtaining an optimized packing result that takes into account space efficiency, low carbon effect and transportation stability.

[0051] The above are merely embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for optimizing low-carbon food packaging across multiple temperature zones based on a hybrid intelligent algorithm, characterized in that: Includes the following steps: S1. Obtain the size, weight, temperature requirements, and fragility attributes of the food to be packed, form a cargo set according to temperature requirements and compatibility relationships, and form a candidate set of partitions according to the internal dimensions of the container and movable partitions. S2. Based on the cargo set and the partition candidate set, group and sort the food, generate a packing sequence, and generate a partition allocation scheme for food that is compatible with multiple temperature zones. S3. Construct an available space set based on the packing sequence and partitioning allocation scheme, and perform posture matching for each food item to obtain candidate packing schemes; S4. Calculate the space utilization rate, refrigerant usage and weight distribution balance according to the candidate packing schemes, and form the evaluation results by combining temperature layer matching, boundary constraints, overlap constraints and load constraints. S5. Based on the evaluation results, iteratively optimize the packing sequence, attitude coding, and partition allocation scheme, and output the optimized packing results.

2. The method for optimizing low-carbon food packaging across multiple temperature zones based on a hybrid intelligent algorithm according to claim 1, characterized in that: The candidate set of partitions in step S1 includes a frozen partition, a refrigerated partition, and a room temperature partition. The specific steps for forming the candidate set of partitions are as follows: the information of the food to be packed includes the length, width, height, weight, temperature requirements, and whether it is a fragile food; according to the temperature requirements and compatibility, the food is divided into frozen food, frozen / refrigerated compatible food, refrigerated food, refrigerated / room temperature compatible food, and room temperature food; according to the internal dimensions of the container and the movable partitions, a candidate set of partitions for the frozen partition, refrigerated partition, and room temperature partition is formed.

3. The method for optimizing low-carbon food packaging across multiple temperature zones based on a hybrid intelligent algorithm according to claim 2, characterized in that: The method for generating the packing sequence in step S2 is as follows: S201. Based on the cargo set and the partition candidate set, the food is divided into frozen food group, frozen / refrigerated compatible food group, refrigerated food group, refrigerated / room temperature compatible food group and room temperature food group according to the temperature category to which the food belongs. Volume information, weight information and vulnerability attributes are extracted in each food group to form a sorting base set corresponding to each food group. S202. Based on the sorting base set, arrange the food in each food group in a group according to temperature priority and volume descending order. The temperature priority is used to allow food groups with stricter temperature requirements to enter the arrangement process before food groups with more lenient temperature requirements. The volume descending order is used to allow food with a volume greater than a preset threshold to enter the arrangement process before food with a volume not exceeding the preset threshold, thus forming the group sorting result corresponding to each food group. S203. Based on the sorting results within the group and combined with the temperature layer connection relationship between adjacent food groups, the sorting results within the group are spliced ​​between groups to obtain an initial packing sequence. The temperature layer connection relationship is used to maintain the continuity of temperature layer changes between the end position of the previous food group and the start position of the next food group, and to reserve a sequence segment for the insertion and adjustment of compatible foods. S204. Based on the initial packing sequence and the sequence segment, insert and test the compatible foods in the frozen / refrigerated compatible food group and the refrigerated / room temperature compatible food group to form a transitional arrangement between the compatible foods and the adjacent temperature layer foods in terms of volume and weight distribution, thereby obtaining the packing sequence. The transitional arrangement is used to reduce temperature layer jumps and local off-center loading during subsequent partitioning.

4. The method for optimizing low-carbon food packaging across multiple temperature zones based on a hybrid intelligent algorithm according to claim 3, characterized in that: The steps for generating the partition allocation scheme in step S2 are as follows: S205. Based on the packing sequence, extract compatible food and its corresponding adjacent temperature zone food from the packing sequence to form a set to be allocated, and determine the current capacity margin, current load margin and current refrigerant load status of the frozen zone, refrigerated zone and room temperature zone respectively according to the partition candidate set to form an allocation base set corresponding to the set to be allocated. S206. Based on the allocation base set, determine the space occupancy change value, refrigerant usage change value, and weight distribution change value for each compatible food in the set to be allocated after entering different allowed zones. Then, process the space occupancy change value, the refrigerant usage change value, and the weight distribution change value according to a preset combination rule to obtain a comprehensive change value. Finally, generate the zone adaptation order based on the comprehensive change value. S207. According to the partition adaptation order, each compatible food is sequentially assigned to the corresponding allowed partition. After each compatible food is assigned, the current capacity margin, the current load margin, and the current refrigerant load status are updated to form a stage allocation result. The next compatible food in the stage allocation result continues to be assigned using the updated current capacity margin, current load margin, and current refrigerant load status, so that the partition allocation process remains continuously adjusted. S208. Based on the stage allocation results, the compatible foods in each allowed zone are reviewed. If the space utilization rate, refrigerant usage, or weight distribution balance of any allowed zone containing a compatible food does not meet the corresponding preset requirements, the compatible food is moved to the next allowed zone in the zone adaptation order and the stage allocation results are updated again until each compatible food is only allocated to one allowed temperature zone, and the zone allocation scheme is output.

5. The method for optimizing low-carbon food packaging across multiple temperature zones based on a hybrid intelligent algorithm according to claim 4, characterized in that: In step S3, an available space set is constructed based on the packing sequence and the partitioning allocation scheme, and posture matching and trial packing are performed on each food item in the available space set. The posture matching may include checking the various orthogonal placement postures of the food item one by one, and selecting a position from the space positions that meet the preset placement conditions for trial packing to obtain a candidate packing scheme.

6. The method for optimizing low-carbon food packaging across multiple temperature zones based on a hybrid intelligent algorithm according to claim 5, characterized in that: The space utilization rate in step S4 is evaluated by the ratio of the total volume of the food to be packed to the effective volume of the container. In step S4, the amount of refrigerant used is evaluated by the cumulative value of the food volume allocated to the frozen zone and the refrigerated zone and the corresponding unit refrigerant consumption coefficient. The amount of refrigerant used in the room temperature zone is not included. The weight distribution balance in step S4 is evaluated by dividing the bottom surface of the container into multiple statistical regions and calculating the discrete index composed of the weight deviation of each statistical region.

7. The method for optimizing low-carbon food packaging across multiple temperature zones based on a hybrid intelligent algorithm according to claim 6, characterized in that: Step S4 further includes applying preset support constraints and fragility constraints to the candidate packing schemes, and penalizing or eliminating candidate packing schemes that have insufficient bottom support or whose fragile food exceeds the preset pressure limit.

8. The method for optimizing low-carbon food packaging across multiple temperature zones based on a hybrid intelligent algorithm according to claim 7, characterized in that: Step S5 employs a preset non-dominated sorting genetic optimization strategy to jointly optimize the space utilization, refrigerant usage, and weight distribution balance.

Images