Time-temperature dynamic control method and device for after-ripening fruit refrigeration

Abstract

The invention provides a time-temperature dynamic control method and device for after-ripening type fruit refrigeration, and relates to the technical field of control, and the method comprises the steps: generating a 3D service window of each fruit order based on the requirements of the fruit orders of a fixed retailer; generating a fruit order service list according to the delivery time of each fruit order; generating a temperature control constraint condition based on the current information of the refrigeration house, the 3D service window of each fruit order and the fruit order service list; based on temperature control constraint conditions, a 3D service window of each fruit order and a fruit order service list, constructing a dual-target dynamic optimization model based on comprehensive service quality satisfaction and total energy consumption; and generating a temperature control path of the refrigeration house based on the double-target dynamic optimization model, so that the temperature control equipment dynamically controls the temperature of the refrigeration house according to the temperature control path. On the basis of the temperature control mode provided by the technical scheme, the energy consumption can be minimized on the premise of ensuring the fruit quality.

Description

Technical Field

[0001] This invention relates to the field of control, specifically to a method and apparatus for dynamic time-temperature control of post-ripening fruit cold storage. Background Technology

[0002] With the development of cold chain logistics, more and more fruits can be transported to various regions through cold chain logistics. Agricultural cold chain logistics mainly uses artificial refrigeration during the circulation process to maintain the post-harvest quality of perishable agricultural products and reduce losses and pollution during the process.

[0003] Currently, comprehensive environmental monitoring of the generation, storage, transportation, sales, and consumption of agricultural products in the cold chain can improve cold chain efficiency and product market competitiveness. Temperature is a core environmental factor affecting the quality of agricultural products, and Time-Temperature Management (TTM) is an important means of controlling temperature in the cold chain process. Currently, TTM optimization is mainly based on single objectives, such as maintaining food quality and reducing process losses, or focusing on energy conservation and emission reduction.

[0004] Because there is an inverse relationship between maintaining quality and energy consumption under cold chain conditions, current strategies cannot achieve a dynamic balance. Therefore, how to balance quality and energy consumption is an urgent problem to be solved in the cold chain logistics of fruits. Summary of the Invention

[0005] The purpose of this invention is to address the problem that existing cold chain time-temperature management strategies cannot dynamically balance quality maintenance and energy consumption. It provides a method and device for dynamic time-temperature control of post-ripening fruit cold storage, which can balance quality and energy consumption in fruit cold chain logistics. While meeting the quality requirements of each order, it can minimize energy consumption in fruit cold chain logistics.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0007] Firstly, a time-temperature dynamic control method for cold storage of post-ripening fruits is provided. This method includes: generating a three-dimensional (3D) service window for each fruit order based on the requirements of a fixed retailer's fruit orders; the 3D service window for each fruit order includes a maturity window, a delivery time window, and a quantity window; generating a fruit order service list according to the delivery time of each fruit order; generating temperature control constraints based on current cold storage information, the 3D service window for each fruit order, and the fruit order service list; the temperature control constraints include at least one of the following: the total quantity of fruit orders meets the total inventory constraint, the constraint on the completion of each fruit order within the 3D service window, the cold storage temperature constraint, and the overall service quality satisfaction deviation constraint; constructing a dual-objective dynamic optimization model based on overall service quality satisfaction and total energy consumption based on the temperature control constraints, the 3D service window for each fruit order, and the fruit order service list; and generating a temperature control path for the cold storage based on the dual-objective dynamic optimization model, so that the temperature control equipment dynamically controls the temperature of the cold storage according to the temperature control path.

[0008] Optionally, the overall service quality satisfaction deviation constraint includes: the deviation between the maturity window of the fruit order delivery and the center value of the maturity window required by the fruit order is greater than or equal to zero, and the outbound waiting time of the fruit orders in the fruit order service list is greater than or equal to zero.

[0009] Optionally, an evolutionary model of the fruit ripening process is constructed; based on the fruit ripening process evolutionary model and the refrigeration temperature of the cold storage, the maturity of the fruit is predicted; based on the deviation between the predicted delivery maturity and the center value of the required maturity window, and the outbound waiting time of the fruit order, the service quality deviation of each fruit order is generated; based on the cargo quantity of each fruit order and the service quality deviation of each fruit order, the comprehensive service quality deviation of all fruit orders in the fruit order service list is determined; based on the temperature difference inside and outside the cold storage and the surface area of ​​the cold storage, a dynamic energy consumption determination model is constructed to determine the power consumption within the temperature control cycle; a dual-objective dynamic optimization model for minimizing the comprehensive service quality deviation and the total energy consumption is constructed.

[0010] Optionally, the first temperature control path is a temperature control path for a determined first control period. The method further includes: during the process of dynamically controlling the temperature according to the first temperature control path in the first control period, predicting the maturity of the fruit and the cold storage energy consumption in the second control period based on the fruit ripening process evolution model and the temperature indicated by the first temperature control path; determining the comprehensive service deviation and total cold storage energy consumption of all fruit orders in the fruit order service list within the second control period based on the predicted maturity and cold storage energy consumption in the second control period; if the comprehensive service deviation and total energy consumption within the second control period meet the temperature control conditions, then continue to dynamically control the temperature according to the first temperature control path within the second control period, and the temperature control conditions indicate that the shipment meets the 3D service window of each fruit order; if the comprehensive service deviation and total energy consumption do not meet the temperature control conditions, then update the first temperature control path to the second temperature control path based on the current information of the cold storage, the current maturity of the fruit, the fruit order service list within the second control period, and the bi-objective dynamic optimization model.

[0011] Optionally, the method further includes: if the overall service deviation and total energy consumption in the second control period meet the temperature control conditions, then based on the overall service deviation and total energy consumption in the second control period, planning the delivery time of the fruit orders in the fruit order service list in the second control period.

[0012] Optionally, the method further includes: during the process of the temperature control device dynamically controlling the temperature according to the temperature control path, if a first fruit order from a dynamic retailer is received, determining the 3D service window of the first fruit order; determining whether the current cold storage conditions can supply the goods within the 3D service window of the first fruit order; if the goods can be supplied within the 3D service window of the first fruit order, adding the first fruit order to the fruit order service list; continuing to dynamically control the temperature according to the temperature control path, and supplying the goods within the 3D service window of the first fruit order.

[0013] Optionally, the method further includes: after adding the first fruit order to the fruit order service list to obtain an updated fruit order service list, updating the temperature control conditions based on the 3D service window of the first fruit order and the updated fruit order service list, wherein the temperature control conditions further include: an insertion list constraint for the fruit orders of the dynamic retailer; the insertion list constraint indicates that the maturity window required by the inserted fruit order covers the maturity window of the fruit in the cold storage within the outbound time window of the inserted fruit order under the current temperature control path; updating the bi-objective dynamic optimization model based on the updated temperature control constraints, the updated fruit order service list, and the 3D service window of each fruit order in the updated fruit order service list; and updating the temperature control path based on the updated bi-objective dynamic optimization model.

[0014] Optionally, according to preset rules, weight coefficients of two objective functions are assigned to fruit orders from fixed retailers and fruit orders from dynamic retailers respectively, and the bi-objective dynamic optimization model is updated.

[0015] Optionally, during the process of solving the bi-objective dynamic optimization model, when initializing and generating the initial solution set at the (k+1)th time step, half of the leading solution set obtained at the kth time step is merged with half of the initial solution set generated at the (k+1)th time step to form the initial solution set at the (k+1)th time step.

[0016] Secondly, embodiments of the present invention provide a time-temperature dynamic control device for cold storage of post-ripening fruits. This device includes: a 3D service window generation module, an order service list generation module, a temperature control constraint generation module, a control model construction module, and a temperature control path generation module. The 3D service window generation module is used to generate a three-dimensional 3D service window for each fruit order based on the requirements of fruit orders from fixed retailers. Each fruit order's 3D service window includes: a maturity window, a delivery time window, and a quantity window. The order service list generation module is used to generate a fruit order service list according to the delivery time of each fruit order. The temperature control constraint generation module generates a time-temperature dynamic control path based on the current information of the cold storage and the requirements of each fruit order. The system generates temperature control constraints for each 3D service window and fruit order service list. These constraints include at least one of the following: total fruit order quantity meeting total inventory constraints, constraints on the completion of each fruit order within the 3D service window, cold storage temperature constraints, and overall service quality satisfaction deviation constraints. A control model construction module is used to construct a dual-objective dynamic optimization model based on overall service quality satisfaction and total energy consumption, using the temperature control constraints, the 3D service window for each fruit order, and the fruit order service list. A temperature control path generation module is used to generate a temperature control path for the cold storage based on the dual-objective dynamic optimization model, enabling temperature control equipment to dynamically control the cold storage temperature according to the temperature control path.

[0017] Thirdly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the time-temperature dynamic control method for cold storage of post-ripening fruits as described in the first aspect.

[0018] Fourthly, embodiments of the present invention provide a computer device, a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein when the computer-readable instructions are executed by the processor, they implement the time-temperature dynamic control method for cold storage of post-ripening fruits as described in the first aspect.

[0019] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0020] First, based on the requirements of fruit orders from fixed retailers, a 3D service window is generated for each fruit order, resulting in a maturity window, a delivery time window, and a quantity window for each order. Second, a fruit order service list is generated according to the delivery time of each fruit order. Third, temperature control constraints are generated based on the current information of the cold storage, the 3D service window of each fruit order, and the fruit order service list. Then, based on the temperature control constraints, the 3D service window of each fruit order, and the fruit order service list, a dual-objective dynamic optimization model based on comprehensive service quality satisfaction and total energy consumption is constructed. Finally, based on the dual-objective dynamic optimization model, a temperature control path for the cold storage is generated, enabling the temperature control equipment to dynamically control the temperature of the cold storage according to the temperature control path. In other words, during the process of determining the temperature control path, a dual-objective optimization model is constructed based on the actual needs of each order and the deviation of the overall service quality satisfaction. This model aims to minimize the power consumption of the cold storage while meeting the needs of each order. It can balance quality and energy consumption in fruit cold chain logistics, and minimize energy consumption in fruit cold chain logistics while meeting the delivery quality of each fruit order. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of a time-temperature dynamic control system provided in an embodiment of the present invention.

[0022] Figure 2 This is a flowchart illustrating a method for dynamic time-temperature control of cold storage for post-ripening fruits, provided in an embodiment of the present invention.

[0023] Figure 3 This is a schematic diagram of a Pareto front solution at different times based on an improved NSGA-II, provided as an embodiment of the present invention.

[0024] Figure 4 This is a schematic diagram illustrating the robustness of simultaneous insertion of dynamic order requirements, as provided in an embodiment of the present invention.

[0025] Figure 5 This is a schematic diagram illustrating the robustness of a dynamic order demand distributed insertion method as provided in an embodiment of the present invention.

[0026] Figure 6 This is a schematic diagram illustrating the temperature control and quality maturity change path of a constant 0°C TTM strategy provided in an embodiment of the present invention.

[0027] Figure 7 This is a schematic diagram illustrating the temperature control and quality maturity change path of a constant 4°C TTM strategy provided in an embodiment of the present invention.

[0028] Figure 8 This is a schematic diagram illustrating the temperature control and quality maturity change path of a constant 8°C TTM strategy provided in an embodiment of the present invention.

[0029] Figure 9 This is a schematic diagram illustrating the temperature control and quality maturity change path of an optimized TTM strategy provided in an embodiment of the present invention.

[0030] Figure 10 A hardware structure diagram of the computer equipment containing the time-temperature dynamic control device for the cold storage of post-ripening fruits according to an embodiment of the present invention.

[0031] Figure 11 This invention provides a time-temperature dynamic control device for the cold storage of post-ripening fruits. Detailed Implementation

[0032] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.

[0033] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular forms “a,” “the,” and “the” used in this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0034] It should be understood that although the terms first, second, third, etc., may be used in this invention to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first information may also be referred to as second information without departing from the scope of this invention, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0035] The embodiments of the present invention will now be described in detail.

[0036] Figure 1 This is a schematic diagram of a time-temperature dynamic control system provided in an embodiment of the present invention. Figure 1As shown, the time-temperature dynamic control system 100 includes: an order receiving device 101, a server 102, a cold storage group 103, and a cold storage temperature control device 104; the cold storage group 103 includes P cold storages, where P is a positive integer; the order receiving device 101 is used to receive fruit orders from retailers (fixed retailers and / or dynamic retailers); the server 102 is used to construct a dual-objective dynamic optimization model for fruit quality and energy consumption based on the retailers' fruit order requirements (fruit type, fruit maturity, fruit delivery time, fruit quantity) and the current information of each cold storage in the cold storage group 103 (stored fruit type, maturity of each type of fruit, inventory of each type of fruit, current temperature inside and outside the cold storage) and generate a temperature control path based on the dual-objective dynamic optimization model; the cold storage temperature control device 104 is used to control the temperature of post-ripening fruit in each cold storage in the cold storage group 103 according to the temperature control path.

[0037] Post-ripening fruits typically refer to fruits that are not fully ripe when picked and need to be left to ripen under specific conditions (usually room temperature) for a period of time to undergo physiological and biochemical changes to achieve their optimal flavor, texture, and sweetness. Examples include kiwifruit, bananas, persimmons, mangoes, and avocados.

[0038] The time-temperature dynamic control system for cold storage of post-ripening fruits provided in this invention can be applied to the temperature control process of cold storage of post-ripening fruits. It generates temperature control constraints, a 3D service window for fruit orders, and a service list for fruit orders based on fruit order requirements and current cold storage information. Based on this information, it constructs a dual-objective dynamic optimization model based on comprehensive service quality satisfaction and total energy consumption. Then, it generates a temperature control path based on this model, and subsequently performs cold storage temperature control based on this path. This results in a temperature control method that satisfies both fruit quality requirements and cold storage energy consumption requirements. It can balance fruit quality and energy consumption in fruit cold chain logistics, minimizing energy consumption while meeting the fruit quality requirements of each order.

[0039] Figure 2 This is a flowchart illustrating a method for dynamic time-temperature control of cold storage for post-ripening fruits, as provided in an embodiment of the present invention. Figure 2 As shown, the method includes the following steps S201 to S205.

[0040] S201. Based on the requirements of fruit orders from fixed retailers, generate a 3D service window for each fruit order.

[0041] The 3D service window for each fruit order includes: maturity window, outbound time window, and quantity window.

[0042] For example, after the time-temperature dynamic control system receives an order for post-ripening fruit, it first constructs a maturity window, a time window, and a quantity window for the fruit to be shipped out, based on the requirements of the fruit order, thereby obtaining a 3D service window for each fruit order. Then, based on the current situation of the cold storage and the 3D service window of the fruit order, it can determine whether the maturity and inventory of the fruit in the cold storage can meet the needs of each fruit order.

[0043] For example, suppose the set of fruit orders is Θ, and the i-th fruit order is o i ∈Θ, This represents the i-th fruit order. i The 3D service window, where i is a positive integer, i = 1, ..., N k N k This indicates the total quantity of fruit orders; This represents the i-th fruit order. i The maturity window for outbound shipments, l i This represents the i-th fruit order. i Required lower limit of maturity, h i This represents the i-th fruit order. i Required upper limit of maturity; t i This represents the i-th fruit order. i Outbound time; v i This represents the i-th fruit order. i The demand for the quantity of goods, v i ∈[v i ,v i +κ], where κ represents the positive offset of the quantity of goods.

[0044] In this embodiment of the invention, the time-temperature dynamic control system can simultaneously control the temperature of at least one cold storage facility. When controlling multiple cold storage facilities, the system can combine the inventory of each cold storage facility and select the appropriate cold storage facility for temperature control based on order demand.

[0045] S202. Generate a list of fruit order services based on the outbound time of each fruit order.

[0046] Specifically, the time-temperature dynamic control system can schedule the outbound time of each order according to the order in the fruit order service list.

[0047] In this embodiment of the invention, fruit orders can be used as nodes, and all fruit orders in the fruit order service list can be sequentially linked according to the irreversible change path of fruit maturity in a time sequence to form an ordered unidirectional graph.

[0048] Among them, for the same type of fruit, the change in fruit ripeness is affected by the temperature control path, and the energy required is different when using different temperature control paths.

[0049] For example, the maturity change r is influenced by the temperature control path.

[0050] For example, if the cold storage temperature is T1, the energy required for the first fruit to change from maturity r1 to maturity r2 from time t1 to time t2 is E1; if the cold storage temperature is T2, the energy required for the first fruit to change from maturity r1 to maturity r3 from time t1 to time t2 is E2.

[0051] It should be noted that the inventory in the cold storage is guaranteed to meet at least the fruit order requirements of a fixed retailer.

[0052] After the temperature control path is planned based on this scheme, subsequent fruit orders from dynamic retailers can only be added to (updated) the fruit order service list if the fruit category in the dynamic retailer's fruit order belongs to the category of fruit currently stored in the cold storage.

[0053] S203. Based on the current information of the cold storage, the 3D service window of each fruit order, and the fruit order service list, generate temperature control constraints.

[0054] The current information of the cold storage includes at least one of the following: the type of fruit in the cold storage, the total inventory of each type of fruit, the maturity of each type of fruit, and the current temperature of the cold storage.

[0055] The temperature control constraints include at least one of the following: the total quantity of fruit orders meets the total inventory constraint, the constraint that each fruit order is completed within the 3D service window, the cold storage temperature constraint, and the overall service quality satisfaction deviation constraint.

[0056] (1) The total quantity of fruit orders meets the total inventory constraint.

[0057]

[0058] Among them, t k This represents the k-th time. N represents the inventory of the cold storage at time k. k This represents the total number of orders at time k. This indicates that the product should be delivered to N at time k. k The total volume of fruit orders placed by a retailer.

[0059] (2) Constraints on the completion of each fruit order within the 3D service window

[0060]

[0061] (3) Cold storage temperature constraints

[0062] T l ≤γ≤T h Formula (3)

[0063] Among them, T l T represents the lowest refrigeration temperature in a cold storage facility. h This indicates the highest refrigeration temperature in the cold storage.

[0064] By controlling the temperature in cold storage, physiological damage to fruits can be avoided due to excessively high or low temperatures.

[0065] Optionally, the overall service quality satisfaction deviation constraint includes: the deviation between the maturity window of the fruit order delivery and the center value of the maturity window required by the fruit order is greater than or equal to zero, and the outbound waiting time of the fruit orders in the fruit order service list is greater than or equal to zero.

[0066] (4) Constraints on the deviation of overall service quality satisfaction (i.e., constraints on decision variables)

[0067] ξ i ≥0,w i ≥0 Formula (4)

[0068] Where, ξ i This represents the i-th fruit order. i The estimated delivery maturity and the i-th fruit order o i The deviation of the required maturity window center value, ξ i =|r ti -(l i +h i ) / 2|,w i w represents the waiting time for the i-th fruit order to be shipped out. i =t i -e i e i This indicates the required shipping time for the i-th fruit order.

[0069] Optionally, if the orders in the fruit order service list change, such as adding a new fruit order or deleting a fruit order (the order has been completed, or the original order has not been completed and has been cancelled), the temperature control constraints can be updated again.

[0070] Optionally, the temperature control conditions can be updated periodically, before the next control cycle begins. For example, the temperature control path can be updated daily if no new orders are added.

[0071] S204. Based on temperature control constraints, construct a dual-objective dynamic optimization model for each fruit order's 3D service window and fruit order service list, based on comprehensive service quality satisfaction and total energy consumption.

[0072] It is understandable that, given that the fruit order service list consists entirely of fruit orders from fixed retailers, the bi-objective dynamic optimization model under the above four constraints can be used to characterize the TTM (Time to Market) of refrigerated post-ripening fruits with a 3D service window.

[0073] It is understandable that the decision variable in the bi-objective dynamic optimization model is the deviation between the actual delivery and the required maturity window center value.

[0074] S205. Based on the dual-objective dynamic optimization model, generate the temperature control path for the cold storage so that the temperature control equipment can dynamically control the temperature of the cold storage according to the temperature control path.

[0075] It is understood that, in the embodiments of the present invention, the temperature control path indicates the fruit refrigeration temperature used in the cold storage at different time periods.

[0076] For example, the temperature control path can be to use a first temperature in a first time period, a second temperature in a second time period, and a third temperature in a third time period.

[0077] Specifically, the control temperature T at the k-th time step k ∈γ needs to be decided for sustainable TTM.

[0078] This invention provides a time-temperature dynamic control method for cold storage of post-ripening fruits. First, based on the requirements of fruit orders from fixed retailers, a three-dimensional (3D) service window is generated for each fruit order, resulting in a maturity window, a delivery time window, and a quantity window for each order. Second, a fruit order service list is generated according to the delivery time of each fruit order. Third, temperature control constraints are generated based on the current information of the cold storage, the 3D service window of each fruit order, and the fruit order service list. Then, based on the temperature control constraints, the 3D service window of each fruit order, and the fruit order service list, a dual-objective dynamic optimization model based on comprehensive service quality satisfaction and total energy consumption is constructed. Finally, based on the dual-objective dynamic optimization model, a temperature control path for the cold storage is generated, enabling the temperature control equipment to dynamically control the temperature of the cold storage according to the temperature control path. In other words, during the process of determining the temperature control path, a dual-objective optimization model is constructed based on the actual needs of each order and the deviation of the overall service quality satisfaction. This model aims to minimize the power consumption of the cold storage while meeting the needs of each order. It can balance quality and energy consumption in fruit cold chain logistics, and minimize energy consumption in fruit cold chain logistics while meeting the delivery quality of each fruit order.

[0079] Optionally, in the time-temperature dynamic control method for cold storage of post-ripening fruits provided in the embodiments of the present invention, the above-mentioned S204 may specifically include the following S204a to S204d.

[0080] S204a. Construct an evolutionary model of the fruit ripening process.

[0081] Specifically, in this embodiment of the invention, a model of fruit maturity change can be constructed based on the basic dynamic response theory using formula (5).

[0082]

[0083] Where ε represents the dynamic coefficient of change, and n represents the reaction order of the model.

[0084] Specifically, in this embodiment of the invention, a zero-order reaction model is used as a model for the change in fruit maturity to predict the quantitative quality maturity of fruit at the time of shipment. For the i-th fruit order, when the reaction order n = 0, the shipment time t of the i-th fruit order is predicted using the zero-order reaction model. i Quantitative quality and ripeness of fruit in, This represents the quantitative quality maturity of the fruit at time k.

[0085] In this embodiment of the invention, the real-time influence of fruit temperature on dynamic parameters can be modeled based on the Arrhenius equation, and the predicted temperature control path T can be dynamically obtained. k rate of change Where A represents the Arrhenius exponent, Γ represents the activation energy of the reaction (in kJ / mol), and R represents the universal gas constant, R = 8.314 J / (molgK). Indicates temperature control path T k The absolute temperature, where the unit conversion relationship between Celsius and absolute temperature can be shown below. (K indicates that the unit is Kelvin).

[0086] S204b: Based on the fruit ripening process evolution model and refrigeration temperature, predict the maturity of fruit.

[0087] Specifically, based on the fruit ripening process evolution model indicated by the above formula, the ripeness of fruit at different refrigeration temperatures and various time periods can be predicted.

[0088] S204c: Based on the deviation between the predicted delivery maturity and the center value of the required maturity window, and the outbound waiting time of fruit orders, generate the service quality deviation for each fruit order; based on the quantity of goods in each fruit order and the service quality deviation of each fruit order, determine the comprehensive service quality deviation of all fruit orders in the fruit order service list.

[0089] It is understandable that service quality is related to delivery maturity and delivery time. The greater the deviation in maturity at the time of delivery, the greater the deviation in service quality; the longer the waiting time for outbound shipment, the greater the deviation in service quality.

[0090] S204d: Based on the temperature difference between inside and outside the cold storage and the surface area of ​​the cold storage, a dynamic energy consumption determination model is constructed to determine the power consumption during the temperature control cycle.

[0091] For example, in an embodiment of the present invention, the energy consumption of the refrigeration unit at the k-th time step is calculated based on the effective time-temperature combination of the dynamic temperature difference indicated by formula (6) driving the improved heat load formula.

[0092]

[0093] Among them, E k This represents the energy consumption of the refrigeration unit at the k-th time step, where the starting time of the k-th time step is time t. k The end time is the (k+1)th time t. k+1 u represents the heat transfer coefficient, s represents the surface area of ​​the cold storage, λ represents the coefficient of performance (COP), and ΔT kT represents the temperature difference between the inside and outside of the cold storage at time k. a The temperature outside the cold storage is represented by T. k =T a -△T k By adjusting the temperature of the temperature control path inside the cold storage, the specific energy consumption required for different temperature control paths can be obtained.

[0094] S204e, Construct a dual-objective dynamic optimization model that minimizes the deviation of integrated service quality and the total energy consumption.

[0095] Specifically, the first objective function indicates the minimization of the overall service quality deviation, and the second objective function indicates the minimization of total energy consumption.

[0096]

[0097] Where f1 represents N k The minimum overall service deviation of a retailer's fruit order, f2 represents the minimum energy consumption at the k-th time step, and s i This represents a single retailer's fruit order. i Service deviation, s i =p1·ξ' i +p2·w' i p1, p2 represent weighting coefficients, ξ' i Indicates an order for fruit. i The deviation ξ between actual delivery and the required maturity window center value i Normalize, w' i Indicates an order for fruit. i Outbound waiting time w i Normalize,

[0098] Specifically, the refrigeration unit adjusts the temperature according to the currently optimized temperature control target path until a new decision is made at the next time step.

[0099] The above-mentioned dynamic calculation model of energy consumption E k It can provide a key control parameter for TTM sustainable decision-making.

[0100] Based on this scheme, the fruit ripening process evolution model is constructed to predict the maturity of the fruit and generate the service quality deviation for each fruit order; the power consumption during the temperature control cycle is determined by constructing a dynamic energy consumption determination model; finally, a dual-objective dynamic optimization model that minimizes both the overall service quality deviation and the total energy consumption can be constructed.

[0101] Optionally, in the time-temperature dynamic control method for cold storage of post-ripening fruits provided in the embodiments of the present invention, the above-mentioned bi-objective dynamic optimization model can be solved by some improved initial solution algorithms. For example, the bi-objective dynamic optimization model can be solved based on the Non-dominated Sorting Genetic Algorithm II (NSGA-II). Therefore, the above-mentioned S205 may specifically include the following S205a.

[0102] S205a. In the process of solving the bi-objective dynamic optimization model based on the non-dominated sorting genetic algorithm II, when generating the initial solution set at the (k+1)th time step, half of the front solution set obtained at the kth time step is merged with half of the initial solution set generated at the (k+1)th time step to form the initial solution set at the (k+1)th time step.

[0103] It is understandable that the method for generating the initial solution set of the existing NSGA-II has been improved in the process of solving the problem based on the existing NSGA-II.

[0104] Specifically, based on the memory-enhanced NSGA-II, when calculating the solution at the (k+1)th time step, an initial solution set is randomly generated, and the leading edge solution at the kth time step is used as prior knowledge of the potential search region. Half of the initial solutions are randomly generated, and half of the previous optimized solutions and half of the new random solutions are combined to generate the initial solution at the kth time step, thereby enabling the optimal temperature control path to be obtained quickly and accurately.

[0105] During the solution process, the fitness of candidate solutions is evaluated with the optimization objectives of comprehensive service deviation and energy consumption. Two parent solutions are selected using the binary tournament selection operator for offspring solution propagation. A niche strategy without external parameters is adopted, assigning different ranks to solutions with the same non-dominated rank based on solution density to distinguish solutions with the same non-dominated rank.

[0106] Specifically, through binary encoding, two parent solutions are used to generate two child solutions y. 1j (k) and y 2j (k).

[0107]

[0108] Where, γ j The distribution index represents the affinity between parents and offspring; the larger the distribution index, the closer the parents and offspring are. α j η represents a random number generated in the range [0-1]. c For custom positive or negative exponents.

[0109] Specifically, the distribution of the solutions generated by the parent generation can be controlled based on the distribution index.

[0110] In this embodiment of the invention, a polynomial mutation operator with perturbation parameters can be used to mutate neighboring solutions (offspring) from the current solution (parent generation) based on formula (10). This process increases population diversity to explore promising regions in the search space. The search space indicates the range of solutions to the optimal solution across different dimensions.

[0111] y j+1 (k)=y j (k)+β j Formula (11)

[0112] Among them, y j (k) represents the parent solution, y j+1 (k) represents the offspring solution obtained by mutation, β j Indicates the disturbance parameter. η m A custom positive or negative exponent is used to adjust the magnitude of random fluctuations in the perturbation parameters.

[0113] When adjusting the temperature of the cold storage at the k-th time step, the comprehensive loss function of formula (12) can be used as the decision criterion during the solution process, which can enable the improved NSGA-II to quickly search a set of Pareto front solutions, where each solution corresponds to a control path.

[0114]

[0115] Where Ψ(·) represents the comprehensive loss function, y * (k) represents the definite solution at the k-th time step, Y Front (k) represents the frontier solution set at the k-th time step, ρ1 represents the service satisfaction loss coefficient of the first objective function, and ρ2 represents the energy consumption loss coefficient corresponding to the second objective function.

[0116] Based on this scheme, the improved non-dominated sorting genetic algorithm II can be used to solve the dual-objective dynamic optimization model. By combining half of the previous optimized solution and half of the new random solution, an initial solution for the current time step can be generated, thereby balancing efficiency and population diversity. The resulting temperature control path is more in line with the dual-objective requirements mentioned above.

[0117] Optionally, in the time-temperature dynamic control method for cold storage of post-ripening fruits provided in the embodiments of the present invention, the first temperature control path is a temperature control path for a determined first control cycle, and after the above-mentioned S205, it may also include the following S206 to S208a or S206 to S208b.

[0118] S206. During the process of dynamically controlling the temperature in the first control cycle according to the first temperature control path, the maturity of the fruit and the energy consumption of the cold storage in the second control cycle are predicted based on the fruit ripening process evolution model and the temperature indicated by the first temperature control path.

[0119] Optionally, the temperature indicated by the temperature control path can be the same temperature or multiple different temperatures, and the embodiments of the present invention do not specifically limit this.

[0120] It is understandable that during the temperature control process using the first temperature control path generated in the previous control cycle, if it is detected that the next cycle (i.e., the second control cycle) is about to begin, the ripeness of the fruit and the energy consumption of the cold storage can be predicted based on the temperature used in this cycle if the temperature is continued to be used to refrigerate the fruit in the cold storage in the next control cycle.

[0121] S207. Based on the predicted maturity and cold storage energy consumption of the second control period, determine the comprehensive service deviation and total cold storage energy consumption of all fruit orders in the fruit order service list within the second control period.

[0122] It is understandable that after predicting the maturity of the fruit in the cold storage during the second control cycle, the comprehensive service deviation corresponding to the second control cycle can be determined based on the predicted maturity, and the total energy consumption of the cold storage during the second control cycle can be determined based on the first temperature control path generated using the previous control cycle.

[0123] S208a. If the overall service deviation and total energy consumption in the second control cycle meet the temperature control conditions, then the temperature will continue to be dynamically controlled in the second control cycle according to the first temperature control path.

[0124] S208b If the overall service deviation or total energy consumption in the second control cycle does not meet the temperature control conditions, then based on the current information of the cold storage, the current maturity of the fruit, the fruit order service list in the second control cycle, and the dual-objective dynamic optimization model, the first temperature control path is updated to the second temperature control path.

[0125] It is understandable that if the overall service deviation of the cold storage refrigerated fruit in the second control cycle using the first temperature control path is greater than the first deviation value, it means that the temperature control path of the previous cycle cannot be used. Otherwise, the fruit maturity may be too high or too low, and there may be orders in the fruit order service list that cannot be completed in the 3D service window.

[0126] Specifically, the temperature control constraints can be updated first based on the current information of the cold storage, the current maturity of the fruit, and the fruit order service list in the second control period. Then, the two objective functions of the bi-objective dynamic optimization model can be updated. Based on the updated bi-objective dynamic optimization model and the temperature control constraints, the temperature control path for the second control period can be generated.

[0127] Based on this scheme, the time-temperature dynamic control system can predict the fruit ripeness in future time periods during temperature control, thereby determining whether the temperature control path needs to be adjusted. If no adjustment is needed, the original temperature control path can continue to be used. If adjustment is required, the temperature control path can be updated based on the current information of the cold storage, the current ripeness of the fruit, the fruit order service list in the second control cycle, and the bi-objective dynamic optimization model, thus making temperature control more flexible and adaptable.

[0128] Optionally, in the time-temperature dynamic control method for cold storage of post-ripening fruits provided in the embodiments of the present invention, after S207 above, the method may further include S209 below.

[0129] S209. If the overall service deviation and total energy consumption in the second control period meet the temperature control conditions, then based on the overall service deviation and total energy consumption in the second control period, plan the delivery time of the fruit orders in the fruit order service list in the second control period.

[0130] It is understandable that if the overall service deviation within the second control period is less than or equal to the first deviation value, it means that the ripeness and delivery time of the fruit orders in the fruit order service list can be met, and the fruit orders with the delivery time window within the second control period can be delivered according to the order requirements.

[0131] Based on this scheme, if the predicted comprehensive service deviation in the future time period meets the deviation requirements, the delivery time of fruit orders can be planned according to the predicted comprehensive service deviation, so that the fruit maturity meets the order requirements as much as possible and the waiting time is shortened as much as possible, thereby dynamically reducing the overall power consumption.

[0132] Optionally, in the time-temperature dynamic control method for cold storage of post-ripening fruits provided in the embodiments of the present invention, after S205 above, S210 to S213 may also be included below.

[0133] S210. During the process of the temperature control equipment dynamically controlling the temperature according to the temperature control path, if the first fruit order from the dynamic retailer is received, the 3D service window of the first fruit order is determined.

[0134] It is understandable that in practical applications, users can have both fixed retailers and dynamic retailers.

[0135] S211. Determine whether the current cold storage conditions allow for supply within the 3D service window of the first fruit order.

[0136] After generating temperature control paths based on orders from fixed retailers, if a dynamic retailer generates a fruit order, the ripeness of the fruit can be predicted based on the temperature control paths in the cold storage and the fruit ripening process evolution model, thus determining whether the fruit can be supplied within the dynamic retailer's 3D service window.

[0137] S212. If the goods can be supplied within the 3D service window of the first fruit order, then add the first fruit order to the fruit order service list.

[0138] S213. Continue to dynamically control the temperature according to the temperature control path, and supply the goods within the 3D service window of the first fruit order.

[0139] Based on this scheme, after a new order is placed, the 3D service window for the new order can be determined based on the order information. The ripeness of the fruit can be predicted based on the current temperature control path and the fruit ripening process evolution model to determine whether the needs of the new order can be met. If the needs of the new order are met, the temperature of the cold storage can continue to be controlled according to the current temperature control path, and the fruit can be supplied within the 3D service window of the first fruit order.

[0140] Optionally, in the time-temperature dynamic control method for cold storage of post-ripening fruits provided in the embodiments of the present invention, after S212 above, S214 to S216 may also be included below.

[0141] S214. After adding the first fruit order to the fruit order service list and obtaining the updated fruit order service list, update the temperature control conditions based on the 3D service window of the first fruit order and the updated fruit order service list.

[0142] The temperature control conditions mentioned above also include: an insertion list constraint for fruit orders from dynamic retailers; the insertion list constraint indicates that the maturity window required by the inserted fruit order covers the maturity window of the fruit in the cold storage within the outbound time window of the inserted fruit order under the current temperature control path.

[0143] (5) Insert list constraint for fruit orders of dynamic retailers.

[0144] Dynamic retailer fruit orders d A 3D service window that can insert a list of fruit order services. d (r d ,t d ,vd The constraint is expressed by the following formula (13).

[0145]

[0146] Among them, fruit orders o d Maturity window r d ∈[l d ,h d The current temperature path T should be covered. k and time window t d ∈[e d ,μ d Possible maturity value And the quantity requirement V of goods d Less than the current remaining inventory.

[0147] S215. Based on the updated temperature control constraints, the updated fruit order service list, and the 3D service window of each fruit order in the updated fruit order service list, update the bi-objective dynamic optimization model.

[0148] In this scenario, the bi-objective dynamic optimization model under the above five constraints can be used to characterize the TTM (Time to Tear) of post-ripening fruits with 3D service windows.

[0149] S216. Update the temperature control path based on the updated bi-objective dynamic optimization model.

[0150] Based on this scheme, if some new orders are added during the temperature control process, after it is determined that they can be added to the fruit order service list, the temperature control conditions and the objective function of the bi-objective dynamic optimization model can be updated. Then, based on the updated bi-objective model, the improved NSGA-II algorithm can be used again to solve the problem and generate a new temperature control path. The temperature control path can be updated again to adapt to the needs of the fruit orders in the fruit order service list.

[0151] Optionally, in the time-temperature dynamic control method for cold storage of post-ripening fruits provided in the embodiments of the present invention, the above-mentioned S215 can be specifically executed by the following S215a.

[0152] S215a. According to the preset rules, assign weight coefficients of two objective functions to fruit orders from fixed retailers and fruit orders from dynamic retailers respectively, and update the bi-objective dynamic optimization model.

[0153] The preset rules indicate the weight coefficients of the two objective functions at different time periods.

[0154] It is understood that if fixed retailers are satisfied first and dynamic retailers are satisfied later, then fixed retailers can be assigned a larger weight and dynamic retailers a smaller weight. The specific settings can be made according to actual needs, and this embodiment of the invention does not impose specific limitations on this.

[0155] For example, in the first time period, the fruit is mainly sold by fixed retailers, with fewer dynamic retailers. The weighting coefficients for fixed retailers and dynamic retailers are 0.8 and 0.2, respectively. In the second time period, the fruit is mainly sold by fixed retailers, but the order volume from dynamic retailers also increases dramatically. In this case, the weighting coefficients for fixed retailers and dynamic retailers can be set to 0.6 and 0.4, respectively.

[0156] Based on this scheme, different weight coefficients can be assigned to different types of fruit orders according to the preset rules, and the bi-objective dynamic optimization model can be updated, thereby dynamically adjusting the importance of different types of retailers at different time periods.

[0157] Example:

[0158] In the simulation experiment of the time-temperature dynamic control method for cold storage of post-ripening fruits provided in this invention, a kiwi fruit order instance was designed, including 12 fixed retailers and 6 dynamic retailers. A cold storage facility was used as an example for the simulation experiment. The orders from fixed retailers were processed first, followed by the orders from dynamic retailers. The simulation was conducted based on the established bi-objective dynamic optimization model and the improved NSGA-II algorithm.

[0159] 3-1. Order Request Information

[0160] Table 1 is a list of kiwi fruit order demand information for mixed retailers. Table 1 shows the detailed order demand information for each kiwi fruit order, including order attributes, maturity window, outbound time window, and quantity, with a time step of 1 day.

[0161] As can be seen from Table 1, the maturity window and the outbound time window are different for each order.

[0162] Table 1 - List of Kiwi Fruit Order Demands from Mixed Retailers

[0163]

[0164] 3-2 Model Parameter Configuration

[0165] The model parameters are mainly derived from the established bi-objective dynamic optimization model and the parameters required in the improved NSGA-II algorithm. Table 2 is a schematic table of the parameter configuration of a bi-objective dynamic optimization model provided by the present invention, wherein the detailed parameter configuration of the bi-objective dynamic optimization model is shown in Table 2. To ensure consistency with actual cold storage scenarios, the empirical data used were obtained through industry surveys or reference materials, and the relevant sources have been noted in Table 2.

[0166] Table 3 is a schematic table of parameter configuration for a dual-objective dynamic optimization model provided by this invention. The parameter settings for the improved NSGA-II algorithm use the optimal combination of main parameters suggested in existing literature: population size PS = 40, crossover rate Pc = 1, and mutation rate Pm = 0.01. Simultaneously, after balancing algorithm efficiency and performance, the maximum number of generations, max_Gen, is set to 1000. The custom exponent is set to η. c =20 and η m =20. In addition, based on empirical estimates, the loss coefficients are ρ1 = 0.5 and ρ2 = 0.5, respectively.

[0167] Table 2 - Parameter Configuration of the Bi-objective Dynamic Optimization Model

[0168] variable definition value References <![CDATA[r0]]> Initial maturity (FF) 11.89 kg·cm⁻² Sun, et al., 2020 A Arrhenius index 1.546×107 Sun, et al., 2020 Γ Activation energy of reaction 4.229×104 Sun, et al., 2020 R Universal gas constant 8.314 Sun, et al., 2020 u heat transfer coefficient 0.0007kW / m2 / ℃ Fanetal., 2021 s cold storage surface area 270m2 Fanetal., 2021 λ Coefficient of performance (COD) 1 Fanetal., 2021 <![CDATA[T a ]]> external temperature 20℃ - <![CDATA[T l ]]> Minimum storage temperature 0℃ Asichee et al., 2017 <![CDATA[T h ]]> Maximum storage temperature 20℃ Asichee et al., 2017 <![CDATA[V t0 ]]> Initial inventory 50 - <![CDATA[p1]]> Maturity weighting coefficient 0.5 -

[0169] Table 1 - Parameter Configuration of Improved NSGA-II Algorithm

[0170] variable meaning value PS Population size 40 Pc Cross rate 1 Pm Variation rate 0.01 max_Gen Maximum number of iterations 1000 <![CDATA[η c ]]> Custom cross index 20 <![CDATA[η m ]]> Custom Mutation Index 20 <![CDATA[ρ1]]> Service satisfaction loss coefficient 0.5 <![CDATA[ρ2]]> Energy loss coefficient 0.5

[0171] Figure 3 This diagram illustrates a Pareto front solution at different times based on an improved NSGA-II algorithm, as provided in an embodiment of the present invention. The improved NSGA-II algorithm is applied to a dynamic bi-objective TTM for kiwifruit cold storage, enabling real-time decision-making for optimized temperature control paths. Figure 3 As shown in the figure, the improved NSGA-II is included at different time steps (t). k＝0 ,t k＝10 The Pareto front solution under () is used to determine t by comparing the solution sets. k＝10 The quality of the optimized solution is better than t k＝0 Specifically, t k＝0 The f1 value at time t is slightly lower than that at time t. k＝10 , and t k＝0 The value of f2 at time t is much higher than that at time t. k＝10 In other words, in the early stages of TTM, there is more room to optimize maturity changes, and as the service window approaches, the effective range of temperature control also expands.

[0172] 3-3 Robustness verification of different insertion scenarios for dynamic orders

[0173] The robustness of the method of this invention is designed by analyzing the cost changes caused by dynamic order insertion. With the emergence of new demand, smaller fluctuations are expected in the previous temperature control path, indicating that the model has stronger dynamic adaptability. Among them, robustness is defined based on the custom formula (14), where Λ∈(0,1], and the closer it is to 1, the more applicable the TTM strategy of this invention is to the dynamic demand scenario of retailers.

[0174]

[0175] Where, N d This is the total number of dynamically inserted orders. and This represents the d-th dynamic order o d The combined cost loss before and after insertion.

[0176] Two case studies were implemented to verify the robustness under different dynamic order insertion scenarios.

[0177] Case 1: Simultaneous insertion of dynamic orders

[0178] Table 4 shows an example row table for inserting different quantities of dynamic orders simultaneously. This can be based on the data N in Table 4. d Robustness verification was performed using the numbers 1, 2, 3, 4, 5, 6.

[0179] Table 2 - Simultaneous Insertion of Different Quantities of Dynamic Orders

[0180] <![CDATA[N d ]]> Inserted dynamic orders 1 13 2 13，14 3 13，14，15 4 13，14，15，16 5 13，14，15，16，17 6 13，14，15，16，17，18

[0181] For different insertion times (t) k＝2 ,t k＝5 ,t k＝10 The robustness variation curve with the amount of insertion is shown in the figure. Figure 4 As shown. Figure 4 This is a robustness variation curve diagram for simultaneous insertion of dynamic order demands, provided as an embodiment of the present invention. Based on calculation results under different insertion times, Λ generally falls between 0.82 and 1, indicating that the method of the present invention has strong robustness. Figure 4 In the data, all curves show a downward trend, indicating that the more orders inserted, the worse the robustness. This is because a single temperature control path cannot accurately meet multiple dynamic demands simultaneously, leading to accumulated service deviations. Meanwhile, from different curves (t... k＝2 ,t k＝5 ,t k＝10 Comparing the two models, the insertion time has almost no impact on robustness, which further reveals the strong adaptability of the bi-objective dynamic optimization model.

[0182] Case 2: Dynamic Order Scattered Insertion

[0183] Table 5 shows a sample row table of inserting the same number of dynamic orders at different time points. The first row inserts all orders at once; the second row inserts them in two separate rows of three orders each; and the third row inserts them in three separate rows of two orders each.

[0184] Table 3 - Dynamic orders with the same number inserted at different times

[0185] Number of order insertions <![CDATA[t k＝2 ]]> <![CDATA[t k＝5 ]]> <![CDATA[t k＝10 ]]> c＝1 13，14，15，16，17，18 - - c＝2 13，14，15 16，17，18 - c＝3 13，14 14，15 17，18

[0186] The robustness values ​​for different insertion counts (c = 1, 2, 3) are shown in the box plots. Figure 5 This is a schematic diagram illustrating the robustness of a dynamic order demand distributed insertion method according to an embodiment of the present invention, as shown below. Figure 5 As shown, Λ is 0.837, 0.915, and 0.942 for c=1, c=3, respectively. The results indicate that distributed insertion improves the model's robustness because concentrated demand makes it more difficult to simultaneously coordinate the overall optimal solution. Therefore, in practical cold storage environments, a unified demand-driven TTM is beneficial for balancing multiple objectives and improving overall sustainability performance.

[0187] 3-4: Comparison of sustainability with constant temperature TTM

[0188] To verify the sustainability of the proposed method, its sustainability was compared with that of traditional constant-temperature refrigeration at different temperature gradients (0°C, 4°C, and 8°C) in terms of service satisfaction and energy consumption. Table 6 shows the sustainability energy conversion results for different temperature gradients.

[0189] Table 4 - Sustainability Results of Different TTM Strategies

[0190] Temperature path Number of completed orders <![CDATA[f1]]> <![CDATA[f2]]> Ψ(·) 0℃ 16 / 18 17.655 3.780 10.718 4℃ 18 / 18 7.197 3.024 5.111 8℃ 15 / 18 6.633 2.268 4.451 Optimize path 18 / 18 4.138 2.609 3.374

[0191] As shown in Table 6, the TTM method provided in this application demonstrates a significant advantage in service satisfaction, providing satisfactory service to all 18 retailers at the lowest cost, f1 = 4.138. In terms of energy consumption, the energy consumption of dynamic temperature storage is only slightly higher than that of constant 8°C low-temperature storage, remaining competitive to a certain extent. Compared with commonly used constant temperature management, the optimized TTM method generally improves sustainability, with the lowest overall loss value being Ψ(·) = 3.374.

[0192] Figure 6 This is a schematic diagram illustrating the temperature control and quality maturity change path of a constant 0°C TTM strategy provided in an embodiment of the present invention. Figure 7 This is a schematic diagram illustrating the temperature control and quality maturity change path of a constant 4°C TTM strategy provided in an embodiment of the present invention. Figure 8 This is a schematic diagram illustrating the temperature control and quality maturity change path of a constant 8°C TTM strategy provided in an embodiment of the present invention. Figure 9 This diagram illustrates the temperature control and quality maturity change path of an optimized TTM strategy provided in this embodiment of the invention. The temperature control paths (solid yellow lines) and their corresponding maturity index changes (dashed blue lines) under different TTM strategies are shown below. Figures 6 to 9 As shown, the black and red two-dimensional squares represent the complete and incomplete service windows related to quality maturity and outbound time, respectively. It is clearly evident that the constant temperature warehousing model struggles to flexibly meet the service needs of different orders.

[0193] like Figure 6 As shown, under constant conditions of 0℃, the maturity change path cannot pass through all blocks, indicating that two orders (No.1 and No.14) cannot be completed as required. Similarly, under constant conditions of 8℃, Figure 8 There are 3 unfulfilled orders (No. 5, No. 16, and No. 18). Although the constant TTM at 4°C meets... Figure 7 All order requirements are met, but a linear quality maturity path is difficult to dynamically respond to various random requirements.

[0194] like Figure 9 As shown, the proposed dynamic TTM strategy can effectively achieve online autonomous optimization decision-making under design constraints. However, Figure 9 The maturity index changes along a path where the actual flesh firmness (FFs) of later-stage orders generally approaches the lower limit of the corresponding window. This is because orders are gradually completed over time, leading to a decrease in the overall weight of remaining service orders. Conversely, the importance of low consumption increases relatively as inventory decreases to achieve sustainable energy efficiency. Therefore, the temperature is controlled at a higher level in the later stages of TTM, approximately 11°C. After the final order is completed, the storage temperature is adjusted to a minimum of 0°C.

[0195] In summary, the dynamic bi-objective optimization method proposed in this study can effectively balance service satisfaction and energy consumption, and achieve sustainable temperature management for the cold storage of post-ripening fruits.

[0196] Corresponding to the embodiments of the foregoing methods, the present invention also provides embodiments of the apparatus and the computer equipment on which it is applied.

[0197] Embodiments of the device of this invention can be applied to computer equipment, such as servers or terminal devices. The device embodiments can be implemented through software, hardware, or a combination of both. Taking software implementation as an example, a logically defined device is formed by the processor of the time-temperature dynamic control system for post-ripening fruit cold storage loading the corresponding computer program instructions from non-volatile memory into memory for execution. From a hardware perspective, such as... Figure 10 The diagram shown is a hardware structure diagram of the computer equipment containing the time-temperature dynamic control device for post-ripening fruit cold storage according to an embodiment of the present invention. Except for... Figure 10 In addition to the processor 1010, memory 1030, network interface 1020, and non-volatile memory 1040 shown, the server or electronic device where the time-temperature dynamic control device 1031 for the refrigeration of post-ripening fruits in the embodiment is located may also include other hardware depending on the actual function of the computer device, which will not be described in detail here.

[0198] Figure 10 This invention provides a time-temperature dynamic control device for the cold storage of post-ripening fruits. The time-temperature dynamic control device 1000 includes: a 3D service window generation module 1001, an order service list generation module 1002, a temperature control constraint generation module 1003, a control model construction module 1004, a temperature control path generation module 1005, and a temperature control module 1006. The 3D service window generation module 1001 is used to generate a three-dimensional 3D service window for each fruit order based on the requirements of a fixed retailer's fruit order. The 3D service window for each fruit order includes: a maturity window, a delivery time window, and a quantity window. The order service list generation module 1002 is used to generate a fruit order service list according to the delivery time of each fruit order. The temperature control constraint generation module 1006... 03. Based on the current information of the cold storage, the 3D service window of each fruit order, and the fruit order service list, generate temperature control constraints. The temperature control constraints include at least one of the following: the total quantity of fruit orders meets the total inventory constraint, the constraint of each fruit order being completed within the 3D service window, the cold storage temperature constraint, and the comprehensive service quality satisfaction deviation constraint. The control model construction module 1004 is used to construct a dual-objective dynamic optimization model based on comprehensive service quality satisfaction and total energy consumption based on the temperature control constraints, the 3D service window of each fruit order, and the fruit order service list. The temperature control path generation module 1005 is used to generate the temperature control path of the cold storage based on the dual-objective dynamic optimization model, so that the temperature control module 1006 controls the temperature control equipment to dynamically control the temperature of the cold storage according to the temperature control path.

[0199] Optionally, the overall service quality satisfaction deviation constraint includes: the deviation between the maturity window of the fruit order delivery and the center value of the maturity window required by the fruit order is greater than or equal to zero, and the outbound waiting time of the fruit orders in the fruit order service list is greater than or equal to zero.

[0200] Optionally, the control model construction module is used to: construct an evolution model of the fruit ripening process; predict the maturity of the fruit based on the fruit ripening process evolution model and the refrigeration temperature of the cold storage; generate the service quality deviation for each fruit order based on the deviation between the predicted delivery maturity and the center value of the required maturity window, and the outbound waiting time of the fruit order; determine the comprehensive service quality deviation of all fruit orders in the fruit order service list based on the cargo quantity of each fruit order and the service quality deviation of each fruit order; construct a dynamic energy consumption determination model based on the temperature difference inside and outside the cold storage and the surface area of ​​the cold storage to determine the power consumption within the temperature control cycle; and construct a dual-objective dynamic optimization model that minimizes the comprehensive service quality deviation and the total energy consumption.

[0201] Optionally, the first temperature control path is a determined temperature control path for a first control period; the time-temperature dynamic control device further includes: a prediction module, a determination module, and an update module; the prediction module is used to predict the maturity of the fruit and the cold storage energy consumption for the second control period based on the fruit ripening process evolution model and the temperature indicated by the first temperature control path during the process of dynamically controlling the temperature according to the first temperature control path in the first control period; the determination module is used to determine the comprehensive service deviation and total cold storage energy consumption of all fruit orders in the fruit order service list within the second control period based on the predicted maturity and cold storage energy consumption of the second control period; the temperature control module is used to continue to dynamically control the temperature according to the first temperature control path within the second control period if the comprehensive service deviation and total energy consumption within the second control period meet the temperature control conditions, and the temperature control conditions indicate that the shipment meets the 3D service window of each fruit order; the update module is used to update the first temperature control path to the second temperature control path if the comprehensive service deviation and total energy consumption do not meet the temperature control conditions, based on the current information of the cold storage, the current maturity of the fruit, the fruit order service list within the second control period, and the bi-objective dynamic optimization model.

[0202] Optionally, the time-temperature dynamic control device further includes a shipment planning module; the shipment planning module is used to plan the shipment time of fruit orders in the fruit order service list within the second control period based on the comprehensive service deviation and total energy consumption within the second control period, if the comprehensive service deviation and total energy consumption within the second control period meet the temperature control conditions.

[0203] Optionally, the 3D service window generation module is also used to determine the 3D service window of the first fruit order if a first fruit order from a dynamic retailer is received during the process of the temperature control equipment dynamically controlling the temperature according to the temperature control path; the order service list generation module is used to determine whether the current cold storage conditions can supply the goods within the 3D service window of the first fruit order; if the goods can be supplied within the 3D service window of the first fruit order, the first fruit order is added to the fruit order service list; the temperature control module is used to continue to dynamically control the temperature according to the temperature control path and supply the goods within the 3D service window of the first fruit order.

[0204] Optionally, the update module is further configured to, after adding the first fruit order to the fruit order service list to obtain an updated fruit order service list, update the temperature control conditions based on the 3D service window of the first fruit order and the updated fruit order service list. The temperature control conditions also include: an insertion list constraint for the dynamic retailer's fruit orders; an insertion list constraint indicating that the maturity window required by the inserted fruit order covers the maturity window of the fruit in the cold storage within the outbound time window of the inserted fruit order under the current temperature control path; based on the updated temperature control constraints, the updated fruit order service list, and the 3D service window of each fruit order in the updated fruit order service list, update the bi-objective dynamic optimization model; and based on the updated bi-objective dynamic optimization model, update the temperature control path.

[0205] Optionally, the update module is specifically used to: assign weight coefficients of two objective functions to fruit orders from fixed retailers and fruit orders from dynamic retailers according to preset rules, and update the bi-objective dynamic optimization model.

[0206] Optionally, the temperature control path generation module is specifically used to, during the process of solving the bi-objective dynamic optimization model based on the non-dominated sorting genetic algorithm II, when initializing and generating the initial solution set at the (k+1)th time step, merge half of the leading solution set from the leading solution set solved at the k-th time step with half of the initial solution set generated at the (k+1)th time step as the initial solution set for the (k+1)th time step.

[0207] This invention provides a time-temperature dynamic control device for cold storage of post-ripening fruits. First, based on the requirements of fruit orders from fixed retailers, a three-dimensional (3D) service window is generated for each fruit order, resulting in a maturity window, a delivery time window, and a quantity window for each order. Second, a fruit order service list is generated according to the delivery time of each fruit order. Third, temperature control constraints are generated based on the current information of the cold storage, the 3D service window of each fruit order, and the fruit order service list. Then, based on the temperature control constraints, the 3D service window of each fruit order, and the fruit order service list, a dual-objective dynamic optimization model based on comprehensive service quality satisfaction and total energy consumption is constructed. Finally, based on the dual-objective dynamic optimization model, a temperature control path for the cold storage is generated, enabling the temperature control equipment to dynamically control the temperature of the cold storage according to the temperature control path. In other words, during the process of determining the temperature control path, a dual-objective optimization model is constructed based on the actual needs of each order and the deviation of the overall service quality satisfaction. This model aims to minimize the power consumption of the cold storage while meeting the needs of each order. It can balance quality and energy consumption in fruit cold chain logistics, and minimize energy consumption in fruit cold chain logistics while meeting the delivery quality of each fruit order.

[0208] Accordingly, the present invention also provides a time-temperature dynamic control system for cold storage of post-ripening fruits. This system includes a processor and a memory for storing processor-executable instructions. The processor is configured to: generate a three-dimensional (3D) service window for each fruit order based on the requirements of a fixed retailer's fruit order; each fruit order's 3D service window includes a ripeness window, a delivery time window, and a quantity window; generate a fruit order service list according to the delivery time of each fruit order; and generate temperature control constraints based on the current cold storage information, the 3D service window of each fruit order, and the fruit order service list. The temperature control constraints include at least one of the following: the total quantity of fruit orders meets the total inventory constraint, the constraint that each fruit order is completed within the 3D service window, the cold storage temperature constraint, and the overall service quality satisfaction deviation constraint; based on the temperature control constraints, a dual-objective dynamic optimization model based on the overall service quality satisfaction and total energy consumption is constructed for each fruit order's 3D service window and the fruit order service list; based on the dual-objective dynamic optimization model, a temperature control path for the cold storage is generated so that the temperature control equipment dynamically controls the temperature of the cold storage according to the temperature control path.

[0209] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the various steps in the above embodiment of the time-temperature dynamic control method for cold storage of post-ripening fruits.

[0210] The present invention also provides a computer device, the computer device including a memory, a processor and computer-readable instructions stored in the memory and executable on the processor, wherein when the computer-readable instructions are executed by the processor, they implement the various steps in the above embodiment of a time-temperature dynamic control method for cold storage of post-ripening fruits.

[0211] The specific implementation process of the functions and roles of each module in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0212] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of the present invention according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0213] The foregoing has described specific embodiments of the invention. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0214] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention claimed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not claimed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0215] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

[0216] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for dynamic time-temperature control of cold storage for post-ripening fruits, characterized in that, The method includes: Based on the requirements of fruit orders from fixed retailers, a 3D service window is generated for each fruit order; the 3D service window for each fruit order includes: ripeness window, outbound time window, and quantity window; Generate a list of fruit order services based on the outbound time of each fruit order; Temperature control constraints are generated based on the current information of the cold storage, the 3D service window of each fruit order, and the fruit order service list. The temperature control constraints include at least one of the following: the total quantity of fruit orders meets the total inventory constraint, the constraint of each fruit order being completed within the 3D service window, the cold storage temperature constraint, and the overall service quality satisfaction deviation constraint. Based on the temperature control constraints, a dual-objective dynamic optimization model based on comprehensive service quality satisfaction and total energy consumption is constructed for each fruit order's 3D service window and the fruit order service list. Based on the aforementioned dual-objective dynamic optimization model, a temperature control path for the cold storage is generated, enabling the temperature control equipment to dynamically control the temperature of the cold storage according to the temperature control path.

2. The method according to claim 1, characterized in that, The overall service quality satisfaction deviation constraint includes: The deviation between the maturity window for fruit order delivery and the center value of the maturity window required by the fruit order is greater than or equal to zero, and the outbound waiting time for fruit orders in the fruit order service list is greater than or equal to zero.

3. The method according to claim 1, characterized in that, The construction of a dual-objective dynamic optimization model based on comprehensive service quality satisfaction and total energy consumption includes: Construct an evolutionary model of the fruit ripening process; Based on the fruit ripening process evolution model and the cold storage temperature, the ripeness of the fruit is predicted. Based on the deviation between the predicted delivery maturity and the required maturity window center value, and the outbound waiting time of fruit orders, the service quality deviation for each fruit order is generated; based on the cargo quantity of each fruit order and the service quality deviation of each fruit order, the overall service quality deviation of all fruit orders in the fruit order service list is determined. A dynamic energy consumption determination model is constructed based on the temperature difference between inside and outside the cold storage and the surface area of ​​the cold storage to determine the power consumption during the temperature control cycle. Construct a dual-objective dynamic optimization model that minimizes both the overall service quality deviation and total energy consumption.

4. The method according to claim 3, characterized in that, The first temperature control path is a temperature control path defined for a first control cycle; the method further includes: During the process of dynamically controlling the temperature in the first control cycle according to the first temperature control path, the maturity of the fruit and the energy consumption of the cold storage in the second control cycle are predicted based on the fruit ripening process evolution model and the temperature indicated by the first temperature control path. Based on the predicted maturity and cold storage energy consumption of the second control period, the comprehensive service deviation and total cold storage energy consumption of all fruit orders in the fruit order service list within the second control period are determined. If the overall service deviation and total energy consumption in the second control cycle meet the temperature control conditions, then the temperature will continue to be dynamically controlled in the second control cycle according to the first temperature control path. The temperature control conditions indicate that the shipment meets the 3D service window of each fruit order. If the overall service deviation and total energy consumption do not meet the temperature control conditions, the first temperature control path will be updated to the second temperature control path based on the current information of the cold storage, the current maturity of the fruit, the fruit order service list within the second control period, and the dual-objective dynamic optimization model.

5. The method according to claim 4, characterized in that, The method further includes: If the overall service deviation and total energy consumption within the second control period meet the temperature control conditions, then based on the overall service deviation and total energy consumption within the second control period, the shipping time of the fruit orders in the fruit order service list within the second control period is planned.

6. The method according to claim 1, characterized in that, The method further includes: During the process of the temperature control equipment dynamically controlling the temperature according to the temperature control path, if the first fruit order from the dynamic retailer is received, the 3D service window of the first fruit order is determined. Determine whether the current cold storage conditions allow for delivery within the 3D service window of the first fruit order; If the fruit can be supplied within the 3D service window of the first fruit order, then add the first fruit order to the fruit order service list. The temperature is continuously and dynamically controlled according to the temperature control path, and the goods are supplied within the 3D service window of the first fruit order.

7. The method according to claim 6, characterized in that, The method further includes: After adding the first fruit order to the fruit order service list to obtain the updated fruit order service list, the temperature control conditions are updated based on the 3D service window of the first fruit order and the updated fruit order service list. The temperature control conditions also include: the insertion list constraint of the dynamic retailer's fruit order; the insertion list constraint indicates that the maturity window required by the inserted fruit order covers the maturity window of the fruit in the cold storage within the outbound time window of the inserted fruit order under the current temperature control path. Based on the updated temperature control constraints, the updated fruit order service list, and the 3D service window of each fruit order in the updated fruit order service list, the bi-objective dynamic optimization model is updated. The temperature control path is updated based on the updated dual-objective dynamic optimization model.

8. The method according to claim 7, characterized in that, The process of updating the bi-objective dynamic optimization model based on the updated temperature control constraints, the updated fruit order service list, and the 3D service window of each fruit order in the updated fruit order service list includes: According to preset rules, weight coefficients of two objective functions are assigned to fruit orders from fixed retailers and fruit orders from dynamic retailers respectively, and the bi-objective dynamic optimization model is updated.

9. The method according to claim 1, characterized in that, The temperature control path generated based on the dual-objective dynamic optimization model includes: In the process of solving the dual-objective dynamic optimization model, when initializing and generating the initial solution set at the (k+1)th time step, half of the frontier solutions in the frontier solution set obtained at the kth time step are merged with half of the initial solution set generated at the (k+1)th time step to form the initial solution set for the (k+1)th time step, where k is a positive integer.

10. A time-temperature dynamic control device for the cold storage of post-ripening fruits, characterized in that, The device includes: a 3D service window generation module, an order service list generation module, a temperature control constraint generation module, a control model construction module, and a temperature control path generation module; The 3D service window generation module is used to generate a three-dimensional 3D service window for each fruit order based on the requirements of the fruit orders from fixed retailers; the 3D service window for each fruit order includes: a maturity window, a delivery time window, and a quantity window; The order service list generation module is used to generate a fruit order service list according to the outbound time of each fruit order; The temperature control constraint generation module generates temperature control constraints based on the current information of the cold storage, the 3D service window of each fruit order, and the fruit order service list. The temperature control constraints include at least one of the following: the total quantity of fruit orders meets the total inventory constraint, the constraint of each fruit order being completed within the 3D service window, the cold storage temperature constraint, and the overall service quality satisfaction deviation constraint. The control model construction module is used to construct a dual-objective dynamic optimization model based on comprehensive service quality satisfaction and total energy consumption, based on the temperature control constraints, the 3D service window of each fruit order, and the fruit order service list. The temperature control path generation module is used to generate a temperature control path for the cold storage based on the dual-objective dynamic optimization model, so that the temperature control equipment can dynamically control the temperature of the cold storage according to the temperature control path.

Images