A seed cotton grading, warehousing and processing collaborative scheduling method and system based on digital twinning

Abstract

This invention discloses a method and system for collaborative scheduling of graded storage and processing of seed cotton based on digital twins. The method uses single-bale quality risk to drive closed-loop control, establishes a digital twin benchmark model for storage, collects information on cotton bale attributes, environment, storage location, and handling status, and binds these information with unique identifiers to construct a twin object. The system discretizes the layout into candidate storage locations and layers, uses a dynamic storage location allocation model to comprehensively evaluate factors such as handling distance, mixed storage, and quality risk, generates allocation schemes under multiple constraints, and performs feasibility decoding and rolling freeze re-optimization through a hybrid genetic algorithm. Virtual pre-execution is performed, and after operation, three-source verification and correction are conducted using RFID, vision, and positioning. Upon outbound processing, drying, cleaning, and feeding parameters are recommended downstream based on a homogeneous batch quality profile, and traceability is established. This invention can improve the timeliness of response to high-risk seed cotton, the consistency and stability of storage locations, the homogeneity of outbound processing, and the matching of processing parameters.

Description

Technical Field

[0001] This invention relates to the fields of cotton warehousing and logistics, digital twins, unmanned handling scheduling and processing control, specifically to a method and system for graded storage and collaborative scheduling of seed cotton based on digital twins, and particularly to dynamic storage location allocation, rolling freeze optimization and collaborative processing control for seed cotton bales. Background Technology

[0002] Seed cotton is characterized by its dispersed sources, large differences in moisture regain, large differences in impurity content, significant fluctuations in quality grade, and dynamic changes in arrival time during the acquisition, temporary storage, stacking, and processing processes. Ginning mills typically need to complete the identification, unloading, grading, stacking, warehousing, and feeding of large quantities of seed cotton bales within a short period. Relying solely on manual experience for storage location arrangements can easily lead to problems such as long waiting times for high-moisture bales, mixing of bales of different quality grades or moisture regain, conflicting retrieval sequences between upper and lower layers within the same storage location, inconsistencies between storage location records and actual stacking, and significant fluctuations in processing batches.

[0003] While existing digital warehousing systems can achieve partial automation using RFID, barcodes, storage location coding, or unmanned forklifts, their focus is primarily on inventory registration, route execution, or storage location coordinate management. They lack a virtual-physical linkage mechanism addressing issues such as seed cotton moisture regain, mold risk, grading and zoning, layer obstruction, dynamic entry, and the homogeneity of processing batches. Especially in scenarios with limited outdoor storage space and continuous arrival of cotton bales, traditional static storage rules struggle to simultaneously accommodate requirements such as close-range handling of high-risk cotton bales, physical constraints of three-layer stacking, strict prohibition of mixed storage, and minimal movement of existing cotton bales.

[0004] Therefore, a method is needed to uniformly map, virtually simulate, dynamically allocate storage locations, and perform closed-loop verification of the physical storage status, quality risk status, storage location status, handling execution status, and downstream processing status of seed cotton, so as to achieve coordinated optimization of the entire process of seed cotton from warehousing, stacking, transfer, and delivery to processing and feeding. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for coordinated scheduling of seed cotton grading storage and processing based on digital twins, which can solve the problems of insufficient single-bale-level state mapping, untimely response to high-risk cotton bales, lack of hierarchical constraints and freezing mechanisms in dynamic storage location allocation, lack of feasibility simulation before unmanned forklifts are executed, lack of consistency verification after the operation is completed, and insufficient linkage between storage outbound and downstream processing parameters in existing seed cotton storage.

[0006] This invention proposes a digital twin-based method for the coordinated scheduling of graded storage and processing of seed cotton, comprising: establishing a digital twin benchmark model of the physical space of seed cotton storage; collecting multi-source information and binding it with the unique identifier of the cotton bale; constructing twin objects of cotton bales, storage locations, handling equipment, and processing interfaces; calculating a quality risk index based on moisture regain, waiting time, storage area environmental parameters, stacking levels, shading status, and downstream processing requirements; discretizing the factory layout map or 3D grid into candidate storage locations and layers; constructing a dynamic storage location allocation model based on cotton bales-storage locations-layers, virtually rehearsing candidate storage locations, handling paths, and stacking actions, and generating a dynamic storage location allocation scheme under constraints such as unique allocation, storage capacity, layer continuity, graded and classified allocation, high-risk proximity, strict prohibition of mixed storage, and frozen cotton bales; performing consistency verification and updating or correcting the digital twin status after the operation is completed; forming homogeneous processing batches and dynamic outbound sorting schemes during the outbound stage, and synchronizing the batch quality profile to the downstream processing control system.

[0007] This invention also provides a seed cotton grading, storage, and processing collaborative scheduling system based on digital twins, comprising a physical layer, a multi-source sensing layer, a digital twin layer, a simulation optimization layer, an execution control layer, and a processing collaboration layer. The physical layer corresponds to the back-end stacking area, the front-end temporary storage area, the processing area, the cotton feeding port, unmanned forklifts, and warehouse passageways; the multi-source sensing layer collects RFID, 3D vision, moisture regain, temperature and humidity, and forklift positioning data; the digital twin layer maintains various twin objects; the simulation optimization layer performs layout digitization, candidate storage location screening, hard constraint verification, comprehensive cost evaluation, dynamic storage location allocation using a hybrid genetic algorithm, rolling freeze re-optimization, execution path pre-simulation, and stacking action pre-simulation; the execution control layer generates and executes storage operation instructions; and the processing collaboration layer transmits the quality profile of the outgoing batch to the processing control system and establishes a traceability relationship for the finished cotton products.

[0008] According to the specific embodiments provided in this application, the following technical effects are disclosed: By modeling cotton bale twin objects, storage location twin objects, handling equipment twin objects, and processing interface twin objects, real-time mapping of the spatial location, stacking level, quality risk, and processing requirements of a single seed cotton bale is achieved, reducing the risk of inconsistency between digital inventory and actual stacking; by using a dynamic storage location allocation model of cotton bale-storage location-layer, handling distance, expected shading, mixed storage, high moisture regain storage risk, and dynamic movement disturbance are uniformly incorporated into the storage location optimization process, improving the executability and stability of the storage location results; By freezing existing cotton bales and only partially optimizing newly added or affected bales, the total amount of unloading and ineffective handling is reduced; through a virtual pre-simulation mechanism, unreachable storage locations, incorrect stacking, path conflicts, and upper-level obstruction are eliminated before physical handling, improving the success rate of unmanned forklifts; through a consistency verification closed loop, RFID, 3D vision, and handling equipment positioning feedback are used to correct the digital twin of seed cotton storage; and by linking homogenized processing batches and processing parameters, frequent switching of downstream processing technology is reduced, energy consumption and fiber damage are lowered, and quality traceability from farmers' plots to finished cotton products is achieved. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart of the collaborative scheduling method for seed cotton storage and processing based on digital twins according to the present invention.

[0011] Figure 2 This is a diagram of the digital twin system architecture for seed cotton storage of the present invention.

[0012] Figure 3 This is a diagram showing the relationship between digital twin objects in the repository area of ​​this invention.

[0013] Figure 4 This is a flowchart of the dynamic storage location allocation and consistency verification process of the present invention.

[0014] Figure 5 This is a flowchart of the dynamic storage allocation model and DSA-HGA solution of the present invention.

[0015] Figure 6 This is a flowchart illustrating the dynamic outbound sorting and processing parameter linkage of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0017] In one embodiment, refer to Figure 2 At the acquisition entrance, RFID readers, barcode scanners, and license plate recognition devices are installed to read the unique identifier of the seed cotton bales, vehicle information, and entry time. Weighing equipment, data acquisition terminals, moisture regain detection devices, and quality grading input terminals are installed on the weighbridge to collect vehicle gross weight, tare weight, cotton bale weight, weighing time, moisture regain, and quality grade. 3D vision equipment is installed in the unloading area to collect the outer boundary of the cotton bales and unloading coordinates. Temperature and humidity sensors, RFID readers for storage locations, 3D vision equipment, and warehouse positioning base stations are installed in the front temporary storage area and the rear stacking area to collect environmental parameters, storage space occupancy, stacking height, and actual placement coordinates. RFID readers, feeding status detection devices, and processing interface data terminals are installed in the waiting-to-process area and the cotton feeding port to confirm the arrival of cotton bales, the status of the cotton feeding port, and downstream production capacity. The unmanned forklifts are equipped with positioning units, load detection units, and task execution feedback units to transmit current position, handling trajectory, pick-up and place status, and termination position.

[0018] Reference Figure 3 The cotton bale twin, storage location twin, handling equipment twin, and processing interface twin establish data relationships through the unique identifier of the cotton bale, storage location coordinates, task number, and processing batch number. An occupancy and location mapping relationship is formed between the cotton bale twin and the storage location twin. The handling equipment twin transmits retrieval and placement results back to the cotton bale twin and the storage location twin, while the processing interface twin provides feedback on processing requirements and batch status to the cotton bale twin.

[0019] The system assigns or retrieves a unique identifier for each cotton bale and binds this identifier to the farmer's identity, plot number, vehicle license plate number, weighing information, moisture regain, quality grade, storage time, and initial operating status. The bound data forms a cotton bale twin object, which records not only the inventory quantity but also the current storage location coordinates, stacking level, maximum allowable waiting time, shading status, quality risk index, and processing requirements.

[0020] A baseline model of the storage area is established based on a three-dimensional mesh. The coordinates of the storage location can be represented as (x, y, z), where x and y represent the ground mesh position and z represents the stacking level. Each storage location twin object is configured with capacity, allowed number of stacking layers, current occupied quantity, attribute lock label, accessibility status, distance to the processing area or feed inlet, transfer cost, functional zoning, and environmental parameters. The attribute lock label can be used to indicate that the storage location has been locked for a specific quality grade, moisture regain range, impurity content range, farmer batch, or batch to be processed.

[0021] The material handling equipment twin records the unmanned forklift's ID, current location, accessible area, load status, executable tasks, path, task priority, and fault status. The processing interface twin records the downstream processing control system's capacity, current drying parameters, cleaning parameters, feeding parameters, batches awaiting processing, and cotton feed status. These objects are linked through unique cotton bale identifiers, storage location coordinates, and task numbers, forming a real-time updatable digital twin of seed cotton storage.

[0022] In one embodiment, the system updates the quality risk index when events such as cotton bale warehousing, transfer, environmental sampling, outbound sorting, and processing feed occur. As one possible implementation, for the i-th seed cotton bale, its quality risk index... It can be calculated using the following formula: In the formula, This indicates the risk factor for moisture regain. This indicates the waiting time risk factor. Indicates environmental risk factors in the reservoir area. This indicates stacking levels or occlusion risk factors. Indicating the urgency factor of downstream processing demand, These are the weighting coefficients.

[0023] Moisture regain risk factor It can be obtained by normalizing the deviation between the actual moisture regain rate and the safe moisture regain rate range; waiting time risk factor. It can be obtained by normalizing the ratio of actual waiting time to maximum allowable waiting time; reservoir area environmental risk factor. This can be determined based on the degree of deviation between the temperature and humidity of the storage area and the preset suitable range; stacking levels or shading risk factors. This can be determined based on the layer of the cotton bale, the number of upper layers obstructing it, and the accessibility of the goods; the urgency factor of downstream processing demand. It can be obtained based on downstream capacity, the number of batches to be processed, and the planned material input time. Each factor can be normalized to the range of 0 to 1, and the weighting coefficients can be adjusted according to the season, the process requirements of the cotton ginning plant, or management strategies.

[0024] when Greater than the preset risk threshold When this occurs, the system marks the corresponding seed cotton bale as a high-risk bale and writes this status into the bale's twin object. High-risk bales are preferentially allocated to storage locations near the processing area, feeding port, with good accessibility, and lower expected shading costs during dynamic storage allocation; in dynamic outbound sorting, they are given priority for the earliest feasible homogeneous processing batch. The above formula is one embodiment; in practical applications, segmented scoring, rule-based scoring, or other risk calculation methods that reflect moisture regain, waiting time, environmental conditions, shading status, and processing requirements can also be used.

[0025] In one embodiment, refer to Figure 4 and Figure 5 Once the system identifies seed cotton bales to be stored or transferred, it first generates a set of candidate storage locations based on the bale's quality grade, moisture regain range, impurity content range, location zone, target operation stage, storage location occupancy status, and high-risk scheduling constraints. The system then removes storage locations that are insufficient in capacity, have mismatched attribute lock tags, exceed the allowed stacking layer, have no lower support, have non-removable obstructions on the upper layer, are inaccessible to unmanned forklifts, or do not meet the distance constraints for high-risk cotton bales.

[0026] Candidate storage locations can be automatically generated from plant layout diagrams, CAD drawings, or 3D meshes of the storage area. The system reads layout dimensions, storage area boundaries, access nodes, feed inlets or main operating points, and boundaries of each functional zone, discretizing the storage yard into multiple numberable storage locations. For each storage location, it generates a storage location number, storage location index, ground coordinates, candidate layer, capacity, functional zone, allowable grade or moisture regain range, inlet point, access distance, and distance thresholds related to high-risk disposal. Access distances can be determined based on the shortest path distance, Manhattan distance, or a weighted average of both from the storage area access map.

[0027] The dynamic reservoir allocation model takes cotton bale-reservoir-stratum as the core decision-making objects. For cotton bale i, reservoir u, and stratum l, allocation variables are set. The value is 1 when cotton bale i is assigned to the l-th layer of storage location u, and 0 otherwise. The model outputs the target storage location, target layer, accessibility status, expected occlusion risk, mixed placement risk, and high-risk cotton bale storage risk for each cotton bale, and writes the above outputs into the cotton bale twin object and the storage location twin object.

[0028] In one possible implementation, the overall cost F can be expressed as: .in, This indicates the cost of transporting goods or passing through aisles. This indicates the estimated cost of obstruction incurred when a lower layer of cotton bales is expected to be shipped out first but is obstructed by an upper layer of cotton bales. This indicates a penalty for mixed storage due to mismatches in different grades, moisture regain categories, impurity content ranges, or functional zones. This indicates that high moisture regain or high-risk cotton bales pose storage risks due to increased distance, environment, or stacking time. This represents the cost of movement disturbance caused by changes in the storage location or layer location of existing cotton bales relative to the previous cycle during rolling freeze re-optimization. The above formula is used to illustrate one method of calculating the overall cost and does not limit the specific weight, dimensions, or normalization method of each cost item.

[0029] The dynamic storage allocation model must satisfy at least the following constraints: each cotton bale participating in this round of allocation must be assigned to one storage location and layer; at most one cotton bale can be stored in the same storage location and layer, and the total number of occupied bales cannot exceed the capacity; stacks must be formed continuously from bottom to top, and there cannot be a situation where there are cotton bales on the upper layer and the lower layer is empty; cotton bales can only enter storage locations that match their grade, moisture regain category, impurity content range, or functional zone; high-risk cotton bales are assigned within a preset maximum distance range or are subject to high-weight risk penalties; combinations of cotton bales that are strictly prohibited from being mixed must not be located in the same storage location; existing cotton bales frozen in this round retain their original storage location and layer.

[0030] In one implementation, the system employs a dynamic storage location allocation hybrid genetic algorithm (DSA-HGA) to solve the aforementioned storage location allocation problem. The algorithm uses a cotton bale allocation priority sequence as chromosomes, generating an initial population based on high-risk markers, risk weights, expected outbound or latest processing time, quality grade, and random perturbations. Feasibility decoding is performed on each chromosome to obtain specific storage location and stratum allocation schemes. Fitness is then evaluated based on comprehensive costs, and the population is iteratively updated using elite retention, sequential crossover, two-point exchange, interval reversal, single-point insertion, and local search.

[0031] The feasibility decoder pre-generates candidate storage locations according to grade zone, functional zone, and channel distance during the decoding process. For high-risk cotton bales, it prioritizes scanning high-risk proximity zones, followed by regular and isolation zones; for ordinary cotton bales, it prioritizes scanning storage locations that match their quality grade or attribute locking labels. During the scanning process, it skips storage locations that are full, do not meet strict mixing requirements, do not meet high-risk distance requirements, do not match attribute locking, or do not meet layer continuity requirements, and selects target storage locations based on incremental distance, incremental risk, incremental mixing, incremental occlusion, and zone penalty.

[0032] In dynamic rolling scenarios, the system writes cotton bales that have been stably stored in the previous cycle and are unaffected by current events into the decoder as frozen cotton bales, maintaining their original storage location and layer position. Newly arrived cotton bales, cotton bales updated under high-risk conditions, cotton bales affected by occlusion, or cotton bales affected by manual adjustments are locally re-optimized. This reduces the number of times existing cotton bales are moved while maintaining constraints, avoiding a complete re-layout every time a new cotton bale enters the field.

[0033] Once the target storage location is determined, the system generates a forklift execution task. The task includes the pickup point, target storage location, target level, recommended route, task priority, estimated execution time, and exception handling strategy. The exception handling strategy may include replanning the route if it is occupied, reselecting a candidate storage location if the target storage location is temporarily occupied, and triggering manual review or re-reading the unique identifier of the cotton bale if pickup fails.

[0034] In one embodiment, refer to Figure 4 After the unmanned forklift completes its inbound, transfer, or outbound operations, the system acquires RFID reading results, 3D vision recognition results, and handling equipment positioning results. RFID reading results are used to confirm the ID of the cotton bale actually handled or placed; 3D vision recognition results are used to confirm the cotton bale boundaries, stacking height, and actual placement coordinates; and handling equipment positioning results are used to confirm the handling trajectory, stopping position, and termination position.

[0035] The system compares the actual cotton bale ID, actual storage location coordinates, actual stacking level, actual boundary position, and forklift termination position with the expected cotton bale ID, expected storage location, and expected layer recorded in the seed cotton storage digital twin. If the actual state matches the expected state, the system updates the current position, stacking level, operation status, and timestamp of the cotton bale twin object, and simultaneously updates the occupied quantity, occupied cotton bale list, reachability status, and environmental association information of the storage location twin object.

[0036] If the actual state differs from the expected state, the system will set the corresponding storage location as an abnormal storage location, pending verification, or locked, and trigger relocation, manual review, corrective handling, or rescheduling based on the type of deviation. For example, when the cotton bale ID read by RFID does not match the task target, the system will suspend subsequent outbound tasks for that storage location and require review; when the deviation between the 3D visual recognition coordinates and the target coordinates exceeds the allowable range, the system will generate a corrective handling task; when the forklift's termination position does not match the target storage location and the path is blocked, the system will re-perform path rehearsal and storage location allocation.

[0037] In one embodiment, refer to Figure 6 During the outbound phase, the system reads the status of the seed cotton storage digital twin and retrieves cotton bales that can be shipped. Cotton bales ready for shipment must meet certain conditions, including completion of inbound verification, availability of storage location, absence of anomaly markers, absence of being locked by other tasks, and compliance with the current processing plan. The system generates candidate homogeneous processing batches based on quality grade, moisture regain range, impurity content range, process requirements, and downstream capacity.

[0038] The system uses a rolling time-domain approach to generate dynamic outbound sorting schemes. Within each rolling time-domain, the system comprehensively considers factors such as the number of process switching times, quality risk index, maximum allowable waiting time overdue penalty, pickup accessibility, and handling distance. For high-moisture or high-risk seed cotton bales, if their waiting time is close to the maximum allowable waiting time, the system prioritizes inserting them into the earliest feasible homogeneous processing batch; if they are currently obscured by upper cotton bales, the system simultaneously generates necessary transfer or inventory dumping tasks and feeds back the estimated inventory dumping cost to the storage location allocation optimization module.

[0039] Before the cotton bales are delivered to the cotton feeder, the system reads the unique identifier of each bale again and retrieves the moisture regain, quality grade, quality risk index, homogeneous processing batch, and recommended process parameters from the bale's twin. This information is synchronized to the downstream processing control system, which adjusts the drying intensity, cleaning parameters, feeding speed, or other relevant process parameters accordingly. After processing, the system links the finished cotton product identifier with the original bale's twin, farmer identity, plot number, quality grade, and processing batch, thus forming a quality traceability chain from warehousing to the finished cotton product.

[0040] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0041] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for collaborative scheduling of seed cotton grading, storage, and processing based on digital twins, characterized in that, Includes the following steps: S100, Establish a digital twin benchmark model of the physical space of seed cotton storage. The digital twin benchmark model includes the back-end stacking area, the front-end temporary storage area, the waiting-to-process area, the cotton feeding port, the warehouse passage, the gridded storage location, the stacking level, the operating boundary of the handling equipment, and the downstream processing interface. S200 collects traceability information, physical attribute information, real-time operation information, environmental information, storage location status information, and handling equipment status information of seed cotton bales, and binds the information with the unique identifier of the cotton bale to form a single bale-level status record; the physical attribute information includes at least the quality grade, moisture regain, impurity content, arrival time, and maximum allowable waiting time; S300, based on the unique identifier of the cotton bale, construct a cotton bale twin object, a storage location twin object, a handling equipment twin object, and a processing interface twin object respectively to form a seed cotton storage digital twin, and maintain the association relationship between cotton bales, storage locations, layers, handling tasks, and processing batches in the seed cotton storage digital twin; S400 calculates the quality risk index based on the degree of deviation of the cotton bale moisture regain rate from the safe moisture regain rate range, the ratio of waiting time to the maximum allowable waiting time, the degree of deviation of the temperature and humidity of the warehouse area from the suitable range, the number of stacking levels or upper layer obstructions and the accessibility of picking up goods, and the downstream planned feeding time. When the quality risk index exceeds the preset risk threshold, the corresponding seed cotton bales are marked as high-risk bales, and a scheduling constraint is generated to keep the high-risk bales available or prioritized for transfer and delivery within a preset time window. S500 generates candidate storage locations and candidate layers based on the factory layout diagram or the three-dimensional grid of the warehouse area. It combines the quality grade, moisture regain range, impurity content range, location zone, target operation stage, storage location occupancy status, and the scheduling constraints to form a set of candidate storage locations. It then performs a virtual pre-simulation of the candidate storage locations, candidate layers, handling paths, and stacking actions in the seed cotton warehouse digital twin. The virtual pre-simulation verifies at least path conflicts, storage capacity, attribute locking, layer continuity, stacking support, upper layer obstruction, forklift accessibility, and the accessibility of high-risk cotton bales within a preset time window. Only candidate solutions that pass the virtual pre-simulation are retained. S600: Calculate the comprehensive cost of candidate schemes obtained through virtual pre-simulation. The comprehensive cost includes at least the handling distance cost, the expected occlusion cost, the penalty for mixed placement, the quality risk cost, and the dynamic movement disturbance cost. Under the constraints of unique allocation of cotton bales, storage capacity, layer continuity, graded and classified allocation, high-risk proximity, strict prohibition of mixed placement, and freezing of existing cotton bales, select candidate schemes whose comprehensive cost satisfies the preset optimization rules to generate a dynamic storage allocation scheme including target storage location and target layer. Issue warehousing operation instructions to the handling equipment, including the pickup point, target storage location, target layer, recommended path, task priority, expected execution time, and exception handling strategy. S700: After the operation is completed, the RFID reading results, 3D visual recognition coordinates, stacking height, and handling equipment positioning results are compared with the expected cotton bale ID, expected storage location, expected layer, and expected termination location in the seed cotton storage digital twin. When the verification deviation exceeds the corresponding allowable threshold, the abnormal storage location is set to a pending verification or locked state, and relocation, manual review, deviation correction handling, or rescheduling is triggered. Based on the updated seed cotton storage digital twin, quality risk index, accessibility status, and downstream processing capacity, a homogeneous processing batch and dynamic outbound sorting scheme are generated through a rolling time domain method. The seed cotton bales are read a second time before being sent to the cotton feeding port, and the batch quality profile and recommended process parameters are synchronized to the downstream processing control system to adjust the drying, cleaning, or feeding process parameters in a coordinated manner.

2. The method according to claim 1, characterized in that, The cotton bale twin objects include a unique cotton bale identifier, farmer identity information, plot number, vehicle license plate information, moisture regain, impurity content, quality grade, warehousing time, current storage location coordinates, stacking level, location zone, operating status, maximum allowable waiting time, estimated outbound or latest processing time, shading status, accessibility status, and quality risk index. The storage location twin objects include storage location coordinates, candidate layers, storage location capacity, current occupied quantity, attribute lock tag, accessibility status, distance from the processing area or feed inlet, transfer cost, allowable stacking layers, functional zones, and storage location environmental parameters. The handling equipment twin objects include equipment number, current location, load status, executable tasks, path, and fault status. The processing interface twin objects include downstream capacity, current process parameters, batch to be processed, and feed inlet status.

3. The method according to claim 1, characterized in that, The candidate storage locations and candidate storage layers are generated as follows: read the factory layout map, CAD drawing, or 3D grid of the storage area; discretize the back-end stacking area, front-end temporary storage area, emergency treatment area, isolation area, and regular storage area into numberable storage locations; record the storage location number, coordinates, entry point, functional zoning, allowable grade or moisture regain range, layer capacity, and passage distance to the flower feed inlet or main operation point for each storage location; and mark the high-risk close-range area, regular area, and isolation area according to distance zoning, functional zoning, or risk disposal requirements.

4. The method according to claim 1, characterized in that, The quality risk index is determined by normalized weighting factors including moisture regain risk factor, waiting time risk factor, warehouse environment risk factor, stacking level or shading risk factor, and urgency factor of downstream processing demand. When the quality risk index exceeds the preset risk threshold, the system marks the corresponding seed cotton bales as high-risk bales and increases their priority in storage allocation, transfer, or warehousing.

5. The method according to claim 1, characterized in that, The virtual pre-simulation includes simulating the transportation path of seed cotton bales from their current location to candidate storage locations, processing areas, or feeding ports in the seed cotton storage digital twin, and verifying path conflicts, storage capacity, attribute matching, layer continuity, stacking support, upper layer obstruction, forklift accessibility, and time window constraints for high-risk cotton bales; only when the virtual pre-simulation passes will the corresponding physical execution instructions be generated.

6. The method according to claim 1, characterized in that, The dynamic storage location allocation scheme satisfies at least one of the following constraints: each cotton bale participating in this round of allocation is assigned to a unique storage location and layer; at most one cotton bale is stored in the same storage location and layer, and the total number of occupied storage locations does not exceed the capacity; stacking layers are formed continuously from bottom to top; cotton bales are only allocated to storage locations that match their grade, moisture regain category, impurity content range, or functional zone; high-risk cotton bales are allocated to a preset maximum distance range or are subject to high-weight risk penalties; combinations of cotton bales that are strictly prohibited from being mixed must not be located in the same storage location; existing frozen cotton bales retain their original storage location and layer.

7. The method according to claim 1, characterized in that, The comprehensive cost is determined at least based on the handling distance, expected obstruction, mixed storage, quality risk, and dynamic movement disturbance. Expected obstruction is determined based on the time difference when the expected outbound or latest processing time of the lower layer cotton bales in the same storage location is earlier than that of the upper layer cotton bales. Mixed storage is determined based on quality grade, moisture regain category, impurity content difference, or functional zoning overflow. Dynamic movement disturbance is determined based on the number or weight of existing cotton bales that have changed storage location or layer relative to the previous cycle.

8. The method according to claim 1, characterized in that, The dynamic storage allocation scheme is generated using a rolling freeze re-optimization mechanism that combines event triggering and periodic triggering. Triggering events include new cotton bales entering the site, high-risk cotton bales arriving, storage space release, manual adjustment, cotton bale attribute update, or reaching a fixed rolling cycle. After each trigger, the current stockpile status is read to determine the newly added cotton bales, movable cotton bales, and frozen cotton bales. Local re-optimization is performed only on the newly added cotton bales and the affected cotton bales, and the storage location, layer, accessibility status, expected shading risk, and storage risk of high-risk cotton bales are output for each cotton bale.

9. The method according to claim 1, characterized in that, The dynamic outbound sorting scheme is obtained through rolling time-domain optimization. The objectives within the rolling time-domain include at least reducing the number of process changeovers, reducing quality risks, reducing the maximum allowable waiting time overdue, improving pickup accessibility, and reducing handling distances. High-moisture or high-risk seed cotton bales are prioritized for inclusion in the earliest feasible homogeneous processing batch.

10. A seed cotton grading, storage, and processing collaborative scheduling system based on digital twins, characterized in that, The system for implementing the method according to any one of claims 1 to 9, the system comprising: The multi-source sensing module is used to collect cotton bale traceability information, physical attribute information, warehouse environment information, storage location status information, and handling equipment status information; The digital twin modeling module is used to build and maintain digital twin objects for cotton bales, storage locations, handling equipment, processing interfaces, and seed cotton storage. The risk assessment module is used to calculate the quality risk index and identify high-risk cotton bales; The simulation optimization module is used to perform layout digitization, candidate storage location screening, hard constraint verification, comprehensive cost evaluation, dynamic storage location allocation using a hybrid genetic algorithm, rolling freeze re-optimization, storage location allocation result output, and pre-simulation of candidate storage locations, transport paths, and stacking actions. The warehouse management module is used to generate instructions for inbound, transfer, sorting, and outbound operations. The forklift dispatch module is used to control the handling equipment to perform operations and to transmit positioning and execution status back. The consistency verification module is used to verify and correct the twin state based on RFID, 3D vision, and positioning feedback from handling equipment. The processing collaboration module is used to synchronize homogenized processing batches, batch quality profiles, and recommended process parameters to the downstream processing control system, and to establish traceability relationships for finished cotton products.

Images