Stacker multi-machine collaborative scheduling method and device based on dynamic carbon footprint accounting, computer equipment and storage medium

Abstract

The application discloses a stacking machine multi-machine collaborative scheduling method and device based on dynamic carbon footprint accounting, computer equipment and a storage medium. The method comprises the following steps: constructing an energy consumption calculation model of a single handling operation of a stacking machine to obtain total power consumption generated by executing a single handling task; accounting for carbon emissions of the single handling task; constructing a multi-lane multi-stacking machine collaborative scheduling optimization model with the optimization goal of minimizing the total carbon emission cost of the system; solving the optimization model to generate a carbon optimal scheduling scheme; simulating and verifying the carbon optimal scheduling scheme; and after the simulation verification is passed, converting the carbon optimal scheduling scheme into a control instruction and issuing it to the stacking machine for execution. Through the implementation of the method of the application, model self-calibration and algorithm continuous evolution can be realized, so that the global carbon efficiency and long-term low-carbon operation ability of the automated stereoscopic warehouse are improved while the operation efficiency is ensured.

Description

Technical Field

[0001] This invention relates to the field of multi-machine collaborative scheduling technology for stacker cranes, and more specifically to a method, apparatus, computer equipment, and storage medium for multi-machine collaborative scheduling of stacker cranes based on dynamic carbon footprint accounting. Background Technology

[0002] In automated storage and retrieval systems (AS / RS), stacker cranes, as the core equipment for performing goods storage and retrieval operations, are the main energy-consuming unit of the system. Their operation and scheduling strategies directly determine the energy consumption level and total carbon emissions of the entire warehousing system.

[0003] Currently, industry research on stacker crane scheduling optimization primarily focuses on improving operational efficiency, such as minimizing task completion time, maximizing system throughput, or optimizing operational paths to enhance warehouse operational efficiency. However, these traditional methods generally lack direct and refined consideration of carbon emissions, a key environmental indicator. In recent years, although some studies have begun to incorporate energy consumption as one of the optimization objectives, significant technical shortcomings remain in carbon emission accounting and scheduling decisions.

[0004] First, in terms of carbon emission accounting methods, existing technologies typically use a fixed, annually averaged grid emission factor to calculate carbon emissions from electricity consumption. This method fails to reflect the true fluctuations in carbon intensity at the power generation end over time (e.g., on a minute or hourly scale), especially given the large-scale grid integration of renewable energy sources (such as wind and solar power), where real-time carbon intensity variations in the grid are significant. Using a fixed factor for accounting not only leads to severely distorted carbon emission measurement results but also prevents dispatch systems from detecting and responding to low-carbon periods with abundant "green electricity," thus missing crucial opportunities to reduce carbon footprints using clean energy.

[0005] Secondly, in terms of energy consumption model construction, the energy consumption models relied upon by existing scheduling algorithms are often oversimplified. They typically treat the stacker crane as a complete black box, failing to fully consider the independent physical characteristics of its three core actions: horizontal movement (X-axis), vertical lifting (Y-axis), and fork extension / retraction (Z-axis). More importantly, these models ignore the significant impact of dynamic factors such as load changes and speed change curves during acceleration and deceleration on energy consumption, resulting in insufficient accuracy in calculating the carbon cost of a single handling operation, making it difficult to support refined low-carbon scheduling decisions.

[0006] Finally, at the system coordination level, existing scheduling optimization is often disconnected from upstream storage location allocation strategies and the park's energy management system (such as distributed energy sources like photovoltaics and energy storage). Storage location allocation does not consider the correlation between cargo turnover rate and carbon efficiency, and scheduling decisions fail to link with the real-time status of energy supply (such as green electricity output). This separation of "energy-equipment-task" hinders the formation of an integrated, globally optimal carbon reduction decision-making loop, limiting further improvements in the overall carbon efficiency of the warehousing system.

[0007] In summary, there is an urgent need for a new stacker crane scheduling technology that can overcome the above-mentioned shortcomings. This technology should be able to achieve high-precision carbon cost accounting and, on this basis, perform multi-objective collaborative optimization, thereby effectively reducing the system's direct and indirect carbon emissions while ensuring the efficiency of warehousing operations. Summary of the Invention

[0008] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method, device, computer equipment and storage medium for multi-machine collaborative scheduling of stacker cranes based on dynamic carbon footprint accounting.

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] A multi-machine collaborative scheduling method for stacker cranes based on dynamic carbon footprint accounting includes:

[0011] Construct an energy consumption calculation model for a single handling operation of a stacker crane to obtain the total electrical energy consumption generated by performing a single handling task;

[0012] Establish a minute-level dynamic grid carbon emission factor access mechanism, and calculate the carbon emission of a single transportation task based on the total power consumption and the dynamic grid carbon emission factor corresponding to the task plan execution time.

[0013] The total carbon emission cost of the system is the sum of the carbon emissions of all the handling tasks to be performed. To minimize the total carbon emission cost of the system, a multi-lane multi-stacking crane collaborative scheduling optimization model is constructed.

[0014] Design a two-layer collaborative scheduling solution strategy of storage location and scheduling, and solve the optimization model to generate a carbon-optimal scheduling scheme;

[0015] The carbon-optimal scheduling scheme was verified through simulation.

[0016] After the simulation verification is passed, the carbon-optimal scheduling scheme is converted into control commands and sent to the stacker crane for execution. During the execution process, the actual energy consumption data of the equipment is collected in real time, and the actual carbon emissions are calculated based on the dynamic power grid carbon emission factor. Finally, the actual carbon emissions are fed back to the energy consumption calculation model and scheduling solution strategy for dynamic calibration of the model and continuous optimization of the algorithm.

[0017] This invention also provides a multi-machine collaborative scheduling device for stacker cranes based on dynamic carbon footprint accounting, comprising:

[0018] A data acquisition unit is constructed to build an energy consumption calculation model for a single handling operation of a stacker crane, so as to obtain the total electrical energy consumption generated by performing a single handling task;

[0019] A calculation unit is established to establish a minute-level dynamic grid carbon emission factor access mechanism, and to calculate the carbon emission of a single transportation task based on the total power consumption and the dynamic grid carbon emission factor corresponding to the task plan execution time.

[0020] A construction unit is used to sum the carbon emissions of all the handling tasks to be performed as the total carbon emission cost of the system, and to construct a multi-lane multi-stacking crane collaborative scheduling optimization model with the goal of minimizing the total carbon emission cost of the system.

[0021] The design solver unit is used to design a two-layer collaborative scheduling solution strategy for storage location and scheduling, and to solve the optimization model to generate a carbon-optimal scheduling scheme.

[0022] The simulation unit is used to simulate and verify the carbon optimal scheduling scheme.

[0023] The execution feedback unit is used to convert the carbon-optimal scheduling scheme into control commands and send them to the stacker crane for execution after the simulation verification is passed. During the execution process, the actual energy consumption data of the equipment is collected in real time, and the actual carbon emissions are calculated based on the dynamic power grid carbon emission factor. Finally, the actual carbon emissions are fed back to the energy consumption calculation model and scheduling solution strategy for dynamic calibration of the model and continuous optimization of the algorithm.

[0024] The present invention also provides a computer device, the computer device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the above-described method.

[0025] The present invention also provides a storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0026] The advantages of this invention compared to existing technologies are as follows: By establishing a minute-level dynamic power grid carbon emission factor access mechanism and combining it with a precisely constructed energy consumption calculation model for a single stacker crane handling operation, the actual carbon emissions of each handling task at the planned execution time can be accurately calculated. This abandons the traditional coarse calculation method of fixed annual average factors, making the carbon cost measurement results highly synchronized with the actual carbon intensity fluctuations at the power generation end, providing a real and reliable data foundation for subsequent low-carbon scheduling decisions. Furthermore, since the scheduling optimization model directly minimizes the sum of carbon emissions of all tasks to be executed (i.e., the total carbon emission cost of the system), and the calculation of this cost depends on the dynamic carbon factor at the task execution time, the resulting scheduling scheme intelligently prioritizes high-energy-consuming tasks to be executed in time windows with lower power grid carbon intensity (i.e., a higher proportion of "green electricity"). This enables the warehousing system to proactively track and utilize clean energy, reducing indirect carbon emissions caused by electricity consumption. Furthermore, by designing a two-layer collaborative solution strategy of storage location and scheduling, the upstream storage location allocation (affecting handling distance and frequency) and the downstream multi-machine collaborative scheduling (determining task execution timing and path) are jointly optimized. Simultaneously, the entire optimization process uses dynamic carbon cost as the core link, tightly coupling external energy supply (dynamic power grid) and internal equipment operation (stacking crane energy consumption). This integrated decision-making model breaks down the barriers between different links in traditional methods, achieving optimal carbon emission control at the system-wide level. Moreover, this technical solution not only completes optimization before scheduling but also collects actual equipment energy consumption data in real time during the solution execution process, calculating the actual carbon emissions accordingly. This actual data is fed back into the energy consumption calculation model and scheduling solution strategy for dynamic calibration of model parameters and continuous optimization of the algorithm. This closed-loop mechanism ensures that the system can adapt to real-world factors such as equipment aging and changes in operating conditions, maintaining high-precision carbon accounting capabilities and efficient low-carbon scheduling performance over the long term.

[0027] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a schematic diagram illustrating an application scenario of the stacker crane multi-machine collaborative scheduling method based on dynamic carbon footprint accounting provided in an embodiment of the present invention.

[0030] Figure 2This is a flowchart illustrating the multi-machine collaborative scheduling method for stacker cranes based on dynamic carbon footprint accounting provided in an embodiment of the present invention.

[0031] Figure 3 A schematic block diagram of a stacker crane multi-machine collaborative scheduling device based on dynamic carbon footprint accounting provided in an embodiment of the present invention;

[0032] Figure 4 A schematic block diagram of a computer device provided for an embodiment of the present invention. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0035] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0036] Please see Figure 1 and Figure 2 , Figure 1 This is a schematic diagram illustrating an application scenario of the stacker crane multi-machine collaborative scheduling method based on dynamic carbon footprint accounting provided in an embodiment of the present invention. Figure 2 This is a schematic flowchart illustrating a multi-machine collaborative scheduling method for stacker cranes based on dynamic carbon footprint accounting, provided in an embodiment of the present invention. This method is applied to a server that interacts with terminals to achieve high-precision real-time calculation of carbon emissions from handling tasks. With the goal of minimizing the total carbon emission cost of the system, a warehouse location-scheduling collaborative optimization model is constructed, enabling the scheduling scheme to proactively respond to "green electricity" periods, effectively reducing indirect carbon emissions. Simultaneously, through an execution-feedback-optimization closed-loop mechanism, model self-calibration and continuous algorithm evolution are achieved, thereby improving the overall carbon efficiency and long-term low-carbon operation capability of automated storage and retrieval systems while ensuring operational efficiency.

[0037] Figure 2This is a flowchart illustrating the multi-machine collaborative scheduling method for stacker cranes based on dynamic carbon footprint accounting provided in an embodiment of the present invention. Figure 2 As shown, the method includes the following steps S110 to S160.

[0038] S110. Construct an energy consumption calculation model for a single handling operation of a stacker crane to obtain the total electrical energy consumption generated by performing a single handling task;

[0039] Specifically, the energy consumption calculation model does not employ empirical formulas or fixed energy consumption values, but is constructed based on the physical dynamics principles of the stacker crane. First, a complete handling task is decomposed into three independent but ordered motion stages: horizontal movement (X-axis), vertical lifting (Y-axis), and fork extension / retraction (Z-axis). For each motion axis, a separate energy consumption sub-model is established. The core input parameters of the model include: the coordinates of the task's start and end points, the actual weight of the goods, the maximum speed / acceleration limits of the equipment on each axis, and a preset speed-time curve (such as an S-curve). By integrating the energy consumed by the motors on each axis during different motion stages such as acceleration, constant speed, deceleration, and braking (considering energy feedback efficiency), the energy consumption of the three axes is summed to obtain the total energy consumption required to perform the specific handling task. This model can be encapsulated as a function module that receives task parameters and outputs accurate energy consumption values.

[0040] In other words, this implementation method overcomes the shortcomings of traditional simplified models that ignore the dynamic characteristics of equipment, and improves the accuracy of energy consumption calculation for a single task. High-precision energy consumption data is a prerequisite for subsequent real carbon emission accounting, ensuring the reliability of carbon cost assessment and laying the foundation for refined scheduling.

[0041] In one embodiment, the step of constructing an energy consumption calculation model for a single handling operation of a stacker crane to obtain the total electrical energy consumption generated by performing a single handling task includes:

[0042] A single handling task is broken down into three basic motion stages: horizontal movement, vertical lifting, and fork extension / retraction.

[0043] Based on the structural parameters, motor performance, and motion control characteristics of the stacker crane, the theoretical energy consumption of each stage is estimated.

[0044] The theoretical electrical energy consumption is calibrated by combining historical operating data or on-site measurement results;

[0045] The energy consumption after calibration in the three stages is summarized and output to obtain the total power consumption of this handling task.

[0046] Specifically, upon receiving a specific handling task instruction (including the starting cargo location coordinates (X1, Y1, Z1) and the target cargo location coordinates (X2, Y2, Z2), as well as the cargo weight W), the system first logically decomposes the entire handling process into three independent and sequentially executed basic motion stages:

[0047] Horizontal travel phase: The stacker crane moves from X1 to X2 along the aisle track;

[0048] Vertical lifting phase: The loading platform is raised (or lowered) from height Y1 to Y2;

[0049] Fork extension / retraction phase: The forks extend (or retract) from the initial position from Z1 to Z2 to complete the storage or retrieval of goods.

[0050] The distance (ΔX, ΔY, ΔZ) and direction of movement at each stage can be directly calculated from the coordinate difference.

[0051] For each of the three stages mentioned above, a theoretical energy consumption sub-model based on physical principles is established. The construction of these models relies on pre-entered detailed parameters of the stacker crane in the system.

[0052] Structural parameters include the stacker crane's own weight, the weight of the loading platform, the weight of the forks, the transmission ratio of each motion axis, the coefficient of friction of the rollers, and the inclination of the guide rails.

[0053] Motor performance: including rated power, rated speed, efficiency curve (ηvs. load / speed), and moment of inertia of each axis drive motor;

[0054] Motion control characteristics: These refer to the system's preset standard speed-time (St) and acceleration-time (at) curves, typically smooth S-shaped curves to ensure stable operation. For each stage, the system calculates the required torque and speed of the motor during acceleration, constant speed, and deceleration based on the travel distance, cargo weight W, and preset speed curve using dynamic equations (such as F=ma). Combined with the motor's efficiency curve, mechanical work is converted into electrical energy consumption. For example, during the vertical lifting phase, the work done against gravity (mgh) needs to be calculated in detail, considering the different efficiencies of the motor during lifting (power consumption) and descent (potentially generating electricity for feedback).

[0055] Purely theoretical models, by neglecting numerous complex factors in actual operation (such as wind resistance, bearing aging, voltage fluctuations, and differences in control algorithms), may result in systematic biases in their estimations. Therefore, this invention introduces a data-driven calibration process. The system utilizes massive amounts of historically accumulated operational data or specially organized field measurement data, which includes precise inputs (coordinates, weight) and corresponding actual energy consumption (collected via high-precision meters) for each task. Through machine learning methods (such as multiple linear regression, support vector machines, or neural networks), a mapping relationship or calibration factor between "theoretical energy consumption" and "actual energy consumption" is established. This calibration model can be global or segmented for different operating conditions (such as no-load / full-load, long-distance / short-distance). Whenever the energy consumption of each stage of a new task is estimated using the theoretical model, it is input into the calibration model to obtain a calibrated energy consumption value that more closely reflects reality.

[0056] The total energy consumption (in kWh) for this specific handling task is obtained by algebraically summing the calibrated energy consumption values ​​for the three stages: horizontal movement, vertical lifting, and fork extension. This total energy consumption value is passed to the subsequent carbon emission accounting module as a key attribute of the task.

[0057] In other words, by decomposing tasks and deeply binding them to the physical characteristics of equipment, this model overcomes the crudeness of traditional "one-size-fits-all" energy consumption estimation. In particular, the introduction of a calibration process based on real-world data effectively bridges the gap between theory and practice, resulting in highly accurate energy consumption predictions for individual tasks. Furthermore, the model can dynamically respond to changes in the task itself. For example, the energy consumption of the vertical lifting phase of moving heavy goods to a higher storage level is much higher than that of moving light goods to a lower storage level; the energy consumption of long-distance horizontal movement is also significantly higher than that of short-distance movement. This dynamic responsiveness to key variables such as load and distance ensures the comparability and fairness of carbon cost assessments for different tasks.

[0058] S120. Establish a minute-level dynamic grid carbon emission factor access mechanism, and calculate the carbon emission of a single transportation task based on the total power consumption and the dynamic grid carbon emission factor corresponding to the task plan execution time.

[0059] Specifically, the system accesses real-time regional power grid minute-level dynamic carbon emission factor (gCO2 / kWh) data streams published by power grid operators or authoritative third-party organizations via API interfaces or data subscription services. This factor reflects the real-time carbon intensity of the power generation mix (thermal power, hydropower, wind power, photovoltaic, etc.) within each minute. When scheduling an execution time for a transport task (e.g., planned for execution at T+15 minutes), the scheduling system queries the dynamic carbon factor corresponding to that future time. Subsequently, by multiplying the total energy consumption (kWh) of the task calculated in step S110 with the dynamic carbon factor at that time, the expected carbon emissions (kgCO2) of the task can be accurately calculated.

[0060] In other words, this mechanism completely abandons the practice of using a fixed annual average factor, enabling carbon emission accounting to truly reflect the spatiotemporal variation characteristics of electricity production. This gives the dispatch system the ability to "sense green electricity" and identify the "green window period" with the lowest carbon intensity during the day, providing a direct basis for guiding tasks to migrate to low-carbon periods.

[0061] In one embodiment, the establishment of a minute-level dynamic grid carbon emission factor access mechanism, and the calculation of the carbon emission of a single transport task based on the total power consumption and the dynamic grid carbon emission factor corresponding to the task plan execution time, includes:

[0062] Access to minute-level regional power grid carbon emission factor data streams;

[0063] Based on the planned execution time period of the task, the carbon emission factor at the corresponding time is matched and weighted according to the actual running time to obtain the effective carbon emission factor of the task.

[0064] The total electrical energy consumption is multiplied by the effective carbon emission factor to calculate the carbon emissions of a single transport mission.

[0065] Specifically, the system interfaces with grid carbon emission data services provided by external authoritative organizations through standardized network communication interfaces (such as RESTful APIs, MQTT messaging protocols, or WebSocket real-time channels). These organizations include national or regional grid dispatch centers, power trading centers, or professional third-party carbon data platforms. They continuously publish grid carbon emission intensity data for their covered areas, updated every minute, typically in grams of carbon dioxide per kilowatt-hour (gCO2 / kWh). The system has a dedicated data receiving and caching module that can acquire and store carbon emission factor sequences arranged minute by minute for a future period (e.g., 24 hours), forming a timestamp-indexed carbon intensity database for the dispatch system to access at any time.

[0066] During the scheduling scheme generation process, each handling task is not only assigned a specific start time but also has a clearly defined expected runtime due to factors such as movement distance and speed planning. Since the actual execution of a task may span two or more minute time intervals (for example, a task starting at 10:15:30 and lasting 95 seconds would cover the 10:15, 10:16, and 10:17 time intervals), a carbon emission factor at a single time point cannot be simply used. Therefore, the system divides the complete execution time of the task into several sub-time intervals based on minute boundaries. For each sub-time interval, the system queries the corresponding minute's carbon emission factor value. Then, it calculates the proportion of each sub-time interval in the total task runtime, using this as the weight of the carbon factor for that minute. Finally, the carbon emission factors for each minute are multiplied by their corresponding weights and summed to obtain a comprehensive "effective carbon emission factor" representing the average carbon intensity of the entire task's execution process. This weighted processing ensures that the carbon factor value accurately reflects the actual electricity consumption distribution during different carbon intensity periods.

[0067] After obtaining the total power consumption (in kilowatt-hours) of the transportation task and the effective carbon emission factor (in grams of carbon dioxide per kilowatt-hour) calculated above, the system multiplies the two to obtain the total carbon dioxide emissions generated by the task during the entire execution process, in grams. This result is the dynamic carbon footprint of the task under the current power grid structure and scheduling arrangement, and serves as the carbon cost attribute of the task, used in the objective function calculation for subsequent multi-task collaborative scheduling optimization.

[0068] In other words, traditional methods typically use annual or monthly average carbon emission factors, which fail to reflect the drastic changes in grid carbon intensity caused by fluctuations in renewable energy output such as wind and solar power within a single day. For example, the carbon intensity during the midday peak of solar power generation may be far lower than during the nighttime period when coal-fired power generation is the primary source of energy. This implementation method, by accessing minute-level updated carbon data and combining it with the actual execution time of tasks for refined weighting, ensures that the carbon emissions of each handling task accurately reflect the actual cleanliness of the grid at the time of electricity consumption, avoiding the distortion problem of "using green electricity but being calculated as high-carbon." Furthermore, since the carbon cost of each task is no longer a fixed value but dynamically changes depending on the time of day it is scheduled for execution, the scheduling algorithm can clearly compare the carbon emission differences resulting from scheduling the same task in the morning, noon, or evening. This dynamic carbon cost signal enables the system to proactively identify and utilize "low-carbon windows" (such as the midday peak of solar power generation), thereby strategically migrating high-energy-consuming tasks to the time of day with the lowest grid carbon intensity, maximizing the use of clean energy, and reducing the overall indirect carbon emissions of the warehouse. Furthermore, by weighting the carbon factor at the minute level according to the actual running time, it is ensured that even if a task happens to cross a point of sudden change in carbon intensity, its carbon cost can still be calculated reasonably and continuously, avoiding evaluation bias or unfairness caused by mismatch in time granularity. This high-resolution accounting method makes the comparison of carbon performance between different tasks and different scheduling schemes more scientific, accurate, and reliable.

[0069] S130. The sum of carbon emissions of all the handling tasks to be performed is taken as the total carbon emission cost of the system. With minimizing the total carbon emission cost of the system as the optimization objective, a multi-lane multi-stacking crane collaborative scheduling optimization model is constructed.

[0070] Specifically, all pending storage and retrieval tasks (task sets) in the warehousing system are taken as input. The decision variables of the optimization model include: which stacker crane each task is assigned to, the execution order of the tasks, and the start time of the tasks. The objective function of the model is explicitly defined as the sum of the carbon emissions of all tasks, i.e., the total carbon emission cost of the system. Constraints include: physical kinematic constraints of the stacker cranes (such as speed and acceleration limits), logical constraints between tasks (such as first-in, first-out), the exclusivity constraint that a stacker crane can only execute one task at a time, and collision avoidance constraints between stacker cranes in different aisles. This model is a complex mixed-integer nonlinear programming (MINLP) problem.

[0071] In other words, by shifting the optimization objective from the traditional "shortest time" to "lowest carbon emissions," the scheduling logic is fundamentally changed. This model can globally find the task allocation and execution sequence scheme with the lowest total carbon emissions, while satisfying all operational requirements and physical constraints, thereby maximizing carbon reduction at the system level.

[0072] In one embodiment, the sum of carbon emissions from all pending handling tasks is taken as the total carbon emission cost of the system. To minimize this total carbon emission cost, a multi-aisle, multi-stall crane collaborative scheduling optimization model is constructed, comprising:

[0073] The estimated carbon emissions of each handling task to be executed on different selectable execution time windows and different available stacker cranes are used as the carbon cost indicator of the task in the scheduling decision.

[0074] The total carbon emission cost of the system is the sum of the carbon cost indicators allocated to all tasks, and the optimization objective of the scheduling scheme is to minimize this total cost.

[0075] The scheduling feasibility constraints are introduced, including: each stacker crane can only execute one task at a time, aisle access is mutually exclusive, tasks must be completed before their deadline, and equipment acceleration and travel capacity are limited.

[0076] Based on the optimization objectives and constraints, a collaborative scheduling optimization model for multiple lanes and multiple stacker cranes, which includes three-dimensional decision variables of task, equipment, and time, is constructed.

[0077] Specifically, before scheduling optimization begins, the system first preprocesses all pending handling tasks. For each task, the system enumerates all feasible execution options, mainly including two dimensions: first, the available stacker cranes (usually limited to available equipment in its own lane or adjacent lanes); and second, the selectable execution time window (e.g., several discrete time periods from the current moment to the task's deadline). For each "task-stacker-time window" combination, the system invokes the aforementioned energy consumption calculation model and minute-level dynamic carbon factor mechanism to pre-calculate the carbon emissions that the task will generate under that combination. This estimated value serves as the "carbon cost indicator" for the task under a specific scheduling choice and is stored in a three-dimensional index table for quick lookup by the subsequent optimization model.

[0078] The core objective of scheduling optimization is clearly defined as: minimizing the total carbon emissions generated by the entire warehousing system in completing all pending tasks, while satisfying all operational requirements and physical constraints. To this end, the system accumulates the carbon cost indicators corresponding to all tasks after they are assigned to specific stacker cranes and execution times; the sum of these is the "total carbon emission cost of the system" for this scheduling scheme. The goal of the optimization algorithm is to find the scheme that minimizes this total cost from all possible scheduling combinations.

[0079] To ensure that the generated scheduling scheme is physically executable and logically conflict-free, multiple key constraints are embedded in the model:

[0080] Equipment exclusivity constraint: It stipulates that any stacker crane can only perform one handling task at a time and cannot process multiple instructions simultaneously;

[0081] Lane mutual exclusion constraint: Since stacker cranes occupy physical space when running in lanes, multiple machines in the same lane must maintain a safe distance to avoid collisions. Therefore, the model must ensure that at any given time, no two stacker cranes in the same lane will have conflicting running paths;

[0082] Task timeliness constraints: Each handling task usually has a clear deadline (such as order delivery time requirements), and the scheduling plan must ensure that the task is completed before its deadline;

[0083] Equipment dynamics constraints: Consider the actual movement capabilities of the stacker crane, including maximum travel / lifting speed, acceleration limits, minimum turning / start-stop distance, etc. The model must ensure that the planned task execution time is sufficient for the equipment to physically complete the specified journey, avoiding the generation of unrealistic "overspeed" scheduling instructions.

[0084] Based on the aforementioned objectives and constraints, a structured mathematical optimization model is constructed: a multi-aisle, multi-stall crane collaborative scheduling optimization model. The core of this model is a set of three-dimensional decision variables, each representing whether a task is assigned to a stacker crane for execution within a specific time window. Through these variables, the model can comprehensively describe the matching relationship between tasks, equipment, and time. The entire problem is formalized as a large-scale combinatorial optimization problem, the objective of which is to find a set of decision variable values ​​that minimizes the sum of the carbon cost indices of all selected task items, while satisfying all hard constraints. This model fully reflects the collaborative coupling relationship between multiple aisles, multiple devices, and multiple tasks.

[0085] In other words, by explicitly shifting the optimization objective to "minimizing the total carbon emission cost of the system," the scheduling logic changes from simply pursuing speed to considering environmental impact, driving the warehousing system towards a green and low-carbon direction. Furthermore, by assigning different carbon costs to the execution options of each task on different equipment and at different times, the model can identify "high-carbon paths" and "low-carbon paths." For example, a heavy-load task executed at night by an older stacker crane may have a high carbon cost, while executing it at midday by a newer, more efficient machine has a lower carbon cost. The optimization algorithm will prioritize the latter, thus achieving the optimal carbon emission configuration at the global level. In addition, the four types of constraints introduced cover key dimensions such as equipment physical limits, spatial safety, and business timeliness, ensuring that the generated "carbon-optimal" solution is not a theoretical concept but a safe and reliable sequence of control commands that can be directly issued and executed. This avoids the risk of sacrificing system stability or operational reliability for carbon reduction.

[0086] S140. Design a two-layer collaborative scheduling solution strategy for storage location and scheduling, and solve the optimization model to generate a carbon-optimal scheduling scheme.

[0087] Specifically, to efficiently solve the complex optimization model described above, a hierarchical solution strategy is adopted. The upper layer is the storage location allocation layer, which initially determines the target storage locations for goods to be stored based on information such as the frequency of goods entering and leaving the warehouse (ABC classification), their relevance, and carbon efficiency principles (e.g., high-frequency goods should be allocated to storage locations closer to the entrance / exit and with lower handling energy consumption). The lower layer is the task scheduling layer, which optimizes the task sequence and execution time of multiple stacker cranes based on the storage location allocation results determined by the upper layer, using heuristic algorithms (such as genetic algorithms and simulated annealing) or reinforcement learning methods to minimize the total carbon cost. There is information feedback between the two layers. For example, if the scheduling layer finds that certain storage location allocations lead to excessively high carbon costs, it can propose adjustment suggestions to the upper layer, forming a collaborative iteration.

[0088] In other words, this strategy effectively decouples the two tightly coupled challenges of storage location allocation and task scheduling, reducing the overall problem-solving complexity. More importantly, it incorporates the upstream decision of storage location allocation into the scope of carbon optimization, achieving end-to-end carbon coordination from "where to store" to "how to move," thus avoiding ineffective high-carbon transportation caused by unreasonable storage locations.

[0089] In one embodiment, the design of a two-layer collaborative scheduling strategy involving storage location and scheduling, and the solution of the optimization model to generate a carbon-optimal scheduling scheme, includes:

[0090] At the upper level, based on the prediction of the frequency of goods entering and leaving the warehouse, high-frequency goods are preferentially allocated to low-energy-consumption storage locations near the entrance and exit, forming a low-carbon storage location layout.

[0091] Using the low-carbon storage location layout as input for the lower-level scheduling, the starting and ending points of each handling task are determined, and the carbon emission data of the task under different stacker cranes and execution periods are calculated accordingly.

[0092] Based on carbon emission data, under the premise of meeting equipment resource constraints and task timeliness requirements, an intelligent optimization algorithm is used to solve the feasible scheduling scheme with the lowest total carbon emissions.

[0093] Output the carbon-optimal scheduling scheme, which includes task-equipment allocation, execution order, and timing, based on the feasible scheduling scheme with the lowest total carbon emissions.

[0094] Specifically, the system first analyzes historical warehousing operation data, using time series forecasting and machine learning methods to predict the frequency of inbound and outbound operations for various types of goods within a future period (such as the next week or month). Based on the prediction results, the system executes an optimized storage location allocation strategy: "high-frequency goods" with high expected inbound and outbound frequencies are prioritized for storage in locations near warehouse entrances / exits, aisle ends, or the initial parking position of stacker cranes; while low-frequency goods are arranged in locations further away, at higher vertical heights, or deep within the warehouse. The core logic of this layout is that high-frequency goods, due to their frequent handling, will have reduced cumulative carbon emissions over the entire cycle if each handling trip is short and energy consumption is low. This process outputs an optimized static or quasi-static storage location allocation scheme as the basic input for the lower-level scheduling module.

[0095] After obtaining the aforementioned low-carbon storage location layout, the starting and ending coordinates (i.e., source and target storage locations) of all pending inbound and outbound tasks are uniquely determined. Based on this, the scheduling system invokes the aforementioned energy consumption calculation model and minute-level grid carbon factor mechanism to pre-calculate the carbon emissions of each task under different feasible execution conditions. These conditions include the available stacker cranes (usually limited by lane assignment), multiple possible execution time periods (which must meet the task deadline), and the dynamic carbon intensity matched to the corresponding equipment operating during that time period. Thus, each task forms a candidate set containing multiple "equipment-time" combinations and their corresponding carbon emission values, providing a quantitative basis for subsequent scheduling decisions.

[0096] With complete task carbon emission data, the system initiates a scheduling optimization solver. This solver aims to minimize the total carbon emissions of all tasks while strictly adhering to a series of hard constraints, including: each stacker crane can only handle one task at any given time; equipment paths within the same lane must not conflict; all tasks must be completed before their deadlines; and limitations on the equipment's acceleration and maximum travel capacity. Considering the large problem size, strong variable coupling, and complex solution space, the system typically employs intelligent optimization algorithms for efficient solving, such as improved genetic algorithms, simulated annealing, tabu search, or hybrid heuristics combining reinforcement learning. These algorithms can search for the lowest total carbon emission from a vast pool of feasible scheduling schemes within a reasonable computation time.

[0097] After optimization, the system generates a structured scheduling and execution plan, namely the carbon-optimal scheduling plan. This plan explicitly specifies which stacker crane will perform each handling task, the execution order of tasks, and the specific start and end times for each task. This plan not only meets all operational functionality and safety requirements but also minimizes the overall carbon emissions of the system under the current warehouse layout and grid carbon intensity conditions. This plan can be directly deployed to the warehouse control system to drive the stacker cranes to execute operations along a low-carbon path.

[0098] In other words, traditional methods often treat warehouse location allocation and task scheduling as two independent processes. This leads to situations where, even with the best scheduling algorithms, high-energy-consuming long-distance handling may be forced due to poor storage locations. This implementation method, through a two-layer collaborative mechanism, proactively ensures that warehouse location layout serves the goal of low-carbon scheduling, compressing the physical journey of high-frequency tasks from the source. This reduces carbon emissions in both the "storage" and "retrieval" stages, achieving deep decarbonization throughout the entire lifecycle of warehousing operations. Furthermore, high-frequency goods are handled far more frequently than low-frequency goods. Although the energy savings per handling are small, multiplied by the high frequency, the cumulative carbon-saving benefits are considerable. By placing high-frequency goods at the front end, the system continuously gains a structural energy consumption advantage without increasing equipment investment, enabling the warehouse to maintain a low carbon emission level per order in long-term operation. In addition, upper-level warehouse location optimization provides a high-quality initial solution space for the lower-level scheduling problem. Because the start and end points of tasks have been optimized, the inherent carbon cost baseline of the tasks themselves is lower. The scheduling algorithm does not need to struggle with a large number of high-carbon task combinations, making it easier to find a globally better or even near-optimal low-carbon scheduling scheme, while shortening the solution time and meeting real-time or near-real-time scheduling requirements. In addition, the low-carbon warehouse layout is not static, but is updated regularly (e.g., weekly or daily) based on rolling predictions of future inbound and outbound frequencies. This dynamic adjustment mechanism enables the warehouse to quickly respond to changes in cargo flow patterns brought about by peak sales seasons, promotional activities, or supply chain changes, always keeping warehouse allocation in line with actual operational needs, avoiding ineffective handling rebounds caused by fixed warehouse locations, and ensuring the continuous and stable low-carbon benefits.

[0099] S150. Simulation verification of the carbon optimal scheduling scheme is performed.

[0100] Specifically, before putting the scheduling scheme into actual operation, it was first validated in a high-fidelity digital twin simulation environment. The simulation model reproduced the physical layout of the warehouse, the dynamic characteristics of the stacker crane, and the task flow, and was used to verify the feasibility, safety, and expected carbon reduction effect of the scheme.

[0101] In other words, the safety and reliability of the solution can be guaranteed through simulation verification.

[0102] In one embodiment, the simulation verification of the optimal carbon scheduling scheme includes:

[0103] The carbon-optimal scheduling scheme is used as input data and imported into a digital twin simulation platform that integrates the stacker crane dynamic motion model, energy consumption model, and real-time power grid carbon emission factor.

[0104] On the digital twin simulation platform, the system drives the virtual stacker crane to perform all handling tasks according to the scheduling instructions, calculates the running time and path energy consumption of each action in real time, and dynamically accumulates the total carbon emissions in combination with the carbon emission factor of the current period. At the same time, it detects whether there are path conflicts or resource contention between equipment.

[0105] After the simulation is completed, the output will include the task completion time, equipment conflict records, total system energy consumption, and total carbon emissions.

[0106] If all indicators meet the preset constraints, the scheme is deemed feasible and proceeds to the actual execution stage; otherwise, the bottleneck information from the simulation feedback is extracted to correct the model parameters or adjust the constraints, and the improved carbon-optimal scheduling scheme is re-solved.

[0107] Specifically, the system first establishes a digital twin simulation environment that perfectly corresponds to the actual automated storage and retrieval system (AS / RS). The digital twin simulation platform integrates three core models: first, a dynamic motion model of the stacker crane, capable of accurately simulating the equipment's horizontal movement, vertical lifting, and fork extension / retraction within the aisles, including physical characteristics such as acceleration, deceleration, maximum speed, and start-stop inertia; second, a refined energy consumption model, which can calculate the instantaneous power and cumulative energy consumption in real time based on the stacker crane's speed, load, direction of movement, and motor efficiency at every moment; and third, a real-time grid carbon emission factor data interface updated every minute, ensuring that the carbon intensity values ​​used in the simulation are synchronized with the actual grid conditions.

[0108] The carbon-optimal scheduling scheme generated by the aforementioned optimization algorithm is used as input and fully imported into the simulation platform. This scheme includes which stacker crane each task is assigned to, the order in which tasks are executed, and the start time accurate to the second. Based on these instructions, the simulation engine sequentially drives the corresponding virtual stacker cranes to perform handling operations. During execution, the system continuously tracks the real-time location, speed, load status, and action stage of each device, and synchronously records its running path, time consumption, and energy consumption.

[0109] At each time step of the task execution, the simulation platform multiplies the instantaneous energy consumption of the current stacker crane by the carbon emission factor obtained from the power grid interface at the corresponding moment to obtain the carbon emission contribution at that moment, and continuously accumulates them to form the total carbon emissions of the system. Simultaneously, the platform's built-in conflict detection module monitors the spatial position and movement trajectory of all equipment in real time, determining whether there are violations of physical safety rules such as two stacker cranes intersecting paths in the same aisle, equipment spacing being too close, or shared resources (such as entrance / exit buffer zones) being simultaneously occupied. Once a conflict is detected, the system immediately records the time of the conflict, the equipment involved, and the type of conflict.

[0110] Once all handling tasks are completed in the simulation, the platform automatically generates a comprehensive performance report, including: the actual completion time of all tasks (to verify whether deadlines are met), a list of equipment conflict events (if any), the total power consumption of the system throughout the entire operation cycle, and the final total carbon dioxide emissions. The system compares these indicators with preset constraint thresholds: for example, whether all tasks were completed on time, whether there were any unacceptable equipment conflicts, and whether the total carbon emissions are indeed lower than the baseline scheme. If all indicators meet the requirements, the scheduling scheme is deemed "feasible" and can be implemented in the actual warehouse control system; if there are conflicts, timeouts, or abnormal carbon emissions, it is considered "infeasible."

[0111] For infeasible solutions, the system does not discard them directly, but instead conducts an in-depth analysis of the bottlenecks exposed during the simulation. For example, if conflicts are concentrated in a specific roadway during a certain period, it may indicate that the equipment density in that area is too high or the scheduling intervals are too frequent; if a task frequently times out, it may be that the energy consumption model underestimates its actual running time. This feedback information will be used to correct relevant parameters in the original optimization model (such as minimum safe distance between equipment, task execution time margin, energy consumption estimation coefficient, etc.) or adjust constraints (such as increasing the mutual exclusion period of roadways and relaxing the deadlines of some tasks). Subsequently, the system calls the optimization solver again to generate an improved carbon-optimal scheduling scheme, and sends it back to the simulation platform for verification, forming a closed-loop iterative process of "optimization-simulation-feedback-re-optimization" until a final scheme that is both low-carbon and safe to execute is obtained, namely the improved carbon-optimal scheduling scheme.

[0112] In other words, theoretical optimization results are often based on simplification assumptions and may ignore details of equipment dynamics or the complexity of spatial interactions. High-fidelity simulation can expose potential risks of equipment collisions, deadlocks, or timeouts in scheduling schemes in advance without interfering with actual production. This avoids production line downtime or safety accidents caused by blind execution, fundamentally ensuring the unity of "low carbon" and "reliability." Furthermore, the simulation uses dynamic energy consumption and grid carbon factors consistent with reality, making the calculated total carbon emissions much closer to reality than static estimates. This not only verifies the carbon reduction effect of the scheme but also provides a reliable data source for subsequent carbon performance assessment. Simultaneously, simulation results can also back-calibrate the energy consumption model, continuously improving its predictive accuracy. Moreover, directly testing scheduling schemes in a physical warehouse can lead to material delays, equipment damage, and even manual intervention costs if conflicts or failures occur. Digital twin simulation can complete the simulation of a full day's tasks within minutes, quickly verifying multiple schemes and significantly shortening the cycle from algorithm generation to on-site deployment, thus improving the engineering practicality of low-carbon scheduling technology. Furthermore, by feeding back the deviations and bottleneck information from each simulation to the optimization model, the system has the ability to "accumulate experience." Over time, the model's understanding of equipment behavior, task characteristics, and environmental changes becomes increasingly accurate, and the quality of the generated scheduling schemes gradually improves, forming an intelligent mechanism for long-term low-carbon operation.

[0113] S160. After the simulation verification is passed, the carbon-optimal scheduling scheme is converted into control commands and sent to the stacker crane for execution. During the execution process, the actual energy consumption data of the equipment is collected in real time, and the actual carbon emissions are calculated based on the dynamic power grid carbon emission factor. Finally, the actual carbon emissions are fed back to the energy consumption calculation model and scheduling solution strategy for dynamic calibration of the model and continuous optimization of the algorithm.

[0114] Specifically, after successful verification, the scheduling engine parses the solution into specific PLC control instructions and sends them to each stacker crane for execution. During execution, the actual energy consumption data of each stacker crane is collected in real time through the electricity meters or energy consumption monitoring modules built into the drivers on the equipment. Combined with the real dynamic carbon factor during task execution, the actual carbon emissions are calculated. Finally, the "actual energy consumption / carbon emissions" are compared with the "model prediction value," and machine learning methods (such as online learning and parameter regression) are used to dynamically adjust key parameters in the energy consumption calculation model (such as friction coefficient and motor efficiency), and the heuristic rules of the scheduling algorithm are fine-tuned to achieve continuous evolution of the model and algorithm.

[0115] In other words, execution monitoring ensures the effective implementation of the theoretical solution, while the closed-loop feedback calibration mechanism gives the system a strong adaptive capability, which can automatically compensate for model deviations caused by factors such as equipment wear and environmental changes, ensuring that the system maintains high-precision carbon accounting capabilities and optimal scheduling performance in long-term operation, forming a sustainable and optimized intelligent low-carbon operation system.

[0116] In one embodiment, after the simulation verification is passed, the carbon-optimal scheduling scheme is converted into control commands and issued to the stacker crane for execution; during execution, the actual energy consumption data of the equipment is collected in real time, and the actual carbon emissions are calculated based on the dynamic grid carbon emission factor; finally, the actual carbon emissions are fed back to the energy consumption calculation model and scheduling solution strategy for dynamic model calibration and continuous algorithm optimization, including:

[0117] The verified optimal carbon scheduling scheme is converted into control commands and sent to the stacker crane control system for execution.

[0118] During the execution process, the actual energy consumption data and operating status data of the equipment are collected in real time through the Internet of Things;

[0119] By combining the dynamic carbon emission factor, actual energy consumption, and operational status data at the time of mission execution, the actual carbon emissions of each mission are calculated.

[0120] The actual carbon emissions are fed back to the energy consumption calculation model and scheduling solution strategy to calibrate model parameters and optimize algorithm logic, thereby achieving continuous adaptive improvement of the system.

[0121] Specifically, after the digital twin simulation verifies the feasibility of the scheduling scheme, the system performs an engineering transformation of this carbon-optimal scheduling scheme. Specifically, the high-level task descriptions originally used for optimization and simulation (such as "Task A is executed by stacker crane 3 at 10:15") are parsed into a sequence of control instructions recognizable by the lower-level equipment. These instructions include precise motion target coordinates (cargo location address), action type (picking / placing), speed curve settings, start / stop timing, etc., conforming to the stacker crane controller's communication protocol (such as PLC instruction sets or industrial bus standards). After the transformation, the instructions are sent in real-time through the industrial network to the stacker crane control system of the corresponding aisle, driving the physical equipment to perform handling operations according to the low-carbon scheduling plan.

[0122] During the stacker crane's operation, various sensors and smart meters deployed on the equipment continuously operate. Through IoT technologies (such as industrial Ethernet, wireless sensor networks, or edge computing gateways), the system collects key operational status data of the stacker crane in real time, including current location, operating speed, acceleration, motor current, voltage, power factor, and cumulative power consumption. This data is uploaded to the central data platform at a high frequency (e.g., multiple times per second), forming a measured operating trajectory and energy consumption record that corresponds one-to-one with the scheduling plan.

[0123] The system synchronously acquires the regional power grid carbon emission factor data stream updated every minute during task execution. For each completed transport task, the system time-aligns and matches the second-by-second or minute-by-minute energy consumption data collected during its actual execution time with the corresponding power grid carbon emission factor. Then, the energy consumption value of each time segment is multiplied by its corresponding carbon emission factor, and the results are summed to accurately calculate the actual carbon dioxide emissions generated by the task throughout its execution. This result is closer to reality than the estimates from the simulation phase, reflecting the impact of real-world factors such as equipment aging, load fluctuations, and environmental temperature and humidity.

[0124] The actual carbon emissions, actual running time, and actual energy consumption of all tasks are systematically archived and compared with the predicted values ​​before scheduling. If a systematic deviation is found (e.g., a certain type of task always consumes 10% more energy than the model predicts), a model calibration process is triggered: relevant parameters in the energy consumption calculation model (such as friction coefficient, motor efficiency curve, no-load / full-load power consumption baseline, etc.) are adjusted to better reflect the actual performance of the current equipment. Simultaneously, this measured data is also used to optimize the scheduling solution strategy itself—for example, introducing correction weights based on historical errors into the objective function of the intelligent optimization algorithm, or adding time margins for tasks with high uncertainty to the constraints. Through this closed-loop mechanism of "execution-observation-learning-improvement," the entire scheduling system possesses the ability to continuously adapt and evolve.

[0125] In other words, traditional scheduling systems typically only generate scheduling plans, lacking tracking and feedback on execution results. This implementation method re-injects actual operational data into the optimization front-end, enabling the system not only to "plan for low carbon emissions" but also to "act accurately and learn quickly," forming a complete green intelligent control closed loop and improving the effectiveness of low-carbon strategies. Furthermore, as operational data accumulates, the energy consumption model can dynamically adapt to real-world factors such as equipment performance degradation, changes in maintenance status, or alterations in the operating environment, preventing scheduling plans from deviating from the optimal path due to model inaccuracies. In the long term, the continuously improving accuracy of carbon emission accounting provides a highly reliable data foundation for enterprise carbon management, carbon disclosure, and carbon trading. Moreover, the actual warehousing environment contains many unpredictable factors, such as cargo weight deviations, minor track deformations, and temporary obstacles. By continuously training and adjusting the scheduling logic with real data, the algorithm gradually learns to identify high-risk tasks, reserve reasonable buffers, and allocate resources more rationally in multi-objective trade-offs, thereby improving system stability and task completion rates while ensuring low carbon emissions.

[0126] The aforementioned multi-machine collaborative scheduling method for stacker cranes based on dynamic carbon footprint accounting establishes a minute-level dynamic grid carbon emission factor access mechanism and combines it with a precisely constructed energy consumption calculation model for a single stacker crane handling operation. This allows for accurate calculation of the actual carbon emissions of each handling task at the planned execution time, abandoning the traditional coarse-grained calculation method based on fixed annual average factors. It ensures that the carbon cost measurement results are highly synchronized with the actual carbon intensity fluctuations at the power generation end, providing a real and reliable data foundation for subsequent low-carbon scheduling decisions. Furthermore, since the scheduling optimization model directly minimizes the sum of carbon emissions from all tasks to be executed (i.e., the total carbon emission cost of the system), and the calculation of this cost depends on the dynamic carbon factor at the task execution time, the resulting scheduling scheme intelligently prioritizes high-energy-consuming tasks during time windows with lower grid carbon intensity (i.e., a higher proportion of "green electricity"). This enables the warehousing system to proactively track and utilize clean energy, reducing indirect carbon emissions caused by electricity consumption. Furthermore, by designing a two-layer collaborative solution strategy of storage location and scheduling, the upstream storage location allocation (affecting handling distance and frequency) and the downstream multi-machine collaborative scheduling (determining task execution timing and path) are jointly optimized. Simultaneously, the entire optimization process uses dynamic carbon cost as the core link, tightly coupling external energy supply (dynamic power grid) and internal equipment operation (stacking crane energy consumption). This integrated decision-making model breaks down the barriers between different links in traditional methods, achieving optimal carbon emission control at the system-wide level. Moreover, this technical solution not only completes optimization before scheduling but also collects actual equipment energy consumption data in real time during the solution execution process, calculating the actual carbon emissions accordingly. This actual data is fed back into the energy consumption calculation model and scheduling solution strategy for dynamic calibration of model parameters and continuous optimization of the algorithm. This closed-loop mechanism ensures that the system can adapt to real-world factors such as equipment aging and changes in operating conditions, maintaining high-precision carbon accounting capabilities and efficient low-carbon scheduling performance over the long term.

[0127] Figure 3 This is a schematic block diagram of a stacker crane multi-machine collaborative scheduling device 300 based on dynamic carbon footprint accounting, provided in an embodiment of the present invention. Figure 3 As shown, corresponding to the above-described stacker crane multi-machine collaborative scheduling method based on dynamic carbon footprint accounting, the present invention also provides a stacker crane multi-machine collaborative scheduling device 300 based on dynamic carbon footprint accounting. This stacker crane multi-machine collaborative scheduling device 300 includes a unit for executing the above-described stacker crane multi-machine collaborative scheduling method based on dynamic carbon footprint accounting, and the device can be configured in a server. Specifically, please refer to... Figure 3 The stacker crane multi-machine collaborative scheduling device 300 based on dynamic carbon footprint accounting includes:

[0128] The acquisition unit 301 is used to construct an energy consumption calculation model for a single handling operation of the stacker crane in order to obtain the total electrical energy consumption generated by performing a single handling task.

[0129] A calculation unit 302 is established to establish a minute-level dynamic grid carbon emission factor access mechanism, and to calculate the carbon emission of a single transportation task based on the total power consumption and the dynamic grid carbon emission factor corresponding to the task plan execution time.

[0130] Construction unit 303 is used to sum the carbon emissions of all the handling tasks to be performed as the total carbon emission cost of the system, and to construct a multi-lane multi-stacking crane collaborative scheduling optimization model with minimizing the total carbon emission cost of the system as the optimization objective.

[0131] Design solver unit 304 is used to design a two-layer collaborative scheduling solution strategy of storage location and scheduling, and solve the optimization model to generate a carbon-optimal scheduling scheme.

[0132] Simulation unit 305 is used to simulate and verify the carbon optimal scheduling scheme;

[0133] The execution feedback unit 306 is used to convert the carbon-optimal scheduling scheme into control commands and send them to the stacker crane for execution after the simulation verification is passed; during the execution process, the actual energy consumption data of the equipment is collected in real time, and the actual carbon emissions are calculated based on the dynamic power grid carbon emission factor; finally, the actual carbon emissions are fed back to the energy consumption calculation model and scheduling solution strategy for dynamic calibration of the model and continuous optimization of the algorithm.

[0134] In one embodiment, the construction acquisition unit 301 includes:

[0135] The decomposition module is used to break down a single handling task into three basic motion stages: horizontal walking, vertical lifting, and fork extension / retraction.

[0136] The estimation module is used to estimate the theoretical energy consumption of each stage based on the stacker crane's structural parameters, motor performance, and motion control characteristics.

[0137] The calibration module is used to calibrate the theoretical electrical energy consumption by combining historical operating data or on-site measurement results;

[0138] The summary output module is used to summarize and output the energy consumption after calibration in the three stages to obtain the total power consumption of this handling task.

[0139] In one embodiment, the accounting unit 302 includes:

[0140] The access module is used to access minute-level regional power grid carbon emission factor data streams;

[0141] The matching and weighting module is used to match the carbon emission factor at the corresponding time according to the planned execution time of the task, and weight it according to the actual running time to obtain the effective carbon emission factor of the task.

[0142] The calculation module is used to multiply the total power consumption by the effective carbon emission factor to calculate the carbon emissions of a single transport task.

[0143] In one embodiment, the building unit 303 includes:

[0144] The module is used to estimate the carbon emissions of each handling task to be executed on different selectable execution time windows and different available stacker cranes, and use this as a carbon cost indicator for the task in scheduling decisions.

[0145] The configuration module is used to set the total carbon emission cost of the system as the sum of the carbon cost indicators allocated to all tasks, and to set the optimization objective of the scheduling scheme as minimizing this total cost.

[0146] The module introduces scheduling feasibility constraints, including: each stacker crane can only execute one task at a time, aisle access is mutually exclusive, tasks must be completed before their deadline, and equipment acceleration and travel capacity limits.

[0147] The module is used to construct a collaborative scheduling optimization model for multiple lanes and multiple stacker cranes, which includes three-dimensional decision variables of task, equipment, and time, based on optimization objectives and constraints.

[0148] In one embodiment, the design solving unit 304 includes:

[0149] The prediction module is used at the upper level to predict the frequency of goods entering and leaving the warehouse, and to prioritize the allocation of high-frequency goods to low-energy storage locations near the entrance and exit, thus forming a low-carbon storage location layout.

[0150] The determination module is used to take the low-carbon warehouse layout as input to the lower-level scheduling, determine the start and end points of each handling task, and calculate the carbon emission data of the task under different stacker cranes and execution periods.

[0151] The solution module is used to solve the feasible scheduling scheme with the lowest total carbon emissions based on carbon emission data, under the premise of meeting equipment resource constraints and task timeliness requirements;

[0152] The output module is used to output a carbon-optimal scheduling scheme, including task-equipment allocation, execution order, and timing arrangement, based on the feasible scheduling scheme with the lowest total carbon emissions.

[0153] In one embodiment, the simulation unit 305 includes:

[0154] The import module is used to take the carbon-optimal scheduling scheme as input data and import it into a digital twin simulation platform that integrates the stacker crane dynamic motion model, energy consumption model and real-time power grid carbon emission factor.

[0155] The execution detection module is used on the digital twin simulation platform to drive the virtual stacker crane to perform all handling tasks according to the scheduling instructions. It calculates the running time and path energy consumption of each action in real time, and dynamically accumulates the total carbon emissions in combination with the carbon emission factor of the current period. At the same time, it detects whether there are path conflicts or resource contention between devices.

[0156] The output module is used to output the task completion time, equipment conflict records, total system energy consumption and total carbon emissions after the simulation is completed.

[0157] The decision correction module is used to determine the feasibility of the scheme if all indicators meet the preset constraints, and then proceed to the actual execution stage; otherwise, it extracts the bottleneck information from the simulation feedback, uses it to correct the model parameters or adjust the constraints, and re-solves to generate an improved carbon-optimal scheduling scheme.

[0158] In one embodiment, the execution feedback unit 306 includes:

[0159] The conversion module is used to convert the verified carbon-optimal scheduling scheme into control commands and send them to the stacker crane control system for execution.

[0160] The data acquisition module is used to collect real-time data on the actual energy consumption and operating status of the device via the Internet of Things during the execution process.

[0161] The calculation module is used to calculate the actual carbon emissions for each task by combining the dynamic carbon emission factor, actual energy consumption and operating status data at the time of task execution.

[0162] The feedback optimization module is used to feed back the actual carbon emissions to the energy consumption calculation model and scheduling solution strategy, in order to calibrate the model parameters and optimize the algorithm logic, so as to achieve continuous adaptive improvement of the system.

[0163] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the stacker crane multi-machine collaborative scheduling device 300 based on dynamic carbon footprint accounting and each unit can be referred to the corresponding description in the foregoing method embodiments. For the sake of convenience and brevity, it will not be repeated here.

[0164] The aforementioned stacker crane multi-machine collaborative scheduling device 300 based on dynamic carbon footprint accounting can be implemented as a computer program, which can be used in various ways, such as... Figure 4 It runs on the computer device shown.

[0165] Please see Figure 4 , Figure 4 This is a schematic block diagram of a computer device provided in an embodiment of this application. The computer device 500 can be a server, wherein the server can be a standalone server or a server cluster composed of multiple servers.

[0166] See Figure 4 The computer device 500 includes a processor 502, a memory, and a network interface 505 connected via a system bus 501. The memory may include a non-volatile storage medium 503 and internal memory 504.

[0167] The non-volatile storage medium 503 may store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions that, when executed, cause the processor 502 to perform a stacker crane multi-machine collaborative scheduling method based on dynamic carbon footprint accounting.

[0168] The processor 502 provides computing and control capabilities to support the operation of the entire computer device 500.

[0169] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute a stacker crane multi-machine collaborative scheduling method based on dynamic carbon footprint accounting.

[0170] This network interface 505 is used for network communication with other devices. Those skilled in the art will understand that... Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device 500 to which the present application is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0171] The processor 502 is used to run a computer program 5032 stored in the memory to perform the following steps:

[0172] A calculation model for the energy consumption of a single handling operation of a stacker crane is constructed to obtain the total electrical energy consumption generated by executing a single handling task. A minute-level dynamic grid carbon emission factor access mechanism is established, and the carbon emission of a single handling task is calculated based on the total electrical energy consumption and the dynamic grid carbon emission factor corresponding to the task plan execution time. The sum of the carbon emissions of all handling tasks to be executed is taken as the total carbon emission cost of the system. With minimizing the total carbon emission cost of the system as the optimization objective, a multi-lane multi-stacking crane collaborative scheduling optimization model is constructed. A storage location-scheduling two-layer collaborative scheduling solution strategy is designed, and the optimization model is solved to generate a carbon-optimal scheduling scheme. The carbon-optimal scheduling scheme is simulated and verified. After the simulation verification is passed, the carbon-optimal scheduling scheme is converted into control commands and issued to the stacker crane for execution. During the execution process, the actual energy consumption data of the equipment is collected in real time, and the actual carbon emission is calculated based on the dynamic grid carbon emission factor. Finally, the actual carbon emission is fed back to the energy consumption calculation model and the scheduling solution strategy for dynamic calibration of the model and continuous optimization of the algorithm.

[0173] It should be understood that in the embodiments of this application, the processor 502 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0174] It will be understood by those skilled in the art that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program includes program instructions and can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiments of the above methods.

[0175] Therefore, the present invention also provides a storage medium. This storage medium can be a computer-readable storage medium. The storage medium stores a computer program, wherein when executed by a processor, the computer program causes the processor to perform the following steps:

[0176] A calculation model for the energy consumption of a single handling operation of a stacker crane is constructed to obtain the total electrical energy consumption generated by executing a single handling task. A minute-level dynamic grid carbon emission factor access mechanism is established, and the carbon emission of a single handling task is calculated based on the total electrical energy consumption and the dynamic grid carbon emission factor corresponding to the task plan execution time. The sum of the carbon emissions of all handling tasks to be executed is taken as the total carbon emission cost of the system. With minimizing the total carbon emission cost of the system as the optimization objective, a multi-lane multi-stacking crane collaborative scheduling optimization model is constructed. A storage location-scheduling two-layer collaborative scheduling solution strategy is designed, and the optimization model is solved to generate a carbon-optimal scheduling scheme. The carbon-optimal scheduling scheme is simulated and verified. After the simulation verification is passed, the carbon-optimal scheduling scheme is converted into control commands and issued to the stacker crane for execution. During the execution process, the actual energy consumption data of the equipment is collected in real time, and the actual carbon emission is calculated based on the dynamic grid carbon emission factor. Finally, the actual carbon emission is fed back to the energy consumption calculation model and the scheduling solution strategy for dynamic calibration of the model and continuous optimization of the algorithm.

[0177] The storage medium can be any computer-readable storage medium capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), magnetic disk, or optical disk.

[0178] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0179] In the several embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For example, the division of each unit is merely a logical functional division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0180] The steps in the method of this invention can be adjusted, merged, or reduced in order according to actual needs. The units in the device of this invention can be merged, divided, or reduced according to actual needs. Furthermore, the functional units in the various embodiments of this invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0181] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0182] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A multi-machine collaborative scheduling method for stacker cranes based on dynamic carbon footprint accounting, characterized in that, include: Construct an energy consumption calculation model for a single handling operation of a stacker crane to obtain the total electrical energy consumption generated by performing a single handling task; Establish a minute-level dynamic grid carbon emission factor access mechanism, and calculate the carbon emission of a single transportation task based on the total power consumption and the dynamic grid carbon emission factor corresponding to the task plan execution time. The total carbon emission cost of the system is the sum of the carbon emissions of all the handling tasks to be performed. To minimize the total carbon emission cost of the system, a multi-lane multi-stacking crane collaborative scheduling optimization model is constructed. Design a two-layer collaborative scheduling solution strategy of storage location and scheduling, and solve the optimization model to generate a carbon-optimal scheduling scheme; The carbon-optimal scheduling scheme was verified through simulation. After the simulation verification is successful, the carbon optimal scheduling scheme is converted into control commands and sent to the stacker crane for execution; During execution, the actual energy consumption data of the equipment is collected in real time, and the actual carbon emissions are calculated based on the dynamic power grid carbon emission factor. Finally, the actual carbon emissions are fed back to the energy consumption calculation model and scheduling solution strategy for dynamic calibration of the model and continuous optimization of the algorithm. The method involves summing the carbon emissions of all pending handling tasks as the total carbon emission cost of the system, and minimizing this total carbon emission cost as the optimization objective. A multi-aisle, multi-stall crane collaborative scheduling optimization model is constructed, including: The estimated carbon emissions of each handling task to be executed on different selectable execution time windows and different available stacker cranes are used as the carbon cost indicator of the task in the scheduling decision. The total carbon emission cost of the system is the sum of the carbon cost indicators allocated to all tasks, and the optimization objective of the scheduling scheme is to minimize the total carbon emission cost of the system. The scheduling feasibility constraints are introduced, including: each stacker crane can only execute one task at a time, aisle access is mutually exclusive, tasks must be completed before their deadline, and equipment acceleration and travel capacity are limited. Based on the optimization objective and constraints, a multi-lane multi-stall crane collaborative scheduling optimization model with three-dimensional decision variables of task, equipment and time is constructed. The design employs a two-layer collaborative scheduling strategy involving storage location and scheduling, and solves the optimization model to generate a carbon-optimal scheduling scheme, including: At the upper level, based on the prediction of the frequency of goods entering and leaving the warehouse, high-frequency goods are preferentially allocated to low-energy-consumption storage locations near the entrance and exit, forming a low-carbon storage location layout. Using the low-carbon storage location layout as input for the lower-level scheduling, the starting and ending points of each handling task are determined, and the carbon emission data of the task under different stacker cranes and execution periods are calculated accordingly. Based on carbon emission data, under the premise of meeting equipment resource constraints and task timeliness requirements, an intelligent optimization algorithm is used to solve the feasible scheduling scheme with the lowest total carbon emissions. Output the carbon-optimal scheduling scheme, which includes task-equipment allocation, execution order, and timing, based on the feasible scheduling scheme with the lowest total carbon emissions.

2. The stacker crane multi-machine collaborative scheduling method based on dynamic carbon footprint accounting according to claim 1, characterized in that, The construction of the energy consumption calculation model for a single handling operation of the stacker crane, to obtain the total electrical energy consumption generated by performing a single handling task, includes: A single handling task is broken down into three basic motion stages: horizontal movement, vertical lifting, and fork extension / retraction. Based on the structural parameters, motor performance, and motion control characteristics of the stacker crane, the theoretical energy consumption of each stage is estimated. The theoretical electrical energy consumption is calibrated by combining historical operating data or on-site measurement results; The energy consumption after calibration in the three stages is summarized and output to obtain the total power consumption of this handling task.

3. The stacker crane multi-machine collaborative scheduling method based on dynamic carbon footprint accounting according to claim 1, characterized in that, The establishment of a minute-level dynamic grid carbon emission factor access mechanism, and the calculation of the carbon emission of a single transportation task based on the total power consumption and the dynamic grid carbon emission factor corresponding to the task plan execution time, includes: Access to minute-level regional power grid carbon emission factor data streams; Based on the planned execution time period of the task, the carbon emission factor at the corresponding time is matched and weighted according to the actual running time to obtain the effective carbon emission factor of the task. The total electrical energy consumption is multiplied by the effective carbon emission factor to calculate the carbon emissions of a single transport mission.

4. The stacker crane multi-machine collaborative scheduling method based on dynamic carbon footprint accounting according to claim 1, characterized in that, The simulation verification of the optimal carbon scheduling scheme includes: The carbon-optimal scheduling scheme is used as input data and imported into a digital twin simulation platform that integrates the stacker crane dynamic motion model, energy consumption model, and real-time power grid carbon emission factor. On the digital twin simulation platform, the system drives the virtual stacker crane to perform all handling tasks according to the scheduling instructions, calculates the running time and path energy consumption of each action in real time, and dynamically accumulates the total carbon emissions in combination with the carbon emission factor of the current period. At the same time, it detects whether there are path conflicts or resource contention between equipment. After the simulation is completed, the output will include the task completion time, equipment conflict records, total system energy consumption, and total carbon emissions. If all indicators meet the preset constraints, the scheme is deemed feasible and proceeds to the actual execution stage; otherwise, the bottleneck information from the simulation feedback is extracted to correct the model parameters or adjust the constraints, and the improved carbon-optimal scheduling scheme is re-solved.

5. The stacker crane multi-machine collaborative scheduling method based on dynamic carbon footprint accounting according to claim 1, characterized in that, After successful simulation verification, the optimal carbon scheduling scheme is converted into control commands and issued to the stacker crane for execution. During execution, real-time data on actual energy consumption of the equipment is collected, and the actual carbon emissions are calculated based on the dynamic grid carbon emission factor. Finally, the actual carbon emissions are fed back into the energy consumption calculation model and scheduling solution strategy for dynamic model calibration and continuous algorithm optimization, including: The verified optimal carbon scheduling scheme is converted into control commands and sent to the stacker crane control system for execution. During the execution process, the actual energy consumption data and operating status data of the equipment are collected in real time through the Internet of Things; By combining the dynamic carbon emission factor, actual energy consumption, and operational status data at the time of mission execution, the actual carbon emissions of each mission are calculated. The actual carbon emissions are fed back to the energy consumption calculation model and scheduling solution strategy to calibrate model parameters and optimize algorithm logic, thereby achieving continuous adaptive improvement of the system.

6. A stacker crane multi-machine collaborative scheduling device based on dynamic carbon footprint accounting, based on the method of any one of claims 1 to 5, characterized in that, include: A data acquisition unit is constructed to build an energy consumption calculation model for a single handling operation of a stacker crane, so as to obtain the total electrical energy consumption generated by performing a single handling task; A calculation unit is established to establish a minute-level dynamic grid carbon emission factor access mechanism, and to calculate the carbon emission of a single transportation task based on the total power consumption and the dynamic grid carbon emission factor corresponding to the task plan execution time. A construction unit is used to sum the carbon emissions of all the handling tasks to be performed as the total carbon emission cost of the system, and to construct a multi-lane multi-stacking crane collaborative scheduling optimization model with the goal of minimizing the total carbon emission cost of the system. The design solver unit is used to design a two-layer collaborative scheduling solution strategy for storage location and scheduling, and to solve the optimization model to generate a carbon-optimal scheduling scheme. The simulation unit is used to simulate and verify the carbon optimal scheduling scheme. The execution feedback unit is used to convert the carbon optimal scheduling scheme into control commands and send them to the stacker crane for execution after the simulation verification is passed; During execution, the actual energy consumption data of the equipment is collected in real time, and the actual carbon emissions are calculated based on the dynamic power grid carbon emission factor. Finally, the actual carbon emissions are fed back to the energy consumption calculation model and scheduling solution strategy for dynamic calibration of the model and continuous optimization of the algorithm. The building unit includes: The module is used to estimate the carbon emissions of each handling task to be executed on different selectable execution time windows and different available stacker cranes, and use this as a carbon cost indicator for the task in scheduling decisions. The configuration module is used to set the total carbon emission cost of the system as the sum of the carbon cost indicators allocated to all tasks, and to set the optimization objective of the scheduling scheme as minimizing the total carbon emission cost of the system. The module introduces scheduling feasibility constraints, including: each stacker crane can only execute one task at a time, aisle access is mutually exclusive, tasks must be completed before their deadline, and equipment acceleration and travel capacity limits. The module is used to build a multi-lane, multi-stall crane collaborative scheduling optimization model that includes three-dimensional decision variables of task, equipment, and time, based on optimization objectives and constraints. The design solution unit includes: The prediction module is used at the upper level to predict the frequency of goods entering and leaving the warehouse, and to prioritize the allocation of high-frequency goods to low-energy storage locations near the entrance and exit, thus forming a low-carbon storage location layout. The determination module is used to take the low-carbon warehouse layout as input to the lower-level scheduling, determine the start and end points of each handling task, and calculate the carbon emission data of the task under different stacker cranes and execution periods. The solution module is used to solve the feasible scheduling scheme with the lowest total carbon emissions based on carbon emission data, under the premise of meeting equipment resource constraints and task timeliness requirements; The output module is used to output a carbon-optimal scheduling scheme, including task-equipment allocation, execution order, and timing arrangement, based on the feasible scheduling scheme with the lowest total carbon emissions.

7. A computer device, characterized in that, The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 5.

8. A storage medium, characterized in that, The storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 5.

Images