Multi-floor crown block dynamic scheduling method and system based on pressure balance algorithm

Abstract

The application discloses a multi-floor crane dynamic scheduling method and system based on a pressure balance algorithm, relates to the technical field of industrial automation, and comprises the following steps: acquiring state data of a target floor, forming a global state vector, and outputting a pressure field of the target floor; calculating pressure differences between floors according to a distribution result of the pressure field, combining a trend value to construct a migration driving force matrix, and outputting a migration trend value; determining the state of the floor according to the migration trend value, and dynamically adjusting a scheduling route and task allocation of the crane; and the application realizes multi-floor crane dynamic optimization and load balancing through pressure balance scheduling.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial automation, more particularly, the present application relates to a multi-floor crane dynamic scheduling method and system based on a pressure balance algorithm. BACKGROUND

[0002] With the development of modern logistics, intelligent warehousing and automation production, multi-floor warehousing systems have gradually become an important means to improve space utilization and operational efficiency. In these systems, cranes or stackers serve as the main handling equipment, taking on the task of transporting and handling goods between different floors. However, as the scale of the warehouse and the amount of tasks increase, how to efficiently and reliably schedule multi-floor cranes has become a core problem that needs to be solved in the warehouse management system.

[0003] For example, the invention patent with publication number CN120097170A discloses an intelligent building elevator traffic scheduling method and system for energy optimization. The method includes collecting real-time data of building sensor devices and elevator monitoring systems, constructing a prediction model to obtain the traffic heat distribution of future floors, designing a scheduling strategy based on multi-objective reinforcement learning, including target floor allocation and dynamic running path of elevators, formulating an energy consumption optimization execution control strategy to minimize global energy consumption, and generating optimized control instructions. Through the deep integration of traffic prediction, scheduling optimization and energy consumption management, the system adaptability, flexibility and energy efficiency are improved, and the passenger experience is significantly improved, providing an intelligent and efficient elevator scheduling solution for modern high-rise buildings.

[0004] For example, the invention patent with publication number CN120672063A discloses a production line load balancing method based on an intelligent optimization algorithm. The method includes the following steps: S1, real-time collection of data from each station of the production line and data validity verification; S2, multi-dimensional correlation analysis of real-time dynamic data sets; S3, multi-scale feature extraction through the Transformer self-attention mechanism; S4, real-time optimization of target weight coefficients through dynamic self-adaptive adjustment; S5, iterative optimization of production task loads of each station through the whale optimization algorithm, and real-time generation of adjustment sequences for task allocation of each station; S6, output of scheduling instructions through task load active exchange strategies between stations; S7, dynamic regulation of production tasks of each station according to the scheduling instructions, forming a real-time closed-loop feedback production line load dynamic balancing system. Through the adaptive adjustment and real-time feedback correction mechanism, production efficiency and equipment utilization are improved, and energy consumption is reduced.

[0005] In the above-mentioned technical solutions, at least the following technical problems exist:

[0006] The traditional multi-floor crane scheduling method is usually based on fixed rules, lacks dynamic response to real-time floor state, and the traditional method usually only relies on static task amount or experience rule to judge the floor state, cannot quantify the migration, migration or stable state of the floor, and cannot systematically consider the task migration trend and pressure difference between floors, so that the scheduling decision is easy to deviate from the optimal state.

[0007] In view of the above problems, the present application provides a solution. SUMMARY

[0008] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a multi-floor crane dynamic scheduling method and system based on pressure balance algorithm, which realizes multi-floor crane dynamic optimization and load balancing through pressure balance scheduling.

[0009] In order to achieve the above object, the present application provides the following technical scheme:

[0010] The multi-floor crane dynamic scheduling method based on pressure balance algorithm comprises: acquiring state data of target floors, forming a global state vector, and outputting a pressure field of the target floors; calculating the pressure difference between each floor according to the distribution result of the pressure field, combining the trend value to construct a migration driving force matrix, and outputting the migration trend value; determining the state of the floor according to the migration trend value, and dynamically adjusting the scheduling route and task allocation of the crane.

[0011] In a preferred embodiment, the acquisition of the state data of the target floor, the formation of the global state vector, and the output of the pressure field of the target floor are as follows: acquiring the state data of the target floor, and preprocessing the state data to form a global state vector; performing time window normalization processing on the global state vector, and generating a weighted state set with sampling credibility as the weight; constructing a multi-objective comprehensive pressure function according to the weighted state set, and outputting the static comprehensive pressure value of each floor to form a static pressure set; mapping the static pressure set to a virtual fluid space to form a virtual fluid initial state; abstracting the transmission process of the task between floors into a pressure diffusion phenomenon based on the virtual fluid initial state, and establishing a discrete Laplace dynamic equation; solving the discrete Laplace dynamic equation by a fourth-order Runge-Kutta numerical integration method, outputting a continuous time pressure evolution sequence and a pressure difference between floors; and collecting the continuous time pressure evolution sequences of all floors to form a pressure field under continuous time.

[0012] In a preferred embodiment, the pressure difference between each floor is calculated according to the distribution result of the pressure field, a migration driving force matrix is constructed in combination with the trend value, and a migration trend value is output, specifically as follows: according to the distribution result of the pressure field, the pressure difference between each floor is calculated; the historical migration amount, migration frequency and load proportion of each floor are obtained, a migration trend vector is constructed, and a dynamic trend value is generated through an exponential smoothing method; a migration driving force matrix is constructed according to the trend value in combination with the pressure difference between each floor; the migration driving force matrix is mapped to a migration amount through a nonlinear mapping function, and a migration trend value is output according to the migration amount; according to the change direction of the migration trend value within a continuous time step, a trend stability score is output; a time series of the migration trend value is analyzed for volatility, and a standard deviation is output as a volatility score; the trend stability score and the volatility score are comprehensively analyzed through an analytic hierarchy process to form a confidence level; the migration trend value and the confidence level are used to determine the migration-in, migration-out and stable states of the floors, and the dispatching route and task allocation of the head sheave are dynamically adjusted.

[0013] In a preferred embodiment, the migration trend value is output according to the migration amount, specifically as follows: the migration amount matrix between floors is obtained within each time step, and the net migration amount of each floor is output; the migration trend component is extracted according to the net migration amount; each trend component is dynamically normalized based on historical statistics of a rolling window, and the normalized trend component is weighted and synthesized with a preset weight to obtain an original trend score; the original trend score is converted to a migration trend value through a smoothing mapping function.

[0014] In a preferred embodiment, the state of the floor is determined according to the migration trend value, and the dispatching route and task allocation of the head sheave are dynamically adjusted, specifically as follows: a task migration demand model is constructed according to the floor state determination result, and the migration-in and migration-out demands between floors are output; the migration-in and migration-out demands between floors are matched and calculated, the task migration amount between floors is output, a task migration matrix is constructed, and the task migration matrix is dynamically corrected through a nonlinear mapping function to obtain a corrected effective migration matrix; the task interaction relationship of each floor is obtained according to the effective migration matrix, and the task density, priority and corresponding head sheave travel distance of each migration path are calculated; a head sheave task allocation sequence is generated according to the task density and priority of the migration path, and a scheduling operation is performed according to the head sheave task allocation sequence; during the scheduling execution process, the running state, task completion rate and floor task load change of each head sheave are collected in real time, and an actual task execution sequence is generated; the actual task execution sequence is compared with the predetermined head sheave task allocation sequence to obtain a scheduling deviation sequence; the update parameters of the migration trend value are recalibrated according to the scheduling deviation sequence, and the task migration matrix is re-output according to the corrected migration trend value and the latest pressure difference data.

[0015] In a preferred embodiment, the task interaction relationship of each floor is obtained according to the effective migration matrix, and the task density, priority and corresponding crane travel distance of each migration path are calculated, specifically as follows: according to the numerical value of each element in the effective migration matrix, the task density on each migration path is calculated, and the task density is normalized to generate a path load value; according to the task density and the path load value, the path priority is output; the target floor space coordinate data is obtained, the crane travel distance matrix is calculated according to the actual distance of the floor; the path load value and the crane travel distance matrix are comprehensively calculated based on the TOPSIS multi-objective approximation method to generate a path scheduling complexity score; according to the path scheduling complexity score, the priority of all migration paths is sorted to obtain a path priority execution sequence.

[0016] The system of the multi-floor crane dynamic scheduling method based on the pressure balance algorithm comprises a pressure field module, a migration trend value module and an adjustment module, and there is a connection between the modules; the pressure field module is used for obtaining state data of target floors, forming a global state vector and outputting a pressure field of the target floors; the migration trend value module is used for calculating pressure differences between floors according to a distribution result of the pressure field, combining trend values to construct a migration driving force matrix and outputting migration trend values; and the adjustment module is used for judging the state of the floors according to the migration trend values, and dynamically adjusting the scheduling route and task allocation of the crane.

[0017] The technical effects and advantages of the multi-floor crane dynamic scheduling method and system based on the pressure balance algorithm are as follows:

[0018] 1. The present application maps multi-dimensional indexes such as task quantity, task waiting time, track occupancy rate and energy consumption intensity into floor pressure values by constructing a multi-objective comprehensive pressure function, and simulates the task transmission process between floors by using a virtual fluid model and a discrete Laplace dynamic equation to form a pressure field under continuous time. This pressure field modeling method can accurately depict the load difference and energy transmission relationship between floors, solve the problem of isolated task state and insufficient coupling in traditional scheduling algorithms, and provide a measurable and evolving dynamic load evaluation basis for multi-floor systems.

[0019] 2. The present application realizes the dynamic balance of pressure gradient driving and historical trend inertia by nonlinearly weighting and coupling the pressure difference matrix between floors and the time evolution trend value matrix. This mechanism can quickly respond to changes in pressure difference when the task load mutates, and adaptively enhance the trend smoothness in the system stabilization stage, thereby effectively suppressing migration oscillation and scheduling delay, and improving the dynamic stability and prediction accuracy of the system. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 It is a flowchart of the multi-floor crane dynamic scheduling method based on the pressure balance algorithm of the present application.

[0021] Figure 2 The system structure schematic diagram of the multi-floor crane dynamic scheduling method based on the pressure balance algorithm of the application is shown in the figure. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the application will be clearly and completely described with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only a part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the application.

[0023] Embodiment 1, Figure 1 The multi-floor crane dynamic scheduling method based on the pressure balance algorithm of the application is given, including the following steps:

[0024] S1, obtaining the state data of a target floor, forming a global state vector, and outputting the pressure field of the target floor;

[0025] In this embodiment, the state data of the target floor is obtained, a global state vector is formed, and the pressure field of the target floor is outputted, specifically as follows:

[0026] The state data of the target floor is obtained, the time error of data sampling of each floor is ensured to be within a threshold interval through a clock alignment algorithm, the detected abnormal data is removed using the three-sigma criterion, the missing values are interpolated using the moving average method, and a global state vector is formed. The state data includes the number of tasks, task waiting time, track occupancy rate, crane energy consumption and running position;

[0027] The global state vector is normalized by a time window, and a weighted state set is formed by taking the sampling credibility as a weight coefficient;

[0028] According to the weighted state set, a multi-objective comprehensive pressure function is constructed, and the static comprehensive pressure value of each floor is outputted to form a static pressure set. The multi-objective comprehensive pressure function integrates four types of indexes including task density, waiting time, track utilization rate and energy consumption intensity into a single pressure value, and the weights of the indexes are determined by a multi-objective optimization algorithm (such as a particle swarm or a genetic algorithm);

[0029] The static pressure set is mapped to a virtual fluid space to form an initial state of the virtual fluid;

[0030] Based on the initial state of the virtual fluid, the task transmission process between floors is abstracted as a pressure diffusion phenomenon, and a discrete Laplace dynamic equation is established;

[0031] The discrete Laplace dynamic equation is solved by a fourth-order Runge-Kutta numerical integration method, and a continuous time pressure evolution sequence and a pressure difference between each floor are output.

[0032] The continuous time pressure evolution sequence of all floors is summarized to form a pressure field under continuous time.

[0033] The multi-objective comprehensive pressure function is specifically as follows:

[0034]

[0035] The mapping of the static pressure set to the virtual fluid space is specifically as follows:

[0036]

[0037] In the formula: is the initial pressure value of the i-th floor, is a weight coefficient with a sampling credibility, is the number of task or job requests of the i-th floor, is the power consumption per unit time of the i-th floor equipment group, represents the proportion of the occupied area of the i-th floor, represents the overall energy consumption intensity of the i-th floor, , , and is a static weight coefficient, representing the contribution proportion of the task, power, occupancy rate and energy consumption to the floor pressure, is the initial state of the virtual fluid, represents the geometric coordinates of the i-th floor in the building structure, is a topological mapping function, projecting the floor pressure data to the virtual fluid model coordinate system.

[0038] The discrete Laplace dynamic equation is specifically as follows:

[0039]

[0040] The fourth-order Runge-Kutta numerical integration method for solving the discrete Laplace dynamic equation is specifically as follows:

[0041]

[0042] In the formula: represents the initial pressure value of the i-th floor, represents the current time t, is a proportional constant of the task or energy transferable rate between floors, , denotes a set of floors directly connected with the i-th floor or having task interaction, is a disturbance term, representing the influence of burst tasks, random noise or external interference on the stress, denotes the stress at the next time instant, is the time step, denotes the stress value of the i-th floor at time t.

[0043] S2, according to the distribution result of the stress field, calculates the stress difference between each floor, combines the trend value to construct the migration driving force matrix, and outputs the migration trend value;

[0044] In this embodiment, according to the distribution result of the stress field, the stress difference between each floor is calculated, the trend value is combined to construct the migration driving force matrix, and the migration trend value is output, which is as follows:

[0045] According to the distribution result of the stress field, the stress difference value between any two floors is calculated, and a stress difference matrix is constructed;

[0046] The historical migration amount, migration frequency and load proportion of each floor are obtained, a migration trend vector is constructed, and a dynamic trend value is generated through an exponential smoothing method;

[0047] The stress difference matrix between each floor and the trend value matrix are coupled according to a nonlinear weighting model, so as to form a migration driving force matrix;

[0048] The migration driving force matrix is mapped to the migration amount through a nonlinear mapping function, and the migration trend value is output according to the migration amount;

[0049] According to the change direction of the migration trend value in the continuous time step, it is judged whether the trend symbol is consistent, and a trend stability score is output;

[0050] The time series of the migration trend value is analyzed for volatility, and a standard deviation is output as a volatility score;

[0051] The trend stability score and the volatility score are integrated through an analytic hierarchy process to form a confidence level;

[0052] According to the migration trend value and the confidence level, the migration, migration and stable state of the floor are determined, the dispatching route and task allocation of the head sheave are dynamically adjusted, and the real-time balance of the task load of each floor is realized;

[0053] If the trend value is higher than the upper threshold and the confidence level meets the minimum requirement, and the duration reaches the set number of steps, it is in the migration state;

[0054] If the trend value is lower than the lower threshold and the confidence level meets the minimum requirement, and the duration reaches the set number of steps, it is in the migration state;

[0055] If the floor does not meet the migration or migration conditions, it is a stable state.

[0056] In this embodiment, the migration trend value is output according to the migration amount, specifically as follows:

[0057] In each time step, the migration amount matrix between floors is obtained, and the net migration amount of each floor (the total migration amount minus the total migration amount) is calculated;

[0058] According to the net migration amount, the migration trend component is extracted, including but not limited to: migration direction, amplitude, migration event frequency component, average migration rate component and fluctuation degree component;

[0059] Each trend component is dynamically normalized based on rolling window historical statistics, and the normalized trend component is weighted and synthesized with a preset weight to obtain an original trend score;

[0060] The original trend score is converted into a migration trend value by a smoothing mapping function (such as Sigmoid), and the numerical range is normalized between 0 and 1.

[0061] S3, according to the migration trend value, the state of the floor is determined, and the dispatching route and task allocation of the overhead crane are dynamically adjusted;

[0062] In this embodiment, according to the migration trend value, the state of the floor is determined, and the dispatching route and task allocation of the overhead crane are dynamically adjusted, specifically as follows:

[0063] According to the floor state determination result, a task migration demand model is constructed, including:

[0064] For the migration-out floor, a task migration-out demand value is generated to determine the releasable task capacity;

[0065] For the migration-in floor, a task receiving demand value is generated to determine the upper limit of the task that can be carried;

[0066] For the stable floor, the current task load is maintained or the task allocation is fine-tuned according to the state of the adjacent layer;

[0067] The migration-out and migration-in demands between floors are matched and calculated to output the task migration amount between floors, construct a task migration matrix, and dynamically correct the task migration matrix through a nonlinear mapping function to obtain a corrected effective migration matrix;

[0068] According to the effective migration matrix, the task interaction relationship of each floor is obtained, and the task density, priority and corresponding overhead crane travel distance of each migration path are calculated;

[0069] According to the task density and priority of the migration path, a task allocation sequence of the overhead crane is generated, so that the overhead crane preferentially performs high-weight migration tasks in the adjacent floors in the direction of the migration trend, so as to reduce empty running and waiting time;

[0070] According to the task allocation sequence of the overhead crane, a scheduling operation is performed, and the overhead crane is controlled to complete the tasks in sequence according to the migration direction and priority determined in the sequence, so as to realize real-time balancing of multi-floor task loads;

[0071] During the scheduling execution process, the running state, task completion rate and floor task load change of each overhead crane are collected in real time, and an actual task execution sequence is generated;

[0072] The actual task execution sequence is compared with the predetermined task allocation sequence of the overhead crane to obtain a scheduling deviation sequence;

[0073] According to the scheduling deviation sequence, the update parameters of the migration trend value are recalibrated, and a task migration matrix is re-output based on the corrected migration trend value and the latest pressure difference data.

[0074] In this embodiment, according to the effective migration matrix, the task interaction relationship of each floor is obtained, and the task density, priority and corresponding overhead crane driving distance of each migration path are calculated, as follows:

[0075] Non-zero elements are extracted from the effective migration matrix to identify the floor pair relationship with task flow, and a task interaction graph between floors is constructed, in which each directed edge represents the task migration relationship from floor i to floor j;

[0076] The task interaction graph is topologically analyzed, and the in-degree and out-degree of each floor are counted, which are used to measure the role strength of the floor as a task convergence point or distribution point;

[0077] According to the numerical value of each element in the effective migration matrix, the task density on each migration path is calculated, which is used to reflect the concentration degree of task migration per unit time on the path, and the task density is normalized by combining the spatial distance between floors, the task type and the historical scheduling efficiency, to generate a path load value, so as to maintain comparability in subsequent calculations;

[0078] According to the task density and the path load value, the path priority is output;

[0079] The spatial coordinate data of the target floor is obtained, and the overhead crane driving distance matrix is calculated according to the actual distance of the floor, wherein the matrix element represents the shortest feasible distance of the overhead crane from the starting floor to the target floor;

[0080] Based on the TOPSIS multi-objective approximation method, the path load value and the overhead crane driving distance matrix are comprehensively calculated to generate a path scheduling complexity score;

[0081] According to the path scheduling complexity score, all migration paths are prioritized to obtain a path priority execution sequence.

[0082] Embodiment 2, Figure 2 The system of the multi-floor crane dynamic scheduling method based on the pressure balance algorithm is characterized in that it comprises a pressure field module, a migration trend value module and an adjustment module, and there is a connection between the modules.

[0083] The pressure field module is configured to obtain state data of a target floor, form a global state vector, and output a pressure field of the target floor.

[0084] The migration trend value module is configured to calculate a pressure difference between floors according to a distribution result of the pressure field, construct a migration driving force matrix in combination with a trend value, and output a migration trend value.

[0085] The adjustment module is configured to determine the state of the floor according to the migration trend value, and dynamically adjust the scheduling route and task allocation of the crane.

[0086] The above formulas are all dimensionless values, and the formulas are obtained by collecting a large amount of data to simulate a formula of the nearest real situation, and the preset parameters in the formula are set by a person skilled in the art according to the actual situation.

[0087] The above embodiments can be realized wholly or partially by software, hardware, firmware or any combination thereof. When realized by software, the above embodiments can be realized in the form of a computer program product wholly or partially.

[0088] Those skilled in the art can realize that the modules and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized by hardware or software depends on the specific application and design constraints of the technical solutions. A person skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0089] In addition, each functional module in each embodiment of the present application can be integrated in one processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0090] The above is merely a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0091] Finally: the above only for the preferred embodiments of the present application, and not for limiting the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application, should be included in the scope of protection of the present application.

Claims

1. A dynamic scheduling method for multi-story overhead cranes based on a pressure balance algorithm, characterized in that, include: Acquire the state data of the target floor, form a global state vector, and output the pressure field of the target floor; The pressure difference between each floor is calculated based on the distribution of the pressure field. The migration driving force matrix is ​​constructed by combining the trend value and the migration trend value is output. The status of each floor is determined based on the migration trend value, and the scheduling route and task allocation of the overhead crane are dynamically adjusted. The pressure difference between each floor is calculated based on the distribution of the pressure field, and a migration driving force matrix is ​​constructed by combining the trend value. The migration trend value is then output, as detailed below: Calculate the pressure difference between each floor based on the pressure field distribution results; Obtain the historical migration volume, migration frequency, and load percentage for each floor, construct a migration trend vector, and generate dynamic trend values ​​using an exponential smoothing method; Based on the trend values ​​and the pressure differences between floors, a migration driving force matrix is ​​constructed; The migration driving force matrix is ​​mapped to migration quantity through a nonlinear mapping function; At each time step, obtain the migration matrix between floors and output the net migration of each floor; Extract migration trend components based on net migration; Each trend component is dynamically normalized based on historical statistics from a scrolling window, and the normalized trend components are weighted and synthesized with preset weights to obtain the original trend score. The original trend score is converted into a migration trend value using a smoothing mapping function; Output the trend stability score based on the direction of change of the migration trend value over consecutive time steps; Volatility analysis is performed on the time series of migration trend values, and the standard deviation is output as a volatility score. The trend stability score and volatility score are combined using the analytic hierarchy process to form a confidence level. The migration trend value and confidence level determine the inflow, outflow and stable status of the floors, and dynamically adjust the scheduling route and task allocation of the overhead cranes. The process of determining the floor status based on migration trend values ​​and dynamically adjusting the overhead crane's scheduling route and task allocation is as follows: Based on the floor status determination results, construct a task migration requirement model and output the migration out and migration in requirements between each floor. The outbound and inbound requirements of each floor are matched and calculated, the task migration amount between floors is output, the task migration matrix is ​​constructed, and the task migration matrix is ​​dynamically corrected by a non-linear mapping function to obtain the corrected effective migration matrix. Based on the effective migration matrix, obtain the task interaction relationship of each floor, and calculate the task density, priority and corresponding crane travel distance of each migration path; Based on the task density and priority of the migration path, generate the crane task allocation sequence, and execute the scheduling operation according to the crane task allocation sequence; During the scheduling and execution process, the operating status of each crane, task completion rate, and floor task load changes are collected in real time, and the actual task execution sequence is generated. The actual task execution sequence is compared with the predetermined overhead crane task allocation sequence to obtain the scheduling deviation sequence; Based on the scheduling deviation sequence, the update parameters of the migration trend value are recalibrated, and the task migration matrix is ​​re-output based on the corrected migration trend value and the latest pressure difference data.

2. The multi-story crane dynamic scheduling method based on the pressure balance algorithm according to claim 1, characterized in that, The process of acquiring the state data of the target floor, forming a global state vector, and outputting the pressure field of the target floor is as follows: Acquire the status data of the target floor and preprocess the status data to form a global status vector; The global state vector is normalized by a time window, and a weighted state set is generated by using the sampling confidence as the weight. Based on the weighted state set, a multi-objective comprehensive pressure function is constructed, and the static comprehensive pressure value of each floor is output to form a static pressure set; The static pressure set is mapped to the virtual fluid space to form the initial state of the virtual fluid. Based on the initial state of the virtual fluid, the process of transferring tasks between floors is abstracted as a pressure diffusion phenomenon, and a discrete Laplace dynamic equation is established. The discrete Laplace dynamic equations are solved using the fourth-order Runge-Kutta numerical integration method, outputting a continuous-time pressure evolution sequence and pressure differences between floors. The continuous-time pressure evolution sequences of all floors are summarized to form a pressure field under continuous time.

3. The multi-story overhead crane dynamic scheduling method based on the pressure balance algorithm according to claim 2, characterized in that, The process involves obtaining the task interaction relationships for each floor based on the effective migration matrix, and calculating the task density, priority, and corresponding crane travel distance for each migration path, as detailed below: Based on the numerical values ​​of each element in the effective migration matrix, the task density on each migration path is calculated, and the task density is normalized to generate the path load value. Output path priority based on task density and path load value; Obtain the spatial coordinate data of the target floor and calculate the crane travel distance matrix based on the actual distance between the floors; Based on the TOPSIS multi-objective approximation method, the path load value and the crane travel distance matrix are comprehensively calculated to generate a path scheduling complexity score. Based on the path scheduling complexity score, all migration paths are prioritized to obtain the path priority execution sequence.

4. A system using the multi-story overhead crane dynamic scheduling method based on the pressure balance algorithm as described in any one of claims 1-3, characterized in that, It includes a pressure field module, a migration trend value module, and an adjustment module, and the modules are interconnected. The pressure field module is used to acquire the state data of the target floor, form a global state vector, and output the pressure field of the target floor. The migration trend value module is used to calculate the pressure difference between each floor based on the distribution results of the pressure field, construct the migration driving force matrix by combining the trend value, and output the migration trend value. The adjustment module is used to determine the status of the floors based on the migration trend value and dynamically adjust the scheduling route and task allocation of the overhead crane.

Images