Intelligent warehouse energy consumption collaborative optimization method and system based on digital twinning

Abstract

This invention relates to the field of warehouse management technology, specifically to a smart warehouse energy consumption collaborative optimization method and system based on digital twins. The method includes the following steps: extracting humidity deviations to obtain a compensation temperature and stripping away precipitated water mass to obtain the true density; comparing the densities of adjacent nodes to generate an abnormal buoyancy warning vector; combining the parameters of the guided vehicle to obtain the water film thickness and establish a global energy consumption optimization trajectory; extracting the enthalpy value of the refrigeration group to obtain a tolerance drift heat assessment item; and updating the load based on the tolerance drift heat to generate an energy consumption collaborative power flow state. In this invention, the equivalent compensation temperature is obtained by analyzing temperature and humidity deviations to update the true density; a buoyancy warning is generated based on the node density comparison and associated with the surface water film thickness; the linear velocity of physical movement is calculated to establish a global optimization trajectory to avoid ineffective losses; the tolerance heat is assessed in conjunction with the spatial enthalpy to reduce the microgrid power; and the transient load data is updated to generate a collaborative power flow state to achieve dynamic energy balance.

Description

Technical Field

[0001] This invention relates to the field of warehouse management technology, and in particular to a method and system for collaborative optimization of energy consumption in smart warehousing based on digital twins. Background Technology

[0002] Warehouse management technology involves the storage, scheduling, and circulation of materials within the supply chain. This field includes specific operations such as receiving, inventory management, order sorting, and outbound delivery. It establishes a full lifecycle material ledger and physical tracking mechanism by comprehensively managing personnel, equipment, and space resources within the warehouse. Traditional smart warehousing energy consumption optimization methods refer to the scheduling and operation of power consumption for energy-consuming components such as temperature control equipment, lighting facilities, and logistics handling equipment. Traditional approaches typically involve setting the operating temperature and start / stop times of air conditioners or fans in different time periods, combined with manual on-site inspections or timed circuit switches to control the on / off status of lighting circuits in various warehouse zones. For handling equipment such as stacker cranes and automated guided vehicles, movement commands are issued according to the order of received inbound and outbound orders or pre-set fixed physical guidance routes in the workshop, and a fixed schedule is implemented for rotating charging plans.

[0003] Traditional warehouse management models employ segmented temperature control equipment start / stop settings and manual inspection to control lighting. Handling equipment follows a fixed sequence and physical route to execute movement and shift charging plans. This rigid operating mechanism lacks dynamic response capabilities when environmental parameters change abruptly, making it difficult to cope with the additional energy consumption caused by temperature and humidity fluctuations. Fixed routes make it easy for mobile equipment to generate ineffective energy consumption behavior on complex terrain. Manual inspection and timed control suffer from information lag, making it impossible to achieve coordinated energy supply and demand scheduling at the microgrid level. This can easily lead to excessive energy consumption in the overall warehouse and a decline in scheduling efficiency. Summary of the Invention

[0004] To address the technical problems existing in the prior art, this invention provides a smart warehouse energy consumption collaborative optimization method based on digital twins, comprising the following steps: S1: Collect the dry bulb and dew point temperature values ​​of the storage temperature and humidity sensor. When the dew point temperature is lower than the dry bulb temperature, extract the humidity deviation. Use the preset temperature compensation mapping model to convert the deviation to the latent heat constant and obtain the node equivalent compensation temperature set. S2: Call the node equivalent compensation temperature set to convert the theoretical density, strip the equivalent mass density of water precipitated from the theoretical density, collect the density of adjacent nodes in the physical storage space and compare it with the real density, and when the density of adjacent nodes is greater than the real density, combine the density difference with the product of the gravity constant to extract the acceleration conversion amount and generate an abnormal buoyancy warning vector. S3: Based on the anomalous buoyancy warning vector, extract the equivalent vertical lift factor in the vector, and use the equivalent vertical lift factor to correct the wheel-ground normal pressure of the physical mobile guided vehicle. Collect the ground surface temperature and dew point temperature of the physical mobile guided vehicle. In the over-limit state, combine the condensation conversion rate to obtain the water film thickness. Extract the wheel angular velocity to convert the theoretical linear velocity and compare it with the chassis movement speed to establish a global energy consumption optimization trajectory. S4: Call the global energy consumption optimization trajectory, collect the active power flow values ​​of the physical microgrid branches, compare the active power flow values ​​with the thermal stability limit values, extract the spatial enthalpy values ​​of the physical refrigeration unit in the over-limit state, compare the critical enthalpy values ​​of fresh food with the spatial enthalpy values, and obtain the tolerance drift heat assessment item. S5: Call the tolerance drift heat assessment item, combine specific heat capacity, mass flow rate and control duration to calculate the power reduction, remove the active load to reduce the power and update the transient active load, combine the power factor to calculate the transient reactive load, and generate the energy consumption coordinated power flow state of the warehouse microgrid.

[0005] As a further aspect of the present invention, the node equivalent compensation temperature set includes sensible heat correction index, vaporization phase change energy consumption benchmark, and virtual temperature scale difference value; the anomalous buoyancy early warning vector includes spatial convection mutation mode length, upward turbulence direction angle, and air mass instability gradient; the global energy consumption optimization trajectory includes anti-slip travel curvature, traction force extreme value coordinate point, and endurance enhancement posture sequence; the tolerance drift heat assessment item includes storage capacity insulation margin, goods preservation buffer zone, and quality loss-free temperature rise range; and the energy consumption collaborative power flow state of the warehouse microgrid includes node voltage phasor distribution, feeder network loss matrix, and source-storage mutual balance degree.

[0006] As a further aspect of the present invention, the specific steps of S1 are as follows: S101: Collects dry bulb and dew point temperature values ​​from warehouse temperature and humidity sensors, performs difference calculations on the dry bulb and dew point temperature values ​​for comparison, and extracts the temperature difference deviation obtained from the subtraction operation when the dew point temperature does not reach the dry bulb temperature limit. Then, performs a serialization splicing operation on the temperature difference deviation to generate a humidity deviation feature vector. S102: Call the humidity deviation feature vector, monitor the latent heat constant value associated with the regional environment, perform multiplication operation on each vector element and latent heat constant value in the humidity deviation feature vector, extract the potential energy compensation mapping value obtained by multiplication operation, and arrange all potential energy compensation mapping values ​​in an array according to the sensor spatial node position to obtain the humidity latent heat conversion matrix. S103: Perform mean smoothing operation on each conversion element in the humidity latent heat conversion matrix, extract the compensation benchmark value, convert the compensation benchmark value into a temperature scale correction amount through a preset equivalent conversion coefficient, and perform an accumulation operation on the original dry bulb temperature value and the temperature scale correction amount to obtain the single node compensation value. Collect all single node compensation values ​​according to the corresponding relationship of the nodes to obtain the node equivalent compensation temperature set.

[0007] As a further aspect of the present invention, the specific steps of S2 are as follows: S201: Call the equivalent compensation temperature set of the node to convert the theoretical density, collect the preset precipitated water mass parameter, convert the precipitated water mass parameter into the equivalent density of precipitated water of the corresponding volume, perform a subtraction operation on the theoretical density and the equivalent density of precipitated water, extract the scalar density residual value obtained by the subtraction operation, and perform a matrix-based recombination and arrangement operation on the associated density residual values ​​of all nodes according to the grid coordinates to obtain the true density matrix of the nodes; S202: Collect the density of adjacent nodes, call the true density matrix of the nodes, perform a size comparison operation between the density of adjacent nodes and the related elements in the true density matrix of the nodes, and extract the corresponding excess deviation value when the density of adjacent nodes exceeds the limit of the related elements. Perform a one-dimensional vector transformation and splicing operation on all excess deviation values ​​to obtain the density jump sequence. S203: Based on the density jump sequence, monitor the environmental gravitational acceleration constant, perform a multiplication operation on each out-of-limit deviation value in the density jump sequence and the gravitational acceleration constant, divide the result by the air reference density, extract the node acceleration conversion value obtained by the division operation, perform a mapping and arrangement operation on all node acceleration conversion values ​​according to the topological relationship, and generate an abnormal buoyancy warning vector.

[0008] As a further aspect of the present invention, the specific steps of S3 are as follows: S301: Based on the anomalous buoyancy warning vector, collect the vehicle surface temperature and dew point temperature, perform a subtraction operation on the vehicle surface temperature and dew point temperature, extract the deviation value when the difference exceeds the condensation limit, obtain the condensation conversion rate constant, perform a multiplication operation on the deviation value and the conversion rate constant, arrange the product sequence, and generate an anomalous water film thickness matrix. S302: Call the anomalous water film thickness matrix, collect the wheel angular velocity parameter and wheel diameter constant, perform product conversion on the two parameters to extract the theoretical linear velocity, detect the chassis running speed, perform a subtraction comparison operation on the linear velocity and the running speed to obtain the linear velocity difference, identify the offset term by mapping the linear velocity difference according to the coordinates in the thickness matrix, and obtain the slip loss feature sequence. S303: Based on the slip loss feature sequence, detect the energy consumption weight constant of the grid nodes, perform a weighted operation on the offset term and the energy consumption weight constant in the feature sequence to extract the node resistance cost, sort all costs in ascending order to extract the first and second order paths, connect the endpoints of the first and second paths according to the topological relationship, and establish a global energy consumption optimization trajectory.

[0009] As a further aspect of the present invention, the result of multiplying the deviation values ​​of each point by the condensation conversion rate constant is arranged into a product sequence. Based on the horizontal and vertical dimension data of the sampling grid, the product sequence is mapped and configured into a two-dimensional array, and the two-dimensional array is used as the anomalous water film thickness matrix.

[0010] As a further aspect of the present invention, the specific steps of S4 are as follows: S401: Call the global energy consumption optimization trajectory, collect the active power flow values ​​of the microgrid branches, detect the thermal stability limit values, perform difference calculation and comparison between the active power flow values ​​and the thermal stability limit values, determine that the active power flow values ​​are in an over-limit state, extract the over-limit deviation value of the power flow, map all the over-limit deviation values ​​of the power flow according to the node coordinates of the optimization trajectory, and generate the branch load overflow matrix. S402: Call the branch load overflow matrix, collect the spatial enthalpy value of the refrigeration unit group, retrieve the fresh food critical enthalpy value parameter, perform a subtraction comparison operation on the spatial enthalpy value and the fresh food critical enthalpy value, extract the corresponding difference result of the two values, and perform a weighted fusion operation on the node load status flag bit in the branch load overflow matrix to obtain the node enthalpy drift deviation set; S403: Based on the node enthalpy drift deviation set, collect the environmental heat transfer coefficient constant, perform product integration operation on the deviation element and the environmental heat transfer coefficient constant to obtain the heat fluctuation scalar, arrange all heat fluctuation scalars according to the time sequence step relationship, collect them to form a converged data attribute feature distribution item, and obtain the tolerance drift heat assessment item.

[0011] As a further aspect of the present invention, the initial difference is extracted by subtracting the thermal stability limit value from the active power flow value of the microgrid branch, and the initial difference is compared with the zero constant. When the initial difference is greater than the zero constant, the active power flow value of the microgrid branch is determined to be in an over-limit state, and the initial difference is used as the power flow over-limit deviation value.

[0012] As a further aspect of the present invention, the specific steps of S5 are as follows: S501: Call the tolerance drift heat assessment item, collect the air specific heat capacity constant and mass flow rate parameter, divide the heat fluctuation scalar in the tolerance drift heat assessment item by the assessment time step, and perform thermoelectric conversion logic operation in combination with the specific heat capacity constant and mass flow rate parameter, extract the corresponding cooling power adjustment offset value of each node, perform serialization summation and aggregation operation on all offset values, and arrange the aggregation result according to the branch index mapping to generate a thermoelectric power reduction matrix; S502: Call the thermoelectric power reduction matrix, collect the original active load values ​​of the physical branch, perform a stripping subtraction operation on the original active load values ​​and the corresponding power reduction elements in the thermoelectric power reduction matrix, extract the load difference data obtained by the subtraction operation, and perform a spatiotemporal dimension correlation mapping operation on all load difference data to obtain the transient active load distribution set. S503: Based on the transient active load distribution set, collect the branch power factor constant, perform positive switching multiplication logic operation on each load element and the power factor constant in the transient active load distribution set to obtain the transient reactive load scalar, and perform complex field vector synthesis operation on the transient reactive load scalar and the corresponding element in the transient active load distribution set to generate the energy consumption coordinated power flow state of the warehouse microgrid.

[0013] A smart warehouse energy consumption collaborative optimization system based on digital twins includes: The parameter mapping module collects the dry bulb and dew point temperature values ​​from the warehouse temperature and humidity sensor. When the dew point temperature is lower than the dry bulb temperature, it extracts the humidity deviation and uses a preset temperature compensation mapping model to convert the deviation to the latent heat constant, thereby obtaining the node equivalent compensation temperature set. The density analysis module calls the equivalent compensation temperature set of the nodes to convert the theoretical density, strips the equivalent mass density of the precipitated water from the theoretical density, collects the density of adjacent nodes in the physical storage space and compares it with the actual density. When the density of adjacent nodes is greater than the actual density, it combines the density difference with the product of the gravity constant to extract the acceleration conversion amount and generate an abnormal buoyancy warning vector. The optimization and deduction module extracts the equivalent vertical lift factor from the anomalous buoyancy warning vector and uses the equivalent vertical lift factor to correct the wheel-ground normal pressure of the physical mobile guided vehicle. It collects the ground surface temperature and dew point temperature of the physical mobile guided vehicle, and combines the condensation conversion rate to obtain the water film thickness in the over-limit state. It extracts the wheel angular velocity, converts it into theoretical linear velocity, and compares it with the chassis movement speed to establish a global energy consumption optimization trajectory. The enthalpy assessment module calls the global energy consumption optimization trajectory, collects the active power flow values ​​of the physical microgrid branches, compares the active power flow values ​​with the thermal stability limit values, extracts the spatial enthalpy values ​​of the physical refrigeration unit in the over-limit state, compares the critical enthalpy values ​​of fresh food with the spatial enthalpy values, and obtains the tolerance drift heat assessment item. The energy consumption coordination module calls the tolerance drift heat assessment item, combines specific heat capacity, mass flow rate and control duration to calculate the power reduction, removes active load to reduce power and updates transient active load, combines power factor to calculate transient reactive load, and generates the energy consumption coordination power flow state of the warehouse microgrid.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, the equivalent compensation temperature is obtained by extracting the environmental dry-bulb and dew point temperature values ​​and converting the humidity deviation. The theoretical density is then converted by deeply combining the compensation temperature and the mass of the precipitated water is stripped to obtain the true density. An abnormal buoyancy warning vector is generated by comparing the density of adjacent nodes in the physical space. This vector is then correlated with the surface temperature parameters to obtain the water film thickness and calculate the wheel linear velocity, thereby establishing a global energy consumption optimization trajectory to avoid ineffective slippage losses. Simultaneously, the active power flow of the microgrid is collected and compared with the thermal stability limit value. Combined with the spatial enthalpy of the refrigeration unit, the tolerable drift heat term is evaluated. Based on this, the power is reduced and the transient load data is updated, directly generating the energy consumption coordinated power flow state to achieve dynamic balance and fine scheduling of energy supply and demand in the storage facility. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is the main flowchart of the intelligent warehousing energy consumption collaborative optimization of the present invention; Figure 2 This is a schematic diagram of temperature and humidity compensation and node density mapping in this invention; Figure 3 This is a schematic diagram illustrating the generation of the abnormal buoyancy early warning vector in this invention; Figure 4 This is a schematic diagram of the water film sliding and energy consumption trajectory of the mobile guide vehicle of the present invention; Figure 5 This is a schematic diagram illustrating the enthalpy drift assessment and microgrid power flow coordination of the present invention. Figure 6 This is a block diagram of the intelligent warehousing energy consumption collaborative optimization system of the present invention. Detailed Implementation

[0017] The following embodiments further illustrate the intelligent warehousing energy consumption collaborative optimization method and system based on digital twins. The digital twin in this specification is used to establish a virtual-physical correspondence between the physical warehousing space, the operating status of the mobile guided vehicle, the enthalpy status of the refrigeration unit group, and the power flow status of the warehousing microgrid. This enables the data collected on the physical side to form traceable, callable, and collaboratively predictable state objects in the virtual mapping space. In subsequent embodiments, the "dry-bulb temperature value" is the temperature value collected by the warehousing temperature and humidity sensor and used to characterize the sensible heat state of the environment; the "dew point temperature value" is the dew point value collected within the same collection area to characterize the air humidity state; the "humidity deviation" is the deviation state quantity extracted from the difference between the dew point temperature and the dry-bulb temperature when the dew point temperature is lower than the dry-bulb temperature; and the "node equivalent compensation temperature set" is the node-level compensation temperature result formed based on the humidity deviation, latent heat constant, and a preset temperature compensation mapping model, which is used for subsequent theoretical density conversion. The equivalent compensation temperature set of nodes serves as the virtual temperature scale mapping result of the physical acquisition nodes in the digital twin space, ensuring that the humidity and heat state, density state, and subsequent energy consumption adjustment state of the same storage node can be transmitted along a unified data link. In the embodiment, the "abnormal buoyancy warning vector" characterizes the buoyancy anomaly direction, intensity, and instability state formed when the density of adjacent nodes exceeds the node's true density; the "global energy consumption optimization trajectory" characterizes the low-energy-consumption travel path formed by the mobile guide vehicle under the combined effects of water film, slippage, and node resistance costs; the "tolerance drift heat assessment term" characterizes the adjustable heat state of the refrigeration unit's spatial enthalpy relative to the critical enthalpy of fresh produce when the active power flow of the microgrid branch exceeds the limit; and the "storage microgrid energy consumption collaborative power flow state" characterizes the collaborative power flow result formed by transient active load, transient reactive load, and the source-storage mutual support relationship. All of the above state objects can be maintained in the digital twin mapping relationship according to the physical node, branch index, trajectory node, and time step relationship, avoiding data disconnection between the environmental side, vehicle side, refrigeration side, and grid side. The figure numbers and S-type step numbers appearing in this embodiment are only used for process and figure identification and do not constitute a range of parameter values, quantities, sequences, or performance limitations. This embodiment does not add any entity component numbers not configured in the original disclosure, and the system module names correspond only to the module responsibilities described in the claims.

[0018] Please see Figures 1 to 5This embodiment provides a smart warehouse energy consumption collaborative optimization method based on digital twins. This method is applied to the mapping relationship between the physical warehouse space and its digital twin, continuously associating temperature and humidity states, node density states, mobile guided vehicle operating states, refrigeration enthalpy states, and microgrid power flow states. This allows warehouse energy consumption regulation to no longer rely solely on a single refrigeration power or a single vehicle path, but rather to achieve collaborative optimization across the humid and hot environment, ground adhesion states, and branch load states. In this method, the digital twin is not an independent display model separate from the physical warehouse operation, but rather serves as a carrier of relationships between physical data collection, virtual node states, trajectory extrapolation results, and microgrid power flow update results, enabling the intermediate results generated in each step to be continuously invoked by subsequent steps.

[0019] S1: Collect dry-bulb and dew-point temperature values ​​from the warehouse temperature and humidity sensors. When the dew-point temperature is lower than the dry-bulb temperature, extract the humidity deviation. Use a preset temperature compensation mapping model to convert the deviation to the latent heat constant, obtaining the node equivalent compensation temperature set. During the digital twin mapping process, this node equivalent compensation temperature set is used to synchronize the humidity and heat acquisition status in the physical warehouse environment to the virtual node temperature scale status, thus providing a unified node input for subsequent density analysis.

[0020] In this step, the digital twin receives dry-bulb and dew-point temperature values ​​from various collection nodes in the physical storage space. After the dry-bulb and dew-point temperatures enter the comparison link of the same node, it first determines whether the two types of collected values ​​have a relationship based on the same source node, the same collection object, and a temporal correspondence. Collected values ​​that can be mapped to the same storage space node enter the humidity deviation extraction process; collected values ​​that cannot form a node correspondence are not used as the basis for generating the compensation temperature for that node, and their state is only retained as the collection state to be supplemented, without outputting new compensation results to subsequent density conversion steps. Therefore, the virtual nodes in the digital twin space will not receive temperature and humidity states with inconsistent sources or mismatched temporal sequences, ensuring that the basic data for synchronizing virtual and physical states has a clear source.

[0021] S101: Collects dry-bulb and dew-point temperature values ​​from the warehouse temperature and humidity sensor. Performs a difference calculation on the dry-bulb temperature value and the dew-point temperature value for comparison. If the dew-point temperature does not reach the dry-bulb temperature limit, extracts the temperature difference deviation obtained from the subtraction operation. Performs a serialization and splicing operation on the temperature difference deviation to generate a humidity deviation feature vector.

[0022] The humidity deviation feature vector is a data object formed by organizing the temperature difference deviations of each acquisition node according to the node relationship. The input sources for this feature vector are the dry-bulb temperature and dew point temperature values ​​within the same region. It processes the difference state when the dew point temperature is below the dry-bulb temperature limit, and the output is used for latent heat conversion. Nodes with dew point temperatures not lower than the dry-bulb temperature do not trigger humidity deviation extraction, and these nodes do not generate new humidity deviation elements in this round of processing to avoid bringing acquisition values ​​that do not meet the condensation correlation prerequisite into subsequent compensation mappings. On the digital twin side, the humidity deviation feature vector is also used to identify the set of nodes in the virtual storage grid that have humidity and heat compensation requirements, enabling subsequent compensation temperature and density states to be expanded along the same node sequence.

[0023] S102: Call the humidity deviation feature vector, monitor the latent heat constant value associated with the regional environment, perform multiplication operation on each vector element and latent heat constant value in the humidity deviation feature vector, extract the potential energy compensation mapping value obtained by multiplication operation, and arrange all potential energy compensation mapping values ​​in an array according to the sensor spatial node position to obtain the humidity latent heat conversion matrix.

[0024] The humidity latent heat conversion matrix is ​​node mapping data obtained by transforming the humidity deviation feature vector and the latent heat constant value associated with the regional environment. The latent heat constant value is used as a fixed parameter in the preset temperature compensation mapping model, and its role is to transform the humidity deviation state into a latent heat compensation mapping state associated with the energy consumption of vaporization phase change. Each latent heat compensation mapping value is arranged in an array according to the sensor spatial node position, so that subsequent steps can maintain the data position relationship consistent with the storage space grid nodes. This array arrangement allows the virtual storage grid in the digital twin space to take over the physical sensor node positions, preventing the latent heat compensation result from deviating from the actual spatial nodes.

[0025] S103: Perform mean smoothing operation on each conversion element in the humidity latent heat conversion matrix, extract the compensation benchmark value, convert the compensation benchmark value into a temperature scale correction amount through a preset equivalent conversion coefficient, and perform an accumulation operation on the original dry bulb temperature value and the temperature scale correction amount to obtain the single node compensation value. Collect all single node compensation values ​​according to the corresponding relationship of the nodes to obtain the node equivalent compensation temperature set.

[0026] The equivalent compensation temperature set for each node includes a sensible heat correction index, a vaporization phase change energy consumption benchmark, and a virtual temperature scale difference. The sensible heat correction index characterizes the sensible heat state of the dry-bulb temperature after compensation. The vaporization phase change energy consumption benchmark bridges the gap between the latent heat constant and humidity deviation. The virtual temperature scale difference indicates the direction of the compensation temperature change of each node in the digital twin space. Mean smoothing aims to suppress direct disturbances to the compensation benchmark values ​​caused by abrupt changes in the acquisition states between nodes, ensuring that the compensation results can be stably used in subsequent theoretical density conversions. Through this processing, the node temperature scale states in the digital twin space retain both the physical acquisition source and form a compensation temperature input suitable for continuous simulation.

[0027] S2: The equivalent compensation temperature set of the nodes is used to convert the theoretical density. The equivalent mass density of precipitated water in the theoretical density is removed. The density of adjacent nodes in the physical storage space is collected and compared with the actual density. When the density of an adjacent node is greater than the actual density, the acceleration conversion is extracted by combining the density difference with the gravitational constant, and an abnormal buoyancy warning vector is generated. In the digital twin density analysis link, the equivalent compensation temperature set of the nodes serves as the virtual node thermal state, and the theoretical density and the node actual density matrix serve as the virtual density state. The two together support the mapping and identification of air mass anomalies in the physical space.

[0028] In this step, the node equivalent compensation temperature set serves as input for density analysis. The digital twin side generates a theoretical density based on the compensation temperature state, and the equivalent density of precipitated water, calculated from the precipitated water mass parameter, is separated from the theoretical density. This processing ensures that the node's true density matrix not only reflects the temperature-compensated air density state but also eliminates the influence of the equivalent mass of precipitated water on air mass density judgment, providing a benchmark for identifying density jumps between adjacent nodes. Therefore, the true density state in the digital twin space is not simply a temperature conversion result but a consistent virtual-real density benchmark including the removed precipitated water.

[0029] S201: Call the equivalent compensation temperature set of the node to convert the theoretical density, collect the preset precipitated water mass parameters, convert the precipitated water mass parameters into the equivalent density of precipitated water of the corresponding volume, perform a subtraction operation on the theoretical density and the equivalent density of precipitated water, extract the scalar density residual value obtained by the subtraction operation, and perform a matrix-based recombination and arrangement operation on the associated density residual values ​​of all nodes according to the grid coordinates to obtain the true density matrix of the nodes.

[0030] The input to the node true density matrix comes from the node equivalent compensation temperature set and the preset precipitated water mass parameter. The output is used for density comparison between adjacent nodes. Each associated element of this matrix corresponds to a storage space grid node, and the scalar density margin value is used to characterize the true density state after removing the influence of precipitated water. If a node does not have a corresponding compensation temperature state, then that node does not participate in the formation of a new density margin value, and subsequent comparisons are only performed on the density states of nodes that have completed association. Based on this, the digital twin side maintains a one-to-one mapping between the node true density matrix and the physical storage grid, so that subsequent density jump sequences can accurately point back to the physical space location.

[0031] S202: Collect the density of adjacent nodes, call the actual density matrix of the nodes, perform a size comparison operation between the density of adjacent nodes and the related elements in the actual density matrix of the nodes, and extract the corresponding excess deviation value when the density of adjacent nodes exceeds the limit of the related elements. Perform a one-dimensional vector transformation and splicing operation on all excess deviation values ​​to obtain the density jump sequence.

[0032] The density jump sequence describes the set of deviations where the density of adjacent nodes exceeds the node's true density. Its input consists of the adjacent node density and the associated elements within the node's true density matrix, and its output is processed by acceleration conversion. Nodes whose adjacent node densities do not exceed the associated element limits do not form density jump elements; nodes whose adjacent node densities exceed the associated element limits enter the buoyancy anomaly determination link. Through this filtering, the anomalous buoyancy warning vector is generated only from nodes with density jump conditions. In digital twin simulations, the density jump sequence is used to filter nodes in the virtual warehouse space that need to enter the buoyancy warning calculation, reducing the interference of invalid nodes on vehicle path and energy consumption coordination judgments.

[0033] S203: Based on the density jump sequence, monitor the environmental gravitational acceleration constant, perform a multiplication operation on each out-of-limit deviation value in the density jump sequence and the gravitational acceleration constant, divide the result by the air reference density, extract the node acceleration conversion value obtained by the division operation, perform a mapping and arrangement operation on all node acceleration conversion values ​​according to the topological relationship, and generate an abnormal buoyancy warning vector.

[0034] The anomalous buoyancy warning vector includes the spatial convection mutation mode length, the upward turbulence direction angle, and the air mass instability gradient. The spatial convection mutation mode length represents the anomalous intensity of density jumps in the nodal space, the upward turbulence direction angle characterizes the directional state of anomalous buoyancy, and the air mass instability gradient reflects the expansion direction of instability trends between adjacent nodes. The output of this vector is used for subsequent corrections to the operating status of the mobile guided vehicle, enabling the ground adhesion risk caused by air mass instability to be incorporated into the vehicle's energy consumption optimization process. In the digital twin space, the anomalous buoyancy warning vector also serves as a bridge between environmental density anomalies and vehicle motion risks, allowing changes in the humid and hot environment to be transformed into a basis for correcting the guide vehicle's wheel-ground state.

[0035] S3: Based on the anomalous buoyancy warning vector, extract the equivalent vertical lift factor from the vector, and use the equivalent vertical lift factor to correct the wheel-ground normal pressure of the physical mobile guided vehicle. Collect the surface temperature and dew point temperature of the physical mobile guided vehicle, and obtain the water film thickness by combining the condensation conversion rate in the over-limit state. Extract the wheel angular velocity, convert it to theoretical linear velocity, and compare it with the chassis movement speed to establish a global energy consumption optimization trajectory. In the digital twin vehicle simulation link, the anomalous buoyancy warning vector, water film thickness, slip loss, and node resistance cost together constitute the basis for evaluating the virtual guided vehicle path.

[0036] In this step, the anomalous buoyancy warning vector is not directly output as the path, but is first converted into an equivalent vertical lift factor that affects the wheel-to-ground contact state of the moving guide wheel. This factor is used to correct the wheel-to-ground normal pressure, so that water film and slip recognition are not only based on surface temperature and dew point temperature, but also reflect the impact of air mass instability on wheel adhesion. The corrected wheel-to-ground contact state is then used to determine slip loss, ultimately affecting the path selection for the global energy consumption optimization trajectory. Through this virtual-real coupling relationship, the digital twin side maps the impact of abnormal physical environment on vehicle running resistance into the trajectory optimization process in advance.

[0037] S301: Based on the anomalous buoyancy warning vector, collect the vehicle surface temperature and dew point temperature, perform a subtraction operation on the vehicle surface temperature and dew point temperature, and extract the deviation value when the difference exceeds the condensation limit. Obtain the condensation conversion rate constant, perform a multiplication operation on the deviation value and the conversion rate constant, arrange the product sequence, and generate an anomalous water film thickness matrix.

[0038] The inputs to the anomalous water film thickness matrix include the vehicle-mounted surface temperature, dew point temperature, condensation conversion rate constant, and wheel-to-ground contact correction state derived from the anomalous buoyancy warning vector. When the difference exceeds the condensation limit, the deviation value and the condensation conversion rate constant together form a product sequence related to the water film thickness. The product results at each point are mapped into a two-dimensional array according to the horizontal and vertical dimensions of the sampling grid. This two-dimensional array serves as the anomalous water film thickness matrix. Each position in the matrix corresponds to the operating area of ​​the physical mobile guided vehicle and is used for subsequent linear velocity difference mapping. This matrix corresponds to the physical surface condensation risk distribution in the digital twin space, enabling virtual trajectory simulation to identify differences in adhesion conditions at different nodes.

[0039] S302: Call the anomalous water film thickness matrix, collect the wheel angular velocity parameter and wheel diameter constant, perform product conversion on the two parameters to extract the theoretical linear velocity, detect the chassis running speed, perform subtraction comparison operation on the linear velocity and the running speed to obtain the linear velocity difference, identify the offset term according to the coordinate mapping of the thickness matrix of the linear velocity difference, and obtain the slip loss feature sequence.

[0040] The slip loss feature sequence is used to represent the offset between the theoretical linear velocity of the wheel and the chassis operating speed. Its inputs include an anomalous water film thickness matrix, wheel angular velocity parameters, wheel diameter constant, and chassis operating speed. The processing order is as follows: first, the theoretical linear velocity is generated; then, it is compared with the chassis operating speed; finally, an offset term mapping is established according to the coordinates of the water film thickness matrix. The more accurately the offset term reflects the inconsistency between the linear velocity and the actual speed, the more easily it is identified by subsequent node resistance costs as a travel state that needs to be avoided or have its weight reduced. In digital twin trajectory extrapolation, the slip loss feature sequence is used to transform the motion deviation of the physical wheel into the resistance characteristics of virtual path nodes.

[0041] S303: Based on the slip loss feature sequence, detect the energy consumption weight constant of the grid nodes, perform a weighted operation on the offset term and the energy consumption weight constant in the feature sequence to extract the node resistance cost, sort all costs in ascending order to extract the first and second order paths, connect the endpoints of the first and second paths according to the topological relationship, and establish a global energy consumption optimization trajectory.

[0042] The global energy consumption optimization trajectory includes the skid avoidance curvature, traction force extreme coordinate points, and endurance enhancement pose sequence. The skid avoidance curvature represents the turning state of the path under water film and slip constraints, the traction force extreme coordinate points indicate the locations of nodes with prominent traction force demands, and the endurance enhancement pose sequence is used to support the driving posture after connecting low-resistance nodes. This trajectory outputs path node coordinates to subsequent microgrid branch power flow assessments, enabling the correlation between vehicle energy consumption changes and the load status of the warehouse microgrid branches. As a virtual operating path in the digital twin space, the global energy consumption optimization trajectory also transmits local energy consumption changes of the mobile guided vehicle to the microgrid branch load assessment stage.

[0043] S4: Invoke the global energy consumption optimization trajectory, collect the active power flow values ​​of the physical microgrid branches, compare the active power flow values ​​with the thermal stability limit values, extract the spatial enthalpy values ​​of the physical chiller group in the over-limit state, compare the critical enthalpy values ​​of fresh produce with the spatial enthalpy values, and obtain the tolerable drift heat assessment item. In the digital twin microgrid mapping link, the global energy consumption optimization trajectory provides path node association, the branch active power flow values ​​provide electrical load status, and the spatial enthalpy values ​​of the chiller group provide the heat-side adjustable status.

[0044] In this step, the global energy consumption optimization trajectory provides the node coordinates that need to be associated with the load, the active power flow values ​​of the microgrid branches provide the branch load status, and the thermal stability limit values ​​provide the branch over-limit judgment boundaries. Only when the active power flow values ​​are in an over-limit state does the method call the spatial enthalpy values ​​of the refrigeration unit cluster and compare them with the critical enthalpy values ​​of fresh produce to determine whether there is a tolerable enthalpy drift space on the refrigeration side. The digital twin side uses this conditional triggering relationship to make the over-limit state of the electrical side correspond to the adjustment space of the thermal side in the same virtual node link.

[0045] S401: Call the global energy consumption optimization trajectory, collect the active power flow values ​​of the microgrid branches, detect the thermal stability limit values, perform difference calculation and comparison between the active power flow values ​​and the thermal stability limit values, determine that the active power flow values ​​are in an over-limit state, extract the over-limit deviation value of the power flow, map all the over-limit deviation values ​​of the power flow according to the node coordinates of the optimization trajectory, and generate the branch load overflow matrix.

[0046] The branch load overflow matrix describes the active power flow exceeding the limit state of branches at the associated nodes of the optimization trajectory. An initial difference is formed by subtracting the thermal stability limit value from the active power flow value of the microgrid branches. When the initial difference is greater than a zero constant, the active power flow value of the microgrid branches is determined to be in an exceeding limit state, and the initial difference is used as the power flow exceeding limit deviation value. Branches not in an exceeding limit state do not trigger enthalpy drift assessment, thus avoiding generating power reduction requests for branches that do not require adjustment. In the digital twin space, the branch load overflow matrix represents the mapping state between branch load anomalies and trajectory nodes, limiting the triggering range for enthalpy drift assessment.

[0047] S402: Call the branch load overflow matrix, collect the spatial enthalpy value of the refrigeration unit group, retrieve the fresh food critical enthalpy value parameter, perform a subtraction comparison operation on the spatial enthalpy value and the fresh food critical enthalpy value, extract the corresponding difference result of the two values, and perform a weighted fusion operation on the node load status flag bit in the associated branch load overflow matrix to obtain the node enthalpy drift deviation set.

[0048] The node enthalpy drift deviation set is an evaluation data set jointly formed by the spatial enthalpy value, the critical enthalpy value for fresh produce, and the branch load status flag. The spatial enthalpy value reflects the thermal state of the space in which the refrigeration unit operates, the critical enthalpy value for fresh produce reflects the freshness limit of the goods, and the branch load status flag reflects whether there are any over-limit conditions in the microgrid branches that require load reduction. The fusion result of these three is used to determine whether the reduction in refrigeration power can be tolerated within the temperature rise range that does not damage the quality of the goods. Through the thermoelectric state correlation of digital twins, the node enthalpy drift deviation set can simultaneously express the thermal margin of the refrigeration space and the electrical adjustment requirements of the microgrid branches.

[0049] S403: Based on the node enthalpy drift deviation set, collect the environmental heat transfer coefficient constant, perform product integration operation on the deviation element and the environmental heat transfer coefficient constant to obtain the heat fluctuation scalar, arrange all heat fluctuation scalars according to the time sequence step relationship, collect them to form a converged data attribute feature distribution item, and obtain the tolerance drift heat assessment item.

[0050] The tolerance for heat loss assessment includes storage capacity insulation margin, product preservation buffer zone, and quality-loss-free temperature rise range. Storage capacity insulation margin characterizes the thermal inertia load of the storage space under short-term power regulation; product preservation buffer zone characterizes the state in which fresh produce can maintain its quality boundary during enthalpy drift; and quality-loss-free temperature rise range constrains refrigeration power reduction to prevent it from exceeding product preservation requirements. This assessment item serves as a direct input to the thermoelectric power reduction matrix, creating a closed loop between microgrid regulation and refrigeration enthalpy states. In the digital twin environment, the tolerance for heat loss assessment also limits the executable boundaries of virtual power reduction simulations, preventing electrical-side regulation from deviating from physical product preservation constraints.

[0051] S5: Invoke the aforementioned tolerable drift heat assessment item, and calculate the power reduction by combining specific heat capacity, mass flow rate, and control duration. Remove the active load from the power reduction calculation to update the transient active load, and combine the power factor to calculate the transient reactive load, generating the energy consumption coordination power flow state of the storage microgrid. In the digital twin power flow coordination link, the power reduction, transient active load, and transient reactive load jointly update the virtual microgrid operating state and provide feedback on the storage energy consumption coordination results.

[0052] In this step, the tolerable drift heat assessment term is used to define the source and boundary of the power reduction, while the air specific heat capacity constant, mass flow rate parameter, and control duration are used to establish the thermoelectric conversion relationship from the heat fluctuation scalar to the cooling power adjustment offset value. Transient active and reactive loads are not generated in isolation, but rather formed as complex domain vector synthesis results based on the correlation processing between the original active load, reduced power, and power factor. The digital twin side then uses this to form a closed-loop deduction from the thermal side's tolerable drift to the electrical side's power flow state update, enabling cooling load adjustments to be mapped into the coordinated operation state of the warehouse microgrid.

[0053] S501: Call the tolerance drift heat assessment item, collect the air specific heat capacity constant and mass flow rate parameter, divide the heat fluctuation scalar in the tolerance drift heat assessment item by the assessment time step, and perform thermoelectric conversion logic operation in combination with the specific heat capacity constant and mass flow rate parameter, extract the corresponding cooling power adjustment offset value of each node, perform serialization summation and aggregation operation on all offset values, and arrange the aggregation result according to the branch index mapping to generate a thermoelectric power reduction matrix.

[0054] The thermoelectric power reduction matrix represents the executable cooling power reduction states of each branch. Its inputs include a tolerance drift heat assessment term, air specific heat capacity constant, mass flow rate parameter, and assessment time step; its output is a transient active load update. Cooling power adjustment offset values ​​are only formed within the range supported by the tolerance drift heat assessment term. If the heat fluctuation scalar cannot meet the constraints of the product preservation buffer zone or the temperature rise range without quality loss, then that node is not considered a valid source of power reduction. The thermoelectric power reduction matrix is ​​arranged in the digital twin space according to branch index mapping, enabling the virtual power reduction to correspond to the physical microgrid branches.

[0055] S502: Call the thermoelectric power reduction matrix, collect the original active load values ​​of the physical branches, perform a stripping subtraction operation on the original active load values ​​and the corresponding power reduction elements in the thermoelectric power reduction matrix, extract the load difference data obtained by the subtraction operation, and perform a spatiotemporal dimension correlation mapping operation on all load difference data to obtain the transient active load distribution set.

[0056] The transient active load distribution set represents the distribution state of the original active load of a branch after power reduction. This distribution set maintains consistency with the branch index and time step, ensuring that subsequent transient reactive load conversion corresponds to the same branch and the same time sequence. Branches that do not form effective power reduction elements in the thermoelectric power reduction matrix do not generate new reduction effects in this step; they only maintain the corresponding position of the original active load entering the transient load distribution. The transient active load distribution set, as a phased load state in the digital twin microgrid, is used to absorb the impact of the thermoelectric power reduction matrix on the physical branch load.

[0057] S503: Based on the transient active load distribution set, collect the branch power factor constant, perform positive switching multiplication logic operation on each load element and the power factor constant in the transient active load distribution set to obtain the transient reactive load scalar, and perform complex field vector synthesis operation on the transient reactive load scalar and the corresponding element in the transient active load distribution set to generate the energy consumption coordinated power flow state of the warehouse microgrid.

[0058] The energy consumption collaborative power flow state of the warehouse microgrid includes node voltage phasor distribution, feeder network loss matrix, and source-storage balance. The node voltage phasor distribution characterizes the node electrical state after the combined effect of transient active and reactive loads; the feeder network loss matrix characterizes the loss state after branch load updates; and the source-storage balance characterizes the collaborative relationship between the source side, energy storage side, and load side within the microgrid. Thus, the method completes a closed-loop process from temperature and humidity acquisition, density anomaly identification, mobile guided vehicle path optimization, enthalpy drift assessment to microgrid power flow updates. The energy consumption collaborative power flow state of the warehouse microgrid also serves as the final operating state in the digital twin space, reflecting the consistent virtual-real results after the physical warehouse energy consumption collaborative optimization.

[0059] Please see Figure 1 and Figure 6 This embodiment provides a smart warehouse energy consumption collaborative optimization system based on digital twins. This system implements the aforementioned smart warehouse energy consumption collaborative optimization method based on digital twins, and completes the data flow loop through sequential calls between the parameter mapping module, density analysis module, optimization deduction module, enthalpy assessment module, and energy consumption coordination module. The system uses digital twin mapping relationships as the data connection semantics between modules, enabling environmental parameters, density status, vehicle trajectories, enthalpy assessment, and power grid flow to circulate within the same virtual-real correspondence framework.

[0060] The parametric mapping module is used to collect dry-bulb and dew-point temperature values ​​from the warehouse temperature and humidity sensors, and extracts the humidity deviation when the dew-point temperature is lower than the dry-bulb temperature. This module takes dry-bulb temperature, dew-point temperature, and latent heat constant as input, and outputs the node equivalent compensation temperature set. The node equivalent compensation temperature set is then sent to the density analysis module as input for theoretical density conversion. In the digital twin system, the parametric mapping module serves as the entry point for converting the physical temperature and humidity data acquisition state to the virtual node temperature scale state.

[0061] The density analysis module is used to convert the theoretical density by calling the equivalent compensation temperature set of the nodes and to remove the equivalent mass density of precipitated water from the theoretical density. This module further collects the density of adjacent nodes in the physical storage space and compares it with the related elements in the node's true density matrix. When the density of an adjacent node is greater than the true density, an anomalous buoyancy warning vector is generated. The anomalous buoyancy warning vector is output to the optimization and deduction module to correct the wheel-ground normal pressure of the moving guide vehicle and to determine subsequent water film slippage. In the digital twin system, the density analysis module undertakes the intermediate analysis responsibility for converting the physical density state into the virtual buoyancy warning state.

[0062] The optimization and deduction module extracts the equivalent vertical lift factor from the anomalous buoyancy warning vector and uses this factor to correct the wheel-ground normal pressure of the physical mobile guided vehicle. This module collects the surface temperature, dew point temperature, wheel angular velocity, and chassis speed of the physical mobile guided vehicle. Under condensation exceeding limits, it forms an anomalous water film thickness matrix and generates a slip loss characteristic sequence based on a comparison of theoretical linear velocity and chassis speed, thereby establishing a global energy consumption optimization trajectory. The global energy consumption optimization trajectory is output to the enthalpy assessment module for node mapping of the branch load overflow matrix. In the digital twin system, the optimization and deduction module is responsible for deducing the transformation from the physical vehicle's operating state to the virtual energy consumption trajectory state.

[0063] The enthalpy assessment module collects active power flow values ​​from the physical microgrid branches and compares them with thermal stability limit values. In cases of exceeding limits, it retrieves the spatial enthalpy values ​​of the physical refrigeration unit cluster and compares them with the critical enthalpy value for fresh produce. This module outputs a tolerance-based heat loss assessment item, establishing a correspondence between the adjustable heat loss on the refrigeration side and the exceeding-limit states of the microgrid branches. In the digital twin system, the enthalpy assessment module is responsible for the collaborative assessment between the physical refrigeration thermal state and the virtual electrical exceeding-limit states.

[0064] The energy consumption coordination module is used to invoke the tolerable drift heat assessment item and calculate the reduced power by combining specific heat capacity, mass flow rate, and control duration. This module extracts the reduced power from the original active load of the physical branch to obtain the transient active load distribution set, and then combines it with the power factor to calculate the transient reactive load, finally generating the energy consumption coordination power flow state of the storage microgrid. The inputs and outputs of each module in the system correspond one-to-one with the aforementioned method steps, without introducing any additional unclaimed electronic devices, storage media, or program product subjects. In the digital twin system, the energy consumption coordination module is responsible for outputting the mapping of the virtual power reduction simulation results to the power flow coordination state of the physical microgrid.

[0065] Summary of Examples and Declaration of Protection: The above examples, focusing on warehouse temperature and humidity compensation, node density analysis, anomalous buoyancy early warning, water film slippage identification of mobile guided vehicles, global energy consumption trajectory optimization, refrigeration enthalpy drift assessment, and microgrid transient power flow coordination in a digital twin environment, disclose the method steps and the input sources, processing order, intermediate results, output destinations, and collaborative relationships between system modules. Through the virtual-real synchronization and node mapping of digital twins, the processing results of the environment side, vehicle side, refrigeration side, and power grid side in the above examples can form a continuous data flow closed loop. The specific processes, field states, judgment conditions, parameter sources, module responsibilities, data mapping, and processing order described in the examples are only used to explain the possible implementations of the present invention and should not limit the present invention to the specific examples listed. Without departing from the scope of the claims and the original disclosure, any equivalent substitutions, equivalent modifications, equivalent combinations, order adjustments, module-to-module substitutions, equivalent transformations of field names, equivalent succession of the executing entity, or equivalent changes in the carrier form that can be conceived by a person skilled in the art shall fall within the scope of protection of this patent; however, it shall not be extended to unclaimed subjects, nor shall the substantive correspondence of the technical objects be altered by changing the drawing number, step number, or label.

Claims

1. A method for collaborative optimization of energy consumption in smart warehousing based on digital twins, characterized in that, Includes the following steps: S1: Collect the dry bulb and dew point temperature values ​​of the storage temperature and humidity sensor. When the dew point temperature is lower than the dry bulb temperature, extract the humidity deviation. Use the preset temperature compensation mapping model to convert the deviation to the latent heat constant and obtain the node equivalent compensation temperature set. S2: Call the node equivalent compensation temperature set to convert the theoretical density, strip the equivalent mass density of water precipitated from the theoretical density, collect the density of adjacent nodes in the physical storage space and compare it with the real density, and when the density of adjacent nodes is greater than the real density, combine the density difference with the product of the gravity constant to extract the acceleration conversion amount and generate an abnormal buoyancy warning vector. S3: Based on the anomalous buoyancy warning vector, extract the equivalent vertical lift factor in the vector, and use the equivalent vertical lift factor to correct the wheel-ground normal pressure of the physical mobile guided vehicle. Collect the ground surface temperature and dew point temperature of the physical mobile guided vehicle. In the over-limit state, combine the condensation conversion rate to obtain the water film thickness. Extract the wheel angular velocity to convert the theoretical linear velocity and compare it with the chassis movement speed to establish a global energy consumption optimization trajectory. S4: Call the global energy consumption optimization trajectory, collect the active power flow values ​​of the physical microgrid branches, compare the active power flow values ​​with the thermal stability limit values, extract the spatial enthalpy values ​​of the physical refrigeration unit in the over-limit state, compare the critical enthalpy values ​​of fresh food with the spatial enthalpy values, and obtain the tolerance drift heat assessment item. S5: Call the tolerance drift heat assessment item, combine specific heat capacity, mass flow rate and control duration to calculate the power reduction, remove the active load to reduce the power and update the transient active load, combine the power factor to calculate the transient reactive load, and generate the energy consumption coordinated power flow state of the warehouse microgrid.

2. The smart warehousing energy consumption collaborative optimization method based on digital twins according to claim 1, characterized in that, The node equivalent compensation temperature set includes sensible heat correction index, vaporization phase change energy consumption benchmark, and virtual temperature scale difference value. The anomalous buoyancy early warning vector includes spatial convection mutation mode length, upward turbulence direction angle, and air mass instability gradient. The global energy consumption optimization trajectory includes anti-slip travel curvature, traction force extreme value coordinate point, and endurance enhancement posture sequence. The tolerance drift heat assessment item includes storage capacity insulation margin, goods preservation buffer zone, and quality loss-free temperature rise range. The energy consumption collaborative power flow state of the warehouse microgrid includes node voltage phasor distribution, feeder network loss matrix, and source-storage mutual balance.

3. The method for collaborative optimization of energy consumption in smart warehousing based on digital twins according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: Collects dry bulb and dew point temperature values ​​from warehouse temperature and humidity sensors, performs difference calculations on the dry bulb and dew point temperature values ​​for comparison, and extracts the temperature difference deviation obtained from the subtraction operation when the dew point temperature does not reach the dry bulb temperature limit. Then, performs a serialization splicing operation on the temperature difference deviation to generate a humidity deviation feature vector. S102: Call the humidity deviation feature vector, monitor the latent heat constant value associated with the regional environment, perform multiplication operation on each vector element and latent heat constant value in the humidity deviation feature vector, extract the potential energy compensation mapping value obtained by multiplication operation, and arrange all potential energy compensation mapping values ​​in an array according to the sensor spatial node position to obtain the humidity latent heat conversion matrix. S103: Perform mean smoothing operation on each conversion element in the humidity latent heat conversion matrix, extract the compensation benchmark value, convert the compensation benchmark value into a temperature scale correction amount through a preset equivalent conversion coefficient, and perform an accumulation operation on the original dry bulb temperature value and the temperature scale correction amount to obtain the single node compensation value. Collect all single node compensation values ​​according to the corresponding relationship of the nodes to obtain the node equivalent compensation temperature set.

4. The method for collaborative optimization of energy consumption in smart warehousing based on digital twins according to claim 3, characterized in that, The specific steps of S2 are as follows: S201: Call the equivalent compensation temperature set of the node to convert the theoretical density, collect the preset precipitated water mass parameter, convert the precipitated water mass parameter into the equivalent density of precipitated water of the corresponding volume, perform a subtraction operation on the theoretical density and the equivalent density of precipitated water, extract the scalar density residual value obtained by the subtraction operation, and perform a matrix-based recombination and arrangement operation on the associated density residual values ​​of all nodes according to the grid coordinates to obtain the true density matrix of the nodes; S202: Collect the density of adjacent nodes, call the true density matrix of the nodes, perform a size comparison operation between the density of adjacent nodes and the related elements in the true density matrix of the nodes, and extract the corresponding excess deviation value when the density of adjacent nodes exceeds the limit of the related elements. Perform a one-dimensional vector transformation and splicing operation on all excess deviation values ​​to obtain the density jump sequence. S203: Based on the density jump sequence, monitor the environmental gravitational acceleration constant, perform a multiplication operation on each out-of-limit deviation value in the density jump sequence and the gravitational acceleration constant, divide the result by the air reference density, extract the node acceleration conversion value obtained by the division operation, perform a mapping and arrangement operation on all node acceleration conversion values ​​according to the topological relationship, and generate an abnormal buoyancy warning vector.

5. The smart warehousing energy consumption collaborative optimization method based on digital twins according to claim 4, characterized in that, The specific steps for S3 are as follows: S301: Based on the anomalous buoyancy warning vector, collect the vehicle surface temperature and dew point temperature, perform a subtraction operation on the vehicle surface temperature and dew point temperature, extract the deviation value when the difference exceeds the condensation limit, obtain the condensation conversion rate constant, perform a multiplication operation on the deviation value and the conversion rate constant, arrange the product sequence, and generate an anomalous water film thickness matrix. S302: Call the anomalous water film thickness matrix, collect the wheel angular velocity parameter and wheel diameter constant, perform product conversion on the two parameters to extract the theoretical linear velocity, detect the chassis running speed, perform a subtraction comparison operation on the linear velocity and the running speed to obtain the linear velocity difference, identify the offset term by mapping the linear velocity difference according to the coordinates in the thickness matrix, and obtain the slip loss feature sequence. S303: Based on the slip loss feature sequence, detect the energy consumption weight constant of the grid nodes, perform a weighted operation on the offset term and the energy consumption weight constant in the feature sequence to extract the node resistance cost, sort all costs in ascending order to extract the first and second order paths, connect the endpoints of the first and second paths according to the topological relationship, and establish a global energy consumption optimization trajectory.

6. The method for collaborative optimization of energy consumption in smart warehousing based on digital twins according to claim 5, characterized in that, The product sequence is obtained by multiplying the deviation values ​​at each point by the condensation conversion rate constant. The product sequence is then mapped and configured into a two-dimensional array based on the horizontal and vertical dimensions of the sampling grid. This two-dimensional array is then used as the anomalous water film thickness matrix.

7. The method for collaborative optimization of energy consumption in smart warehousing based on digital twins according to claim 5, characterized in that, The specific steps of S4 are as follows: S401: Call the global energy consumption optimization trajectory, collect the active power flow values ​​of the microgrid branches, detect the thermal stability limit values, perform difference calculation and comparison between the active power flow values ​​and the thermal stability limit values, determine that the active power flow values ​​are in an over-limit state, extract the over-limit deviation value of the power flow, map all the over-limit deviation values ​​of the power flow according to the node coordinates of the optimization trajectory, and generate the branch load overflow matrix. S402: Call the branch load overflow matrix, collect the spatial enthalpy value of the refrigeration unit group, retrieve the fresh food critical enthalpy value parameter, perform a subtraction comparison operation on the spatial enthalpy value and the fresh food critical enthalpy value, extract the corresponding difference result of the two values, and perform a weighted fusion operation on the node load status flag bit in the branch load overflow matrix to obtain the node enthalpy drift deviation set; S403: Based on the node enthalpy drift deviation set, collect the environmental heat transfer coefficient constant, perform product integration operation on the deviation element and the environmental heat transfer coefficient constant to obtain the heat fluctuation scalar, arrange all heat fluctuation scalars according to the time sequence step relationship, collect them to form a converged data attribute feature distribution item, and obtain the tolerance drift heat assessment item.

8. The method for collaborative optimization of energy consumption in smart warehousing based on digital twins according to claim 7, characterized in that, The initial difference is extracted by subtracting the thermal stability limit value from the active power flow value of the microgrid branch. The initial difference is compared with the zero constant. If the initial difference is greater than the zero constant, the active power flow value of the microgrid branch is determined to be in an over-limit state. The initial difference is used as the power flow over-limit deviation value.

9. The method for collaborative optimization of energy consumption in smart warehousing based on digital twins according to claim 7, characterized in that, The specific steps of S5 are as follows: S501: Call the tolerance drift heat assessment item, collect the air specific heat capacity constant and mass flow rate parameter, divide the heat fluctuation scalar in the tolerance drift heat assessment item by the assessment time step, and perform thermoelectric conversion logic operation in combination with the specific heat capacity constant and mass flow rate parameter, extract the corresponding cooling power adjustment offset value of each node, perform serialization summation and aggregation operation on all offset values, and arrange the aggregation result according to the branch index mapping to generate a thermoelectric power reduction matrix; S502: Call the thermoelectric power reduction matrix, collect the original active load values ​​of the physical branch, perform a stripping subtraction operation on the original active load values ​​and the corresponding power reduction elements in the thermoelectric power reduction matrix, extract the load difference data obtained by the subtraction operation, and perform a spatiotemporal dimension correlation mapping operation on all load difference data to obtain the transient active load distribution set. S503: Based on the transient active load distribution set, collect the branch power factor constant, perform positive switching multiplication logic operation on each load element and the power factor constant in the transient active load distribution set to obtain the transient reactive load scalar, and perform complex field vector synthesis operation on the transient reactive load scalar and the corresponding element in the transient active load distribution set to generate the energy consumption coordinated power flow state of the warehouse microgrid.

10. A smart warehousing energy consumption collaborative optimization system based on digital twins, characterized in that, The system is used to implement the smart warehouse energy consumption collaborative optimization method based on digital twins as described in any one of claims 1-9, and the system includes: The parameter mapping module collects the dry bulb and dew point temperature values ​​from the warehouse temperature and humidity sensor. When the dew point temperature is lower than the dry bulb temperature, it extracts the humidity deviation and uses a preset temperature compensation mapping model to convert the deviation to the latent heat constant, thereby obtaining the node equivalent compensation temperature set. The density analysis module calls the equivalent compensation temperature set of the nodes to convert the theoretical density, strips the equivalent mass density of the precipitated water from the theoretical density, collects the density of adjacent nodes in the physical storage space and compares it with the actual density. When the density of adjacent nodes is greater than the actual density, it combines the density difference with the product of the gravity constant to extract the acceleration conversion amount and generate an abnormal buoyancy warning vector. The optimization and deduction module extracts the equivalent vertical lift factor from the anomalous buoyancy warning vector and uses the equivalent vertical lift factor to correct the wheel-ground normal pressure of the physical mobile guided vehicle. It collects the ground surface temperature and dew point temperature of the physical mobile guided vehicle, and combines the condensation conversion rate to obtain the water film thickness in the over-limit state. It extracts the wheel angular velocity, converts it into theoretical linear velocity, and compares it with the chassis movement speed to establish a global energy consumption optimization trajectory. The enthalpy assessment module calls the global energy consumption optimization trajectory, collects the active power flow values ​​of the physical microgrid branches, compares the active power flow values ​​with the thermal stability limit values, extracts the spatial enthalpy values ​​of the physical refrigeration unit in the over-limit state, compares the critical enthalpy values ​​of fresh food with the spatial enthalpy values, and obtains the tolerance drift heat assessment item. The energy consumption coordination module calls the tolerance drift heat assessment item, combines specific heat capacity, mass flow rate and control duration to calculate the power reduction, removes active load to reduce power and updates transient active load, combines power factor to calculate transient reactive load, and generates the energy consumption coordination power flow state of the warehouse microgrid.

Images