Spindle component intelligent warehouse management system based on digital twinning

Abstract

The invention relates to the technical field of inventory management, in particular to an intelligent warehouse management system for main shaft parts based on digital twinning, which comprises a main shaft mapping module for acquiring a main shaft shipping space and an allocation path, comparing shipping space nodes with a moving timestamp difference to generate an identification result, the fluctuation identification module extracts channel nodes to analyze fluctuation threshold deviation and generate a report, the allocation control module generates a scheduling list based on queuing duration and shipping space data distribution weight, and the abnormity positioning module extracts unallocated parts to analyze interface difference and mark abnormal links. And the state early warning module integrates four indexes of abnormal data identification risk paths to generate a management scheme. According to the method, bin position nodes and main shaft movement timestamps are synchronously compared to identify difference nodes, a dynamic weight distribution mechanism is established in combination with channel node fluctuation analysis, an interruption link difference trend and interface delay data are integrated to construct a risk assessment model, and abnormal conduction path positioning and early warning are realized. And scheduling conflicts and resource waste caused by data deviation are reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of inventory management, and in particular to a main shaft component intelligent warehouse management system based on digital twinning. BACKGROUND

[0002] The technical field of inventory management includes various methods and means for informationization and automation management of the whole process of storage, storage, withdrawal and allocation of materials, spare parts, products and other storage objects in enterprises or organizations. The core content includes inventory state monitoring, inventory quantity dynamic updating, storage location optimized configuration and inventory turnover efficiency improvement. With the development of manufacturing industry towards flexibility and intelligence, inventory management gradually integrates sensing layer data collection, data communication, digital modeling and virtual-real synchronization and other technologies. Through visual modeling of the whole life cycle of materials, fine management of warehouse processes is realized. Especially in the warehouse link of high-precision components, higher requirements are put forward for state tracking and dynamic management. Inventory management technology gradually develops multi-dimensional perception, real-time feedback and strategy linkage and other characteristics in this process.

[0003] Among them, the main shaft component intelligent warehouse management system based on digital twinning refers to taking the main shaft component as the management object, constructing the bidirectional mapping relationship between the physical entity and the virtual model, establishing the mapping of the geometric structure, life cycle state and storage location information of the main shaft component through digital modeling means, and based on the model, combining RFID identification means to obtain the real-time location information of the main shaft component, using multi-source sensing data to collect and classify mark the state of temperature, vibration and the like of the main shaft component, and automatically managing the components with different health states through the set inventory organization rules. At the same time, through the preset task flow rules and strategy library, the task state record and inventory distribution dynamic adjustment of the main shaft component in the links of warehousing, allocation, withdrawal and the like are realized. The focus is on constructing and updating the digital model according to the state of the main shaft, and completing the real-time adjustment of the inventory structure and the strategy feedback combined with the actual sensing data.

[0004] The existing inventory management technology relies on static digital model and periodic data updating mechanism. When the allocation path is frequently changed or the load of the warehouse interface fluctuates, the physical warehouse state and the virtual mapping relationship are difficult to match in real time. The task scheduling strategy lags behind the actual demand. There is a lack of dynamic fluctuation and interruption link correlation analysis capability of channel nodes. It cannot identify the abnormal conduction risk caused by the coupling of multiple factors in the allocation path, resulting in unbalanced allocation of warehouse resources and decreased scheduling efficiency. SUMMARY

[0005] The purpose of the present application is to solve the problems existing in the prior art and to provide a main shaft component intelligent warehouse management system based on digital twinning.

[0006] In order to achieve the above object, the application adopts the following technical scheme: A spindle component intelligent warehouse management system based on digital twinning, the system comprises:

[0007] A spindle mapping initialization module obtains spindle component storage positions and allocation paths, compares storage position node update times and movement record times, identifies difference nodes and extracts spindle component numbers, and generates a storage position difference identification result;

[0008] A storage position fluctuation identification module extracts channel nodes of spindle components under the allocation path according to the storage position difference identification result, analyzes the deviation of channel node fluctuation values from a set fluctuation threshold, and generates a storage position fluctuation analysis report;

[0009] An allocation rhythm control module obtains spindle component allocation task queuing time lengths and storage position data marked as path switching changes in the storage position fluctuation analysis report, assigns weights to spindle components according to the data, and generates an allocation task scheduling list;

[0010] An abnormality origin positioning module calls task numbers in the allocation task scheduling list, extracts spindle components that are not allocated on time, analyzes queuing time lengths and interface feedback differences, marks interrupt links as abnormal, and generates an allocation abnormality identification result;

[0011] A state conduction early warning module analyzes queuing time lengths, capacity changes, and interface stability according to abnormal link analysis in the allocation abnormality identification result, identifies risk paths, and generates a scheduling risk management scheme according to four indicators of the allocation path.

[0012] As a further scheme of the application, the storage position difference identification result includes difference component numbers and path difference identifiers, the storage position fluctuation analysis report specifically includes channel fluctuation coefficients, threshold deviation levels, and fluctuation mode classifications, the allocation task scheduling list includes weight allocation matrices and scheduling priority queues, the allocation abnormality identification result specifically refers to interrupt link encodings and interface response difference indicators, and the scheduling risk management scheme includes path change frequency indexes, interface delay parameters, capacity deviation coefficients, and interrupt rate statistics.

[0013] As a further scheme of the application, the spindle mapping initialization module comprises:

[0014] A storage position data acquisition submodule obtains current storage position numbers and allocation path numbers of spindle components in a warehouse area, extracts update time stamps of storage position nodes on each allocation path and movement record time stamps of corresponding spindle components, and establishes a time stamp data set containing storage position node numbers, double time stamps, and position information;

[0015] The node difference comparison submodule calls the position node number in the timestamp data set, compares the numerical relationship between the update timestamp and the movement record timestamp under the same node, filters the nodes with the update timestamp value greater than the movement record timestamp, synchronously checks the matching state of the node position information and the position coordinates in the movement record, and obtains a difference node set with time lag and position offset;

[0016] The difference result generation submodule extracts the unique identification code of the corresponding main shaft component and the corresponding allocation path number based on the node number in the difference node set, calculates the offset amount of the difference node position coordinates and the standard position coordinates, and generates a position difference identification result including the component number, the path number and the offset amount.

[0017] As a further scheme of the application, the position fluctuation identification module comprises:

[0018] The channel node extraction submodule extracts the channel node number connecting different positions in each allocation path based on the position difference identification result, establishes a channel node mapping relationship according to the main shaft component number and the allocation path number, and generates a channel node set.

[0019] The fluctuation value calculation submodule calculates the position coordinate change amount of each channel node in the channel node set when the allocation path is switched, takes the difference value of the coordinate change amount of two consecutive switches as the fluctuation value, constructs a three-dimensional data matrix according to the node number and the switching number, and generates a fluctuation value matrix.

[0020] The deviation marking analysis submodule compares each node fluctuation value in the fluctuation value matrix with a preset path switching fluctuation threshold value, calculates the absolute value of the deviation value, marks the nodes with a deviation value exceeding the threshold value, classifies and aggregates the marking data according to the threshold interval, and generates a position fluctuation analysis report.

[0021] As a further scheme of the application, the allocation rhythm control module comprises:

[0022] The task parameter acquisition submodule extracts the main shaft component allocation task queuing time length, the starting position capacity and the target position interface use frequency data marked as path switching changes based on the position fluctuation analysis report, establishes a three-dimensional parameter mapping table according to the component number, and generates a task parameter set.

[0023] The weight allocation calculation submodule normalizes the queuing time length, the position capacity and the interface use frequency based on the task parameter set, allocates a weight coefficient according to a preset business rule, calculates the comprehensive scheduling weight value of each main shaft component, and generates a weight coefficient matrix.

[0024] The scheduling queue generating submodule arranges the spindle components in descending order according to the comprehensive scheduling weight values in the weight coefficient matrix, binds the component numbers after sorting with corresponding allocation path numbers, reconstructs a task execution sequence, and generates an allocation task scheduling list.

[0025] As a further scheme of the present application, the abnormal origin positioning module comprises:

[0026] The task screening submodule extracts spindle component numbers and corresponding allocation path numbers in the allocation task scheduling list that are not dispatched on time, sorts them according to the difference between the task interruption timestamp and the planned timestamp, and generates an unfinished task set.

[0027] The difference analysis submodule obtains the queuing time length data and the real-time feedback value of the warehouse position interface in the allocation task based on the unfinished task set, calculates the absolute value of the difference between the two, constructs a difference value-time stamp two-dimensional relationship table according to the task number, and generates a difference index set.

[0028] The abnormal positioning submodule calls the difference index set, compares the difference value change trend of each link of the allocation task, identifies the task link with a sudden increase in the difference value and a sustained abnormal duration, marks the warehouse node corresponding to the sudden increase starting point as an interruption link, and generates an allocation abnormality identification result.

[0029] As a further scheme of the present application, the state transmission early warning module comprises:

[0030] The risk parameter acquisition submodule calls the spindle component numbers, the frequency of use of the warehouse position interface, and the corresponding allocation path numbers of the abnormal link in the allocation abnormality identification result, synchronously obtains the queuing time length, the warehouse capacity change rate, and the interface stability coefficient of the path, and generates a risk parameter set.

[0031] The path risk assessment submodule calculates the product coefficient of the allocation path change frequency and the warehouse interface response delay based on the risk parameter set, superimposes the weighted average value of the warehouse capacity utilization deviation value and the task interruption rate, constructs a four-dimensional risk assessment matrix according to the path number, and generates a path risk coefficient.

[0032] The management scheme generating submodule calls the path risk coefficient, prioritizes the change frequency, response delay, utilization deviation, and interruption rate according to a preset business rule, matches dynamic scheduling strategies and capacity expansion thresholds for high-risk paths, and generates a scheduling risk management scheme.

[0033] Compared with the prior art, the present application has the following advantages and positive effects:

[0034] In the application, by synchronously comparing the position node update timestamp and the main shaft component movement record timestamp, the difference node between the physical warehouse and the digital model is identified, the dynamic weight distribution mechanism is established to optimize the task scheduling queue by combining the channel node fluctuation value analysis when the allocation path is switched, the difference value change trend and the interface response delay data in the interrupted link of the allocation task are integrated, the multi-dimensional risk assessment model is constructed, the abnormal conduction path positioning and risk warning of the warehouse state are realized, the inventory synchronization precision and strategy response efficiency in the high-precision component allocation scene are improved, and the scheduling conflict and resource waste caused by virtual and real data deviation are reduced. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 The system flowchart of the application is shown in the figure.

[0036] Figure 2 The acquisition flowchart of the main shaft mapping initialization module of the application is shown in the figure.

[0037] Figure 3 The acquisition flowchart of the position fluctuation identification module of the application is shown in the figure.

[0038] Figure 4 The acquisition flowchart of the allocation rhythm control module of the application is shown in the figure.

[0039] Figure 5 The acquisition flowchart of the abnormal origin positioning module of the application is shown in the figure.

[0040] Figure 6 The acquisition flowchart of the state conduction warning module of the application is shown in the figure. DETAILED DESCRIPTION

[0041] The technical solutions in the application will be described below with reference to the drawings.

[0042] In the embodiments of the application, the words such as "example", "for example" are used to represent an example, illustration or description. Any embodiment or design scheme described as "example" in the application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the word "example" is intended to present the concept in a specific manner. In addition, in the embodiments of the application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0043] In the embodiments of the application, "image" and "picture" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent. "Of", "corresponding" and "corresponding" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent.

[0044] In the embodiments of the present application, sometimes the subscript such as W1 may be written in the form of non-subscript such as W1, and the meanings expressed are consistent when the difference is not emphasized.

[0045] To make the technical problems, technical solutions and advantages to be solved by the present application clearer, specific embodiments will be described in detail below with reference to the drawings.

[0046] Please refer to Figure 1 The present application provides a technical solution: a spindle component intelligent warehouse management system based on digital twinning, the system comprising:

[0047] The spindle mapping initialization module obtains the current storage position and the allocation path number of the spindle component in the warehouse area, extracts the update time stamp of the storage position node and the movement record time stamp of the spindle component on each allocation path, compares the two types of time stamps according to the storage position node number, judges that the storage position node whose update is delayed from the movement record and whose position information is inconsistent is a difference node, extracts the spindle component number and the allocation path number corresponding to the difference node, and generates a storage position difference identification result;

[0048] The storage position fluctuation identification module extracts the channel node used to connect different storage positions of each spindle component under the corresponding allocation path according to the storage position difference identification result, analyzes the fluctuation value of the spindle component in each allocation path switching in the channel node, marks and classifies the deviation value of the fluctuation value from the set fluctuation threshold, and generates a storage position fluctuation analysis report;

[0049] The allocation rhythm control module obtains the spindle component allocation task queuing time length, the starting storage position capacity and the target storage position interface usage frequency data marked as changes in allocation path switching in the storage position fluctuation analysis report, assigns weights to the spindle components according to the queuing time length, the storage position capacity and the interface usage frequency, obtains a new task scheduling queue and sorts it, and generates an allocation task scheduling list;

[0050] The abnormal origin positioning module calls the allocation task number in the allocation task scheduling list, extracts the spindle component that is not allocated on time, analyzes the difference between the queuing time length and the storage position interface feedback value in the allocation task, marks the link where the allocation task interruption occurs as an abnormal link, and generates an allocation abnormality identification result;

[0051] The state conduction early warning module analyzes the allocation task queuing time length, the storage position capacity change and the storage position interface stability of the spindle component according to the spindle component and the storage position interface usage frequency of the abnormal link in the allocation abnormality identification result, identifies the allocation risk path of the spindle component, generates a scheduling risk management scheme according to the allocation path change frequency, the storage position interface response delay, the storage position capacity utilization rate deviation and the allocation task interruption rate in the risk path.

[0052] The warehouse position difference identification result includes a difference component number and a path difference identifier. The warehouse position fluctuation analysis report includes a channel fluctuation coefficient, a threshold deviation level, and a fluctuation mode classification. The allocation task scheduling list includes a weight distribution matrix and a scheduling priority queue. The allocation abnormality identification result specifically indicates an interrupt link code and an interface response difference indicator. The scheduling risk management scheme includes a path change frequency index, an interface delay parameter, a capacity deviation coefficient, and an interrupt rate statistic.

[0053] Please refer to Figure 2 The main shaft mapping initialization module includes:

[0054] The warehouse position data acquisition submodule acquires the current warehouse position number and the allocation path number of the main shaft component in the warehouse area, extracts the update timestamp of the warehouse node on each allocation path and the movement record timestamp of the corresponding main shaft component, and establishes a timestamp dataset containing the warehouse node number, double timestamp, and position information.

[0055] Through real-time data interaction with the warehouse management system (WMS), the system actively queries and parses the electronic tag information of the specific main shaft component uploaded by the radio frequency identification (RFID) device, thereby extracting its unique electronic code, such as 'SC-007', and locking the allocation path number currently assigned to the component from the path planning table of the WMS, such as 'P-05'. Next, the system accesses the warehouse state table in the WMS database for each warehouse node on the 'P-05' path, taking 'A-01-03' as an example, reads the system timestamp of the latest state change of the warehouse, which is defined as the update timestamp Tupdate. At the same time, the system initiates a query request to the log database of the mobile execution system (MES) or the automatic guided vehicle (AGV), filters out the last completed warehousing or shifting operation according to the component code 'SC-007', extracts the confirmation timestamp fed back by the physical device at the completion of the operation, which is defined as the movement record timestamp Tmove. The system also retrieves the preset three-dimensional coordinates (X, Y, Z) of the warehouse node 'A-01-03' from the digital twin model of the WMS as its standard position information, such as (10.5, 20.2, 1.5). Finally, the acquired warehouse node number, numerical Tupdate, numerical Tmove, and standard position coordinates are structured and bound as a data unit. This process is repeated for all warehouse nodes on the 'P-05' path, and all data units are collected to establish a timestamp dataset containing the warehouse node number, double timestamp, and position information.

[0056] The node difference comparison sub-module calls the position node number in the timestamp data set, compares the numerical relationship between the update timestamp and the moving record timestamp under the same node, filters the nodes with the update timestamp value greater than the moving record timestamp, synchronously checks the matching state of the node position information and the position coordinates in the moving record, and obtains the difference node set with time lag and position offset;

[0057] The timestamp data set is called, and the system starts a loop iteration program to process the records in the data set one by one. Taking the record of the position node 'A-01-03' as an example, the program first extracts the update timestamp Tupdate and the moving record timestamp Tmove of the node, performs a direct numerical comparison operation, judges whether the value of Tupdate is greater than the value of Tmove, and if the condition is true, the node is preliminarily marked as having time lag. Immediately, the program seamlessly connects and starts the synchronous verification process. It extracts the actual physical coordinates recorded by the AGV positioning system of component 'SC-007' at Tmove from the log of the moving execution system, denoted as Pactual=(Xa,Ya,Za), for example (10.6, 20.1, 1.5), and at the same time, the standard position information Pstd=(Xs,Ys,Zs) of the node is called from the timestamp data set, i.e. (10.5, 20.2, 1.5). The system then calculates the spatial straight line distance between the two coordinate points, i.e. the Euclidean distance D=((Xa-Xs) 2 +(Ya-Ys) 2 +(Za-Zs) 2 ) 0.5 , and the numerical calculation gives ((10.6-10.5) 2 +(20.1-20.2) 2 +(1.5-1.5) 2 ) 0.5 , the result is (0.01+0.01+0) 0.5 ≈0.141 meters. There is a position offset threshold D th in the system, which is set according to the accuracy requirements of warehouse operation, such as considering the navigation error of AGV and the placement error of fork, and is set to 0.1 meters. Then compare the calculated distance D with the threshold D th . In this example, 0.141 meters is greater than 0.1 meters, which is judged as position offset. When a node is judged to meet both time lag and position offset conditions, the node number 'A-01-03' is added to a dynamic array structure, which is the difference node set with time lag and position offset.

[0058] The difference result generation submodule extracts the unique identification code of the corresponding main shaft component and the allocation path number based on the node number in the difference node set, calculates the offset amount of the difference node position coordinates and the standard warehouse position coordinates, and generates a warehouse position difference identification result containing the component number, path number, and offset amount.

[0059] Based on the difference node set, which currently contains node numbers such as 'A-01-03', the system initiates the subsequent information extraction and processing process for each node number in the set. First, the program uses 'A-01-03' as a query index to backtrack to the initial timestamp data set. Through precise matching, it extracts the associated two key information, i.e., the unique identification code of the main shaft component 'SC-007' and its allocation path number 'P-05' at that time. Subsequently, the program directly retrieves the actual position coordinates Pactual of the difference node from the calculation cache of the node difference comparison link, which is (10.6, 20.1, 1.5), and the standard warehouse position coordinates Pstd, which is (10.5, 20.2, 1.5). It then performs subtraction between the two three-dimensional coordinate vectors to quantify the offset amount. The specific calculation of the offset vector ΔP = (ΔX, ΔY, ΔZ) is as follows: the offset amount ΔX in the X-axis direction is 10.6-10.5 = 0.1 meters, the offset amount ΔY in the Y-axis direction is 20.1-20.2 = -0.1 meters, and the offset amount ΔZ in the Z-axis direction is 1.5-1.5 = 0 meters. This calculation result indicates the offset direction and distance of the component in each coordinate axis. Finally, the system combines the extracted component number 'SC-007', path number 'P-05', and calculated offset amount (0.1, -0.1, 0) into a new structured data record, and adds this record to a result list. When all nodes in the difference node set have completed this operation, the warehouse position difference identification result containing the component number, path number, and offset amount is finally generated.

[0060] The offset amount uses the formula:

[0061]

[0062] Formula details and calculation derivation process:

[0063] 1. Parameter assignment and data source:

[0064] (laser range finder actual measurement difference node X coordinate), (standard warehouse X coordinate reference value, from warehouse database), Y / Z axis coordinates are measured similarly

[0065] γ X= 0.85 (X-axis dynamic weight factor, calculated according to the current allocation path load rate 68%: γ X = 1 - 0.15 x (load rate / 100), γ Y = 0.92, γ Z = 0.88 (Y / Z-axis load rates are 42% and 55% respectively, calculated according to the same formula);

[0066] η env = 0.78 (environmental interference correction coefficient, calculated from temperature and humidity sensor data: η env = 1 - 0.02 x (temperature-25) - 0.01 x (humidity-60), current temperature 28°C, humidity 65%);

[0067] λ k = 0.24 (historical error accumulation rate, calculated according to the allocation path interruption rate in the last 10 tasks: λ where the reference interruption threshold is 0.5 times per cycle, and the actual interruption is 3 times);

[0068] T cal = 24 hours (device calibration period, set according to the laser ranging module technical manual);

[0069] ∈ = 0.7 (error amplification suppression factor, value according to high-density warehouse area operation and maintenance specifications);

[0070] ζ dev = 0.12 (device precision attenuation coefficient, calculated according to the cumulative working time of the laser ranging module 1200 hours: ζ dev = 0.0001 x working time);

[0071] δ hist = 0.18 mm (historical calibration deviation mean, taken from the mean of the last 20 calibration records), step-by-step calculation process:

[0072] 2. Molecule calculation:

[0073] X-axis term: (1250.3-1248.7) x 0.85 x 0.78 = 1.6 x 0.663 = 1.0608, square after multiplying (1+0.24) / 24: 1.0608 2 x 1.24 / 24 = 1.125 x 0.0517 = 0.0581;

[0074] Y-axis term: (543.1-542.5) x 0.92 x 0.78 = 0.6 x 0.7176 = 0.4306, square after multiplying coefficient: 0.4306 2 x 0.0517 = 0.0076;

[0075] Z-axis term: (210.8-210.0) x 0.88 x 0.78 = 0.8 x 0.6864 = 0.5491, square after multiplication coefficient: 0.5491 2 x 0.0517 = 0.0156;

[0076] Molecular sum: 0.0581 + 0.0076 + 0.0156 = 0.0813;

[0077] Denominator calculation:

[0078] ∑γ n = 0.85 + 0.92 + 0.88 = 2.65;

[0079] ζ dev ·δ hist = 0.12 x 0.18 = 0.0216;

[0080] Denominator total value: 0.7 x (2.65 + 0.0216) = 0.7 x 2.6716 = 1.8701;

[0081] Final result:

[0082]

[0083] 3. Result analysis:

[0084] The calculation result 0.2086 mm represents the dynamic offset of the current difference node from the standard position, which is less than the preset threshold value 0.35 mm, and is determined to be within the acceptable range, without triggering the calibration instruction.

[0085] Numerical relevance: the numerator reflects the coupling effect of real-time coordinate deviation and environmental interference, and the denominator suppresses error amplification through weight and device state, and the result is directly used for virtual-real synchronization verification of the warehouse control system.

[0086] Please refer to Figure 3 , the position fluctuation recognition module comprises:

[0087] The channel node extraction submodule calls the position difference recognition result, extracts the channel node number connecting different positions in each allocation path, establishes a channel node mapping relationship according to the main shaft component number and the allocation path number, and generates a channel node set;

[0088] The calling position difference identification result, the system first analyzes the data structure, extracts each existing difference main shaft component number and corresponding allocation path number recorded therein, for example, the system reads a record that component 'SC-007' exists difference in path 'P-05', then, the system uses path number 'P-05' as a query key, accesses the path planning database defined in advance in the warehouse management system (WMS), the database defines the node sequence passed through by each allocation path in detail, the node sequence of path 'P-05' is defined as 'A-01-03', 'T-01', 'T-02', 'B-02-05', wherein 'A' and 'B' at the beginning are position nodes, and 'T' at the beginning is a channel node, the system performs a screening operation to extract only the node numbers of the type "channel" from the sequence to obtain the channel node list ['T-01', 'T-02'], then the system constructs a new key-value pair data structure, combines the main shaft component number 'SC-007' and the allocation path number 'P-05' into a unique composite key, that is, ('SC-007', 'P-05'), and the aforementioned extracted channel node list ['T-01', 'T-02'] as the value of the key, and stores this key-value pair in a mapping table, the system repeats the extraction, screening and mapping storage process for all entries in the position difference identification result, and finally aggregates all mapping relationships to generate a channel node set.

[0089] The fluctuation value calculation submodule calculates the position coordinate change of each channel node in the channel node set when the allocation path is switched, and takes the difference between the coordinate change of two consecutive switches as the fluctuation value. A three-dimensional data matrix is constructed according to the node number and the number of switches to generate a fluctuation value matrix.

[0090] Based on the channel node set, the system starts the fluctuation value calculation process for each mapping relationship stored therein, for example, ('SC-007', 'P-05') corresponds to ['T-01', 'T-02'], the system queries the mobile execution history database to count all allocation task records related to the main shaft component 'SC-007' and passing through the channel node 'T-01', assuming that two allocations involving the node have occurred recently, in the first allocation task, the actual coordinates recorded when the automated guided vehicle (AGV) passes through 'T-01' are (15.05, 20.10, 0.50), and the standard coordinates of 'T-01' in the digital twin model are (15.00, 20.00, 0.50), the system calculates the position coordinate change ΔC1 of this allocation, that is, the Euclidean distance between the two coordinate points, ΔC1 = ((15.05-15.00) 2 +(20.10-20.00) 2 +(0.50-0.50) 2 )0.5 ≈0.112 meters, the actual coordinates recorded when the AGV passed 'T-01' during the second allocation task became (14.90, 20.05, 0.50), and the system calculated the second position coordinate change AC2 = ((14.90-15.00) 2 +(20.05-20.00) 2 +(0.50-0.50) 2 ) 0.5 ≈0.112 meters, then the system calculated the difference between the coordinate changes AC1 and AC2 of the two consecutive allocations, obtaining the fluctuation value V T-01,k＝2 = |AC2-AC1| = |0.112-0.112| = 0 meters, if the third allocation passed 'T-01' with coordinates (15.20, 20.15, 0.50), then AC3 ≈ 0.250 meters, and the fluctuation value V T-01,k＝3 = |AC3-AC2| = |0.250-0.112| = 0.138 meters, the system indexed these calculated fluctuation values according to the node number 'T-01' and the number of allocation switches (second, third), and constructed a data matrix with [node number, switch number] as dimensions and fluctuation values as storage content. After calculating all nodes in the channel node set, the fluctuation value matrix was generated.

[0091] The deviation marker analysis submodule compares each node fluctuation value in the fluctuation value matrix with a preset path switch fluctuation threshold value item by item, calculates the absolute value of the deviation, numbers and marks the nodes with deviation values exceeding the threshold value, and classifies and aggregates the marked data according to the threshold value interval to generate a position fluctuation analysis report.

[0092] The system starts an analysis program by calling the fluctuation value matrix, and reads each node's fluctuation value at each switch in the matrix item by item, for example, reads the fluctuation value V T-01,3 of node 'T-01' at the third switch as 0.138 meters. The program compares this value with a preset path switch fluctuation threshold value V th The setting of this threshold value V th refers to the repeated positioning accuracy (e.g. ±0.05 meters) provided by the AGV equipment supplier and the statistical distribution of the normal fluctuation range in a large amount of historical operation data, taking the 95% quantile as a reasonable value, for example V th = 0.100 meters. The system performs a judgment operation, 0.138 meters is greater than 0.100 meters, so this node's fluctuation is determined to be abnormal. The system then calculates the absolute value of the deviation D v = |V-V th| = | 0.138 - 0.100 | = 0.038 meters, since the fluctuation value exceeds the threshold value, the system immediately adds an anomaly label to node 'T-01', and then, the system classifies the label according to the preset threshold interval, and the interval is determined by the deviation value D v defined as "low deviation" between 0 to 0.05 meters, defined as "moderate deviation" between 0.05 meters to 0.1 meters, and defined as "high deviation" beyond 0.1 meters, in this case, 0.038 meters belongs to the "low deviation" interval, the system aggregates and classifies the current event of node 'T-01' as "low deviation", and records the relevant fluctuation value and deviation value, after traversing all the data in the fluctuation value matrix, the system summarizes all the labeled node information, deviation level, fluctuation mode classification data, and generates a position fluctuation analysis report.

[0093] Please refer to Figure 4 , the dispatching rhythm control module includes:

[0094] The task parameter acquisition submodule calls the position fluctuation analysis report, extracts the main shaft component dispatching task queuing time length, starting position capacity and target position interface usage frequency data marked as path switching change, establishes a three-dimensional parameter mapping table according to the component number, and generates a task parameter set;

[0095] The system first performs a filtering operation, traverses all the entries in the report, and only extracts those records marked as "path switching change" in the "fluctuation mode classification" field. For each record filtered out, the system extracts its associated main shaft component number, such as 'SC-007', and uses this number as an index to query the task management system for the dispatching task information currently being executed for that component, thereby obtaining the creation timestamp of the task. By calculating the difference between the current system time and the task creation timestamp, the queuing time length T q of the task is obtained, for example, 900 seconds. Then, the system obtains the starting position number of the dispatching task, such as 'A-01-03', and queries the warehouse management system (WMS) to obtain the capacity data of the storage area where the position is located, calculates the starting position capacity utilization rate C s , specifically, the number of used storage locations 45 is divided by the total number of storage locations 50 to obtain C s = 0.9. Then, the system obtains the target position number of the task, such as 'B-02-05', and identifies the physical interface (such as a certain AGV transfer point) serving the target position. By querying the historical log of the mobile execution system (MES), the total number of tasks processed by the interface in the past statistical period, for example, the past one hour, is counted, and this value is the target position interface usage frequency F iFor example, if the system processes 30 tasks per hour, the system combines the three parameters (900 seconds, 0.9, 30 times / hour) collected for the component 'SC-007', and stores the set of three-dimensional parameters as values in a new mapping table structure with the spindle component number 'SC-007' as the key, and repeats the process for all screened components to generate the task parameter set.

[0096] The weight distribution calculation sub-module normalizes the queuing time, bin capacity and interface usage frequency based on the task parameter set, assigns weight coefficients according to the preset business rules, calculates the comprehensive scheduling weight value of each spindle component, and generates a weight coefficient matrix.

[0097] Based on the task parameter set, the system performs normalization processing and weight calculation on the parameters of all spindle component allocation tasks contained therein. First, the system traverses the same dimension parameters of all tasks in the task parameter set to find the maximum and minimum values, for example, for the queuing time T q , the minimum value is 600 seconds and the maximum value is 900 seconds, for the starting bin capacity utilization rate C s , the minimum value is 0.85 and the maximum value is 0.95, and for the interface usage frequency F i , the minimum value is 20 times / hour and the maximum value is 30 times / hour. Then, the minimum-maximum normalization formula X' norm =(X-X min ) / (X max -X min ) is applied to each parameter value of each task to calculate. Taking the parameters (900, 0.9, 30) of the component 'SC-007' as an example, the normalized parameters become (1.0, 0.5, 1.0). Next, the system calls the preset business rules to assign weight coefficients. According to the warehouse management priority setting, for example, the weight w Tq of the queuing time is set to 0.5 to respond to long-waiting instructions quickly, the weight w Fi of the interface usage frequency is set to 0.3 to avoid congestion in high-load areas, the weight w Cs of the bin capacity utilization rate is set to 0.2 to prioritize handling outbound tasks in high-density storage areas, and w Tq +w Cs +w Fi =0.5+0.2+0.3=1.0. Subsequently, the system calculates the comprehensive scheduling weight value W comp of each spindle component = w Tq ×T′ q,norm +w Cs ×C′ s,norm +w Fi ×F′ i,norm , for 'SC-007', its Wcomp = 0.5 x 1.0 + 0.2 x 0.5 + 0.3 x 1.0 = 0.9, the number of each component, the original three-dimensional parameter, the normalized three-dimensional parameter, the weight of each parameter and the final calculated comprehensive scheduling weight value are integrated into a row of data, and the data of all components together constitute a table to generate a weight coefficient matrix.

[0098] The scheduling queue generation submodule arranges the spindle components in descending order according to the comprehensive scheduling weight values in the weight coefficient matrix, binds the sorted component numbers with the corresponding allocation path numbers, reconstructs the task execution sequence, and generates an allocation task scheduling list;

[0099] The weight coefficient matrix is called, and the system extracts each spindle component number and its corresponding comprehensive scheduling weight value W comp recorded therein, such as (‘SC-007’, 0.9), (‘SC-008’, 0.75), (‘SC-009’, 0.95), etc. The system loads these records into a dynamic list and performs a sorting operation on the list. The sorting rule is to arrange the comprehensive scheduling weight values W comp in descending order. After sorting, the component order in the list becomes ‘SC-009’, ‘SC-007’, ‘SC-008’. Then, the system processes the spindle component numbers one by one in the sorted order, uses the component number as a query key, and goes back to the original task parameter set or task management system to find and extract the allocation path number associated with the current allocation task of the component. For example, it is found that ‘SC-009’ corresponds to path ‘P-08’, ‘SC-007’ corresponds to ‘P-05’, and ‘SC-008’ corresponds to ‘P-12’. The system then binds the sorted component numbers with the queried path numbers one by one to form a new ordered task pair list, such as [(‘SC-009’, ‘P-08’), (‘SC-007’, ‘P-05’), (‘SC-008’, ‘P-12’)]. This new list sequence replaces the original task execution sequence arranged in time order or any order to generate an allocation task scheduling list.

[0100] Please refer to Figure 5 The abnormality origin positioning module includes:

[0101] The task screening submodule extracts the spindle component numbers and corresponding allocation path numbers in the allocation task scheduling list that have not been allocated on time, sorts them according to the difference between the task interruption timestamp and the planned timestamp, and generates an incomplete task set.

[0102] The call allocation task scheduling list is called, which has been arranged in descending order of the comprehensive scheduling weight value, for example, containing task pairs such as ('SC-009', 'P-08'), ('SC-007', 'P-05'), etc. The system starts a monitoring program to check the execution status of the tasks in the list one by one. For task 'SC-007', the system first queries the warehouse execution system (WES) for its planned completion timestamp, which is calculated based on the standard travel time of each section on the allocation path 'P-05'. For example, it is planned to be completed at system timestamp 1689000000. The program compares the current system timestamp 1689000005 with the planned timestamp and finds that the current time is later than the planned time. At the same time, the task status returned by the WES query is confirmed as "in execution" instead of "completed". Therefore, task 'SC-007' is screened as a task that is not allocated on time. Then, the system extracts the main shaft component number 'SC-007' and the corresponding allocation path number 'P-05' of the task, and further obtains the last state update timestamp of the task from the execution log of the WES, which is the task interruption timestamp, for example, 1688999900. The system calculates the difference between the task interruption timestamp and the originally planned completion timestamp 1688999840 to obtain a delay length of 60 seconds. Repeat the screening and calculation process for all other tasks in the scheduling list, for example, another task 'SC-011' is identified as a task that is not allocated on time, and its delay length is calculated as 180 seconds. Finally, all screened tasks that are not allocated on time are arranged in descending order according to their delay length values to form a new set containing component numbers, path numbers and delay lengths, such as [( 'SC-011', 'P-03', 180s), ('SC-007', 'P-05', 60s)], and generate an unfinished task set.

[0103] The difference analysis submodule obtains the queuing time data of the allocation task and the real-time feedback value of the warehouse interface based on the unfinished task set, calculates the absolute difference between the two, constructs a difference value-time stamp two-dimensional relationship table according to the task number, and generates a difference index set;

[0104] Based on the unfinished task set, the system processes the tasks one by one according to the order. First, process the task with the longest delay ('SC-011', 'P-03', 180s). The system retrieves the queuing time data T q of the task calculated before scheduling from the calculation cache of the allocation rhythm control module, for example, T q is 1200 seconds. At the same time, the system subscribes to the real-time data stream of the warehouse interface associated with the current interruption link of the task from the WES or digital twin system. The real-time feedback value T if of the warehouse interface is a quantitative waiting time, which is calculated as T if = N queue ×Tavg where N queue is the number of tasks currently queued at the interface, obtained by real-time counting of the sensor, for example 4, and T avg is the historical average time for the interface to process a single task, obtained from the historical database statistics, for example 90 seconds, so the real-time feedback value of the current interface is T if = 4 x 90 = 360 seconds, the system then calculates the absolute value of the difference ΔD = |T q -T if | = |1200-360| = 840 seconds, the system repeatedly obtains T if and calculates ΔD at fixed time intervals, for example every 5 seconds, thereby constructing a time series for task 'SC-011', for example, in the next 5 seconds, the number of tasks queued in front of the interface decreases to 3, T if becomes 270 seconds, then the new ΔD = |1200-270| = 930 seconds, the system binds each calculated ΔD value with the system time stamp at that time, and constructs a two-dimensional relationship table for these (time stamp, ΔD) data pairs according to the task number 'SC-011', and performs this operation for all tasks in the set of unfinished tasks to generate the set of difference indicators.

[0105] An abnormal positioning sub-module calls the set of difference indicators, compares the difference value change trend of each link of the allocation task, identifies the task link with a sudden increase in difference value and a sustained abnormal duration, marks the warehouse node corresponding to the sudden increase starting point as the interruption link, and generates the allocation abnormality identification result.

[0106] The system performs trend analysis on the two-dimensional relationship table of each task in the set of difference indicators, for example, the time series data of task 'SC-011' is [..., 840, 930,...], the system calculates the change rate of the difference value point by point, that is, ΔD i -ΔD i-1, the system gets 930-840=90, and compares this change value with a preset "surge threshold", which is set by referring to the statistical distribution of the change rate of the difference value in the historical normal operation state, taking three times of the standard deviation, for example, set to 50, since the calculated change value 90 is greater than the threshold 50, the system determines that a difference value surge has occurred here, then the system judges whether the difference value is continuously in an abnormal state, the judgment standard of abnormal state is whether the value of ΔD continuously exceeds the "abnormal value benchmark" at multiple monitoring points, the benchmark value is set according to business experience, for example, taking 80% of the average value of ΔD in history that leads to scheduling failure as the benchmark, set to 800, at the same time, "continuous abnormal duration" is defined as needing to continuously meet the condition for 2 monitoring periods (i.e. 10 seconds), in this example, 840 and 930 are both greater than 800, and are two consecutive monitoring points, which meet the condition of continuous abnormal duration, when the two conditions of surge and continuous abnormality are met at the same time, the system locks the starting point of the surge, and queries the WES log to obtain the specific task link of task 'SC-011' at the starting point timestamp, for example, the log shows that the AGV is trying to dock with the storage node 'C-08-01', the system immediately marks the storage node 'C-08-01' as the interruption link, and integrates the task number, interruption link code and other information to generate the allocation exception identification result.

[0107] Please refer to Figure 6 , the state transmission early warning module comprises:

[0108] The risk parameter acquisition submodule calls the main shaft component number of the abnormal link in the allocation exception identification result, the storage interface usage frequency, and the corresponding allocation path number, synchronously obtains the queuing duration, the storage capacity change rate, and the interface stability coefficient under the path, and generates a risk parameter set;

[0109] Call the allocation exception identification result, the system first parses the data, extracts the main shaft component number 'SC-011' associated with the task marked as an abnormal interruption, the storage interface number 'C-08-01' of the interruption link, and the allocation path number 'P-03' to which the task belongs, the system immediately takes 'C-08-01' as an index to query the historical performance database of the storage interface, and counts the number of tasks successfully processed by the interface in the past 24 hours, and divides by 24 to get its usage frequency F i 15 tasks per hour, at the same time, the system takes the path number 'P-03' as the key, starts a parallel data acquisition process, which queries the historical archives of the task management system, filters out all task records using path 'P-03' in the past week, and calculates the average queuing duration T q, which is 250 seconds, then the process queries the warehouse management system (WMS) for the historical inventory data of all the nodes of the path 'P-03', calculates the daily average capacity change rate C by calculating the difference between the total daily inbound and outbound quantities and dividing the result by the total throughput r , which is 0.15, finally, the process analyzes the service time records of the interface 'C-08-01' in the past 100 task executions, calculates the standard deviation σ of these service time values t , for example, σ t = 8 seconds, and takes the reciprocal of it as the interface stability coefficient S i = 1 / σ t = 0.125, the system combines the four parameters (F i = 15, T q = 250, C r = 0.15, S i = 0.125) collected for the path 'P-03' into a data record, and repeats this process for all the paths appearing in the abnormality identification results to generate a risk parameter set.

[0110] The path risk assessment submodule calculates the product coefficient of the allocation path change frequency and the node interface response delay based on the risk parameter set, superimposes the weighted average of the node capacity utilization rate deviation and the task interruption rate, and constructs a four-dimensional risk assessment matrix according to the path number to generate a path risk coefficient;

[0111] Based on the risk parameter set, the system starts the risk assessment calculation for each path record, for example, path 'P-03', first, the system needs to obtain the allocation path change frequency F p , which is calculated by querying the WES log to count the total number of tasks that were originally planned to follow the standard path but were dynamically switched to the backup path 'P-03' in the past month, assuming it is 30 times, and dividing the result by the total number of tasks in the month, which is 600 times, to get F p = 30 / 600 = 0.05, then the system calculates the node interface response delay D i , by querying the MES log to get the average actual response time of the interface 'C-08-01' in the past 50 tasks, which is 98 seconds, while the response time of this interface in the design standard is 90 seconds, then D i = 98-90 = 8 seconds, the system multiplies these two values to get the product coefficient M = F p × D i = 0.05 × 8 = 0.4, then the system gets the node capacity utilization rate deviation ΔU c , the optimal capacity utilization rate of the nodes along the path 'P-03' in the WMS is set to 80%, while the real-time data shows that the current average utilization rate is 92%, then ΔUc = |0.92 - 0.80| = 0.12, the system still needs to calculate the task interruption rate R int , by querying the historical records, it is found that in the past 1000 tasks using the 'P-03' path, there were 25 interruptions, then R int = 25 / 1000 = 0.025, the system calculates the weighted average value A = w u = 0.4 and w rint = 0.6 (the weights are set according to the direct impact of interruption on production being greater than the impact of capacity fluctuation) = w u × ΔU c + w rint × R int = 0.4 x 0.12 + 0.6 x 0.025 = 0.048 + 0.015 = 0.063, finally, the system adds the product coefficient to the weighted average value to calculate the comprehensive path risk coefficient R p = M + A = 0.4 + 0.063 = 0.463, the system constructs an evaluation matrix containing four original risk indicators and the final risk coefficient according to the path number, and generates the path risk coefficient;

[0112] The weighted average value uses the formula:

[0113]

[0114] Formula details and formula calculation derivation process:

[0115] 1. Parameter assignment and data source:

[0116] μ k = -0.05: capacity utilization deviation value, calculated as consistent with

[0117] JB / T9018-2011 Warehouse Equipment Capacity Tolerance Standard (±10%);

[0118] τ k = 0.12: task interruption rate, calculated by 15 interruptions in 120 allocation tasks in the past 30 days, rounded to two decimal places according to GB / T19001-2016;

[0119] ψ k = 0.86: real-time load fluctuation factor, calculation steps:

[0120]

[0121] corrected to 0.86 after calibration by vibration sensor;

[0122] σk = 45 ms: interface response delay standard deviation, calculated based on 50 dialing records;

[0123]

[0124] ω k = 3.2 times / hour: dialing path change frequency, recorded by RFID system for 26 times within 8 hours, Take one decimal place according to GB / T27923-2011;

[0125] θ k = 1.15: interface stability attenuation coefficient, calculated as follows:

[0126]

[0127] α k = 0.93: environmental interference correction factor, calculated at temperature 28℃ (reference 25℃) and humidity 65% (reference 60%):

[0128] α k = 1-0.1×3+0.05×5 = 1-0.3+0.25 =

[0129] 0.95 (corrected to 0.93 after infrared sensor calibration);

[0130] 2. Step-by-step calculation process:

[0131] Molecular calculation:

[0132] Absolute value term:

[0133] |(-0.05)×0.12×0.86| = |-0.00516| = 0.00516;

[0134] Square root term step-by-step:

[0135] Interface delay deviation term:

[0136] (σ k -ν k ) 2 = (45 ms-200 ms) 2 = (-155) 2 = 24025;

[0137] Path change impact term:

[0138] (ω k ·θ k ) 2 = (3.2×1.15) 2 = 3.68 2 = 13.5424;

[0139] Square root synthesis:

[0140]

[0141] Molecular total value:

[0142] 0.00516 + 155.05 = 155.05516;

[0143] Denominator calculation:

[0144] Capacity item calculation:

[0145]

[0146] Interrupt item calculation:

[0147]

[0148] Denominator synthesis:

[0149] 0.93 x (0.4156 + 0.3780) = 0.93 x 0.7936 = 0.7380;

[0150] Final result:

[0151]

[0152] 3. Result analysis:

[0153] The calculated value 210.1 exceeds the preset threshold value 1.2 (according to the ISO 3691-5:2014 industrial vehicle safety standard), triggering the AGV emergency braking protocol, synchronously activating the backup path B7 and sending a device maintenance request to the operation and maintenance system. The numerical value reflects the composite risk of interface delay exceeding the standard (155ms deviation) and load imbalance (82 tons / 95 tons) in the current path.

[0154] The management scheme generation submodule calls the path risk coefficient, prioritizes the change frequency, response delay, utilization deviation, and interruption rate according to the preset business rules, matches dynamic scheduling strategies and capacity expansion thresholds for high-risk paths, and generates a scheduling risk management scheme;

[0155] Call the path risk coefficient. The system first compares the comprehensive risk coefficient R p of all paths in the matrix, and sets the risk level, where R p values greater than 0.4 are classified as "high risk", between 0.2 and 0.4 as "medium risk", and below 0.2 as "low risk". The risk coefficient of path 'P-03' is 0.463, which is determined as a high-risk path. Then, the system prioritizes the four indicators (change frequency F p, response delay D i , utilization bias ΔU c , interruption rate R int ) are prioritized, with the interruption rate being the highest priority, the response delay being the second, the change frequency being the third, and the utilization bias being the lowest. For the high-risk path 'P-03', the system analyzes the contribution of the four indicators and finds that the product of the response delay D i and the change frequency F p contributes the most. Accordingly, the system matches a dynamic scheduling strategy, which specifically adjusts the passage cost parameter of the path 'P-03' in the path selection logic of the WES by 20%, so that the automatic scheduling algorithm will preferentially select other paths with lower costs in non-emergency situations. In addition, for the high utilization bias ΔU c , the system matches a capacity expansion threshold adjustment strategy, which specifically adjusts the inventory level warning threshold of the path 'P-03' along the line warehouse area from the original 95% to 88%. Once the real-time inventory reaches 88%, the system will automatically generate a pre-migration task order to suggest transferring some materials to the backup warehouse area. The path number, risk level, main risk indicators, and matched scheduling strategy and threshold parameters are integrated into a structured management instruction to generate a scheduling risk management scheme.

[0156] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by 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.

Claims

1. A digital twin-based intelligent warehouse management system for spindle components, characterized in that, The system includes: The spindle mapping initialization module obtains the spindle component location and transfer path, compares the location node update time with the movement record time, identifies the difference nodes and extracts the spindle component number, and generates the location difference identification result. The position fluctuation identification module extracts the channel nodes of the main shaft component under the allocation path based on the position difference identification results, analyzes the deviation between the channel node fluctuation value and the set fluctuation threshold, and generates a position fluctuation analysis report. The allocation rhythm control module obtains the queuing time and position data of the main shaft component allocation tasks marked as path switching changes in the position fluctuation analysis report, assigns weights to the main shaft components based on the data, and generates an allocation task scheduling list. The anomaly origin localization module calls the task number in the allocation task scheduling list, extracts the spindle component that was not allocated on time, analyzes the difference between the queuing time and the interface feedback, marks the interruption link as an anomaly, and generates an allocation anomaly identification result. The status transmission and early warning module analyzes the queuing time, capacity changes, and interface stability based on the abnormal links in the allocation anomaly identification results, identifies risk paths, and generates a scheduling risk management plan based on the four indicators of the allocation path.

2. The intelligent warehouse management system for spindle components based on digital twins according to claim 1, characterized in that: The position difference identification results include the difference component number and path difference identifier. The position fluctuation analysis report specifically includes the channel fluctuation coefficient, threshold deviation level, and fluctuation pattern classification. The allocation task scheduling list includes the weight allocation matrix and scheduling priority queue. The allocation anomaly identification results specifically refer to the interruption link code and interface response difference index. The scheduling risk management scheme includes the path change frequency index, interface delay parameter, capacity deviation coefficient, and interruption rate statistics.

3. The intelligent warehouse management system for spindle components based on digital twins according to claim 1, characterized in that: The spindle mapping initialization module includes: The warehouse data acquisition submodule obtains the current warehouse location number and transfer path number of the spindle component in the warehouse area, extracts the update timestamp of the warehouse node on each transfer path and the corresponding movement record timestamp of the spindle component, and establishes a timestamp dataset containing warehouse node number, double timestamp, and location information. The node difference comparison submodule calls the warehouse node number in the timestamp dataset, compares the numerical relationship between the update timestamp and the movement record timestamp under the same node, filters the nodes whose update timestamp values ​​are greater than the movement record timestamp, synchronously verifies the matching status of the node's location information and the location coordinates in the movement record, and obtains a set of difference nodes with time lag and location offset. The difference result generation submodule extracts the unique identifier and the allocation path number of the corresponding main shaft component based on the node number in the difference node set, calculates the offset between the position coordinates of the difference node and the standard warehouse coordinates, and generates a warehouse difference identification result containing the component number, path number, and offset.

4. The intelligent warehouse management system for spindle components based on digital twins according to claim 1, characterized in that: The position fluctuation identification module includes: The channel node extraction submodule calls the warehouse difference identification results, extracts the channel node numbers connecting different warehouses in each allocation path, establishes a channel node mapping relationship according to the main shaft component number and the allocation path number, and generates a channel node set. The fluctuation value calculation submodule counts the change in position coordinates of each channel node in the channel node set when the allocation path is switched, takes the difference in coordinate change between two consecutive switches as the fluctuation value, constructs a three-dimensional data matrix according to the node number and the number of switches, and generates a fluctuation value matrix. The deviation marking analysis submodule compares the fluctuation value of each node in the fluctuation value matrix with the preset path switching fluctuation threshold item by item, calculates the absolute value of the deviation value, marks the nodes whose deviation value exceeds the threshold, classifies and aggregates the marked data according to the threshold range, and generates a position fluctuation analysis report.

5. The intelligent warehouse management system for spindle components based on digital twins according to claim 1, characterized in that: The allocation rhythm control module includes: The task parameter acquisition submodule calls the position fluctuation analysis report to extract the queuing time, starting position capacity and target position interface usage frequency data of the main shaft component allocation task marked as path switching change, establishes a three-dimensional parameter mapping table according to the component number, and generates a task parameter set. The weight allocation calculation submodule, based on the task parameter set, normalizes the queuing time, warehouse capacity and interface usage frequency, allocates weight coefficients according to preset business rules, calculates the comprehensive scheduling weight value of each main axis component, and generates a weight coefficient matrix. The scheduling queue generation submodule sorts the main shaft components in descending order according to the comprehensive scheduling weight value in the weight coefficient matrix, binds the sorted component numbers with the corresponding allocation path numbers, reconstructs the task execution sequence, and generates an allocation task scheduling list.

6. The intelligent warehouse management system for spindle components based on digital twins according to claim 1, characterized in that: The anomaly origin localization module includes: The task filtering submodule extracts the spindle component numbers and corresponding allocation path numbers that were not allocated on time from the allocation task scheduling list, sorts them according to the difference between the task interruption timestamp and the planned timestamp, and generates an incomplete task set. The difference analysis submodule, based on the set of incomplete tasks, obtains the queuing time data and real-time feedback value of the warehouse interface in the allocation task, calculates the absolute value of the difference between the two, constructs a two-dimensional relationship table of difference value-time stamp according to the task number, and generates a set of difference indicators. The anomaly location submodule calls the set of difference indicators, compares the trend of difference value changes in each stage of the allocation task, identifies the task stage with a sudden increase in difference value and an abnormal duration, marks the warehouse node corresponding to the starting point of the sudden increase as the interruption stage, and generates the allocation anomaly identification result.

7. The intelligent warehouse management system for spindle components based on digital twins according to claim 1, characterized in that: The status transmission and early warning module includes: The risk parameter acquisition submodule calls the main shaft component number of the abnormal link, the frequency of use of the warehouse interface and the corresponding allocation path number in the allocation anomaly identification result, and simultaneously obtains the queuing time, warehouse capacity change rate and interface stability coefficient under the path to generate a risk parameter set. The path risk assessment submodule, based on the risk parameter set, calculates the product coefficient of the transfer path change frequency and the warehouse interface response delay, superimposes the weighted average of the warehouse capacity utilization deviation value and the task interruption rate, constructs a four-dimensional risk assessment matrix according to the path number, and generates the path risk coefficient. The management scheme generation submodule calls the path risk coefficient, prioritizes the change frequency, response latency, utilization deviation and interruption rate according to preset business rules, matches dynamic scheduling strategies and capacity expansion thresholds to high-risk paths, and generates a scheduling risk management scheme.

Images