Production line management and control method and system based on flexible production

Abstract

The present application relates to the technical field of production process management and control, in particular to a production line management and control method and system based on flexible production, comprising the following steps: calculating the delivery time and width deviation to generate edge weight constant, constructing a graph, extracting batch sequence, generating angle control data according to heating and circumference constant, establishing a discrete matrix combining displacement and time consumption parameters to solve the interpolation position equation, deriving the equation and updating the time length when the jump degree exceeds the limit to reconstruct the curve control data, and integrating and packaging into flexible production line management and control records. In the present application, the space-time cost is extracted to use the graph theory mechanism to relieve production congestion, deduce the operation phase to avoid parameter guessing hazards, project the displacement demand to a multi-dimensional matrix to constrain the trajectory equation to guide smooth handover, identify the instantaneous extreme value and expand the time domain span to suppress mechanical fatigue, aggregate production and servo action across levels to form a strict control closed loop, replace manual intervention and ensure that the equipment maintains continuous operation and healthy service.

Description

Technical Field

[0001] This invention relates to the field of production process control technology, and in particular to production line control methods and systems based on flexible production. Background Technology

[0002] The field of production process control technology mainly involves the collaborative interaction between the planning and scheduling at the workshop level and the machine execution level in manufacturing enterprises. It covers core matters such as production scheduling, work order dispatch, on-site material flow tracking, real-time collection of machine operating parameters, and archiving of product quality inspection data. The aim is to establish a refined management mechanism for the entire process from order receipt to finished product delivery. Traditional production line control methods and systems refer to the technical aspects of allocating continuous manufacturing tasks across multiple processes in the workshop and coordinating the actions of underlying machines. In the corrugated cardboard production process, customer orders for different flute types and widths are typically collected manually. Then, the base paper layout plan is calculated manually and paper work orders are issued to the workshop. Based on the specifications and dimensions on the paper work orders, the on-site operators manually adjust the meshing gap of the corrugating rollers of the single-face corrugator, adjust the opening of the steam valve of the preheating cylinder, and set the glue coating amount of the glue coater by rotating physical knobs and pressing buttons on the control panel. At the same time, when changing to different batches of orders, the production line is stopped manually, the fasteners of the longitudinal cutting and creasing machine are loosened manually to move the blade spacing, and the cutting length of the cross-cutting machine is re-entered on the control panel. In this way, the heating, corrugating, gluing, cutting, and stacking processes of the cardboard are completed in batches in sequence.

[0003] Traditional manufacturing production line management relies on manual order collection and production scheduling, leading to planning delays and delivery delays. The transfer of paper records on-site hinders data exchange at the lower levels and restricts the smooth flow of production line operations. Workers frequently stop machines to manually adjust knobs and fasteners based on subjective experience, affecting continuous operation. Mechanical transmission components experience increased wear and fatigue due to abrupt starts and stops and rough intervention. The lack of smooth linkage and verification between lower-level operating parameters increases displacement deviation and component wear, restricting the collaborative efficiency of multiple links in the workshop and ensuring continuous processing. Summary of the Invention

[0004] To achieve the above objectives, the present invention adopts the following technical solution: a production line management method based on flexible manufacturing, comprising the following steps: S1: Collect the target delivery time value and the paper width value of the orders to be scheduled for production and calculate the deviation respectively. Multiply the obtained delivery time difference value and the width difference value to generate the edge weight constant. S2: Construct a global order directed graph based on the edge weight constant, remove directed lines that exceed the boundary reference value, trace path connectivity, and combine the extracted closed graph region nodes to generate a continuous scheduling batch sequence. S3: Call the continuous scheduling batch sequence, use the product of the target heating time value and the instantaneous linear velocity value to obtain the theoretical heating arc length value, combine the actual heating arc length value to calculate the deviation, and generate drive angle control data based on the preheating cylinder circumference value and the circumferential angle constant. S4: Based on the driving angle control data, generate a benchmark order change time parameter, calculate the difference between the standby and actual absolute coordinate values ​​of the corresponding continuous scheduling batch sequence to obtain the total spatial displacement value, construct a spatiotemporal discrete node matrix and solve the polynomial coefficients to generate a benchmark interpolation position equation. S5: Obtain the real-time instantaneous jump value by calculating the third-order time derivative of the benchmark interpolation position equation. When the maximum allowable jump limit constant is exceeded, update the order change duration parameter by extending the time step value. Combine the total spatial displacement value to reconstruct the curve and generate curve control data. Encapsulate the continuous scheduling batch sequence, the drive angle control data and the curve control data into a data format to generate a flexible production line management record.

[0005] As a further embodiment of the present invention, the edge weight constant specifically refers to the delivery time urgency coefficient and the width penalty factor; the continuous scheduling batch sequence includes an order production sorting table, same-order exchange details, and material grouping queue; the driving angle control data specifically refers to compensation pulse commands, target rotation phase, and servo wrap angle parameters; the reference interpolation position equation specifically refers to displacement fitting coefficients, motion trajectory polynomials, and node coordinate matrices; and the flexible production line management record includes a jump limit log, a duration update report, and an overload alarm code.

[0006] As a further aspect of the present invention, the step of obtaining the edge weight constant specifically includes: S101: Receive order scheduling records from the scheduling server, parse the order scheduling records to obtain the orders to be scheduled, collect the target delivery time value and the original paper width value corresponding to the orders to be scheduled, and integrate the target delivery time value and the original paper width value to obtain the original parameter set of the order. S102: Call the original set of order parameters, perform time deviation calculation based on the target delivery time value to obtain the delivery time difference value, perform size deviation calculation based on the original paper width value to obtain the width difference value, and integrate the delivery time difference value and the width difference value to obtain the order deviation parameter set; S103: Call the order deviation parameter set, perform a product operation on the delivery time difference value and the width difference value, and generate the edge weight constant.

[0007] As a further aspect of the present invention, the step of obtaining the continuous scheduling batch sequence specifically includes: S201: Obtain the order network aggregation instruction, parse the order network aggregation instruction, call the orders to be scheduled to construct topology nodes, construct directed connections for the topology nodes, assign values ​​to the directed connections according to the edge weight constants, and construct a global order directed graph; S202: Obtain the aggregate truncation reference boundary value, remove the directed lines that exceed the boundary in the global order directed graph based on the aggregate truncation reference boundary value, and for the removed global order directed graph, trace the path connectivity of adjacent nodes along the remaining directed lines to extract the graph regions where the internal nodes are connected to each other as closed graph regions. S203: Call the closed graph region, perform combination operations on the topological nodes within the closed graph region, and generate a continuous scheduling batch sequence.

[0008] As a further aspect of the present invention, the process of obtaining the order network aggregation instruction specifically includes: Monitor the cache queue of the data communication port, extract the current production stack quantity value of the cache queue, read the maximum throughput limit of the server hardware to generate the aggregation trigger threshold constant, perform a numerical comparison operation between the production stack quantity value and the aggregation trigger threshold constant, and when it is determined that the production stack quantity value is greater than the aggregation trigger threshold constant, use the underlying communication protocol to perform message encapsulation operation to obtain the order network aggregation instruction; The process of performing combination operations on topological nodes within a closed graph region to generate a continuous scheduling batch sequence is as follows: The association feature mapping data of the topological nodes in the closed graph region is extracted. Based on the association feature mapping data, the order ranking weight values ​​of the topological nodes in the closed graph region are read. According to the order ranking weight values, the topological nodes in the closed graph region are sorted by unidirectional linearization to establish a one-dimensional node queue. The arrangement index position of each node in the one-dimensional node queue is extracted and converted to generate node time sequence mapping parameters. According to the node time sequence mapping parameters, a structured splicing operation is performed on the one-dimensional node queue to generate a continuous scheduling batch sequence.

[0009] As a further aspect of the present invention, the step of acquiring the drive angle control data specifically includes: S301: Obtain the heating control configuration of the preheating cylinder and the servo motor of the guide roller of the single-facer, parse the heating control configuration, call the continuous scheduling batch sequence, obtain the preheating cylinder circumference value and the target heating time value, collect the instantaneous linear velocity value of the encoder of the single-facer main drive shaft, perform a product operation on the target heating time value and the instantaneous linear velocity value to obtain the theoretical heating arc length value, and integrate the preheating cylinder circumference value and the theoretical heating arc length value to obtain the heating arc length parameter set; S302: Call the set of heated arc length parameters, collect the actual heated arc length value of the paper guide roller servo motor, perform deviation calculation on the theoretical heated arc length value and the actual heated arc length value to obtain the arc length deviation value, and perform ratio calculation on the arc length deviation value and the preheating cylinder circumference value to obtain the number of revolutions ratio value. S303: Call the stated number of revolutions ratio value to obtain the circumferential angle constant, perform a product operation on the number of revolutions ratio value and the circumferential angle constant to obtain the rotation angle compensation value, and generate drive angle control data based on the rotation angle compensation value.

[0010] As a further aspect of the present invention, the step of obtaining the reference interpolation position equation specifically includes: S401: Receive the lateral displacement planning command of the slitting and creasing machine cutter holder, evaluate the response time of the guide roller based on the drive angle control data, and use the response time of the guide roller as the benchmark order change time parameter. S402: Collect the current actual absolute coordinate value of the cutting blade holder of the longitudinal cutting and creasing machine, obtain the standby coordinate value corresponding to the continuous scheduling batch sequence, perform difference calculation on the standby coordinate value and the current actual absolute coordinate value, and obtain the total spatial displacement value. S403: Based on the total spatial displacement length and the reference exchange time parameter, construct a spatiotemporal discrete node matrix, perform polynomial coefficient calculations on the spatiotemporal discrete node matrix, and construct a reference interpolation position equation.

[0011] As a further aspect of the present invention, the process of evaluating the response time of the guide roller is specifically as follows: Extract the rated angular velocity operating parameters of the guide roller servo motor, call the drive angle control data, perform division calculation on the drive angle control data and the rated angular velocity operating parameters to obtain the basic rotation time value, read the equipment's underlying control board to obtain the mechanical transmission delay compensation constant, sum the basic rotation time value and the mechanical transmission delay compensation constant to obtain the guide roller response time; The process of constructing the spatiotemporal discrete node matrix, performing polynomial coefficient operations on the spatiotemporal discrete node matrix, and constructing the benchmark interpolation position equation is as follows: The initial zero-velocity boundary conditions and the final zero-velocity boundary conditions of the lateral motion are obtained. The total spatial displacement length, the reference change time parameter, the initial zero-velocity boundary conditions and the final zero-velocity boundary conditions are synchronously mapped to a two-dimensional data array to construct a spatiotemporal discrete node matrix. Matrix inversion is performed on the spatiotemporal discrete node matrix to obtain the polynomial coefficient solution set. Based on the polynomial coefficient solution set and the time independent variable characteristic term, an algebraic recombination operation is performed to construct the reference interpolation position equation.

[0012] As a further aspect of the present invention, the steps for obtaining the flexible production line control records are specifically as follows: S501: Extract the hardware overload protection threshold of the ball screw transmission mechanism, extract the reference interpolation position equation, perform third-order time derivative calculation on the reference interpolation position equation at the time domain discrete node to obtain the real-time instantaneous jump value, obtain the maximum allowable jump limit constant of the ball screw transmission mechanism, and integrate the real-time instantaneous jump value and the maximum allowable jump limit constant to obtain the jump comparison parameter set. S502: Call the set of jump comparison parameters, perform numerical comparison based on the real-time instantaneous jump value and the maximum allowable jump limit constant, when it is determined that there is a real-time instantaneous jump value greater than the maximum allowable jump limit constant, obtain the time step extension value, and perform a summation operation on the benchmark order change duration parameter and the time step extension value to obtain the updated order change duration parameter; S503: Recalculate the polynomial coefficients based on the update changeover duration parameter and the total spatial displacement value, reconstruct the interpolation displacement curve to generate curve control data, encapsulate the data format of the continuous scheduling batch sequence, the drive angle control data and the curve control data, and generate flexible production line management records.

[0013] A production line management system based on flexible manufacturing includes: Order edge weight fusion module: Collects the target delivery time value and the original paper width value of the orders to be scheduled for production and calculates the deviation respectively. Multiplies the obtained delivery time difference value and the width difference value to generate the edge weight constant. Directed graph batch aggregation module: Constructs a global order directed graph based on the edge weight constant, removes directed lines that exceed the boundary reference value, tracks path connectivity, and combines the extracted closed graph region nodes to generate a continuous scheduling batch sequence; Preheating cylinder angle calculation module: calls the continuous scheduling batch sequence, uses the product of the target heating time value and the instantaneous linear velocity value to obtain the theoretical heating arc length value, calculates the deviation in combination with the actual heating arc length value, and generates drive angle control data based on the preheating cylinder circumference value and the circumferential angle constant. Interpolation equation construction module: Based on the driving angle control data, a benchmark changeover time parameter is generated, the difference between the standby and actual absolute coordinate values ​​of the corresponding continuous scheduling batch sequence is obtained to get the total spatial displacement value, a spatiotemporal discrete node matrix is ​​constructed and the polynomial coefficients are solved to generate the benchmark interpolation position equation. Constraint curve smoothing module: Obtain the real-time instantaneous jump value by calculating the third-order time derivative of the benchmark interpolation position equation. When the maximum allowable jump limit constant is exceeded, the changeover duration parameter is updated by extending the time step value. The curve is reconstructed by combining the total spatial displacement value to generate curve control data. The continuous scheduling batch sequence, the drive angle control data and the curve control data are encapsulated in data format to generate flexible production line management records.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, a cost map is established by extracting order time and size features, and graph theory connectivity is used to separate related batches to resolve sorting deadlocks. Combined with the target heating requirements, the compensation arc length and operation phase are deduced to eliminate the hidden dangers of human parameter guessing. Spatial displacement and operation rhythm are synchronously projected into the multi-dimensional matrix constraint to reconstruct the continuous trajectory equation to guide smooth handover. Furthermore, the instantaneous impact extreme value of the equipment is identified by multi-order time differentiation, and the time domain span is adaptively expanded in the overload critical zone to suppress wear and fatigue. The overall task timing and electromechanical action parameters are aggregated across levels to form a stable control closed loop. The obstacles of the underlying paper flow are eliminated and the original extensive shutdown intervention is replaced. The manufacturing barriers are broken down and the equipment is guaranteed to maintain continuous operation and long-term healthy service. 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 a schematic diagram of the steps of the present invention; Figure 2 This is a detailed schematic diagram of S1 of the present invention; Figure 3 This is a detailed schematic diagram of S2 of the present invention; Figure 4 This is a detailed schematic diagram of S3 of the present invention; Figure 5 This is a detailed schematic diagram of S4 of the present invention; Figure 6 This is a detailed schematic diagram of S5 of the present invention; Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0018] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0019] Please see Figure 1 This invention provides a production line management method based on flexible manufacturing, comprising the following steps: S1: Collect the target delivery time value and the paper width value of the orders to be scheduled for production and calculate the deviation respectively. Multiply the obtained delivery time difference value and the width difference value to generate the edge weight constant. S2: Construct a global order directed graph based on edge weight constants, remove directed lines that exceed the boundary reference values, trace path connectivity, and combine the extracted closed graph region nodes to generate a continuous scheduling batch sequence. S3: Call the continuous scheduling batch sequence, use the product of the target heating time value and the instantaneous linear velocity value to obtain the theoretical heating arc length value, combine the actual heating arc length value to calculate the deviation, and generate drive angle control data based on the preheating cylinder circumference value and the circumferential angle constant. S4: Generate benchmark order change time parameters based on drive angle control data, calculate the difference between the standby and actual absolute coordinate values ​​of the corresponding continuous scheduling batch sequence to obtain the total spatial displacement value, construct a spatiotemporal discrete node matrix and solve the polynomial coefficients to generate benchmark interpolation position equations. S5: Obtain the real-time instantaneous jump value by calculating the third-order time derivative of the benchmark interpolation position equation. When the maximum allowable jump limit constant is exceeded, update the order change duration parameter by extending the time step value. Combine the total spatial displacement value to reconstruct the curve and generate curve control data. Encapsulate the continuous scheduling batch sequence, drive angle control data and curve control data into a data format to generate flexible production line management records.

[0020] The edge weight constants are specifically the delivery urgency coefficient and width penalty factor. The continuous scheduling batch sequence includes the order production sorting table, the same-order exchange details, and the material grouping queue. The drive angle control data specifically refers to the compensation pulse command, the target rotation phase, and the servo wrap angle parameter. The benchmark interpolation position equation is specifically the displacement fitting coefficient, the motion trajectory polynomial, and the node coordinate matrix. The flexible production line control records include the jump limit log, the duration update report, and the overload alarm code.

[0021] Please see Figure 2 The specific steps for obtaining the edge weight constants are as follows: S101: Receive order scheduling records from the scheduling server, parse the order scheduling records to obtain the orders to be scheduled, collect the target delivery time value and the original paper width value corresponding to the orders to be scheduled, and integrate the target delivery time value and the original paper width value to obtain the original parameter set of the order. Establish a network socket communication link with the scheduling server, listen for TCP packets on the specified communication port, and receive order scheduling record messages encapsulated in JSON format. Perform deserialization on the received messages and extract the header identifier bits to verify data integrity. After successful verification, traverse the body nodes of the messages and filter out the pending production orders with active status bits based on the order status tags. For each filtered pending production order, read its delivery time field to obtain the target delivery time value, which is represented in ten-digit Unix timestamp format, specifically extracted as 1715472000. Simultaneously, read the physical size field to obtain the original paper width value, which is measured in millimeters, specifically extracted as 1200 millimeters. Create a memory cache space based on a key-value pair structure, bind and store the extracted target delivery time value and original paper width value as attribute values, using the order's unique identifier as the primary key, and perform data serialization and merging operations. This generates and writes a set of original order parameters containing structured feature attributes to the control platform's memory, providing a standardized data source support for subsequent deviation calculations.

[0022] S102: Call the original set of order parameters, calculate the time deviation based on the target delivery time value to obtain the delivery time difference value, calculate the size deviation based on the original paper width value to obtain the width difference value, and integrate the delivery time difference value and the width difference value to obtain the order deviation parameter set. The system retrieves the serialized set of original order parameters from the memory cache, reads the clock chip register on the main control board to obtain the current clock reference time value, which is 1715000000. Substituting the target delivery time value (1715472000) and the current clock reference time value into the time difference calculation logic, the system subtracts the current clock reference time value from the target delivery time value, calculating a delivery time difference of 472000 seconds. The system then reads the actual width sensor reading on the single-facer machine via the industrial fieldbus, obtaining the current actual width value as 1000 mm. Substituting the original paper width value (1200 mm) and the current actual width value into the size deviation calculation logic, the system subtracts the current actual width value from the original paper width value and takes the positive difference, calculating a width difference value of 200 mm. A new two-dimensional array structure is constructed, and the calculated delivery time difference value of 472,000 seconds and the width difference value of 200 mm are stored in the consecutive memory addresses of the two-dimensional array. This completes the horizontal binding mapping of the two-dimensional parameters and obtains the order deviation parameter set, thereby unifying the multi-source heterogeneous physical and time deviation data into the same calculation dimension.

[0023] S103: Call the order deviation parameter set, perform a product operation on the delivery time difference value and the width size difference value, and generate the edge weight constant; The system analyzes the order deviation parameter set located in a two-dimensional array in memory, extracting the delivery date difference and width difference values. Considering the significant physical isolation between the time unit of seconds and the physical size unit of millimeters, the system extracts the factory-set time normalization adjustment coefficient and width normalization adjustment coefficient, which are then burned into read-only memory. The time normalization adjustment coefficient is fixed at 0.0001, and the width normalization adjustment coefficient is fixed at 0.5. Multiplying the delivery date difference of 472000 seconds by the time normalization adjustment coefficient yields a time scale value of 47.2. Multiplying the width difference of 200 millimeters by the width normalization adjustment coefficient yields a size scale value of 100. Subsequently, a product operation is performed on the corrected time scale value and size scale value, i.e., multiplying 47.2 by 100, resulting in a product of 4720. The product result is defined as a punitive quantitative indicator to characterize the overall cost of order replacement, and given a high-precision floating-point data format, thereby generating an edge weight constant of 4720. This constant intuitively reflects the dual impact of the order on the urgency of delivery and the physical cost of equipment replacement in the current production environment, providing a quantitative basis for the connection strength of the network topology of subsequent orders.

[0024] Please see Figure 3 The specific steps for obtaining the batch sequence of continuous scheduling are as follows: S201: Obtain the order network aggregation instruction, parse the order network aggregation instruction, call the orders to be scheduled to build topology nodes, build directed connections for the topology nodes, assign values ​​to the directed connections according to the edge weight constants, and build a global order directed graph; The system listens for interrupt request signals on the underlying communication bus. When a high-level trigger state is detected, it retrieves the order network aggregation instruction. The received instruction data packet is binary-coded and parsed to extract the start flag and scope parameters. All pending production orders in the database are traversed, and each pending order is instantiated as a virtual topology node object. A unique memory handle and node identifier number are allocated to each topology node. Based on the forward order switching logic in the production process, directed connections representing processing switching paths are constructed pairwise for all topology nodes using a Cartesian product permutation. Pre-calculated edge weight constants are extracted, and based on the mapping relationship between the start and end points of the directed connections, the corresponding edge weight constants are written into the attribute field of the directed connection as path penalty weights, thus assigning values ​​to the directed connections. All instantiated topology nodes and directed connections with attached edge weight constants are loaded into the graph database engine for structured storage, constructing a global directed order graph containing complete production dependency network relationships, realizing the transformation of production scheduling business logic into a graph theory mathematical model.

[0025] Table 1. Node Connection Weight Assignment Mapping Table 1 2 4720 2 3 3500 3 1 8900 Table 1 shows the weight assignment results of some node connections and their corresponding connectivity.

[0026] S202: Obtain the aggregate truncation reference boundary value, remove the directed lines that exceed the boundary in the global order directed graph based on the aggregate truncation reference boundary value, and for the removed global order directed graph, trace the path connectivity of adjacent nodes along the remaining directed lines to extract the graph regions where the internal nodes are connected as closed graph regions. The system reads the configuration text document from the production control platform, retrieves the aggregate truncation reference boundary value pre-evaluated and written to read-only memory by the workshop production process engineer, and sets this value to a constant 5000. It then traverses all directed connections within the global order directed graph, reads the edge weight constant from the attribute field of each connection, and performs a one-way numerical comparison operation with the aggregate truncation reference boundary value of 5000. When an edge weight constant of a directed connection is found to be greater than 5000, a memory cleanup function is triggered to destroy the data object of that directed connection, thus completing the forced removal of directed connections exceeding the boundary reference value. For the global order directed graph after the removal operation, a depth-first search algorithm is used to trace the connectivity of adjacent nodes in the remaining topological nodes. Specifically, starting from any isolated node, the system recursively visits subsequent nodes along the direction of the directed connection, pushes the visited node identifier onto the search stack, and marks the node's visited state as visited. When a group of nodes is found to form a loop or a bidirectional connected closed loop through connections, it is determined that the group of nodes satisfies the definition of a strongly connected component in graph theory. Subsequently, the group of nodes and its internal connections are extracted from the global order directed graph and encapsulated and isolated as an independent closed graph region with independent memory addresses.

[0027] S203: Call the closed graph region, perform combination operations on the topological nodes within the closed graph region, and generate a continuous scheduling batch sequence; The system retrieves a closed graph region, independently encapsulated within the memory address pool, and reads the associated feature mapping data inherent in the structure of each topological node within this region. This mapping data contains quantifiable similarity indicators of orders at the process and material level. Based on this associated feature mapping data, a hash function is used to match and read the order ranking weight value corresponding to each topological node within the closed graph region. All obtained order ranking weight values ​​are used as sorting keys. A quicksort algorithm mechanism is introduced, setting a benchmark comparison pivot. Through two-pointer movement towards each other and element swapping operations, a unidirectional linear sorting process is performed on all topological nodes within the closed graph region. The sorting process follows the principle of increasing weight values ​​from small to large, thereby establishing a one-dimensional node queue with sequential execution logic that eliminates the network entanglement structure. This one-dimensional node queue is traversed, and the actual arrangement index position value of each node in the underlying array is extracted. This value is then directly converted into node timing mapping parameters used to guide the order change operation time of on-site equipment. Based on the node's timing mapping parameters, the head and tail structured splicing and data message merging operations are performed on each topology order object in the one-dimensional node queue to generate a continuous scheduling batch sequence with a strict execution order and prohibiting interruption. This sequence data is then serialized and pushed into the buffer of the production execution queue to await invocation.

[0028] Please see Figure 4 The specific steps for obtaining drive angle control data are as follows: S301: Obtain the heating control configuration of the preheating cylinder and the servo motor of the guide roller of the single-facer, parse the heating control configuration, call the continuous scheduling batch sequence, obtain the preheating cylinder circumference value and the target heating time value, collect the instantaneous linear velocity value of the encoder of the single-facer main drive shaft, perform a product operation on the target heating time value and the instantaneous linear velocity value to obtain the theoretical heating arc length value, and integrate the preheating cylinder circumference value and the theoretical heating arc length value to obtain the heating arc length parameter set; By accessing the register address block of the production line's programmable logic controller via the industrial Ethernet interface, the system obtains the heating control configuration data packet for the preheating cylinder process of the single-facer and the servo motor of the guide roller. Bitwise mask parsing is performed on this configuration data packet to obtain the heating process constraint indicators contained in the current execution item of the continuous scheduling batch sequence. From this, the preheating cylinder circumference is fixed at 3140 mm, and the target heating time required for the current paper is set to 2 seconds. A high-resolution photoelectric rotary encoder mounted on the main drive shaft of the single-facer is simultaneously activated, capturing pulse signals with a sampling period of 10 milliseconds. After digital-to-analog conversion, the instantaneous linear velocity value at the current moment is calculated, and this instantaneous linear velocity value is read as stable at 500 mm / s. The target heating time value of 2 seconds and the instantaneous linear velocity value of 500 mm / s are substituted into the kinematic multiplication unit for product calculation, yielding a theoretical heating arc length of 1000 mm. This value represents the physical length of the paper that should adhere to the surface of the preheating cylinder to meet the heat exchange time required for the paper pasting process. Subsequently, a continuous dedicated memory range is allocated, and the preheating cylinder circumference value of 3140 mm and the calculated theoretical heating arc length value of 1000 mm are written into this memory range for binding and latching, thereby obtaining the set of heating arc length parameters for direct addressing and calling by the subsequent servo closed-loop control algorithm.

[0029] S302: Call the set of heating arc length parameters, collect the actual heating arc length value of the paper guide roller servo motor, perform deviation calculation on the theoretical heating arc length value and the actual heating arc length value to obtain the arc length deviation value, and perform ratio calculation on the arc length deviation value and the preheating cylinder circumference value to obtain the number of revolutions ratio value. An interrupt handler is triggered to call the set of heated arc length parameters stored in the read-only memory area. Simultaneously, the actual position feedback pulse of the paper guide roller servo motor is read in real time through the feedback communication channel of the servo driver, multiplied by the transmission ratio coefficient, and the current actual heated arc length value is collected and calculated to be 800 mm. The theoretical heated arc length value of 1000 mm stored in the set and the actual heated arc length value of 800 mm obtained from the feedback are input into the deviation calculation function model. Subtraction logic is executed, subtracting the actual heated arc length value from the theoretical heated arc length value to calculate the arc length deviation value as positive 200 mm, indicating that the current paper bonding length is insufficient and the motor needs to perform compensation. The preheating cylinder circumference value of 3140 mm is extracted from the heated arc length parameter set. The calculated arc length deviation value of 200 mm is used as the dividend, and the preheating cylinder circumference value of 3140 mm is used as the divisor, performing a high-precision floating-point division ratio operation. The calculated rotation ratio is 0.0637. This ratio physically represents the relative rotation ratio of the number of rotations the guide roller servo motor needs to drive the robotic arm to rotate around the center of the preheating cylinder in order to compensate for the 200 mm heating arc length gap. This lays the mathematical foundation for the subsequent direct command conversion of motor angle.

[0030] S303: Call the revolution ratio value to obtain the circumferential angle constant, multiply the revolution ratio value and the circumferential angle constant to obtain the rotation angle compensation value, and generate drive angle control data based on the rotation angle compensation value. The underlying constant register of the servo motion control card is accessed, and the revolution ratio value of 0.0637 retained from the previous calculation cycle is retrieved. Simultaneously, standard circular geometry mapping feature values ​​are extracted from the core firmware dictionary, i.e., the circular angle constant is obtained, which is fixed at 360 degrees in the physical rules. The revolution ratio value of 0.0637 and the circular angle constant of 360 degrees are input together into the kinematic product conversion logic node, and a scalar multiplication operation is performed. The rotation angle compensation value is calculated to be 22.932 degrees using floating-point multiplication instructions. This calculation result is physically represented as the amount of physical angular displacement required for the servo motor output shaft to rotate clockwise. This rotation angle compensation value of 22.932 degrees is substituted into the pulse equivalent conversion formula specific to the servo driver, and combined with the encoder physical resolution of 10,000 pulses per revolution of the servo motor, the physical angle is converted into the number of underlying electrical execution signal pulses. These converted control data are encapsulated into standard industrial motion control bus protocol data frames, with added checksums, verification codes, and execution timestamp tags. This generates drive angle control data that can be directly read, parsed, and executed by the servo amplifier. The generation of this control data signifies that the thermodynamic process parameters based on production scheduling requirements have been transformed into microscopic motion commands for the underlying electrical drive system.

[0031] Please see Figure 5 The specific steps for obtaining the benchmark interpolation position equation are as follows: S401: Receives the lateral displacement planning command of the slitting and creasing machine cutter holder, evaluates the response time of the guide roller based on the drive angle control data, and uses the response time of the guide roller as the benchmark order change time parameter. The production line's programmable logic controller (PLC) receives lateral displacement planning commands from the slitting and creasing machine's cutter holder via its internal shared data block. To achieve precise synchronization of the upstream and downstream equipment's operating cycles, the rated angular velocity operating parameter of the guide roller servo motor is extracted. This parameter is directly read from the built-in chip in the motor's nameplate configuration table, and its rated angular velocity operating parameter is determined to be constant at 45 degrees per second. The physical angular displacement of 22.932 degrees from the drive angle control data output from the previous control flow is retrieved, and a division mathematical calculation is performed on this drive angle control data and the rated angular velocity operating parameter. Dividing the physical angular displacement of 22.932 degrees by the angular velocity of 45 degrees per second yields a basic rotation time of 0.5096 seconds. The non-volatile memory built into the equipment's underlying control board is read, and the mechanical transmission delay compensation constant, calibrated and solidified through forced vibration experiments, is retrieved and set to 0.2 seconds. The base rotation time of 0.5096 seconds and the mechanical transmission delay compensation constant of 0.2 seconds are fed into the accumulator register for summation, resulting in a total action time of 0.7096 seconds. This guide roller response time of 0.7096 seconds is directly mapped and statically assigned as the baseline order change time parameter, serving as the insurmountable time axis boundary standard for the slitting and creasing machine to switch the tool's lateral movement position.

[0032] S402: Collect the current actual absolute coordinate value of the cutting blade holder of the longitudinal cutting and creasing machine, obtain the standby coordinate value of the corresponding continuous scheduling batch sequence, perform difference calculation on the standby coordinate value and the current actual absolute coordinate value, and obtain the total spatial displacement value. The laser position measuring instrument installed on the side of the precision linear guide of the slitting and creasing machine is activated. It acquires the current actual absolute coordinate value of the cutting tool holder of the slitting and creasing machine via the device's serial bus at a refresh rate of 1 millisecond. The currently acquired actual absolute coordinate value is accurately determined to be 450 mm. The pre-push-in continuous scheduling batch sequence is analyzed. Based on the production requirements of the width specifications of subsequent orders for the current scheduled batch, the tool holder standby coordinate value for the corresponding continuous scheduling batch sequence is obtained through a lookup table addressing matching method. This standby coordinate value is accurately determined to be 1350 mm. The standby coordinate value of 1350 mm and the current actual absolute coordinate value of 450 mm are substituted into the one-dimensional spatial displacement scalar calculation logic unit. The standby coordinate value is directly subtracted from the current actual absolute coordinate value. After performing a pure algebraic difference operation, the total spatial displacement length is calculated to be 900 mm in the positive direction. The total spatial displacement value clarifies the straight-line physical distance that the cutter holder must traverse within the upcoming order change cycle. This value, closely combined with the time boundary parameters obtained from the aforementioned calculations, constitutes an indispensable underlying geometric prerequisite in the mathematical equation planning of the cutter holder's three-dimensional spatiotemporal motion trajectory, ensuring that the tool can accurately cut into the production preparation position for the next order.

[0033] S403: Based on the total spatial displacement length and the benchmark exchange time parameter, construct a spatiotemporal discrete node matrix, perform polynomial coefficient calculations on the spatiotemporal discrete node matrix, and construct the benchmark interpolation position equation. To obtain the rigid mechanical constraint limits that the tool holder's lateral kinematics must adhere to, it is determined that the tool holder must remain physically stationary at the start and end of the movement. This is achieved by extracting initial zero-velocity boundary conditions specifying a starting speed of 0 and an end-effector zero-velocity boundary condition specifying a stopping speed of 0. The calculated total spatial displacement of 900 mm, the baseline changeover time of 0.7096 seconds, and the aforementioned initial and end-effector zero-velocity boundary conditions are synchronously mapped as input constraint parameters into a continuous, isolated space in computer memory. This constructs a 4x4 dimension spatiotemporal discrete node matrix data array representing the correspondence between the polynomial's undetermined coefficients and the motion state. The underlying linear algebra mathematical calculation function library is invoked to perform high-order matrix inversion and Cramer's rule elimination calculations on the boundary values ​​within this spatiotemporal discrete node matrix, accurately solving for an independent polynomial coefficient solution set containing four elements: constant, linear, quadratic, and cubic terms. Based on the extracted polynomial coefficient solution set, combined with the time independent variable feature term, an algebraic recombination operation is performed at the software algorithm layer. Each coefficient value is multiplied by different order terms of the time independent variable in turn and summed. Finally, a complete benchmark interpolation position equation is constructed, providing the cutter holder with a high-order smooth and continuous displacement guide trajectory at the mathematical level.

[0034] Please see Figure 6 The specific steps for obtaining flexible production line control records are as follows: S501: Extract the hardware overload protection threshold of the ball screw transmission mechanism, extract the reference interpolation position equation, perform third-order time derivative calculation on the reference interpolation position equation at the time domain discrete node to obtain the real-time instantaneous jump value, obtain the maximum allowable jump limit constant of the ball screw transmission mechanism, and integrate the real-time instantaneous jump value and the maximum allowable jump limit constant to obtain the jump comparison parameter set. The read-only database of the factory safety inspection report of the ball screw drive mechanism installed on the drive side of the longitudinal indentation machine is read, and the overload threshold data block limited by the yield strength of the material structure itself is extracted. The reference interpolation position equation generated in the aforementioned calculation steps is called, and discretization slicing is performed in the time domain with a fixed step size of 10 milliseconds to extract each isolated time-domain discrete node. In the calculus calculation logic unit, three consecutive algebraic derivative operations are performed on the reference interpolation position equation. The first derivative is used to obtain the instantaneous velocity value, the second derivative to obtain the instantaneous acceleration value, and then the third time derivative is used to calculate the real-time instantaneous jump value characterizing the drastic change in acceleration. The real-time instantaneous jump value calculated at the 0.3-second slice node is determined to be 150 mm / s. The maximum permissible jump limit constant limited in the above overload threshold data block is extracted. This constant is determined by the transient impact physical withstand limit of the screw-nut pair, and the maximum permissible jump limit constant is fixed at 120 mm / s. Create a comparison structure data object, and write the calculated real-time instantaneous jump value of 150 mm per cubic second and the maximum allowable jump limit constant of 120 mm per cubic second into different parameter fields of the structure, thereby integrating and generating a jump comparison parameter set for subsequent security boundary verification comparison.

[0035] S502: Call the jump comparison parameter set, compare the values ​​based on the real-time instantaneous jump value and the maximum allowable jump limit constant, and when it is determined that there is a real-time instantaneous jump value that is greater than the maximum allowable jump limit constant, obtain the time step extension value, and perform a summation operation on the baseline order change duration parameter and the time step extension value to obtain the updated order change duration parameter. The system analyzes the set of jump comparison parameters in the memory cache pool, extracting the real-time instantaneous jump value and the maximum permissible jump limit constant. It then drives the logic arithmetic execution unit within the central processing chip to perform a strict numerical comparison operation based on the extracted real-time instantaneous jump value of 150 mm / s and the maximum permissible jump limit constant of 120 mm / s. After the comparator circuit determines that the value 150 is greater than the value 120, it immediately triggers the over-limit interrupt flag. The main control program then immediately determines that there is a real-time instantaneous jump value in the current motion trajectory planning that exceeds the maximum permissible jump limit constant, indicating that forcibly maintaining the original speed will damage the mechanical transmission components. The control flow then switches to adaptive protection deceleration control flow, reading the set relaxation factor step size for each iteration from the flexible configuration document area, obtaining the time step extension value, and fixing this extension value at 0.1 seconds. The previously calculated baseline order change duration parameter of 0.7096 seconds is extracted. This baseline order change duration parameter, along with the acquired time step extension value of 0.1 seconds, is then substituted into a floating-point adder for mathematical summation, yielding a final result of 0.8096 seconds. This summation result is then directly overwritten in memory to replace the original time variable address, thus obtaining an effectively widened updated order change duration parameter. This lengthens the action cycle in the time dimension to dilute the instantaneous mechanical impact.

[0036] S503: Recalculate the polynomial coefficients based on the updated order change duration parameters and the total spatial displacement length, reconstruct the interpolated displacement curve to generate curve control data, encapsulate the data format for continuous scheduling batch sequence, drive angle control data and curve control data, and generate flexible production line management records. The newly calculated update and replacement time parameter of 0.8096 seconds is used to replace the original time-domain calculation boundary parameter. Combined with the original value of 900 mm for the total spatial displacement, the underlying algorithm engine for high-order matrix inversion is reactivated. A new set of deceleration polynomial coefficients is recalculated through a second full-scale solution operation. Based on this newly acquired set of coefficients, a flexible interpolation displacement curve mathematical function model is reconstructed. This mathematical model is densely sampled at 1-millisecond clock cycles and converted into a servo absolute position following instruction string, thereby generating the final curve control data that directly drives the cutter holder's movement. Subsequently, the continuous scheduling batch sequence that determines the order processing order, the drive angle control data that determines the thermodynamic bonding state, and the newly generated secure curve control data are retrieved. JSON object serialization conversion technology is used to encapsulate and merge the data format of the above three sets of core control flow data, and an anti-tamper-proof digital encrypted signature and runtime timestamp tag are added. By merging the parameters from these three dimensions and writing them as a whole into the non-volatile database storage array of the manufacturing execution platform, a flexible production line control record with full-link traceability and anti-counterfeiting capabilities is generated, thus closing the control process from top-level scheduling to bottom-level electromechanical adaptive execution.

[0037] Table 2 Flexible Production Line Control Record Packaging Structure Table SequenceData Continuous scheduling batch sequence 1024 AngleControlData Drive angle control data 256 CurveMotionData Curve control data 2048 Table 2 shows the final generated data packet structure containing multi-dimensional control hierarchy management records.

[0038] Please see Figure 7 A production line management system based on flexible manufacturing includes: Order edge weight fusion module: Collects the target delivery time value and the original paper width value of the orders to be scheduled for production and calculates the deviation respectively. Multiplies the obtained delivery time difference value and the width difference value to generate the edge weight constant. Directed Graph Batch Aggregation Module: Constructs a global order directed graph based on edge weight constants, removes directed connections that exceed boundary reference values, tracks path connectivity, and combines extracted closed graph region nodes to generate a continuous scheduling batch sequence; Preheating cylinder angle calculation module: calls the continuous scheduling batch sequence, uses the product of the target heating time value and the instantaneous linear velocity value to obtain the theoretical heating arc length value, calculates the deviation by combining the actual heating arc length value, and generates drive angle control data based on the preheating cylinder circumference value and the circumferential angle constant. Interpolation equation construction module: Based on the driving angle control data, a benchmark order change time parameter is generated, the difference between the standby and actual absolute coordinate values ​​of the corresponding continuous scheduling batch sequence is obtained to obtain the total spatial displacement value, a spatiotemporal discrete node matrix is ​​constructed and the polynomial coefficients are solved to generate the benchmark interpolation position equation. Constraint Curve Smoothing Module: Obtains real-time instantaneous jump values ​​by calculating the third-order time derivative of the benchmark interpolation position equation. When the maximum allowable jump limit constant is exceeded, the changeover duration parameter is updated by extending the time step value. The curve is reconstructed by combining the total spatial displacement value to generate curve control data. The continuous scheduling batch sequence, drive angle control data and curve control data are encapsulated in data format to generate flexible production line management records.

[0039] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A production line control method based on flexible manufacturing, characterized in that, Includes the following steps: S1: Collect the target delivery time value and the paper width value of the orders to be scheduled for production and calculate the deviation respectively. Multiply the obtained delivery time difference value and the width difference value to generate the edge weight constant. S2: Construct a global order directed graph based on the edge weight constant, remove directed lines that exceed the boundary reference value, trace path connectivity, and combine the extracted closed graph region nodes to generate a continuous scheduling batch sequence. S3: Call the continuous scheduling batch sequence, use the product of the target heating time value and the instantaneous linear velocity value to obtain the theoretical heating arc length value, combine the actual heating arc length value to calculate the deviation, and generate drive angle control data based on the preheating cylinder circumference value and the circumferential angle constant. S4: Based on the driving angle control data, generate a benchmark order change time parameter, calculate the difference between the standby and actual absolute coordinate values ​​of the corresponding continuous scheduling batch sequence to obtain the total spatial displacement value, construct a spatiotemporal discrete node matrix and solve the polynomial coefficients to generate a benchmark interpolation position equation. S5: Obtain the real-time instantaneous jump value by calculating the third-order time derivative of the benchmark interpolation position equation. When the maximum allowable jump limit constant is exceeded, update the order change duration parameter by extending the time step value. Combine the total spatial displacement value to reconstruct the curve and generate curve control data. Encapsulate the continuous scheduling batch sequence, the drive angle control data and the curve control data into a data format to generate a flexible production line management record.

2. The production line control method based on flexible manufacturing according to claim 1, characterized in that, The edge weight constants specifically include the delivery urgency coefficient and the width penalty factor. The continuous scheduling batch sequence includes the order production sorting table, the same-order exchange details, and the material grouping queue. The drive angle control data specifically refers to the compensation pulse command, the target rotation phase, and the servo wrap angle parameters. The reference interpolation position equation specifically includes the displacement fitting coefficient, the motion trajectory polynomial, and the node coordinate matrix. The flexible production line management and control records include the jump limit log, the duration update report, and the overload alarm code.

3. The production line control method based on flexible manufacturing according to claim 1, characterized in that, The specific steps for obtaining the edge weight constants are as follows: S101: Receive order scheduling records from the scheduling server, parse the order scheduling records to obtain the orders to be scheduled, collect the target delivery time value and the original paper width value corresponding to the orders to be scheduled, and integrate the target delivery time value and the original paper width value to obtain the original parameter set of the order. S102: Call the original set of order parameters, perform time deviation calculation based on the target delivery time value to obtain the delivery time difference value, perform size deviation calculation based on the original paper width value to obtain the width difference value, and integrate the delivery time difference value and the width difference value to obtain the order deviation parameter set; S103: Call the order deviation parameter set, perform a product operation on the delivery time difference value and the width difference value, and generate the edge weight constant.

4. The production line control method based on flexible manufacturing according to claim 1, characterized in that, The specific steps for obtaining the continuous scheduling batch sequence are as follows: S201: Obtain the order network aggregation instruction, parse the order network aggregation instruction, call the orders to be scheduled to construct topology nodes, construct directed connections for the topology nodes, assign values ​​to the directed connections according to the edge weight constants, and construct a global order directed graph; S202: Obtain the aggregate truncation reference boundary value, remove the directed lines that exceed the boundary in the global order directed graph based on the aggregate truncation reference boundary value, and for the removed global order directed graph, trace the path connectivity of adjacent nodes along the remaining directed lines to extract the graph regions where the internal nodes are connected to each other as closed graph regions. S203: Call the closed graph region, perform combination operations on the topological nodes within the closed graph region, and generate a continuous scheduling batch sequence.

5. The production line control method based on flexible manufacturing according to claim 1, characterized in that, The process of obtaining the order network aggregation instruction is as follows: Monitor the cache queue of the data communication port, extract the current production stack quantity value of the cache queue, read the maximum throughput limit of the server hardware to generate the aggregation trigger threshold constant, perform a numerical comparison operation between the production stack quantity value and the aggregation trigger threshold constant, and when it is determined that the production stack quantity value is greater than the aggregation trigger threshold constant, use the underlying communication protocol to perform message encapsulation operation to obtain the order network aggregation instruction; The process of performing combination operations on topological nodes within a closed graph region to generate a continuous scheduling batch sequence is as follows: The association feature mapping data of the topological nodes in the closed graph region is extracted. Based on the association feature mapping data, the order ranking weight values ​​of the topological nodes in the closed graph region are read. According to the order ranking weight values, the topological nodes in the closed graph region are sorted by unidirectional linearization to establish a one-dimensional node queue. The arrangement index position of each node in the one-dimensional node queue is extracted and converted to generate node time sequence mapping parameters. According to the node time sequence mapping parameters, a structured splicing operation is performed on the one-dimensional node queue to generate a continuous scheduling batch sequence.

6. The production line control method based on flexible manufacturing according to claim 1, characterized in that, The specific steps for acquiring the drive angle control data are as follows: S301: Obtain the heating control configuration of the preheating cylinder and the servo motor of the guide roller of the single-facer, parse the heating control configuration, call the continuous scheduling batch sequence, obtain the preheating cylinder circumference value and the target heating time value, collect the instantaneous linear velocity value of the encoder of the single-facer main drive shaft, perform a product operation on the target heating time value and the instantaneous linear velocity value to obtain the theoretical heating arc length value, and integrate the preheating cylinder circumference value and the theoretical heating arc length value to obtain the heating arc length parameter set; S302: Call the set of heated arc length parameters, collect the actual heated arc length value of the paper guide roller servo motor, perform deviation calculation on the theoretical heated arc length value and the actual heated arc length value to obtain the arc length deviation value, and perform ratio calculation on the arc length deviation value and the preheating cylinder circumference value to obtain the number of revolutions ratio value. S303: Call the stated number of revolutions ratio value to obtain the circumferential angle constant, perform a product operation on the number of revolutions ratio value and the circumferential angle constant to obtain the rotation angle compensation value, and generate drive angle control data based on the rotation angle compensation value.

7. The production line control method based on flexible manufacturing according to claim 1, characterized in that, The specific steps for obtaining the reference interpolation position equation are as follows: S401: Receive the lateral displacement planning command of the slitting and creasing machine cutter holder, evaluate the response time of the guide roller based on the drive angle control data, and use the response time of the guide roller as the benchmark order change time parameter; S402: Collect the current actual absolute coordinate value of the cutting blade holder of the longitudinal cutting and creasing machine, obtain the standby coordinate value corresponding to the continuous scheduling batch sequence, perform difference calculation on the standby coordinate value and the current actual absolute coordinate value, and obtain the total spatial displacement value. S403: Based on the total spatial displacement length and the reference exchange time parameter, construct a spatiotemporal discrete node matrix, perform polynomial coefficient calculations on the spatiotemporal discrete node matrix, and construct a reference interpolation position equation.

8. The production line control method based on flexible manufacturing according to claim 1, characterized in that, The process for evaluating the response time of the paper guide roller is as follows: Extract the rated angular velocity operating parameters of the guide roller servo motor, call the drive angle control data, perform division calculation on the drive angle control data and the rated angular velocity operating parameters to obtain the basic rotation time value, read the equipment's underlying control board to obtain the mechanical transmission delay compensation constant, sum the basic rotation time value and the mechanical transmission delay compensation constant to obtain the guide roller response time; The process of constructing the spatiotemporal discrete node matrix, performing polynomial coefficient operations on the spatiotemporal discrete node matrix, and constructing the benchmark interpolation position equation is as follows: The initial zero-velocity boundary conditions and the final zero-velocity boundary conditions of the lateral motion are obtained. The total spatial displacement length, the reference change time parameter, the initial zero-velocity boundary conditions and the final zero-velocity boundary conditions are synchronously mapped to a two-dimensional data array to construct a spatiotemporal discrete node matrix. Matrix inversion is performed on the spatiotemporal discrete node matrix to obtain the polynomial coefficient solution set. Based on the polynomial coefficient solution set and the time independent variable characteristic term, an algebraic recombination operation is performed to construct the reference interpolation position equation.

9. The production line control method based on flexible manufacturing according to claim 1, characterized in that, The specific steps for obtaining the flexible production line control records are as follows: S501: Extract the hardware overload protection threshold of the ball screw transmission mechanism, extract the reference interpolation position equation, perform third-order time derivative calculation on the reference interpolation position equation at the time domain discrete node to obtain the real-time instantaneous jump value, obtain the maximum allowable jump limit constant of the ball screw transmission mechanism, and integrate the real-time instantaneous jump value and the maximum allowable jump limit constant to obtain the jump comparison parameter set. S502: Call the set of jump comparison parameters, perform numerical comparison based on the real-time instantaneous jump value and the maximum allowable jump limit constant, when it is determined that there is a real-time instantaneous jump value greater than the maximum allowable jump limit constant, obtain the time step extension value, and perform a summation operation on the benchmark order change duration parameter and the time step extension value to obtain the updated order change duration parameter; S503: Recalculate the polynomial coefficients based on the update changeover duration parameter and the total spatial displacement value, reconstruct the interpolation displacement curve to generate curve control data, encapsulate the data format of the continuous scheduling batch sequence, the drive angle control data and the curve control data, and generate flexible production line management records.

10. A production line control system based on flexible manufacturing, characterized in that, The system is used to implement the production line control method based on flexible production as described in any one of claims 1-9, and the system includes: Order edge weight fusion module: Collects the target delivery time value and the original paper width value of the orders to be scheduled for production and calculates the deviation respectively. Multiplies the obtained delivery time difference value and the width difference value to generate the edge weight constant. Directed graph batch aggregation module: Constructs a global order directed graph based on the edge weight constant, removes directed lines that exceed the boundary reference value, tracks path connectivity, and combines the extracted closed graph region nodes to generate a continuous scheduling batch sequence; Preheating cylinder angle calculation module: calls the continuous scheduling batch sequence, uses the product of the target heating time value and the instantaneous linear velocity value to obtain the theoretical heating arc length value, calculates the deviation in combination with the actual heating arc length value, and generates drive angle control data based on the preheating cylinder circumference value and the circumferential angle constant. Interpolation equation construction module: Based on the driving angle control data, a benchmark changeover time parameter is generated, the difference between the standby and actual absolute coordinate values ​​of the corresponding continuous scheduling batch sequence is obtained to get the total spatial displacement value, a spatiotemporal discrete node matrix is ​​constructed and the polynomial coefficients are solved to generate the benchmark interpolation position equation. Constraint curve smoothing module: Obtain the real-time instantaneous jump value by calculating the third-order time derivative of the benchmark interpolation position equation. When the maximum allowable jump limit constant is exceeded, the changeover duration parameter is updated by extending the time step value. The curve is reconstructed by combining the total spatial displacement value to generate curve control data. The continuous scheduling batch sequence, the drive angle control data and the curve control data are encapsulated in data format to generate flexible production line management records.

Images