Injection molding scheduling method and system based on multi-dimensional fitting

Abstract

The application relates to the technical field of injection molding scheduling, and particularly discloses an injection molding scheduling method and system based on multidimensional fitting, which comprises the following steps: generating a thermal recovery characteristic curve according to a thermal history operation record set of an injection molding device; determining a preliminary order attribution relationship and a planned total mold number according to a scheduling basic data set; generating a time elasticity coefficient interval sequence of each sequential position according to the planned total mold number and the thermal recovery characteristic curve; performing thermal state adaptability distribution on orders in the preliminary order attribution relationship according to the time elasticity coefficient interval sequence to obtain an initial scheduling plan data set, and determining a real-time thermal state offset according to a real-time cycle time of an executed mold number during the execution of the scheduling; and triggering dynamic reconstruction of an unexecuted task queue when the absolute value of the real-time thermal state offset of any device exceeds a preset threshold value. The application can improve the self-adaptive capacity of a scheduling plan to a thermal state.

Description

Technical Field

[0001] This invention relates to the field of injection molding scheduling technology, and in particular to an injection molding scheduling method and system based on multi-dimensional fitting. Background Technology

[0002] In the field of injection molding scheduling, existing scheduling methods generally treat the capacity of injection molding equipment as a constant fixed parameter, assuming that the same equipment has the same single-piece production cycle time in all production periods. Based on this assumption, traditional methods first generate a fixed task execution sequence for each equipment in the initial stage of scheduling, and then assign a uniform time parameter to each task in the sequence. This approach abstracts the equipment capacity into a constant that is independent of the specific position of the task in the sequence, making it easy to ignore the objective impact of the order position of the task on the actual production cycle when the scheduling model is first built.

[0003] When injection molding equipment restarts after a shutdown, its actual single-piece production cycle gradually shortens with the increase of the number of runs until it tends to stabilize. The actual production capacity faced by tasks in different sequences on the same equipment has systematic differences. However, the fixed task sequence generated by the existing scheduling method cannot effectively reflect this difference. When the actual production progress deviates from the expected progress due to fluctuations in the thermal state of the equipment, the traditional method passively compensates for the deviation by adjusting the time parameters of each subsequent task. This compensation method is difficult to change the sequence relationship between tasks, which leads to the irrational arrangement in the initial scheduling stage being solidified and the deviation being passed down step by step along the fixed sequence, causing a continuous deviation between the scheduling plan and the actual execution result. Therefore, how to improve the adaptability of the scheduling plan to thermal state has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides an injection molding production scheduling method and system based on multi-dimensional fitting to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides an injection molding scheduling method based on multi-dimensional fitting, comprising: C1. Generate thermal recovery characteristic curves for the injection molding equipment based on the thermal history operation record set of the injection molding equipment; C2. Based on the production scheduling data set of the injection molding equipment, determine the preliminary order attribution relationship and planned total number of molds for the injection molding equipment in the current production scheduling period. Based on the planned total number of molds and the heat recovery characteristic curve, generate the time elasticity coefficient interval sequence of the injection molding equipment at each sequential position in the current production scheduling period. C3. Based on the time elasticity coefficient interval sequence, perform thermal state adaptability allocation on the orders in the preliminary order attribution relationship to obtain the initial production schedule dataset. During the production scheduling process based on the initial production schedule dataset, determine the real-time thermal state offset of the injection molding equipment in the current production scheduling period based on the real-time cycle time of the molds already executed on the injection molding equipment. C4. When the absolute value of the real-time thermal state offset of any device in the injection molding equipment exceeds the preset offset threshold, the unexecuted task queue of the injection molding equipment is dynamically reconstructed.

[0006] To address the above problems, the present invention also provides an injection molding scheduling system based on multi-dimensional fitting, the system comprising: The heat recovery curve generation module is used to generate heat recovery characteristic curves for the injection molding equipment based on the thermal history operation record set of the injection molding equipment. The elastic interval sequence generation module is used to determine the preliminary order attribution relationship and planned total number of molds for the injection molding equipment in the current production scheduling period based on the production scheduling basic dataset of the injection molding equipment, and generate the time elasticity coefficient interval sequence of the injection molding equipment at each sequential position in the current production scheduling period based on the planned total number of molds and the heat recovery characteristic curve. The adaptation and allocation monitoring module is used to perform thermal state adaptation allocation on the orders in the preliminary order affiliation relationship according to the time elasticity coefficient interval sequence, to obtain the initial production scheduling plan dataset, and to determine the real-time thermal state offset of the injection molding equipment in the current production scheduling period according to the real-time cycle time of the molds already executed on the injection molding equipment during the production scheduling process based on the initial production scheduling plan dataset. The dynamic queue reconstruction module is used to trigger the dynamic reconstruction of the unexecuted task queue of the injection molding equipment when the absolute value of the real-time thermal state offset of any device in the injection molding equipment exceeds a preset offset threshold.

[0007] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention generates a thermal recovery characteristic curve for injection molding equipment, quantifies the actual periodic change pattern of each mold position after cold start, and reconstructs the equipment capacity as a function of sequential position. Based on this curve, a time elasticity coefficient interval sequence for each sequential position is generated, allowing the production scheduling model to consider the impact of task position on cycle time during the matching phase, providing flexible constraints corresponding to the equipment's thermal state for order allocation. In the initial matching, the process time requirements of orders in the preliminary order attribution relationship are matched with the elasticity interval of each position to minimize the deviation, so that the determination of the production order reflects the capacity difference of the equipment at different positions, avoiding the separation of sorting and timing. During execution, the thermal state offset is accumulated for each mold, and the offset reflects the degree of deviation between the actual thermal recovery progress of the equipment and the curve prediction. When the offset exceeds the threshold, the remaining elasticity interval is corrected as a whole by using the opposite number of the offset as the translation amount, so that the remaining orders are re-matched to positions, and the deviation response is changed from adjusting the time value to adjusting the task execution order.

[0008] 2. This invention decouples task sequences from fixed ones, allowing for dynamic adjustments to the execution order based on actual production feedback. This enables the location distribution of subsequent tasks to be re-optimized based on thermal state deviations, reducing cumulative time offsets caused by thermal state fluctuations and ensuring that production scheduling keeps pace with the equipment's thermal recovery status. The thermal recovery characteristic curve is constructed based on the equipment's own history, eliminating the need for general experience values ​​to form a personalized production scheduling benchmark. This allows production scheduling to adapt to equipment of different ages and maintenance conditions, improving the consistency between production scheduling plans and actual production. Attached Figure Description

[0009] Figure 1 This is a flowchart illustrating an injection molding production scheduling method based on multi-dimensional fitting, provided in an embodiment of the present invention. Figure 2 This is a functional block diagram of an injection molding production scheduling system based on multi-dimensional fitting, provided in an embodiment of the present invention.

[0010] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0011] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0012] This application provides an injection molding scheduling method based on multi-dimensional fitting. The execution entity of this multi-dimensional fitting injection molding scheduling method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the multi-dimensional fitting injection molding scheduling method can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0013] Reference Figure 1 The diagram shown is a flowchart illustrating an injection molding scheduling method based on multi-dimensional fitting according to an embodiment of the present invention. In this embodiment, the injection molding scheduling method based on multi-dimensional fitting includes: C1. Generate thermal recovery characteristic curves for the injection molding equipment based on the thermal history operation record set of the injection molding equipment; In this embodiment of the invention, the thermal history operation record set is used to store the complete time sequence data of each injection molding equipment in each continuous operation stage during the actual production process. The record set is indexed by the equipment identifier and stores the cold start time of each equipment, the continuous operation termination time after each cold start, and the actual cycle time of each mold recorded in the order of mold number starting from the start time.

[0014] For a single cold start, the record set corresponds to a continuous running cycle sequence. Each data point in the sequence contains the sequence number of the module in this start and the actual number of seconds consumed from the start of mold closing to the completion of mold opening.

[0015] Extract all continuous operating cycle sequences after each cold start of a certain injection molding machine from the record set. Each start sequence is stored independently and there is no cross-mixing between the starts. This results in the start-up operating cycle set of the machine for each cold start. This start-up operating cycle set contains all cycle time data of each cold start of the machine in history, from the first mold to the end of the start.

[0016] For each startup in the startup cycle set, the startup cycle sequence is checked one by one from the last module backwards. When the fluctuation of the actual cycle time of a certain range of modules relative to the average cycle time of that range does not exceed plus or minus 5%, the modules in that range are considered to be in a thermal equilibrium stable state. Then, the actual cycle times of all modules in that range are averaged to obtain the thermal equilibrium average cycle time of that startup.

[0017] The method for determining whether a certain module is in a thermal equilibrium stable state is as follows: starting from the last module in the current startup cycle sequence, check each module backwards. Each module is included in the candidate stable segment and the average cycle time of the candidate stable segment is recalculated. When the fluctuation range of the average cycle time of the candidate stable segment does not exceed ±5% for the continuously included modules, the segment is expanded forward until the first time a module causes the fluctuation range of the average cycle time of the candidate stable segment to exceed ±5%, at which point the expansion stops. All modules in the candidate stable segment determined before the expansion stops are considered to be in a thermal equilibrium stable state.

[0018] Divide the actual cycle time of any specified module position in this startup by the above-mentioned average cycle time of thermal equilibrium. The resulting quotient is the single startup cycle ratio of that module position in this startup. This ratio reflects the relative multiple relationship between the actual cycle time of that module position and the steady-state cycle time of the equipment after thermal equilibrium.

[0019] The ratio of single start-up cycles for the same injection molding machine at the same mold position in all historical cold starts is collected to obtain the set of ratios corresponding to that mold position. The value in the middle position after being sorted by numerical value from this set of ratios is selected as the heat recovery characteristic value of the machine at that mold position.

[0020] For example, if there is historical data of five starts at a certain module position, the third value of the five single start cycle ratios is taken as the thermal recovery characteristic value of that position after arranging them in ascending order; for the case of an even number of starts, the arithmetic mean of the two middle values ​​is taken as the thermal recovery characteristic value of that position.

[0021] The median, rather than the average, was used for aggregation because a single startup may be affected by unforeseen factors such as temporary mold cleaning or material batch fluctuations, resulting in extreme ratios deviating from the normal range. The median is not sensitive to these extreme values ​​and can more accurately reflect the typical heat recovery state of the equipment at that mold position. The above operation was performed on all mold positions one by one to obtain the correspondence between each mold position and each heat recovery characteristic value.

[0022] Following the order of increasing mold positions starting from 1, each mold position and its corresponding heat recovery characteristic value are arranged into a two-dimensional sequence. The horizontal axis of this sequence represents the mold position, and the vertical axis represents the heat recovery characteristic value. This two-dimensional sequence is used as the heat recovery characteristic curve of the injection molding equipment.

[0023] The maximum value on the horizontal axis of the thermal recovery characteristic curve represents the maximum number of consecutive operating cycles achieved during a single cold start in the equipment's historical records.

[0024] When the planned total number of molding operations for a certain injection molding equipment exceeds the maximum value of the horizontal axis of the equipment's thermal recovery characteristic curve, the predicted cycle ratio of the sequence position corresponding to the maximum value of the horizontal axis is all set to one. The basis for this value is that after the number of molding operations exceeds the maximum number of molding operations required to reach a stable thermal equilibrium state in the historical records, the equipment is already in a state of complete thermal equilibrium, and the cycle time of each subsequent molding operation is consistent with the steady-state cycle time.

[0025] The above method generates a unique heat recovery characteristic curve for each injection molding machine, and curve data is not shared between different machines.

[0026] As another optional implementation of this embodiment, for newly commissioned injection molding equipment, when there is no data on previous cold starts in its thermal history operation record set, the thermal recovery characteristic curve cannot be generated through historical data statistics. In this scenario, the type test data recorded under standard operating conditions before the equipment leaves the factory is used as the initial thermal history operation record set. This type test data contains a complete cycle time series of the equipment from cold start to thermal equilibrium.

[0027] The type test data is used to extract the start-up operation cycle set, calculate the ratio of single start-up cycles, and take the median to generate the initial thermal recovery characteristic curve of the equipment. After the equipment has accumulated a sufficient number of cold start records in actual production, the subsequently generated actual operation data is gradually replaced or weighted and integrated into the type test data. This allows the thermal recovery characteristic curve to gradually approach the true thermal recovery characteristics of the equipment as actual operating experience accumulates, thus avoiding the inability to use this method for new equipment due to a lack of historical data.

[0028] As another optional implementation of this embodiment, for each startup data in the historical cold startup record set, before calculating the ratio of single startup cycles at each module position, the consistency of the complete cycle time series of each startup is first checked.

[0029] The specific verification method is as follows: plot the complete cycle time series of each startup as a curve according to the module position, calculate the dynamic time warping distance between the curve and the other startup curves of the same device, and when the warping distance of a startup curve exceeds three times the average warping distance of all startup curves, the complete cycle time series of that startup is judged as an abnormal startup and excluded from subsequent statistical calculations.

[0030] By excluding abnormal startup data caused by occasional factors such as improper mold replacement operations, abnormal material batches, and sudden equipment failures, the remaining normal startup data can more accurately represent the thermal recovery law of the equipment under typical operating conditions, thereby improving the reliability and stability of the thermal recovery characteristic curve.

[0031] C2. Based on the production scheduling data set of the injection molding equipment, determine the preliminary order attribution relationship and planned total number of molds for the injection molding equipment in the current production scheduling period. Based on the planned total number of molds and the heat recovery characteristic curve, generate the time elasticity coefficient interval sequence of the injection molding equipment at each sequential position in the current production scheduling period. In this embodiment of the invention, the production scheduling basic dataset includes order information for all orders to be scheduled during the current production scheduling period, available capacity information for each injection molding equipment, and mold compatibility information for each injection molding equipment. The order information includes the order identifier, the product model of the required product, the product quantity, and the delivery time. The available capacity information includes the maximum number of molds that each equipment can use during the current production scheduling period. The mold compatibility information includes the mold models that can be installed and used on each equipment and the range of product models that the mold can produce.

[0032] The product models required by each order to be scheduled are read from the production scheduling base dataset. The product models are compared with the mold compatibility information of each injection molding equipment item by item. All injection molding equipment that can install the corresponding mold model to produce the product are selected. These equipment constitute the available injection molding equipment set for the order.

[0033] For each order to be scheduled for production, the available capacity information of each equipment is read from the set of available injection molding equipment. When the number of molds that can be produced by a certain equipment is greater than or equal to the number of products required by the order, the entire order is allocated to that equipment. If multiple equipment meet the capacity requirements, the equipment with the smallest remaining difference between the currently allocated number of molds and the upper limit of available capacity is selected for allocation.

[0034] After processing each pending production order, all the orders allocated to each injection molding machine constitute the initial order set for that machine.

[0035] The preliminary order set is associated with the device identification information of the device in the form of key-value pairs and stored to generate the preliminary order attribution relationship.

[0036] For each injection molding machine, the product quantities required by each order in its initial order set are summed up. The summation result is the planned total number of molds for that machine in the current production period. The unit of the planned total number of molds is the number of molds, which is consistent with the unit of counting the number of molds in injection molding production.

[0037] The total number of sequential positions for the equipment within the current production schedule is determined based on the planned total number of mold runs. Sequential positions are numbered sequentially from 1 up to the value corresponding to the planned total number of mold runs. Each sequential position represents one mold run that the equipment will produce at that position. For each sequential position, the vertical axis value corresponding to the horizontal axis position with the same number is read from the thermal recovery characteristic curve of the equipment. This value is the prediction cycle ratio for that sequential position. This prediction cycle ratio is a positive real number greater than 1. The reciprocal of this ratio, resulting in a positive real number less than 1, is used as the baseline coefficient for that sequential position.

[0038] For each sequential position, a lower tolerance coefficient and an upper tolerance coefficient are set. The lower tolerance coefficient is 0.95 and the upper tolerance coefficient is 1.05. These two values ​​are determined based on the fact that the normal fluctuation range of the single-mold cycle time of the injection molding equipment under thermal equilibrium and stable conditions is ±5%. This fluctuation range is determined by the superposition effect of the pressure fluctuation of the hydraulic system of the injection molding machine control system and the temperature fluctuation of the mold temperature control system. The boundary value of the tolerance coefficient is set to be consistent with the normal fluctuation range, so that the time elasticity coefficient range formed by multiplying the base coefficient by the tolerance coefficient can cover the upper and lower limits of the cycle time fluctuation of the equipment under normal operating conditions. The difference between the two tolerance coefficients is 0.10, and this difference constitutes the floating bandwidth of that position.

[0039] The product of the baseline coefficient and the lower tolerance coefficient of 0.95 is used as the lower limit coefficient for that sequence position, and the product of the baseline coefficient and the upper tolerance coefficient of 1.05 is used as the upper limit coefficient for that sequence position. These two coefficients form a numerical interval, which is the time elasticity coefficient interval for that sequence position. After performing the above operation on all sequence positions, the time elasticity coefficient intervals for each position are arranged into a time elasticity coefficient interval sequence according to the increasing order of sequence position from 1 to the total planned modulus.

[0040] Each position in the sequence corresponds to an independent elastic range. The larger the position number, the closer the benchmark coefficient is to 1, and the closer the center value of the corresponding elastic range is to 1. The smaller the position number, the further the benchmark coefficient is from 1, and the smaller the corresponding elastic range is overall.

[0041] As another optional implementation of this embodiment, for each injection molding equipment whose preliminary order attribution relationship has been determined, before generating the time elasticity coefficient interval sequence based on the planned total number of molds, the planned total number of molds for each injection molding equipment is first checked for balance.

[0042] The specific verification method is as follows: calculate the average and standard deviation of the planned total number of molds for all injection molding equipment. When the planned total number of molds for a certain equipment deviates from the average by more than twice the standard deviation, the orders on that equipment that exceed the average are redistributed to other available injection molding equipment with a planned total number of molds lower than the average, until the deviation of the planned total number of molds for all equipment is controlled within two times the standard deviation.

[0043] By performing balance checks, the total number of planned modulo operations for each device is distributed relatively evenly, preventing a single device from bearing an excessive production load and causing the scheduling period to span the long operating range after the device's thermal balance is established. This makes the time span distribution of the production schedule on each device more balanced.

[0044] As another optional implementation of this embodiment, when generating the time flexibility coefficient interval sequence for each sequential position, for product models corresponding to orders with particularly urgent delivery times, the floating bandwidth of the interval is further reduced when constructing the flexibility coefficient interval for the sequential position to which the order may be assigned. Specifically, the reduction method is as follows: the difference between the lower limit tolerance coefficient and the upper limit tolerance coefficient under normal operating conditions is reduced to half of the original difference, that is, the lower limit tolerance coefficient is raised to the midpoint between the original lower limit tolerance coefficient and the upper limit tolerance coefficient, and the upper limit tolerance coefficient is lowered to the same midpoint, so that the flexibility coefficient interval for that position is reduced to a narrower range.

[0045] By narrowing the bandwidth of the elastic range, the matching of the process time requirement coefficient of urgent orders at that position is more accurate, reducing the risk that urgent orders will be assigned to inappropriate positions due to excessive time elasticity at that position, and ensuring the certainty of the production scheduling time for urgent orders.

[0046] C3. Based on the time elasticity coefficient interval sequence, perform thermal state adaptability allocation on the orders in the preliminary order attribution relationship to obtain the initial production schedule dataset. During the production scheduling process based on the initial production schedule dataset, determine the real-time thermal state offset of the injection molding equipment in the current production scheduling period based on the real-time cycle time of the molds already executed on the injection molding equipment. In this embodiment of the invention, the preliminary order set corresponding to each injection molding equipment is extracted from the preliminary order attribution relationship according to the equipment identifier of the injection molding equipment. The process time requirement coefficient of each order is determined according to the delivery time information of each order in the preliminary order set of the current equipment. Specifically, all orders in the preliminary order set of the current equipment are sorted from earliest to latest according to their delivery time. The process time requirement coefficient of the order with the earliest delivery time is taken as the lower limit of the elastic coefficient interval of the first sequential position in the time elastic coefficient interval sequence of the equipment. The process time requirement coefficient of the order with the latest delivery time is taken as the upper limit of the elastic coefficient interval of the last sequential position in the time elastic coefficient interval sequence of the equipment. The process time requirement coefficients of the remaining orders are linearly corresponding to the lower limit and upper limit of the interval according to their relative position between the earliest and latest delivery times.

[0047] The process time requirement coefficient is a positive real number greater than 0. Its dimension is a dimensionless multiplier coefficient, which represents the expected cycle multiplier of the product required by the order based on the steady-state cycle time.

[0048] All orders in the initial order set are arranged from earliest to latest according to their delivery time to obtain an initial order queue that reflects the initial delivery date priority. Then, all orders in the initial order queue and their order are completely copied to generate a matching queue. Each order in the matching queue carries the process time requirement coefficient for that order, and the order of the orders in the matching queue is completely consistent with that of the initial order queue.

[0049] The matching process begins with the first order in the initial sequence queue and the first sequential position in the time elasticity coefficient interval sequence. The order currently being processed in the initial sequence queue is defined as the current order, and the sequential position currently being processed in the time elasticity coefficient interval sequence is defined as the current position. The process time requirement coefficient of the current order is then extracted and compared with the elasticity coefficient interval of the current position.

[0050] If the process time requirement coefficient of the current order is greater than or equal to the lower limit of the elasticity coefficient range of the current position and less than or equal to its upper limit, then the order falls within the range of the current position. At this time, the current order is assigned to the current position and removed from the matching queue to indicate that the order has been assigned. Then, the next order in the initial sequence queue is taken as the new current order, and the next sequence position in the time elasticity coefficient range sequence is taken as the new current position, and the matching operation continues.

[0051] If the process time demand coefficient of the current order is less than the lower limit of the elasticity coefficient range of the current position, then find the order with the largest process time demand coefficient from the remaining unassigned orders in the matching queue, and swap the position of the current order with that order in the initial order queue. After the swap, the current order that was originally in that position is moved to the back of the matching queue, while the largest demand order is moved to the current position. Then, the largest demand order that is now in the current position is taken as the new current order and compared again with the elasticity coefficient range of the current position.

[0052] If the process time demand coefficient of the current order is greater than the upper limit of the elasticity coefficient range of the current position, then find the order with the smallest process time demand coefficient from the remaining unassigned orders in the matching queue, and swap the position of the current order with that order in the initial order queue. After the swap, the current order that was originally in that position is moved to the back of the matching queue, while the order with the smallest demand coefficient is moved to the current position. Then, the order with the smallest demand coefficient that is now in the current position is taken as the new current order and compared with the elasticity coefficient range of the current position again.

[0053] After each swap operation, the order order order order sequence in the queue to be matched is no longer consistent with the initial sequence queue. At this point, the complete order of all orders in the initial sequence queue after the swap is copied to the queue to be matched, ensuring that the order of the queue to be matched is completely synchronized with the initial sequence queue. Both queues always contain the same set of orders with the exact same order. The only difference is that orders already assigned in the queue to be matched are marked as assigned, and these assigned orders are automatically skipped when searching for unassigned orders later. The process of re-comparing after a swap is a cyclical process; a location may undergo multiple swaps before obtaining an order whose process time requirement coefficient falls within the elastic range of that location.

[0054] Process each order sequentially starting from the first sequential position as described above. Once a position is matched, the order at that position is fixed and removed from the matching queue. Continue processing the next position until all orders have been assigned to their corresponding sequential positions.

[0055] After all orders are allocated, for each assigned order at a specific sequence position, the process time requirement coefficient for that order is compared with the lower limit of the elasticity coefficient range for that position. The larger of the two values ​​is taken as the planning cycle coefficient for that order at that sequence position. This planning cycle coefficient is a dimensionless positive real number, representing the ratio of the planned cycle to the steady-state cycle when the equipment produces that order at that position.

[0056] Multiply the planned cycle coefficient of each assigned order in the sequence position by the steady-state cycle time of the product required by that order. The steady-state cycle time is in seconds, and the planned cycle coefficient is dimensionless. The corrected planned single-piece cycle time obtained by multiplying the two is also in seconds. This time is the planned single-piece production time consumed by the order in that sequence position.

[0057] The production order of each order within each injection molding machine is determined based on the order of each order in its sequential position and the corrected planned cycle time of each order. That is, the order that is listed first is produced first, and the order that is listed later is produced later. The order of each order on the same machine is determined by the size of the sequential position number.

[0058] Starting from the start time of the injection molding equipment's production schedule within the current production period, the order is progressed sequentially from smallest to largest position. The estimated start time of production for the first order in the sequence is the start time of that production schedule, and the estimated end time of production is the start time of that production schedule plus the adjusted planned cycle time per unit for that order in that sequence position.

[0059] For the second and subsequent sequence positions, the estimated production start time of the order at the current sequence position is the estimated production end time of the order at the previous sequence position, and the estimated production end time of the order at the current sequence position is the estimated production start time of the order plus the adjusted planned single-piece cycle time of the order at that sequence position.

[0060] This process is repeated position by position until the estimated start and end times for all orders at each sequential position are calculated. The production sequence, estimated start and end times, and planning cycle coefficient for each order are then stored in association. The set of this information for all orders constitutes the initial production scheduling dataset for the equipment.

[0061] After each exchange operation during the matching process, the estimated production end time after the exchange is calculated for the two orders involved in the exchange. For the two orders after the exchange, the planned single-piece cycle time is accumulated and corrected position by position from the start of production scheduling to that position to obtain the estimated production end time for each order after the exchange, and the unit of time is seconds.

[0062] The cumulative number of seconds corresponding to the expected production end time after the exchange, from the start time of the production schedule, is compared with the cumulative number of seconds corresponding to the delivery time, from the same start time of the production schedule. The same start time of the production schedule is the moment when the injection molding equipment starts producing the first order in the current production period. Both time quantities are converted with the start time of the production schedule as the zero point of timing. After conversion, the unit of both is seconds. They belong to the same time dimension and have the same zero point of timing, so they can be directly compared.

[0063] If the estimated production completion time of both orders after the exchange is no later than their respective delivery times, the exchange is valid and the matching operation continues. If the estimated production completion time of either order after the exchange is later than its delivery time, the exchange is invalid, the exchange operation is immediately cancelled, both orders are restored to their pre-exchange positions, and the current order before the exchange is forcibly and fixedly assigned to the current sequence position. No further exchange attempts will be made at this position, and this sequence position is marked as a non-exchangeable position, which will be skipped in subsequent matching processes.

[0064] During the actual production process based on the initial production schedule dataset, for each mold cycle that has been completed on each injection molding machine, the real-time cycle time of that mold cycle is obtained by reading the actual number of seconds consumed from the start of mold closing to the completion of mold opening through the injection molding machine's timer.

[0065] Divide the real-time cycle time by the steady-state cycle time of the device, where the steady-state cycle time is in seconds. The resulting quotient is a dimensionless real-time cycle ratio. Read the vertical axis value corresponding to the horizontal axis position with the same sequence position number as the executed module from the device's thermal recovery characteristic curve. This vertical axis value is the predicted cycle ratio for that sequence position. Subtract the predicted cycle ratio from the real-time cycle ratio; the difference is the single-module thermal state deviation value for the executed module. This difference can be positive, zero, or negative.

[0066] At the start of the current production schedule, the initial value of the real-time thermal state offset is set to zero. Starting from this start time, after each production cycle is completed by the equipment, the single-cycle thermal state deviation value of that cycle is added to the accumulated value of the current real-time thermal state offset. The result of the addition is used as the updated real-time thermal state offset. This process is repeated for each cycle until the current time. The unit of this offset is dimensionless, with the same ratio to the cycle. Its positive or negative sign reflects the overall deviation direction of the actual thermal recovery progress of the equipment from the predicted progress of the thermal recovery characteristic curve since the start of production.

[0067] As another optional implementation of this embodiment, before allocating orders in the preliminary order set for each injection molding equipment based on thermal adaptability, the process time requirement coefficient of each order in the preliminary order set is first corrected according to the mold changeover time required by each order.

[0068] The specific correction method is as follows: For adjacent orders that need to be continuously produced on the same equipment in the initial order set, when the product models required by the two orders are different and the mold needs to be changed, the process time requirement coefficient of the latter order will be increased by a mold switching compensation amount. The compensation amount is equal to the quotient obtained by dividing the average time required for the equipment to switch from the previous product mold to the current product mold by the steady-state cycle time of the order.

[0069] By increasing the mold changeover compensation, orders that require frequent mold changes are more likely to be assigned to positions with lower time flexibility coefficients during the matching process. This avoids the production schedule from being affected by too much time spent on frequent mold changes, and allows the production schedule to consider the impact of mold changeover time on the actual production cycle during the matching phase.

[0070] As another optional implementation of this embodiment, when calculating the expected production end time of the orders involved in the exchange after performing the exchange operation, in addition to accumulating the corrected planned single-piece cycle time of each sequential position, the proportion of sprue material remixing used in each order is also added as an additional time correction item to the accumulation result.

[0071] The specific method of inclusion is as follows: When an order requires the use of injection molding materials containing sprue remixing, the corresponding single-piece cycle time extension factor is obtained by referring to a table based on the sprue remixing ratio for that order. This extension factor is then multiplied by the corrected planned single-piece cycle time for that order and added to the calculation of the order's estimated production end time. By introducing the time correction based on the sprue remixing ratio, the calculation of the estimated production end time more closely reflects the actual cycle time extension characteristics when using remixed materials in injection molding production, thus improving the accuracy of delivery date constraint verification.

[0072] C4. When the absolute value of the real-time thermal state offset of any device in the injection molding equipment exceeds the preset offset threshold, the unexecuted task queue of the injection molding equipment is dynamically reconstructed.

[0073] In this embodiment of the invention, the preset offset threshold is the product of the steady-state cycle time of the injection molding equipment divided by the predicted cycle ratio of the first mold position of the thermal recovery characteristic curve, and then multiplied by 5%. This product is a dimensionless number and is used to compare with the real-time thermal state offset in the same dimension.

[0074] The value is determined based on the allowable fluctuation limit of the single-mold cycle ratio of the injection molding equipment under normal production conditions. This fluctuation limit is jointly determined by the minimum resolution of the timer in the injection molding equipment control system and the fluctuation range of the mold closing pressure.

[0075] The absolute value of the real-time thermal state offset is compared with the preset offset threshold. When the absolute value of the real-time thermal state offset is greater than 5% of the steady-state cycle time, it is determined that the actual thermal recovery progress of the device deviates from the predicted progress of the thermal recovery characteristic curve beyond the normal fluctuation range, and dynamic reconstruction needs to be triggered.

[0076] Extract all unexecuted orders for the equipment from the initial production schedule dataset. For each unexecuted order, extract its order identifier and process time requirement coefficient. Arrange all extracted orders in ascending order of their original position in the initial production schedule dataset. The resulting sequence serves as the unexecuted task queue for the equipment. Each task unit in this queue contains two fields: order identifier and process time requirement coefficient. The original order position information of each order after the arrangement is only used for this sorting and is not passed on with the task units.

[0077] The above information is arranged in ascending order of its original position to form the unexecuted task queue of the equipment. Each task unit in the queue contains three fields: order identifier, process time requirement coefficient, and original position number.

[0078] Obtain the currently unused sequential positions of the device and their corresponding elasticity coefficient ranges. Unused sequential positions refer to all positions from the next position after the currently executed module to the position of the planned total number of modules.

[0079] The negative of the real-time thermal state offset of the equipment during the current production period is used as the translation amount. For each unused sequential position, the translation amount is added to the lower limit and upper limit of its elastic coefficient range. The translation amount can be positive, zero, or negative. After translation, the sequential position obtains a new lower limit and a new upper limit. The new lower limit and the new upper limit constitute the corrected elastic coefficient range of the sequential position.

[0080] If the lower limit of the corrected elastic coefficient interval at a certain sequential position is greater than the upper limit, that is, the upper and lower limits of the interval are reversed after translation, then the lower and upper limits of the sequential position are reset to the arithmetic mean of the lower and upper limits before correction. This arithmetic mean is equal to the center value of the original elastic coefficient interval plus the translation amount. At this time, the lower and upper limits are equal, and the corrected elastic coefficient interval shrinks to a point. The value of this point is the center value plus the translation amount.

[0081] After performing the above translation and reversal correction operations on all unused sequential positions, arrange each position and its corresponding corrected elastic coefficient interval in ascending order of sequential position to generate a sequence of corrected elastic coefficient intervals.

[0082] Arrange the orders in the unexecuted task queue according to their original order position to generate the initial reconstruction sequence queue.

[0083] Starting from the first order in the initial reconstructed sequence queue and the first sequential position in the corrected elastic coefficient interval sequence, all orders in the unexecuted task queue are reassigned to their respective sequential positions in the corrected elastic coefficient interval sequence using the same deviation absolute value minimization matching method as the initial matching.

[0084] The specific matching method is as follows: compare the process time requirement coefficient of the current order with the adjusted elasticity coefficient range of the current sequence position. If it falls within the range, it is fixedly assigned. If it is less than the lower limit, it is swapped with the unassigned order in the queue with the largest process time requirement coefficient. If it is greater than the upper limit, it is swapped with the unassigned order in the queue with the smallest process time requirement coefficient. After the swap, the comparison is repeated, and the process is repeated until all orders are assigned.

[0085] For each reassigned sequence position, the process time requirement coefficient of the order at that position is compared with the lower limit of the adjusted elasticity coefficient range for that position. The larger of the two values ​​is taken as the adjusted planned cycle coefficient for that order at that restructured position. This adjusted planned cycle coefficient is then multiplied by the steady-state cycle time of the product required by the order to obtain the adjusted planned unit cycle time for each order in the restructured execution sequence.

[0086] Starting from the last completion time of the portion of the injection molding equipment has executed within the current production schedule, the corrected planned single-piece cycle time of each order is accumulated sequentially according to the order of each order in the reconstructed execution sequence to determine the expected production start time and expected production end time of each order after the reconstruction.

[0087] The information of each order in the initial production schedule dataset that has been executed remains unchanged. The information of each order in the reconstructed execution sequence is concatenated after the executed part according to its reconstructed order. The reconstructed complete execution sequence replaces the corresponding unexecuted part in the original initial production schedule dataset, generating a dynamically reconstructed production schedule dataset.

[0088] After dynamic reconstruction, the executed part of the production schedule data and the unexecuted part after reconstruction are continuous in time. The estimated production start time of the first order after reconstruction is equal to the estimated production end time of the last order in the executed part.

[0089] As another optional implementation of this embodiment, when the absolute value of the real-time thermal state offset exceeds the preset offset threshold and triggers dynamic reconstruction, during the process of reallocating unexecuted orders to the corrected elastic coefficient interval sequence, a hard constraint condition is added to the order with the most urgent delivery time in the unexecuted task queue. The order with the most urgent delivery time is the order with the highest delivery time in the unexecuted task queue. The basis for determining this proportion is to ensure that the delivery time of urgent orders is prioritized while reserving sufficient thermal state adaptation adjustment space for the remaining 80% of orders, so as to avoid the optimization effect of dynamic reconstruction being weakened due to the excessively large range of hard constraints.

[0090] The specific hard constraint is that the order with the most urgent delivery time must be in the reconstructed execution sequence no later than its original order position in the initial production scheduling dataset; that is, the order cannot be swapped to a position later than its original position.

[0091] By imposing this hard constraint, we ensure that urgent orders are not rescheduled to later production positions during dynamic reconfiguration in pursuit of optimal thermal adaptation, thus guaranteeing that the delivery dates of urgent orders are prioritized during dynamic reconfiguration.

[0092] As another optional implementation of this embodiment, when there are multiple consecutive unexecuted task queues for dynamic reconstruction of the injection molding equipment, that is, the equipment is triggered two or more times in the same production scheduling period, the number of consecutive triggers for reconstruction is recorded. When the number of consecutive triggers reaches the preset reconstruction number threshold, the heat recovery characteristic curve currently used by the equipment is shifted downward by the preset curve correction step size.

[0093] The specific translation method is as follows: the thermal recovery characteristic value at each module position on the thermal recovery characteristic curve is multiplied by a correction coefficient, which is equal to the product of the ratio of the number of consecutively triggered reconstructions to the preset reconstruction number threshold and 0.02.

[0094] The value of this correction coefficient increases the overall ratio of the prediction period at each position of the corrected curve, and the increase increases with the number of consecutive reconstruction triggers. The upper limit of the correction amount does not exceed 10% of the thermal recovery characteristic value itself, enabling the thermal recovery characteristic curve to achieve adaptive adjustment at the curve level when multiple reconstructions still cannot eliminate the deviation. The corrected thermal recovery characteristic curve is used as the input replacement curve for step C1 in subsequent production scheduling, and the corrected curve is directly used as the thermal recovery characteristic curve in the next production scheduling.

[0095] like Figure 2 The diagram shown is a functional block diagram of an injection molding production scheduling system based on multi-dimensional fitting provided in an embodiment of the present invention.

[0096] The injection molding scheduling system 100 based on multi-dimensional fitting described in this invention can be installed in an electronic device. Depending on the functions implemented, the injection molding scheduling system 100 based on multi-dimensional fitting may include a heat recovery curve generation module 101, an elastic interval sequence generation module 102, an adaptation allocation monitoring module 103, and a dynamic queue reconstruction module 104. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0097] In this embodiment, the functions of each module / unit are as follows: The heat recovery curve generation module 101 is used to generate heat recovery characteristic curves for the injection molding equipment based on the thermal history operation record set of the injection molding equipment. The elastic interval sequence generation module 102 is used to determine the preliminary order attribution relationship and planned total number of molds for the injection molding equipment in the current production scheduling period based on the production scheduling basic dataset of the injection molding equipment, and generate the time elasticity coefficient interval sequence of the injection molding equipment at each sequential position in the current production scheduling period based on the planned total number of molds and the heat recovery characteristic curve. The adaptation allocation monitoring module 103 is used to perform thermal state adaptation allocation on the orders in the preliminary order affiliation relationship according to the time elasticity coefficient interval sequence, to obtain the initial production schedule dataset, and to determine the real-time thermal state offset of the injection molding equipment in the current production schedule period according to the real-time cycle time of the molds already executed on the injection molding equipment during the production scheduling process based on the initial production schedule dataset. The dynamic queue reconstruction module 104 is used to trigger the dynamic reconstruction of the unexecuted task queue of the injection molding equipment when the absolute value of the real-time thermal state offset of any device in the injection molding equipment exceeds a preset offset threshold.

[0098] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0099] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0100] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0101] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0102] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An injection molding production scheduling method based on multi-dimensional fitting, characterized in that, The method includes: C1. Generate thermal recovery characteristic curves for the injection molding equipment based on the thermal history operation record set of the injection molding equipment; C2. Based on the production scheduling data set of the injection molding equipment, determine the preliminary order attribution relationship and planned total number of molds for the injection molding equipment in the current production scheduling period. Based on the planned total number of molds and the heat recovery characteristic curve, generate the time elasticity coefficient interval sequence of the injection molding equipment at each sequential position in the current production scheduling period. C3. Based on the time elasticity coefficient interval sequence, perform thermal state adaptability allocation on the orders in the preliminary order attribution relationship to obtain the initial production schedule dataset. During the production scheduling process based on the initial production schedule dataset, determine the real-time thermal state offset of the injection molding equipment in the current production scheduling period based on the real-time cycle time of the molds already executed on the injection molding equipment. C4. When the absolute value of the real-time thermal state offset of any device in the injection molding equipment exceeds the preset offset threshold, the unexecuted task queue of the injection molding equipment is dynamically reconstructed.

2. The injection molding production scheduling method based on multi-dimensional fitting as described in claim 1, characterized in that, The step of generating thermal recovery characteristic curves for the injection molding equipment based on its thermal history operating record set includes: Extract the continuous operating cycle sequence after each cold start of the injection molding equipment from the thermal history operation record set of the injection molding equipment to obtain the start-up operation cycle set of each cold start; Divide the actual cycle time of the module position in the startup operation cycle set by the average cycle time of the module at the end of the startup operation cycle set that is in a thermal equilibrium stable state to obtain the single startup cycle ratio of the module position. The median of the single-start cycle ratio of the same mold position in all historical cold starts of the injection molding equipment is used to obtain the thermal recovery characteristic value of the injection molding equipment at the corresponding mold position. Based on the increasing order of the mold positions, the mold positions and their corresponding heat recovery characteristic values ​​are arranged to generate the heat recovery characteristic curve of the injection molding equipment.

3. The injection molding production scheduling method based on multi-dimensional fitting as described in claim 1, characterized in that, The step of determining the preliminary order attribution and planned total number of molding cycles for the injection molding equipment within the current production scheduling period based on the production scheduling dataset of the injection molding equipment includes: Based on the product models required by the orders to be scheduled in the production scheduling base dataset corresponding to the injection molding equipment, the set of available injection molding equipment is selected from the mold compatibility information of the injection molding equipment in the production scheduling base dataset. Based on the available capacity information of injection molding equipment in the production scheduling base dataset, the orders to be scheduled are initially allocated to injection molding equipment whose available capacity meets the product quantity requirements of the orders to be scheduled, thus obtaining the preliminary order set of the injection molding equipment. The preliminary order set is associated with and stored with the equipment identification information of the corresponding injection molding equipment to generate a preliminary order attribution relationship; Based on the correspondence between the product quantity and the number of molding cycles required by each order in the preliminary order set, the product quantities required by each order in the preliminary order set are summed, and the sum is used as the planned total number of molding cycles for the injection molding equipment.

4. The injection molding production scheduling method based on multi-dimensional fitting as described in claim 3, characterized in that, The step of generating a time elasticity coefficient interval sequence for each sequential position of the injection molding equipment within the current production schedule, based on the planned total number of molding cycles and the heat recovery characteristic curve, includes: Based on the planned total number of molding cycles, determine the sequential position of the injection molding equipment within the current production time period; Based on the sequential position of the injection molding equipment in the current production period, determine the predicted cycle ratio on the heat recovery characteristic curve corresponding to the current sequential position, and use the reciprocal of the predicted cycle ratio as the benchmark coefficient of the current sequential position; Multiply the benchmark coefficient by the corresponding lower limit tolerance coefficient to obtain the lower limit coefficient of the current sequence position, and multiply the benchmark coefficient by the corresponding upper limit tolerance coefficient to obtain the upper limit coefficient of the current sequence position; Based on the lower limit coefficient and the upper limit coefficient, a time elasticity coefficient interval for the current sequential position is constructed, and the time elasticity coefficient intervals for each sequential position are arranged to obtain a time elasticity coefficient interval sequence.

5. The injection molding production scheduling method based on multi-dimensional fitting as described in claim 1, characterized in that, The step of performing hot-state adaptive allocation of orders in the preliminary order attribution relationship based on the time elasticity coefficient interval sequence to obtain the initial production scheduling plan dataset includes: The orders in the preliminary order set in the preliminary order attribution relationship are sorted from earliest to latest according to the delivery time of the orders. Based on the sorting result, the values ​​of the elastic coefficient intervals corresponding to each sequential position in the time elastic coefficient interval sequence are taken as the process time demand coefficients. The process time requirement coefficient is matched with the elastic coefficient interval of each sequential position in the time elastic coefficient interval sequence by minimizing the absolute value of the deviation, and the matching result is used to determine the internal production order of the order in each of the injection molding machines, as well as the planning cycle coefficient at each sequential position. Multiply the planned cycle coefficient of each sequential position by the steady-state cycle time of the product required by the order to obtain the corrected planned single-piece cycle time at each sequential position; According to the internal production order, starting from the start time of the injection molding equipment in the current production period, the corrected planned single-piece cycle time is accumulated in ascending order of each sequence position. The accumulated time up to the end of the previous position corresponding to the current sequence position is the expected production start time, and the accumulated time up to the end of the current sequence position is the expected production end time. The production sequence, the expected production start time, the expected production end time, and the planning cycle coefficient are associated and stored to generate an initial production scheduling dataset.

6. The injection molding production scheduling method based on multi-dimensional fitting as described in claim 5, characterized in that, The step of matching the process time requirement coefficient with the elasticity coefficient intervals at each sequential position in the time elasticity coefficient interval sequence by minimizing the absolute value of the deviation, and using the matching result to determine the internal production order of the order in each of the injection molding machines, and the planning cycle coefficient at each sequential position, includes: Based on the delivery time sequence of the orders in the preliminary order set, the preliminary order set is arranged. Based on the initial order queue obtained by the arrangement, a matching queue is generated. In the initial order queue, the process time requirement coefficient of the current order and the elasticity coefficient range of the current order position are compared. If the process time requirement coefficient of the current order falls within the elasticity coefficient range of the current sequence position, then the current order will be assigned to the current sequence position, and the next order and the next sequence position in the queue to be matched will continue to be processed. If the process time requirement coefficient of the current order is less than the lower limit of the elasticity coefficient range of the current sequence position, then the order with the largest process time requirement coefficient from the unallocated orders in the matching queue is swapped with the current order in the initial sequence queue. The order in the current position after the swap is taken as the new current order and the comparison is repeated. If the process time requirement coefficient of the current order is greater than the upper limit of the elasticity coefficient range of the current sequence position, then the order with the smallest process time requirement coefficient from the unallocated orders in the matching queue will be swapped with the current order in the initial sequence queue. The order in the current position after the swap will be used as the new current order, and the comparison will be repeated. Once all orders for each injection molding machine have been assigned to their corresponding sequence positions, the larger of the process time requirement coefficient of the order and the lower limit of the elasticity coefficient range corresponding to the sequence position will be used as the planning cycle coefficient of the order at the corresponding sequence position. The product of the planned cycle coefficient of the order at each sequential position and the steady-state cycle time of the product required by the order is used as the corrected planned single-piece cycle time. According to the order of the orders in the sequential positions, the corrected planned single-piece cycle times at each sequential position are arranged to generate the internal production order of the orders in the injection molding equipment.

7. The injection molding production scheduling method based on multi-dimensional fitting as described in claim 6, characterized in that, The comparison of the process time requirement coefficient of the current order with the elasticity coefficient range of the current sequence position also includes: After the exchange process is executed for the order with the largest process time requirement coefficient, or after the exchange process is executed for the order with the smallest process time requirement coefficient, the cumulative time of the two orders after the exchange, from the start time of production scheduling to the end time of each position, is taken as the expected production end time of the two orders involved in the exchange process after the exchange. The estimated production end time after the exchange is compared with the delivery time of the two orders respectively; If the expected production end time of an order after the exchange is later than the corresponding delivery time, the exchange is deemed invalid, the exchange is cancelled, and the current order before the exchange is fixedly assigned to the current sequence position, and the current sequence position is marked as a position that can no longer be exchanged. If the estimated production completion time of both orders after the exchange is no later than the corresponding delivery time, the exchange is deemed valid.

8. The injection molding production scheduling method based on multi-dimensional fitting as described in claim 1, characterized in that, During the production scheduling process based on the initial production scheduling plan dataset, determining the real-time thermal state offset of the injection molding equipment within the current production scheduling period based on the real-time cycle time of the molded cycles already executed on the injection molding equipment includes: Divide the real-time cycle time of the completed molding cycle on the injection molding equipment by the steady-state cycle time of the injection molding equipment to obtain the real-time cycle ratio of the completed molding cycle. Read the ratio of the prediction period to the sequential position corresponding to the executed module from the thermal recovery characteristic curve; Subtracting the predicted cycle ratio from the real-time cycle ratio yields the single-mode thermal state deviation value of the executed mode. The thermal state deviation values ​​of all single-mold cycles executed between the start time of the current production scheduling period of the injection molding equipment and the current time are summed to obtain the real-time thermal state offset of the injection molding equipment in the current production scheduling period.

9. The injection molding production scheduling method based on multi-dimensional fitting as described in claim 1, characterized in that, The triggering of dynamic reconstruction of the unexecuted task queue of the injection molding equipment includes: Using the negative of the real-time thermal state offset as the translation amount, the translation amount is added to the upper and lower limits of the elastic coefficient interval corresponding to each unused sequential position in the injection molding equipment to generate the corrected elastic coefficient interval for each sequential position. Arrange the unused sequential positions and their corresponding corrected elastic coefficient intervals to generate a corrected elastic coefficient interval sequence; The objective is to minimize the absolute value of the deviation between the process time requirement coefficient of each order in the unexecuted task queue and the upper and lower limits of the corrected elastic coefficient interval corresponding to each sequential position in the corrected elastic coefficient interval sequence. Each order in the queue of unexecuted tasks is reassigned to its corresponding position in the corrected elasticity coefficient interval sequence; Based on the reallocation results, the corrected planned cycle time for each order in the reconstructed execution sequence is generated. The reconstructed execution sequence replaces the corresponding unexecuted part in the initial production schedule dataset, generating a dynamically reconstructed production schedule dataset.

10. An injection molding production scheduling system based on multi-dimensional fitting, characterized in that, The system for implementing the injection molding scheduling method based on multi-dimensional fitting as described in claim 1 includes: The heat recovery curve generation module is used to generate heat recovery characteristic curves for the injection molding equipment based on the thermal history operation record set of the injection molding equipment. The elastic interval sequence generation module is used to determine the preliminary order attribution relationship and planned total number of molds for the injection molding equipment in the current production scheduling period based on the production scheduling basic dataset of the injection molding equipment, and generate the time elasticity coefficient interval sequence of the injection molding equipment at each sequential position in the current production scheduling period based on the planned total number of molds and the heat recovery characteristic curve. The adaptation and allocation monitoring module is used to perform thermal state adaptation allocation on the orders in the preliminary order affiliation relationship according to the time elasticity coefficient interval sequence, to obtain the initial production scheduling plan dataset, and to determine the real-time thermal state offset of the injection molding equipment in the current production scheduling period according to the real-time cycle time of the molds already executed on the injection molding equipment during the production scheduling process based on the initial production scheduling plan dataset. The dynamic queue reconstruction module is used to trigger the dynamic reconstruction of the unexecuted task queue of the injection molding equipment when the absolute value of the real-time thermal state offset of any device in the injection molding equipment exceeds a preset offset threshold.

Images