Injection molding energy consumption dynamic optimization control method and device for injection molding machine
By acquiring historical operating records and analyzing digital twin models of two-platen direct-pressure injection molding machines, and combining this with distributed sensing devices for real-time energy consumption monitoring and optimization, the problem of lagging energy consumption optimization in injection molding machines has been solved. This has enabled accurate energy consumption prediction and adaptive optimization, thereby reducing energy consumption.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NANTONG SIZE PLASTIC CO LTD
- Filing Date
- 2025-12-18
- Publication Date
- 2026-04-10
AI Technical Summary
The current energy consumption optimization of injection molding machines lacks dynamic real-time monitoring and prediction capabilities, resulting in inaccurate energy efficiency control, lagging optimization, and difficulty in adapting to rapidly changing working conditions.
By acquiring valid historical operating records of a two-platen direct-pressure injection molding machine, simulation analysis is performed using a digital twin model, and dynamic monitoring is conducted using distributed sensing devices to predict energy consumption in real time. When the energy consumption does not meet the predetermined energy consumption threshold, optimization plans are retrieved for energy consumption optimization.
It achieves accurate energy consumption prediction and adaptive optimization during the key operating stages of injection molding machines, significantly reducing energy consumption and improving production efficiency and equipment reliability.
Smart Images
Figure CN121821735A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the technical field of energy consumption optimization for injection molding machines, specifically to a method and device for dynamic optimization and control of injection molding energy consumption for injection molding machines. Background Technology
[0002] Injection molding machines, as core equipment in the plastics processing industry, often suffer from low energy efficiency and unnecessary energy waste due to unreasonable process parameter settings, insufficient equipment status monitoring, or limited dynamic adjustment capabilities. This is especially true in two-platen direct-pressure injection molding machines, where the dynamic performance of the mold clamping mechanism significantly impacts energy consumption. Due to their complex mechanical structure and variable operating conditions, traditional methods struggle to achieve precise energy consumption control. Digital twins, by constructing virtual mapping models of physical equipment and combining them with real-time data-driven simulation analysis, can achieve dynamic monitoring, prediction, and optimization of equipment operating status. However, existing digital twin modeling for injection molding machines lacks refined energy consumption analysis for key operational stages such as mold clamping, injection, and holding pressure, resulting in limited energy consumption optimization effects and difficulty adapting to the rapidly changing operating conditions of injection molding machines. Consequently, it fails to reduce energy consumption while maintaining production efficiency.
[0003] Therefore, current technologies for injection molding machines suffer from a lack of dynamic real-time monitoring and prediction capabilities, leading to inaccurate energy efficiency control and delayed optimization. Summary of the Invention
[0004] This application provides a method and device for dynamic optimization and control of injection molding energy consumption for injection molding machines. It solves the technical problem in the prior art that the lack of dynamic real-time monitoring and prediction capabilities for energy consumption optimization of injection molding machines leads to inaccurate energy efficiency control and lagging optimization. It achieves the technical effect of accurate energy consumption prediction and adaptive optimization in the key operating stages of injection molding machines, and significantly reduces energy consumption.
[0005] This application provides a method for dynamic optimization and control of injection molding energy consumption for injection molding machines. The method includes: acquiring valid historical operating records of a two-platen direct-pressure injection molding machine, wherein the valid historical operating records include multiple sets of operating data from multiple cycles; acquiring a target stage in a predetermined operating phase and matching the target operating information of the target stage with the multiple sets of operating data; preprocessing and analyzing the target operating information to obtain target simulation data of the target stage, and activating a digital twin model to simulate and analyze the target simulation data to obtain a target energy consumption index; dynamically and periodically monitoring the mold closing mechanism of the two-platen direct-pressure injection molding machine through distributed sensing devices to obtain dynamic sensing data; comparing the dynamic sensing data with the target simulation data to obtain a real-time predicted energy consumption index; and when the real-time predicted energy consumption index does not meet the target predetermined energy consumption threshold, retrieving an energy consumption optimization plan to optimize the energy consumption of the two-platen direct-pressure injection molding machine.
[0006] In a possible implementation, the injection molding energy consumption dynamic optimization and control method for injection molding machines also performs the following processing: the predetermined operation phase includes the mold opening and closing phase, the mold locking phase, and the ejection phase of the two-platen direct-pressure injection molding machine.
[0007] In a possible implementation, the injection molding energy consumption dynamic optimization and control method for injection molding machines further performs the following processing: extracting arbitrary operating data corresponding to any period in the target operating information, wherein the arbitrary operating data includes first operating data corresponding to a first operating index; performing random sampling and fitting on the first operating data to obtain a first spline curve; extracting verification operating data corresponding to the verification period in the target operating information, and obtaining a first verification curve of the first operating index based on the verification operating data; if the curve similarity between the first verification curve and the first spline curve reaches a predetermined similarity limit, then obtaining a first fitting polynomial of the first spline curve; obtaining first simulated data of the first operating index based on the first fitting polynomial, and forming the target simulated data.
[0008] In a possible implementation, the dynamic optimization and control method for injection molding energy consumption of injection molding machines further performs the following processing: the first operating index refers to any one of the predetermined operating condition indicators, which at least includes the clamping cylinder pressure, the moving cylinder displacement, the flow rate, and the motor power.
[0009] In a possible implementation, the injection molding energy consumption dynamic optimization and control method for injection molding machines further performs the following processing: matching the first sensing data corresponding to the first operating index in the dynamic sensing data; calculating the first distance value from the first sensing data to the first spline curve; determining whether the first distance value meets the predetermined distance value constraint; if not, issuing an operating abnormality signal, and providing a real-time operating abnormality warning for the two-platen direct-pressure injection molding machine based on the operating abnormality signal.
[0010] In a possible implementation, the injection molding energy consumption dynamic optimization and control method for injection molding machines further performs the following processing: if the first distance value satisfies the predetermined distance value constraint, the dynamic sensing data and the target simulation data are compared to obtain the real-time predicted energy consumption index; wherein, the process includes: sequentially vectorizing the dynamic sensing data and the target simulation data to obtain the dynamic sensing vector and the target simulation vector respectively; comparing the cosine similarity between the dynamic sensing vector and the target simulation vector, and normalizing the results to obtain the real-time predicted energy consumption index.
[0011] In a possible implementation, the injection molding energy consumption dynamic optimization and control method for injection molding machines further performs the following processing: based on the first sensing data and the first simulation data, a first operating data pair is formed for the first operating index; based on the first operating data pair, multiple sets of operating data pairs are formed, wherein the multiple sets of operating data pairs include arbitrary point pairs; a predetermined distance grid is obtained, and arbitrary spatial distances of the arbitrary point pairs are filled into the predetermined distance grid to obtain an initial filled grid; an update filling strategy is obtained, and the initial filled grid is updated and filled according to the update filling strategy to obtain a target filled grid; the filling value of the predetermined grid in the target filled grid is taken as the Fraser distance; and the real-time predicted energy consumption index is verified by the normalized Fraser distance.
[0012] In a possible implementation, the injection molding energy consumption dynamic optimization and control method for injection molding machines further performs the following processing: partitioning the initial filling grid according to the update filling strategy to obtain partitioning results; obtaining update blocks in the partitioning results and extracting arbitrary grids in the update blocks, wherein the arbitrary grids correspond to arbitrary initial filling values; constructing an arbitrary reference grid set for the arbitrary grids and filtering the arbitrary reference grid set to obtain an arbitrary maximum grid filling value; replacing the arbitrary initial filling value with the sum of the arbitrary maximum grid filling value and the arbitrary initial filling value as the arbitrary update filling value of the arbitrary grids; and forming the target filling grid based on the arbitrary update filling value.
[0013] In a possible implementation, the injection molding energy consumption dynamic optimization and control method for injection molding machines further performs the following processing: the arbitrary reference grid set includes the left neighbor grid, the upper neighbor grid, and the upper left neighbor grid of the arbitrary grid.
[0014] This application also provides a dynamic optimization and control device for injection molding energy consumption of injection molding machines. The device includes: an effective historical operation record acquisition module for acquiring effective historical operation records of a two-platen direct-pressure injection molding machine, wherein the effective historical operation records include multiple sets of operation data for multiple cycles; a target operation information matching module for acquiring a target stage in a predetermined operation phase and matching the target operation information of the target stage in the multiple sets of operation data; a target energy consumption index acquisition module for preprocessing and analyzing the target operation information to obtain target simulation data of the target stage, and activating a digital twin model to simulate and analyze the target simulation data to obtain a target energy consumption index; a dynamic sensing data acquisition module for dynamically and periodically monitoring the mold closing mechanism of the two-platen direct-pressure injection molding machine through distributed sensing devices to obtain dynamic sensing data; a real-time predicted energy consumption index acquisition module for comparing the dynamic sensing data with the target simulation data to obtain a real-time predicted energy consumption index; and an energy consumption optimization module for retrieving an energy consumption optimization plan to optimize the energy consumption of the two-platen direct-pressure injection molding machine when the real-time predicted energy consumption index does not meet the target predetermined energy consumption threshold.
[0015] This application proposes a method and device for dynamic optimization and control of injection molding energy consumption in injection molding machines. The method involves acquiring valid historical operating records of a two-platen direct-pressure injection molding machine; obtaining target stages within a predetermined operating phase and matching target operating information; preprocessing and analyzing target simulation data and activating a digital twin model for simulation analysis to obtain a target energy consumption index; dynamically and periodically monitoring the mold clamping mechanism using distributed sensing devices to obtain dynamic sensing data; comparing this data to obtain a real-time predicted energy consumption index; and when the real-time predicted energy consumption index does not meet the target predetermined energy consumption threshold, retrieving an energy consumption optimization plan to optimize the energy consumption of the two-platen direct-pressure injection molding machine. This solves the technical problem in existing technologies where injection molding machine energy consumption optimization lacks dynamic real-time monitoring and prediction capabilities, leading to inaccurate energy efficiency control and delayed optimization. It achieves accurate energy consumption prediction and adaptive optimization in key operating stages of the injection molding machine, significantly reducing energy consumption. Attached Figure Description
[0016] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the apparatus according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.
[0017] Figure 1 This is a schematic diagram of the dynamic optimization and control method for injection molding energy consumption of injection molding machines provided in the embodiments of this application.
[0018] Figure 2 This is a schematic diagram of the dynamic optimization and control device for injection molding energy consumption of injection molding machines provided in an embodiment of this application.
[0019] Explanation of reference numerals in the attached diagram: 10 for obtaining valid historical operation records, 20 for matching target operation information, 30 for obtaining target energy consumption index, 40 for obtaining dynamic sensing data, 50 for obtaining real-time predicted energy consumption index, and 60 for energy consumption optimization. Detailed Implementation
[0020] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.
[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0022] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or apparatuses. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.
[0023] This application provides a method for dynamic optimization and control of injection molding energy consumption for injection molding machines, such as... Figure 1 As shown, the method includes: Step S100: Obtain the valid historical operation records of the two-platen direct-pressure injection molding machine, wherein the valid historical operation records include multiple sets of operation data from multiple cycles.
[0024] Preferably, stable operating phases of the injection molding machine are selected from the historical operation records of the two-platen direct-pressure injection molding machine. These are operation data recorded under normal, fault-free conditions. Operation records of abnormal shutdowns, maintenance, or large parameter fluctuations are excluded. Valid historical operation records are determined, including multiple complete production cycles, such as injection molding cycles. Each cycle contains multiple sets of operating parameters reflecting the equipment status, which may include mechanical parameters such as clamping cylinder pressure, mold moving cylinder displacement, and platen position; hydraulic parameters such as flow rate, oil temperature, and system pressure; electrical parameters such as motor power, current, and servo drive signals; and duration information for each stage of mold closing, injection, pressure holding, and ejection. The two-platen direct-pressure injection molding machine is a high-efficiency injection molding equipment that adopts a two-platen mold closing structure and a direct-pressure clamping structure. The two-platen mold closing structure consists of only two main parts: a moving platen and a fixed platen. The direct-pressure clamping structure directly applies and precisely controls the clamping force through a high-response servo hydraulic system, resulting in a more uniform clamping force distribution.
[0025] Step S200: Obtain the target stage in the predetermined operation phase, and match the target operation information of the target stage in the multiple sets of operation data.
[0026] Step S200 further includes the predetermined operating phase including the mold opening and closing phase, the mold locking phase, and the ejection phase of the two-platen direct-pressure injection molding machine.
[0027] Preferably, operational information for predetermined operational stages is accurately extracted from the multi-cycle effective historical operation records of a two-platen direct-pressure injection molding machine. These predetermined operational stages include the mold opening and closing stage, the mold locking stage, and the ejection stage. Specifically, the mold opening and closing stage refers to the process of the mold platen opening and closing, including parameters such as the displacement and speed of the mold-moving cylinder; the mold locking stage refers to the application of locking force after the mold platen closes to ensure a tight mold fit, with key parameters including the locking cylinder pressure and hydraulic pressure stability; the ejection stage refers to the ejection of the ejector pins after the injection-molded product is formed, monitoring the ejection speed and pressure. Then, one of the predetermined operational stages is selected as the target stage, and the corresponding operational data is accurately matched from multiple sets of operational data according to timestamps or process signals. For example, in the mold locking stage, the data segment of the locking cylinder pressure rising from 0 to its peak value is extracted; in the ejection stage, the timing data from the start of ejector pin displacement to the completion of reset is extracted. Finally, the target operational information is composed, ensuring that energy consumption optimization targets the key parameters of a specific stage, improving analysis efficiency and accuracy.
[0028] Step S300: Preprocess and analyze the target operation information to obtain target simulation data for the target stage, and activate the digital twin model to perform simulation analysis on the target simulation data to obtain the target energy consumption index.
[0029] Step S300 further includes step S310, extracting arbitrary running data corresponding to any period in the target running information, wherein the arbitrary running data includes first running data corresponding to the first running index; step S320, performing random sampling and fitting on the first running data to obtain a first spline curve; step S330, extracting verification running data corresponding to the verification period in the target running information, and obtaining a first verification curve of the first running index based on the verification running data; step S340, if the curve similarity between the first verification curve and the first spline curve reaches a predetermined similarity limit, then obtaining a first fitting polynomial of the first spline curve; step S350, obtaining first simulated data of the first running index based on the first fitting polynomial, and forming the target simulated data.
[0030] Furthermore, step S310 also includes the following: the first operating index refers to any one of the predetermined operating condition indexes, and the predetermined operating condition indexes include at least the clamping cylinder pressure, the moving cylinder displacement, the flow rate, and the motor power.
[0031] Preferably, the operating data corresponding to any cycle of the target stage (such as the 5th injection cycle) of the valid historical operating records is randomly extracted, i.e., arbitrary operating data, including the first operating data corresponding to the first operating index. The first operating index refers to any one of the predetermined operating condition indicators, which at least include the clamping cylinder pressure, the mold moving cylinder displacement, the flow rate, and the motor power. The clamping cylinder pressure refers to the locking force applied to the mold by the hydraulic system. Too high a pressure leads to energy waste, while too low a pressure may cause flash. The mold moving cylinder displacement refers to the real-time position of the moving platen. The displacement accuracy affects the mold closing speed and energy consumption, and deviation may cause mechanical wear. The flow rate refers to the volume of hydraulic oil flowing per unit time. Too high a flow rate increases the pump's energy consumption, while too low a flow rate reduces the response speed. The motor power refers to the real-time electrical power driving the hydraulic pump and injection unit, which directly reflects the energy consumption level. The optimization goal is to reduce peak power and improve energy efficiency. The first operating data may be a data sequence of the clamping cylinder pressure changing over time.
[0032] Preferably, the first operating data is randomly sampled, i.e., non-uniform sampling, to avoid overfitting. For example, operating data corresponding to several time points are randomly selected and fitted to generate a smooth first identical line curve, representing the trend of index change. Then, the verification operating data corresponding to the verification period is extracted from the target operating information. The verification period may be any period other than the one excluding any other period. The verification operating data is the operating data of the same first operating index as the first operating data. The verification operating data is randomly sampled and fitted to obtain the first verification curve of the first operating index. Then, the curve similarity between the first verification curve and the first identical line curve is calculated by dynamic time warping distance or cosine similarity. If the curve similarity reaches a predetermined similarity limit (e.g., 90%), the fitting is considered effective. The fitting polynomial is then derived from the first identical line curve to obtain the first fitting polynomial. The first simulated data of the first operating index is then obtained based on the first fitting polynomial. That is, the first fitting polynomial is applied to the entire time range to generate continuous and smooth first simulated data. Similarly, simulated data of all predetermined operating condition indices are generated to form complete target simulated data, ensuring that discrete historical operating data is transformed into a high-fidelity digital twin input.
[0033] Preferably, the digital twin model is activated to perform simulation analysis on the target simulation data, i.e., standardized energy efficiency scoring, to obtain the target energy consumption index, such as a value in the range of 0% to 100%, where a higher value indicates better energy efficiency. Specifically, a physical simulation model of a two-platen direct-pressure injection molding machine is constructed based on ANSYS or MATLAB and determined as a digital twin model. The target simulation data (such as clamping cylinder pressure data, mold moving cylinder displacement data, etc.) is input into the digital twin model for multi-physics simulation, including analyzing the coupling relationship of hydraulic system energy flow, mechanical structure dynamics, motor efficiency mapping, etc., and then calculating information such as hydraulic oil work, motor power consumption, and mechanical friction loss during the clamping stage. Then, the energy consumption index of each indicator is calculated according to the formula: energy consumption index = (theoretical minimum energy consumption ÷ actual simulation energy consumption) × 100%, and a weighted sum is performed to obtain the target energy consumption index, which characterizes the energy consumption efficiency of the two-platen direct-pressure injection molding machine in a specific operating stage.
[0034] Step S400: Dynamic periodic monitoring of the mold closing mechanism of the two-platen direct-pressure injection molding machine is performed using distributed sensing devices to obtain dynamic sensing data.
[0035] Preferably, a distributed sensing device installed on the mold clamping mechanism of a two-platen direct-pressure injection molding machine is used to dynamically and periodically monitor the mold clamping mechanism, collecting physical data related to the mold clamping action in real time to reflect the operating status of the equipment. Displacement sensors are arranged at the four corner points of the mold clamping mechanism to achieve parallelism closed-loop control. The distributed sensing device includes strain gauges for monitoring template deformation; magnetostrictive displacement sensors for obtaining the position of the moving template; vibration sensors for monitoring mechanical impact; and hydraulic pressure sensors for monitoring the clamping cylinder. Dynamic periodic monitoring is performed using the distributed sensors at a fixed frequency, for example, collecting clamping cylinder pressure data at a high frequency of 200Hz during the clamping stage and reducing the frequency to 10Hz during the ejection stage, thereby obtaining dynamic sensing data, which may include time-mold movement cylinder displacement / clamping cylinder pressure curves, flow rate, and motor power data, etc.
[0036] Furthermore, step S400 also includes step S410 matching the first sensing data corresponding to the first operating index in the dynamic sensing data; step S420 calculating the first distance value from the first sensing data to the first spline curve; step S430 determining whether the first distance value meets the predetermined distance value constraint; and step S440 issuing an operating abnormality signal if it does not meet the constraint, and providing a real-time operating abnormality warning for the two-platen direct-pressure injection molding machine based on the operating abnormality signal.
[0037] Preferably, from the dynamic sensing data collected by the distributed sensing devices, such as the clamping cylinder pressure curve, the first sensing data corresponding to the same first operating index as when modeling the digital twin is matched and extracted, that is, the real-time clamping cylinder pressure data sequence of the current period, is compared with the first line curve pre-generated by the digital twin. The first distance value between the two is calculated by dynamic time warping or Euclidean distance to quantify the degree of deviation. Then, it is determined whether the first distance value meets the predetermined distance value constraint (e.g., 10). If the first distance value does not meet the predetermined distance value constraint, it is determined to be an operating abnormality, and an operating abnormality signal is issued. Based on the operating abnormality signal, a real-time operating abnormality warning is issued for the two-platen direct pressure injection molding machine, that is, a warning signal is triggered. If the actual pressure curve is detected to be delayed by 50ms compared with the ideal curve and the amplitude exceeds the limit, it indicates that the hydraulic system response is lagging. The alarm is pushed to the operation interface in real time through the industrial Internet of Things platform, and the pre-plan is automatically executed, such as reducing the mold closing speed or stopping the machine for inspection.
[0038] Step S500: Compare the dynamic sensing data with the target simulation data to obtain the real-time predicted energy consumption index.
[0039] Step S500 further includes step S510, which involves sequentially vectorizing the dynamic sensing data and the target simulation data to obtain the dynamic sensing vector and the target simulation vector, respectively; step S520, which involves comparing the cosine similarity between the dynamic sensing vector and the target simulation vector and normalizing the results to obtain the real-time predicted energy consumption index.
[0040] Preferably, the real-time collected dynamic sensing data such as clamping cylinder pressure and mold moving cylinder displacement are converted into feature vectors. For example, statistical quantities such as mean, variance, and peak value are extracted through a sliding window to form a dynamic sensing vector. The target simulation data is also vectorized to ensure that the vector dimensions are consistent, forming a target simulation vector. Then, the cosine similarity between the dynamic sensing vector and the target simulation vector is calculated. That is, the degree of deviation between the dynamic sensing data and the target simulation data is evaluated by the cosine value of the vector angle. The result range is [-1, 1]. The closer it is to 1, the closer the energy consumption characteristics are to the ideal state. Then, normalization is performed to map the cosine value to the interval [0, 100] to determine the real-time predicted energy consumption index, providing quantitative and executable decision support for energy saving and consumption reduction of injection molding machines.
[0041] Furthermore, step S520 also includes step S521, forming a first operational data pair for the first operational index based on the first perceived data and the first simulated data; step S522, forming multiple sets of operational data pairs based on the first operational data pair, wherein the multiple sets of operational data pairs include arbitrary point pairs; step S523, obtaining a predetermined distance grid and filling the predetermined distance grid with arbitrary spatial distances of the arbitrary point pairs to obtain an initial filled grid; step S524, obtaining an update filling strategy and updating and filling the initial filled grid according to the update filling strategy to obtain a target filled grid; step S525, taking the filling value of the predetermined grid in the target filled grid as the Fraser distance; and step S526, verifying the real-time predicted energy consumption index using the normalized Fraser distance.
[0042] Preferably, based on the first operational indicator, corresponding first sensing data and first simulation data are extracted from the dynamic sensing data and target simulation data, respectively. These are aligned by time points to form first operational data pairs. Multiple expansions are performed on these first operational data pairs to generate multiple operational data pairs, meaning each indicator independently generates an operational data pair. These multiple operational data pairs include arbitrary point pairs. A two-dimensional grid with a predetermined distance is constructed, where the horizontal axis represents simulation data values and the vertical axis represents sensing data values. Arbitrary spatial distances between arbitrary point pairs are filled into the predetermined distance grid. Grid cells store the spatial distances of corresponding data points. The distance values of all data pairs are filled into the corresponding positions in the grid to determine the initial filled grid. Then, an update filling strategy is obtained, including partition filtering and neighborhood propagation. Partition filtering divides the grid into several blocks, prioritizing high-density areas, such as pressure peak intervals. Neighborhood propagation involves taking the maximum filling value of each grid cell's left, top, and top-left neighbors and accumulating it with the current value for updating. Finally, the initial filled grid is updated and filled according to the update filling strategy to determine the target filled grid. The filling value of the predetermined grid in the target filling grid is taken, that is, the minimum and maximum path values from the starting point to the ending point are extracted from the target filling grid as the Fraser distance, which represents the maximum local deviation of the two curves in the spatiotemporal dimension. It is better than the global average of Euclidean distance. For example, if the Fraser distance = 15, it means that the peak pressure-time deviation between the measured curve and the simulated curve at a certain moment reaches 15 units. The Fraser distance is normalized, including dividing by the historical maximum deviation value to map the Fraser distance to the interval [0, 1]. Finally, the real-time predicted energy consumption index is verified. If the normalized value ≤ 0.1, the real-time energy consumption index is confirmed to be reliable and the current optimization strategy is maintained. If 0.1 < value ≤ 0.3, the digital twin model is recalibrated. If the value > 0.3, it is judged as a serious deviation, the optimization is frozen and an alarm is triggered.
[0043] Furthermore, step S524 also includes step a, partitioning the initial filling grid according to the update filling strategy to obtain a partitioning result; step b, obtaining the update block in the partitioning result and extracting any grid in the update block, wherein the arbitrary grid corresponds to an arbitrary initial filling value; step c, constructing an arbitrary reference grid set for the arbitrary grid and filtering the arbitrary reference grid set to obtain an arbitrary maximum grid filling value; step d, replacing the arbitrary initial filling value with the sum of the arbitrary maximum grid filling value and the arbitrary initial filling value as the arbitrary update filling value of the arbitrary grid; and step e, forming the target filling grid based on the arbitrary update filling value.
[0044] Furthermore, step c also includes the arbitrary reference grid set including the left neighbor grid, the top neighbor grid, and the top-left neighbor grid of the arbitrary grid.
[0045] Preferably, according to the update filling strategy, the initial filling grid is divided into several blocks according to data density or gradient changes to obtain partitioning results, such as a pressure rapidly rising area, a stable area, etc. Then, the area with drastic data changes in the partitioning results is extracted as the update block, such as the block near the pressure peak, and the grid in the update block is randomly selected as an arbitrary grid and the corresponding initial filling value is obtained, where the arbitrary grid corresponds to an arbitrary initial filling value; for the arbitrary grid to be updated, its three adjacent grids are selected as references, including the left neighbor grid (the previous column in the same row), the top neighbor grid (the previous row in the same column), and the top-left neighbor grid (the previous row and column). Next, select the maximum fill value of three adjacent grids from any reference grid set as the maximum fill value of any grid. Then, calculate the sum of the maximum fill value and the initial fill value, and use it to replace the initial fill value as the updated fill value of any grid. For example, if the initial value of the current grid is 2.4 and the maximum fill value of the adjacent grid is 3.1, then the updated fill value of any grid is 5.5. Finally, iterate through the grids of all blocks to update the fill value, and finally form the target filled grid. Backtrack from the end point of the target filled grid (the last row and the last column) to the starting point to find the minimum and maximum value path. The peak value is the final Fraser distance.
[0046] Step S600: When the real-time predicted energy consumption index does not meet the target predetermined energy consumption threshold, the energy consumption optimization plan is retrieved to optimize the energy consumption of the two-platen direct-pressure injection molding machine.
[0047] Preferably, if the real-time predicted energy consumption index does not meet the target predetermined energy consumption threshold, energy consumption optimization is automatically triggered. Specifically, an energy consumption optimization plan matching the current abnormal mode is retrieved from the pre-stored optimization strategy library. For example, if the real-time predicted energy consumption index deviates slightly, a parameter fine-tuning strategy is invoked, such as reducing the clamping pressure by 5%; if the real-time predicted energy consumption index deviates significantly, a composite strategy is activated, such as adjusting the clamping cylinder pressure and hydraulic oil cooling. Then, the energy consumption of the two-platen direct-pressure injection molding machine is optimized using the energy consumption optimization plan, that is, the equipment parameters are adjusted in real time through the PLC or servo system, such as adjusting the proportional valve opening to optimize the flow rate, reducing the motor speed and enabling the energy feedback mode, or extending the holding time to compensate for pressure loss. Finally, the operating data is re-collected to calculate the real-time predicted energy consumption index until it reaches the target, ultimately realizing intelligent dynamic optimization of the energy consumption of the two-platen direct-pressure injection molding machine, significantly reducing energy consumption, improving production efficiency, and enhancing the reliability of the injection molding machine operation.
[0048] In the above text, refer to Figure 1 This paper describes in detail a dynamic optimization and control method for injection molding energy consumption of injection molding machines according to embodiments of the present invention. Next, we will refer to... Figure 2 This invention describes a dynamic optimization and control device for injection molding energy consumption of injection molding machines according to an embodiment of the present invention.
[0049] The injection molding energy consumption dynamic optimization and control device for injection molding machines according to embodiments of the present invention is used to solve the technical problems in the prior art where the energy consumption optimization of injection molding machines lacks dynamic real-time monitoring and prediction capabilities, resulting in inaccurate energy efficiency control and lagging optimization. It achieves the technical effect of realizing accurate energy consumption prediction and adaptive optimization in key operating stages of injection molding machines, significantly reducing energy consumption. Figure 2 As shown, the injection molding energy consumption dynamic optimization and control device for injection molding machines includes: an effective historical operation record acquisition module 10, a target operation information matching module 20, a target energy consumption index acquisition module 30, a dynamic sensing data acquisition module 40, a real-time predicted energy consumption index acquisition module 50, and an energy consumption optimization module 60.
[0050] The system includes: a valid historical operation record acquisition module 10, used to acquire valid historical operation records of the two-platen direct-pressure injection molding machine, wherein the valid historical operation records include multiple sets of operation data for multiple cycles; a target operation information matching module 20, used to acquire a target stage in a predetermined operation phase and match the target operation information of the target stage in the multiple sets of operation data; a target energy consumption index acquisition module 30, used to preprocess and analyze the target operation information to obtain target simulation data of the target stage, and activate a digital twin model to simulate and analyze the target simulation data to obtain a target energy consumption index; a dynamic sensing data acquisition module 40, used to dynamically and periodically monitor the mold closing mechanism of the two-platen direct-pressure injection molding machine through distributed sensing devices to obtain dynamic sensing data; a real-time predicted energy consumption index acquisition module 50, used to compare the dynamic sensing data with the target simulation data to obtain a real-time predicted energy consumption index; and an energy consumption optimization module 60, used to retrieve an energy consumption optimization plan to optimize the energy consumption of the two-platen direct-pressure injection molding machine when the real-time predicted energy consumption index does not meet the target predetermined energy consumption threshold.
[0051] The specific configuration of the target operation information matching module 20 will be described in detail below. The target operation information matching module 20 further includes: the predetermined operation stage includes the mold opening and closing stage, the mold locking stage, and the ejection stage of the two-platen direct-pressure injection molding machine.
[0052] The specific configuration of the target energy consumption index acquisition module 30 will be described in detail below. The target energy consumption index acquisition module 30 further includes: extracting arbitrary operating data corresponding to any period in the target operating information, wherein the arbitrary operating data includes first operating data corresponding to a first operating indicator; performing random sampling and fitting on the first operating data to obtain a first spline curve; extracting verification operating data corresponding to the verification period in the target operating information, and obtaining a first verification curve of the first operating indicator based on the verification operating data; if the curve similarity between the first verification curve and the first spline curve reaches a predetermined similarity limit, then obtaining a first fitting polynomial of the first spline curve; obtaining first simulated data of the first operating indicator based on the first fitting polynomial, and assembling the target simulated data.
[0053] The specific configuration of the target energy consumption index acquisition module 30 will be described in detail below. The target energy consumption index acquisition module 30 further includes: the first operating index refers to any one of the predetermined operating condition indicators, which at least include the clamping cylinder pressure, the moving cylinder displacement, the flow rate, and the motor power.
[0054] The specific configuration of the dynamic sensing data acquisition module 40 will be described in detail below. The dynamic sensing data acquisition module 40 further includes: matching the first sensing data corresponding to the first operating index in the dynamic sensing data; calculating the first distance value from the first sensing data to the first spline curve; determining whether the first distance value meets the predetermined distance value constraint; if not, issuing an operating abnormality signal, and providing a real-time operating abnormality warning for the two-platen direct-pressure injection molding machine based on the operating abnormality signal.
[0055] The specific configuration of the real-time predicted energy consumption index acquisition module 50 will be described in detail below. The real-time predicted energy consumption index acquisition module 50 further includes: if the first distance value satisfies the predetermined distance value constraint, then comparing the dynamic sensing data with the target simulation data to obtain the real-time predicted energy consumption index; wherein, this includes: sequentially vectorizing the dynamic sensing data and the target simulation data to obtain a dynamic sensing vector and a target simulation vector respectively; comparing the cosine similarity between the dynamic sensing vector and the target simulation vector, and normalizing the results to obtain the real-time predicted energy consumption index.
[0056] The specific configuration of the real-time predicted energy consumption index acquisition module 50 will be described in detail below. The real-time predicted energy consumption index acquisition module 50 further includes: forming a first operating data pair based on the first sensed data and the first simulated data to form the first operating index; constructing multiple sets of operating data pairs based on the first operating data pair, wherein the multiple sets of operating data pairs include arbitrary point pairs; acquiring a predetermined distance grid and filling the predetermined distance grid with arbitrary spatial distances of the arbitrary point pairs to obtain an initial filled grid; acquiring an update filling strategy and updating and filling the initial filled grid according to the update filling strategy to obtain a target filled grid; taking the filling value of the predetermined grid in the target filled grid as the Fraser distance; and verifying the real-time predicted energy consumption index using the normalized Fraser distance.
[0057] The following will describe in detail the specific configuration of the real-time predicted energy consumption index acquisition module 50. The real-time predicted energy consumption index acquisition module 50 further includes: partitioning the initial filling grid according to the update filling strategy to obtain partitioning results; obtaining update blocks from the partitioning results and extracting arbitrary grids from the update blocks, wherein the arbitrary grids correspond to arbitrary initial filling values; constructing an arbitrary reference grid set for the arbitrary grids and filtering the arbitrary reference grid set to obtain an arbitrary maximum grid filling value; replacing the arbitrary initial filling value with the sum of the arbitrary maximum grid filling value and the arbitrary initial filling value as the arbitrary update filling value for the arbitrary grid; and forming the target filling grid based on the arbitrary update filling value.
[0058] The specific configuration of the real-time predicted energy consumption index acquisition module 50 will be described in detail below. The real-time predicted energy consumption index acquisition module 50 further includes: the arbitrary reference grid set includes the left neighbor grid, the upper neighbor grid, and the upper left neighbor grid of the arbitrary grid.
[0059] The injection molding energy consumption dynamic optimization and control device for injection molding machines provided in this embodiment of the invention can execute the injection molding energy consumption dynamic optimization and control method for injection molding machines provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.
[0060] Although this application makes various references to certain modules in the apparatus according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not intended to limit the scope of protection of this invention.
[0061] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.
Claims
1. A method for dynamic optimization and control of injection molding energy consumption for injection molding machines, characterized in that, include: Obtain valid historical operation records of a two-platen direct-pressure injection molding machine, wherein the valid historical operation records include multiple sets of operation data from multiple cycles; Obtain the target stage in the predetermined operation phase, and match the target operation information of the target stage in the multiple sets of operation data; The target operation information is preprocessed and analyzed to obtain target simulation data for the target stage, and the digital twin model is activated to simulate and analyze the target simulation data to obtain the target energy consumption index. Dynamic sensing data is obtained by dynamically and periodically monitoring the mold closing mechanism of the two-platen direct-pressure injection molding machine through distributed sensing devices. The real-time predicted energy consumption index is obtained by comparing the dynamic sensing data with the target simulation data. When the real-time predicted energy consumption index does not meet the target predetermined energy consumption threshold, the energy consumption optimization plan is retrieved to optimize the energy consumption of the two-platen direct-pressure injection molding machine.
2. The method for dynamic optimization and control of injection molding energy consumption for injection molding machines as described in claim 1, characterized in that, The predetermined operating phases include the mold opening and closing phase, the mold locking phase, and the ejection phase of the two-platen direct-pressure injection molding machine.
3. The method for dynamic optimization and control of injection molding energy consumption for injection molding machines as described in claim 1, characterized in that, Preprocessing and analyzing the target operation information to obtain target simulation data for the target stage includes: Extract arbitrary running data corresponding to any period from the target running information, wherein the arbitrary running data includes first running data corresponding to the first running indicator; Randomly sample and fit the first running data to obtain the first identical curve; Extract the verification operation data corresponding to the verification cycle from the target operation information, and obtain the first verification curve of the first operation indicator based on the verification operation data; If the curve similarity between the first verification curve and the first spline curve reaches a predetermined similarity limit, then the first fitting polynomial of the first spline curve is obtained. The first simulated data of the first operating index is obtained based on the first fitting polynomial, and the target simulated data is formed.
4. The method for dynamic optimization and control of injection molding energy consumption for injection molding machines as described in claim 3, characterized in that, The first operating indicator refers to any one of the predetermined operating condition indicators, which at least include the clamping cylinder pressure, the moving cylinder displacement, the flow rate, and the motor power.
5. The method for dynamic optimization and control of injection molding energy consumption for injection molding machines as described in claim 3, characterized in that, After dynamically and periodically monitoring the mold closing mechanism of the two-platen direct-pressure injection molding machine using distributed sensing devices to obtain dynamic sensing data, the process includes: Match the first sensing data corresponding to the first operational indicator in the dynamic sensing data; Calculate the first distance value from the first sensed data to the first spline curve; Determine whether the first distance value meets the predetermined distance value constraint; If the conditions are not met, an operational anomaly signal is issued, and a real-time operational anomaly warning is issued to the two-platen direct-pressure injection molding machine based on the operational anomaly signal.
6. The method for dynamic optimization and control of injection molding energy consumption for injection molding machines as described in claim 5, characterized in that, If the first distance value satisfies the predetermined distance value constraint, then the dynamic sensing data and the target simulation data are compared to obtain the real-time predicted energy consumption index; This includes: The dynamic sensing data and the target simulation data are sequentially vectorized to obtain the dynamic sensing vector and the target simulation vector, respectively. The real-time predicted energy consumption index is obtained by comparing the cosine similarity between the dynamic sensing vector and the target simulation vector and then normalizing the results.
7. The method for dynamic optimization and control of injection molding energy consumption for injection molding machines as described in claim 5, characterized in that, After comparing the cosine similarity between the dynamic sensing vector and the target simulation vector, and normalizing the result to obtain the real-time predicted energy consumption index, the method further includes: Based on the first sensed data and the first simulated data, a first operational data pair is formed to constitute the first operational indicator; Based on the first set of running data, multiple sets of running data pairs are constructed, wherein the multiple sets of running data pairs include any point pairs; Obtain a predetermined distance grid and fill the predetermined distance grid with any spatial distance between any pair of points to obtain an initial filled grid; Obtain the update fill strategy, and update and fill the initial fill grid according to the update fill strategy to obtain the target fill grid; The fill value of the predetermined grid in the target filled grid is taken as the Fraser distance; The real-time predicted energy consumption index is verified by the normalized Fréchet distance.
8. The method for dynamic optimization and control of injection molding energy consumption for injection molding machines as described in claim 7, characterized in that, Obtaining an update fill strategy and updating and filling the initial fill grid according to the update fill strategy to obtain a target fill grid includes: The initial filling grid is partitioned according to the update filling strategy to obtain the partitioning results; Obtain the updated block from the partitioning result, and extract any grid from the updated block, wherein the arbitrary grid corresponds to any initial fill value; Construct an arbitrary reference mesh set for the arbitrary mesh, and filter the arbitrary reference mesh set to obtain an arbitrary maximum mesh fill value; The sum of the arbitrary maximum grid fill value and the arbitrary initial fill value is used as the arbitrary update fill value for the arbitrary grid. The target filled grid is formed based on the arbitrary updated fill value.
9. The method for dynamic optimization and control of injection molding energy consumption for injection molding machines according to claim 8, characterized in that, The arbitrary reference grid set includes the left neighbor grid, the top neighbor grid, and the top-left neighbor grid of the arbitrary grid.
10. A dynamic optimization and control device for injection molding energy consumption of injection molding machines, characterized in that, The device is used to implement the dynamic optimization and control method for injection molding energy consumption of injection molding machines according to any one of claims 1 to 9, and the device includes: The effective historical operation record acquisition module is used to acquire the effective historical operation records of a two-platen direct-pressure injection molding machine, wherein the effective historical operation records include multiple sets of operation data for multiple cycles; The target operation information matching module is used to obtain the target stage in the predetermined operation stage and match the target operation information of the target stage in the multiple sets of operation data; The target energy consumption index acquisition module is used to preprocess and analyze the target operation information to obtain the target simulation data of the target stage, and activate the digital twin model to simulate and analyze the target simulation data to obtain the target energy consumption index. The dynamic sensing data acquisition module is used to dynamically and periodically monitor the mold closing mechanism of the two-platen direct-pressure injection molding machine through distributed sensing devices to obtain dynamic sensing data. A real-time predicted energy consumption index acquisition module is used to compare the dynamic sensing data with the target simulation data to obtain a real-time predicted energy consumption index. The energy consumption optimization module is used to retrieve the energy consumption optimization plan to optimize the energy consumption of the two-platen direct-pressure injection molding machine when the real-time predicted energy consumption index does not meet the target predetermined energy consumption threshold.