Sectional conveying platform for assembling plastic molds

Abstract

The present application relates to the field of mold conveying, in particular to a segmented conveying platform for plastic mold assembly. The platform is composed of multiple conveying platform segments arranged in process sequence, mold trays and a control system. Each platform segment is provided with independent drive and head and tail detection, and is used as a work station segment or a buffer segment that can be independently started, stopped and adjusted in speed. The control system collects conveying speed, start-stop acceleration and deceleration, mold arrival rhythm and buffer occupation state, and provides assembly operation after the mold tray is stopped and stabilized in the work station segment. The finished signal triggers the judgment of downstream work stations and buffer conditions, and releases the tray according to the conditions. Based on the impact posture related parameters and rhythm-buffer related parameters, the comprehensive index is calculated, and the speed setting value, acceleration and deceleration parameters and release rhythm of each segment are jointly generated to realize the safe, smooth and work station rhythm advancing assembly control of heavy molds on the segmented conveying platform.

Description

Technical Field

[0001] This invention relates to the field of mold conveying, specifically a segmented conveying platform for assembling plastic molds. Background Technology

[0002] The assembly and maintenance of plastic molds typically involves multiple processes, including post-machining inspection of the mold base, mold assembly, accessory installation, guide pillar and guide assembly, connection of hot runners and cooling water systems, and pre-trial inspection and maintenance. Molds are large and heavy, and their center of gravity often changes depending on the opening and closing state and the installation of accessories. Existing factories generally use overhead cranes, forklifts, and ground-rail trolleys for point-to-point transport, moving molds between the mold repair area, assembly area, and injection molding machine. In some cases, parallel rails and adjustable-height trolleys are installed on one side of the injection molding machine to facilitate the connection between the mold and the clamping device. The common characteristics of these solutions are: logistics are mainly based on single-point transport and fixed-station operations. Heavy molds rely heavily on manual command and operational experience during transport, and the cycle time is significantly affected by lifting, repositioning, and fluctuations in manual operation time, making it difficult to form a stable and controllable cycle chain.

[0003] In the fields of general logistics and final assembly, technologies such as pallet conveyors and segmented chain conveyors are relatively mature. By dividing the conveyor line into several standard sections and combining them with simple section start / end detection, basic workstation segmentation, stacking buffering, and cycle time control can be achieved. These systems are widely used in multi-station assembly and testing of light-load materials in home appliances, automotive parts, and other industries. However, publicly available information shows that their control logic is mostly limited to the level of "start when there is a pallet, stop when the front is occupied." There is a lack of systematic quantitative and closed-loop control of factors such as the start / stop acceleration / deceleration parameters of each section, the impact magnitude during the section switching process, and the stability of the pallet and the load posture. For heavy, high-center-of-gravity, and high-inertia plastic molds, simply applying these segmented conveyor lines can easily lead to large start / stop impacts and posture swaying during section switching, increasing the risk of mold slippage, loosening of clamps, or even overturning. On the other hand, it is difficult to balance operational safety and assembly cycle time, resulting in the continued reliance on overhead cranes / forklifts in practice, with segmented conveyors only being used in limited areas.

[0004] Regarding cycle time and buffer management, existing segmented conveyor systems typically rely on the length of the accumulation zone and simple congestion logic to roughly absorb cycle time differences, lacking a refined cycle time and buffer coordination mechanism for heavy-duty assembly scenarios. Actual workstation operation times often fluctuate, with long queues easily forming at slower workstations and prolonged idle time at faster workstations. On-site adjustments often rely on temporary adjustments by the team leader and manual intervention by experienced operators. Attempting to use a continuously driven conveyor line to move the mold synchronously with the line results in frequent start-stop cycles due to inconsistent operation times across processes and high operational safety requirements, further reducing cycle time stability and output efficiency. Publicly available literature lacks a systematic solution for heavy-duty mold assembly scenarios that quantifies workstation cycle time deviations, buffer occupancy, queue lengths, etc., into calculable indicators and dynamically adjusts release rhythm and buffer strategies accordingly.

[0005] Furthermore, from an overall control perspective, existing technologies often handle safety-related physical quantities and cycle time and buffer status in a fragmented manner. On the one hand, physical quantities closely related to safety, such as start-stop impacts, mold posture changes, and variable acceleration, are often only roughly considered during equipment selection and experience-based debugging, lacking real-time monitoring and parameter self-adaptation capabilities after deployment. On the other hand, cycle time and buffer control are mostly based on simple timing logic and manual experience, failing to form a unified mathematical model that balances safety and efficiency. Existing segmented conveying systems typically do not introduce similar comprehensive indicators to jointly optimize key parameters such as speed settings, acceleration / deceleration times, and release intervals for each segment, making it difficult to find a balance point that can be automatically adjusted according to changing operating conditions between reducing impact and posture risks and improving cycle time and buffer utilization. These technical deficiencies are particularly prominent in applications such as plastic molds, which are characterized by high weight, high center of gravity, numerous processes, and large cycle time fluctuations, directly hindering the upgrade of mold assembly lines from traditional point-to-point transfer modes to safe, stable, and flexible segmented conveying modes. Summary of the Invention

[0006] The purpose of this invention is to provide a segmented conveyor platform for assembling plastic molds, so as to solve the technical problems mentioned in the background art.

[0007] Based on the above ideas, the present invention provides the following technical solution: A segmented conveyor platform for assembling plastic molds includes the following steps: The system includes multiple conveyor platform sections arranged sequentially along the mold assembly process, a tray for supporting the plastic mold, and a control system electrically connected to each of the conveyor platform sections. The control system includes at least a programmable logic controller (PLC), a drive control unit, and a human-machine interface. The control system is characterized by being configured as follows: S1. Divide each of the conveying platform segments into workstation segments and buffer segments. Each conveying platform segment is equipped with a drive motor and segment head and end position detection sensors. The segment head and end position detection signals are collected to the programmable controller so that each conveying platform segment can be used as a control unit that can be independently started, stopped and speed adjusted. S2. When the mold pallet is detected by the section head position detection sensor of a certain work station section to enter the work station section, the work station section is controlled to run to the preset stopping position according to the preset acceleration curve, and the work station section is stopped after the mold pallet reaches the preset stopping position, so that the mold pallet is stationary in the work station section, so that the operator can complete the corresponding mold assembly process in the stationary state of the mold pallet. S3. During the operation of the mold pallet on each of the conveying platform sections, the conveying speed of each workstation section, the acceleration and deceleration during the start and stop process, the attitude change related to the height of the mold center of gravity, the arrival time of the mold pallet entering the workstation section, and the occupancy status of the buffer sections before and after the workstation section are collected. The conveying speed and the arrival time are used as common time parameters, the attitude change and the impact related to the start and stop process are used as impact attitude parameters, and the occupancy rate of the buffer section and the time deviation of the workstation section are used as buffer time parameters. S4. Based on the shared cycle time parameters and the impact posture parameters, calculate the comprehensive impact posture index at the entrance of each workstation section to characterize the impact of inter-section switching and the stability of the mold posture. Then, dynamically adjust the start-stop acceleration and deceleration curves, target conveying speed and maximum allowable number of start-stop cycles of the corresponding workstation section according to the comprehensive impact posture index, so that the impact of the mold pallet during the inter-section transfer process is controlled within a preset range and the instability of the mold posture on the pallet is suppressed. S5. Based on the shared cycle time parameters and the buffer cycle time parameters, calculate the cycle time buffer comprehensive index, which characterizes the degree of cycle time matching and the degree of buffer saturation, in the buffer areas arranged before and after the key work station section. According to the cycle time buffer comprehensive index, dynamically adjust the release rhythm of the upstream work station section, the number of buffer sections allowed to occupy, and the target cycle time of the downstream work station section to buffer the cycle time difference between fast and slow processes and avoid serious accumulation or idleness before and after the key work station. S6. The comprehensive impact attitude index and the comprehensive cycle buffer index calculated for each workstation segment and buffer segment are used as input variables for joint optimization. A joint optimization objective function is constructed with the goal of reducing inter-segment impact and attitude risks, improving assembly cycle stability and buffer utilization. Based on the optimization results of the joint optimization objective function, speed setpoints, acceleration / deceleration curve parameters and release rhythm parameters for each of the conveying platform segments are generated and sent to the corresponding drive control unit to realize mold pallet conveying and assembly control according to the workstation rhythm on the entire segmented conveying platform.

[0008] By clearly dividing the conveyor platform into workstation sections and buffer sections, and treating each section as an independent start / stop and speed control unit, the original rhythm of the heavy mold conveyor, which followed the entire line, is transformed into a rhythm mode that advances according to the workstation's pace. Those skilled in the art can directly equip each section with a motor, frequency converter / servo drive, and sensors at the beginning and end of the section. Then, a PLC can collect signals uniformly, enabling rhythm control that allows work to resume only after the workstation has stopped and to release the mold only after the work is completed. In this way, the heavy mold remains completely stationary for most of the assembly time, allowing operators to perform mold assembly, accessory installation, and connection of cooling water or hot runners without relative displacement, significantly improving safety compared to continuous drive lines. Simultaneously, by setting buffer sections before and after key processes, and coordinating with the upstream release rhythm and downstream receiving conditions, the rhythm differences between fast and slow processes can be absorbed within the buffer area, avoiding long queues at slow workstations and waiting times at fast workstations. This significantly improves the overall line capacity utilization and line balance. In addition, the control system quantifies speed, acceleration / deceleration, cycle time, buffer status, etc. into parameters, and then distributes them to each segment through a joint optimization objective function to achieve comprehensive optimization of cycle time and safety. This can be achieved by those skilled in the art using a conventional PLC / industrial Ethernet. Preferably, the comprehensive impact posture index is used to characterize the start-stop impact and mold posture stability of a single workstation entry point within a conveying cycle. Specifically, the control system is configured to calculate the comprehensive impact posture index as follows: Accelerometers installed on the pallet bearing surface or workstation frame collect the acceleration time series of the mold pallet during the start-up and stop process, and extract the impact peak along the conveying direction during the start-up and stop process and the attitude decay characteristics of the mold pallet in the lateral swaying direction after the start-up and stop. By combining the preset mold weight and mold center of gravity height parameters, the impact peak and attitude attenuation characteristics are normalized to obtain the inter-segment start-stop impact level and attitude stability level of the workstation under the current operating parameters. Collect the target conveying speed parameters of the current workstation section and the arrival cycle parameters of the corresponding assembly workstation. Use the target conveying speed parameters and the arrival cycle parameters as common cycle parameters, and input them together with the inter-section start-stop impact level and attitude stability level into the impact attitude comprehensive index calculation model to obtain the impact attitude comprehensive index used to constrain the start-stop control of the workstation section. When the comprehensive impact posture index exceeds the preset upper limit, the acceleration and deceleration time of the workstation section is automatically extended, the target conveying speed is reduced, the number of consecutive start-stop cycles is limited, or the proportion of uniform speed sections in the transition distance between sections is increased, so as to reduce the risk of instantaneous impact and posture instability at the section switching point.

[0009] The start-stop impact and mold attitude stability are quantified into calculable, closed-loop control indicators. This section uses an accelerometer to collect the acceleration curves during the start-stop process, extracts the impact peak along the conveying direction and the attenuation characteristics of lateral sway, and normalizes them by combining mold weight and center of gravity height, allowing molds of different specifications and weights to be compared in impact intensity and attitude stability on a uniform scale. In this way, the problem of whether or not there is sway, which was originally only a matter of experience, becomes a control target with numerical standards. Combined with the target conveying speed and arrival cycle parameters of the workstation, the first comprehensive indicator can reflect whether the start-stop is too abrupt, and whether the mold is prone to displacement or overturning under the current speed / cycle configuration. When this indicator exceeds the limit, the system automatically extends the acceleration and deceleration time, reduces the speed, and limits the number of consecutive start-stop cycles, essentially establishing an adaptive deceleration control based on actual impact and attitude feedback for each section of the conveying platform. This not only reduces the instantaneous impact during inter-section switching, avoiding the risks of mold micro-movement on the pallet, fixture loosening, or even mold collapse, but also reduces the probability of tooling, fixtures, and measuring equipment mounted on the mold falling, significantly improving operational safety and equipment lifespan.

[0010] Preferably, the calculation equation for the comprehensive impact attitude index X is: ; Where σ is the absolute value of the maximum deceleration or acceleration of the mold pallet during the start-up and stop process of the workstation section, σ lim To determine the deceleration limit based on the allowable impact of the mold system, θ rms Let θ be the root mean square value of the tilt angle of the mold pallet around the lateral axis during start-up, shutdown, and inter-segment switching. lim J represents the maximum allowable tilt angle limit determined based on the mold's center of gravity height and clamping allowance, where J is the peak value of the variable acceleration of the mold pallet during start-up and shutdown. lim The variable acceleration limit is determined based on the pallet stiffness and mold mounting stiffness, where v is the currently set target conveying speed for this workstation section. ref q represents the reference conveying speed under no-load or light-load conditions, and q is the target arrival cycle time for the corresponding assembly station. ref The reference cycle time is given for the design process. α1, α2, α3, and α4 are dimensionless weighting coefficients that are pre-set according to the sensitivity of different workstations to impact, posture, and cycle time. The control system adjusts the target conveying speed v, the start / stop acceleration / deceleration time, and the shape of the acceleration / deceleration curve to keep the calculated X below a preset impact posture safety threshold.

[0011] The peak impact value σ and the root mean square angle of inclination θ rmsThe three physical quantities strongly related to human and equipment safety—variable acceleration J, variable acceleration J, and variable acceleration J—are combined into a single dimensionless index. Simultaneously, conveying speed v and cycle time q are introduced as shared parameters, reflecting the physical principle that the more aggressive the speed / cycle time, the higher the impact posture index. This definition of X has the following advantages: each term is the square of the measured / permissible quantity; a ratio greater than 1 indicates exceeding the limit, directly reflecting the safety margin; through the weighting of α1, α2, and α3, impact reduction or posture stabilization can be targeted for different workstations; for example, α2 can be increased for workstations with a high center of gravity; through α4(v / v... ref )(q ref The / q) term explicitly superimposes the risks of high speed and tight cycle time onto X, making the system automatically more conservative under high-speed, high-cycle conditions. The X value output by the formula is directly used by the control system as a constraint: when X is less than a threshold, higher speed and shorter acceleration / deceleration times are allowed to increase cycle time; when X approaches or exceeds the threshold, the drive parameters are automatically adjusted to lengthen acceleration / deceleration times, reduce speed, and decrease continuous start-stop cycles, so that speed cycle time and impact attitude safety are balanced under the same index. Those skilled in the art only need to collect acceleration and attitude data, set limit values ​​and weights, and then implement real-time calculation and closed-loop control in a PLC or industrial PC based on this formula.

[0012] Preferably, the cycle time buffer comprehensive index is used to characterize the cycle time matching degree and buffer occupancy status of the buffer areas before and after the workstation section. The control system is specifically configured to calculate the cycle time buffer comprehensive index as follows: Statistically analyze the actual completion cycle time of each work station segment within the preset time window, and calculate the cycle time deviation between the actual cycle time of each work station segment and the corresponding target cycle time. Collect occupancy status signals on the buffer sections before and after the critical workstation section, calculate the real-time occupancy quantity and occupancy rate of each buffer section within the time window, and obtain the maximum allowable occupancy rate of the buffer section based on the total number of buffer sections; The target conveying speed parameter and the assembly station target beat parameter are collected and used in the calculation of the impact attitude comprehensive index. The target conveying speed parameter and the assembly station target beat parameter are used as common beat parameters and are input into the beat buffer comprehensive index calculation model together with the beat deviation and the buffer occupancy rate to obtain the beat buffer comprehensive index used to guide the upstream release rhythm and buffer scheduling strategy. When the overall index of the cycle time buffer exceeds the preset upper limit and the corresponding buffer segment occupancy rate is close to saturation, the release frequency of the upstream workstation segment is reduced, the allowed waiting time of the upstream workstation segment is extended, and if necessary, cycle time limiting is implemented on several upstream workstation segments; when the overall index of the cycle time buffer is lower than the preset lower limit and the corresponding buffer segment is idle for a long time, the release frequency of the upstream workstation segment is increased or priority is given to releasing to the buffer segment arranged in front of the critical workstation segment, so as to improve the start-up rate of the critical workstation.

[0013] By quantifying cycle time deviation and buffer occupancy, queue length and buffer saturation, which were previously managed based on experience, are transformed into automatically adjustable control variables. The cycle time deviation ΔT is obtained by statistically analyzing the actual completion cycle time of each workstation within a specific time window, and this deviation is then combined with the buffer occupancy rate ρ and the target occupancy rate ρ. opt Queue size L and maximum queue size L max Based on the data, the system can make real-time judgments: Which workstations are consistently slower than the target cycle time, creating cycle time bottlenecks? Which buffers are nearing saturation, becoming accumulation points and posing a risk of congestion? Which buffers are idle for extended periods, wasting line resources?

[0014] With this indicator model, the control system can automatically reduce the upstream release frequency and extend the waiting time to protect the saturated buffer. It can also increase the upstream release frequency when the buffer is consistently empty, ensuring that critical workstations always have work to do. Compared to the simple logic of releasing workstations as soon as they are available and stopping when the buffer is full, this invention can adjust the release rhythm at a finer granularity, preventing the cycle time from being locked by a single slow workstation, and instead allowing upstream and downstream to work collaboratively through the buffer. For multi-variety, small-batch scenarios where cycle time fluctuates significantly, the second comprehensive indicator can be used to dynamically adjust the target cycle time of workstations and the buffering strategy, making the entire production line more robust to process changes.

[0015] Preferably, the calculation equation for the comprehensive index Y of the clock cycle buffer is: ; Where ΔT is the average cycle time deviation of the workstation segment within the preset time window, that is, the absolute value of the time difference between the actual completion cycle time and the target cycle time q, q is the target arrival cycle time of the corresponding assembly station, and ρ is the average buffer occupancy rate of the corresponding buffer area within the time window. opt To set a target occupancy rate based on the optimal cache utilization allowed for this region, L is the average number of queued trays in this cache region within the time window. max The maximum allowable queue size is determined based on the number of physical segments in the buffer and the safety distance, where v is the target delivery speed. ref With q ref These are reference values ​​for the target conveying speed and the target cycle time, respectively. β1, β2, β3, and β4 are dimensionless weighting coefficients preset according to the process requirements for cycle time stability, buffer balance, and line utilization. The control system adjusts the release interval of the upstream workstation, the allowed number of buffer sections, and the target cycle time of the critical workstation to constrain the calculated Y within the preset cycle time buffer control range.

[0016] Key logistical characteristics such as cycle time deviation, buffer occupancy deviation ρ, and queue length are normalized and superimposed onto a dimensionless index Y, while allowing vvv and qqq to be shared with X, achieving "the entanglement of cycle time / speed parameters between impact safety and buffer utilization." The specific effects of this formula include: (ΔT / q) 2 This allows the cycle time deviation to be measured on a relative scale, which is not affected by the absolute value of the cycle time at different workstations, making it more suitable for multi-workstation comparison. (ρ ρ opt ) / ρ opt This ensures that the cache is neither overly saturated nor overly idle, guiding the system to operate the cache in a comfortable range. (L / L max ) 2 This amplifies the risk of queuing lengths approaching their limits, causing the system to rapidly reduce the release rate when it approaches its physical limits, thus preventing overflow and on-site congestion. The last term still uses v and q to reflect the risk of running too fast or having too tight a pace in Y, forming a common constraint direction with X.

[0017] The calculated value of Y directly drives the release and buffering strategies: when Y is too high, mitigation is achieved by reducing the upstream release frequency, decreasing the number of pallets allowed to enter simultaneously, and relaxing the cycle time of critical workstations; when Y is low, the system can appropriately increase the release frequency and shorten the waiting time to improve the overall line throughput. Through this metric, the system maintains the utilization rate of the buffer segment within a reasonable range while ensuring no line jams or prolonged idle runs. This improves the line load rate while avoiding extreme congestion, enhancing the overall production line smoothness and adaptability to process changes.

[0018] Preferably, when the control system performs joint optimization using the impact attitude comprehensive index and the cycle buffer comprehensive index, it is specifically configured as follows: Within each preset optimization cycle, the comprehensive impact posture index and the comprehensive cycle buffer index corresponding to all workstations and buffer sections are collected to form an index set for the entire segmented conveying platform. Through the human-machine interface or pre-configured process recipes, different impact posture weights and cycle time buffer weights are set for different areas or workstations of different process importance, so as to prioritize cycle time stability in critical workstations and prioritize posture safety in high-risk workstations. Based on the set of indicators and their corresponding weights, the target conveying speed, acceleration and deceleration time, and upstream release interval time of the workstation section are taken as control variables to be optimized. Constraints such as the upper limit of the maximum conveying speed, the lower limit of the minimum acceleration and deceleration time, and the number of available buffer sections in each area are applied to each of the control variables. Based on the joint optimization results, the speed setpoints, acceleration / deceleration curve parameters, and release rhythm parameters of each of the aforementioned conveyor platform segments are updated. In the next optimization cycle, the comprehensive impact posture index and the comprehensive cycle buffer index are recalculated based on the new operating data to achieve rolling optimization control of the entire segmented conveyor platform.

[0019] By extending the X and Y categories of indicators to the entire production line optimization dimension, and introducing optimization cycles, weight configurations, and constraints, a workable rolling optimization framework is formed. Those skilled in the art can, based on the on-site process, mark certain workstations as critical, hazardous, or ordinary workstations, and then configure different impact-attitude weights and cycle time-buffer weights for these areas: for example, in areas adjacent to manual operation zones or high-center-of-gravity mold areas, the X weight can be increased to prioritize safety; in bottleneck workstations or cycle time-sensitive workstations, the Y weight can be increased to prioritize cycle time.

[0020] Based on this, the system treats the target speed, acceleration / deceleration time, and release interval of each workstation as variables to be optimized, imposing constraints such as speed upper limits, acceleration / deceleration time lower limits, and the number of available buffer segments. An optimization algorithm solves for a set of parameters to minimize the overall Z value. The rolling optimization mechanism ensures that when the process or order structure changes (e.g., a temporary addition of a process to a workstation, or an increase in the batch size of a certain mold), the control parameters can be adaptively updated, rather than remaining at their initial setpoints indefinitely. In this way, the entire segmented conveyor platform is no longer a static configuration, but an adaptive line capable of adjusting its cycle time and safety margin based on real-time data, significantly improving stability, flexibility, and energy efficiency during long-term operation.

[0021] Preferably, the objective value Z of the joint optimization objective function is defined as: ; Where X is the calculated comprehensive impact attitude index, and Y is the calculated comprehensive clock cycle buffer index. ref Y is the impact posture reference value determined based on experience at the mold assembly site or simulation results. ref The reference value for the cycle buffer is determined based on the requirements of the overall cycle balance and buffer utilization. γ1, γ2, and γ3 are joint optimization weight coefficients that can be configured by the user according to the process scenario. Among them, γ1 and γ2 are non-negative, and γ3 is used to adjust the interaction between the impact posture and the cycle buffer. The control system is designed to satisfy X≤X max Y≤Y maxUnder the safety constraints, minimizing Z is the optimization objective. The target conveying speed, acceleration / deceleration time, and release interval parameters for each conveying platform segment are generated using a rolling optimization method. In subsequent optimization cycles, Z is recalculated based on the updated impact attitude comprehensive index and cycle buffer comprehensive index, thereby achieving dynamic joint optimization control of the entire segmented conveying platform.

[0022] Unify the two types of indicators, X and Y, into a single overall objective function Z, and introduce a reference X. ref ,Y ref The interaction term γ3 makes the optimization not just a simple linear weighting, but a nonlinear characteristic with safe beat interaction. Its beneficial effect is reflected in: through X / X... ref and Y / Y ref The normalized square of Z can simultaneously reflect how much the current impact attitude level deviates from the target level and how much the current beat / buffer level deviates from the target level; the cross term (X / X ref (Y / Y) ref ) is used to characterize the combined effect when both safety and cycle time indicators deviate significantly, so that the control system can more actively reduce speed and widen cycle time when both are not ideal, rather than adjusting only one side; the configuration of γ1, γ2, and γ3 allows users to choose between prioritizing safety or prioritizing cycle time according to different production strategies, and make dynamic adjustments at different stages.

[0023] Minimizing Z is directly used to drive the optimization of the entire production line parameters: In each optimization cycle, the control system uses the real-time calculation results of X and Y as input to evaluate the size of Z, and then reduces Z by adjusting variables such as speed, acceleration / deceleration, and release interval of each work station.

[0024] The technical solution of the present invention may include the following beneficial effects: This invention constructs a segmented conveyor platform for plastic mold assembly lines, consisting of workstation sections and buffer sections. Each platform section has independent start / stop and speed adjustment capabilities, upgrading the traditional point-to-point transfer mode relying on overhead cranes and forklifts to a streamlined conveyor mode that progresses according to the rhythm of each workstation. Mold pallets complete complex operations such as mold assembly, accessory installation, and hot / cold water circuit connection in a stationary state within the workstation sections, smoothly transferring only short distances between sections under controlled conditions. This effectively reduces the risk of swaying, displacement, or even overturning of heavy molds during movement, improving the human and machine safety environment on-site. Simultaneously, by setting adjustable buffer sections before and after key processes, the entire line can actively absorb the rhythm differences between fast and slow processes, avoiding large-scale waiting and localized congestion caused by a single bottleneck workstation, achieving a balance in mold assembly rhythm and an overall improvement in capacity utilization. Building upon this foundation, the present invention further incorporates a first comprehensive index X and a second comprehensive index Y, unifying key physical quantities and logistics characteristics such as start-stop impact, posture stability, cycle time deviation, and buffer occupancy rate into a calculable and constrainable mathematical model. The first comprehensive index X, with physical quantities such as peak impact, root mean square tilt angle, and variable acceleration as its core, combined with conveyor speed and target cycle time, achieves quantitative evaluation and adaptive amplitude limiting control of inter-segment switching impact and mold posture. The second comprehensive index Y characterizes the queuing and release status before and after the workstation from dimensions such as cycle time deviation, buffer deviation, and queue length, enabling dynamic adjustment of buffer utilization and cycle time matching. Through these indices, safety control and cycle time coordination, which originally relied on experience and manual observation, are transformed into a closed-loop control process that can be calculated and corrected in real time in a PLC or industrial PC. This allows those skilled in the art to achieve refined management of the conveyor platform's operating status with explicit parameters and formulas. Furthermore, this invention utilizes a joint optimization objective function Z to normalize and nonlinearly combine the two types of indicators X and Y, and combines this with the weight configuration and safety constraints of each workstation to implement rolling optimization control across the entire production line. By periodically solving and updating the target speed, acceleration / deceleration time, and release interval of each workstation segment, the system can automatically find a balance between impact / posture safety and cycle time / buffer efficiency for different mold specifications, different process cycles, and different order structures, thereby maintaining high cycle time stability and line flexibility during long-term operation. Compared with traditional control methods that rely on experience-based tuning and are difficult to dynamically adjust, this invention not only significantly reduces safety risks and maintenance costs in the heavy mold assembly process, but also makes it easier for production lines to achieve energy saving, consumption reduction, and multi-product switching through parameter optimization. It has good promotional value for the automation and intelligent transformation of plastic mold assembly workshops. Attached Figure Description

[0025] Figure 1 This is an operation flowchart of a segmented conveyor platform for assembling plastic molds and its operating system according to the present invention. Detailed Implementation

[0026] Example 1 like Figure 1 , A segmented conveyor platform for assembling plastic molds includes the following steps: The system includes multiple conveyor platform sections arranged sequentially along the mold assembly process, a tray for supporting the plastic mold, and a control system electrically connected to each of the conveyor platform sections. The control system includes at least a programmable logic controller (PLC), a drive control unit, and a human-machine interface. The system is characterized in that it is configured as follows: S1. Divide each of the conveying platform segments into workstation segments and buffer segments. Each conveying platform segment is equipped with a drive motor and segment head and end position detection sensors. The segment head and end position detection signals are collected to the programmable controller so that each conveying platform segment can be used as a control unit that can be independently started, stopped and speed adjusted. S2. When the mold pallet is detected by the section head position detection sensor of a certain work station section to enter the work station section, the work station section is controlled to run to the preset stopping position according to the preset acceleration curve, and the work station section is stopped after the mold pallet reaches the preset stopping position, so that the mold pallet is stationary in the work station section, so that the operator can complete the corresponding mold assembly process in the stationary state of the mold pallet. S3. During the operation of the mold pallet on each of the conveying platform sections, the conveying speed of each workstation section, the acceleration and deceleration during the start and stop process, the attitude change related to the height of the mold center of gravity, the arrival time of the mold pallet entering the workstation section, and the occupancy status of the buffer sections before and after the workstation section are collected. The conveying speed and the arrival time are used as common time parameters, the attitude change and the impact related to the start and stop process are used as impact attitude parameters, and the occupancy rate of the buffer section and the time deviation of the workstation section are used as buffer time parameters. S4. Based on the shared cycle time parameters and the impact posture parameters, calculate the comprehensive impact posture index at the entrance of each workstation section to characterize the impact of inter-section switching and the stability of the mold posture. Then, dynamically adjust the start-stop acceleration and deceleration curves, target conveying speed and maximum allowable number of start-stop cycles of the corresponding workstation section according to the comprehensive impact posture index, so that the impact of the mold pallet during the inter-section transfer process is controlled within a preset range and the instability of the mold posture on the pallet is suppressed. S5. Based on the shared cycle time parameters and the buffer cycle time parameters, calculate the cycle time buffer comprehensive index, which characterizes the degree of cycle time matching and the degree of buffer saturation, in the buffer areas arranged before and after the key work station section. According to the cycle time buffer comprehensive index, dynamically adjust the release rhythm of the upstream work station section, the number of buffer sections allowed to occupy, and the target cycle time of the downstream work station section to buffer the cycle time difference between fast and slow processes and avoid serious accumulation or idleness before and after the key work station. S6. The comprehensive impact attitude index and the comprehensive cycle buffer index calculated for each workstation segment and buffer segment are used as input variables for joint optimization. A joint optimization objective function is constructed with the goal of reducing inter-segment impact and attitude risks, improving assembly cycle stability and buffer utilization. Based on the optimization results of the joint optimization objective function, speed setpoints, acceleration / deceleration curve parameters and release rhythm parameters for each of the conveying platform segments are generated and sent to the corresponding drive control unit to realize mold pallet conveying and assembly control according to the workstation rhythm on the entire segmented conveying platform.

[0027] By clearly dividing the conveyor platform into workstation sections and buffer sections, and treating each section as an independent start / stop and speed control unit, the original rhythm of the heavy mold conveyor, which followed the entire line, is transformed into a rhythm mode that advances according to the workstation's pace. Those skilled in the art can directly equip each section with a motor, frequency converter / servo drive, and sensors at the beginning and end of the section. Then, a PLC can collect signals uniformly, enabling rhythm control that allows work to resume only after the workstation has stopped and to release the mold only after the work is completed. In this way, the heavy mold remains completely stationary for most of the assembly time, allowing operators to perform mold assembly, accessory installation, and connection of cooling water or hot runners without relative displacement, significantly improving safety compared to continuous drive lines. Simultaneously, by setting buffer sections before and after key processes, and coordinating with the upstream release rhythm and downstream receiving conditions, the rhythm differences between fast and slow processes can be absorbed within the buffer area, avoiding long queues at slow workstations and waiting times at fast workstations. This significantly improves the overall line capacity utilization and line balance. In addition, the control system quantifies speed, acceleration / deceleration, cycle time, buffer status, etc. into parameters, and then distributes them to each segment through a joint optimization objective function to achieve comprehensive optimization of cycle time and safety. This can be achieved by those skilled in the art using a conventional PLC / industrial Ethernet. In one specific embodiment, the conveyor platform consists of 12 heavy-duty chain conveyor units connected in series, with 8 sections serving as workstations and 4 sections as buffer sections. Each section is approximately 2.5 meters long and is equipped with a steel structure pallet to support the mold. The pallet has four height-adjustable support pads to accommodate plastic molds of different sizes. A geared motor and chain drive mechanism are installed below each conveyor unit, with the motor controlled by an independent frequency converter or servo drive. A pair of photoelectric sensors are installed at the entrance of each section to detect the pallet entering, and a pair of photoelectric sensors are installed at the exit to detect the pallet reaching the end of the section. The entire line is controlled collaboratively by a programmable logic controller (PLC) and an industrial PC. The PLC is responsible for acquiring signals at the beginning / end of each section and for starting and stopping the drive, while the industrial PC is responsible for calculating the cycle time and overall performance indicators. A human-machine interface (HMI) is installed on one side of the line, displaying the status of each section, workstation number, buffer occupancy, and providing operations such as workstation completion confirmation and manual release. After the system is powered on, the programmable controller (PLC) first reads the status of all sensors to identify whether each segment has a pallet. It then marks each segment as a workstation segment or a buffer segment according to a preset parameter table, and associates each segment with its corresponding assembly station, target cycle time, and access permissions. When the current segment's exit detects a pallet leaving and the downstream segment's entrance is empty, the downstream segment starts at a set speed, transports the pallet to near the end of the segment, and stops, achieving a stationary stop. Operators can then complete tasks such as mold assembly, accessory installation, and water connection on the stationary pallet. After the workstation is completed, the operator sends a completion confirmation signal via the nearest button box or barcode scanner. The PLC determines whether to release the pallet based on this signal and the idle status of the downstream workstation and buffer segments. If the downstream segment is ready to receive the pallet, the current segment starts and completes the pallet transfer with the downstream segment. If the downstream segment is full, the current workstation remains locked, and the pallet is temporarily held in the workstation or the preceding buffer segment. For critical processes, safety railings and pull-rope emergency stop switches are added to both sides. When the safety devices activate, several adjacent segments automatically stop and are prohibited from moving, ensuring personnel safety. With this structure and control arrangement, those skilled in the art can realize segmented delivery according to the rhythm of the workstation based on independently controllable workstation sections and buffer sections, and provide the necessary detection and control interfaces for subsequent comprehensive index calculation.

[0028] Specifically, the comprehensive impact posture index is used to characterize the start-stop impact and mold posture stability of a single workstation entry point within a conveying cycle. The control system is specifically configured to calculate the comprehensive impact posture index as follows: Accelerometers installed on the pallet bearing surface or workstation frame collect the acceleration time series of the mold pallet during the start-up and stop process, and extract the impact peak along the conveying direction during the start-up and stop process and the attitude decay characteristics of the mold pallet in the lateral swaying direction after the start-up and stop. By combining the preset mold weight and mold center of gravity height parameters, the impact peak and attitude attenuation characteristics are normalized to obtain the inter-segment start-stop impact level and attitude stability level of the workstation under the current operating parameters. Collect the target conveying speed parameters of the current workstation section and the arrival cycle parameters of the corresponding assembly workstation. Use the target conveying speed parameters and the arrival cycle parameters as common cycle parameters, and input them together with the inter-section start-stop impact level and attitude stability level into the impact attitude comprehensive index calculation model to obtain the impact attitude comprehensive index used to constrain the start-stop control of the workstation section. When the comprehensive impact posture index exceeds the preset upper limit, the acceleration and deceleration time of the workstation section is automatically extended, the target conveying speed is reduced, the number of consecutive start-stop cycles is limited, or the proportion of uniform speed sections in the transition distance between sections is increased, so as to reduce the risk of instantaneous impact and posture instability at the section switching point.

[0029] The start-stop impact and mold attitude stability are quantified into calculable, closed-loop control indicators. This section uses an accelerometer to collect the acceleration curves during the start-stop process, extracts the impact peak along the conveying direction and the attenuation characteristics of lateral sway, and normalizes them by combining mold weight and center of gravity height, allowing molds of different specifications and weights to be compared in impact intensity and attitude stability on a uniform scale. In this way, the problem of whether or not there is sway, which was originally only a matter of experience, becomes a control target with numerical standards. Combined with the target conveying speed and arrival cycle parameters of the workstation, the first comprehensive indicator can reflect whether the start-stop is too abrupt, and whether the mold is prone to displacement or overturning under the current speed / cycle configuration. When this indicator exceeds the limit, the system automatically extends the acceleration and deceleration time, reduces the speed, and limits the number of consecutive start-stop cycles, essentially establishing an adaptive deceleration control based on actual impact and attitude feedback for each section of the conveying platform. This not only reduces the instantaneous impact during inter-section switching, avoiding the risks of mold micro-movement on the pallet, fixture loosening, or even mold collapse, but also reduces the probability of tooling, fixtures, and measuring equipment mounted on the mold falling, significantly improving operational safety and equipment lifespan.

[0030] In one embodiment, workstations located near the manual operation area and with a high center of gravity of the mold are selected as key monitoring workstations. Three-axis accelerometer modules are installed on the bottom surface of the pallet corresponding to these workstations, near the mold's center of gravity. These modules integrate amplification and filtering circuits and communicate with an industrial PC via a bus. Whenever the programmable controller issues a start or stop command to a certain section, it simultaneously sends "start / stop window start" and "start / stop window end" markers to the industrial PC. The industrial PC collects acceleration curves along the conveying direction within the marked time period to extract the magnitude of the start / stop impact. Shortly after the start / stop is completed, it continues to collect three-axis acceleration to calculate the change in the pallet's tilt angle around the transverse axis over time, and obtains the strength and attenuation characteristics of the swaying posture through statistical algorithms. To estimate variable acceleration, differential calculations can be performed on the acceleration time series along the conveying direction to observe the rate of acceleration change, thereby determining whether the speed change during start / stop is too abrupt. The mold weight and center of gravity height can be pre-imported into the industrial PC through process data or the mold management system. The software then correlates this information with the actually measured impact and posture data to form a unified evaluation scale. After each start-up and shutdown, the industrial PC stores the impact level, attitude stability level, and variable acceleration level of that segment in a database and feeds it back to the programmable controller via a network interface. This data is then used to adjust the start-up, shutdown, speed, and cycle time of each segment. For ordinary workstations without sensors, estimated impact and attitude levels can be provided based on mold weight, center of gravity height, and drive parameters, combined with debugging experience. These simplified versions are then input into the subsequent model.

[0031] Specifically, the calculation equation for the comprehensive impact attitude index X is as follows: ; Where σ is the absolute value of the maximum deceleration or acceleration of the mold pallet during the start-up and stop process of the workstation section, σ lim To determine the deceleration limit based on the allowable impact of the mold system, θ rms Let θ be the root mean square value of the tilt angle of the mold pallet around the lateral axis during start-up, shutdown, and inter-segment switching. lim J represents the maximum allowable tilt angle limit determined based on the mold's center of gravity height and clamping allowance, where J is the peak value of the variable acceleration of the mold pallet during start-up and shutdown. lim The variable acceleration limit is determined based on the pallet stiffness and mold mounting stiffness, where v is the currently set target conveying speed for this workstation section. ref q represents the reference conveying speed under no-load or light-load conditions, and q is the target arrival cycle time for the corresponding assembly station. ref The reference cycle time is given for the design process. α1, α2, α3, and α4 are dimensionless weighting coefficients that are pre-set according to the sensitivity of different workstations to impact, posture, and cycle time. The control system adjusts the target conveying speed v, the start / stop acceleration / deceleration time, and the shape of the acceleration / deceleration curve to keep the calculated X below a preset impact posture safety threshold.

[0032] The peak impact value σ and the root mean square angle of inclination θ rms The three physical quantities strongly related to human and equipment safety—variable acceleration J, variable acceleration J, and variable acceleration J—are combined into a single dimensionless index. Simultaneously, conveying speed v and cycle time q are introduced as shared parameters, reflecting the physical principle that the more aggressive the speed / cycle time, the higher the impact posture index. This definition of X has the following advantages: each term is the square of the measured / permissible quantity; a ratio greater than 1 indicates exceeding the limit, directly reflecting the safety margin; through the weighting of α1, α2, and α3, impact reduction or posture stabilization can be targeted for different workstations; for example, α2 can be increased for workstations with a high center of gravity; through α4(v / v... ref )(q ref The / q) term explicitly superimposes the risks of high speed and tight cycle time onto X, making the system automatically more conservative under high-speed, high-cycle conditions. The X value output by the formula is directly used by the control system as a constraint: when X is less than a threshold, higher speed and shorter acceleration / deceleration times are allowed to increase cycle time; when X approaches or exceeds the threshold, the drive parameters are automatically adjusted to lengthen acceleration / deceleration times, reduce speed, and decrease continuous start-stop cycles, so that speed cycle time and impact attitude safety are balanced under the same index. Those skilled in the art only need to collect acceleration and attitude data, set limit values ​​and weights, and then implement real-time calculation and closed-loop control in a PLC or industrial PC based on this formula.

[0033] In one embodiment, the industrial PC is pre-configured with impact limits, tilt angle limits, and variable acceleration limits for each type of workstation, along with corresponding weighting parameters to reflect the sensitivity of different workstations to impact and attitude. The software receives the start-stop impact level, attitude stability level, and variable acceleration level provided in Embodiment 2, and simultaneously obtains the conveying speed and target cycle time of the current workstation segment from the programmable controller. These quantities are normalized, and their respective proportions are calculated using the limits as a reference. These proportions are then superimposed according to preset weights to form a single comprehensive value, which serves as the evaluation result for the start-stop safety of that workstation segment. To reflect the aggressiveness of the operating mode, the current conveying speed is compared with a reference speed, and the current target cycle time is compared with a reference cycle time. When the speed is higher than the reference value or the cycle time is shorter than the reference value, this degree of "compression margin" is included in the comprehensive value, so that the comprehensive value naturally increases under high-speed, high-cycle conditions. The programmable controller (PLC) compares the comprehensive values ​​returned by the industrial PC with the upper limit safety threshold set for that workstation. If the comprehensive value remains high for a prolonged period, it automatically adjusts the target speed and acceleration / deceleration time of the drive segment, such as reducing the conveyor speed, lengthening the acceleration / deceleration time, and reducing the frequency of continuous start-stop operations. Conversely, if the comprehensive value remains low while the overall production line cycle time is high, the speed of that workstation segment can be appropriately increased or the acceleration / deceleration time shortened to recover some cycle time margin. In this way, the comprehensive value does not directly reveal the formula, but serves as a unified, dimensionless evaluation result, guiding the PLC to adaptively adjust the start-stop parameters of that workstation segment, thereby achieving safety control based on physical quantities.

[0034] Specifically, the cycle time buffer comprehensive index is used to characterize the cycle time matching degree and buffer occupancy status of the buffer areas before and after the workstation section. The control system is specifically configured to calculate the cycle time buffer comprehensive index as follows: Statistically analyze the actual completion cycle time of each work station segment within the preset time window, and calculate the cycle time deviation between the actual cycle time of each work station segment and the corresponding target cycle time. Collect occupancy status signals on the buffer sections before and after the critical workstation section, calculate the real-time occupancy quantity and occupancy rate of each buffer section within the time window, and obtain the maximum allowable occupancy rate of the buffer section based on the total number of buffer sections; The target conveying speed parameter and the assembly station target beat parameter are collected and used in the calculation of the impact attitude comprehensive index. The target conveying speed parameter and the assembly station target beat parameter are used as common beat parameters and are input into the beat buffer comprehensive index calculation model together with the beat deviation and the buffer occupancy rate to obtain the beat buffer comprehensive index used to guide the upstream release rhythm and buffer scheduling strategy. When the overall index of the cycle time buffer exceeds the preset upper limit and the corresponding buffer segment occupancy rate is close to saturation, the release frequency of the upstream workstation segment is reduced, the allowed waiting time of the upstream workstation segment is extended, and if necessary, cycle time limiting is implemented on several upstream workstation segments; when the overall index of the cycle time buffer is lower than the preset lower limit and the corresponding buffer segment is idle for a long time, the release frequency of the upstream workstation segment is increased or priority is given to releasing to the buffer segment arranged in front of the critical workstation segment, so as to improve the start-up rate of the critical workstation.

[0035] By quantifying cycle time deviation and buffer occupancy, queue length and buffer saturation, which were previously managed based on experience, are transformed into automatically adjustable control variables. The cycle time deviation ΔT is obtained by statistically analyzing the actual completion cycle time of each workstation within a specific time window, and this deviation is then combined with the buffer occupancy rate ρ and the target occupancy rate ρ. opt Queue size L and maximum queue size L max Based on the data, the system can make real-time judgments: Which workstations are consistently slower than the target cycle time, creating cycle time bottlenecks? Which buffers are nearing saturation, becoming accumulation points and posing a risk of congestion? Which buffers are idle for extended periods, wasting line resources?

[0036] With this indicator model, the control system can automatically reduce the upstream release frequency and extend the waiting time to protect the saturated buffer. It can also increase the upstream release frequency when the buffer is consistently empty, ensuring that critical workstations always have work to do. Compared to the simple logic of releasing workstations as soon as they are available and stopping when the buffer is full, this invention can adjust the release rhythm at a finer granularity, preventing the cycle time from being locked by a single slow workstation, and instead allowing upstream and downstream to work collaboratively through the buffer. For multi-variety, small-batch scenarios where cycle time fluctuates significantly, the second comprehensive indicator can be used to dynamically adjust the target cycle time of workstations and the buffering strategy, making the entire production line more robust to process changes.

[0037] In one embodiment, the programmable controller (PLC) configures an arrival detection signal and a completion confirmation signal for each workstation segment. The former is triggered by the photoelectric sensor at the segment's exit, and the latter is triggered by a human-machine interface button or barcode scanner next to the workstation. The control program records the arrival time when the pallet arrives and the completion time when the operator presses the completion confirmation button. The difference between the two is the cycle time required for the workstation to complete one operation. The control program accumulates multiple cycle time records within a certain time window (e.g., 10 minutes), calculates the average value, and compares it with the target cycle time set by the process to obtain the cycle time deviation. For buffer segments before and after the workstation, the PLC counts whether each buffer segment is occupied by a pallet at fixed time intervals. By averaging the occupancy flags over a period of time, the average occupancy rate of the buffer area is calculated. Simultaneously, the number of pallets in the buffer area is counted within the same time interval to obtain the average queue length. These cycle time deviation, occupancy rate, and queue length data are periodically uploaded to an industrial PC via Ethernet. In each optimization cycle, the industrial PC calculates the area before and after each critical workstation, and combines the magnitude of the cycle time deviation, the degree to which the cache utilization rate deviates from the target value, and the degree to which the queue length approaches the physical limit into a cycle time-cache index, which is used as a quantitative basis for the release policy and cache policy of that area.

[0038] Specifically, the calculation equation for the comprehensive index Y of the clock cycle buffer is as follows: ; Where ΔT is the average cycle time deviation of the workstation segment within the preset time window, that is, the absolute value of the time difference between the actual completion cycle time and the target cycle time q, q is the target arrival cycle time of the corresponding assembly station, and ρ is the average buffer occupancy rate of the corresponding buffer area within the time window. opt To set a target occupancy rate based on the optimal cache utilization allowed for this region, L is the average number of queued trays in this cache region within the time window. max The maximum allowable queue size is determined based on the number of physical segments in the buffer and the safety distance, where v is the target delivery speed. ref With q ref These are reference values ​​for the target conveying speed and the target cycle time, respectively. β1, β2, β3, and β4 are dimensionless weighting coefficients preset according to the process requirements for cycle time stability, buffer balance, and line utilization. The control system adjusts the release interval of the upstream workstation, the allowed number of buffer sections, and the target cycle time of the critical workstation to constrain the calculated Y within the preset cycle time buffer control range.

[0039] Key logistical characteristics such as cycle time deviation, buffer occupancy deviation ρ, and queue length are normalized and superimposed onto a dimensionless index Y, while allowing vvv and qqq to be shared with X, achieving "the entanglement of cycle time / speed parameters between impact safety and buffer utilization." The specific effects of this formula include: (ΔT / q) 2 This allows the cycle time deviation to be measured on a relative scale, which is not affected by the absolute value of the cycle time at different workstations, making it more suitable for multi-workstation comparison. (ρ ρ opt ) / ρ opt This ensures that the cache is neither overly saturated nor overly idle, guiding the system to operate the cache in a comfortable range. (L / L max ) 2 This amplifies the risk of queuing lengths approaching their limits, causing the system to rapidly reduce the release rate when it approaches its physical limits, thus preventing overflow and on-site congestion. The last term still uses v and q to reflect the risk of running too fast or having too tight a pace in Y, forming a common constraint direction with X.

[0040] The calculated value of Y directly drives the release and buffering strategies: when Y is too high, mitigation is achieved by reducing the upstream release frequency, decreasing the number of pallets allowed to enter simultaneously, and relaxing the cycle time of critical workstations; when Y is low, the system can appropriately increase the release frequency and shorten the waiting time to improve the overall line throughput. Through this metric, the system maintains the utilization rate of the buffer segment within a reasonable range while ensuring no line jams or prolonged idle runs. This improves the line load rate while avoiding extreme congestion, enhancing the overall production line smoothness and adaptability to process changes.

[0041] The industrial PC reads recently collected parameters into memory for the buffer zone before each critical workstation, including cycle time deviation, current target cycle time, current average buffer occupancy rate, target occupancy rate, average queue length, and maximum allowable queue length. The software uses the ratio of cycle time deviation to target cycle time, the degree of deviation between buffer occupancy rate and target occupancy rate, and the ratio of queue length to maximum queue length as primary inputs. These parameters, combined with the current workstation's conveyor speed and cycle time configuration, are weighted and summed to generate a cycle time-buffer index value. Setting different weight parameters for different buffer zones can highlight the cycle time and buffer status before and after critical bottleneck workstations. After this indicator is calculated, it is compared with the upper and lower limits set for the area. If the indicator is greater than the upper limit, it indicates that the cycle time deviation in the area is large and the buffer is approaching saturation. The industrial PC will reduce the release frequency of the upstream workstation, for example, by extending the minimum interval between two releases or reducing the number of pallets allowed to enter the buffer area at the same time, so that the buffer area gradually returns to a reasonable occupancy range. If the indicator is lower than the lower limit and there is a waiting period for materials at the downstream workstation, the release frequency of the upstream workstation will be increased, the release interval will be shortened, or the number of pallets allowed to enter the buffer area will be increased, so that the buffer is filled to a certain extent. For buffer areas within the normal range, the existing release strategy remains unchanged. Through this implementation, those skilled in the art can directly convert the statistical results of cycle time and buffer into the automatic adjustment logic of the upstream release strategy without frequent manual intervention.

[0042] Specifically, when the control system performs joint optimization using the impact attitude comprehensive index and the cycle buffer comprehensive index, it is specifically configured as follows: Within each preset optimization cycle, the comprehensive impact posture index and the comprehensive cycle buffer index corresponding to all workstations and buffer sections are collected to form an index set for the entire segmented conveying platform. Through the human-machine interface or pre-configured process recipes, different impact posture weights and cycle time buffer weights are set for different areas or workstations of different process importance, so as to prioritize cycle time stability in critical workstations and prioritize posture safety in high-risk workstations. Based on the set of indicators and their corresponding weights, the target conveying speed, acceleration and deceleration time, and upstream release interval time of the workstation section are taken as control variables to be optimized. Constraints such as the upper limit of the maximum conveying speed, the lower limit of the minimum acceleration and deceleration time, and the number of available buffer sections in each area are applied to each of the control variables. Based on the joint optimization results, the speed setpoints, acceleration / deceleration curve parameters, and release rhythm parameters of each of the aforementioned conveyor platform segments are updated. In the next optimization cycle, the comprehensive impact posture index and the comprehensive cycle buffer index are recalculated based on the new operating data to achieve rolling optimization control of the entire segmented conveyor platform.

[0043] By extending the X and Y categories of indicators to the entire production line optimization dimension, and introducing optimization cycles, weight configurations, and constraints, a workable rolling optimization framework is formed. Those skilled in the art can, based on the on-site process, mark certain workstations as critical, hazardous, or ordinary workstations, and then configure different impact-attitude weights and cycle time-buffer weights for these areas: for example, in areas adjacent to manual operation zones or high-center-of-gravity mold areas, the X weight can be increased to prioritize safety; in bottleneck workstations or cycle time-sensitive workstations, the Y weight can be increased to prioritize cycle time.

[0044] Based on this, the system treats the target speed, acceleration / deceleration time, and release interval of each workstation as variables to be optimized, imposing constraints such as speed upper limits, acceleration / deceleration time lower limits, and the number of available buffer segments. An optimization algorithm solves for a set of parameters to minimize the overall Z value. The rolling optimization mechanism ensures that when the process or order structure changes (e.g., a temporary addition of a process to a workstation, or an increase in the batch size of a certain mold), the control parameters can be adaptively updated, rather than remaining at their initial setpoints indefinitely. In this way, the entire segmented conveyor platform is no longer a static configuration, but an adaptive line capable of adjusting its cycle time and safety margin based on real-time data, significantly improving stability, flexibility, and energy efficiency during long-term operation.

[0045] In one embodiment, the industrial PC is set to a fixed optimization cycle, for example, performing a line-wide optimization every 30 seconds. At the beginning of each cycle, the industrial PC retrieves the start-stop combined values ​​of each workstation segment and the takt-buffer index of each buffer area from the database or shared memory in the previous cycle, and weights them according to the "safety weight" and "takt weight" configured for each workstation by the process engineer in the parameter table to obtain the overall safety level and overall takt level of the entire line. For each workstation segment, the industrial PC uses the target conveyor speed, acceleration / deceleration time, and upstream release interval as variables to be adjusted, and reads the maximum allowable speed, minimum allowable acceleration / deceleration time, and associated maximum queuing limit of the buffer area for that workstation segment as constraints. The optimization algorithm can adopt a stepwise adjustment approach, that is, in each cycle, it only attempts to increase or decrease the speed, acceleration / deceleration time, or release interval of a certain workstation segment by a small amount, and calculates the impact of this small adjustment on the overall line index. If the overall line index improves without violating the safety threshold and buffer constraints, the adjustment is accepted and the new parameters are written to the programmable controller; if no adjustment improves the overall index, the current parameters are kept unchanged, and the system waits for new data in the next cycle. In this way, during long-term operation, the operating parameters of each workstation gradually approach the optimal point around multiple objectives. The overall production line cycle time and safety level can be slowly and adaptively adjusted based on actual operating data, without the need to solve a complex mathematical optimization problem all at once. Those skilled in the art can implement this gradual adjustment logic using standard industrial PC programming languages ​​(such as C++ or Python) and then exchange parameters with the PLC via OPC or industrial Ethernet.

[0046] Specifically, the objective value Z of the joint optimization objective function is defined as: ; Where X is the calculated comprehensive impact attitude index, and Y is the calculated comprehensive clock cycle buffer index. ref Y is the impact posture reference value determined based on experience at the mold assembly site or simulation results. ref The reference value for the cycle buffer is determined based on the requirements of the overall cycle balance and buffer utilization. γ1, γ2, and γ3 are joint optimization weight coefficients that can be configured by the user according to the process scenario. Among them, γ1 and γ2 are non-negative, and γ3 is used to adjust the interaction between the impact posture and the cycle buffer. The control system is designed to satisfy X≤X max Y≤Y max Under the safety constraints, minimizing Z is the optimization objective. The target conveying speed, acceleration / deceleration time, and release interval parameters for each conveying platform segment are generated using a rolling optimization method. In subsequent optimization cycles, Z is recalculated based on the updated impact attitude comprehensive index and cycle buffer comprehensive index, thereby achieving dynamic joint optimization control of the entire segmented conveying platform.

[0047] Unify the two types of indicators, X and Y, into a single overall objective function Z, and introduce a reference X. ref ,Y ref The interaction term γ3 makes the optimization not just a simple linear weighting, but a nonlinear characteristic with safe beat interaction. Its beneficial effect is reflected in: through X / X... ref and Y / Y ref The normalized square of Z can simultaneously reflect how much the current impact attitude level deviates from the target level and how much the current beat / buffer level deviates from the target level; the cross term (X / X ref (Y / Y) ref ) is used to characterize the combined effect when both safety and cycle time indicators deviate significantly, so that the control system can more actively reduce speed and widen cycle time when both are not ideal, rather than adjusting only one side; the configuration of γ1, γ2, and γ3 allows users to choose between prioritizing safety or prioritizing cycle time according to different production strategies, and make dynamic adjustments at different stages.

[0048] Minimizing Z is directly used to drive the optimization of the entire production line parameters: In each optimization cycle, the control system uses the real-time calculation results of X and Y as input to evaluate the size of Z, and then reduces Z by adjusting variables such as speed, acceleration / deceleration, and release interval of each work station.

[0049] In one embodiment, during the system debugging phase, the industrial PC first selects a set of operating parameters considered safe and with reasonable cycle time as a baseline configuration. The entire production line is then run under this configuration for a period of time, and the average value of the start-stop comprehensive index for each workstation and the average value of the cycle time-buffer index for each buffer zone are recorded as reference levels for comprehensive evaluation. Based on different production strategies, such as safety priority, capacity priority, or balanced mode, three different sets of weight parameters are set for the system to combine safety-related and cycle time-related indicators to form a total evaluation value for the entire production line. During operation, the industrial PC calculates the deviation of the current safety-related and cycle time-related indicators from the reference level in each optimization cycle and synthesizes them into a total evaluation value according to the currently selected strategy weights, minimizing the total evaluation value while ensuring that safety constraints are not breached. The human-machine interface allows operators to select different strategies for different production batches. For example, a safety priority strategy is used for trial molding or high-risk mold batches, while a capacity priority or balanced strategy is used when capacity is ramping up or delivery is tight. After each strategy switch, the combined weights are updated, the comprehensive evaluation method is adjusted accordingly, and the optimization algorithm automatically searches for suitable speed, cycle time, and buffer configurations in the new target direction. By combining this strategy formulation with evaluation values, those skilled in the art can adapt to different production stages and operational preferences within the same hardware and software framework by adjusting parameters, without modifying the underlying control structure.

Claims

1. A segmented conveyor platform for assembling plastic molds, characterized in that, Includes the following steps: The system includes multiple conveyor platform sections arranged sequentially along the mold assembly process, a tray for supporting the plastic mold, and a control system electrically connected to each of the conveyor platform sections. The control system includes at least a programmable logic controller (PLC), a drive control unit, and a human-machine interface. The system is characterized in that it is configured as follows: S1. Divide each of the conveying platform segments into workstation segments and buffer segments. Each conveying platform segment is equipped with a drive motor and segment head and end position detection sensors. The segment head and end position detection signals are collected to the programmable controller so that each conveying platform segment can be used as a control unit that can be independently started, stopped and speed adjusted. S2. When the mold pallet is detected by the section head position detection sensor of a certain work station section to enter the work station section, the work station section is controlled to run to the preset stopping position according to the preset acceleration curve, and the work station section is stopped after the mold pallet reaches the preset stopping position, so that the mold pallet is stationary in the work station section, so that the operator can complete the corresponding mold assembly process in the stationary state of the mold pallet. S3. During the operation of the mold pallet on each of the conveying platform sections, the conveying speed of each workstation section, the acceleration and deceleration during the start and stop process, the attitude change related to the height of the mold center of gravity, the arrival time of the mold pallet entering the workstation section, and the occupancy status of the buffer sections before and after the workstation section are collected. The conveying speed and the arrival time are used as common time parameters, the attitude change and the impact related to the start and stop process are used as impact attitude parameters, and the occupancy rate of the buffer section and the time deviation of the workstation section are used as buffer time parameters. S4. Based on the shared cycle time parameters and the impact posture parameters, calculate the comprehensive impact posture index at the entrance of each workstation section to characterize the impact of inter-section switching and the stability of the mold posture. Then, dynamically adjust the start-stop acceleration and deceleration curves, target conveying speed and maximum allowable number of start-stop cycles of the corresponding workstation section according to the comprehensive impact posture index, so that the impact of the mold pallet during the inter-section transfer process is controlled within a preset range and the instability of the mold posture on the pallet is suppressed. S5. Based on the shared cycle time parameters and the buffer cycle time parameters, calculate the cycle time buffer comprehensive index, which characterizes the degree of cycle time matching and the degree of buffer saturation, in the buffer areas arranged before and after the key work station section. According to the cycle time buffer comprehensive index, dynamically adjust the release rhythm of the upstream work station section, the number of buffer sections allowed to occupy, and the target cycle time of the downstream work station section to buffer the cycle time difference between fast and slow processes and avoid serious accumulation or idleness before and after the key work station. S6. The comprehensive impact attitude index and the comprehensive cycle buffer index calculated for each workstation segment and buffer segment are used as input variables for joint optimization. A joint optimization objective function is constructed with the goal of reducing inter-segment impact and attitude risks, improving assembly cycle stability and buffer utilization. Based on the optimization results of the joint optimization objective function, speed setpoints, acceleration / deceleration curve parameters and release rhythm parameters for each of the conveying platform segments are generated and sent to the corresponding drive control unit to realize mold pallet conveying and assembly control according to the workstation rhythm on the entire segmented conveying platform.

2. The segmented conveyor platform for assembling plastic molds according to claim 1, characterized in that, The comprehensive impact posture index is used to characterize the start-stop impact and mold posture stability of a single workstation entry point within a conveying cycle. The control system is specifically configured to calculate the comprehensive impact posture index as follows: Accelerometers installed on the pallet bearing surface or workstation frame collect the acceleration time series of the mold pallet during the start-up and stop process, and extract the impact peak along the conveying direction during the start-up and stop process and the attitude decay characteristics of the mold pallet in the lateral swaying direction after the start-up and stop. By combining the preset mold weight and mold center of gravity height parameters, the impact peak and attitude attenuation characteristics are normalized to obtain the inter-segment start-stop impact level and attitude stability level of the workstation under the current operating parameters. Collect the target conveying speed parameters of the current workstation section and the arrival cycle parameters of the corresponding assembly workstation. Use the target conveying speed parameters and the arrival cycle parameters as common cycle parameters, and input them together with the inter-section start-stop impact level and attitude stability level into the impact attitude comprehensive index calculation model to obtain the impact attitude comprehensive index used to constrain the start-stop control of the workstation section. When the comprehensive impact posture index exceeds the preset upper limit, the acceleration and deceleration time of the workstation section is automatically extended, the target conveying speed is reduced, the number of consecutive start-stop cycles is limited, or the proportion of uniform speed sections in the transition distance between sections is increased, so as to reduce the risk of instantaneous impact and posture instability at the section switching point.

3. The segmented conveyor platform for assembling plastic molds according to claim 2, characterized in that, The equation for calculating the comprehensive impact attitude index X is as follows: ； Where σ is the absolute value of the maximum deceleration or acceleration of the mold pallet during the start-up and stop process of the workstation section, σ lim To determine the deceleration limit based on the allowable impact of the mold system, θ rms Let θ be the root mean square value of the tilt angle of the mold pallet around the lateral axis during start-up, shutdown, and inter-segment switching. lim J represents the maximum allowable tilt angle limit determined based on the mold's center of gravity height and clamping allowance, where J is the peak value of the variable acceleration of the mold pallet during start-up and shutdown. lim The variable acceleration limit is determined based on the pallet stiffness and mold mounting stiffness, where v is the currently set target conveying speed for this workstation section. ref q represents the reference conveying speed under no-load or light-load conditions, and q is the target arrival cycle time for the corresponding assembly station. ref The reference cycle time is given for the design process. α1, α2, α3, and α4 are dimensionless weighting coefficients that are pre-set according to the sensitivity of different workstations to impact, posture, and cycle time. The control system adjusts the target conveying speed v, the start / stop acceleration / deceleration time, and the shape of the acceleration / deceleration curve to keep the calculated X below the preset impact posture safety threshold.

4. The segmented conveyor platform for assembling plastic molds according to claim 3, characterized in that, The cycle time buffer comprehensive index is used to characterize the cycle time matching degree and buffer occupancy status of the buffer areas before and after the workstation section. The control system is specifically configured to calculate the cycle time buffer comprehensive index as follows: Statistically analyze the actual completion cycle time of each work station segment within the preset time window, and calculate the cycle time deviation between the actual cycle time of each work station segment and the corresponding target cycle time. Collect occupancy status signals on the buffer sections before and after the critical workstation section, calculate the real-time occupancy quantity and occupancy rate of each buffer section within the time window, and obtain the maximum allowable occupancy rate of the buffer section based on the total number of buffer sections; The target conveying speed parameter and the assembly station target beat parameter are collected and used in the calculation of the impact attitude comprehensive index. The target conveying speed parameter and the assembly station target beat parameter are used as common beat parameters and are input into the beat buffer comprehensive index calculation model together with the beat deviation and the buffer occupancy rate to obtain the beat buffer comprehensive index used to guide the upstream release rhythm and buffer scheduling strategy. When the overall index of the cycle time buffer exceeds the preset upper limit and the corresponding buffer segment occupancy rate is close to saturation, the release frequency of the upstream workstation segment is reduced, the allowed waiting time of the upstream workstation segment is extended, and if necessary, cycle time limiting is implemented on several upstream workstation segments; when the overall index of the cycle time buffer is lower than the preset lower limit and the corresponding buffer segment is idle for a long time, the release frequency of the upstream workstation segment is increased or priority is given to releasing to the buffer segment arranged in front of the critical workstation segment, so as to improve the start-up rate of the critical workstation.

5. A segmented conveyor platform for assembling plastic molds according to claim 4, characterized in that, The equation for calculating the comprehensive index Y of the clock buffer is as follows: ； Where ΔT is the average cycle time deviation of the workstation segment within the preset time window, that is, the absolute value of the time difference between the actual completion cycle time and the target cycle time q, q is the target arrival cycle time of the corresponding assembly workstation, and ρ is the average buffer occupancy rate of the corresponding buffer area within the time window. opt To set a target occupancy rate based on the optimal cache utilization allowed for this region, L is the average number of queued trays in this cache region within the time window. max The maximum allowable queue size is determined based on the number of physical segments in the buffer and the safety distance, where v is the target delivery speed. ref With q ref These are reference values ​​for the target conveying speed and the target cycle time, respectively. β1, β2, β3, and β4 are dimensionless weighting coefficients preset according to the process requirements for cycle time stability, buffer balance, and line utilization. The control system adjusts the release interval of the upstream workstation, the allowed number of buffer sections, and the target cycle time of the critical workstation to constrain the calculated Y within the preset cycle time buffer control range.

6. The segmented conveyor platform for assembling plastic molds according to claim 5, characterized in that, When the control system performs joint optimization using the impact attitude comprehensive index and the clock cycle buffer comprehensive index, it is specifically configured as follows: Within each preset optimization cycle, the comprehensive impact posture index and the comprehensive cycle buffer index corresponding to all workstations and buffer sections are collected to form an index set for the entire segmented conveying platform. Through the human-machine interface or pre-configured process recipes, different impact posture weights and cycle time buffer weights are set for different areas or workstations of different process importance, so as to prioritize cycle time stability in critical workstations and prioritize posture safety in high-risk workstations. Based on the set of indicators and their corresponding weights, the target conveying speed, acceleration and deceleration time, and upstream release interval time of the workstation section are taken as control variables to be optimized. Constraints such as the upper limit of the maximum conveying speed, the lower limit of the minimum acceleration and deceleration time, and the number of available buffer sections in each area are applied to each of the control variables. Based on the joint optimization results, the speed setpoints, acceleration / deceleration curve parameters, and release rhythm parameters of each of the aforementioned conveyor platform segments are updated. In the next optimization cycle, the comprehensive impact posture index and the comprehensive cycle buffer index are recalculated based on the new operating data to achieve rolling optimization control of the entire segmented conveyor platform.

7. A segmented conveyor platform for assembling plastic molds according to claim 6, characterized in that, The objective value Z of the joint optimization objective function is defined as: ； Where X is the calculated comprehensive impact attitude index, and Y is the calculated comprehensive clock cycle buffer index. ref Y is the impact posture reference value determined based on experience at the mold assembly site or simulation results. ref The reference value for the cycle buffer is determined based on the requirements of the overall cycle balance and buffer utilization. γ1, γ2, and γ3 are joint optimization weight coefficients that can be configured by the user according to the process scenario. Among them, γ1 and γ2 are non-negative, and γ3 is used to adjust the interaction between the impact posture and the cycle buffer. The control system is designed to satisfy X≤X max Y≤Y max Under the safety constraints, minimizing Z is the optimization objective. The target conveying speed, acceleration / deceleration time, and release interval parameters for each conveying platform segment are generated using a rolling optimization method. In subsequent optimization cycles, Z is recalculated based on the updated impact attitude comprehensive index and cycle buffer comprehensive index, thereby achieving dynamic joint optimization control of the entire segmented conveying platform.

Images