Accurate temperature regulation and control method for flexible mold

Abstract

The invention relates to the technical field of injection molding temperature control, in particular to a flexible mold temperature accurate regulation and control method which comprises the steps that logic areas of a current mold are obtained, temperature time sequence data sets of all the logic areas are collected, and temperature characteristics of all the logic areas under a stage boundary are analyzed based on the number of offset points of all the logic areas; delimiting event types and association rules corresponding to the temperature features; the temperature features of all the logic areas serve as input, the temperature similarity of all the logic areas is calculated, and the logic areas needing to be adjusted are marked as areas to be adjusted; generating a temperature change signal according to the duration time and the control period of the to-be-adjusted region at the boundary of each stage, and extracting an abnormal label of the temperature change signal; and taking the temperature change signal as the basis of strategy comparison, taking the data of the temperature change signal under the boundary of each stage as the node of the strategy tree, and determining a target temperature regulation strategy according to the path parameter value of the strategy tree. And the response rate and the accuracy of temperature regulation and control are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of injection molding temperature control, and in particular to a method for accurately controlling the temperature of a flexible mold. Background Art

[0002] In thermal processing processes such as injection molding and die-casting, precise mold temperature control is crucial for ensuring product quality and improving production efficiency. However, in mold temperature control scenarios, fixed partitioning can lead to uneven mold temperature control, resulting in delayed temperature control strategy responses and reduced mold production performance.

[0003] For example, Chinese patent publication number CN118809988A discloses an automatic injection molding machine operation monitoring system and method. During the injection molding process, due to the inclusion of multiple stages such as mold filling, pressure holding, and cooling, the mold temperature and injection pressure will fluctuate with the stage changes. In order to accurately monitor abnormal conditions, real-time and historical mold temperature and injection pressure time series data are obtained. By analyzing the similarities between the real-time time series data and the historical time series data of the injection pressure, the stage time periods of the current injection molding process are divided. Considering the impact of changes in production conditions on the duration of the stage, it is necessary to analyze the difference in duration between the real-time and historical data within the stage time period to accurately locate the comparison moment.

[0004] For example, Chinese patent publication number CN114734604A discloses a method for online mold temperature control during the injection molding process. To overcome the problem of poor temperature control stability caused by the inability to eliminate the time lag of temperature changes in existing technologies that adjust temperature through real-time temperature detection, the present invention includes the following steps: dividing the area from the cavity gate to the cavity end into several detection zones; determining the temperature control mode of the detection zone based on the temperature change rate per unit time, the real-time temperature value, and the stage of the injection molding process; each detection zone executes the corresponding temperature control mode, and a model is trained based on the inlet temperature, outlet temperature, water flow rate, and water flow pressure of the cooling / heating water circuit. The model is used to estimate the temperature change of the detection zone and control the water supply of the cooling / heating water circuit; and the mold injection molding process is completed.

[0005] In the existing technology, the temperature change trend is identified by observing the temperature difference at the comparison moment through the difference in the duration of the injection molding process, and the temperature change of the current injection molding process is viewed through the water flow circuit; however, when identifying the temperature in the existing technology, it is easy to ignore local abnormalities on the mold surface area, resulting in the mold having a certain complex shape. It is impossible to dynamically adjust the partition according to the mold, and timely check the event type in each partition, so as to complete the precise control of the local area on the mold. Summary of the Invention

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: a method for precise temperature control of a flexible mold, including: S1, obtaining a three-dimensional model of the current mold, partitioning the current mold according to shape features, and forming various logical areas.

[0007] S2, collects the temperature time series data set of each logical area, analyzes the temperature characteristics of each logical area under the stage boundary based on the number of offset points of each logical area, and defines the event type and association rules corresponding to the temperature characteristics.

[0008] S3, taking the temperature characteristics of each logic area as input, calculating the temperature similarity of each logic area, determining the relationship between each logic area and the temperature characteristics, and marking the logic area that needs to be adjusted as the area to be adjusted.

[0009] S4, analyzing the temperature operation status of the area to be adjusted, generating a temperature change signal based on the duration and control cycle of the area to be adjusted at the boundary of each stage, and extracting an abnormal label of the temperature change signal.

[0010] S5, taking the temperature change signal as the basis for strategy comparison, taking the data of the temperature change signal at the boundary of each stage as the node of the strategy tree, and determining the target temperature control strategy according to the path parameter value of the strategy tree.

[0011] The beneficial effects of the present invention are as follows: 1. The present invention divides the mold surface into multiple logical areas, identifies the data points with offsets based on the temperature change rate in each logical area, and then, starting from arbitrary temperature features, aligns the temperature deviation with the process stage boundary, calculates the time offset and generates conditional probability association rules to describe the spatiotemporal correlation of the current temperature features under the process stage, providing data support for subsequent strategic decision-making.

[0012] 2. The present invention calculates temperature similarity and classifies outlier regions based on the calculated temperature similarity. It distinguishes the types of anomalies by the mold opening temperature difference and the cooling rate difference to define the areas with temperature anomalies in the current processing process. The problematic areas are then queried to determine the anomaly labels that exist when the current areas are clustered, thereby realizing the location and identification of anomalies in local areas.

[0013] 3. The present invention uses temperature values, abnormal labels, and logical areas as input nodes, associates stage boundaries with temperature control strategies, and selects the optimal strategy through path weight optimization to obtain the target strategy in the current scenario and improve the response speed and accuracy of temperature control. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The present invention will be further described below with reference to the accompanying drawings and examples.

[0015] Figure 1It is a flow chart of a method for precisely controlling the temperature of a flexible mold.

[0016] Figure 2 The figure is a flow chart of step S2 of a method for precisely controlling the temperature of a flexible mold.

[0017] Figure 3 The figure is a flow chart of step S3 of the method for precisely controlling the temperature of a flexible mold.

[0018] Figure 4 The figure is a flow chart of step S4 of the method for precisely controlling the temperature of a flexible mold.

[0019] Figure 5 The figure is a flow chart of step S5 of the method for precisely controlling the temperature of a flexible mold. DETAILED DESCRIPTION

[0020] The following embodiments of the present invention are described in detail. The embodiments described below are exemplary and are only used to explain the present invention, and are not to be construed as limiting the present invention. Where specific techniques or conditions are not specified in the embodiments, the techniques or conditions described in the literature in the art or in the product specifications shall be followed.

[0021] See Figure 1 A method for precisely controlling the temperature of a flexible mold includes: S1, obtaining a three-dimensional model of the current mold, partitioning the current mold according to shape features, and forming logical areas.

[0022] S2, collects the temperature time series data set of each logical area, analyzes the temperature characteristics of each logical area under the stage boundary based on the number of offset points of each logical area, and defines the event type and association rules corresponding to the temperature characteristics.

[0023] S3, taking the temperature characteristics of each logic area as input, calculating the temperature similarity of each logic area, determining the relationship between each logic area and the temperature characteristics, and marking the logic area that needs to be adjusted as the area to be adjusted.

[0024] S4, analyzing the temperature operation status of the area to be adjusted, generating a temperature change signal based on the duration and control cycle of the area to be adjusted at the boundary of each stage, and extracting an abnormal label of the temperature change signal.

[0025] S5, taking the temperature change signal as the basis for strategy comparison, taking the data of the temperature change signal at the boundary of each stage as the node of the strategy tree, and determining the target temperature control strategy according to the path parameter value of the strategy tree.

[0026] When forming each logical region, the area represented by the mold surface is divided into logical regions for the current temperature recognition processing, for example, based on the wall thickness variation at multiple locations of the current mold, cavity complexity, gate location, hot runner distribution, cooling water channel layout, and other factors. In this case, the division is based on the distance of the shape features, and after covering the current mold with multiple dense grids, multiple regions with different description forms are formed. At this time, the region division can be based on pre-set regions. After that, temperature sensors are deployed in each logical region to ensure that each region has sufficient temperature data. At this time, the point with the largest temperature gradient in each logical region, as well as points close to the mold cavity surface and points close to the cooling water channel, can be selected to obtain the current mold temperature.

[0027] That is, the implementation method of step S1 includes: S11, using the shape features of the current model as input to perform geometric topological analysis to obtain the key features of the mold. The key features will represent multiple areas with similar shapes, such as wall thickness mutation areas, surface curvature, gate / hot runner positions, and cooling water channel layouts. If there is a conflict in the description of the area, the larger area in the conflict is used as the division standard to divide the mold surface into multiple areas. Then, the temperature change of each area is queried to find out whether the current mold has uneven heating.

[0028] S12: Divide the mold surface into multiple logical partitions based on the mold's key features. The logical partitions are determined based on the key features. Each key feature's area is circled and considered a logical partition. If more precise control of the mold is required, the area corresponding to each key feature can be divided into multiple directional grids of equal size to form the currently monitored logical partitions. Array temperature sensors can be used to monitor the temperature of each logical partition at multiple locations on the mold surface.

[0029] In one embodiment of the present invention, the temperature time series data set will contain all the data of each area after completing a mold production cycle. At this time, sufficient temperature data will be obtained based on the batch of mold production to complete the adjustment and control of the current mold.

[0030] The aforementioned stage boundaries include the temperature values ​​at the start and end positions of the preheating stage, injection stage, holding stage, and cooling stage to describe whether the current mold needs to be heated to compensate for the temperature or adjust the cooling temperature during production. This is used to indicate whether there are abnormal temperature changes at the adjacent start and end positions of each stage, and the event type and association rules of the temperature characteristics in each stage are annotated with the temperature values ​​at the start and end positions of each stage. The extracted temperature features will represent the starting temperature, end temperature, average temperature, temperature change, temperature peak, and cooling rate of each stage. These data are combined into the currently extracted temperature features for subsequent identification and processing of the changing trend of the current mold production temperature.

[0031] The cooling rate here is used to illustrate the changes in the temperature reduction of the current mold in multiple stages of production to prevent problems such as excessive or slow temperature reduction. At this time, water flow at 25°C or other temperatures will be injected to adjust its temperature in the form of cooling water channels and reduced heating rates.

[0032] Therefore, the implementation of step S2 further includes: using the temperature characteristics of the temperature time series data set at the stage boundary to perform feature recognition on the temperature values ​​within each stage boundary, and obtain the temperature change rate of each data point.

[0033] Based on the temperature change rate of each data point, cluster analysis is performed on each data point of the temperature time series data set to obtain at least one set of temperature change rate offset points and neighborhood temperature change offset points.

[0034] When the number of temperature change rate offset points and neighborhood temperature change offset points exceeds a preset number threshold, the corresponding logical area is output to obtain the number of offset points in each logical area.

[0035] At this time, check whether the temperature values ​​in each logical area are offset in multiple stages, and output the part where the offset point is greater than the preset threshold to illustrate the offset; the temperature change rate offset point and the neighborhood temperature change offset point are used as the offset points for identifying each logical area.

[0036] The above-mentioned temperature change rate offset point is the average value of the temperature value of the historical data at the same stage plus three standard deviations. It serves as the threshold of the temperature change rate offset point at this time. Here, we tend to judge whether the temperature change rate per unit time exceeds the threshold of the temperature change rate in the historical data to obtain the current temperature accumulation situation.

[0037] The neighborhood temperature change offset point indicates that there is a data point whose temperature change rate is inconsistent with that of the adjacent logical area data points. The corresponding set of data points is regarded as the offset point identified at this time to illustrate the current temperature change rate.

[0038] As for the preset number threshold of temperature change rate offset points, the average number used when the offset point alarm is performed in the historical data is used as the preset number threshold at this time. As for the preset number threshold of neighborhood temperature change offset points, the average number of neighborhood temperature change offset points that appear continuously when the alarm is performed in the historical data is used as the preset number threshold.

[0039] When there are too many temperature change rate offset points and neighborhood temperature change offset points, it can be known that there is an abnormality in the current logical area. Then, according to the part of the temperature change rate offset points and neighborhood temperature change offset points identified at this time, the output logical area is used as the part for subsequent temperature feature analysis to explain what event type exists in the logical area where heating fluctuations and temperature change abnormalities exist, and what association rules need to be set for the relevant logical areas to view the relevant abnormalities that may exist in the corresponding areas.

[0040] Preferably, when using the number of temperature change rate offset points and the number of neighborhood temperature change offset points to obtain the number of offset points in each logical area, it also includes: intercepting the time points corresponding to the number of offset points, checking the distribution time period of each deviation point in the temperature time series data set, and outputting the temperature time series data set in sequence based on the number of offset points contained in the distribution time period and with the time period with the largest number of offset points as the center.

[0041] The intercepted time points are used to describe the distribution of offset points, indicating which time periods are more likely to experience abnormal temperature changes. This helps us understand the trends and related patterns of temperature anomalies. Then, with the time period with the largest number of offset points as the center, we can focus on the period when problems are most likely to occur. The temperature time series datasets can be analyzed and processed in order from the largest to the smallest number of offset points.

[0042] At the same time, when describing its temperature characteristics, it is also necessary to use sensors to check the length of time the real-time temperature of the mold continues to cool and heat up under the same historical data, and associate these data with the currently obtained temperature characteristics to facilitate subsequent retrieval queries.

[0043] like Figure 2As shown, the implementation method of step S2 also includes: S21, based on any temperature feature under the stage boundary, the temperature value corresponding to each data point is clustered with the data of the temperature feature having deviations at each moment, and the temperature identifier is set with the temperature value of each cluster cluster after clustering, and the event type of each temperature feature under the temperature value is identified. The temperature identifier set at this time is used to illustrate the temperature value taken by the temperature feature in the cluster cluster after clustering, and then the temperature features of these clusters are marked to see whether there is a corresponding event type, such as the content of a sudden increase in temperature in a certain area in the clustered cluster, or the temperature value that suddenly decreases. These event types are used to illustrate whether the temperature feature can indicate whether there is a potential abnormal event in the logical area divided by the current mold after clustering; as for the clustering method, any temperature feature under the current stage boundary can be clustered by clustering methods such as K-means and DBSCAN to identify the temperature pattern existing in the current mold production process.

[0044] The above event types include but are not limited to sudden temperature rise and fall, abnormal periodic fluctuations, abnormal temperature deviation from the baseline value, local overheating and abnormal uniformity, etc., which are used to indicate whether corresponding temperature anomalies exist in the multiple logical areas divided by the current mold.

[0045] S22, align the first triggering moment of each event type with the time series corresponding to the stage boundary, calculate the time offset of the triggering moment of each event type relative to the stage boundary, and use it as the triggering time of each event type. At this time, the triggering time is used to indicate the time point when the temperature characteristic deviates. The time point at which the corresponding time type can appear is associated with the corresponding stage of the current mold production. For example, the first triggering time of the event type is associated with the time of the injection start, the pressure holding end, and other stages. Then, the time difference between the occurrence of the corresponding event type and the corresponding stage is described to illustrate how long it takes to maintain the corresponding temperature value for a certain time, lag for a certain time, or the corresponding temperature value changes, and how long the corresponding abnormality may occur.

[0046] The trigger time above indicates the time point at which each event type can persist and be included after being first recognized by the system within the current time period.

[0047] S23, check the conditional probability of each temperature feature and event type, generate association rules corresponding to each temperature feature based on the conditional probability and trigger time corresponding to each temperature feature and event type, and add the association rules of each temperature feature to the boundaries of each stage; for example, the association rule can be expressed as follows: if the temperature value of the pressure holding stage is A and the trigger time is <5 seconds, there is a 90% probability of a sudden temperature increase; at this time, the possible event type and temperature value, the trigger time of the temperature value change, the duration of reaching the temperature value, etc. are comprehensively described to express what form the abnormality will take when it exists in the current logical partition.

[0048] The conditional probability of a temperature feature and event type is used to describe the probability of a temperature value and the corresponding event type that satisfy the current deviation. To calculate the conditional probability, the temperature feature's temperature value is calculated, along with the frequency of the associated event type. The probability of each temperature feature and event type is then calculated using the conditional probability formula.

[0049] In one embodiment of the present invention, the temperature similarity calculation focuses on identifying the starting temperature in the preset stage, the temperature variation in the injection stage, the temperature peak and average temperature in the holding stage, and the corresponding cooling rate in the cooling stage.

[0050] The temperature characteristics of the logical area are combined with the differences in the identified content at the boundaries of each stage to calculate the temperature similarity of each logical area at the boundary conditions of different stages. The temperature similarity can be calculated using the cosine similarity method to obtain the temperature similarity of the current corresponding logical area. At the same time, it is necessary to determine whether there are abnormal parts between the temperature characteristics that need to be adjusted, and mark these abnormal parts as parts that need to be adjusted to regulate the temperature conditions of the flexible mold at various positions.

[0051] In step S3, based on the event types and association rules found after the temperature feature analysis, multiple logical areas are associated and analyzed to illustrate the relationships between the logical areas when a single area is abnormal. Based on the obtained mutual areas, the area that currently needs to be adjusted is found to complete the identification and processing of the specific temperature under the current mold production. The main method is to use the obtained event types and association rules to transfer the currently acquired temperature features in a label transmission manner, map the temperature features to each other, and then perform an expected comparison with the mapped data to check whether the current logical area needs to be marked as an area to be adjusted.

[0052] like Figure 3 As shown, the implementation method of step S3 includes: S31, using the event type and association rules corresponding to the temperature feature, the temperature similarity of the temperature feature of the current logical partition is calculated, and the temperature similarity calculation is performed in sequence using the temperature mean, standard deviation and range in each logical partition; at this time, similarity processing is performed on the static features of each logical area, and similar patterns under different time series are retained in the form of time series; as for the event type and association rules, when processing the dynamic features related to the temperature change rate, the analysis contents under the static features and dynamic features are combined and processed, so that more comprehensive analysis data of the current logical area can be obtained, thereby improving the recognition of relevant data in the area.

[0053] It should be noted that after performing temperature-related calculations, if there are different dimensions or large differences in values, a standardization process can be used to analyze all temperature values ​​as normalized values.

[0054] Based on the temperature similarity obtained by the temperature mean, standard deviation, and range, the average value of the temperature similarity is selected to represent the correlation between the current logical partition and other partitions under the partial temperature features corresponding to the specific event type and association rule.

[0055] S32, after the temperature similarity calculation is completed, hierarchical clustering is performed based on the temperature similarity between the logical areas to obtain multiple clusters. At this time, a bottom-up merging strategy is adopted during clustering. The data after the temperature similarity calculation is set as the similarity matrix. The data of each logical area in the temperature similarity calculation is used as an initial cluster. The two clusters with the highest similarity are gradually merged to finally complete the hierarchical clustering and obtain multiple clusters after the current clustering is completed.

[0056] S33, outlier region detection is performed on the logic regions within each cluster. The outlier regions that exist after clustering indicate that the temperature behavior pattern of the region is significantly different from the overall temperature change of the mold, usually resulting in local circulation abnormalities or cooling failures. Based on the outlier detection results, the corresponding logic region is marked as an area to be adjusted.

[0057] When performing outlier area detection on the logical areas within each cluster, the implementation method also includes: if there are outlier areas, the outlier areas are classified according to the mold opening temperature difference and cooling rate, and the abnormal mold opening temperature difference area and the abnormal cooling rate area corresponding to the current temperature characteristics are divided; at the same time, the relevant type of the current logical area is also checked. For example, if the current logical area is a cooling channel, there is no need to directly check its outlier areas and other contents, and the abnormal mold opening temperature difference area and the abnormal cooling rate area are marked as areas to be adjusted.

[0058] The abnormal mold opening temperature difference area indicates that the area corresponding to a certain cluster has a temperature that exceeds the allowable deviation when the mold is opened, which will cause product warping or demolding damage. This area is regarded as the abnormal mold opening temperature difference area; the abnormal cooling rate area indicates low cooling efficiency, and there are any phenomena such as too low water flow rate, too high water temperature and poor contact, which indicate a decrease in cooling rate. At this time, by describing the rate of change of the temperature value of the cluster after clustering in the cooling stage, we can check the outlier area that meets the insufficient cooling rate, and then regard the area corresponding to these data as the abnormal cooling rate area.

[0059] For example, the outlier area detection method uses the algorithm of local outlier factor for identification, and identifies the k nearest neighbor data points for the data point p in each cluster. At this time, the Euclidean distance between each data point is calculated by normalizing the temperature value of each data point in the cluster, and the Euclidean distance of the kth nearest neighbor data point is used as the k distance of the data point p. After that, the k distance is used to calculate the reachable distance from the data point to other data points, and the local reachable density of each data point is judged to finally obtain the local outlier factor of each data point. Then, the existing outlier areas are screened according to the value of the local outlier factor, and the part with a local outlier factor value greater than 1.5 is regarded as the currently required outlier area to represent the difference relative to the whole, or other forms of local outlier factor values ​​are used, which can be set based on the current calculation needs; the specific calculation formula can be directly viewed in the existing technology, so it will not be described in detail here. As for the nearest neighbor data points obtained, they can be selected based on 3 to obtain data points that are relatively close to the current data point, and the local outlier factor represented by each data point can be identified using the outlier detection calculation method, thereby selecting local data that are obviously different.

[0060] In one embodiment of the present invention, when it is determined that the area to be adjusted has temperature anomalies related to the mold opening temperature difference and cooling rate, a temperature operation analysis is performed based on the average temperature, standard deviation, lag time, overshoot, and number of abnormal event types corresponding to the stage boundary of the area to be adjusted. This data is combined into a sequence at different stage boundaries to illustrate the specific stage of the temperature anomaly at the current stage boundary and the relative changes presented. The lag time is the time required for the current temperature to reach the target value after receiving a command or at the beginning of the stage. It reflects the response speed of the system. The shorter the lag time, the faster the system response. The overshoot indicates that the system response exceeds the target value. The difference ratio between the lag time and the preset response time is calculated as the overshoot value. As for the number of abnormal event types, based on the event types corresponding to the temperature characteristics, the event types represented as abnormal in the current area to be adjusted are counted to describe the relative situation of the area to be adjusted.

[0061] like Figure 4 As shown, the implementation of step S4 includes: S41, extracting the average temperature, standard deviation, lag time, overshoot and number of abnormal event types of the area to be adjusted based on the duration of each stage boundary and the control cycle; the duration of the stage boundary indicates the duration of the preheating stage, injection stage, holding stage and cooling stage; the control cycle determines how often the system checks the current temperature and adjusts its temperature control according to the current temperature.

[0062] The multiple values ​​extracted at this time will be based on the duration and temperature values ​​under the corresponding control cycle, and the average temperature, standard deviation, lag time, overshoot, and number of abnormal event types will be extracted. If the control cycle is a smaller value scenario, these values ​​are obtained based on the duration.

[0063] S42 , the average temperature, standard deviation, lag time, overshoot, and number of abnormal event types of the area to be adjusted are spliced ​​according to the boundaries of each stage to form a temperature change signal.

[0064] At this time, the data to be processed is mainly spliced, the values ​​in the temperature change signal are continuously updated, and the updated values ​​are added to the temperature change signal. At the same time, the number of abnormal event types currently identified is recorded to describe the abnormal pattern of the current temperature change signal, and the temperature change signal is threshold detected to identify the abnormal pattern of the temperature change signal.

[0065] S43, using the average temperature, standard deviation, lag time, overshoot and number of abnormal event types of the area to be adjusted as query rules, and combining each rule for retrieval query in turn. If there is an abnormal pattern corresponding to the current rule value in the database, the area to be adjusted is marked and output as an abnormal label of the temperature change signal.

[0066] The above abnormal pattern can be expressed as follows: when the average temperature is greater than 185°C or less than 17°C, a temperature out-of-bounds alarm is triggered; or when the standard deviation is greater than 1°C, a temperature fluctuation abnormality alarm is triggered. At this time, the five values ​​extracted from the area to be adjusted will be queried in a one-to-one or many-to-one form to find out whether there is a related abnormal pattern in the current area to be adjusted. At this time, compared with the obtained event type, it tends to describe more specifically the abnormal conditions that may occur at the boundary of each stage in the area to be adjusted, and mark these abnormalities with as specific labels as possible to indicate whether the data contained in the temperature change signal will affect the overall mold production.

[0067] In one embodiment of the present invention, step S5 is used to use the data queried in the temperature change signal as a node to set a strategy tree. The nodes of the strategy tree include not only abnormal labels and temperature values, but also the location of the logical area corresponding to the temperature value, event type, association rules and other processing content, and the nodes of the strategy tree can all perform strategy matching based on abnormal labels and specific temperature values, so that the sequence data covered by the temperature change signal is used as the root node of the decision tree, and the abnormal labels and stages of the temperature change signal are used as its branch nodes to describe the strategies adopted by the leaf nodes under multiple branch nodes, to illustrate the part of the current temperature change signal that needs to be responded to and processed first; form a path of temperature change signal-abnormal label-logical area-stage boundary-temperature value situation-temperature control strategy to implement the temperature control strategy for processing some specific temperature anomalies; at this time, weights are assigned to the strategy tree nodes according to the content contained in the nodes in the strategy tree, such as setting weights based on the content described by the abnormal labels, and then setting weights for the parts with problems in the logical area, stage boundary and temperature value in turn, and finally The path with the highest priority or high weight under its path is selected as the temperature control strategy for priority processing and implementation; as for the weight of the abnormal label of each node on the strategy tree, it is set based on the flag value of the severity of different abnormal labels in the database. This abnormal label will show different degrees of impact due to different description content. At this time, its weight is mainly set based on the flag content of the relevant abnormal situation in the database; the node of the logical area uses the ratio of the frequency of abnormalities in the logical area to the total abnormality frequency as its weight. For the node describing the current stage, such as the stage boundary, the ratio of the frequency of abnormalities in each stage to the total abnormality frequency is used as its weight; as for the node related to the temperature value, the ratio of its value to the total value in the current sampling and analysis data is used as its weight to illustrate the path corresponding to the currently executed temperature control strategy. The temperature value situation will include multiple value contents such as temperature value, standard deviation, variance, etc. covered in the temperature change signal. The current temperature-related values ​​are all regarded as temperature value situations, and the content of each temperature value combination is adapted to the content in the stable control strategy.

[0068] like Figure 5 As shown, the implementation method of step S5 includes: S51, taking the temperature value, abnormal label and corresponding logical area in the temperature change signal as the decision basis for strategy comparison; taking the stage boundary associated with the temperature change signal as the decision constraint, and taking the control parameters, control strategy pointer and strategy coordination weight of the temperature control strategy in the database as the decision mapping, and constructing a strategy tree with the decision basis, decision constraints and decision mapping.

[0069] S52, connect the nodes in the strategy tree to form multiple strategy paths, compare the path parameter values ​​of each strategy path with the temperature control strategy that the strategy path ultimately points to, and take the path parameter value with the largest value as the output target temperature control strategy.

[0070] The target temperature control strategy represents a temperature control strategy corresponding to the maximum path parameter value under the current anomaly label. This temperature control strategy is adjusted according to the control strategy pointer contained therein. The path parameter value is represented as the sum of the weights of the nodes included in the strategy path. It should be noted that the weight obtained here does not include the strategy coordination weight corresponding to the temperature control strategy.

[0071] The control parameters of the above-mentioned temperature control strategy represent the target temperature value adjusted for the current temperature change signal and the allowable temperature fluctuation range. These data are used to describe the values ​​that the strategy tree needs to adjust and achieve in response to the corresponding abnormalities in the temperature change signal; the control strategy pointer indicates the identifier pointing to the specific strategy execution, such as PID parameter adjustment, heating power limiting and cooling water flow compensation, which directly point to the content of the temperature control strategy execution method; the strategy coordination weight represents the weight set for the specific content executed by the current temperature control strategy, which is used to illustrate the execution order under the current temperature control strategy that needs to execute multiple control strategy policies. This weight is set in the database in advance.

[0072] Preferably, the implementation method of the strategy tree also includes: using the currently input temperature change signal as the root node of the strategy tree, using the abnormal label, logical area and stage boundary of the temperature change signal as branch nodes of the root node in sequence, using the temperature value of the temperature change signal and the temperature control strategy as secondary nodes of the branch node, and using the control strategy pointer and strategy coordination weight of the temperature control strategy as leaf nodes to form a strategy tree.

[0073] Preferably, when forming a strategy tree, it is also necessary to perform decision mapping between the temperature value of the temperature change signal and the temperature control strategy, and compare the temperature value of the current temperature change signal with the control parameters of the temperature control strategy in turn. When the control parameters of the temperature control strategy are met, the control strategy pointer and strategy coordination weight of the current temperature control strategy are used as the output target temperature control strategy.

[0074] When comparing the temperature change signal with the control parameters of the temperature control strategy, by comparing the temperature value and the temperature fluctuation range, only when the current temperature control strategy meets the abnormal label, logical area and stage boundary of the temperature change signal and corresponds to the temperature value of the temperature change signal, the relevant content in the temperature control strategy will be executed according to the control strategy pointer it contains, and gradually implemented from large to small through the size of the strategy coordination weight.

[0075] If it is not satisfied, the temperature control strategy of the nearest neighbor in the historical data will be used as the target temperature control strategy currently being executed. At this time, when the temperature value of the current temperature change signal is used for querying, if it does not correspond to the temperature fluctuation range of the temperature control strategy, then the content executed for this situation in the historical decision will be used to implement the corresponding strategy. This means that when there is a strategy that cannot completely match the current situation, the strategy path corresponding to the current temperature change signal in the strategy tree is converted into a vector form, and the cosine similarity is calculated with other temperature control strategies in the historical data in the vector form, and the strategy with the largest cosine similarity is used as the temperature control strategy of its nearest neighbor, or the temperature control strategy that is closest to the current time sequence in time is used as the target temperature control strategy to implement the strategy fallback processing method to complete the dynamic strategy control for the current mold production.

[0076] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention, which are still covered by the scope of protection of the present invention.

Claims

1. A method for precisely controlling the temperature of a flexible mold, characterized in that: include: S1, obtaining a three-dimensional model of the current mold, partitioning the current mold according to shape features to form logical areas; S2: Collect the temperature time series data set of each logical area, analyze the temperature characteristics of each logical area under the stage boundary based on the number of offset points of each logical area, and define the event type and association rules corresponding to the temperature characteristics; S3, using the temperature characteristics of each logic area as input, calculating the temperature similarity of each logic area, determining the relationship between each logic area and the temperature characteristics, and marking the logic area that needs to be adjusted as the area to be adjusted; S4, analyzing the temperature operating status of the area to be adjusted, generating a temperature change signal based on the duration and control period of the area to be adjusted at each stage boundary, and extracting an abnormal label of the temperature change signal; S5, taking the temperature change signal as the basis for strategy comparison, taking the data of the temperature change signal at the boundary of each stage as the node of the strategy tree, and determining the target temperature control strategy according to the path parameter value of the strategy tree.

2. A method for precise temperature control of a flexible mold according to claim 1, characterized in that: The implementation of step S1 includes: S11, performing geometric topology analysis using the shape features of the current model as input to obtain key features of the mold; S12, dividing the mold surface into multiple logical partitions based on the key features of the mold.

3. The method for precise temperature control of a flexible mold according to claim 1, characterized in that: The implementation of step S2 further includes: Based on the temperature characteristics of the temperature time series data set at the stage boundary, the temperature values ​​within each stage boundary are identified to obtain the temperature change rate of each data point; Based on the temperature change rate of each data point, cluster analysis is performed on each data point of the temperature time series data set to obtain at least one set of temperature change rate offset points and neighborhood temperature change offset points; When the number of temperature change rate offset points and neighborhood temperature change offset points exceeds the preset number threshold, the corresponding logical area will be output to obtain the number of offset points of each logical area, and the temperature change rate offset points and neighborhood temperature change offset points will be used as offset points for identifying each logical area.

4. A method for precise temperature control of a flexible mold according to claim 3, characterized in that: The implementation method of obtaining the number of offset points of each logical area also includes: Intercept the time points corresponding to the number of offset points, view the distribution time period of each deviation point in the temperature time series dataset, and output the temperature time series dataset in sequence based on the number of offset points contained in the distribution time period and the time period with the largest number of offset points as the center.

5. The method for precise temperature control of a flexible mold according to claim 1, characterized in that: The implementation of step S2 further includes: S21, based on any temperature feature at the stage boundary, clustering the temperature values ​​corresponding to each data point with the data having deviations in the temperature feature at each moment, setting a temperature identifier with the temperature value of each cluster after clustering, and identifying the event type of each temperature feature under the temperature value; S22, aligning the time series corresponding to the first triggering moment of each event type with the phase boundary, and calculating the time offset of the triggering moment of each event type relative to the phase boundary as the triggering time of each event type; S23, checking the conditional probability of each temperature feature and event type, generating association rules corresponding to each temperature feature based on the conditional probability and trigger time corresponding to each temperature feature and event type, and adding the association rules of each temperature feature to the boundaries of each stage.

6. The method for precise temperature control of a flexible mold according to claim 1, characterized in that: The implementation of step S3 includes: S31, calculating the temperature similarity of the temperature features of the current logical partition based on the event type and association rules corresponding to the temperature features, and performing temperature similarity calculations in sequence using the temperature mean, standard deviation, and range within each logical partition; S32, after the temperature similarity calculation is completed, hierarchical clustering is performed based on the temperature similarity between the logical areas to obtain multiple clusters; S33 , performing outlier region detection on the logical regions within each cluster, and marking the corresponding logical regions as regions to be adjusted based on the outlier detection results.

7. A method for precise temperature control of a flexible mold according to claim 6, characterized in that: When outlier region detection is performed on the logical region within each cluster, the implementation method further includes: If there are outlier areas, they are classified according to the mold opening temperature difference and cooling rate, and the abnormal mold opening temperature difference area and cooling rate abnormal area corresponding to the current temperature characteristics are divided, and the abnormal mold opening temperature difference area and cooling rate abnormal area are marked as areas to be adjusted.

8. The method for precise temperature control of a flexible mold according to claim 1, characterized in that: The implementation of step S4 includes: S41, extracting the average temperature, standard deviation, lag time, overshoot and number of abnormal event types of the area to be adjusted based on the duration and control period of each stage boundary; S42, the average temperature, standard deviation, lag time, overshoot, and number of abnormal event types of the area to be adjusted are spliced ​​according to the boundaries of each stage to form a temperature change signal; S43, using the average temperature, standard deviation, lag time, overshoot and number of abnormal event types of the area to be adjusted as query rules, and combining each rule for retrieval query in turn. If there is an abnormal pattern corresponding to the current rule value in the database, the area to be adjusted is marked and output as an abnormal label of the temperature change signal.

9. The method for precise temperature control of a flexible mold according to claim 1, characterized in that: The implementation of step S5 includes: S51: The temperature value, anomaly label, and corresponding logical region in the temperature change signal are used as a decision basis for strategy comparison; the stage boundary associated with the temperature change signal is used as a decision constraint; the control parameters, control strategy pointers, and strategy coordination weights of the temperature control strategy in the database are used as a decision mapping; and a strategy tree is constructed based on the decision basis, decision constraints, and decision mapping. S52, connect the nodes in the strategy tree to form multiple strategy paths, compare the path parameter values ​​of each strategy path with the temperature control strategy that the strategy path ultimately points to, and take the path parameter value with the largest value as the output target temperature control strategy.

10. A method for precise temperature control of a flexible mold according to claim 9, characterized in that: The implementation of the strategy tree also includes: The currently input temperature change signal is used as the root node of the strategy tree. The abnormal label, logical area and stage boundary of the temperature change signal are used as branch nodes of the root node in sequence. The temperature value of the temperature change signal and the temperature control strategy are used as secondary nodes of the branch node. The control strategy pointer and strategy coordination weight of the temperature control strategy are used as leaf nodes to form a strategy tree.

Images